Engineers Guide to Pharmaceuticals Production

The information in this book is given in good 
faith and belief in its accuracy, but does not 
imply the acceptance of any legal liability or 
responsibility whatsoever, by the Institution, or 
by the editors, for the consequences of its use or 
misuse in any particular circumstances. This 
disclaimer shall have effect only to the extent 
permitted by any applicable law. 
All rights reserved. No part of this publication 
may be reproduced, stored in a retrieval 
system, or transmitted, in any form or by any 
means, electronic, mechanical, photocopying, 
recording or otherwise, without the prior 
permission of the publisher. 

P r e f a c e 
The pharmaceutical industry aims to produce safe and effective medicines with 
efficiency and profitability. In order to achieve these aims, qualified personnel 
from many scientific and commercial disciplines are needed. The industry 
needs specialists with qualifications in biological, chemical, engineering and 
pharmaceutical sciences, but there is also a requirement for a wider knowledge 
of the integral parts of an innovative manufacturing company including 
research, development, manufacturing, distribution, marketing and sales. 
Chapter 1 sets the scene by introducing the essential stages, from the synthesis 
of a new chemical entity through to its development into a licensed medicine. 
Further education and advanced training for staff in the industry is needed 
through in-house or external courses. However, there is a distinct lack of 
detailed texts written by industrial experts. This book overcomes this deficiency 
in the area of pharmaceutical engineering and provides detailed information in 
all principal areas relevant to the manufacture of medicines. It will be a useful 
reference book for information on topics selected from the vast range of 
material covered in Chapters 2 to 11. Comprehensive coverage of each major 
topic, written by experts, provides valuable information for both newcomers 
and experienced personnel working in the pharmaceutical industry. 
Abbreviations and acronyms proliferate throughout the modern world and 
the pharmaceutical industry has its share. Fortunately, the editors have provided 
a list of acronyms and a glossary of terms most commonly used in the industry. 
The book is divided into ten main chapters, each covering specialist areas 
with their principal sub-sections clearly set out in the comprehensive list of 
contents at the beginning of the book. This feature will be very useful for those 
who need rapid access to detailed information in a specific area. 
Chapters 2 to 10 cover all the important aspects of the production of licensed 
medicines, as indicated in the following precis. 
Chapters 5 and 6 cover in detail primary and secondary production from the 
preparation of bulk bioactive substance by chemical synthesis, biotechnology 
and extraction from natural products, through to modern packaging technologies
required for the finished medicine. Chapter 8 deals with the design of utilities 
and services, as well as the associated areas of cleaning and maintenance. 
The design of facilities is continued in Chapter 9 which covers the planning, 
furnishing and provision of services in laboratories, whereas the special requirements 
for process development and pilot plant are presented in Chapter 10. 
Having provided an outline of the chapters dealing with production, we can 
turn towards the beginning of the book for coverage of regulatory matters and 
quality assurance. Chapter 2 is an outline of the main stages in the approval 
process, post-marketing evaluation and the European and US perspectives. 
The concepts and practices embodied in Good Manufacturing Practice are 
covered concisely in Chapter 3 with special reference to engineering aspects of 
pharmaceutical production, whereas validation and safety issues are presented 
in great detail in Chapters 4 and 7. 
Finally, in Chapter 11, the special requirements for the development and 
manufacture of modern bio-pharmaceutical products are dealt with in great 
detail with reference to small scale and pilot facilities. 
After six years working in research and development in the pharmaceutical 
industry, the rest of my career has been in academic pharmacy. Close contact 
with the industry has been maintained through education, training, research, 
consultancy and involvement with the design, delivery, assessment and external 
examinership of postgraduate diploma and MSc courses for advanced training 
of personnel in the industry. Such courses by universities or independent 
consultants provide course material of a high standard, but this should be 
supplemented by texts written by experts working in the industry. The 
Engineering Guide to Pharmaceutical Production provides an authoritative 
and detailed treatment of all major aspects related to the manufacture of 
medicines. 
Geoff Rowley 
Professor of Pharmaceutics, 
Institute of Pharmacy and Chemistry, 
University of Sunderland
L i s t o f a c r o n y m s 
The following is a list of acronyms used in this book. It is followed by a 
glossary of the more important validation terms. 
ADR Adverse Drug Reaction 
AGMP Automated Good Manufacturing Practice 
AGV Automated Guided Vehicles 
AHU Air Handling Unit 
ALARP As Low As Reasonably Practicable 
ANDA Abbreviated New Drug Application 
ANSI American National Standards Institute 
API Active Pharmaceutical Ingredient 
ASME American Society of Mechanical Engineers 
BATNEEC Best Available Techniques Not Entailing Excessive Costs 
BLl Biosafety Level 1 
BL2 Biosafety Level 2 
BL3 Biosafety Level 3 
BL4 Biosafety Level 4 
BMR Batch Manufacturing Record 
BMS Building Management System 
BOD Biological Oxygen Demand 
BP British Pharmacopeia 
BPC Bulk Pharmaceutical Chemical 
BPEO Best Practicable Environmental Option 
BS British Standard 
BSI British Standards Institution 
cAGMP Current Automated Good Manufacturing Practice 
CAMMS Computer Aided Maintenance Management System 
CCTV Closed Circuit Television 
CDER Centre for Drug Evaluation and Research 
CDM Construction (Design and Management) regulations
CFC Chlorofluorocarbons 
CFR Code of Federal Regulations 
CFU Colony Forming Unit 
cGCP Current Good Clinical Practice 
cGLP Current Good Laboratory Practice 
cGMP Current Good Manufacturing Practice 
CHAZOP Computer HAZOP 
CHIP Chemical Hazard Information and Packaging regulations 
CIMAH Control of Industrial Major Accident Hazards regulations 
CIP Clean In Place 
CMH Continuous Motion Horizontal 
COD Chemical Oxygen Demand 
COMAH Control Of Major Accident Hazards regulations 
COSHH Control Of Substances Hazardous to Health 
CPMP Committee on Proprietary Medicinal Products 
CPU Central Processing Unit 
CSS Continuous Sterilization System 
CV Curriculum Vitae 
DAF Dissolved Air Flotation 
DIN Deutsches Institut fur Normung 
DMF Drug Master File 
DNA Deoxyribonucleic Acid 
DOP Dioctyl Phthalate 
DQ Design Qualification 
EC European Community 
EEC European Economic Community 
EMEA European Agency for the Evaluation of Medical Products 
EPA Environmental Protection Agency 
EPDM Ethyl Propylene Diene Terapolymer 
ERP Enterprise Resource Planning 
EU European Union 
FAT Facility Acceptance Testing 
FBD Fluidized Bed Dryer 
FDA Food and Drug Administration 
FMEA Failure Mode Effects Analysis 
FS Functional Specification 
GAMP Good Automated Manufacturing Practice 
GC Gas Chromatograph 
GCP Good Clinical Practice 
GLP Good Laboratory Practice
GLSP Good Large Scale Practice 
GMP Good Manufacturing Practice 
GRP Glass Reinforced Plastic 
GSL General Sales List 
HAZOP Hazard and Operability Study 
HEPA High Efficiency Particulate Arrestor 
HFC Hydrofluorocarbons 
HIC Hydrophobic Interaction Chromatography 
HMAIP Her Majesty's Inspectorate of Air Pollution (now defunct) 
HMSO Her Majesty's Stationery Office 
HPLC High Pressure Liquid Chromatograph 
HS Hazard Study 
HSE Health and Safety Executive 
HSL HAZOP Study Leader 

HVAC Heating Ventilation and Air Conditioning 
IBC Intermediate Bulk Container 
ICH International Conference on Harmonization 
IDF International Diary Foundation 
IEC Ion Exchange Chromatography 
IEEE Institute of Electrical and Electronics Engineers 
IMV Intermittent Motion Vehicle 
IND Investigational New Drug Application 
I/O Inputs and Outputs 
IPA Iso Propyl Alcohol 
IPC Integrated Pollution Control 
IQ Installation Qualification 
ISO International Standards Organization 
ISPE International Society for Pharmaceutical Engineering 
LAAPC Local Authority Air Pollution Control 
LAF Laminar Air Flow 
LIMS Laboratory Information Management System 
LTHW Low Temperature Hot Water 
mAb Monoclonal Antibody 
MCA Medicines Control Agency 
MCB Master Cell Bank 
MCC Motor Control Centre 
MEL Maximum Exposure Limit 
MRA Mutual Recognition Agreement 
MRP Manufacturing Resource Planning 
MSDS Material Safety Data Sheet
NCE New Chemical Entity 
NDA New Drug Application 
NDT Non-Destructive Testing 
NICE National Institute for Chemical Excellence 
NMR Nuclear Magnetic Resonance 
OEL Occupational Exposure Limits 
OES Occupational Exposure Standards 
OQ Operational Qualification 
OSHA Occupational Safety & Health Administration 
OTC Over The Counter 
P Pharmacy only 
PBTB Polybutylene Teraphthalate 
PC Programmable Controller 
PCB Printed Circuit Board 
PDA Personal Digital Assistants 
PEG Polyethylene Glycol 
PFD Process Flow Diagram 
PHA Preliminary Hazard Assessment 
Ph.Eur European Pharmacopeia 
PHS Puck Handling Station 
P&ID Piping and Instrumentation Diagram 
PLA Product Licence Application 
PMI Positive Material Identification 
POM Prescription Only Medicines 
PP Polypropylene 
PPE Personal Protective Equipment 
PQ Performance Qualification 
PSF Performance Shaping Factors 
PTFE Polytetrafluoroethylene 
PV Process Validation 
PVC Polyvinyl Chloride 
PVDF Polyvinylidene Fluoride 
PW Purified Water 
QA Quality Assurance 
QC Quality Control 
QRA Quantitative Risk Assessment 
R&D Research and Development 
RF Radio Frequency 
RH Relative Humidity 
RHS Rolled Hollow Section
RIDDOR Reporting of Injuries, Disease and Dangerous Occurrences 
Regulations 
RP-HPLC Reverse Phase High Performance Liquid Chromatography 
SCADA Supervisory Control And Data Acquisition system 
SEC Size Exclusion Chromatography 
SHE Safety, Health and Environment 
SIP Sterilize In Place/Steam In Place 
SOP Standard Operating Procedure 
SS Suspended Solids 
THERP Technique for Human Error Rate Prediction 
TOC Total Organic Carbon 
TWA Time-Weighted Average 
UK United Kingdom 
UPVC Unplasticized Polyvinyl Chloride 
URS User Requirement Specification 
USA United States of America 
USP United States Pharmacopeia 
UV Ultra Violet 
VDU Visual Display Unit 
VMP Validation Master Plan 
VOC Volatile Organic Compound 
WCB Working Cell Bank 
WFI Water for Injection
G l o s s a r y 
The product specifications and acceptance/rejection 
criteria, such as acceptable quality level and unacceptable 
quality level, with an associated sampling plan, 
that are necessary for making a decision to accept or 
reject a lot or batch (or any other convenient subgroups 
of manufactured units). 
Levels or ranges that may be detrimental to end 
product quality, signalling a drift from normal operating 
conditions. 
Levels or ranges that signify a drift from normal 
operating conditions. These ranges are not perceived 
as being detrimental to end product quality, but 
corrective action should be taken to ensure that 
action levels are not obtained. 
An audit is a formal review of a product, manufacturing 
process, equipment, facility or system for conformance 
with regulations and quality standards. 
Any substance that is represented for use in a drug and 
that, when used in the manufacturing, processing or 
packaging of a drug, becomes an active ingredient or a 
finished dosage form of the drug. The term does not 
include intermediates used in the synthesis of such 
substances. 
Any substance that is intended for use as a component 
in a 'Drug Product', or a substance that is repackaged 
or relabelled for drug use. Such chemicals are usually 
Acceptance criteria 
Action levels 
Alert levels 
Audit 
Bulk drug 
substance 
Bulk pharmaceutical 
chemical
made by chemical synthesis, by processes involving 
fermentation, or by recovery from natural (animal, 
mineral or plant) materials. 
Comparison of a measurement standard or instrument 
of known accuracy with another standard or instrument 
to detect, correlate, report or eliminate by 
adjustment any variation in the accuracy of the item 
being compared. 
Documented statement by qualified authorities that a 
validation event has been done appropriately and that 
the results are acceptable. Certification is also used to 
denote the acceptance of the entire manufacturing 
facility as validated. 
A formal monitoring system by which qualified 
representatives of appropriate disciplines review 
proposed or actual changes that might affect validated 
status and take preventive or corrective action to 
ensure that the system retains its validated state of 
control. 
The validation of computers has been given a particular 
focus by the US FDA. 
Three documents have been published for agency 
and industry guidance. In February 1983, the agency 
published the Guide to Inspection of Computerized 
Systems in Drug Processing; in April 1987, the 
Technical Reference in Software Development Activities 
was published; on 16 April, 1987, the agency 
published Compliance Policy Guide 7132 in Computerized 
Drug Processing: Source Codes for Process 
Control Application Programmes. 
In the inspection guide, attention is called to both 
hardware and software; some key points being the 
quality of the location of the hardware unit as to 
extremes of environment, distances between CPU 
and peripheral devices, and proximity of input devices 
to the process being controlled; quality of signal 
conversion, for example, a signal converter may be 
sending inappropriate signals to a CPU; the need to 
Calibration 
Certification 
Computer validation 
Change control
systematically calibrate and check for accuracy of I/O 
devices; the inappropriateness and compatibility 
within the distributed system of command overrides, 
for example, can an override in one computer controlled 
process inadvertently alter the cycle of another 
process within the distributed system? Maintenance 
procedures are another matter of interest to the agency 
during an inspection. Other matters of concern are 
methods by which unauthorized programme changes 
are prevented, as inadvertent erasures, as well as 
methods of physical security. 
Hardware validation should include verification that 
the programme matches the assigned operational function. 
For example, the recording of multiple lot 
numbers of each component may not be within the 
programme, thus second or third lot numbers of one 
component may not be recorded. The hardware validation 
should also include worse case conditions; for 
example, the maximum number of alphanumeric code 
spaces should be long enough to accommodate the 
longest lot numbering system to be encountered. Software 
validation must be thoroughly documented — 
they should include the testing protocol, results, and 
persons responsible for reviewing and approving the 
validation. The FDA regards source code, i.e., the 
human readable form of the programme written in its 
original programming language, and its supporting 
documentation for application programmes used in 
any drug process control, to be part of the master 
production and control records within the meaning of 
2ICFR parts 210, 211 (Current Good Manufacturing 
Practice Regulations). 
As part of all validation efforts, conditions for 
revalidations are a requirement. 
Establishing documented evidence that the process 
being implemented can consistently produce a 
product meeting its predetermined specifications and 
quality attributes. This phase of validation activities 
typically involves careful monitoring/recording of the 
Concurrent 
validation
process parameters and extensive sampling/testing of 
the in-process and finished product during the initial 
implementation of the process. 
The documented evaluation of the construction or 
assembly of a piece of equipment, process or system 
to assure that construction or assembly agrees with the 
approved specifications, applicable codes and regulations, 
and good engineering practices. The conclusion 
of the evaluation should decidedly state that the 
equipment, process or system was or was not 
constructed in conformance with the specifications. 
Those process variables that are deemed important to 
the quality of the product being produced. 
A 'design review' is performed by a group of specialists 
(such as an Architect, a Quality Assurance 
Scientist, a HVAC Engineer, a Process Engineer, a 
Validation Specialist, a Civil Engineer and a Regulatory 
Affairs Specialist) to review engineering documents 
to ensure that the engineering design complies 
with the cGMPs for the facility. The thoroughness of 
the design review depends upon whether the engineering 
project is a feasibility study, a conceptual design, 
preliminary engineering, or detailed engineering. 
Minutes of all meetings for design review will be 
sent to team members and the client to show the 
compliance of the design to cGMPs. 
Substances recognized in the official USP; substances 
intended for use in the diagnosis, cure, mitigation or 
prevention of disease in man or other animals; 
substances (other than food) intended to affect the 
structure or any function of the body of man or other 
animals; substances intended for use as a component 
of any substances specified above but does not include 
devices or their components, parts or accessories. 
Dynamic attributes are classified into functional, 
operational and quality attributes, which are identified, 
Dynamic attributes 
Drug 
Critical process 
variables 
Design review 
Construction 
qualification
monitored, inspected and controlled during actual 
operation of the system. 
A control or operating parameter value that, if 
exceeded, may have adverse effects on the state of 
control of the process and/or on the quality of the 
product. 
Facilities are areas, rooms, spaces, such as receiving/ 
shipping, quarantine, rejected materials, approved 
materials warehouse, staging areas, process areas, etc. 
Functional attributes are such criteria as controls, 
instruments, interlocks, indicators, monitors, etc., 
that operate properly, are pointing in the correct 
direction, and valves that allow flow in the correct 
sequence. 
The minimum requirements by law for the manufacture, 
processing, packaging, holding or distribution of 
a material as established in Title 21 of the Code of 
Federal Regulations. 
An installation qualification protocol (IQ) contains the 
documented plans and details of procedures that are 
intended to verify specific static attributes of a facility, 
utility/system, or process equipment. Installation 
qualification (IQ), when executed, is also a documented 
verification that all key aspects of the installation 
adhere to the approved design intentions and that 
the manufacturer's recommendations are suitably 
considered. 
Any substance, whether isolated or not, which is 
produced by chemical, physical, or biological action 
at some stage in the production of a bulk pharmaceutical 
chemical and subsequently used at another stage 
in the production of that chemical. 
The time-frame from early stages of development 
until commercial use of the product or process is 
discontinued. 
Edge of failure 
Facilities 
Functional attributes 
Good manufacturing 
practice (GMP) 
Installation 
qualification 
protocol 
Intermediate (drug/ 
chemical) 
Life-cycle
The purpose of a master plan is to demonstrate a 
company's intent to comply with cGMPs and itemizes 
the elements that will be completed between the 
design of engineering and plant start-up. A typical 
master plan may contain, but is not limited to, the 
following elements: approvals, introduction, scope, 
glossary of terms, preliminary drawings/facility 
design, process description, list of utilities, process 
equipment list, list of protocols, list of SOPs, 
equipment matrices, validation schedule, protocol 
summaries, recommended tests, calibration, training, 
manpower estimate, key personnel (organization chart 
and resumes), protocol examples, SOP examples. 
A medical device is defined in the Federal Food Drug 
and Cosmetic Act Section 201(h) as: 
An instrument, apparatus, implement or contrivance 
intended for use in diagnosis, cure, mitigation, 
prevention or other treatment of disease in man or 
other animals, or intended to alter a bodily function or 
structure of man or other animal. 
This is the definition used in the code of Federal 
Regulations 21 parts 800 to 1299. Medical Devices. 
Operational attributes are such criteria as a utility/ 
system's capability to operate at rated ranges, capacities, 
intensities, such as: revolutions per minute, kg 
per square cm, temperature range, kg of steam per 
second, etc. 
An operation qualification (OQ) contains the plan and 
details of procedures to verify specific dynamic attributes 
of a utility/system or process equipment 
throughout its operated range, including worse case 
conditions. Operation qualification (OQ) when 
executed is documented verification that the system 
or subsystem performs as intended throughout all 
anticipated operating ranges. 
A range of values for a given process parameter that 
lie at or below a specified maximum operating value 
and/or at or above a specified minimum operating 
Master plan 
Medical devices 
Operation 
qualification 
protocol 
Operating range 
Operational 
attributes
value, and are specified on the production worksheet 
or the standard operating instruction. 
A process which is sufficient to provide at least a 12 
log reduction of microorganisms having a minimum 
D-Value of 1 minute. 
Process parameters are the properties or features that 
can be assigned values that are used as control levels 
or operating limits. Process parameters assure the 
product meets the desired specifications and quality. 
Examples might be: pressure at 5.2 psig, temperature 
at 37°C±0.5°C, flow rate at 10 ± l.Olmin"1, pH 
at 7.0 ±0.2. 
Process variables are the properties or features of a 
process which are not controlled or which change in 
time or by demand; process variables do not change 
product specifications or quality. 
Establishing documented evidence that provides a 
high degree of assurance that a specific process will 
consistently produce a product meeting its predetermined 
specifications and quality attributes. 
Process validation protocol (PV) is a documented 
plan, and detailed procedures to verify specific 
capabilities of a process equipment/system through 
the use of simulation material, such as the use of a 
nutrient broth in the validation of an aseptic filling 
process. 
A product is considered validated after completion of 
three successive successful lot size attempts. These 
validation lots are saleable. 
Validation conducted prior to the distribution of either 
a new product or a product made under a revised 
manufacturing process, where the revisions may have 
affected the product's characteristics, to ensure that 
the finished product meets all release requirements for 
functionality and safety. 
Overkill sterilization 
process 
Process parameters 
Process validation 
Process validation 
protocol 
Process variables 
Product validation 
Prospective 
validation
A protocol is defined in this book as a written plan 
stating how validation will be conducted. 
The activity of providing evidence that all the information 
necessary to determine that the product is fit 
for the intended use is gathered, evaluated and 
approved. 
Quality attributes refer to those measurable properties 
of a utility, system, device, process or product such as 
resistivity, impurities, particulate matter, microbial 
and endotoxin limits, chemical constituents and 
moisture content. 
The activity of measuring process and product parameters 
for comparison with specified standards to 
assure that they are within predetermined limits and, 
therefore, the product is acceptable for use. 
Validation of a process for a product already in 
distribution based upon establishing documented 
evidence through review/analysis of historical manufacturing 
and product testing data, to verify that a 
specific process can consistently produce a product 
meeting its predetermined specifications and quality 
attributes. In some cases a product may have been on 
the market without sufficient pre-market process validation. 
Retrospective validation can also be useful to 
augment initial pre-market prospective validation for 
new products or changed processes. 
Repetition of the validation process or a specific 
portion of it. 
Document that defines what something is by quantitatively 
measured values. Specifications are used to 
define raw materials, in-process materials, products, 
equipment and systems. 
Written procedures followed by trained operators to 
perform a step, operation, process, compounding or 
other discrete function in the manufacture or produc- 
Protocol 
Quality assurance 
Quality attributes 
Quality control 
Retrospective 
validation 
Revalidation 
Specifications 
Standard 
operating 
procedure (SOP)
tion of a bulk pharmaceutical chemical, biologic, drug 
or drug product. 
A condition in which all process parameters that can 
affect performance remain within such ranges that the 
process performs consistently and as intended. 
Static attributes may include conformance to a 
concept, design, code, practice, material/finish/ 
installation specifications and absence of unauthorized 
modifications. 
Utilities/systems are building mechanical equipment 
and include such things as heating, ventilation and air 
conditioning (HVAC) systems, process water, product 
water (purified water, water for injection), clean 
steam, process air, vacuum, gases, etc. Utilities/ 
systems include electro-mechanical or computerassisted 
instruments, controls, monitors, recorders, 
alarms, displays, interlocks, etc., which are associated 
with them. 
Establishing documented evidence to provide a high 
degree of assurance that a specific process will 
consistently produce a product meeting its predetermined 
specifications and quality. 
The collective activities related to validation. 
Validation protocols are written plans stating how 
validation will be conducted, including test parameters, 
product characteristics, production equipment, 
and decision points on what constitutes 
acceptable test results. There are protocols for installation 
qualification, operation qualification, process 
validation and product validation. When the protocols 
have been executed it is intended to produce documented 
evidence that the system has been validated. 
The scope identifies what is to be validated. In the 
instance of the manufacturing plant, this would 
include the elements that impact critically on the 
Validation scope 
Validation 
programme 
Validation 
protocols 
Validation 
State of control 
Static attributes 
Utilities/ systems
quality of the product. The elements requiring validation 
are facilities, utilities/systems, process equipment, 
process and product. 
A set of conditions (encompassing upper and lower 
processing limits and circumstances including those 
within standard operating procedures), which pose the 
greatest chance of process or product failure when 
compared to ideal conditions. Such conditions do not 
necessarily induce product or process failure. 
Worst case
xxiii This page has been reformatted by Knovel to provide easier navigation. 
Contents 
Preface v 
List of Acronyms ..... vii 
Glossary xiii 
1. Introduction .... 1 
2. Regulatory Aspects ........ 9 
2.1 Introduction ........ 9 
2.2 Key Stages in Drug Approval Process . 10 
2.3 Example of Requirements . 12 
2.4 Post-Marketing Evaluation . 13 
2.5 Procedures for Authorizing Medicinal 
Products in the European Union ........... 14 
2.6 European and US Regulatory Perspectives .......... 14 
3. Good Manufacturing Practice ........... 17 
3.1 Introduction ........ 17 
3.2 GMP Design Requirements . 22 
3.3 GMP Reviews of Design .... 34 
4. Validation ........ 38 
4.1 Introduction ........ 38 
4.2 Preliminary Activities .......... 41 
4.3 Validation Master Planning 44 
4.4 Development of Qualification Protocols and 
Reports .............. 51 
xxiv Contents 
This page has been reformatted by Knovel to provide easier navigation. 
4.5 Design Qualification (DQ) .. 53 
4.6 Installation Qualification (IQ) 55 
4.7 Operational Qualification (OQ) ............. 56 
4.8 Handover and Process Optimization .... 58 
4.9 Performance Qualification (PQ) ............ 59 
4.10 Process Validation (PV) ..... 60 
4.11 Cleaning Validation ............ 61 
4.12 Computer System Validation ................ 68 
4.13 Analytical Methods Validation ............... 71 
4.14 Change Control and Revalidation ......... 71 
5. Primary Production ......... 75 
5.1 Reaction ............ 75 
5.2 Key Unit Operations ........... 85 
5.3 Production Methods and Considerations .............. 96 
5.4 Principles for Layout of Bulk Production 
Facilities ............. 100 
5.5 Good Manufacturing Practice for BPC . 109 
6. Secondary Pharmaceutical Production ............ 111 
6.1 Products and Processes .... 111 
6.2 Principles of Layout and Building Design .............. 154 
6.3 The Operating Environment . 159 
6.4 Containment Issues ........... 176 
6.5 Packaging Operations ........ 177 
6.6 Warehousing and Materials Handling ... 188 
6.7 Automated Production Systems ........... 190 
6.8 Advanced Packaging Technologies ..... 192 
7. Safety, Health and Environment (SHE) ............. 202 
7.1 Introduction ........ 202 
7.2 SHE Management .............. 202 
Contents xxv 
This page has been reformatted by Knovel to provide easier navigation. 
7.3 Systems Approach to SHE 207 
7.4 Inherent SHE ..... 209 
7.5 Risk Assessment ................ 211 
7.6 Pharmaceutical Industry SHE Hazards 236 
7.7 Safety, Health and Environment Legislation ......... 257 
8. Design of Utilities and Services ........ 260 
8.1 Introduction ........ 260 
8.2 Objectives .......... 261 
8.3 Current Good Manufacturing Practice .. 262 
8.4 Design ............... 263 
8.5 Utility and Service System Design ........ 270 
8.6 Sizing of Systems for Batch Production ................ 287 
8.7 Solids Transfer .. 289 
8.8 Cleaning Systems .............. 289 
8.9 Effluent Treatment and Waste Minimization .......... 291 
8.10 General Engineering Practice Requirements ......... 297 
8.11 Installation ......... 299 
8.12 In-House Versus Contractors ............... 300 
8.13 Planned and Preventive Maintenance .. 301 
8.14 The Future? ....... 302 
9. Laboratory Design .......... 304 
9.1 Introduction ........ 304 
9.2 Planning a Laboratory ........ 307 
9.3 Furniture Design 321 
9.4 Fume Cupboards ............... 329 
9.5 Extraction Hoods ................ 336 
9.6 Utility Services ... 337 
9.7 Fume Extraction 337 
9.8 Air Flow Systems ............... 340 
9.9 Safety and Containment .... 344 
xxvi Contents 
This page has been reformatted by Knovel to provide easier navigation. 
10. Process Development Facilities and Pilot 
Plants .............. 346 
10.1 Introduction ........ 346 
10.2 Primary and Secondary Processing ..... 347 
10.3 Process Development ........ 347 
10.4 Small-Scale Pilot Facilities . 352 
10.5 Chemical Synthesis Pilot Plants ........... 361 
10.6 Physical Manipulation Pilot Plants ........ 368 
10.7 Final Formulation, Filling and Packing Pilot 
Plants 369 
10.8 Safety, Health and Environmental Reviews .......... 371 
10.9 Dispensaries ...... 371 
10.10 Optimization ...... 371 
10.11 Commissioning and Validation 
Management ..... 371 
11. Pilot Manufacturing Facilities for the 
Development and Manufacture of Bio- 
Pharmaceutical Products . 372 
11.1 Introduction ........ 372 
11.2 Regulatory, Design and Operating 
Considerations .. 373 
11.3 Primary Production ............. 388 
11.4 Secondary Production ........ 402 
11.5 Design of Facilities and Equipment ...... 417 
11.6 Process Utilities and Services ............... 442 
Index ... 447 
I n t r o d u c t i o n 
i 
Everyone is aware of the potential benefits of medicines and the patient takes 
them on trust expecting them to be fit for the purpose prescribed by the doctor 
or agrees with the claims of the manufacturer on the packaging or on 
advertisements. This book is a general introduction for all those involved in 
the engineering stages required for the manufacture of the active ingredient 
(primary manufacture) and its dosage forms (secondary manufacture). 
All staff working in or for the pharmaceutical industry have a great 
responsibility to ensure that the patient's trust is justified. Medicines made 
wrongly can have a great potential for harm. 
Most of the significant developments of medicines, as we know them, have 
occurred in the last 70 years. 
From ancient times, by a process of trial and error, man has used plants and 
other substances to produce certain pharmacological effects. The best example 
is probably alcohol, which has been developed by every culture. 
Alcohol has a number of well-known effects depending on the dosage 
used. In small amounts it causes flushing of the skin (vasodilatation), larger 
quantities produce a feeling of well being, and if the dose is further 
increased, loss of inhibition occurs leading to signs of aggression. Beyond 
aggression, somnolence occurs and indeed coma can supervene as the 
central nervous system becomes progressively depressed. This well-known 
continuum of effects illustrates very neatly the effect of increasing dosage 
over a period of time with a substance that is metabolized simply at a fairly 
constant rate. It further illustrates that where small quantities of a drug are 
useful, larger quantities are not necessarily better — in fact they are usually 
harmful. 
Using the trial and error technique, the good or harmful properties of various 
other materials were also discovered, for example, coca leaves — cocaine, or 
poppy juice — opium, which contains morphine. 
Today the pharmaceutical industry is faced with escalating research costs to 
develop new products. Once an active product has been discovered and proven
to be medically effective the manufacturer has to produce the active ingredient 
and process it into the most suitable dosage form. 
Speed to market is essential so that the manufacturer can maximize profits 
whilst the product has patent protection. Companies are now concentrating 
products at specific sites to reduce the time-scale from discovery to use, to give 
economics of scale and longer campaign runs. 
The manufacture of the active ingredient is known as primary production (see 
Chapter 5). Well-known examples of synthetic processes are shown in Figures 1.1 
and 1.2 (see pages 3 and 4). The manufacturing process for methylprednisolone 
(a steroid) is complex (see Figure 1.1), but it is relatively simple for 
phenylbutazone (see Figure 1.2). The processing to the final dosage form 
such as tablet, capsule (see Figure 1.3 on page 4), or injection, is known as 
secondary production (see Chapter 6). 
Bringing a mainstream drug to market can cost in excess of .200 million 
(300m US dollars). This involves research, development, manufacturing, 
distribution, marketing and sales. The time cycle from discovery to launch 
takes many years and will probably not be less than four years for a New 
Chemical Entity (NCE). Any reduction in this time-frame improves the 
company's profitability and generates income. 
Many companies conduct the early studies on NCE 's for safety, toxicity 
and blood levels using capsules. This is due to a very small amount of NCE being 
available and the ease of preparing the dosage form without loss of material. 
Only when larger quantities become available is a dosage form formulated as a 
tablet or other form. The product design process must take into account the 
demands of regulatory approval (manufacturing licences, validation), and 
variation in demand requiring flexibility of operation. The treatment of hay 
fever is a good example of a product only being in peak demand in spring and 
early summer. 
All companies will attempt to formulate oral solid dosage forms, such as a 
tablet or capsule, as this is the most convenient form for the patient to take and 
the easiest product to manufacture. An estimated 80-85 percent of the world's 
medicines are produced in this form. Not all products are effective from the oral 
route and other dosage forms such as injections, inhalation products, transdermals 
or suppositories are required. 
The discovery and isolation of a new drug substance and its development 
into a pharmaceutical dosage form is a costly and highly complex task 
involving many scientific disciplines. Figures 1.4 and 1.5 illustrate many of 
the steps involved. 
Figure 1.5 illustrates the various departments and disciplines that need to 
co-operate once it has been decided that the product will be marketed. This
figure assumes that facilities are available for manufacturing the active 
ingredient (primary manufacture). 
Failures by manufacturers led to the establishment of regulatory authorities 
initially in the USA, then in the UK and more recently in Europe. 
In 1938 in America sulphonamide elixir was contaminated by diethylene 
glycol resulting in a large number of deaths. This led to the Food, Drug and 
Hydrocortisone 
Acetate 
Cortisone 
6a-Methylprednisolone 
Figure 1.1 Synthetic route for 6a methylprednisolone
Diethyl 
Malonate 
Hydrazobenzene 
Phenylbutazone 
Figure 1.2 Synthetic route for phenylbutazone 
1 2 3 4 5 
6 7 8 9 10 
Pellet mixture 
Powder granulate 
Tablets 
1 st pellet type 
2nd pellet type 
Capsules 
Paste 
Figure 1.3 Various formulations filled into hard shell capsules
Activity and pre-clinical 
safety 2-4 years 
Development of formulations 
Bioavailability of formulations 
Stability tests on drugs and formulations 
Quality control methods devised 
Process development 
Detailed animal pharmacology 
Synthesis of radio labelled material 
Blood level methods developed 
Acute and 6 month toxicity studies 
Reproduction studies and teratology 
Absorption, excretion and metablism on 
animal species 
Outline clinical trial programme 
Establishment of manufacturing 
processes 
Plant design and buildings 
Development of sales formulation 
Bioavailability studies 
Package development 
Stability studies 
International clinical trials 
Detailed absorption, excretion, and 
metabolism studies in man 
0.5 
year 
2/3 
years 
Approx. 
1 year 
Approx. 
1 year 
Approx. 
2 years 
2/3 
years 
Synthesis of 
active substance 
Pharmacokinetic 
trials 
Approx. 8-10,000 
Potential candidate 
substances screened 
for therapeutic activity 
Toxicity 
trials 
Screened for 
pharmacological 
activity 
Discovery of 
active substance 
Pre-clinical 
trials 
Phase 1 Clinical 
trials 
Phase 2 
Phase 3 
Launch and 
sales 
Registration with 
health authorities 
IV Registration 
and launch 
Figure 1.4 Stages in a new product launch (simplified)
Cosmetic Act coming into force in the USA, followed by the establishment of 
the Food and Drug Administration (FDA). 
In 1962, there was the much publicized Thalidomide tragedy leading to the 
tightening up of the testing of drugs prior to marketing, and eventually to the 
Medicines Act 1968 in the UK. The Medicines Control Agency (MCA) was 
established to police the industry and there is now also the European Medicines 
Evaluation Agency (EMEA) and the National Institute for Chemical Excellence 
(NICE). 
Such legislation (see Chapter 2) has had a considerable impact on the 
design, construction, operation and on-going maintenance of pharmaceutical 
production facilities. 
The FDA, the MCA and European Regulatory Authorities have all issued 
codes of Good Manufacturing Practice, providing basic ground rules to ensure 
adequate patient protection from hazards associated with the poor design of 
manufacturing processes. Chapter 3 provides background knowledge on the 
regulatory framework and constraints on the manufacturer. 
Validation has been introduced in recent years. This was defined by the FDA 
as the act of establishing documentary evidence to provide a high degree of 
assurance that a specific process will consistently produce a product meeting its 
pre-determined specifications and quality attributes. Chapter 4 provides details 
of the documentation required including concepts such as the User Requirement 
Specification (URS), Validation Master Plan (VMP), Design Qualification 
(DQ), Installation Qualification (IQ), Operational Qualification (OQ) and 
Performance Qualification (PQ). 
It is important that the designer understands these requirements because it is 
far easier to collect validation documentation throughout the design process 
rather than to attempt to do so post-design, often known as retrospective 
validation. 
Chapter 5 deals with primary production, or manufacture of the active 
ingredient. For many years designers considered this to be no different to the 
manufacture of any other chemical, but codes of good manufacturing practice 
and validation now apply. Reactions and other key unit operations are discussed 
with ideas for layouts to satisfy good manufacturing practice and other 
regulator requirements. 
Chapter 6 is a comprehensive review of secondary production, turning the 
active ingredient into the dosage form. 
Chapter 7 covering safety, health and environment explains how risks to 
these are managed in the pharmaceutical industry and how effective process 
design can eliminate or control them.
Figure 1.5 Implementation stages of the launch of a new product 
The reader may ask why Chapter 8 has been included as process utilities and 
services are common throughout all industry. This chapter concentrates on 
aspects that are particularly relevant to the pharmaceutical industry. Regulatory 
authority inspectors, when inspecting plants, spend a lot of time looking at 
Other 
launches Launch 
Overall duration 1-5 years 
LfK distribution Export distribution 
4-20 weeks 
v %s^appro\^|>№dti^"":- 
4 weeks 
Obtain, test & approve 
packing materials 
5-28 weeks 
Provide IvTPIPS data 
plan production 
2-5 weeks 
Design/specify pack 
agree launch stocks 
4-29 weeks 
Product appreciation 
and approval 
* Initiate project 
- Capital requirements 
- Initial plant costs 
- Sourcing 
Engineering 
Pharmaceutical 
Development 
Marketing 
Validation 
Production 
Multi-discipline
water supplies, compressed air systems, air conditioning and cleaning systems 
which are all in the designer's control. 
Much of the book is about the production of the active ingredient and dosage 
forms. However, Quality Assurance departments have an important part to play 
in ensuring medicines are of an appropriate quality. In fact, regulatory 
authorities demand that a Qualified Person (usually from the QA department) 
is legally responsible for the release for sale of the manufactured product. 
Chapter 9 focuses on the design of quality control laboratories which form an 
important part of the quality assurance process. 
In a similar way, process development facilities and pilot plants are an 
integral part of the development of the manufacturing process for the active 
ingredient and its dosage form, particularly in the preparation of clinical trials. 
Chapter 10 gives ideas on the design, construction, commissioning and 
validation of these facilities. 
Chapter 11 is a review of the special requirements of Bio-pharmaceutical 
products particularly for pilot-scale manufacture of these products.
2.1 Introduction 
The pharmaceutical industry is distinctive from many other industries in the 
amount of attention paid to it by regulatory authorities. In all industries there 
are regulations relating to safety and the environment, rules and directions for 
services and recommendations from a wide range of authorities about design 
and maintenance of facilities. Engineers in the pharmaceutical industry also 
have to cope with a myriad of medicines regulations throughout the design and 
engineering process. Whilst it is not essential to have a detailed knowledge of 
all aspects of the regulations of medicinal products, facilities and processes, 
engineers should at least recognize that many of these regulations are restrictive 
or impose additional requirements. When products and processes have been 
registered with the regulatory authorities, it can be difficult and time-consuming 
to alter these specifications. This makes it important to be aware of the 
registered processes and quality control systems throughout the design. 
In the UK, medicines are regulated by the Medicines Control Agency 
(MCA). The MCA was launched as an Executive Agency of the UK Department 
of Health in July 1991. The MCA's primary objective is to safeguard 
public health by ensuring that all medicines on the UK market meet appropriate 
standards of safety, quality and efficacy. Safety aspects cover potential or actual 
harmful effects; quality relates to development and manufacture; and efficacy is 
a measure of the beneficial effect of the medicine on patients. The MCA 
operates a system of licensing before the marketing of medicines, monitoring of 
medicines and acting on safety concerns after they have been placed on the 
market, and checking standards of pharmaceutical manufacture and wholesaling. 
The MCA is responsible for enforcing these requirements. It represents 
UK pharmaceutical regulatory interests internationally; publishing quality 
standards for drug substances through the British Pharmacopoeia. 
A medicinal product (also known as a drug product) is any substance or 
article that is administered for a medicinal purpose. This includes treating or 
preventing disease, diagnosing disease, contraception, anaesthesia and preventing 
or interfering with a normal physiological function. 
JOHN WELBOURN 
R e g u l a t o r y a s p e c t s 
2
In all cases, the product must be fit for the purpose for which it is intended. 
From the consumer's point of view this could be a single tablet, but each tablet 
cannot be tested to ensure it is of the correct quality as many of the tests needed 
to demonstrate this are destructive. Manufacturers have to assure quality by 
ensuring all aspects of the process are consistent every time. 
As a result of well-publicized failures, resulting in patients deaths, regulations 
have become more and more stringent. Regulation is now achieved 
through the licensing of both the product and the facilities in which it is 
manufactured and the monitoring of medicines after a licence has been granted. 
The way medicinal products are supplied depends upon the nature and the 
historical experience of the product. Products may be Prescription Only 
Medicines (POM), Pharmacy only (P) or General Sales List (GSL). This 
categorization provides an important element in the control of medicinal 
products. 
In the UK, the Medicines Act 1968 and the Poisons Act 1972, together with 
the Misuse of Drugs Act 1971, regulate all retail and wholesale dealings in 
medicines and poisons. Certain non-medicinal poisons and chemicals are also 
subject to the labelling requirements of the Chemicals Hazard Information and 
Packaging Regulations (CHIP). 
It is important to appreciate at the outset that the Medicines Act 1968 applies 
only to substances where they are used as medicinal products or as ingredients 
in medicinal products. 
2.2 Key stages in drug approval process 
To obtain the evidence needed to show whether a drug is safe and effective, a 
pharmaceutical company will normally embark on a relatively lengthy process 
of drug evaluation and testing. Typically this will begin with studies of the drug 
in animals (preclinical studies) and then in humans (clinical studies). The 
purpose of preclinical testing is two-fold. Firstly, it is used as an aid to assessing 
whether initial human studies will be acceptably safe, and secondly, such 
studies are conducted to predict the therapeutic activity of the drug. If the drug 
looks promising, human clinical studies are proposed. In the USA, for example, 
this requires the submission of an Investigational New Drug Application (IND) 
to the regulatory authority, which in this case would be the Food and Drug 
Administration (FDA). 
The IND must contain sufficient information about the investigational drug 
to show it is reasonably safe to begin human testing. An IND for a drug not 
previously tested in human subjects will normally include the results of
preclinical studies, the protocols for the planned human tests, and information 
on the composition, source and method of manufacture of the drug. 
Provided the IND application is successful, drug testing in humans then 
proceeds progressively through three phases (called Phase 1, 2 and 3). 
Phase 1 includes the initial introduction of an investigational drug into 
humans and consists of short-term studies in a small number of healthy 
subjects, or patients with the target disease, to determine the metabolism and 
basic pharmacological and toxicological properties of the drug, and if 
possible, to obtain preliminary evidence of effectiveness. 
Phase 2 consists of larger, more detailed studies; usually including the first 
controlled clinical studies intended to assess the effectiveness of the drug and 
to determine the common short-term side effects and risks of the drug. 
Phase 3 studies are expanded controlled and uncontrolled trials. They are 
performed after preliminary evidence of effectiveness has been established 
and are designed to gather the additional information necessary to evaluate 
the overall benefit-risk relationship of the drug and to provide an adequate 
basis for professional labelling. 
If the results appear to be favourable at the end of clinical trials and the 
company decides to market the new product, they must first submit an 
application to do this. In the USA the company must submit the results of 
the investigational studies to the FDA in the form of a New Drug Application 
(NDA). The NDA must contain: 
full reports of the studies (both preclinical and clinical) to demonstrate the 
safety and effectiveness of the drug; 
a description of the components, chemical formulation, and manufacturing 
controls; 
samples of the drug itself and of the proposed labelling. 
Many companies choose to prepare a Drug Master File (DMF) to support 
the NDA. A DMF is submitted to the FDA to provide detailed information 
about facilities, processes or articles used in the manufacturing, processing, 
packaging and storage of one or more human drugs. In exceptional cases, a 
DMF may also be used to provide animal or clinical data. A DMF is submitted 
solely at the discretion of the holder, the information being used in support of 
the NDA. 
The application is reviewed. Typically this includes reviews of product 
chemistry, labelling, bio-equivalency, clinical data and toxicity. In addition, and 
of particular relevance to pharmaceutical engineers, the review will also include 
a pre-approval inspection of the facilities in which the drug is manufactured.
The pre-approval inspection will generally consist of a review of the 
facilities, procedures, validation (discussed in Chapter 4) and controls associated 
with formulation development, analytical method development, clinical 
trial manufacturing, manufacturing (if applicable), quality control laboratories, 
bulk chemical sources and contract operations. If the application is successful 
the pharmaceutical company will receive approval to market the product. 
A similar (although not identical) situation exists in Europe. For example, in 
the UK regulation is achieved through a Clinical Trial Certificate, Animal Test 
Certificate and Product Licence (also in certain circumstances Product Licence 
of Right and Reviewed Product Licence) for the product and a Manufacturer's 
Licence, Assembly Only Licence, Special Manufacturer's Licence, Wholesale 
Licence and Wholesale Import Licence for the Manufacturer/Supplier. 
2.3 Example of requirements 
An example of the 'regulatory environment' in the UK is summarized in 
Figure 2.1: 
Figure 2.1: The UK regulatory environment 
GLP is concerned with the organizational processes and the conditions under 
which laboratory studies are planned, performed, monitored, reported and 
recorded. The UK GLP regulations (Statutory Instruments No. 654) came 
into force in April 1997 and are monitored by the UK GLP Monitoring 
Key: 
GLP = Good Laboratory Practice 
GCP = Good Clinical Practice 
GMP = Good Manufacturing Practice 
Distribution / 
Sale and Supply 
Routine 
Production 
Wholesale Dealers 
Licence 
Manufacturers 
Licence 
(Manufacturing 
Authorization) 
Market Launch 
Clinical Trials 
Laboratory Trials 
Initial Research 
Animal Test 
Certificate 
Clinical Trial 
Certificate 
Product Licence 
(Marketing 
Authorization) 
Research 
Pharmacology 
Toxicology 
GLP 
GLP 
Development 
Formulation Development 
Analytical Method 
Clinical Manufacture 
Clinical Studies 
GMP 
GMP 
GMP 
GCP 
Manufacture 
Purchasing 
Production 
Testing 
Storage / Distribution 
Traceability / Recall 
GMP 
GMP 
GMP 
GMP 
GMP
Authority, which is part of the MCA. Currently about 150 test facilities are 
registered under the scheme and are inspected on a two-year cycle. 
GCP is 'a standard for the design, conduct, performance, monitoring, 
auditing, recording, analysis and reporting of clinical trials that provide 
assurance that the data and reported results are credible and accurate, and 
that the rights, integrity, and confidentiality of trial subjects are protected' 
(Definition from the International Conference on Harmonization (ICH) Note 
for Guidance on Good Clinical Practice (CPMP/ICH/135/95)). In the UK, the 
GCP Compliance Unit was established within the Inspection and Enforcement 
Division of the MCA in 1996. GCP inspectors assess compliance with the 
requirements of GCP guidelines and regulations, which involves conducting 
on-site inspections at pharmaceutical sponsor companies, contract research 
organizations' investigational sites and other facilities involved in clinical 
research. 
GMP is 'the part of Quality Assurance (QA) which ensures that products are 
consistently produced and controlled to the quality standards appropriate to 
their intended use and as required by marketing authorization or product 
specification.' (Definition from the EU Guide To Good Manufacturing Practice 
and Good Distribution Practice). GMP is discussed in more detail in Chapter 3. 
2.4 Post-marketing evaluation 
2.4.1 Pharmacovigilance 
No matter how extensive the pre-clinical work in animals and clinical trials in 
patients, certain adverse effects may not be detected until a very large number 
of people have received the new drug product. The conditions under which 
patients are studied pre-marketing do not necessarily reflect the way the new 
drug product will be used in hospitals or in general practice. Pharmacovigilance 
is the process of monitoring medicines as used in everyday practice to: 
identify previously unrecognized (or changes in) patterns of adverse effects; 
assess the risks and benefits of medicines in order to determine what action, 
if any, is necessary to improve their safe use; 
provide information to users to optimize safe and effective use of medicines; 
monitor the impact of any action taken. 
Information from many different sources is used for pharmacovigilance 
including spontaneous adverse drug reaction (ADR) reporting schemes, 
clinical and epidemiological studies, world literature, morbidity and mortality
databases. In the UK the MCA runs the spontaneous adverse drug reaction 
reporting scheme (called the Yellow Card Reporting Scheme) which receives 
reports of suspected drug reactions from doctors, dentists, hospital pharmacists 
and coroners. The scheme provides an early warning of adverse effects of 
medicines. 
2.4.2 Variations and renewal of marketing authorizations 
Drug products may undergo changes over time in relation to production, 
distribution and use. These will require authorization by the licensing agency. 
Also, authorizations are normally renewed on a regular period — marketing 
authorizations are valid for five years in the UK. 
2.5 Procedures for authorizing medicinal products in 
the European Union 
In 1995 a new European system for the authorization of medicinal products 
came into effect, and a new agency was established — the European Medicines 
Evaluation Agency (EMEA) based in London, UK. Two new registration 
procedures for human and veterinarian medicinal products have become 
available. The first system, known as the centralized procedure, is compulsory 
for medicinal products derived from biotechnology and is available at the 
request of companies for other innovative new products. Applications are 
submitted directly to the EMEA who undertake the evaluation and submit their 
opinion to the European Commission. The European Commission then issue a 
single market authorization. 
The second system, known as the decentralized procedure, applies to the 
majority of conventional medicinal products and is based upon the principal of 
mutual recognition of national authorizations. It provides for the extension of 
the marketing authorization granted by one Member State to one or more other 
Member States identified by the applicant. 
2.6 European and US regulatory perspectives 
On the 18 May 1998, the European Union and the USA signed a 'Joint 
Declaration to the agreement on Mutual Recognition between the EU and the 
USA'. This agreement lays down the framework for mutual recognition of 
GMP regulations under the principal of 'equivalence' and the mutual recognition 
of pre-approval and post-approval inspections.
The agreement covers human medicinal products (prescription and nonprescription 
drugs, biologicals including vaccines and immunologicals); veterinary 
pharmaceuticals (prescription and non-prescription drugs premixes and 
preparations for medicated feeds); active pharmaceutical ingredients and 
intermediate product, starting materials, bulk pharmaceuticals. The agreement 
excludes human blood, human plasma, human tissues and organs, veterinary 
immunologicals, human plasma derivatives, investigational medicinal products, 
human radiopharmaceuticals and medicinal gases. 
Reading list 
1. Rules and guidance for pharmaceutical manufacturers and distributors 1997. 
London. The Stationery Office, 1997. ISBN 0 11 321995 4. (Also known as the 
'Orange Guide'). (Incorporating EC Guides to Good Manufacturing Practice and 
Good Distribution Practice; EC GMP Directives (91/356/EEC & 91/412/EEC); 
Code of Practice for Qualified Persons and Guidance for Responsible Persons; 
Standard provisions for manufacturer's licences; Standard provisions for wholesale 
dealers licences; Guidance on reporting defective medicines). 
2. Good Laboratory Practice Regulations 1999 (GLP Regulations); Statutory Instrument 
1999/3106; Department of Health, The United Kingdom Good Laboratory 
Practice Monitoring Authority. 
3. Guide to UK GLP Regulations 1999, Feb 2000, Department of Health, The United 
Kingdom Good Laboratory Practice Monitoring Authority. 
4. International Conference on Harmonization (ICH) Note for Guidance on Good 
Clinical Practice (CPMP/ICH/135/95) 
5. Research Governance in the NHS, Guidance on Good Clinical Practice and Clinical 
Trials in the NHS, Department of Health 
6. Royal Pharmaceutical Society of Great Britain; Medicines, Ethics and Practices, 
A guide for Pharmacists, 18 Edition, July 1997. 
7. US Food and Drug Administration, Centre for Drug Evaluation and Research 
(CDER), Department of Health and Human Services; Code of Federal Regulations 
21 CFR (in particular, but not limited to, Parts 10b, 11, 210, 211, 600, 820). 
Guidance for Industry, including: Guideline For Drug Master Files September 
1989; Content and format of Investigational New Drug Applications (INDs) for 
Phase 1 Studies of Drugs, Including Well-Characterized, Therapeutic Biotechnology-
derived Products; Guideline for the Format and Content of the Microbiological 
Section of An Application (Docket No. 85D-0245); February 1987; Guideline 
for the Format and Content of the Chemistry, Manufacturing and Controls Section 
of An Application; Preparing Data for Electronic Submission in ANDAs [HTML] 
or [PDF], Sep 1999; Regulatory Submissions in Electronic Format; General 
Considerations Jan 1999; Regulatory Submissions in Electronic Format; New 
Drug Applications Jan 1999.
8. Agreement on Mutual Recognition between the European Community and the 
United States; US - EC MRA Pharmaceutical Good Manufacturing Practice 
Annexe; Sectorial Annex For Pharmaceutical Good Manufacturing Practice; 
Signed 18 May 1998; Exchange of Letters 30 October 1998; Published in Official 
Journal L 31, 4 February 1999 
Web Sites 
www.fda.gov/cder/guidance/index.htm 
www.emea.eu.int/ 
www.mca.gov.uk 
www.rpsgb.org.uk/
3.1 Introduction 
This chapter explains what is meant by current Good Manufacturing Practice 
(cGMP) and, in particular, how it applies to the engineering aspects of 
pharmaceutical production. The chapter also shows how it is possible to 
develop the GMP requirements to allow the facility to be engineered, and 
looks at the GMP design review process. 
3.1.1 Definition 
A key part of the control of medicinal products and facilities relates to GMP. 
The EU Guide To Good Manufacturing Practice and Good Distribution 
Practice defines GMP as 'the part of Quality Assurance (QA) which ensures 
that products are consistently produced and controlled to the quality standards 
appropriate to their intended use and as required by marketing authorization or 
product specification.' 
'Engineering for cGMP' may be defined as those activities performed 
throughout the project life-cycle, which ensure that it will be easy and natural 
to operate the completed facility in accordance with current Good Manufacturing 
Practice. 
The 'Project Life-Cycle' means from project inception through feasibility 
studies/conceptual design, engineering, construction, installation, start-up, 
operation, maintenance to final plant decommissioning or modification. 
GMP is controlled by the US Code of Federal Regulation (CFR) 21 in the 
USA. European pharmaceutical companies wishing to supply this market must 
also comply with these regulations. 
The various regulatory authorities produce different types of applicable 
documentation, which broadly fall into two categories: 
directives, rules, regulations, including for example: 
o US Code of Federal Regulations CFR 21 Parts 210 and 211 (Drug 
products) and CFR 21 Parts 600 to 680 (Biological products); 
G o o d m a n u f a c t u r i n g 
p r a c t i c e 
3 JOHN WELBOURN
o EU GMP Directive 91/356/EEC, Commission Directive Laying Down 
The Principles and Guidelines of Good Manufacturing Practice; 
o Rules Governing Medicinal Products For Human Use in the European 
Community, Volume IV; Guide to Good Manufacturing Practice for 
Manufacture of Medicinal Products. 
guides, guidelines, points to consider, including for example: 
o FDA Guide to Inspection of Bulk Pharmaceutical Chemical 
Manufacturing; 
o FDA Guide to Inspection of Validation of Cleaning; 
o FDA Guide to Inspection of Computerized Systems in Drug 
Processing; 
o FDA Guidelines on General Principles of Process Validation. 
Although not necessarily in a strict legal sense, the first category is 
mandatory and must be complied with. The second category, although classed 
as guides or guidelines, is also very important and generally must be complied 
with. 
The US Food and Drug Administration prepares guidelines under 10.90 
(b) of the regulations (21 CFR Part 10) to help with compliance. A 
comprehensive listing of potentially relevant guidelines, guidance and 
points to consider is provided by Center for Drug Evaluation and Research, 
'Guidelines for Regulations that are applicable to the Center for Drug 
Evaluation'. 
As well as the formal documents outlined above, there are other ways that 
cGMPs have evolved. These include the interpretation of the various rules and 
regulations and what is generally considered to be good practice by the 
industry. For example, the US Food and Drug Administration, through the 
freedom of information service, produces reports on inspections and inspection 
failures. These reports are in effect 'legal rulings' or interpretations of the 
regulations, e.g. Form 483. It is important to keep up to date on these 
requirements through publications such as GMP Trends or QC Gold Sheet. 
As a rule of thumb in terms of good practice, if more than 50% of the industry is 
moving over to something then it becomes cGMP. 
In addition to the codes laid down by the various regulatory authorities, there 
are parallel industrial quality standards that are deemed to apply to all 
industries. In Europe these tend to be grouped around ISO 9000, and the US 
equivalent are ANSI standards grouped around Q90. It is obvious that common 
standards should be applied and to this end the International Committees for 
Harmonization of Standards have published relevant recommendations as ICH 
guidelines.
3.1.2 General GMP requirements 
When first embarking on a new pharmaceutical facility, consideration will need 
to be made as to what cGMP requirements will apply to the project and how 
they will impact on the project life-cycle. These may vary. Although the words 
differ, there are common general requirements that run through virtually all the 
cGMPs worldwide. Common elements are: 
the establishment and maintenance of an effective quality assurance 
system; 
control of the process; 
personnel that are suitably qualified, trained and supervised; 
premises and equipment that have been located, designed, installed, operated 
and maintained to suit intended operations; 
maintenance of adequate records of all aspects of the process so that in the 
event of a problem being identified, an investigation can trace the complete 
history of the process, including how, when, and where it was produced, 
under what conditions and by whom (i.e. an audit trail); 
the prevention of contamination from any source, in particular from 
components, environment, premises and equipment by the use of suitable 
premises and equipment and through standard operating procedures. 
3.1.3 Project assessment to determine applicable standards 
Whilst the objectives of most cGMPs are generally the same (i.e. to safeguard 
consumers), the nature of Pharmaceuticals dictate that different sets of specific 
requirements have evolved depending upon the type of product, its stage of 
development or manufacture, and where it will be manufactured and sold. In 
addition the different regulatory authorities have prepared slightly different sets 
of standards, and apply them in different ways. One of the first steps when 
preparing to undertake a new project is to establish under what cGMP 
regulations the plant will operate. An assessment should be made to 
determine the: 
stage of product development; 
stage of production; 
category of the product and production processes employed; 
facility location and location of the markets that the facility will serve. 
Based on these factors a judgment can be made as to applicable standards 
that need be applied.
Stage of product development 
For the purposes of this book, the stage of product development may be divided 
into three parts: 
laboratory trials (pre-clinical animal trials); 
clinical trials; 
routine production. 
Generally speaking, cGMPs regulations do not apply during laboratory 
trials, 'Basic cGMPs' apply during clinical trials, and 'full cGMPs' apply 
during routine production. cGLPs (Current Good Laboratory Practice) may 
apply during laboratory trials and cGCPs (current Good Clinical Practice) may 
apply during clinical trials. 
Essentially GLP is concerned with the organizational processes and the 
conditions under which laboratory studies are planned, performed, monitored, 
reported and recorded. The UK GLP regulations (Statutory Instruments No. 
654) came into force in April 1997 and are monitored by the UK GLP 
Monitoring Authority, which is part of the Medicines Control Agency 
(MCA). Currently about 150 test facilities are registered under the scheme 
and are inspected on a two-year cycle. 
GCP is 'a standard for the design, conduct, performance, monitoring, 
auditing, recording, analysis and reporting of clinical trials to provide assurance 
that the data and reported results are credible and accurate, and that the rights, 
integrity, and confidentiality of trial subjects are protected' (Definition from the 
ICH Note for Guidance on Good Clinical Practice (CPMP/ICH/135/95)). In 
the UK the GCP Compliance Unit was established within the Inspection and 
Enforcement Division of the MCA in Autumn 1996. GCP inspectors assess 
compliance with the requirements of GCP guidelines and regulations, which 
involves conducting on-site inspections at pharmaceutical sponsor companies, 
contract research organizations investigational site and other facilities involved 
in clinical research. 
Stage of production 
The stage of production means what the facility is used for. The stages of 
production can be divided into the following four parts for the purposes of this 
book: 
Bulk Pharmaceutical Chemicals (BPCs) manufacturing; 
finished product manufacturing; 
packaging; 
warehousing/holding.
Different regulatory authorities apply certain specific cGMPs to different stages 
of production. In some cases facilities may be used for more than one stage of 
production, and in such cases more than one set of cGMPs may apply. 
Category of the product and production processes employed 
Broadly speaking most active ingredients are manufactured by one of the 
following routes: 
• chemical synthesis; 
• biotechnology; 
• blood derived; 
• animal or plant extraction. 
By the nature of these routes certain methods of production to produce 
dosage forms have evolved, and in each case specific GMP requirements have 
been developed. From a GMP point of view regulatory authorities categorize 
products as following: 
• sterile medical products: 
o terminally sterilized products; 
o aseptic preparations; 
• biological medical products: 
o microbial cultures, excluding those resulting from r-DNA techniques; 
o microbial cultures, including those resulting from r-DNA or hybridoma 
techniques; 
o extraction from biological tissues; 
o propagation of live agents in embryos or animals; 
• radiopharmaceuticals; 
• veterinary medicinal products; 
• medical gases; 
• herbal medicinal products; 
• liquids, creams and ointments; 
• metered dose aerosols; 
• products derived from blood; 
• tablets and hard gel capsules; 
• soft gel capsules; 
• transdermals; 
• implants. 
Clearly some pharmaceuticals represent a combination of these types. 
Facility location and market location 
cGMPs regulations are produced by a number of different countries or groups 
of countries world-wide, in addition to the World Health Organization. The key
regulations are from the USA, the EC and, to some extent, Japan. However no 
assumption must be made that these are suitable standards to apply. Clarification 
should be sought from the pharmaceutical manufacturer before the design 
commences. 
3.2 GMP design requirements 
Based on an assessment of the regulatory requirements (as described above) we 
can begin to define the GMP requirement for the project. Generally, issues and 
areas to be considered during the conceptual design phase will include: 
process issues: 
o closed or open (Is it to be completely contained with piping and equipment 
at all times or will it be exposed to the surrounding environment? In which 
case, what measures are to be taken to prevent/minimize contamination?); 
o level of batch to batch integrity required (Is simultaneous filling and 
emptying of vessels with different batches in known proportions or limits 
to be permitted? Do systems need to be engineered to be self-emptying? 
Will process systems need to be subject to cleaning, drying or sterilization 
between batches?); 
o level of segregation or containment required (Is it acceptable to manufacture 
product A in the same facilities as product B? Will processes be 
campaigned?); 

o level of production required. 
layout issues: 
o site location and layout (including existing site, brown field, green 
field, overall site layout and its suitability in terms of space, general 
layout); 
o facility layout (including cored versus linear layout; use of transfer 
corridors, segregation of areas, environment, containment strategy, modularization/
expansion, security and access control). 
automation strategy issues: 
o level of technology, use of design tools and models, number of layers — 
hierarchy; 
o availability/redundancy/maintainability, modularization/expansion; 
o instrumentation/cabling/field devices; 
o paperless batch records, electronic signatures. 
flow issues: 
o people (security, access, occupancy level, shift patterns);
o equipment (mobile or fixed, use of hard piping, flexible piping or 
disposable transfer bags, cross-contamination/mix-ups); 
o components/materials (materials handling systems, cross-contamination/
mix-ups). 
regulatory issues: 
o stage of product development, stage of production, category of the product 
and production processes employed, facility location, and location of the 
markets that the facility will serve. 
validation strategy issues: 
o validation required, validation team(s), validation plan(s). 
These basic requirements can then be refined for the various aspects of the 
project to allow the facility to be engineered. The following categories are 
suggested for guidance: 
facilities and environment; 
services and utilities; 
personnel flows; 
material flows; 
equipment flows; 
equipment design; 
computerized systems; 
maintenance and services; 
waste management; 
procedure and documentation. 
The following sections provide guidance to the type of criteria that will need 
to be considered. It may be appropriate to formulate these (and other applicable 
criteria), into a checklist for use during the development of the design and the 
design review. 
3.2.1 Facilities and environment 
These are the buildings, rooms and environment containing the production 
processes. They are of prime concern wherever the product or product 
components may be exposed. Typical criteria include the following: 
General considerations for the entire facility: 
local environmental considerations (including pollution and security); 
suitability/acceptability of physical segregation of processes for manufacturing 
and holding products, (such as segregation of production stages of the 
same, similar and different products and the use of dedicated or shared 
facilities);
overall layout of the facility (including use of cored environmental layout, 
position of technical and other non-production areas with respect to processing 
areas); 
general layout of production processes (logical flow through the facility with 
no/minimal cross-over of processing streams); 
pedestrian and vehicular access; 
pest control. 
Specific considerations for each area: 
available space and ergonomics for operators, equipment, materials and 
products; 
cavities/penetrations and how they are sealed; 
surfaces of walls, floors and ceilings (they should be easily cleanable, low 
particle shedding, minimal dust traps); 
materials of construction of the walls, floors and ceilings and their suitability 
for the intended operations; 
types of doors, windows, light fittings and void closures (for example, flush 
fitting, methods of sealing); 
provision and location of support utilities (both for production and maintenance/
housekeeping purposes); 
provision of suitable electrical outlets and communications systems (electrical 
sockets, telephones, speak through panels, network termination points, 
intercoms); 
furniture (quantity, suitability for the operators, surfaces, cleanability). 
Environment: 
This is of prime concern wherever the product or product components may be 
exposed. Typical criteria include: 
assessment of the environmental classification of the various areas against 
the level of quality required by the product (including non-viable particulate 
and microbiological contamination in both the unmanned and manned 
conditions); 
airflow regime and types of processing operations (turbulent or laminar, 
horizontal or vertical); 
air pressure differentials between areas; 
air change rates per hour; 
location of ventilation ducts relative to processing points and other equipment; 
emissions within the area (water vapour, compressed air, toxic fumes); 
humidity (comfort level, static hazards, growth promoting); 
environmental control at point of access to area;
illumination levels (relative to operations performed); 
adverse operating conditions (start-up/shutdown, dirty filters/blockages, 
power failure, redundancy); 
methods of monitoring, recording and controlling the environment (including 
temperature, pressure, humidity, air flow/velocity, particulate and 
microbiological); 
maintenance and cleaning of environmental systems (such as routine maintenance, 
safe change systems, redundancy). 
GMP requirements may generally be limited for external areas such as 
administration buildings, canteens, plant rooms. Staff will be able to access 
these areas in street clothing or working garments unrestricted by GMP, but 
there may be other reasons why specific garments are required. Personnel 
access should be controlled to all areas within a pharmaceutical facility (by 
access cards or pass-codes, for example). Pest control measures must be 
employed to prevent insect and rodent infestation. 
For areas such as packaging, warehousing, technical areas or where the 
product is fully contained in pipework, typical GMP requirements would 
include that: 
clothing consists of general factory overalls or lab coats and hats, with 
personnel to enter these areas via cloakroom facilities (primary change); 
environment air should be filtered to Eu 3 or above. Air pressure should 
normally be ambient; 
surfaces should be easily cleanable, finished flush and sealed. Equipment 
should be readily accessible for cleaning; 
measures should be taken to minimize the risk of cross-contamination. 
For areas where specific environmental control is required such as in 
secondary pharmaceutical manufacturing where products or ingredients are 
exposed, or for the preparation of solutions and components for terminally 
sterilized products, and in BPC plant areas handling exposed products or 
critical step intermediates, GMP requirements may include the following in 
addition to the above: 
• personnel must enter the area via a secondary change and the area must not 
contain toilets or eating areas; 
• process materials and components should enter via an airlock; 
• filtration and air circulation should achieve EU GMP Guide Grade D or 
equivalent; 
• drains should be sealed during normal operation with air breaks provided 
between sink or equipment outlets and floor drains;
compressed air exhausts should be vented outside the area; 
the preferred material of construction for process equipment is generally 
stainless steel and pipework lagging should be avoided where possible. 
Operators should be protected by mechanical guarding. Separate belt 
conveyors should be in different grade rooms, with dead plates at the wall 
opening. 
Where tight microbial control is required, such as areas used for the 
preparation of solutions to be filtered before aseptic filling, GMP requirements 
may also include that: 
filtration and air circulation should achieve EU GMP Guide Grade C or 
equivalent with pressure positive (typically 15 Pa) to adjacent lower grade 
areas; 
strategically located local environmental protection, such as positive pressure 
Grade A LAF units, should be in place for exposed operations. 
For areas where specific microbial control is to be exercised continually, 
such as for aseptic preparation and filling operations, additional GMP 
requirements will need to be applied such as: 
all operations should be performed aseptically with filtration; 
air circulation should achieve Grade B (EU GMP Guide, Annex 1) at positive 
pressure to lower grade areas; 
any process or equipment drains should be sealed and fitted with a 
sterilizable trap; 
strategic Grade A protection should be provided at all points of product 
exposure. 
3.2.2 Services and utilities 
Services and utilities that come into direct product contact (or form part of the 
product) are of particular concern. Some typical criteria for commonly used 
critical utilities include: 
High purity water systems such as WFI systems: 
assessment of the proposed water quality against the level of quality required 
by the product (in terms of chemical quality, microbiological, pyrogenic, and 
physical particulate contamination); 
materials of construction (including piping, gaskets, valve diaphragms); 
internal surface finishes (Ra ratings, use of electropolishing, passivation); 
water pre-treatment and control (adequacy); 
system sizing (minimum and maximum demand);
key design considerations such as minimum flow rates, minimum deadlegs 
with no cavities, vents and how they are sealed/filtered, drainage air gaps and 
backflow prevention devices; 
use of security devices, such as 0.2 micron sterilizing grade filters, UV 
sterilizers, ozone injection; 
instrumentation and control of critical process parameters (for example, 
temperature, velocity, flow, conductivity control limits and alarms, use of 
dump valves and recirculation of bad quality water, monitoring, recording 
and controlling systems); 
storage (such as storage temperature, maintenance of circulation and wetting 
of all internal surfaces, vent filter integrity and sterilization); 
methods and adequacy of cleaning and sanitization; 
adverse operating conditions (start-up/shutdown, power failure, redundancy, 
etc.); 
proposed method of construction (including procedures, control and inspection 
of material stock, fabrication, welding, field installation, passivation, 
preservation). 
Clean steam systems: 
similar considerations to those described for high purity water systems can 
generally be applied to clean steam systems. 
Gases (such as compressed air, nitrogen, hydrogen and oxygen): 
assessment of the proposed gas quality against the level of quality required 
by the product (in terms of chemical quality, microbiological, pyrogenic, and 
physical particulate contamination); 
materials of construction (including piping, gaskets, valve diaphragms); 
internal surface finishes (Ra ratings, use of electropolishing, passivation); 
system sizing (minimum and maximum demand); 
use of security devices, such as 0.2 micron sterilizing grade; 
instrumentation and control of critical process parameters (for example, 
temperature, pressure and dew point, monitoring, recording and controlling 
systems); 
methods and adequacy of cleaning and sanitization; 
adverse operating conditions (start-up/shutdown, power failure, redundancy, 
etc.); 
proposed method of construction (including procedures, control and inspection 
of material stock, fabrication, welding, field installation, passivation, 
preservation).
Typical GMP criteria for Water for Injection (WFI): 
quality to conform to compendia requirements (such as USP and/or Ph.Eur 
Monographs); 
production to be by distillation (also reverse osmosis allowed in 
some regions) from purified water and to conform to USP and/or Ph.Eur 
Monographs; 
WFI to be sterile and pyrogen free with an action limit set to less than 10 
CFU/100 ml (Colony Forming Units) with a sample size of between 100 
and 300 ml and an endotoxin level of less than 0.25 EU/ml (endotoxin 
units). 
Design of WFI systems: 
Firstly it is important to ensure that there is adequate pre-treatment and 
control of feed water, using methods such as deionization, ultraflltration 
and reverse osmosis. Pre-treatment by deionization alone may prove to be 
unsatisfactory. 
Key features of the WFI system itself include: 
still to be of multi-effect type, heat exchangers of double tube sheet design 
and holding tank employing tube type external jacket; 
WFI system to be fitted with a hydrophobic sterilizing grade vent filter to 
protect system from ingress of non-sterile air; 
vent filter to be jacketed to prevent condensate blocking the filter and to be 
steam sterilizable and integrity tested in place; 
provision for continuous ring main circulation at temperatures over 700C at 
velocities sufficient to achieve a Reynolds number of >25000; 
provision for periodic sterilization of the system; 
provision for sampling at all loop take-offs (the start and end of the loop) 
with take-offs design to prevent re-contamination of the system by airdrying, 
steam locking or trace heating; 
WFI to be stored in a nitrogen atmosphere where appropriate to minimize the 
absorption of oxygen; 
product contact materials be supplied with material certification and PMI 
(Positive Material Identification) and stainless steel contact surfaces to be 
<0.5 um Ra and passivated; 
pipework joints and couplings to be minimized with pipework being orbitally 
welded where possible. Detailed weld records to be supplied with weld logs 
and NDT reports on specified minimum proportions of all welds. Couplings 
and equipment to be crevice free — clamp fittings IDF couplings or similar 
are preferred. Deadlegs in vessels and pipework be minimized, by for
example use of zero deadleg diaphragm valves. System to be designed to 
allow for periodic complete flushing or draining such that all lines will slope 
to low drain points at a slope of greater than 1 in 100. 
3.2.3 Personnel flows 
This includes the influence personnel have on the quality of the product that 
might be caused by their contact with the product. Typical criteria include: 
clothing requirements (suitability of proposed plant clothing against the 
types of operations being performed within that area); 
changing regimes (stages of changing); 
changing facilities (adequacy of changing and washing facilities, doors, step 
over barriers, provision of adequate space for clothing, use of vision panels 
and their position relative to/from production areas); 
security and access control including potential short cuts and back doors; 
types of movements within the area (including passing through, local 
operations, supervisory support); 
occupancy levels; 
shift patterns (what supervisory and maintenance support is available); 
potential points of cross-contamination between personnel (such as transfer 
hatches, changing rooms — gowning/ungowning, finger streak stations); 
activity levels (i.e. sedentary or active and how this compares to the required 
room environment, occupancy level and clothing regime). 
3.2.4 Material flows 
This includes all the movement of materials. Typical criteria include: 
general flow of materials through the area (for example, linear flow through 
with no cross-over of production streams); 
methods of handling and prevention of cross-contamination; 
frequency of movements and available space; 
possible points of cross-contamination between materials (for example, 
temporary storage points, processed and non-process materials, bulk containers); 
identification and segregation of materials; 
storage conditions (refrigerated, toxic, hazardous, filtered). 
3.2.5 Equipment flows 
It is important to consider that not all equipment may be fixed in one position; it 
may either be moved routinely as part of the production process, or at least be
capable of relocation for plant maintenance or reconfiguration. Typical criteria 
include: 
methods of handling and prevention of cross-contamination; 
frequency of movements and available space; 
physical size and weight of equipment against room construction (heavy 
equipment may damage welded sheet vinyl floors or fracture gyprock walls 
— trowelled on epoxy cement or blockwork may be more appropriate); 
possible points of cross-contamination between equipment (such as temporary 
storage points, washing machines and bays); 
identification and segregation of mobile equipment; 
storage conditions (refrigerated, toxic, hazardous, filtered); 
provision of non-routine access, such as removable wall or ceiling panels. 
3.2.6 Equipment design 
The examination of the GMP issues within a machine or system is a 'micro' 
version of those for a facility, and includes many of the same questions such as 
surfaces, flow of materials and personnel issues. The amount of detail will vary 
with the complexity of the equipment and its effect or potential effect on 
product quality. Typical criteria include: 
pedigree of the machine (established for pharmaceutical use, 'off the shelf 
or specially developed prototype); 
pedigree of the manufacturer (specialist supplier to the pharmaceutical 
industry who manufactures more than 50 identical units per year or first 
development machine by a new manufacturer); 
materials of construction and surface finishes of primary and secondary 
contact parts (i.e. primary — direct product contact, secondary — contact 
with local environment); 
equipment sizing (minimum and maximum demand); 
key design considerations (minimum deadlegs with no cavities, all critical 
surfaces accessible and cleanable, drainage air gaps and backflow prevention 
devices); 
instrumentation and control of critical process parameters (temperature, 
pressure, speed control limits and alarms, monitoring, recording and controlling 
systems); 
methods and adequacy of cleaning and sanitization; 
adverse operating conditions (start-up/shutdown, power failure, redundancy); 
proposed method of construction (including procedures, control and inspection 
of material stock, fabrication, field installation);
maintenance (access for maintenance during and outside production, use of 
maintenance free items, requirements for special tools/no tools). 
For equipment and pipework that does not come into contact with the 
product or product components, there are no specific GMP requirements. 
For process pipework and equipment there is no need for sophisticated 
Clean in Place (CIP) or Steam in Place (SIP) but plant washing and flushing 
with water or chemicals may be used. Typical requirements include the 
following: 
dismantling and inspection should be easy and involve minimal use of tools; 
all pipework should slope towards the drain points; 
product contact materials should be supplied with material certification and 
stainless steel surface finishes in contact with the product should be < 1.0 urn 
Ra and passivated. Pipework couplings and equipment should be crevice 
free. Clamp fittings, IDF couplings or similar are preferred and deadlegs in 
vessels and pipework should be minimized. 
For areas where CIP and SIP effectiveness is critical, GMP requirements 
may include, in addition to the above, that joints and couplings are minimized 
with pipework being orbitally welded where possible and that stainless steel 
product contact surfaces are <0.5 urn Ra electropolished. 
For certain types of equipment, specific GMP requirements have been 
issued — one example of this is for sterilization equipment. Typical criteria 
for porous load moist heat sterilizers include: 
the complete chamber space should achieve a uniform temperature distribution 
of less than ± 1°C at the sterilization temperature for the complete 
sterilization period, and the equilibrium time to achieve this distribution 
should be less than 30 seconds; 
the chamber should be resistant to corrosion and the leak rate of the chamber 
should be less than 1.3 mbar per minute; 
monitoring instrumentation and recording charts should be independent of 
control instrumentation and utilize an independent time/temperature and 
pressure chart or equivalent of a suitably large scale to record the sterilization 
process; 
an air detector should be fitted such that a difference in temperature of greater 
than 2° C between the centre of a standard test pack and chamber temperature 
at commencement of equilibrium time is detected; 
drains should be trapped and vented and not connected to other drains which 
could cause a backpressure or obstruction to flow — an air break is 
necessary;
3.2.7 Computerized systems 
The amount of detail will vary with the complexity of the computerized system 
and its effect or potential effect on product quality. In particular the pedigree of 
the manufacturer, type of hardware and type or category of software to be used 
need to be carefully considered. The systems manufacturer is generally 
responsible for providing the validation documentation and ensuring that the 
system complies with GMR Typical criteria include: 
General: 
up to date specifications, including principles, objectives, security measures 
and scope of the system and the main features of the way the system will be 
used and how it interacts with other systems and procedures; 
the development of software in accordance with a system of quality 
assurance; 
system testing including a demonstration that it is capable of achieving the 
intended results; 
procedures for operation and maintenance, calibration, system failure (for 
example, disaster recovery, restarting), recording, authorizing and carrying 
out changes, analysis of errors, performance monitoring; 
pedigree of the machine and manufacturer; 
type of hardware (for example, standard 'off the shelf components from 
reputable suppliers operating a recognized quality system, installed in a 
standard system such as a PC or fully bespoke hardware developed 
specifically for the system); 
type/category of software (operating system, can be configured, bespoke 
software); 
adequacy of system capacity (in terms of memory, I/O, etc.). 
steam used for the sterilization process should have a dryness fraction of 
not less than 0.95 and the superheat measured on expansion of the 
steam to atmospheric pressure should not exceed 250C with the fraction 
of non-condensable gases not exceeding 3.5% by volume. The steam 
generator should be designed to prevent water droplets being carried over 
into the steam and should operate so as to prevent priming. The steam 
delivery system should be fitted with a water separator and traps to 
virtually eliminate condensate build up, and be resistant to corrosion 
with minimum deadlegs to reduce the risk of water collection and biofilm 
formation.
Control/access/security: 
built in checks of the correct entry and processing of data; 
suitable methods of determining unauthorized entry of data such as the use of 
keys, pass cards, passwords and restricted access to computer terminals; 
control of data and amendments to data, including passwords. Records of 
attempts to access by unauthorized persons; 
additional checks of manually entered critical data (such as weight and batch 
number of an ingredient during dispensing); 
entering of data only by persons authorized to do so; 
data storage by physical and electronic means. The accessibility, durability 
and accuracy of stored data. Security of stored data; 
data archiving, remote storage of data; 
recording the identity of operators entering or confirming critical data. 
Amendments to critical data by nominated persons. Recording of such 
changes; 
audit trail for system; 
change control system; 
obtaining clear printed copies of electronically stored data; 
alternative arrangements in the event of system breakdown, including the 
time required to recover critical data; 
positioning of the equipment in suitable conditions where extraneous factors 
cannot interfere with the system; 
form of agreement with suppliers of computerized systems including 
statement of responsibilities, access to information and support; 
release of batches, including records of person releasing batches. 
Personnel/training: 
personnel training in management and use; 
expertise available and used in the design, validation, installation and 
operation of computerized systems. 
Replacement of a manual system: 
replacement of manual systems should result in no decrease in product 
quality or quality assurance; 
during the process of replacement of the manual systems, the two systems 
should be able to operate in parallel; 
reducing the involvement of operators could increase the risk of losing 
aspects of the previous system.
3.2.8 Maintenance and servicing 
This applies to all the facility and everything within it. It is important 
to consider that not all equipment may be fixed in one position, it may 
either be moved routinely as part of the production process or at least be 
capable of relocation for plant maintenance or reconfiguration. Typical criteria 
include: 
methods of handling and prevention of cross-contamination; 
frequency of movements and available space; 
physical size and weight of equipment against room construction (heavy 
equipment may damage welded sheet vinyl floors or fracture gyprock walls; 
trowelled on epoxy cement or blockwork may be more appropriate). 
3.2.9 Procedures and documentation 
In order to support the facility, adequate procedure and documentation are 
required. During the design stage many of the documents required for normal 
operation of the facility may not yet be available. At this stage, it is probably too 
early to consider exactly what documentation will be required, but it is possible 
to begin to consider how documentation will be accommodated and organized. 
Typical criteria include: 
adequate workspace, storage capacity and personnel to control stored 
documentation; 
security of documentation (including access control, fire protection, additional 
remote storage capacity); 
adequate, rapid access to stored data, including suitable provisions for the 
local retrieval of data stored electronically. 
3.3 GMP reviews of design 
To ensure that the project remains in compliance with cGMP as it progresses 
through its life-cycle, periodic GMP design reviews must be undertaken. 
3.3.1 Organizing the GMP design review team 
Reviewing a design for compliance to cGMP requirements can often be a 
daunting prospect. It requires a range of knowledge that no single person is 
likely to possess. For this reason it is often more effective if the review is 
performed by a small team that has an understanding of the basic requirements 
and works methodically. The team should consist of persons selected for both 
their depth of knowledge in a particular area and for general knowledge of
cGMP principles applicable to the project. A good mix for a suitable team 
would be: 
cGMP compliance/validation specialist (knowledge of regulatory, QA, 
validation, etc.); 
architect (knowledge of finishes, layout, personnel/materials flows, etc.); 
process engineer (knowledge of process, equipment, utilities, etc.). 
Depending upon the nature of the facility the architect or process engineer 
could be substituted for more suitable disciplines. For example, the design 
review of an automated high bay warehouse may be better performed using a 
materials handling specialist and an automation specialist. The team would 
normally be lead by the cGMP compliance/validation specialist who would 
organize the team, co-ordinate the review and prepare the report(s). It is 
recommended that the team be kept as small as practicable, since it will be 
able to operate more efficiently and flexibly and be easier to co-ordinate. If 
issues arise that are beyond the combined knowledge of the team then they can 
be referred for further investigation by specialists in the particular subject. 
3.3.2 Information required to perform the review 
Two basic types of information are required to perform an effective review: 
specification of the pharmaceutical product and manufacturing process; 
specification of the equipment and facility. 
Note that some facilities are used for a variety of products that may utilize 
different processes. In this case a separate review of each process may be 
performed. However, often it is possible to base a review on a 'typical' product 
that runs through the entire process. 
As part of the cGMP review all information sources used must be 
documented. Regulatory authorities always demand to see original information. 
It is, therefore, essential that a good record keeping system be established — 
for example, original design calculations must be retained. All engineering 
drawings must be authorized and signed off. 
Specification of the pharmaceutical product and manufacturing process 
General details of the process are required rather than exact details of, say, a 
particular chemical reaction involved. Sources of information may include: 
regulatory documents such as: 
o New Drug Application (NDA), Product Licence Application (PLA), 
Investigational New Drug Application (IND);
o manufacturer's licences such as Product Licence, Wholesale Dealers 
Licence; 
o Drug Master File (DMF); 
technology transfer documents; 
batch manufacturing documentation prepared for similar facilities; 
process description; 
process flow diagrams (PFD). 
The type of information required will typically include: 
description of processing operations including: 
o manual operations such as loading, sampling testing, adjustments; 
o automatic operations such as process unit operations, cleaning cycles and 
materials handling; 
quantities and throughputs; 
components and processing chemicals; 
critical parameters such as temperature, pressure, time and volume; 
batch size and frequency; 
regulatory requirements in original product licence/regulations; 
technical requirements identified during laboratory/pilot scale production. 
Specification of the equipment and facility 
Clearly the review will utilize the GMP design philosophy as a key document, 
but this should be compared with what has actually been specified. General 
details of the equipment and facilities are required. Sources of information may 
include: 
architects/facility engineers; 
process engineers; 
engineers from the various technologies as appropriate — for example, 
mechanical, electrical, civil, control, instruments; 
R&D; 
QC/QA. 
The type of information required will typically include: 
process description, materials and personnel flow diagrams; 
general arrangement drawings, axiometric drawings and room layouts; 
process and instrumentation diagrams (P&IDs); 
HVAC basic layouts, specifications and area classification drawings; 
main equipment items list with specifications; 
utilities list with specifications;
3.3.3 Divide up the facility into manageable sized areas 
The best way to divide up the facility for the review largely depends on the type 
of facility and nature of the process. The following approach is suggested for 
guidance. 
Bulk pharmaceutical chemical manufacturing 
Typically for BPC manufacturing the process is contained within closed vessels 
and pipework arranged as an integrated/interconnected process. In this case it 
is probably easiest to break the cGMP review up into a series of reviews of each 
main P&ID. Each P&ID is then considered by the review team along with any 
associated equipment and utility specifications, control system descriptions 
etc., as a package. 
Secondary manufacturing 
Typically for secondary manufacturing, the process is carried out in a series of 
discrete stages in separate areas such as: 
The best method here may be to perform the review on each area of the 
facility. The review will centre on the room layout drawings along with 
associated environmental classification drawings, equipment and utility specifications 
and control system descriptions, as a package. It may also be possible 
to identify specific areas that have no cGMP implications — these can be 
considered to be 'outside the GMP area' and need not form part of the review 
although any decisions made to include or exclude particular areas should be 
documented. 
In some cases, a combination of both the above methods may be the 
most appropriate. The key point is to break the task down into logical, 
manageable-sized portions, which can then be reviewed. 
Goods in. 
Warehousing. 
Amenities. 
Changing rooms. 
Equipment preparation. 
Dispensing. 
QC testing laboratory. 
Weighing. 
Mixing/blending. 
Filling. 
Sterilizing. 
Labelling. 
Packing. 
Administration area. 
Services and utilities. 
Goods out. 
user requirement specifications; 
control system functional design specifications.
4.1 Introduction 
Validation first started in the 1970s on sterilization processes, when it became 
clear that end product testing alone could not show that every container within 
every batch of product was sterile and the time and cost associated with 
testing each individual container was too great, or the testing was too 
destructive to the product. Validation offered a way of providing evidence 
that the process was capable of consistently producing a product with defined 
specifications. 
This type of work spread gradually through from sterile and aseptic 
processes to non-aseptic processes (tablet manufacture, for example) by the 
mid 1980s. By the late 1980s, the concept of validation was reasonably well 
established. Regulatory authorities and the pharmaceutical industry have 
co-operated to define validation requirements and agree upon the definition. 
The principle is the same for whichever process is being investigated — that is, 
to provide documented proof of GMP compliance. Validation and GMP go 
hand in hand. 
4.1.1 Definition 
Even before the current definitions of validation, industry was operating to 
the concept in the first edition in 1971 of the British Guide to Good 
Pharmaceutical Manufacturing Practice (the 'Orange Guide'), which 
suggested that procedures should undergo a regular critical approach to 
ensure that they are, and remain capable of, achieving the results they are 
intended to achieve. 
Although the US Federal Register does not contain an official definition, US 
CFR Part 211 section 211.100 states that: 
'There should be written procedures for the production and process control 
designed to assure that the drug product has the strength, quality and purity 
they purport or are represented to possess! 
4 
V a l i d a t i o n 
JOHN WELBOURN
The FDA has issued a 'Guideline on General Principles of Process 
Validation' which defines process validation as: 
'Establishing documented evidence which provides a high degree of assurance 
that a specific process will consistently produce a product meeting its 
predetermined specifications and quality attributes.' 
The EU 'Rules Governing Medicinal Products in the European Community' 
VoI IV define validation as: 
'Action of proving, in accordance with the principles of Good Manufacturing 
Practice, that any procedure, process, equipment, material, activity or systems 
actually leads to the expected results! 
The EU Rules also define the term 'Qualification', which arises many times 
within validation work, as: 
'Action of proving that the equipment works correctly and actually leads to 
expected results. The word validation is sometimes widened to incorporate the 
concept of qualification! 
Validation for the engineer is the act of proving with the necessary formal 
documentation that something works. It is advisable to create the documentation 
throughout the design process since it is often expensive and timeconsuming 
to produce retrospective documents. 
4.1.2 The need for validation 
There are three reasons why the pharmaceutical industry is concerned about 
validation: 
government regulation; 
assurance of quality; 
cost reduction. 
Government regulation 
The requirements for validation are now explicitly stated in both the US and 
European regulations (US Code of Federal Regulations US CFR Part 211, 
subpart L, 211.220 and 211.222 and within the EU 'Rules Governing Medicinal 
Products in The European Community' VoI IV, Part 5.21, 5.22, 5.23, 5.24). 
In CFR 211.220 it says: 
'The manufacturer shall validate all drug product manufacturing 
processes ... '
and: 
' . . . validation protocols that identify the product and product specifications 
and specify the procedure and acceptance criteria for the tests to be conducted 
and the data to be collected during process validation shall be developed and 
approved 
and: 
' . . . the manufacturer shall design or select equipment and processes to ensure 
that product specifications are consistently achieved. The manufacturer's 
determination of equipment suitability shall include testing to verify that the 
equipment is operating satisfactorily 
Similar requirements are stated in the EU Rules. 
Assurance of quality 
Without process validation, confidence in the quality of products manufactured 
is difficult to prove. The concepts of GMP and validation are essential to quality 
assurance. Frequently, the validation of a process will lead to quality improvement, 
as well as better consistency. It may also reduce the dependence upon 
intensive in-process and finished product testing. It should be noted that in 
almost all cases end-product testing plays a major role in assuring that quality 
assurance goals are met, i.e. validation and end-product testing are not mutually 
exclusive. 
Cost reduction 
Experience and common sense indicate that a validated process is a more 
efficient process that produces less reworks, rejects, wastage, etc. Process 
validation is fundamentally good business practice. 
In summary, validation should be applied to all aspects of the process, 
including the equipment, computer systems, facilities, utilities/services and 
in-process testing (analytical methods). From the above discussion, the following 
key points have developed: 
documented evidence must be written down (if it's not documented it's not 
done); 
formal documentation — all design documents should be signed off. Signatures, 
page numbering, control copies, storage/retrieval, etc., should be 
installed; 
acceptance criteria — decide what is acceptable before testing; 
repeatable — one-off results are not acceptable;
4.2 Preliminary activities 
Prior to embarking on a validation project, it is necessary to establish an 
organizational framework in which validation resides. This must start with the 
commitment and sponsorship of the senior management within the company, 
for without this commitment to validation any validation project is likely to fail. 
4.2.1 Establishing policies and procedures 
One of the first steps is to establish the policies and procedures that will govern 
the validation project — for example, the development of policies to define 
general concepts involved such as: 
how validation 'fits' within the overall QA structure and its relationship with 
cGMP; 
commitment to cGMP and its reinforcement through validation (i.e. the 
pharmaceutical company's commitment); 
definition of key terms such as critical process step, critical equipment and 
instrumentation, the various qualification activities including DQ, IQ, OQ, 
PQ (more about this later); 
how validation is structured and applied with respect to plant, processes, 
computer systems, analytical methods, etc. (how is it organized, what steps 
are performed in each case and how does it all fit together). 
More specific procedures will need to be generated later for: 
validation documentation preparation (including house style, standard document 
sections, document numbering); 
validation documentation review and approval process; 
validation document change control system; 
validation master plans and final validation reports (preparation, content and 
structure); 
pre-qualification activities; 
cGMP reviews of design; 
vendor assessment and auditing (especially computer systems); 
equipment/computer system protocols and reports (i.e. DQ, IQ, OQ, PQ) 
preparation, content and structure; 
instrumentation and calibration; 
execution of field work; 
validation and qualification — processes are validated whereas the equipment 
used within the process is qualified.
set-up and operation of validation test equipment; 
cleaning validation; 
process optimization and experimental work; 
process validation protocols and reports; 
analytical methods validation; 
documentation filing and management systems. 
Note that it is particularly important at an early stage in the project to agree 
aspects such as document format, structure, content and numbering. This 
agreement needs to be recorded in the project quality plan. 
At this early stage it is a good idea to establish the key validation team 
members and prepare an overall organizational chart. 
Some of the first activities for the validation team to address will include: 
process evaluation to determine validation requirements; 
identification of systems and system boundaries; 
preparation of user requirement specifications; 
development of the validation master plan. 
4.2.2 Process evaluation to determine validation requirements 
Process evaluation involves a review of the process to identify the process steps 
and process variables, to determine how they are controlled/monitored and to 
identify what processing, equipment, utilities, instrumentation and control 
systems are associated with these steps. This should identify which systems 
need to be qualified and which parameters and instrumentation are important to 
the process and will need to be evaluated in the validation study or will become 
'critical instruments.' As part of the development work done on the process, 
much of this should already have been defined, however, the documents where 
this is recorded need to be collated and reviewed. 
The specification and procedures required for the process such as equipment 
operation and maintenance, calibration, set-up, cleaning and in-process testing 
should be identified, since these will need to be prepared for the new facility. 
The various components used to manufacture the product should be 
reviewed to establish that all items have been specified and are under control. 
This may then point to requirements for analytical methods, validation or 
supplier audits, for example. 
Based on an evaluation of the process a decision can be made as to what does 
and does not require validation. To perform such an evaluation requires a 
thorough understanding of the process and may include process components, 
process chemistry, plant (equipment, automation systems, etc.), specifications 
and procedures, in process controls and analytical testing methods.
User requirement specifications (URS) 
These should be prepared by the user to formally document the requirements 
for each system to be qualified in terms of the final process requirements. A 
URS should typically include specific, but non-detailed information relating to, 
for example, quantity, quality, compatibility, performance, environment and 
finishes, in terms of: 
materials of construction; 
cleanability requirements; 
maintenance requirements; 
operator interface requirements; 
performance criteria; 
critical parameters; 
essential design criteria; 
requirements of computerized/automation system; 
training and documentation requirements. 
It should make reference to relevant in-house standards and regulatory 
documents. It is essential that input to the URS includes persons with 'hands 
on' knowledge of the system and persons with a wider knowledge of the overall 
project. 
4.2.3 Identification of systems and system boundaries 
In parallel with process evaluation, systems and system boundaries need to be 
defined. The objective is to break the facility down into logical, manageablesized 
packages of qualification work, and concentrate the validation effort in 
the most important areas to allow structured qualification. 
A system may be an area of the facility (group of rooms), a group of 
functionally related process items, a utility or part of a utility, a HVAC, a 
computerized/automation system or any combination of these. 
Determination of system boundaries involves the evaluation of the proposed 
facility design to establish the boundaries and break points for each package of 
qualification work. It is important that at the earliest stage practicable any 'grey' 
areas are removed, such as overlaps between areas of responsibility, missing 
areas, break points, IT systems interfaces. 
Systems may then be categorized as 'Primary' or 'Secondary', (it may be 
appropriate to develop several more intermediate categories, such as in the case 
of IT systems). For example, primary systems could be defined as 
large, complex, purpose built or configured, generally fixed in place units. 
Examples include an aseptic filling suite, low temperature hot water system, 
water for injection system, electrical power distribution system, a piece of
automated manufacturing equipment or a plant supervisory control and data 
acquisition system (SCADA). 
Secondary systems could be defined as smaller, simple, 'off the shelf, 
generally portable items with no or minimal unique features or configuration, 
such as a bench top balance, filter integrity tester, a pallet-bailing machine and a 
10-litre standard holding tank. Typically these systems may be bought direct 
from a supplier's catalogue. 
Systems may be further categorized as 'critical' or 'non-critical.' Typically 
the following criteria are used to evaluate if a system is critical: 
stage of the process — is it used before, during or after a critical process step; 
effect on product quality; 
contact with product or product components; 
monitoring or controlling elements related to product quality. 
Examples of primary critical systems are an aseptic filling suite, a water for 
injection system, a piece of automated manufacturing equipment, or a plant 
supervisory control and data acquisition system (SCADA). 
Examples of primary non-critical systems are a low temperature hot water 
distribution system or an electrical power distribution system. 
Examples of secondary critical systems are a bench top balance, filter 
integrity tester, and a 10-litre standard holding tank. An example of a secondary 
non-critical system is a pallet-bailing machine. 
All critical systems should be validated. For primary critical systems this 
may involve the development of detailed plans, protocols, reports, certificates; 
for secondary critical systems, however, the use of simple, standard, checksheet 
type documents may be more appropriate. 
Non-critical systems do not require qualification — standard, well-structured 
project documentation is adequate. 
4.3 Validation master planning 
The initial activities described above can be formalized and consolidated into a 
validation master plan (VMP). This is a formal, approved document that 
describes in clear and concise wording the general philosophy, expectations, 
intentions and methods to be adopted for the validation study. Everyone 
involved in a project will have their own interpretation as to what validation 
is and what should be done. The VMP is an agreed document acting as a road 
map or guide for all team members to follow. 
Once complete, it becomes a useful tool to show regulatory bodies that 
compliance with regulations is being sought and that there is a plan describing
in detail the steps and programmes to be implemented to assure a validated and 
compliant facility. 
To prevent the VMP becoming too unwieldy, it is common practice to 
develop separate validation plans for various parts of the overall project such 
as process equipment, utilities, computer systems, process and analytical 
methods. On large projects it may be necessary to have several levels of plans. 
In terms of when to begin to develop the VMP, this will vary from project to 
project but it should normally be in place by the early part of detailed design. 
The VMP will then be a living document, updated regularly and amended 
during the course of the project. At the end of the project the VMP should 
define how the validation was actually performed. 
The VMP, as with all formal validation documents, should be prepared, 
reviewed, approved and controlled under pre-defined company policies and 
procedures with final approval by QA. It must have a document number and a 
document revision history and page numbering must pass the 'drop test' (i.e. it 
is possible to reassemble the document from the page numbering and know that 
all sheets have been accounted for). The number of copies should be controlled. 
4.3.1 Contents of the VMP 
This will differ slightly from project to project and company to company, but 
the following items should usually be included: 
(1) approval page; 
(2) introduction; 
(3) the aim; 
(4) descriptions of: 
facility; 
services/utilities; 
equipment; 
products; 
computer systems; 
(5) validation approach: 
overall; 
detail (matrix of validation documents); 
(6) other documentation. 
Approval Page 
The approval page is the title page to the entire document and should contain the 
name of the company, the title and a space for approval signatures. Usually the 
author and three approvers sign the approval page. The approvals should come
from the people affected by the validation project, such as production, QA and 
engineering functions related to the facility. A development signature may be 
necessary if the project relates to the manufacture of a new product. 
As a general rule it is not a good idea to have too many approvers as there is a 
danger that scrutiny and understanding starts to suffer because each approver 
will be expecting others to have checked certain items. It is important that the 
approvers know what they are signing for. As with all validation documentation, 
the continuity of the dates from the signatures is important. The author 
should sign first, followed by the others, with QA input last. 
Introduction 
The introduction should explain why the project is being undertaken, where it is 
going to be located and the broad timetable. 
Aim 
The aim should explain that this is to be a formal validation study on a specific 
project and show that the approach conforms to cGMP. The aim may point to 
the various company policies and procedures under which the VMP is to be 
prepared and controlled. 
Description 
This section should describe the main features of the project in concise terms, 
picking out particularly critical features or acceptance criteria. 
Facility 
This section of the VMP should outline the facility's intended use, briefly 
discuss how it is to be built and state whether it is an entirely new facility or an 
expansion of an existing one. 
For example, it could describe the size of the facility, the number of floors 
the facility occupies, the processing areas and, if necessary, the segregation for 
contamination; how many HVAC systems there are, and what the classifications 
are; any special gowning procedures or other procedures to be followed. Some 
simple outline drawings will generally be included with the description — 
typical drawings to insert are: 
facility location in relation to site; 
cross section of the facility (if relevant); 
floor plan (one for each floor) with equipment locations; 
HVAC zone identifications; 
personnel flow;
component flow; 
raw material flow; 
product flow. 
Services/ Utilities 
This section may consist of a list of plant utilities and services, such as cold 
potable water, purified water, water for injection, plant air, instrument air, 
nitrogen, chilled water. 
In addition to this listing, there should be a brief description with simple line 
diagrams for each system, which should include any key performance criteria 
such as minimum flow rate or pressure, and quality. However, detailed 
requirements of the systems can be written into individual protocols — this 
helps keep the VMP to a sensible size and makes it easier to control. 
Equipment 
As with the previous section, this could start with a list of all the major items of 
equipment that are going to be installed into the facility, for example, porous 
load steam sterilizer, bench top balance, or powder mixer. It is a good idea to 
divide up the list by facility area or stage in the process. The list that is generated 
should include a unique plant item number for each major piece of equipment 
for reference purposes. For the most important items it is a good idea to include 
a brief description with a simple line diagram with any key performance. 
Products 
In this section, information should be provided about the products that are going 
to be manufactured in the facility in question. For each product this may include: 
batch size; 
ingredients: 
o quantities per unit dose; 
o quantities per batch; 
the steps by which the product is manufactured: 
o process flow diagrams; 
o summary of manufacturing method. 
Computer systems 
This section lists all the computer systems associated with the facility, process 
equipment and utilities as well as IT systems to operate the plant such as LIMS, 
SCADA and MRP systems, and provides descriptions of each system picking 
out any important performance.
Validation approach - overall 
This section of the VMP is used to describe how the validation work is to be 
performed and documented (see Figure 4.1 on page 49). 
It gives the design engineer's viewpoint of the Validation Master Plan. Note 
that it starts with the User Requirement Specifications (URSs), which is usually 
prepared by the user in discussion with the design engineer. This document 
forms the basis for the design. 
This flow chart forms an excellent checklist for the validation process and 
underlines the importance of preparing validation documentation right from the 
issue of the URS to the performance qualification of the plant built to the final 
design. The main aspects of this flow chart, which provide the design engineer 
with a good background to the validation process, are detailed. 
Process evaluation and validation systems 
This section should explain how the facility has been divided up into separate 
systems and how the process has been evaluated to determine what aspects are 
critical to product quality. It should introduce concepts such as 'critical parameters' 
and 'critical instrumentation' and relate these to the validation requirements, 
in line with the method described in Section 4.2.2 and 4.2.3. 
Validation team 
This section defines the role and responsibilities of key personnel involved. It is 
often a good idea to use job titles rather than names since individual personnel 
may change, and to include a project organization chart. In particular, it is 
important to explain the role of QA in the approval processes. 
Validation methodology 
The validation methodology should describe what types of documents will be 
generated within the project (protocols and reports — Design Qualification 
(DQ), Installation Qualification (IQ), Operational Qualification (OQ), Performance 
Qualification (PQ), and Process Validation (PV)) and how they will be 
prepared, reviewed, approved and controlled. This section should draw on 
company policies and procedures, which should define each part in more detail. 
In addition, as appropriate, the methodology should discuss cleaning validation, 
analytical methods validation and computer systems validation (there will 
be more about the various validation activities later in this chapter). 
The section should then describe the execution strategy for the protocols 
including, for example, how results are recorded and how any problems 
encountered are dealt with, and the role of equipment vendors in validation
Design 
Pre-qualification Design qualification 
Construction/delivery/installation/pre-commissioning 
Installation qualification 
Change control Change control 
Commissioning 
Operation qualification 
Hand-over 
Performance qualification 
Maintenance 
Change control Change control 
Validation Master Plan 
URSs 
System boundaries 
Vendor assessment Documentation requirements Documentation submissions 
Room data sheets 
Testing requirements 
Witnessing requirements 
Emphasise documentation requirements 
Test equipment requirements 
Responsibilities 
Emphasise timing of documentation submissions 
cGMP review of design 
Tender design specification 
Tender submissions 
Pre-contract meetings Procurement design specification 
Vendor selection 
Procurement 
Programme user resource requirements 
Submit instrument schedule 
Design qualification 
Programme documentation submissions 
Submit testing, pre-commissioning & commissioning method statements 
Submit calibration procedures 
Factory testing FATs 
Submit O & M manual skeleton 
Design changes 
Submit plant item number list 
cGMP reviews of design changes 
Submit O & M manual draft 
Equipment data sheets 
Calibrate instruments 
Instalation testing 
Test equipment calibration certificates 
Submit 'as-built' drawings list 
Identify 'critical' instruments 
Test equipment calibration certificates 
'As-built' drawings 
Cleaning 
Instalation qualification 
Engineering commissioning SATs Supplement O & M manual Train 
Audit room data sheets 
Draft SOPs 
Process commissioning 
Approve O & M manual 
Train 
Finalise SOPs 
Stress testing 
Reliability testing Range testing Repeatability testing 
Operation qualification 
Performance qualification 
Experiments Media fill trials Validation batches Packing instructions Batch sheets 
Production 
Figure 4.1 Validation flow chart (By kind permission of Validation in Partnership).
(i.e. utilize vendors as much as possible in the preparation and execution of 
validation work or do as much of the work as possible 'in-house'). 
This section can also be used to describe the organization and management 
of project documentation, including document flow and filing (for example, 
documentation filing structure, use of document management systems, IT). 
Validation schedule 
It is often useful (although not obligatory) to include a time schedule in the 
plan. It is probably best to keep this relatively simple, as schedules tend to 
change frequently during a project. The VMP is not intended as a document to 
convey this type of information. 
Validation approach - detail 
This section includes details of which types of documents are going to be 
produced for each system to be qualified and which processes are to be 
validated. This is often done by a validation matrix (see Table 4.1). 
Other documentation required 
This section should establish links to other types of documents that could 
be required at regulatory authority inspections. The type of documents which 
come under this heading include: 
batch production records; 
packing instructions; 
training; 
Table 4.1 Example of a validation matrix 
Item 
Utilities 
HVAC 
WFI 
Equipment 
Tablet Press 
Autoclave 
Product 
Tablet A 
Tablet A, cleaning 
Item no. 
ABC 123 
ABC456 
XYZ789 
XYZ123 
Document type 
DQ IQ OQ PQ PV
4.4 Development of qualification protocols and 
reports 
The VMP defines which systems are to be qualified and how the work is to be 
organized and controlled. The next step involves the preparation of qualification 
protocols and the generation of associated reports. 
4.4.1 Qualification protocols 
There are various different approaches to the format and content of qualification 
protocols — for example, protocols can be developed as stand-alone documents 
or can cross-reference other project engineering documentation. They 
can be designed so that results are recorded within the body of the protocol or 
that all the detail is left for recording in the reports. The former results in bulky 
protocols but brief reports, whereas the latter results in slim protocols and bulky 
reports. As with all validation work the protocols should be developed in 
accordance with company policies and procedures. There should be SOPs for 
protocol preparation, execution and reporting. 
Whatever approach is taken, there are certain key features that the protocol 
must have. These can be summarized as follows: 
formal documents: The protocol must go through a review and approval 
process with final approval by QA; this must be numbered, the number of 
copies must be controlled and have a document revision history, page 
numbering must pass the 'drop test' (see Section 4.3); 
defined scope: The protocol must define what area, equipment, etc., it 
addresses. This may be achieved by, for example, a system description, 
diagram or list of items; 
objective: The protocol should describe the purpose and how this relates to 
the overall validation activity and scope of the protocol; 
test structure: Each test must describe the objective and purpose of the test, 
the test procedure and the method of recording results. This should be in 
sufficient detail so that it could be understood by a third party, and repeated if 
necessary; 
SOPs; 
maintenance and calibration records; 
organizational charts and CVs; 
change control procedure; 
drawings.
acceptance criteria: Each test must have acceptance criteria as to what 
constitutes a pass or a fail. The acceptance criteria must be approved before 
execution of the protocol. 
A typical table of contents for a qualification protocol would consist of the 
following: 
title page; 
revision history; 
table of contents; 
introduction/background; 
purpose; 
scope; 
reference documents; 
system description; 
prerequisites; 
personnel performing the qualification; 
test equipment details; 
method; 
acceptance criteria; 
list of attachments. 
4.4.2 Qualification reports 
Once the protocol has been executed the results should be documented in a 
qualification report. At least one report should be written for each protocol. A 
typical table of contents for a qualification report would consist of the 
following: 
title page; 
revision history; 
table of contents; 
purpose; 
scope; 
executive summary; 
results; 
deficiencies and corrective actions; 
assumptions, exclusions and limitations; 
conclusions; 
appendices (depending on the protocol style adopted, one of the appendices 
may be the complete protocol).
The reports are also formal documents and should follow a similar 
preparation, review and approval process as protocols. 
Deficiencies 
As a general rule the report should be prepared by exclusion; that is, if a test 
was successful with no problems then only a brief mention is required in the 
report. The report should concentrate on the tests that failed and describe what 
remedial action was necessary and what retesting or further work was/is 
required. Examples of deficiencies include: 
conflicts with specifications — for example, the pump seal material was 
viton rubber not EPDM rubber as specified; 
information which is unavailable or incomplete; 
documentation discrepancies (incorrect reference number, issue number). 
Each deficiency should be given a unique identification number and a 
complete list of deficiencies encountered during the execution of the protocol 
should be included in the report. An audit trail should be established to show 
how the deficiency was resolved. 
4.5 Design qualification (DQ) 
The purpose of design qualification is to ensure that the final design: 
accords with all relevant specifications and design intentions; 
meets the requirements of the process, product and user; 
adequately specifies all necessary supporting documentation; 
complies with the requirements and principles of GMP. 
DQ is providing documented evidence that quality is built into the design. 
DQ is an auditing function to provide formal documentation that the facility has 
been designed to meet the requirements of the user and the GMP guidelines. 
DQ activities may include: 
GMP reviews of overall facility design; 
establishing the suitability of vendors and vendor deliverables through 
vendor assessment and auditing where appropriate; 
review and approval of equipment specifications and design documentation 
to ensure user requirement specifications (URS) have been adequately 
interpreted in the design process and that the design is in compliance 
with GMP
DQ comes down to carrying out a formal comparison of what is required 
against the proposed design. There should be DQ documentation for: 
the overall facility; 
each system within the facility. 
4.5.1 GMP reviews of overall facility design 
The GMP review of the overall facility/project design can be defined in the 
same terms as an audit, that is a formal documented review of the design of a 
plant (including facilities, equipment, utilities, computerized/automation 
systems and procedures) to give assurance that: 
it complies with the applicable statutes and associated published current 
Good Manufacturing Practices; 
it complies with applicable regulatory licence(s) and registrations submitted 
for the particular process(es) or product(s) to be manufactured, held or 
stored. 
Note that because of the confidential nature of the process, including licensing 
application details, the second point may be considered separately from the 
first. 
Typically, topics to be dealt with include: 
facility (construction, finishes of walls, floors and ceilings, corners and 
crevices, cleanability, durability, access control, pest control, etc.); 
environment (area classification, temperatures, humidity, air pressures, air 
change rates, viable and non-viable particle levels, etc.); 
personnel flows (access authorization, change regimes, gowning requirements, 
occupancy levels, cross-contamination, etc.); 
materials flows (solids, liquids, gases, toxicity, hazard risk, containers, 
transportation, storage, cross-contamination, etc.); 
equipment flows (size, weight, mobility, cleaning, method of handling, 
cross-contamination, etc.); 
general equipment design (proprietary, purpose built, materials of construction, 
finishes, cleaning, change parts, control systems, etc.); 
automation philosophy (monitoring or controlling, level, protection, environment, 
access control, archive storage and retrieval, electronic signatures, 
disaster recovery, etc.); 
maintenance/servicing (access, space, tools, diagnostic equipment, materials, 
power, lighting, authorization, training, etc.); 
documentation (SOP's, permits, history records, training, log books, etc.); 
waste management (liquids, solids, gases, packaging materials, cleaning, etc).
4.5.2 DQ of each system 
Vendor assessment 
Vendor assessment is the documented evaluation of the suitability and capability 
of the vendor to provide the 'system' to be procured to the quality required 
to fulfil user and cGMP requirements, including all necessary supporting 
documentation. Where appropriate this may include vendor auditing. 
Vendor assessment may stretch over several stages including assessment of 
the vendor's suitability to tender, assessment of preferred vendor and follow up 
vendor audit(s). Vendor assessment would generally involve, for each primary 
critical system including primary critical computer system, sending out selfassessment 
questionnaires and then, where appropriate, auditing vendors prior 
to placement of orders. Subsequent audits may be required throughout the 
design and construction/implementation process depending upon the nature of 
the system and the findings of the assessments and audit. 
DQ of system plant 
Design Qualification (DQ) of system plant (in other words, equipment, piping, 
valves and in-line fittings, field instrumentation, ductwork, insulation etc., or 
combinations of these) is the documented evidence that quality is built into the 
design of the system. It should include verification that the 'system' design 
incorporates the requirements of the user and of cGMP. Typically the DQ 
activities will include. 
cGMP review of design; 
specification review (URS/design specification(s) review); 
compilation of design documents; 
QA/QC review; 
facility acceptance testing (FAT). 
4.6 Installation qualification (IQ) 
Installation qualification is the documented evaluation of the equipment or 
system to establish that it has been installed in accordance with design 
specifications, cGMP requirements and manufacturers recommendations. 
Typically it will consist of various static checks, which may include for 
example: 
• system completion: Check that the system is mechanically complete and all 
critical punch list items have been cleared. Check that all work which should
have been completed and documented during the construction and 
installation of the system has been performed. This will involve checking 
through the various construction check sheets and certificates; 
security /utility connections: Check that the correct connection of utilities 
has been made and that, where appropriate, utilities have been IQed; 
documentation inventory: Check that all necessary supporting documentation 
such as specifications, operation and maintenance manuals are available 
and have been reviewed and approved; 
equipment inventory: Check that installed equipment name plate data 
complies with specification and record equipment serial numbers; 
materials qualification: Check that, where appropriate, contact part materials, 
surface finishes and lubricants are in accordance with the specification. 
This may involve a review of material certificates, chemical data sheets etc., 
or performing physical inspection and testing of materials; 
drawing validation: Perform a P&ID walk-down to check that all main 
components are as shown and in the sequence indicated. Where appropriate 
check pipework slopes (is it free draining?), measure pipework dead legs and 
drainage air gaps, check accessibility of manually operated devices; 
main equipment features: Check that each main component is in accordance 
with the construction drawing, check critical specifications such as 
filter grade, perform any static checks required prior to start up, such as 
checking lubricant levels, drive belt tension and torque settings; 
instrument calibration: Check that all critical instruments have been 
calibrated and that the calibration is traceable to national standards; 
spares and maintenance: Check that adequate spares provision has been 
made and maintenance requirements have been considered. This may 
involve, for example, getting a copy of the spares list reviewed and 
approved by the maintenance department and then checking that all 
spares have been supplied, and checking that the maintenance and calibration 
programme for the system is in place and that equipment log book(s) 
have been prepared. 
4.7 Operational qualification (OQ) 
Operational qualification is the documented evaluation of the system to show 
that it operates as intended throughout the anticipated operating ranges. 
Typically it will consist of various functional checks on the equipment, 
generally performed using inert materials such as water or compressed air 
and in the absence of real product.
Tests should be designed to show that the equipment would perform as 
intended and to specification. The tests should encompass upper and lower 
processing limits and circumstances, including those within normal operating 
conditions, which pose the greatest chance of process or product failure 
compared to ideal conditions. These conditions are widely known as 'worst 
case' or 'most appropriate challenge' conditions. 
For utilities it is important to show that the utility can be delivered within the 
requisite parameters (such as flow rate, temperature, quality, etc.) under conditions 
of maximum diversity (i.e. with the greatest or least preserved normal 
operating demand on the system from the most or least users of the system). 
It is difficult to provide typical examples of tests conducted during OQ 
because they will be dependent upon, and specific to, the system under test, but 
for example the tests on a dispensary area downflow booth could consist of: 
air supply system: 
o downflow and bleed air velocity (check that when correct velocity is 
achieved inside the booth the volumetric flow rate is within range); 
o green zone velocity test (to ensure that the green zone of safe airflow is set 
to correspond to an average filter face velocity of between 0.45 and 
0.55 msec-1); 
o filter pressure differential test (to ensure that the pressure drop across 
each filter is within the correct operating range and to provide a baseline 
clean filter reading); 
o dirty filter simulation test (to ensure that the airflow rate is controlled to 
maintain correct downflow velocity with dirty filters); 
control and indication system: 
o temperature control and indication system (to demonstrate the functionality 
of the temperature control and indication system and show that booth 
temperature can be maintained with specified limits with maximum heat 
load generated in the booth); 
o dehumidiflcation control and indication system (to demonstrate the 
functionality of the dehumidiflcation control and indication system and 
show that booth humidity can be maintained with specified limits with 
maximum moisture load generated in the booth); 
containment systems: 
o HEPA filter integrity testing (check that all HEPA filters are integral and 
pass the DOP test); 
o smoke containment test (to demonstrate using smoke that the booth 
contains emissions generated within the safe working zone at both the 
minimum and maximum safe airflow setting, and that fresh make-up air
drawn in from outside the booth is drawn in and maintained below bench 
top height through to the back of the booth); 
light and sound levels: 
o light levels (to confirm that the lighting levels are within range for an 
industrial working environment); 
o sound levels (to confirm that the sound levels are within range for an 
industrial working environment); 
safety systems: 
o air flow alarm (to demonstrate the functionality of the unsafe flow alarm 
system); 
o emergency stop (to demonstrate the functionality of the emergency stop 
system and check that all devices move to fail safe condition). 
OQ and commissioning 
OQs demonstrate the functionality of the installed system and are often carried 
out as part of commissioning. Engineering commissioning is normally undertaken 
by a 'system' vendor and is geared to starting up the 'system.' OQ work is 
more concerned with the operating parameters of the 'system' and with the 
identification and independent measurement of operating variables over their 
normal operating ranges. 
However, depending on how contracts are let and the responsibilities for the 
'system' testing are specified, the vendor or installer may be requested to carry 
out certain OQ activities as part of commissioning work. For instance, in the 
case of the commissioning of a HVAC system, it may fall within the scope of 
the engineering activities to stimulate certain 'worst case' conditions such as 
the effects on the air pressure regime of a power dip. 
The OQ protocol should require verification of the satisfactory completion 
of all such commissioning activities. 
4.8 Handover and process optimization 
Most projects undergo a period of plant handover following completion of OQ. 
This is normally the time that 'ownership' of the facility is transferred from the 
engineering function to the user function. If a main process contractor is 
running the project then this is often the point that completes their contractual 
responsibilities. 
Generally, before the next stage of the validation can begin, a period of time 
is spent optimizing the process. Process optimization can take various forms 
depending upon the nature of the process and facilities. For example in BPC 
plants this may encompass 'solvent trials', where solvents to be used in the
facility are first introduced. This may require re-tuning of control loops that 
have only previously operated with water. The nitrogen system may now switch 
from running on compressed air over to running with nitrogen. Plant safety is 
clearly of primary concern during this phase. 
Typically during this period operator training will be underway and the 
SOP's required to operate the facility, run the process, and maintain the 
equipment will be finalized. 
4.9 Performance qualification (PQ) 
Prior to commencement of PQ all operators involved must be trained and the 
procedures that will be required during production must be available, since they 
should be used during the PQ. 
Performance qualification is the documented evaluation of the system to 
show that the system operates as intended throughout the anticipated operating 
ranges, under conditions as close as possible to normal production. Typically it 
will consist of various functional checks on the equipment, generally performed 
using actual product. 
PQ work should be performed on systems whose performance or process 
parameters are critical and could affect the quality of the product. Examples of 
the systems requiring PQ work are pieces of process equipment such as a 
production sterilizer and critical utilities such as a WFI system. 
As with an OQ, the critical parameters and acceptance criteria of the system 
under consideration should be defined. Once these have been defined, the test 
that is required to show the parameters are met can be designed. To successfully 
complete PQ work it is necessary to examine a number of consecutive batches 
or runs. One should also consider the variability to be expected to show that it 
does not affect product quality — i.e. 'worst case' conditions. 
Normally any samples taken during PQ testing work will be taken by the 
user's personnel, not by vendors or outside contractors responsible for installing 
and commissioning of the system. 
The contents of a PQ protocol may include for example: 
approval page; 
system description; 
purpose; 
sampling regime; 
testing regime; 
acceptance criteria; 
deviation and corrective action.
4.10 Process validation (PV) 
Process validation is defined as: 
'Establishing documented evidence which provides a high degree of assurance 
that a specific process will consistently produce a product meeting its predetermined 
specifications and quality attributes' 
In essence, a PV is a PQ of the manufacturing process. As with a PQ, the 
critical parameters and acceptance criteria of the process steps should be defined. 
The parameters can be associated with the raw materials used in the process, with 
the equipment used, or with process variables (time, pressure, temperature, etc.). 
Identifying the critical parameters and understanding how each of them can 
adversely affect the finished product is the first step in the validation cycle. 
The second step is to examine the effect of each of the critical parameters on 
the process to ensure that the variability in the parameter anticipated during 
routine production does not adversely affect the quality of the product. This 
procedure of examining the practical limits of the critical parameters is often 
referred to as 'worst case' validation or 'most appropriate challenge' conditions. 
It is essentially examining the robustness of the process. 
The third step to successfully complete PV work is to examine a number of 
consecutive batches (usually three). The sampling and testing of these batches 
should be designed around the critical parameters. This step is what many 
companies have traditionally undertaken to validate their process. It is essentially 
examining the reproducibility of the process, and is acceptable if the 
process being validated is robust; but this is often not the case — hence the 
need for the first two steps. 
The process should be considered as a series of functional steps. Each step 
should have a recognizable end point, or deliver a significant change to the 
material such as an increase in bulk, change of identity, change of physical or 
chemical form, change of container. 
Process validation is associated with the process and not with the product. It 
is the list of instructions that is being qualified. An alternative process that 
produces the same product will be subjected to a separate process validation. 
Each functional step must be examined three times. In many instances a batch 
will comprise a number of sub-lots — it is not necessary to examine every 
functional step in all sub-lots of the three subject batches. 
The protocol is often based on demonstration batches or manufacturing 
batch records. The contents of a typical PV protocol should include: 
approval page; 
system description;
purpose; 
sampling regime; 
testing regime; 
acceptance criteria; 
deviation and corrective action. 
Process validation data is presented as a report. It is important to note that it 
is the review of all the batches involved together, not a series of separate 
individual reviews. 
4.10.1 Retrospective process validation 
When a product has already been manufactured successfully for at least three 
years (and at least twenty batches have been made), a statistical review of all the 
data pertaining to at least the last twenty batches can be carried out. 
No batches may be omitted from this review unless documented reasons are 
included to explain each individual case (examples would include equipment 
failure, or contamination not associated with the process). If more than 20% of 
past batches are omitted, the retrospective process validation should be abandoned, 
as it is likely that influencing systems are not under control. Only when 
these are identified and addressed should the validation project recommence. 
4.10.2 Sterile products 
Process validation for sterile products can be considered in two parts: 
validate the process to gain assurance that the system can deliver a sterile 
product. This would include, for example, themal mapping, thermal commissioning, 
filter integrity testing and control systems testing; 
validate the manufacturing process of the actual product including process 
technology and biological testing. 
4.10.3 Bulk pharmaceutical chemicals (BPC) 
For BPCs process validation starts at the point where the drug substance is 
chemically formed or where other impurities will not be readily removed. 
4.11 Cleaning validation 
The creation and implementation of effective cleaning processes is an essential 
part of any pharmaceutical production process. The two main reasons for this are: 
to ensure that the appropriate level of general cleanliness is maintained in 
order to prevent the accumulation of dirt and microbial contamination which 
could affect the quality of the product;
to minimize the risk of cross-contamination from one active product into the 
subsequent product, which could lead to serious adverse effects on patients. 
Cross-contamination could also result in degradation of the main product 
and loss of potency. 
4.11.1 Choice of cleaning method 
Various approaches can be taken to ensure that cross-contamination levels are 
minimized between two different products. 
The simplest approach is to dedicate a complete facility, its building, 
services and equipment, to a single product. Obviously this is a very 
expensive approach, unless the product is required in sufficient quantity to 
justify a dedicated facility. For very active products such as penicillin, 
cephalosporin and hormones, where cross-contamination at very low levels 
is not acceptable, this is the safest option and is a regulatory requirement. 
In dedicated facilities effective cleaning procedures still need to be 
developed and validated, although the stringent cross-contamination levels 
that are usually applied to multi-product facilities can be relaxed somewhat 
and the emphasis placed on general levels of cleanliness in accordance 
with GMP. 
In most circumstances though, facilities are multi-product and effective 
cleaning processes must be developed and validated by means of sampling and 
measuring the levels of cross-contamination. 
The most common type of cleaning process involves the full or partial 
dismantling of equipment, followed by solvent washing and subsequent drying 
of the separate parts. Water/steam (with or without added detergent) is the most 
common cleaning solvent, but organic solvents can also be utilized. 
Manual cleaning is still used extensively in the pharmaceutical industry but 
'clean-in-place' (CIP) systems are rapidly expanding and 'sterilization-inplace' 
(SIP) is also being introduced. 
It is quite common and also highly desirable to dedicate specific parts of 
the equipment which are difficult to clean, thereby reducing the overall time 
and cost of the cleaning process. Examples of this are the woven fibre filter 
bags used in fluid bed dryers or the rubber/plastic o-rings found in 
pipework. 
These examples illustrate the importance of designing an effective cleaning 
process using a variety of techniques before embarking on any validation work. 
Remember, successful validation will only confirm that the cleaning process is 
effective, it will not make an ineffective one effective!
4.11.2 Measuring the level of cleanliness 
As part of the overall validation programme the actual level of cleanliness that 
has been achieved by the cleaning process must be measured. This involves a 
three-stage process: 
a sampling method to detect and pick up the remaining contaminants; 
a method of analysis to quantify the amount of contaminant remaining; 
a calculation to extrapolate the results. 
The usual sampling methods are: 
swabbing; 
aqueous/solvent rinses; 
non-active product follow through. 
(a) Swabbing 
Swab testing involves the use of dry or solvent impregnated swabs, which are 
wiped over a known area of the processing equipment. The contamination 
picked up is extracted in the laboratory by soaking the swab in a suitable 
solvent, and the solvent is then analysed to give a quantitative result. The total 
quantity of the contamination is calculated by multiplying the total area of the 
equipment by the swabbed area. In practice, the swab is unable to pick up 100% 
of the contamination, but it is possible to run a laboratory test beforehand to 
estimate the percentage pick up. This is done by deliberately contaminating the 
stainless steel plates (or sample of whichever material is in contact with the 
product) with a known quantity of contaminant, usually letting a solution 
evaporate on the plate. The plate can then be swabbed and the swab analysed to 
demonstrate the percentage of the contaminant that has been picked up. The 
analytical method must also be checked to ensure that the swab itself does not 
interfere with the result by running blank swab tests. 
(b) Aqueous/solvent rinses 
Aqueous/ solvent rinses are commonly used in areas where it is difficult to swab 
(such as pipework or a sealed reactor in a bulk chemical plant). The method 
involves rinsing with a known volume of water/solvent and then analysing a 
small quantity of the rinse. The total amount of contaminant is simply: 
Quantity in sample x Total volume of rinse 
Volume sample 
The solvent used must provide sufficient solubility to pick up the contamination 
effectively but must not degrade the contaminant. The contact time must be 
controlled. 
The main drawback of this method is that only material dissolved in the rinse 
water/solvent would be analysed and it would not be possible to find out how
much was left inside the pipework, vessel, etc. The solubility of the contaminant, 
contact time and physical force of the rinse will all affect the final results, 
and it may not be possible to ensure all the areas have been adequately rinsed. 
(c) Non-active product follow-through 
The non-active product follow-through is sometimes used, and involves 
processing a non-active substance through the whole process and then analysing 
samples for the contaminant. The calculation is analogous to that used for 
the rinse method, but this method has the advantage that it mimics the real 
situation of a subsequent batch being processed, and that it covers all the 
equipment involved. However, as with the rinse method, only the contaminant 
that has been picked up can be measured, and not the contaminant left behind. 
Also, in the case of solid dosage forms, the contaminant may not be uniformly 
mixed throughout the non-active substance. 
The swabbing method is generally preferred because it permits the areas 
likely to be most heavily contaminated to be targeted more thoroughly and also 
makes allowance for contamination not recovered, provided the laboratory tests 
are undertaken. Despite all this, it is still prone to variability since no two 
samplers will swab in exactly the same manner. The inherent variability in any 
of the sampling methods is one of the reasons for the use of a 'Safety Factor' 
when calculating the acceptable contamination limit. 
4.11.3 Setting limits 
When a cleaning process is used only between batches of the same product (or 
different lots of the same intermediate in a bulk process), it is normally only 
necessary to meet a criteria of 'visibly clean' for the equipment. Such betweenbatch 
cleaning processes do not normally require validation. 
Chemical cross-contamination limits 
One of the basic concepts of validation is that a process is proven to be capable 
of performing to a pre-defined limit. There is no exception with cleaning 
validation and although agreeing a pre-defined limit can be difficult, it is 
essential to establish one prior to commencing the validation work itself. 
As there are often no obligatory legal or regulatory limits, manufacturers 
have come up with their own viable methods for setting limits. 
The simplest of these methods is to set a blank limit to all products. A typical 
limit would be 1 to 10 ppm. This approach has been used in the bulk pharmaceutical 
chemical production and product development areas where a large 
number of compounds are processed and for many of them relatively little is 
known about their properties. The scientific rationale for limits in the region of 
1 to 10 ppm is that this is somewhere near the limit of detection for suitable
analytical methods for many compounds, and pharmacopoeia limits for heavy 
metals and other adulterants tend to lie in this region. The problem with this 
approach is it makes no allowance for the different pharmacological effects of 
different compounds. This will lead to excessive cleaning and wasted time and 
resources in some cases, whilst in other cases it may leave patients exposed to 
potentially hazardous levels of contamination. 
Several companies have adopted a limit where the maximum amount of 
contaminant (A) that can be ingested by a patient taking the product B, 
manufactured immediately after product A, is one thousandth of the minimum 
normal therapeutic daily dose. The figure 1000 is used as a safety factor, which 
not only reduces the daily dose below pharmacological activity level but also 
allows for the errors inherent in the sampling and testing methods used. 
Finally, the limit of detection for the assay method must be considered. 
Setting a limit of 0.001 mg per swab when the assay limit is 0.01 mg is pointless. 
Either the assay method needs developing, or the limit of assay will have to be 
the acceptance criteria. 
Microbiological cross-contamination limits 
Most cleaning validation protocols do not include sampling and testing 
procedures for microbial contamination. This is because the sterilization 
itself is validated for processes where minimization of microbial contamination 
is important (sterile and aseptic). 
It is important that the cleaning procedure does not actually increase 
the level of microbial contamination. This requires the cleaning agents to have 
a low level of microbial contamination, and the drying procedures to adequately 
remove all traces of water. Storage of equipment is also important — it should 
be kept clean and dry and well covered or wrapped. There should be a maximum 
storage time defined, after which the equipment is cleaned again. 
Where it is felt necessary to confirm that a particular level of microbial 
contamination has been achieved, swabs can be impregnated with a suitable 
growth media. The use of media impregnated swabs or media solutions will itself 
contaminate the equipment, which must be cleaned thoroughly before routine use. 
4.11.4 Validation of CIP systems 
For CIP systems there are several steps to be undertaken before any actual 
sampling and testing is carried out. 
CIP validation cycle 
Assess design of CIP system including analytical method development; 
Experimental work to optimize cycle and cleaning agents and including 
analytical method validation;
Change control system; 
Operational qualification; 
Cleaning validation protocol; 
Cleaning validation report for three successive cleaning cycles. 
CIP systems are usually fitted to large immovable pieces of equipment, such as 
dryers and coaters. Often the CIP system will adequately clean the large flat 
surfaces of the equipment, but will leave excessive amounts of material in the 
corners, crevices, inlet/outlet ports, and around and behind seals and flaps. 
Therefore, before starting with validation protocols, the design of the CIP 
system should be assessed to eliminate (or at least minimize) any obvious weak 
areas. For example, one simple test often performed to determine coverage 
involves coating the item to be cleaned with an appropriate dye, then operating 
the cycle to determine if all the dye can be removed. If alterations to the CIP 
system itself are impractical, then it may be possible to remove part of the 
equipment for separate manual washing. 
The main advantage of a CIP system is that it should provide a reproducible 
cleaning process. This process needs to be effective and optimized to provide 
the best chance of successfully validating the cleaning process. Experimental 
work can be performed using different wash cycles, rinse cycles, detergent 
types, drying conditions, etc. to establish the most effective conditions. If a 
range of products is to be cleaned then experiments should be performed on the 
most difficult to clean product. 
Having established the most effective conditions, the CIP system and 
cleaning cycle should form part of the formal OQ for the equipment, to 
demonstrate that the critical parameters used in the cleaning cycle can be 
satisfactorily achieved and reproduced. 
In parallel to the experimental work and OQ activities, analytical methods 
will have to be established and validated. 
Finally, the cleaning validation/PQ protocol can be written and executed. 
This protocol can be either a stand-alone document or part of the general PQ 
protocol. Either way, the cleaning validation protocol is specific to a particular 
changeover between two products on a specific set of equipment. 
The protocol should include the following sections: 
definition of equipment being used; 
definition of the product(s) being cleaned from the equipment, and the 
product that will subsequently occupy the equipment; 
explanation of the parameters being used in the cleaning process (temperature, 
times, pressures, detergent types and concentrations, etc.); 
sampling regime (sampling method(s), number and location of samples);
testing procedures (description of tests to be performed on samples); 
acceptance criteria (acceptable maximum levels of contamination in each of 
the samples). 
The validation protocol should be performed on at least three successive 
occasions to demonstrate reproducibility. 
When the analysis of the samples is complete, the data should be collated, 
summarized and presented in a validation/PQ report. Comparison of the data to 
the pre-determined acceptance criteria will form the basis of the conclusions. 
Any missing data or data that is outside the acceptance criteria should be 
accompanied by an explanation. If the validation has failed then the cleaning 
process will have to be altered and the work repeated. 
On completion of a successful cleaning programme, the validated cleaning 
procedure must become subject to the plant's change control system. 
4.11.5 Validation of manual cleaning 
Manual cleaning validation cycle: 
Experimental work (optimize cleaning method, drying cycle, etc.); 
Change control system; 
Prepare standard operating procedure (SOP); 
Operator training including retraining/re-evaluation; 
Evaluation of training; 
Cleaning validation protocol; 
Cleaning validation report. 
Most equipment is relatively small, easily dismantled and portable to 
facilitate frequent and rapid cleaning. Operators often dismantle, clean and 
reassemble the equipment. 
Operators are people and are therefore variable. Whilst it is virtually 
impossible to totally eliminate this variability, it can be minimized to an 
acceptable level by the use of clear and concise instructions (SOPs) together 
with regular training and assessment of the operators. Part of the validation of 
any manual cleaning method should involve the evaluation of the process to 
determine the level of variability — a high variability (even if within acceptable 
limits) suggests a process that is poorly controlled. 
The actual validation protocol will be very similar to that used for the CIP 
system validation, but it must refer to any SOPs associated with the cleaning 
procedures.
4.12 Computer system validation 
Automated or computerized systems are validated using the same general 
validation approach identified for equipment and utilities. However the nature 
of computer systems means that certain activities become particularly critical. 
A software programme is not a tangible thing and cannot be tested exhaustively 
(i.e. with large programmes it is impractical to prove the code) since to test 
every possible path through the code under every possible set of circumstances 
would take an inordinate length of time. For this reason the quality and 
confidence must be 'built in'. Software development must be carefully planned 
and controlled under a quality assurance system following a life-cycle 
approach. It should be noted that the term 'computer system' refers to the 
computer hardware and software as well as the interface between the computer 
and the machine/plant/environment. 
Various models have been developed for the validation of computerized 
systems such as that proposed by IEEE (IEEE Standard for Software Verification 
and Validation Plans); the PDA report on the validation of computerrelated 
systems or the GAMP (Good Automated Manufacturing Practice) 
Supplier Guide for Validation of Automated Systems in Pharmaceutical 
Manufacture. All these models are fairly similar. This section will not cover 
in detail the 'engineering' associated with the design, development and testing 
of computer systems but will concentrate on the validation activities associated 
with each stage. 
4.12.1 Assessment of computer systems to determine validation 
requirements 
The necessity for computer system validation is based on several criteria. The 
first of which is that the element in question is to be classified as a computer 
system (for example, some instruments may be programmable and may or may 
not be treated as a computer). The following criteria should help determine 
whether the element is a computer system: 
inputs and outputs (I/O): The presence of physical channels (digital, 
analogue, pulse, serial, etc.) for importing or exporting data that is used or 
has been calculated by the element; 
memory: A means of storing executable code is used; 
Central Processing Unit (CPU): Use of a device for interpreting executable 
code using data accessed from inputs, and presenting the result via outputs. 
If all the above criteria are present then the element can be assumed to be a 
computer system and should be treated as such from a validation point of view.
The next step is to determine if validation is required. This involves a 
process of evaluating the role that the computer system plays. Assessment 
criteria include: 
GMP implication: Generally any computer system with GMP implications 
should be validated. This includes for example critical operations such as 
controlling or monitoring operations that can affect product quality; 
system functionality: If the computer system is only used for supervisory 
tasks, with no computer-generated information being used by or forming part 
of the batch record information then generally the computer system does not 
require validation; 
safety critical systems: Although GMP does not cover safety critical 
systems, there is a good argument for them being treated in the same way; 
system configuration: Although a computer system may be involved with 
critical operations, it might be that another independent system provides a 
full check of the operation of the computer system. In this case the computer 
system does not generally require validation; 
system operability: Although the system may be computerized, the corresponding 
operating procedures may introduce so many manual operations and 
checks that all computer controlled operations are duplicated by the way the 
system is operated. In this case the computer system does not generally 
require validation. 
Once it has been determined that computer system validation is required, the 
detailed validation activities will need to be determined. The extent of computer 
system validation depends upon two main factors — the level of standardization 
and the complexity of the system. A standard system has been largely 
validated by its wide use, so most of the validation effort should go into 
validating the system with respect to the user's particular circumstances. The 
issue of system security (prevention of modification or reconfiguration) must 
also be addressed. Generally the simpler the system, the less validation effort is 
required. There is a risk that because simple systems are easier to understand 
they tend to be more 'fully' validated. Instead increased emphasis should be 
placed onto more complex systems. 
These two criteria should be applied to both the computer hardware and 
software. 
Hardware 
The hardware can be classified as either standard hardware (produced in large 
quantities over an extended period) or application specific (mainly produced for 
the applicable project only). Both will require validation but the approach to
standard hardware is simpler, mainly being concerned with the configuration, 
installation and functional testing aspects. The design and design process must 
also be considered for application specific hardware. This may involve 
assessing the methods employed, critical components, compatibility between 
units, standards used for design and testing, type testing carried out, etc. 
Software 
There are generally three types of software that can be identified for computer 
systems: 
system software: This is the software required to run the computer system 
itself. It includes all the operating systems (the software controlling the CPU, 
memory, I/O, operator interfaces, etc.) as they are configured for a particular 
computer system. Normally this software does not require validation because 
it is classified as 'standard software' (see below). 
configurable software: As the name implies, this type of software would 
normally be standard software, which can easily be adapted to an applicable 
project, such as Lotus 1-2-3 for example. The software purchased from Lotus 
is classed as standard software, which does not require validation (because of 
the wide use of this software), but its use with formulae applicable to a 
specific project must be validated. Configurable software is also sometimes 
referred to as 'canned software'. 
application software: This software is produced or configured specifically 
for the applicable project and must be validated. 
The term 'standard software' is often used as a reason for not performing 
validation. The following criteria may be used to determine if a piece of 
software is standard: 
the supplier's QA system: Ideally this should be a recognized system such 
as ISO9000 or similar and it should demonstrate that development and 
testing of the software is controlled and documented; 
the product being widely used: This is generally interpreted as meaning 
more than 100 similar units. It is of further advantage if the software has been 
applied to a wide range of applications, and thus more thoroughly exercised 
and tested; 
product age: Product history and experience including knowledge of 'bugs' 
will increase with age. Standard software is usually expected to have been in 
wide use for a minimum of twelve months.
version control: Software is usually developed and corrected during its lifecycle. 
The number of software versions can be great, so a system of version 
control must be in place to be able to take all versions into account with 
respect to product age and usage; 
user feedback: The vendor must be able to demonstrate that feedback from 
users is handled and acted upon; 
not application specific: The software cannot be classed as standard if parts 
of it are specific to the particular application. 
If all the above factors are fulfilled then the software can be classed as 
standard and does not require validation. However the computer system may 
still require validation including functional testing. 
The results of the above assessment should be documented and included in 
the Validation Master Plan. 
4.13 Analytical methods validation 
Analytical methods can be validated in a number of ways. Compendial methods 
such as methods appearing in the USP are generally considered validated, but it 
is important to demonstrate that the method works under the actual conditions 
of use. If a compendial method exists but a company elects not to use it, they 
must demonstrate that the in-house method is equivalent or superior to the 
official procedure. 
Validation data from repetitive testing should be consistent, and varying 
concentrations of test solutions should provide linear results. 
4.14 Change control and revalidation 
4.14.1 Change control 
All process and plant subject to validation should be covered by a change 
control system that enables formal reporting and recording of changes, reviews 
the impact of a change on the validation status and permits revalidation 
requirements to be identified. 
Change control standard operating procedures should define which changes 
do and do not require change control. Generally, items subject to change control 
include: 
procedures that contain validated activities or processes (for example, 
cleaning, equipment operation, sterilization);
process equipment and plant; 
facilities; 
utilities; 
production processes; 
commodities (primary packaging components, filters, sterile clothing, disinfectants, 
cleaning agents); 
raw materials; 
computer systems; 
test methods and specifications. 
Standard operating procedures and change control forms should allow all 
proposed changes to be considered, commented upon and approved or rejected 
by relevant experts. These experts generally represent Quality Assurance 
(whose authorization is always required), Production, R&D, and Engineering, 
though other experts may be consulted as necessary. Reviewers should identify 
whether the change needs to be validated and, if so, outline the nature and 
extent of validation required. 
It is recommended that change control forms reference qualification 
protocols in those cases when revalidation is necessary. The date of reintroducing 
the process or plant subject to change into operation should be 
recorded so that it is clear that revalidation, when required, has been completed 
before use. 
On occasions, where an emergency situation occurs, an unplanned change 
may have to be implemented without prior formal consultation. In such cases 
details of the change should be introduced into the change control system as 
soon as possible. 
Where a planned change is not approved, it must not be implemented. Where 
an unplanned change is not approved, the process or plant must immediately be 
returned to its original state. 
4.14.2 Revalidation 
In order to maintain the plant, facilities, systems, procedures, methods and 
processes, once initially qualified, in a state of validation throughout their lifecycle 
there should be continuous review of the need for revalidation and 
implementation of revalidation whenever it is agreed to be necessary. 
Revalidation requirements should be defined based on a technical review 
of the initial qualiflcation(s), change control data and documentation supporting 
the performance of the item subject to validation. Revalidation will be 
undertaken if a change is likely to affect the validated status or if the
performance of the validated system is seen to have deteriorated. Revalidation 
exercises should be built into the Validation Master Plan. 
The need for revalidation may be identified via several mechanisms: 
through a change control procedure; 
by regular review of the performance of a validated item to a predetermined 
schedule; 
by the use of a plant certification system; 
through annual product reviews; 
through internal audits. 
Critical items of the plant are frequently covered by a routine certification 
and re-certification programme. Revalidation intervals and the test to be 
conducted are normally specified at the time of certification. 
Summary 
The key points from this chapter are as follows: 
validation is required to provide documented proof of GMP compliance. 
Validation activities should be organized as a scientific study that follows a 
life-cycle approach; 
validation activities should be conducted in accordance with pre-defined 
company validation policies and procedures under a validation master plan; 
the validation master plan(s) should define what will be validated, describe 
the validation approach to be adopted (this will reference the policies and 
procedures developed) and explain how the validation work will be organized 
and related documentation will be controlled; 
the validation activities should be lead by a validation team, which should 
consist of members from relevant disciplines participating within the project 
including members of the QA/QC function. The team will be responsible for 
organizing the validation activities and reviewing and approving associated 
documentation; 
the processes should be evaluated to determine what aspects are critical and 
require validation. This may include determining critical process steps, 
critical parameters and critical instrumentation and systems; 
in parallel with process evaluation, systems and system boundaries should be 
defined. This allows validation work to be broken down into logical, 
manageable sized packages and concentrates the validation effort in the 
most important areas;
cGMP reviews should be performed at key points in the project life-cycle to 
confirm that the design complies with cGMP requirements and the 
specification; 
User Requirement Specifications (URS) should be prepared by the user for 
each system to be validated to formally document the final process requirements. 
These will form a key part of the basis for subsequent validation 
activities; 
validation activities should be documented and controlled through the use of 
qualification protocols and reports, typically these will fall into categories 
including DQ, IQ, OQ, PQ and PV
This chapter considers the production of the bulk active ingredient or bulk 
pharmaceutical chemical (BPC) that is subsequently converted by physical 
means into the final drug's presentation form. 
This area of the pharmaceutical industry has much in common with fine 
chemical manufacture. The unit operations carried out are similar and many 
fine chemical and speciality chemical manufacturers also manufacture pharmaceutical 
intermediates. 
Traditionally, the bulk production was carried out on a different site to the 
R&D and secondary processing. The style of operation, attention to cGMP and 
culture of a primary site, was more associated with the type of chemistry or 
operation carried out. 
Three main influences are changing the face of the BPC industry: 
regulators, particularly the FDA, are putting greater emphasis on reviewing 
BPC production, and recognize the effect that failure in quality can have on 
the finished dosage form; 
major pharmaceutical companies are focusing on 'Research and Development' 
and 'Marketing and Selling of the finished product'. Secondary 
manufacture to a limited extent, and primary or BPC manufacture to a 
greater extent, is being sub-contracted out to third parties; 
BPCs are becoming more active and tonnage requirements are dropping as a 
result. Linked with this, the size of the equipment used in the manufacture is 
reducing. The increased activity also brings increased handling considerations 
and limits for exposure, which in turn drives towards closed processing 
operations, which is also consistent with improvements to cGMP. 
5.1 Reaction 
The production of the BPC is by three main methods: 
chemical synthesis: Examples of synthetic conversions include aspirin, 
diazepam, ibuprofen. This method produces the largest tonnage; 
5 
P r i m a r y 
p h a r m a c e u t i c a l 
p r o d u c t i o n 
ROGER SHILLITOE, PHIL MASON and FRED SMITH
biotechnology or microbial action: Examples include antibiotics, vaccine 
production, blood plasma products. This method produces the high value 
products; 
extraction: This can be by extraction of natural materials from animal or 
plant material such as the opium alkaloids, dioxin, heparin, insulin (pigs 
pancreas), thyroxine (animal thyroid gland). 
This chapter will concentrate on the first two methods. The extraction 
method for naturally occurring materials was the main source of drugs up to the 
1930s but was being gradually replaced with synthetic routes to products. 
There is resurgence now in extraction techniques linked to the biotechnology 
area, where specifically developed or altered organisms are allowed to grow and 
produce a desired product that is harvested and extracted. This is discussed in 
Section 5.1.2. 
5.1.1 Synthetic chemistry based processes 
Various general synthetic chemical reactions are utilized in the synthesis of 
BPCs. These include simple liquid/liquid reactions, complex liquid reactions 
with catalysis such as Grinards, Freidel Craft, reaction with strong reagents 
such as phosphorous oxychloride, thionyl chloride or elemental halogens such 
as bromine or chlorine. Gas reactions with liquids are common for example 
with hydrogen, hydrogen chloride or phosgene. 
Most reactions in the pharmaceutical industry are carried out on a batch 
basis, in non steady state operation. Continuous processing is occasionally used 
for a few generic tonnage commodity BPCs or where safety can be improved by 
the benefits continuous processing can bring by inventory minimization. 
Conventional batch reactor systems 
The batch reactor is the workhorse of the synthetic BPC industry. Typically 
made from stainless steel or glass lined mild steel, capacities ranges from 500 
litres at the small scale to 16 m3 at the large scale. Some processes employ 
reactors of even greater capacity but this is becoming unusual as the activity of 
new drug substances increases. 
The reactor is typically fitted with an external jacket or half pipe coils so that 
the temperature of the contents can be adjusted. Occasionally if a high heat duty 
is required, further coils can be placed inside the reactor. 
Typical operating conditions are from — 25°C to + 1600C, and full vacuum 
to 6barg. Generally, reactions at elevated pressures above 1 barg are uncommon, 
with the exception of specific gas reactions such as hydrogenation. 
However, more processes are now being developed where working at an
elevated pressure brings benefits — for example, it can allow the selection of 
the ideal solvent for a reaction that could not normally be used at the ideal 
reaction temperature because this would be above its atmospheric boiling point. 
The temperature is normally adjusted by indirect contact with a heating or 
cooling medium circulating through the coil or jacket, but direct heating with 
live steam or quench cooling with water or other materials is possible. The 
medium used for the heating and cooling fall into two main areas: 
multiple fluids: typically steam, cooling water, refrigerated fluid such as 
ethylene glycol or brine. These are applied in sequence to the coil or jacket as 
required; 
single fluids: typically some form of heat transfer oil, heated or cooled by 
indirect contact with steam, cooling water or refrigerant, and blended to 
provide the correct fluid to the coil or jacket. 
Agitation is provided to the reactor to ensure good heat transfer and good 
mixing for reaction. Depending on the process requirements, various agitation 
regimes can be set up using different agitator profiles, speeds and locations. 
Connections are made to both the top and bottom of the reactor to allow 
material to be charged into the reactor, material to be distilled from the reactor, 
and liquids to be drained out. 
Reactors are normally fitted with a manway to allow entry for maintenance 
purposes. Historically, this was also the way in which solids were added to the 
reactor and samples were extracted, but this practice is becoming less common. 
Alternative reactor systems 
Other types of reactor systems exist with each having their own specific 
advantages for specific processes. These include the loop reactor that specializes 
in gas-liquid reactions at elevated pressures, such as hydrogenation, and 
the batch autoclave reactor that specializes in high-pressure reactions of 
100 bar g and higher. 
Materials of construction 
Reaction modules can be constructed from other materials dependant on the 
chemistry being employed and requirements for heat transfer. These include 
glass, plastics and exotic metals such as hastelloy or titanium. 
5.1.2 Biotechnology based processes 
The processes in biotechnology are based on cultivation of micro-organisms, 
such as bacteria, yeast, fungi or animal and plant cells. During the microbial 
process the micro-organisms grow the product, which is either contained within
the cell or excreted into the surrounding liquor. The micro-organisms need 
carbon substrate and nutrient medium for growth and the microbial process is 
normally performed in water. 
There are essentially three steps to biotechnology processing, namely: 
fermentation; 
recovery; 
purification. 
The equipment in which the microbial process is carried out is called the 
fermenter and the process in which micro-organisms grow or format product is 
called fermentation. 
Once the product is formed it is recovered from the biomass or the liquor by 
downstream processing, e.g., centrifugation, homogenization or ultraflltration. 
Purification of the recovered product is then required. Two differing 
techniques are required depending on whether it is for bulk large-scale or for 
small-scale genetically manipulated organisms. Large-scale recovery can be 
likened to bulk chemical organic synthesis operation. 
Fermentation 
The fermenter is the equipment used to produce the micro-organisms. 
Biotechnology applications of fermentations divide conveniently between 
microbial types and mammalian cell culture. Microbial fermentation, which 
can encompass very large-scale antibiotics as well as smaller scale recombinant 
products, is characterized by fast growth rates with accompanying heat and 
mass transfer problems. Mammalian cell culture is characterized by low growth 
rates and high sensitivity to operating conditions. Both techniques have 
common design principles. 
Several different types of vessel are used for large-scale microbiological 
processes, and their degree of sophistication in design, construction and 
operation is determined by the sensitivity of the process to the environment 
maintained in the vessel. 
The following is a brief description of the main types of fermenters: 
(a) Open tank 
The simplest type of fermenter is an open tank in which the organisms are 
dispersed into nutrient liquid. These have been used successfully in the brewing 
industry. In the anaerobic stage of fermentation, a foam blanket of carbon 
dioxide and yeast develops which effectively prevents access of air to the 
process. Cooling coils can be fitted for controlling temperature during fermentation.

(b) Stirred tank 
Stirred-tank fermenters are agitated mechanically to maintain homogeneity, to 
attain rapid dispersion and mixing of injected materials, and to enhance heattransfer 
in temperature control and mass-transfer in dissolving sparingly soluble 
gases such as oxygen. The extent to which these are achieved depends mainly 
on the power dissipated into the medium by the agitator, so that the agitator is 
essentially a power transmission device. The effectiveness of the power input 
depends on the configuration of the agitator and other fermenter components. 
For aerobic fermentations, air is injected through a sparger, a single nozzle 
or a perforated tube arrangement, positioned well below the lowest impeller to 
avoid swamping it with gas. The sparger should have provision for drainage so 
that no culture medium remains in it after the vessel is discharged. 
The rate of air supply must be sufficient to satisfy the oxygen demand of the 
fermentation after allowing for the efficiency of oxygen dissolution achieved. 
Instead of a rotating stirrer, some systems obtain the mechanical power input 
by using a pump to circulate liquid medium from the fermenter vessel through a 
gas entrainer and then back into the fermenter. This separates the liquid 
movement and gas dissolution functions into separate specialized units. Two 
designs have evolved using this principle — the 'loop' fermenter and the 'deep 
jet' fermenter. In the loop fermenter, the gas dissolution device is a subsidiary 
vessel into which gas is injected, and the gas-saturated liquid is recirculated to 
the main growth stage. In the deep-jet system, gas is entrained into a highpower 
jet of liquid injected into the liquid in the fermenter, re-entraining gas 
from the vessel headspace. Exhaust gas is purged partly from the vessel 
headspace and partly from the specially designed circulation pump, from 
which the degassed liquid passes through a supplementary cooler before 
passing to the gas entrainer. This system gives high gas dissolution rate, but 
has correspondingly high power consumption compared to conventional 
systems. The liquid and entrained gas can also be introduced into the fermenter 
through a 'bell', which holds the gas bubbles in contact with the recirculating 
liquid to enhance gas utilization. 
(c) Gas-lift and sparged-tank fermenters 
This design has no mechanical stirrer and the power required for mixing, heattransfer 
and gas dissolution, is provided by the movement of gas through the 
liquid medium. The gas is, therefore, the power transmission system from the 
gas compressors into the vessel. While the relatively low efficiency of gas 
compression seems to make this design unattractive, it has some important 
advantages compared to the stirred-tank system. Firstly, the absence of a 
rotating agitator shaft removes the major contamination risk at its entry point
to the vessel. Secondly, for very large vessels, the required power input for 
agitation is just too large to be transmitted by a single agitator. Thirdly, the 
evaporation of water vapour into the gas stream makes a small contribution to 
cooling the fermentation. The fermenter interior does, however, need careful 
design to ensure that the movement pattern of the gas through the system 
produces satisfactory agitation. 
The various designs of non-mechanically agitated fermenters can be 
grouped broadly into sparged vessels and gas-lift (including air-lift) fermenters. 
Sparged-tank fermenters are usually of high aspect ratio, with gas introduced at 
the bottom through a single nozzle or a perforated or porous distributor plate. 
The gas bubbles rise through the liquid in the vessel and may be redispersed by 
a succession of horizontal perforated baffle-plates sited at intervals up the 
column. In the gas-lift fermenters, internal liquid circulation in the vessel is 
achieved by sparging only part of the vessel with gas. The sparged volume has a 
lower effective density than the bubble-free volume, and the difference in 
hydrostatic pressure between the two sections drives the liquid circulation 
upwards in the sparged section and, after gas disentrainment, downwards in the 
bubble-free section. The two sections may be separated by a vertical draughttube. 
Important design considerations for good fermenter operation 
The following are important design considerations in fermenter operation: 
(a) Aeration and agitation 
Animal cells are shear-sensitive (mild agitation is therefore required) and they 
are often sensitive to air bubbles. These considerations impose significant 
constraints on oxygen transfer design. One way in which this problem has been 
addressed is by the use of gas exchange impellers. Another strategy is to 
circulate medium through the reactor while simultaneously oxygenating it in an 
external loop. A third method is to use silicon tubing through which air diffuses 
into the liquid medium. 
Cell culture medium often contains serum, which has a tendency to cause 
foaming. Since defoamants may inhibit growth, agitation and aeration systems 
must be designed to minimize this potential problem. However, care must be 
taken in the amount of agitation applied because, although it provides good 
oxygen and heat transfer characteristics, it can result in mechanical degradation 
of the cells. Usually systems with gentle agitation also minimize foaming. The 
type of impeller, baffles, and tank dimensions influences the degree of mixing. 
Note that mammalian cell cultures are more easily damaged by these mechanical 
forces than microbial cultures.
(b)PH 
The internal environment of living cells is approximately neutral, yet most 
microbes are relatively insensitive to the external concentrations of hydrogen 
and hydroxyl ions. Many organisms grow well between pH 4 and 9, although 
for any particular organism the required pH range is small and accurate control 
is essential. Note, however, that there are exceptions where growth outside this 
range can occur, 
(c) Sterile design 
The importance of sterile design cannot be over emphasized; even the presence 
of a single contaminant will be disastrous. The fermenter must be designed to 
be easily cleanable (smooth surfaces and no crevices), after which it must be 
sterilized. The most effective form of sterilization is to utilize clean steam to kill 
both the live micro-organisms and their spores. This is usually defined as 
maintaining 1210C for 20 minutes. Shorter times and higher temperatures can 
be used but not vice versa. The quality of the steam supply is important; clean 
steam is required for mammalian cell culture, whereas, plant steam with 
approved additives can be used for large-scale antibiotics. 
If the fermentation design calls for sterility, the following special precautions 
are required: 
air should be provided by an oil free compressor; 
Clean in Place (CIP) and Sterilize in Place (SIP) systems should be 
incorporated into the design; 
the fermenter and all associated piping and vessels should be designed to 
allow sterilization initially by 1.5 bar g steam. Branch connections should be 
minimized. All lines should be free draining and have minimum dead legs 
with the correct type of valves specified. Selection of internal surfaces, 
piping design, and valves is critical in ensuring effective removal of 
unwanted organisms during sterilization and preventing subsequent ingress 
of contaminants from outside the sterilized system; 
many fermentation media, at the large scale, can be sterilized continuously 
by heat. Economies can be achieved by incorporating heat recovery 
exchangers in the system to preheat the feed; 
all seals and instruments must be designed to withstand steam sterilization; 
the equipment should be designed to maintain sterility e.g. to include the use 
of steam seals on agitator inlets, double O-rings for probe insertion and 
steam blocks on transfer lines; 
piping should be stainless steel; 
an integrated approach should be taken to the physical layout, the piping and 
instrumentation (P&ID) flowsheeting and the sequencing to ensure that 
sterility is an integral part of the design;
(d) Temperature control 
The temperature for organism growth ranges from approximately —5°C to 
800C. However, the actual temperature is important, particularly for cell 
cultures, so temperature control is critical. The lower limit is set by the freezing 
point of water, which is lowered by the contents of the cell. The upper limit 
depends on the effect of temperature on the vital constituents of the organisms 
— for example, protein and nucleic acids are destroyed in the temperature 
range 50° to 900C. 
(e) Media sterilization 
Medium ingredients should be controlled through a careful quality assurance 
programme. However, sterilization is also required and there are essentially 
three methods used: 
continuous sterilization for large scale. The time and temperature of the 
continuous sterilizer should be optimized based on the most heat resistant 
contaminant. The hold section of the continuous sterilizer should be 
designed for plug flow to prevent back mixing; 
in-situ batch sterilization by heat for smaller batches; 
sterilization by filtration for heat sensitive products such as cell culture. 
Recovery and purification 
The product separation and purification section is critical to the design of a 
fermentation plant; indeed, the bulk of capital and operating costs for a typical 
plant are often connected with this area. The design of product recovery 
systems encompasses both intracellular and extracellular products from both 
microbial and mammalian cell fermentation broths: 
(a) Large-scale extracellular products 
Technologies for recovering the simpler extracellular products consist of 
conventional unit operations such as vacuum filtration, crystallization, liquidto-
liquid extraction, multi-effect evaporation, precipitation and distillation. 
These are similar to the basic organic synthesis processes detailed earlier in 
this section. 
(b) Recombinant products 
Recombinant therapeutic products can be intra- or extracellular depending 
upon the host micro-organism. Recovery facilities for the more complex 
intracellular protein products involve cell harvesting, debris removal, pellet 
washing and recovery, product concentration, desalting, purification and sterile 
product finishing operations. 
The recovery and purification of protein products from fermentation broths 
involves rapidly-evolving, state-of-the-art unit operations. The complexity of
these operations is increased due to the heat and shear sensitivity of the proteins 
being recovered. 
The use of recombinant-DNA organisms can also affect the design of the 
cell recovery area. If the organisms are not killed in the fermentation area, the 
recovery area handling the live organisms must be designed in accordance with 
applicable guidelines for containment. 
Typical methods for recombinant product isolation and purification include: 
(a) Cell disruption 
For intracellular products the product of interest is inside the cells. The 
objective of cell disruption is to release this product for further separation. 
Cell disruption is usually carried out by mechanical means. This can be by use 
of homogenizers, grinding by beads or by high pressure liquid jet impacting. 
Other methods are use of sound, pressure changes or temperature changes and 
chemical methods. The separation of product from the cell debris after cell 
disruption is usually done by centrifugation. 
(Z?) Centrifugation 
Centrifuges are commonly used for cell harvesting, debris removal, and pellet 
washing operations. Cells can be separated using disc-stack or scroll decanter 
centrifuges. The latter allows cell washing prior to subsequent processing. The 
arrival of steam sterilizable, contained designs have made the use of such 
machines more suitable. 
(c) Ultrafiltration 
Ultrafiltration is widely utilized in the recovery and purification of protein 
products. The main uses of ultrafiltration are as follows: concentrating protein 
products; desalting product solutions by diafiltration; exchanging product 
buffer solutions by diafiltration; and depyrogenating of buffer solutions used 
in the process. Ultraflltration is also finding increasingly wider use in the cell 
harvesting operation. It has an advantage over centrifugation in this situation 
since it subjects the protein to less heat and shear effects. Ultraflltration is 
excellent for processes using cell recycle and in particular for mammalian cell 
applications. 
id) Electrodialysis 
Electrodialysis is sometimes used to remove salts, acids and bases from 
fermentation broths. A unit will consist of compartments separated by alternate 
anion and cation exchange membranes. A direct electric current is then passed 
through the stack to effect the separation.
(e) Chromatography 
Chromatography is the main technique for final purification of the product 
protein. Chromatographic separations take various forms depending on the 
driving force for the separation. There are essentially two basic forms of 
chromatography; partition chromatography (such as gel filtration) and absorption 
chromatography (for example, ion exchange and affinity chromatography). 
Gel filtration, also called molecular sieving, separates molecules based on 
size. It is sometimes used to desalt protein solutions. In this method the product 
and impurities travel at different speeds through the bed effecting the separation. 
Gel filtration is essentially a low capacity technique and not suited for high 
volume processes. 
Absorption chromatography is where the product binds to the matrix in the 
bed and is subsequently eluted by a change in the buffer composition. Common 
forms of separations include ion-exchange chromatography (which separates 
proteins based an electrostatic charge) and affinity chromatography (which 
separates a product or removes an impurity by means of a biospecific attraction 
between the molecule and a liagand attached to the gel or resin). 
In order to achieve the required purity it is necessary to run the chromatographic 
units in series to reach the purity needed. 
Automated programmed chromatography controllers are recommended for 
the reproducibility of their operation and reduced labour requirements. Once 
initiated, the programmed chromatography controller automatically loads the 
product on to the column, washes and elutes the product. 
Scale up of chromatographic systems is reasonably straightforward and 
follows well-documented guidelines. 
Solutions for purification operations 
Solutions required during purification are generally prepared in solution 
preparation areas. Smaller volumes can be filled into portable containers or 
mobile vessels whereas larger volumes are generally piped to the user point. An 
important aspect of buffer preparation is to identify where Water for Injection 
(WFI) is required. In cell culture systems, where endotoxins are not produced by 
the culture, WFI is generally recommended for all buffer solutions so as to 
prevent the introduction of endotoxins, which would then need to be removed in 
a later chromatographic step. In microbial systems where endotoxins are 
produced (such as E. CoIi lipopolysaccharides), WFI may not be needed until 
a later stage where the pyrogens are reduced to low levels or effectively 
eliminated. For very large volumes, storage of diluted buffer solutions is 
impractical. One approach is to make up concentrated solutions and dilute as 
required — this approach can result in significant space and cost savings.
5.2 Key unit operations 
5.2.1 Liquids materials handling 
Materials to be added to a reaction system can come in liquid, solid or gas form. 
However, the easiest to handle are liquids and consequently materials are used 
in the liquid form where possible. If not the natural state at ambient conditions 
then the material can be made liquid either by melting or more commonly by 
making a solution by dissolving in a solvent. 
Liquids can fall into three categories when used in a reaction: 
solvent: this allows the reactant to mix and react and to create a mobile 
mixture that can be controlled for temperature by heat transfer with surface 
contact. Solvent liquids generally form large quantities in a batch make up; 
reactant: the active compound used to react with another material to 
synthesize the desired intermediate or final molecule stage. Use of liquid 
reactants is generally desirable as they can easily be transferred and added to 
the reactor system under controlled conditions; 
catalysts: these are usually required in small amounts. Handling small 
quantities can bring difficulties; it is easy to dispense the correct quantity 
in a laboratory or fume cupboard, but getting it safely into a reactor system 
needs to be carried out via an air lock or charge flask arrangement. 
Liquids are usually handled in drums if the quantities are small or the 
duration of production is short — this typically applies to reactants, particularly 
where there is no source for bulk deliveries. If the material is used in larger 
quantities then bulk delivery in road tankers and storage in a bulk tank system is 
preferred as it minimizes the manual handling requirements, and hence, 
reduces the operator inputs. 
Liquids delivered in bulk quantities from road tankers must be shown to be 
suitable for use in the process — that is they are of the correct purity, strength 
or even the correct chemical composition. This may be by reliance on the 
supplier's audited quality control/assurance system and certificates of conformity, 
or by sampling the road tanker and then analyzing the contents before 
offloading. Alternatively where analysis is lengthy and would incur waiting 
time charges from the delivery company, special quarantine bulk storage tanks 
can be used which allow a segregated offload of the material and then the 
appropriate testing prior to release for use or reject and return. 
With the increasing legislation on Volatile Organic Compound (VOC) 
emissions, it is common to vent the bulk storage tank back to the road tanker 
during offload to avoid release of VOC.
Liquids are charged to the process either by direct pumping from the bulk 
tank or drum into the reaction system or to an intermediate addition vessel such 
as a head tank which allows more accurate determination of quantity and 
greater control over rate of addition. Alternatives to pumping include closed 
vacuum or pressure charging, although these methods are not commonly used 
now because of the safety issues associated with them. 
5.2.2 Solids materials handling 
Solids are most commonly used in processes as reactants but can also be used 
as catalysts, purification agents such as activated charcoal, or seed for crystallization 
process stages. 
One of the main sources of solid is as an intermediate stage in a lengthy 
multi-stage synthesis production operation. 
Solid material is most commonly stored in sacks, plastic drums or lined 
fibreboard kegs. The most important consideration during use is the safe, 
contained dispensing of the required quantity and the charging of this into the 
reactor system. 
Open manway charging used to be the main transfer method but this is now 
considered unacceptable because of the risks of exposing the operators to the 
chemicals inside the reactor. Similarly the risk of exposing the process to crosscontamination 
from surrounding activities is also unacceptable in many 
circumstances. 
Current methods involve creating a protected area for charging, either 
directly to the reactor via a weigh hopper or charge lock, or to an intermediate 
bulk container (IBC). This IBC can then be moved to a docking station to allow 
enclosed charging to the reactor system. The protected area involves controlled 
clean air flows to minimize risk to product and operator by reducing contamination 
and exposure within a purpose designed charge booth. 
The use of split butterfly valves or contained transfer coupling systems is 
now a very popular way of making the connection between the IBC and the 
process system, as it allows the handling of very active materials with increased 
safety and ensures minimal contamination of the reaction process. 
The use of solids in bulk is not very common unless for large tonnage 

products where a dedicated plant with silo storage and transfer techniques such 
as pneumatic transfer, screw feeders or conveyors can be used. 
5.2.3 Liquid/liquid separation techniques 
As part of either the reaction stage or purification stages of the synthesis, it is 
often necessary to separate one liquid from another. There are two main types
of technique available for this, those involving heat and those using other 
properties of the liquids to achieve the separation. 
Thermal processes 
Thermal processes are commonly used for removing materials, such as an 
inhibiting by-product formed during a reaction, typically water, or operations 
where evaporation techniques give an effective and efficient method of 
separation. These can be either single stage such as a flash distillation or 
involve the use of fractional distillation by utilizing distillation packing 
materials in a column. 
Batch distillation is not an easy process to perform due to the unsteady 
composition of the still vessel and the fall in efficiency as volumes drop, and 
therefore, so does contact with the heat transfer surface. A supplementary heat 
transfer surface can be provided by pumped or thermo-syphon circulation 
through a heat exchanger. 
Another problem with thermal processes is that they can result in the 
degradation of product if it is sensitive to heat. To minimize this, the pressure at 
which the distillation is carried out can be reduced by vacuum pump systems to 
allow evaporation at lower temperatures. In the event of particularly sensitive or 
labile materials this can be carried out in small continuous units operating at 
extremely low vacuums known as short path stills. 
An alternative extractive technique is azeotropic distillation. Here an 
additional material is added to create an azeotrope, which will preferentially 
be distilled out achieving an otherwise impossible thermal separation. 
The entrainer is then separated from the removed material and recycled if 
possible. 
Non-thermal processes 
It is a relatively common process to add a liquid to the process into which 
impurities or even the product is preferentially soluble. The added liquid is 
immiscible with the process stream and forms a separate phase, which can then 
be separated by various techniques. This process is commonly carried out with 
water or aqueous solutions and is known as washing. 
The immiscible phases can be separated by allowing the layers to settle in 
the reactor vessel and then running the lower layer out until the interface is seen. 
It is common to run this layer to a receiver; it could be the product layer or if it is 
the waste layer it could be held prior to discharge. 
Interface detection can be difficult. Automatic detection devices have mixed 
success and generally an illuminated tubular sight glass and trained operator is 
the most successful technique.
In large production plants, mechanical techniques such as decanter centrifuge, 
multi-plate disk centrifuge or counter flow liquid-liquid extraction 
devices can be used to increase the efficiency of the separation. 
Techniques that were previously used mainly in the biotechnology field are 
now becoming more available to achieve difficult separations and purifications 
in the synthetic process arena. These include chromatography techniques and 
selective membrane processes, which are becoming more feasible with the 
developments in membrane technology. 
5.2.4 Crystallization 
Most synthetic processes involve the isolation of a solid stage. This can be an 
intermediate stage, a byproduct or most commonly the final active BPC. The 
formation of the solid form can be carried out in several ways: 
crystallization by cooling; 
crystallization by evaporation/concentration and cooling; 
precipitation by reaction or pH change; 
precipitation or crystallization by solvent change. 
This operation can be carried out in the standard or slightly modified batch 
reactor described earlier. The allocation of a specific or dedicated reactor for 
crystallization use is becoming more common and provides a way of avoiding 
contamination of the final product. The need to provide controllable agitation 
with gentle profiles to avoid crystal damage and good heat transfer are the main 
areas addressed along with the rate of addition of precipitant or cooling profiles 
to allow for optimal crystal form and size. In order to promote the desired 
crystal form, seed materials of the desired crystal type can be added at the 
correct stage to initiate crystallization of the appropriate form. 
The crystallization activity is becoming increasingly sophisticated. Known 
as crystal engineering, it is of growing importance especially in tailoring the 
product form of the final BPC to suit the demands of the secondary operations, 
avoiding comminution or granulation to achieve desired product form. 
Most crystallizations are carried out on a batch basis. However, if production 
quantities demand or specific product form/size distribution profiles are 
required then continuous crystallization arrangements can be used. New 
developments involving the use of ultrasound to form a nucleus for crystallization 
(known as Sonocrystallization) have been developed. They can 
produce mono-size distributed slurries accurately engineered for the desired 
property and are of particular interest for sterile production where seed 
introduction is more difficult.
5.2.5 Solids isolation 
Once the solid form has been produced, it needs to be isolated from the liquid or 
mother liquor. 
Separation of solid from liquid generally involves some form of nitration 
since techniques such as sedimentation are not routinely applied in the 
pharmaceutical industry. Filtration involves creating a medium through 
which the liquid can pass but the solid is retained. Once the medium has 
been formed, a driving force to cause the liquid to flow is needed; the way in 
which the driving force is generated is the main area where differences in 
technique or equipment occur and can be created by vacuum, gas pressure, 
mechanical pressure or centrifugal force. 
The other main area which differentiates the filter type is the quantity of 
solid involved and whether it is a by-product to be removed or a product. 
Filters 
Solid impurities in small quantities up to 10 kg can be removed using cartridge, 
bag or multi-plate filters such as the calmic filter. 
The single sheet, nutsche filter is a common unit that has developed greatly. 
The original form was an open box filter that used vacuum in a lower section of 
the box to draw filtrate through a filter medium or cloth. The disadvantage with 
this type is that they offer little to protect the general plant area, contain the 
process to protect the operator or prevent cross-contamination. The other main 
disadvantage is the level of vacuum that can be generated limits the driving force. 
The first development of the nutsche filter was the agitated pressure nutsche 
filter. This unit has an integral pressure chamber above a filter media, typically a 
cloth element. The unit is fitted with an agitation arm that can be used to smooth 
the cake and discharge the damp solid. The driving force for separation is 
generated by either applying vacuum to the filtrate receiver and sucking the 
filtrate out of the slurry to leave a damp cake, or by applying pressure above the 
slurry and forcing the filtrate out to leave a cake. 
Occasionally both pressure and vacuum are used to generate the driving 
force, but it is commonly found that increasing the driving force above 3 bar has 
little benefit on filtration rate due to compression of the cake and the closing off 
of the route by which filtrate can flow out. The pressure is most commonly 
generated by nitrogen and because the materials are typically flammable 
solvents, nitrogen also provides an inert atmosphere. It can be provided 
either once-through from a mains supply leaving via the filtrate receiver or 
recycled taking low pressure nitrogen from the receiver, increasing the 
pressure, then putting it above the cake to displace more filtrate. This has the 
advantage of minimizing the amount of nitrogen used and reducing emissions
to the atmosphere as the nitrogen entering the receiver is laden with solvent 
vapour. The recirculated nitrogen can also be heated prior to entering the filter 
to aid drying of the cake. The nitrogen is then taken directly from below the 
cloth to the compressor package where it is chilled to remove the solvent, then 
repressurized and heated before recirculating back above the cake. 
The cake can be washed in the filter to remove soluble impurities. This is 
done in two ways, either a displacement wash or a reslurry wash. In the 
displacement wash the wash fluid is sprayed onto the cake surface whilst 
vacuum or pressure is applied to cause the wash fluid to quickly pass down 
through the cake, taking out the impurities and out to a receiver. This is 
commonly used where the impurities are very soluble in the wash and can be 
easily removed or where the product cake itself is soluble in the wash so that 
residence time is minimized to avoid losing product with the wash. With a 
reslurry wash, a volume of wash fluid is added to the filter and the agitator is 
used to mix the cake with the wash fluid to form a slurry. By this process the 
impurities can then dissolve into the wash fluid. The resultant slurry is then 
filtered again to remove the wash fluid and the impurities. The wash filtrate is 
often collected in a separate receiver to allow for recovery of product that may 
have been dissolved and lost as well. This is known as second crop recovery. 
Discharge from the filter can be in one of three ways. Most commonly the 
product is discharged as a damp cake; here the agitator is lowered to the cake 
surface and rotated to start to break up the cake. By altering the direction of 
rotation, the cake can be drawn to the outside edge of the filter where an outlet 
hatch is opened to allow discharge of the cake out of the filter to the next 
process unit. As discharge proceeds the agitator is lowered gradually to the 
bottom of the filter to ensure all the cake is discharged. The nature of this 
operation results in slugs of damp cake being discharged as the arm goes past 
the discharge hatch, which may cause problems for the next processing module. 
An alternative approach is to have a central opening in the middle of the filter 
element and dig the cake and bring it to the middle. This provides a continuous 
flow of solid out but reduces the area for filtration and can give problems of 
sealing the central outlet. The other methods of discharge involve either making 
a slurry or solution of the cake in a solvent and charge as per the wash fluid. 
This is then agitated and discharged via a valve and pipe arrangement from the 
side of the filter above the filter cloth. 
The nutsche pressure filter has also been developed into a filter dryer. Here 
heat can be applied to the cake once filtration has occurred via coils on the side 
and top of the filter body and via heating passages through the agitator. A single 
fluid heating medium, often hot water, is circulated through these coils and this 
provides heat to the product to remove the remaining solvent to give a dry solid.
At the same time as the heat is applied, the space above the cake is subjected to 
a vacuum pulled on the system normally via an integral dust filter to avoid any 
losses of product solid with the evaporated filtrate. The filter dryer has proved a 
very successful item of plant and minimizes the exposure of the product during 
its transfer from the filter to another dryer. The disadvantage of the unit is that 
the time taken to filter, wash and dry a batch in the filter dryer is overall rate 
limiting for batch time cycles. 
Other types of filters exist which provide different methods of presenting a 
filtration element and a driving force of pressure to separate solids and liquids 
and then discharge the solid. These include rotary vacuum filters, tube filters, 
disc filters and belt filters, but they are not common in the pharmaceutical 
industry and are used for specialized applications only. 
Centrifuges 
These devices generate a centrifugal force to drive the liquid through the 
separating medium leaving the solid. There are four main types: 
vertical axis — top discharge by basket lift out: This is the traditional 
type and is not commonly used now except in small sizes. The main problem 
is the exposure of the operator when emptying the basket and the risk to the 
product of cross-contamination in the open process; 
vertical axis — bottom plough discharge: This allows contained discharge 
of the solid from the basket by a movable knife or plough that cuts the solid 
out of the basket and down a chute at the bottom of the machine; 
horizontal axis — peeler discharge: This unit has advantages over the 
vertical axis machine in that it can spin at higher speeds, and hence, create a 
higher G-force or driving force for separating the liquid. Discharge of the 
solid is carried out in a similar way by a knife or peeler blade, which is used 
to channel the solid into a chute and away from the machine; 
horizontal axis — inverting bag discharge: This is the most current 
development. It has the benefits of the higher G-force for separation but 
the cake is removed by inverting the filter cloth. It also has the benefit of 
being able to remove the entire heel to ensure ease of further separations and 
minimize batch-to-batch contamination. Most modern centrifuges are automatically 
controlled. This covers inerting and purging cycles, filling, spinning, 
washing and discharge. 
5.2.6 Drying 
The final step for most BPC processes is to dry the intermediate or final 
product. This removes any residual solvent from the solid. Often this is done to
produce a fine free-flowing powder that can easily be handled in the secondary 
processing. Alternatively if the solid is an intermediate then subsequent 
processing often involves the use of a different solvent. Drying reduces the 
moisture level of solvent to an acceptable level, usually to below 1% w/w of the 
solvent present. 
Dryers can be classified into two main types — direct and indirect. With a 
direct dryer, air or more commonly nitrogen is heated and passed through the 
solid. An example of this type of dryer is the batch Fluid Bed Dryer (FBD). 
This unit uses a basket that would be filled either by hand or by gravity from the 
filtration or centrifugation unit. The basket has a perforated base and when 
placed in the fluid bed dryer, the heated air or nitrogen flows up through the 
solid, fluidizing it and evaporating the solvent. The off-gas stream is filtered, 
usually by a cyclone or a bag filter system to prevent loss of product. The 
filtered stream can be cooled to remove the evaporated solvent, then reheated 
and passed back through the basket. Whilst the units are relatively cheap, they 
are not favoured for the following reasons: 
VOC losses are high without the high additional cost of a nitrogen gas 
recycle system; 
there is a high risk of static discharge; 
effective filtration of the heated air stream is required to avoid introducing 
contamination; 
open handling of the cake does not provide a contained system, particularly 
for very active products. 
For these reasons, indirect or enclosed dryers have replaced the direct dryer. 
Many pharmaceutical products tend to be thermally sensitive and as a result 
most are dried under vacuum, since this allows for solvent evaporation at lower 
temperatures. Jacket temperatures of typically 40-1000C are used with hot 
water or a single fluid system as the heat source. A dust filter is installed on the 
dryer body or in the vapour line to prevent loss of product with the vapour 
stream. A vacuum is generated by liquid ring pumps, once-through oil 
lubricated pumps, dry running vacuum pumps or more rarely ejectors. Solvent 
condensing is carried out either before or after the vacuum pump depending on 
the capability of the pump to handle liquids and condensation of the solvent. 
Often this is not desirable for corrosion reasons and all the condensation is 
carried out after the pump. The ideal solution is to use a liquid ring pump with 
the same or compatible solvent, chilled, as the ring fluid, then condensation can 
occur directly into the ring fluid. 
The fundamental principle of the indirect dryer is to provide a heated surface 
and a means to ensure good heat transfer from that surface to the solid, whilst
maintaining a vacuum above the solid to efficiently vaporize the solvent. 
Various designs for achieving this exist and can be categorized by the means 
used to achieve the heat transfer, as follows: 
(a) No agitation 
The vacuum tray dryer is the only example still in routine use under this 
category. Here, solid is laid in thin layers onto trays and placed onto heated 
shelves in a vacuum chamber where heat and vacuum are applied to evaporate 
the solvent. The dryer is not very efficient as it takes a long time to dry the 
product due to the lack of agitation, and hard dried lumps can form because 
there is no agitation to break down agglomeration during drying. The biggest 
failing with the dryer is that it is messy to load and unload the trays, requiring a 
high degree of containment and equipment to protect both the operator and the 
product. It is, however, very popular in R&D environments where its flexibility 
is a benefit, and in instances where mechanical work on the product will 
damage crystal size or shape or cause safety problems such as detonation of a 
shock sensitive solid. 
(b) Horizontal axis agitated vacuum dryers 
This type of dryer, the 'paddle dryer', is most widely used in BPC manufacture. 
It consists of a horizontal cylindrical chamber, the outside of which is fitted 
with heating and cooling jacket or coils. Inside, the dryer is fitted with a slow 
rotating paddle that moves the solid to give good mixing and allows replacement 
of the solid in contact with the heating surface, aiding drying. Horizontal 
axis dryers have high jacket surface area to volume ratios and are efficient 
dryers giving low drying times. Vapour is withdrawn via a dust filter fitted to the 
top of the body, allowing collected powder to be routinely shaken or blown 
back into the batch. They also have low headroom and can be fitted into process 
buildings without adding a full floor whilst utilizing gravity in the isolation 
train. They can be difficult to clean particularly because both shaft seals are 
immersed in the solids. Some designs allow for easy and complete removal of 
the end plate and agitator shaft. 
(c) Vertical axis vacuum dryers 
There are a number of variations of vertical axis, agitated vacuum dryers; the 
main difference between them being the ratio between diameter and depth of 
dryer. Short large-diameter dryers, often referred to as pan dryers, are popular. 
A variant of this utilizes a specially designed agitator that provides a very 
efficient mixing regime giving good heat transfer and efficient drying. This type 
of dryer has been termed a turbo dryer. Some designs allow the lid to be
hydraulically lifted for internal inspection and cleaning. High-speed impellers 
known as lump breakers can be fitted in addition to the main stirrer to break up 
any agglomeration. The drive can be either top or bottom mounted. The bottom 
drive has the disadvantage of requiring a seal in the product contact area, whilst 
the top mounted drive takes up a lot of space on the dryer lid, reducing the 
opportunity for additional nozzles and restricting the opening of the lid. The top 
mounted drive allows for the agitator to be raised and lowered through the solid, 
adding to the range of agitation profiles for drying. A variant of the vertical axis 
vacuum dryer is the filter dryer, referred to in the previous section, which 
combines the functions of a pressure nutsche filter with a vacuum pan dryer. 
The compromise tends to be due to the retention of the filter cloth during the 
drying process and the design of the agitator. 
When the depth of the dryer exceeds the diameter, the dryer is referred to as 
a cone dryer. Deep cone dryers have a double rotating screw inside, which 
performs three functions: wall to centre solids movement for heat transfer by 
horizontal and vertical turning; delumping of solid initially and during drying; 
assisting bottom valve discharge by reversing the screw direction. 
This design is favoured by a number of companies since it offers reasonably 
efficient heat transfer, delumping, relative ease of cleaning by refluxing with 
solvent and caters for variable batch sizes. Top and bottom drive mechanisms 
are available. From a GMP viewpoint, internal drive mechanisms must not shed 
particles. The one disadvantage of these dryers is that they are relatively tall 
compared to the other types and can add a floor to the isolation area, although 
protruding the discharge cone region into the clean pack-off room can 
compensate this. 
5.2.7 Product finishing 
Historically, BPC products were simply packed off from the dryer into fibreboard 
kegs and shipped, via a QC sample and check stage, direct to the 
secondary plant. Here finishing operations such as mixing, comminution or 
milling and granulation were generally carried out. 
However, with the change in the profile of the BPC manufacturer, the end 
user for the BPC is often a different business or group within the same 
pharmaceutical manufacturer, or the BPC manufacturer is a different company 
to the pharmaceutical secondary company. In these instances there is an 
increasing need to provide some of the finishing operations to produce a 
product with specific physical characteristics in addition to the correct chemical 
composition. The increasing demands of 'speed to market' have also caused a 
blurring of the activities traditionally seen as 'secondary operations' and have 
increasingly come to be expected as part of the BPC manufacture.
Milling, sieving and granulation 
Milling is an operation to reduce the particle size of a solid down to an 
acceptable profile or range of sizes typically below a certain maximum size. It is 
best carried out in-line after the dryer to avoid double handling, particularly 
since dryer discharge is often a low rate, semi-controlled process. If carried out 
off-line after quality approval, then a separate milling line in a clean room suite 
is needed. Intermediate bulk containers (IBCs) are usually used for solid 
transfers and act as feed hoppers to the mill feed system. 
There are various types of mill used in the BPC industry, including pin mills, 
hammer mills and more commonly jet mills and micronizing mills. Further 
details are given in Chapter 6 covering secondary processing. 
Sieving is an operation to classify the solid into a range of particle sizes. The 
equipment is often used in-line with the discharge from the dryer. The sieve 
operation consists of passing the solid through a series of screens. The first screen 
removes particles that are larger than the specification; these are discharged and 
recycled to the mill. The second screen then retains particles of the minimum size 
and above. The solids passing through the screen 'fines' is too small and may be 
recycled to the crystallization stage. The material is encouraged to pass through 
the screens by either vibration or by the use of rotating arms. The material that 
does not pass through the screen is removed from the sieve in either a batch or a 
continuous method to be packaged. Oversize and fine material can be reworked 
in some cases, but sometimes has to be destroyed. 
There are some cases where more than two screens are used. This provides a 
series of size fractions that can be used for products that require specific drug 
related release profiles or for filling directly into hard shell gelatin capsules. 
5.2.8 Packaging 
The final packaging of a BPC is carried out in a controlled environment to 
protect the product from contamination by external sources and also to protect 
the operator from exposure to the active material. Most BPCs are solid powders 
and are packaged in sacks, drums or IBCs. A small number of products are 
liquids and these are packaged into the appropriate containers in either a 
manual or automated filling system. 
5.2.9 Solvent recovery 
Solvents are widely used in the production of BPCs and, as previously stated, 
provide several functions including dilution of the reactant concentration and 
mobility to allow good mass and heat transfer. Solvents are important 
in obtaining the correct final product form and in washing the product in 
isolation equipment. When used in a reaction, the solvent generally does not
react or break down to other components. In order to maximize the efficiency of 
the process, solvent remaining after a processing stage can be recovered for use 
in the same process from which it originated. 
Solvent recovery can be either a batch operation or, more commonly if larger 
volumes are involved, a continuous recovery plant. 
The type of recovery used largely depends on the contamination present and 
the properties of the solvent being recovered. Flash stripping is the simplest 
operation and is often sufficient. Fractionation, often by the use of random or 
structured packing, is used where complex mixtures require separating. 
Pre-treatment is often used to allow a simpler recovery. This can involve 
crude solids filtration to more complex precipitations or pre-stripping. 
Most solvent recoveries result in a residue, which will then require further 
treatment or handling — most commonly incineration or landfill. 
5.3 Production methods and considerations 
5.3.1 Production 
Pharmaceutical production is mainly carried out on a batch basis for a number 
of reasons. The main reasons are normally linked to the traceability of the 
product, validation and regulatory issues, but others include the scale of 
operation, the flexibility of operation required, inventory optimization or 
even technology development. 
Production is arranged into three main types of facility: 
dedicated — the facility is designed and built for one specific process; 
multi-purpose — the facility is designed and built to carry out a number of 
known and defined processes, potentially with a minor amount of modification 
to configure the plant to the next process; 
General purpose — the facility is designed to handle a variety of processes, 
both known and envisaged for the future. 
Batch chemical processes with cycle times typically of 16 hours or more 
are most commonly carried out on a 24-hour a day, seven days a week 
operation. 
5.3.2 Automation and control issues 
Any automation system must provide tangible benefits to justify the investment. 
In general, the benefits of automation will derive from: 
higher levels of safety; 
the ability to apply sophisticated control strategies;
consistent product quality; 
higher levels of plant utilization for a given manning level; 
more efficient usage of materials and reduction in waste; 
provision of timely and relevant information. 
The logic and numerical processing capabilities of modern process control 
systems enables operating conditions to be tightly regulated to the specified 
profiles, optimizing processing time, delivering consistent quality of product 
and providing a higher level of safety. 
While the use of properly designed and implemented process automation 
systems enhances the safety of the plant (by improved control and reporting/ 
notification of potential risks) these systems should not be relied upon to ensure 
plant safety. The recently published international standard IEC 61508: Functional 
Safety of Electrical/Electronic/Programmable Electronic Safety Related 
Systems addresses the requirements of safety related systems. 
The key issues to be considered when embarking on automation projects 
include: 
the functionality required; 
the level of automation required; 
the types of systems employed. 
Most primary pharmaceutical manufacturing processes can be classified as 
being either 'continuous' or 'batch' with a few, if any, being categorized as 
'discrete' processes. This section focuses on the requirements of batch type 
operations. 
The requirements of batch operations can generally be considered more 
onerous than those for other types of processing. Batch processing involves the 
sequential modification of process conditions through a predefined regime 
rather than maintenance of established 'steady-state' conditions. 
Batch operations essentially consist of a series of phases that are executed 
sequentially. The execution of a phase is usually dependent on process 
conditions established in a preceding phase; therefore any fault that interrupts 
the execution of a phase may require the processing to be resumed from a point 
in the operation sequence other than that where it was suspended. The process 
automation system must be capable of executing sophisticated exception 
handling procedures. It may require the provision of facilities that enable the 
operator to intervene and manually adjust the point in the sequence at which 
processing is to resume.
System functionality 
The functionality required of the system will principally depend on the 
processing objectives and the method of operation proposed. The plant 
equipment and its connectivity also affects the functionality; the following 
are some possibilities: 
single batch, single stream (one batch at any given time); 
multi-batch, single stream (more than one batch being processed at any given 
time); 
multi-batch, multi-stream, dedicated equipment trains; 
multi-batch, multi-stream, common equipment. 
On plants where a variety of products are regularly manufactured, some form 
of automatic scheduling functionality may be desirable. When equipment is 
required to undergo Clean In Place (CIP) or Sterilize In Place (SIP) routines at 
regular intervals or at product changeover, the CIP/SIP operations may be 
considered as a 'product recipe' and scheduled accordingly. 
The sophistication of the scheduling systems available vary from the basic, 
where queued operations (or batch recipes) are initiated when the necessary 
processing units become available (or predefined constraints are satisfied), to 
others which are capable of developing a production schedule from demand data 
transferred from ERP (Enterprise Resource Planning) or MRPII (Manufacturing 
Resource Planning) systems. The sophisticated systems are capable of 
queuing recipes, calculating the optimum batch sizes to complete a campaign, 
and making changes dynamically as 'demand' changes. (Some form of 
'gateway' to control the transfer of data from ERP systems is recommended 
to prevent disruption of manufacturing operations by sudden changes in 
demand). Other factors that complicate scheduling include the following: 
resources that can be simultaneously allocated to more than one process 
(e.g., cooling fluid circuits, ring-main fed utilities); 
number of streams in the system; 
selection of the best resource to use when several (shared) non-identical units 
are available (requires knowledge of what will happen next); 
operations that are dependent on activities/equipment controlled by external 
systems (which may result in the duration of the operation being unquantifiable). 
The recipe handling requirements of the process control system are affected 
by the type and configuration of the plant. The recipe system may also need to 
be able to cater for variations in the properties of raw materials, which may 
result in a requirement to modify the processing parameters. Any variation in
the processing parameters/formulation, whether for a campaign of batches or 
for an individual lot, needs to be recorded and the appropriate mechanisms and 
facilities need to be provided to enable this. 
As well as the quantity and complexity of the recipes that need to be 
executed, the number of recipes that can be simultaneously active in the system 
(on the plant) needs to be considered. In 'multi-batch' situations, the process 
control system needs to be able to report the impact of a malfunction or process 
deviation on other concurrent activities. 
The exception is handling facilities that are critical to the successful 
operation of a batch plant. In the event of a deviation from the expected pattern 
of occurrences, the operator should be informed and appropriate action should 
be taken promptly. A minimum of three categories of operator message are 
recommended: 
critical alarms generated when there is risk to equipment or personnel; 
process alarms caused by deviations from the expected conditions; 
events which keep the operator aware of actions being performed. 
In the case of critical and process alarms, the process control system will 
normally be expected to take action to put the plant in a safe condition 
automatically. Facilities are also needed to enable the system to restore the 
plant to its prior state as effectively as possible. A good understanding of both 
the process and the control system are required in order to develop the 
necessary procedures and phases. 
The production data, exception reports and alarm information generated 
need to be associated with the appropriate batches and stored to satisfy 
operational as well as regulatory reporting requirements. As in the case of 
the process control software, the definition of the reports requires knowledge of 
operational procedures and company standards. 
The recording and storage of data should be clearly differentiated from the 
reporting function. Justification should be provided for all data that is to be 
recorded because, while it is true that data not recorded is lost forever, recording 
excessive quantities of data can have severe drawbacks. Some systems enable 
data recording to be triggered by events; this enables data collection to be 
restricted to critical phases of an operation (such as during an exothermic 
reaction). 
It is important that the recorded data is stored in a format that allows it to be 
manipulated in the manner required. While the control systems use a variety of 
data compression algorithms to facilitate the storage of large quantities of data, 
this can prevent data export and restrict the processing and manipulation of the 
information to the control system with the consequent limitations.
Interfaces and communication facilities with other systems also need to be 
evaluated when identifying the functionality required of the system and this is 
addressed in a later section. 
Automation levels 
In a processing environment automation should be aimed at removing 
the mundane and repetitive tasks from the operators, freeing them to add 
further value. The numerical processing capabilities of modern control 
systems enable advanced control strategies to be employed to improve 
efficiencies. 
All areas of the plant will not require the same level of automation. There is 
also a trade-off between the manning level reductions available through 
automation and the flexibility available from lower levels of automation. In 
certain areas, such as raw material tank farms, a 'basic' level of automation can 
result in a far more effective system, while other areas benefit from all the 
sophistication available. In the main processing area, manual intervention may 
be restricted to critical operations where heuristic judgment is required or those 
aspects where the necessary facilities to allow automatic execution have not 
been provided. 
As part of the development of the control philosophy, each area of the 
plant should be reviewed and the required automation level established. The 
basis of the justification for automation will vary and could include conditions 
within an operating environment such as physical aspects of the nature 
of the task to be undertaken, the need for an automated record of activities 
performed, etc. 
5.4 Principles for layout of bulk production 
facilities 
Many examples of unplanned developments can be seen on pharmaceutical 
sites throughout the world. Production facilities have grown in many cases in a 
totally uncontrolled manner with decisions made based on the priority of the 
moment with no regard for the future. This has happened due to lack of 
thought, concern for cost and lack of information on the company's future 
marketing plans. The result is a totally random 'notch potch' of buildings 
leading to inefficient operation, potential hazards, questionable use of land, and 
expensive future development of the site. 
Two types of development will now be considered. Green field development 
involves the use of land on which there has been no previous commercial 
developments. Plans for such sites will not generally be restricted by previous
buildings and existing operations. Brown field development may, however, 
have some restrictions due to past or existing operations and freedom of design 
may be curtailed. 
In both instances however, at some stage of design, it is necessary to review 
the impact of the new development on the future use of the site. All these 
principles equally apply to secondary production facilities. 
5.4.1 General considerations 
In the pharmaceutical industry, sites may be laid out for primary production, 
secondary production, research and development, warehousing and distribution 
or administration and head office activities. A single site could cover any 
number of these functions. There is considerable dialogue on the advantages 
and disadvantages of multiple use sites, which will not be discussed in this 
guide, except to point out that all the above activities do not necessarily sit well 
together. Here the guide is aimed at bulk drug primary production site layouts 
only. 
5.4.2 Green field sites 
Site location 
It is assumed in this guide that the new site will consist of multiple production 
units; the first of which is to be built at the time of developing the site 
infrastructure, with others following on at some later time. 
When selecting the site, due consideration will have been given to its 
geographical location with specific attention to road systems, communications, 
ports and airports, availability of skilled labour and adjacent developments. 
Any special environmental requirements and full information on the availability 
and capacity of public utilities will also have been investigated. 
Discussions with all appropriate planning and statutory bodies will have 
been carried out to determine if there are any requirements that would prevent 
the development of the optimum design for the site. It is also necessary to 
ensure that any adjacent developments in the planning stage are compatible 
with a bulk drug operation. For example, an open cast mining site adjacent to a 
plant manufacturing high cost pharmaceuticals would not be ideal. 
It will be necessary to carry out full topographical and geotechnical surveys 
to determine the surface contours and the load bearing characteristics of the 
land. These surveys will provide information on underground obstructions, 
mine-workings and geological faults. Such information could influence the 
positioning of buildings or indicate the need to carry out specific rectification 
work. The land should also be checked for ground contamination. Information
on the ambient climate of the site, including prevailing wind directions, is also 
necessary at this stage. The majority of the above data should be obtained prior 
to the purchase of the land. The above requirements are not exhaustive but 
do indicate typical actions which are required prior to finalizing on a particular 
site. 
Conceptual design 
The project may be divided into two parts. The first part covers site infrastructure, 
including: 
offices and administration buildings; 
operator and staff amenities; 
control and test laboratories (if not in the production plant); 
engineering workshops and stores; 
central warehousing; 
on-site utility generation; 
gate house and security fencing; 
utilities and services distribution; 
roads, road lighting and car parks; 
underground utilities; 
site grading and landscaping. 
The second part will cover production facilities. This, as mentioned 
previously, may be the only production unit or may be the first of a number. 
In this guide it is assumed that the site is to be laid out to accommodate 
a phased development and the design must ensure that future construction 
will not cause interruptions to production. This second part typically will 
include: 
the main reactor and process facility; 
special hazard production units; 
environmentally controlled finishing units; 
bulk raw material tank farm and drum store; 
effluent treatment final conditioning unit; 
control room for the production processes; 
production offices; 
on plot generated services; 
switch rooms and transformers. 
The split of the project into two parts can be advantageous commercially. 
The infrastructure is mainly civil and building engineering and the production
unit is mainly process engineering. More suitable contracts can be negotiated if 
this difference is understood. 
Based on these various elements, it would be normal to look at a number of 
possible layouts to finalize the overall concept before proceeding with detailed 
design. 
Generic production plot layout 
Before proceeding with the layout of the site, it is advantageous to give some 
consideration to possible plot layouts. It is anticipated that the production units, 
which will eventually be constructed on the site, will produce a number of 
products that may benefit from a custom design approach. If the plants are to be 
of a multi-product design then consideration should be given to the maximum 
numbers of reactors to be included in one plant. 
Regardless of the style of production unit, the fully developed site is likely to 
have a number of production buildings each with associated control rooms, onsite 
utility generation, offices and tank farm etc. 
Based on the first production unit to be developed, it is advantageous, before 
considering overall site layout, to develop an outline plot layout that can be the 
basis for all plants on the site. This does not mean that all plots will be identical 
but the main principles will have been identified at this early stage and will have 
some influence on the ongoing development of the site. Typically control room 
positioning, spacing of on-site tank farms, policy for facilities for hazardous 
operations, position of on-site switch room and electrical transformers should 
be identified. 
Whilst the brief for the first production unit may be well defined, subsequent 
developments may be unknown at this stage. It is essential to recognize this and 
to incorporate flexibility into the eventual site layout and to identify which 
production plot parameters could possibly change. Site master plans should not 
be written in tablets of stone but should be reviewed with each new development. 
They should not, however, be changed by default. 
Site layout - master plan - zoning 
The term 'Site Master Plan' has been introduced in the previous paragraph. In 
green field development this is likely to start with an area of land that has no 
structures or building on it. It could be a cornfield, an area of heath land or a 
cleared and level site recovered from some defunct industry. There are likely to 
be several ways to lay it out and the first exercise is to decide on a concept. As 
stated before, there may only be information on the first production unit but the 
positioning of that unit will have a critical influence on the success of the site in 
the future. It is essential to look ahead and prepare a conceptual image of how
this site could look when fully developed to allow a logical expansion of the site 
in future years. 
The first consideration of the master plan is associated with zoning of the 
site — which areas will be allocated to offices, amenities, warehousing, utility 
generation, workshops and production plants. Zoning plans also contribute to 
solutions for the most efficient utilities distribution design and are the first 
stages of development of site logistics. 
Master plan - landscaping 
Having zoned the site, the overall site landscaping strategy can be developed. 
This will be very much dependent on company policy and any particular need 
to screen the plant. The outline site contours will have been decided and any 
necessary planting schemes can be worked out. 
The master plan 
Once the site has been zoned, a generic plot plan has been developed and 
outline landscaping has been decided, it is then appropriate to proceed with the 
overall master plan. The purpose of master planning is to look at how the site 
could be when it is fully developed and then only build the part that is required 
in the first instance — this ensures that what is actually built will fit into a 
logical site development. The master plan should be revisited at the time of each 
future project and modified if necessary to keep in line with changing 
requirements. 
On-site roads 
Discussions with the statutory authorities will have already identified the 
approved entrance and exit from the site, but the on-site road system should 
be developed based on the zoning plan. This must take into consideration gate 
house procedure, off-loading facilities, car parking, restricted access areas, 
emergency access, road vehicle access, forklift truck access and pedestrian 
circulation. The road system must also be capable of progressive development 
as the site expands without disruption to operations. 
Car parking policy can often present major problems. By the very nature of 
the site operation, the site is likely to be away from built up areas and operator 
car parking space is therefore essential. The safest practice is to provide it 
outside the main operational site boundary, but this may not be a popular choice 
on large sites in geographically exposed locations. The main emphasis must 
however be to ensure that private vehicles cannot get within recognized safety 
distances of operating units. Road system designs must recognize this 
requirement.
Public utilities and site generated utilities 
Public utilities are likely to include towns water, electricity, natural gas and 
sewage. Earlier discussions with the supply companies should have identified 
where, on the site boundary, these utilities will be available. It is now necessary 
to decide on the appropriate site interface. In most cases a control booth is 
constructed for piped utilities and a transformer house and switch room for 
electricity is constructed adjacent to the boundary. 
On-site centrally generated utilities will normally include steam and 
compressed air. Refrigeration and recirculated cooling water is normally 
generated on each production plot. 
Utilities, liquid raw material and interplant transfers can be distributed in 
several ways: 
above ground: this will normally involve a pipe bridge and is possibly the 
most convenient way of distribution in that it does not interfere with traffic 
and pedestrian circulation at ground level. An access platform should be 
fitted to the bridge for maintenance purposes; 
below ground (in an open culvert): the culvert walls may be inclined or 
vertical. This has the advantage of easy access for maintenance, but has to be 
bridged at each road crossing and is difficult to keep clean; 
below ground (in closed trench): this is not favoured for bulk drug sites 
because of possible hazards to operating staff and difficulty in maintenance; 
surface run: this method causes problems to traffic and operator circulation. 
The design of the distribution system must allow for future expansion in 
both layout and capacity. The question of ring main capability, which may be 
required in the future if not initially, must be examined. The master plan must 
reserve space on the site for the extension of possible bridgework in the future. 
This design will require an estimate of peak and average usage of utilities when 
the site is fully expanded. This, together with forward assessment of future 
marketing forecasts, will allow an informed decision on the initial sizing of the 
distribution system. 
Site offices, gate house, amenities, laboratories, warehouses 
It is assumed that the site being discussed is for production only. Based on this, 
the general administration offices are likely to be small and can possibly be 
sited in the same building as catering and possibly laboratories, although this 
will depend on the nature of work being carried out in the laboratories. The 
building should be sited adjacent to the entry gate to the site, thus limiting the 
need for visitors and office staff to go through any operational areas. 
The catering facilities are likely to be used by day staff as shift staff associated
with production operation are likely to have their own facilities within the 
control room building of the production unit. The office building will be 
positioned in an unclassified area of the site. 
The procedures for receiving road transport arriving at and leaving the site 
will determine the layout of the gate house area and the final positioning of the 
gate house. Appropriate lay-bys for lorries and weighbridge facilities may need 
to be incorporated in the layout. 
It will always be good practice to minimize vehicular access to the vicinity of 
the operating units. The site warehousing policy will influence this considerably. 
Each production plot can have its own warehouse for raw materials and 
finished goods. This would of course require road transport to have access to 
loading and unloading docks near to operating units. In addition the storage of 
high value, finished products adjacent to chemical reaction operations could 
give rise to a potential financial risk in the case of a hazardous incident 
occurring. It is not possible to generalize on recommendations for positioning 
warehouses but if possible the main warehouse should be positioned in the 
unclassified area of the site and the specific production units could have a small 
storage capacity for finished goods under test and possibly one or two day raw 
material storage. The production plant stores would be supplied by on-site 
forklift trucks. 
Engineering workshops 
Engineering workshops may be directly associated with each production unit or 
may be a site centralized facility — the size of the site will influence the choice. 
In medium to large sites it would be normal to have both a central workshop and 
satellite workshops on each of the production units. Certain engineering 
operations can only be carried out under flame permits or in workshops in 
unclassified zones. 
The production unit 
The discussion on generic production plot layouts identified a number of 
considerations for the individual production plot. The plot will generally house 
the buildings and facilities identified above, but there are no hard and fast rules 
and the requirements for specific products may differ greatly. For the purpose of 
this guide, it is assumed that automated batch reactor plant are being dealt with 
that carry out potentially hazardous processes. Processes that could result in 
explosions and/or use or produce highly active chemicals should be housed in 
a special hazard unit in an isolated area of the site.
The site layout 
With due regard to the above considerations, it is possible to draw up a site 
master plan based on typical processing requirements and information from 
marketing and research and development departments. This can entail some 
guesswork but it gives more logic to the development and hopefully prevents, 
for example, the construction of the site boiler house on the area that might be 
required for a future production unit. The data for the plot layout for the first 
production unit should be available but maybe not those for future units. It is 
normal, however, for a company to be involved in specific types of chemistry 
and this may allow the concept of a typical plot layout to be developed, 
although the concept is unlikely to satisfy the detailed requirements of the next 
factory. The flexible parameters of this master plan are discussed in more detail 
in the next section. 
The master plan suggests that on the area of land under consideration it is 
possible to construct up to, say, five separate production units of a size 
applicable to normal bulk drug facilities. Each unit would have the necessary 
on-plot facilities including a bulk liquid tank farm, the relevant on-plot utility 
generation, a control room and management offices. Depending on the design 
of the main reactor building there could be reactor capacity up to 96,000 litres 
using a variety of reactor sizes. The site infrastructure possibly includes central 
site generation of steam and compressed air and space has been identified for 
engineering workshops, special hazard operation and effluent treatment and 
conditioning. A number of these buildings may be developed in a phased 
manner as the site expands. 
The plan gives a basis for future expansion and allows a logical development 
that is not too restrictive. 
5.4.3 Brown field sites 
There is a wide range of brown field projects — it could begin with a cleared 
area within an existing production site that can be fenced off from adjacent 
operational areas or an area of an existing building that has been cleared for a 
new production unit or it could even be the installation of additional equipment 
in an operating factory. They all have one thing in common — they will all be 
influenced by what is already there. The cleared plot will have to take into 
account the existing site infrastructure; the cleared building will have to take 
into account the potential limitations of the existing structure; the additional 
equipment project will have to recognize the existing utilities and the impact of 
ongoing production operations within the building. For the purposes of this 
guide the discussion will be limited to the cleared site.
It is likely that the brown field project will be equivalent to the production 
plot concept described in green field section. The site boundary will be 
equivalent to the green plot boundary and it should be anticipated that the 
necessary public utilities and centrally generated site utilities would be made 
available at the boundary. The project may or may not include the augmentation 
of these utilities. Considerations for the layout will include: 
process buildings; 
control rooms; 
on-site utility generation; 
tank farm and drum stores; 
switch rooms and transformers; 
warehouse; 
offices and operator amenities. 
In most cases the approach will be similar to a green field production plot 
except for the impact of the surrounding existing site and the restrictions it 
might introduce, both to design and construction activities. 
In some instances integration with the existing site road systems might 
require substantial modification to the existing system. In other examples, the 
new production facility may be required for operation under GMP standards 
when the rest of the site is manufacturing a non-pharmaceutical product. 
The overall approach to the layout of brown field site should follow the same 
general principles as described for green field sites. The overall picture should 
be considered before settling on the layout for the specific plot in question. 
5.4.4 Layout specifics for biotechnology facilities 
Personnel and material flows have to be carefully designed to allow an orderly 
progress of product from fermentation through purification to finishing whilst 
minimizing the risk of cross-contamination. Other factors that are important to 
facility design include constructability, operability and maintainability. The 
latter covers accessibility to equipment for maintenance purposes especially in 
clean rooms; services access can be provided via the interstitial space above 
ceilings or via voids in the walls connecting onto corridors. All these factors 
should be optimized to maximize space utilization and minimize facility cost. 
Due to the changing nature of the biotechnology field, it is important to 
incorporate features into the design to enable expansion, re-use of existing 
space and re-use of equipment. Some of the methods available include: 
mobile vessels that can be moved easily to provide flexibility; 
centralized buffer solution preparation areas;
centralized cleaning areas for mobile vessels, etc.; 
centralized kill systems for liquid/solid wastes. 
However, these methods would have to be reviewed carefully to obviate any 
possibility of cross-product contamination. 
5.5 Good manufacturing practice for BPC 
5.5.1 Regulatory framework 
The manufacture of any pharmaceutical product is subject to regulations 
dependant on the country in which the product is sold. In the case of BPCs, 
the main regulatory body is the Food and Drug Agency in the US. They expect 
BPCs to be manufactured in accordance with the rules laid down in the Code of 
Federal Regulations title 21. Within the EU the manufacture of pharmaceutical 
material is regulated by EU Rules for Pharmaceutical Manufacture, Volume IV 
Current thinking from the FDA is that they expect manufacturers to 'control 
all manufacturing steps, and validate critical process steps'. 
A critical step is not necessarily the last step in manufacture but may be one 
which: 
introduces an essential molecular structural element or results in a major 
chemical transformation; 
introduces significant impurities into the product; 
removes significant impurities from the product. 
Further information on this topic can be found in Chapter 3. 
5.5.2 Good manufacturing practice (GMP) 
The manufacture of BPCs in accordance with GMP ensures that the product has 
a high degree of assurance of meeting its predetermined quality attributes. 
GMP for a BPC is concerned with the manufacturing process, the equipment 
and facility in which it is carried out. 
GMP is all about protecting the product from anything that can cause harm 
to the patient. This covers the processing itself and the avoidance of any 
contamination. 
Modern BPC manufacture is generally carried out in closed process 
equipment so the potential for contamination is greatly reduced. Special 
attention is paid to activities that involve exposure of the product or its raw 
materials or intermediate stages. This involves protection of the operator and 
the process when dispensing, reactor charging, sampling and product packing.
GMP is also concerned with cross-contamination from other sources and 
linked systems. Special attention is paid to hold up within process systems, 
cleanability and the use of Clean In Place techniques, interactions with shared 
systems such as nitrogen and vents. 
GMP is involved with the operating method. Any instruments that record 
critical data have to be calibrated and validated to ensure the integrity of the 
data. The process must be well understood and capable of being controlled. 
5.5.3 Validation 
The validation for BPC follows the same concepts and requirements to those 
detailed in Chapter 4. The main difference for BPC production is the concept of 
a critical step, and the point at which validation and pharmaceutical quality 
assurance have to be applied.
6.1.1 Introduction 
The selection of manufacturing methods for pharmaceuticals is directly related 
to the means by which the active substance is brought into contact with the 
agent responsible for the illness. 
The obvious administration route for the delivery of drug therapy has long 
been via the mouth, perhaps on the basis that the ailment under treatment was 
probably caused by the assimilation of some hostile agent via the same route! 
More localized treatments involving the application of agents to the skin, or the 
insertion of medicament-containing substances into the various body cavities 
was a logical development of oral entry. 
These methods, with enhancements and improvements, remain with us 
today and are still the most widely used, but they have been joined by 
injectable and other transcutaneous routes, inhalations and transdermals. 
A brief description of each of these, together with their associated manufacturing 
procedures, is outlined in the following sections. 
6.1.2 Pills 
One of the earliest forms of oral-dose treatment took the form of manuallyrolled 
gum-based pills. Thomas Beecham, one of the pioneers of pharmaceutical 
formulation, sold his original Tills' in a market at Wigan. These wonders 
were originally produced by mixing the gums with herbal extracts known to 
have pain-killing or laxative properties, and were sufficiently popular that the 
initial production methods needed to be updated and mechanized quite soon in 
the products' life history. Thus, the pill-rolling machine was produced, followed 
by the introduction of quality control in the form of a device which ensured that 
the individual pills were as perfectly spherical and of equal size as the rolling 
machine could produce — rejects being recycled for further processing. 
6 
S e c o n d a r y 
p h a r m a c e u t i c a l 
p r o d u c t i o n 
JIM STRACEY and RALPH TRACY 
6.1 Products and processes
6.1.3 Tablets 
Although a successful formulation, the pill suffered from production output 
restrictions and was overtaken by the modern tablet — produced by mechanically 
compressing suitable mixtures of drug substance and excipients held in a 
cylindrical cavity, or die, by the action of piston-type tools. 
During the early development of the tablet, it was quickly realized that in 
most cases the active drug substance did not lend itself to the formation of a 
reliable compacted entity merely by the application of pressure. The addition of 
binding agents was found to be necessary, together with other excipients 
offering enhanced powder flow, and the following characteristics of wellmade 
tablets were soon established as important: 
the ability to withstand mechanical treatment (packaging, shipping, dispensing); 
freedom from defects; 
reasonable chemical and physical stability; 
the ability to release medicaments in a reproducible and predictable manner; 
the drug and excipients are compressible. 
6.1.4 Granulation 
The process of tablet making using modern machinery involves the blending of 
the drug substance with binders, fillers, colouring materials, lubricants etc., 
followed by a series of operations designed to increase the bulk density and 
uniformity of the mixture and prevent segregation of the drug. These operations 
are known as granulation, and are an important part of modern pharmaceutical 
product manufacture, notably for tablets but also for other products. The 
granulation process is a critical step in reliable drug manufacture, as it often 
involves the relative 'fixing' of several ingredients and must therefore be 
carefully designed and controlled. Regulatory pressures, demanding as they do 
a strict equivalence of product performance before and after development scaleup, 
ensure that during drug research and development the selection of 
granulation methods must be made carefully. This selection, including the 
choice of individual equipment types, can be difficult and costly to change, 
owing to the need for the validation of continued product performance. 
The desired increase in bulk density and uniformity can be achieved by 
compression methods followed by milling, a process known as dry granulation. 
The techniques used for compression include 'slugging', a process not unlike 
tablet making, and roller compaction, which involves the feeding of material 
between a set of closely spaced steel rollers. The former produces tablet-like 
structures, which can then be reduced to granules by milling, whereas the latter
gives rise to a flake-like compact that is first broken into smaller pieces and then 
reduced by milling. In either case, the forces and friction involved are such that 
a lubricating material (such as magnesium stearate) is necessary. To ensure 
good material flow, a material such as Cab-o-Sil (silicon dioxide) is often used. 
Figure 6.1 shows a flow diagram for a dry granulation process. 
The dry granulation process is not very easy to contain in terms of dust 
emission and available equipment suitable for pharmaceutical applications is 
IBC Ingredients 
V-blender 
Lubricant 
addition 
Roller-compactor 
Lubricant 
addition IBC
V-blender 
Dry mill 
IBC IBC 
Compression 
Figure 6.1 Typical dry granulation process
not common. This is mainly due to its greater use in heavy chemical, food and 
fertilizer manufacture. However, all formulation departments will attempt to 
formulate a dry process, as it is cheaper in capital equipment and a simpler 
process. 
Therefore, the process most often used is wet granulation. This operation 
takes the blended materials, adds a suitable wetting agent, mixes the combined 
materials, passes the wet mass through a coarse screen, dries the resultant 
granules using a tray or fluid-bed dryer, and finally reduces the particle size of 
the dry material by passing it through a finer screen. 
Figure 6.2 (see page 115) shows a typical flow diagram for a conventional 
wet granulation process. 
The increasing potency of drug substances has encouraged manufacturers to 
seek granulation methods that are enclosed and free of dust emissions. Thus, a 
number of process equipment manufacturers have developed systems for 
enclosed processing which incorporate several of the granulation steps in a 
single unit. 
The most common of these is the mixer-granulator, which combines the 
powder mixing, wetting, wet massing and cutting operations. These efficient 
machines can perform this set of processes within a matter of minutes, and 
discharge a wet granule which requires only drying, milling and final blending with 
lubricants to produce a tablet compression mix. In most cases, however, the 
discharged wet granule will be further reduced in size by passage through a coarsescreen 
sieve prior to drying, in order to improve drying rates and consistency. 
The key to mixer-granulator operation is the combination of high-shear 
powder mixing with intense chopping of the wet granule. 
Figure 6.3 (see page 116) illustrates a typical mixer-granulator. 
The process steps employed in mixer-granulators are as follows: 
mixing of the dry ingredients with the main impeller and chopper rotating at 
high speed (15ms"1 impeller tip speed and 4000 rpm chopper speed) for, 
typically, 3 minutes; 
addition of a liquid binder solution by pumping, spraying or pouring it onto 
the dry material with the impeller and chopper running at low speed (5 ms"1 
and 1500 rpm) for around 2 minutes; 
wet massing with impeller and cutter running at high speed (2 minutes); 
discharge of the granulated material through a coarse sieve or directly to a dryer. 
The step times indicated will vary according to the product involved, and are 
generally critical in relation to granule consistency. 
There are a number of advantages that combined-processor granulators have 
over conventional methods, as follows:
Figure 6.2 Typical wet granulation process 
IBC 
Compression 
Lubricant 
addition 
V-blender 
Dry mill 
Drying air 
Fluid 
bed 
dryer 
Exhaust 
air 
Mixer-granulator 
Wet mill 
IBC 
Sieve 
IBC 
Dry Mill 
Dispensed 
materials 
Mix solution
Figure 6.3 High shear mixer-granulator with opening lid 
the granulation steps are enclosed in a single unit that can integrate with 
subsequent-stage equipment, thus minimizing dust emissions; 
the process is rapid; 
binder liquid volumes can be reduced; 
granule characteristics can be adjusted easily by changing step times and 
binder addition rates; 
inter-batch cleaning can be performed easily, and can be achieved by use of 
automatic Clean In Place systems. 
However, disadvantages do exist, mainly associated with the high speed and 
energy input provided by the agitators. This can give rise to mechanical 
breakdown of ingredient particles, over-wetting due to compaction producing 
over-sized granules, and chemical degradation of sensitive ingredients due to 
temperature rise. 
Developments of the mixer-granulator include jacketed and heated or cooled 
mixing bowls, which avoid over-heating of the granules or assist in their drying, 
and the use of vacuum to reduce drying times and temperatures. These 'single- 
Motor power monitor 
Product 
discharge 
Chopper blade 
Liquid inlet 
Vent air filter 
Impeller
pot' units aim to provide an efficient and contained operation covering as many 
granulation steps as possible in a single unit. 
Single-pot mixer-granulators using vacuum and heated jackets, but employing 
slightly different configurations of impeller and chopper, include the 
Zanchetta Roto granulator/dryer, which uses a vertical-axis retractable chopper. 
This machine also operates slightly differently in that the bowl is pivoted so 
that the effective heat exchange surface can be maximized for reduced drying 
time. The planes of shear within the powder mass can also be altered at each 
stage of the process for optimum mixing and final size reduction. 
The application of microwave energy for granule drying in-situ has been 
pioneered by Aeromatic-Fielder. The magnetron generators are situated on top 
of a mixer-granulator that operates under vacuum and are energized at the end 
of the wet massing/chopping cycle. 
Figure 6.4 (see page 118) shows a flow diagram for a combined granulation 
process. 
Spray granulation 
A different and somewhat unusual granulation technique is the use of the spray 
dryer. 
Spray granulation requires that all ingredients are soluble or dispersible in a 
common solvent and can be crystallized/combined from that solvent at a 
suitable temperature. The solution or suspension feed stream is passed through 
a nozzle inside the spray dryer chamber, where it immediately comes into 
contact with a co-current or counter-current gas stream at controlled temperature. 
The solvent evaporates rapidly and the resulting solids are separated from 
the air stream by cyclone separators and filters. 
Spray granulation offers a number of advantages over mixer-granulation 
systems. The feed, being a homogeneous liquid, removes concerns over 
blending of liquid binders into dry solids. The resulting granules are homogeneous 
and, regardless of size, contain uniform proportions of the ingredients. 
Temperature control is also more consistent, thus eliminating problems of heatdegradation. 
Finally, the absence of mechanical moving parts generally 
improves cleanability and reduces contamination risks. 
A recent example of this principle is the Spinning Disc Atomization system 
being developed in Switzerland by Prodima SA and EPFL. In this system a 
suspension of the product or a polymer melt is passed between rotating 
concentric conical discs and is released into the gas stream as fine uniform 
droplets, which dry or solidify to produce very spherical and similar granules.
Fluid-bed granulation 
A related process for achieving granulation by spray techniques utilizes the 
mixing action of a fluidized bed to mix powder ingredients in an otherwise 
conventional fluid-bed dryer. The mixture so created is then subjected to a 
Figure 6.4 Typical combined granulation process 
IBC 
Compression 
Mixer-granulator-dryer 
IBC 
Sieve 
IBC 
Dry mill 
MIX SOLUTION 
Dispensed 
materials
sprayed-on binder solution, the evaporation of whose solvent produces an 
intimately-mixed granulate which is then dried by the fluidizing air stream. 
Direct compression 
Some drug substances have characteristics that allow them to be compressed 
without prior granulation, using a process known as 'direct compression'. This 
process avoids the cost and inconvenience of granulation, but often requires the 
use of special binding agents to avoid segregation during mass flow of the mix 
in the tablet compression process. 
Figure 6.5 (see page 120) shows a typical flow diagram for direct compression. 
6.1.5 Tablet compression 
The basic principles of the tablet compression process have remained 
unchanged since their inception. The tablet press compresses the granular or 
powdered material in a die between two punches, each die/punch set being 
referred to as a station. Although many alternative methods have been tried, the 
principle of filling granules into a die and compressing them into a tablet 
between two punches is still the primary method of manufacture for all 
machines used in pharmaceutical manufacturing. 
Developments utilizing a slightly different configuration of punch and 
die are under current examination in Japan and Italy. The primary incentive 
of these developments is to produce an arrangement which can reliably be 
cleaned-in-place, rather than relying on the time-consuming process of 
dismantling the machine to remove product-contact parts for cleaning with 
its attendant risks of operator exposure to active products. 
Tablet machines can be divided into two distinct categories: 
those with a single set of tooling — single station or eccentric presses; 
those with several stations of tooling — multi-station or rotary presses. 
Figure 6.6 (see page 121) illustrates the principles of tablet machine 
operation. 
The former are used primarily in the small-scale product development 
role, while the latter, having higher outputs, are used in production 
operations. Additionally the rotary machines can be classified in several 
ways, but one of the most important is the type of tooling with which 
they are to be used. 
There are basically two types of tooling — 'B' type which is suitable for 
tablets of up to 16 mm diameter or 18 mm length (for elliptical or similar 
shapes), and 'D' type which is suitable for tablets with a maximum diameter or
Figure 6.5 Typical flow diagram for direct compression 
Bulk tablets 
Tablet compressing machine 
IBC 
V-blender 
Sieve 
Dry mill 
IBC IBC Dispensed 
materials
maximum length of 25.4 mm. The 'B' type punches can be used with two types 
of die; the small 'B' die is suitable for tablets up to 9 mm diameter or 11 mm 
maximum length, and the larger 'B' die is suitable for all tablet sizes up to the 
maximum for the 'B' punches. Machines can, therefore, be used with either 4B' 
or 'D' tooling, but not both. 
Machines accepting 'B' type tooling are designed to exert a maximum 
compression force of 6.5 tonnes, and machines accepting 'D' type tooling 
10 tonnes. Special machines are available which are designed for higher 
compression forces. 
The maximum force that can be exerted on a particular size and shape of 
tablet is governed by the size of the punch tip or the maximum force for which 
the machine is designed — whichever is smaller. 
Figure 6.6 Rotary tablet compression machine operation 
1 - Feed frame 
2-Die 
3 - Pull down cam 
4 - Wipe off blade 
5 - Weight control cam 
6 - Lower compression roll 
7 - Upper compression roll 
8 - Raising cam - upper punches 
9 - Raising cam - lower punches 
10 - Ejector cam 
Fill Compress Eject 
10 
9 
6 
5 
3
1
2 
4 
7- 8
Figure 6.7 Some tablet shape possibilities 
Tablets are now available in a range of diameters and thicknesses to suit the 
proportion, active dose and characteristics of the drug substance. Figure 6.7 
shows some examples of tablet shape possibilities. 
Formulation has enabled the production of tablets with special characteristics 
such as: 
effervescent; 
chewable; 
multi-layer; 
delayed or sustained release; 
bolii for veterinary use. 
These examples indicate the extent to which development of the tablet has 
continued since its original introduction. Much effort was expended during the 
first half of the 20th century in establishing the best particle size of the active 
drug and the range and rheology of excipients needed to produce a reliable 
tablet with acceptable dispersion and absorption characteristics. However, the 
technology of tablet compression did not advance significantly during this 
period; reliable and robust machinery was produced and its performance and 
output were considered suitable for the demands of the time. Subsequently, 
improved excipient development by the pharmaceutical industry, based on 
enhanced glidants and micro-crystalline cellulose binding agents, and the 
introduction of reliable sensors coupled with electronic control systems have 
allowed compression technology to advance. 
Flat elliptical Pillow-shaped with breakline Triangular biconvex 
Flat, bevel edge Biconvex, 2-layer Flat toroidal 
Biconcave Biconvex with breakline Biconvex
Whereas the manufacture of a single tablet is simply a matter for formulation 
development, the production of such products at machine speeds in excess of 
300,000 tablets per hour raises additional challenges. The critical stage here is 
the delivery of the granulation into a die on a high-speed rotating disc 
accurately, so that tablets of minimum weight variation can be produced. 
Very high-speed compression machines are now available with built-in 
tablet weight and thickness control and the ability to be self-monitoring from an 
output and quality standpoint. Hence, it has become possible for continuous, 
unmanned operation of the tabletting process to be carried out (the so-called 
'lights out' working). 
More recently, the greater impetus to improve has come from regulatory 
pressures, under which the need for uniformity, consistency and reliability has 
become paramount. The principles of current Good Manufacturing Practice 
(cGMP) and validation have greatly influenced the development of the tablet 
manufacturing process and the materials and methods used therein. 
6.1.6 Coated tablets 
Many tablet products contain active materials that require taste masking or a 
controlled release rate, and a variety of methods have been developed to 
achieve these objectives. A careful choice of excipients can mask the 
unpleasant taste of certain compounds, but a more reliable procedure is to 
coat the tablet with a barrier material. Such coating can be achieved by forming 
a compressed layer around the basic tablet, or core. There are compression 
machines that can accept a previously formed core and surround it with a layer 
of excipient material. An additional and similar use of compression can 
produce layered tablets. 
The traditional method of taste masking is to apply a sugar coating to the 
core, and although this method has largely been superseded by film-coating 
techniques, it is still used. Originally the sugar coating was applied by pouring a 
sugar syrup, usually coloured, onto a bed of pre-varnished tablet cores rotating 
in a steel or copper pan into which warm air was blown. The skill required to 
achieve a successful application of the sugar coat was such that the true art of 
tablet making/coating resided in the hands of a small and respected elite. A key 
feature of the sugar coating process was that the tablet weight increased 
significantly with the sugar coating accounting for typically 60% of total 
tablet weight. 
Subsequently this skill has largely been replaced by a more-automated 
system using mechanized spray/jets of sugar syrup applied in a pre-determined 
and controlled manner to a bed of tablets rotating in a perforated drum and 
warmed with pre-heated air.
A logical development of automated sugar coating was the introduction 
of non-sugar coating materials, based on plastic film-forming solutions/ 
suspensions. This 'film coating' process has largely replaced the original 
sugar coating technique, although the method of application is basically similar. 
Advantages are the removal of food-type materials, a higher speed of throughput 
and a small increase in tablet size/weight, with consequent reductions in 
packaging cost. 
Initially, most film-coating formulations included the use of flammable 
solvents for coating solution/suspension manufacture, and given the relative 
toxicity and safety risks associated with these materials it is not surprising that 
much effort has been expended in developing aqueous-based alternatives. The 
latter now make up the majority of film-coating formulations. 
Figures 6.8 and 6.9 (see pages 124 and 125) are flow diagrams showing the 
stages of the film and sugar coating processes. 
Dispensed 
materials 
Mix 
sealing 
solution 
Mix 
sub-coat 
solution 
Mix 
colour 
coat 
solution 
Mix 
polish 
solution 
Warm 
air 
Warm 
air 
Warm 
air 
Warm 
air 
Warm 
air 
Bulk 
tablet 
cores 
Seal 
cores 
Dry Sub-coat Colour 
coat 
Polish 
To packing 
Inspect 
Figure 6.8 Tablet sugar coating
6.1.7 Capsules 
The encapsulation process is an alternative to tablet compression, which also 
masks unpleasant-tasting actives. It can also have advantages where compression 
could result in a compacted tablet with unacceptably long or short 
dispersion time in the upper alimentary system. As with tablets, the gelatin 
barrier can be further coated with 'enteric' materials which ensure dissolution 
or dispersion only in that part of the system where optimum effect is produced. 
Capsules are generally of two types, made with either hard or soft gelatin. 
Figure 6.9 Tablet film coating 
Inspect 
To packing 
Integrated film coater 
Bulk 
tablet 
cores 
Warm air 
Mix coating 
solution 
Dispensed 
materials
Hard gelatin capsules 
Hard capsules are manufactured from bone gelatin and are produced as empty 
two-part shells supplied to the pharmaceutical manufacturer for filling. The 
capsules are produced in a number of standard sizes designated 5 through 000, 
with larger sizes available for veterinary applications. 
Although originally filled by hand, and later by devices that allowed multiple 
cap/body separation, volumetric filling and reassembly, they are now filled on 
automatic machines. These separate the two parts, fill the body with powder, 
granules, pellets or semi-solids as required by the formulation to a controlled 
level, and reassemble the two parts prior to discharge. One disadvantage of the 
hard capsule is that a number of systems for dosage control have been 
developed by different filling machine manufacturers, so that (unlike tablets) 
the capsule has no standardized filling system. 
The original hard capsule type, which was conceived as long ago as the 
1840s, consisted of two plain-sided cylinders with hemispherical ends, one of 
larger diameter, so that one formed the body and the other the cap. Tolerances 
during manufacture (by dipping pins in molten gelatin) ensured that the 
cap/body clearance was minimized to prevent the possibility of powder 
leakage. Originally designed to deliver powder products, improvements in 
formulations and capsule tolerances have allowed the use of this dosage form 
for delivering oils and pastes. 
Where fine powder escape or simple separation of the two parts proved 
problematic, these capsules were sealed by the application of a band of molten 
gelatin at the cap/body joint. This was achieved using conveyor-type machines, 
which provided space and time for the gelatin band to set, and provided an 
opportunity for visual inspection of the capsules. 
The introduction in the late 1960s of the self-locking capsule, coupled with 
improved dimensional tolerances, largely removed the necessity for band sealing. 
After the initial establishment of hard-shell capsules as a dosage form, 
machines were developed to increase the production rates of filled shells. One 
of the first types, developed by Colton and by Parke-Davis, consisted of a twoplate 
device that simply separated the two halves of the shells, filled the bodies 
volumetrically, and allowed recombination. One of the first commercially 
available machines to automate the process was developed by Hofliger and 
Karg of Germany, and filled at speeds of 150 capsules per minute. This machine 
used the differential diameter of the capsule cap and body to orientate and 
vacuum to separate the two parts, and an auger device to meter the product 
powders or granules and feed them into the capsule bodies. The caps and bodies 
were then re-combined prior to ejection. 
Figure 6.10 (see page 127) illustrates a typical capsule filling process.
1 Powder dosing Vacuum 2 Powder compression 
Vacuum 
3 Ejection Plug 4 
discharge 
Plug 5 
recovery 
Figure 6.10 Details of powder filling on capsule filler 
These techniques for capsule handling have basically been retained in later, 
higher-speed machines, but the dosing system has undergone a divergence in 
design. The original auger type filler is no longer used, mainly because it is not 
capable of high-speed operation without recourse to multiple stations, which 
would give rise to an unacceptably large machine. 
The system developed in the 1960s by the Zanasi brothers in Italy, and still 
used today, employs a plug-forming method to produce the required dose.
A tube is plunged into a container of product having uniform depth, and the 
column of product so contained is compressed in-situ by the downward motion 
of a piston inside the tube. On withdrawal of the tube a cylindrical compact is 
retained within it, and this is then discharged into a capsule body by further 
downward motion of the piston. The dose weight and degree of compression 
(and subsequent dispersion) of the product is capable of adjustment by altering 
the depth of powder/granule in the product container and the extent of 
downward motion of the piston. One advantage of this so-called 'dosator' 
system is that the tube is quite small, so that a number of them can be arranged 
in a dosing module of modest dimensions to give increased output. Original 
machines worked with an intermittent motion, but later versions were designed 
to operate continuously by arranging the capsule feed/handling groups and the 
dosing units on separate rotating turrets, emulating to some extent the 
conventional tablet press. 
To meet the challenge of the higher-speed dosator machines, Hofliger and 
Karg introduced their GKF range of machines, which utilizes the natural 
capacity of the capsule body for controlling product dosing. The capsule 
bodies, having been separated from their caps and fed vertically into cylindrical 
machined holes in a rotating disc, are moved so as to pass under a container of 
product powder/granule (not unlike the feed frame of a tablet compression 
machine), so that the product mix flows into the empty bodies. Before leaving 
the product container, the contents of the capsule bodies are subjected to 
compression by the insertion of pistons to a pre-determined and adjustable 
depth. After compression, the bodies are removed from the dosing zone by the 
rotation of the disc and reunited with their caps. 
This system allowed for a significant speed increase compared with the 
auger type, but was disadvantaged in that the degree of dosage weight and 
compaction control was less than that allowed by the dosator system. A revised 
version was therefore introduced which included an intermediate dosing disc 
which allowed for the formation of a product 'plug', independently of the 
capsule body, which could then be transferred to the body after formation and 
compression. This development permitted the use of dosing discs of different 
thickness to control dose weight. 
Again, the small dimensions of the Hofliger and Karg dosing arrangement 
made it possible to fill capsules at very high speeds of over 2500 filled capsules 
per minute. 
Apart from size considerations, the key to high-speed capsule filling is 
powder flow, which in turn relies on consistent particle size and shape 
distribution. The bulk density of the filling material is of parallel concern, 
and must be uniform if reliable dosage weights are to be achieved. As with
tablet compression, the conditions and processes employed for preparation of 
the filling mix have critical impact on performance. A typical capsule filling 
mix for a high-dose product may contain only the active drug and a lubricant 
(for example, many antibiotic products are formulated in this way), so the 
options for formulation adjustment are limited. 
Products utilizing a lower active dose proportion may also contain a filler 
(such as lactose), flow-aid (for example, silicon dioxide) and surfactant (such as 
sodium lauryl sulphate) and may therefore have superior flow and output 
characteristics. 
Soft gelatin capsules 
Soft gelatin capsules, where the gelatin contains a plasticizer to maintain 
flexibility, were originally developed in France in the 1830s, and are generally 
used where the active product material is liquid or semi-solid, or where the most 
appropriate formulation is in this form. They were originally made in leather 
moulds, which provided an elongated shape and a drawn-out end which could 
be cut off to allow for the insertion of the product liquid, after which the end 
could be sealed with molten gelatin. 
Although less popular than hard-shell capsules, their 'soft' counterparts 
satisfy a different set of product/market criteria, under which the total containment 
of the active principals is a key concern. 
The manufacture of soft-gelatin capsule products is generally regarded 
as more specialized than that of other dosage forms and has been limited 
to a small number of producers. These companies have very much influenced 
the development of the technology employed in the production process. 
R P Scherer developed the modern technology for automated soft-gelatin 
capsule production in the 1930s by designing the Rotary Die Process. The basic 
technique employed in soft-shell filling involves the melting of a gelatin/ 
plasticizer mixture and the extrusion of this between the two halves of a mould 
formed by twin rotating cylinders, while the product liquid or solid is injected 
between the two half-shells thus produced. The continued rotation of the 
cylindrical moulds results in the closing and sealing of the resultant capsule and 
its subsequent ejection. 
6.1.8 Pellets and other extrudates 
A feature of capsules, which can have drug-release benefits, is that they can be 
filled with materials other than powder or granule mixtures. In addition to 
liquids and pastes, which are generally more suited to soft gelatin types, 
product in the form of large granules or pellets can be filled into hard-shell 
capsules.
Whereas 'large' granules can be prepared by the methods already 
described, pellets have their own production technology, based upon extrusion 
and spheronization. The spherical granules, or spheroids, have several 
advantages over conventional granules due to their uniform shape — they 
have superior flow properties, are more easily coated and have more 
predictable active drug release profiles. Dried spheroids may be coated and 
then filled into hard gelatin capsules to provide a sustained release dosage 
form capable of gradually releasing its active constituents into the gastrointestinal 
tract over several hours. 
The process of extrusion has been the subject of much scientific study in the 
polymer, catalyst and metal industries. It may best be described as the process 
of forcing a material from a large reservoir through a small hole, or 'die'. 
Pharmaceutical extrusion usually involves forcing a wet powder mass 
(somewhat wetter than a conventional granulation mix) containing a high 
concentration of the drug substance together with a suitable binder and solvent, 
through cylindrical holes in a die plate or screen. Provided the wet mass is 
sufficiently plastic this produces cylindrical extrudates of uniform crosssection, 
not unlike short strands of spaghetti. These extrudates are loaded 
onto the 'spheronizer', a rotating scored plate at the base of a stationary 
smooth-walled drum. The plate initially breaks the strands into short rods, and 
then propels them outwards and upwards along the smooth wall of the drum 
until their own mass causes them to fall back towards the centre of the plate. 
Each individual granule thus describes a twisted coil pathway around the 
perimeter of the plate, giving the whole mass a doughnut-like shape. This 
movement of the granules over each other combines with the friction of the 
plate to form them into spheres. 
A typical spheronizer arrangement is shown in Figure 6.11 (see page 131). 
The basic core granules for the preparation of controlled release pellets for 
filling into capsules can be prepared by several methods, such as spray coating, 
pan/drum granulation, melt granulation, as well as spheronization. Core 
granules are then coated with a suitable polymer or wax to confer on them 
their controlled-release properties, either by spraying wax-fat solutions onto 
granules tumbling in pans or by spray coating them with polymers or waxes in a 
standard film coating machine. 
The melt-granulation pelletization process is a fairly recent technique, based 
on high-shear mixer-granulator technology. In this process the core material 
(drug substance) is mixed with a suitable low-melting solid excipient (such as 
high molecular weight polyethylene glycol) in a high-shear mixer. The agitation 
is continued until the heat generated melts the excipient, which forms a 
wax-like coating around the core material. Under controlled conditions it is
Blender 
Mixer-granulator 
Extruder 
Exhaust 
air 
Spheronizer 
Fluid bed 
dryer 
Drying 
air 
Product 
Figure 6.11 Typical spheronization process
thus possible to produce coated pellets of reasonably uniform size, which can 
exhibit dissolution or dispersion properties suited to the drug substance 
involved. 
6.1.9 Syrups, elixirs and suspensions 
These dosage forms are basically produced by the dissolution or suspension of 
a drug substance in a suitable solvent/carrier (usually purified water), together 
with appropriate sweeteners, flavours, colours and stabilizing agents. 
The primary use of these products is in paediatric and geriatric treatment, 
where the patient may have difficulty in swallowing solid-dose medicines, 
although they are also valuable where the pre-dissolution or pre-suspension 
of the active drug can enhance therapeutic effect (for example, cough remedies). 
The production of solutions is a relatively straightforward procedure, 
typically using purified water heated to a minimum temperature suitable for 
dissolution of the materials, with the addition of the active and excipients 
followed by a filtration to remove possible haze prior to filling. 
The difficulties inherent in syrup manufacture are associated with product 
stability, for example dissolution and solubility, which may not be adequate at 
normal temperatures and taste masking, which is made more difficult when the 
drug is in solution. 
Suspensions overcome some of these problems for suitable products, but 
other difficulties exist — notably maintaining the product in suspension. This 
latter challenge can only be met by the use of a high-shear dispersion system, or 
homogenizer, which utilizes wet-milling techniques to reduce particle size and 
enable reliable product suspension. 
Elixirs are basically clear, flavoured solutions containing alcohols and 
intended for oral administration. Other ingredients may include glycerin, 
sorbitol, propylene glycol and preservatives. Quite high alcohol contents 
were common to ensure dissolution of certain drug substances, although 
products formulated in this way are becoming unusual. 
The distinction between medicated elixirs and solutions is not altogether 
straightforward, the latter often containing alcohol (for example, up to 4% is 
present in some ephedrine-containing syrups). 
6.1.10 Emulsions 
An emulsion is a two-phase liquid system where one liquid exists in very small 
droplet form (the internal phase), suspended in another (the external phase); the 
two liquids being otherwise insoluble in one another. An emulsifying agent 
contained within the mixture acts on the surface active properties of the two 
liquids such that the emulsion remains stable for a sufficiently long period to
serve its purpose. If necessary, the liquids may be heated in order to enhance the 
stable formation of the emulsion, by reducing its viscosity. The active 
pharmaceutical material may be a solid, which is added to the liquid/liquid 
system, or may be soluble in one of the components. The product is prepared by 
high-shear mixing to reduce droplet sizes, using submerged-head agitation 
devices which draw the mixture through a high-speed rotating impeller 
contained within a close-fitting housing, not unlike a centrifugal pump. 
Most pharmaceutical or cosmetic emulsions contain water and oil as the two 
phases, and may be oil/water or water/oil, depending upon which is the 
internal and which is the external phase. It is possible for emulsions to 'invert'; 
a process in which is the internal and external phases change identity between 
the water and oil ingredients. 
Although more usual in cosmetic topical formulations, pharmaceutical 
emulsions are prepared for topical, oral and parenteral use. Owing to their 
difficulty in preparation, pharmaceutical emulsions are used infrequently and 
only where they exhibit particularly useful characteristics such as drug 
solubility or specific absorption capability. 
6.1.11 Creams, ointments and other semi-solids 
Creams are basically similar to emulsions in that they are two-phase liquid 
systems; however, they exhibit greater physical stability at normal temperatures 
than emulsions and can thus be more useful for topical applications. The 
external phase is often water, while the internal phase is usually a high-viscosity 
oil or semi-solid oleic material. 
Manufacturing involves the heating and stirring together of the two phases 
in the presence of emulsifying agents and other excipients (colour, stabilizers, 
perfume etc.) with the assistance of a high-shear mixing device (colloid mill, 
homogenizer or ultrasonic mixer). The operation is most often carried out at 
slightly elevated temperatures to enhance dispersion. If the active substance is a 
solid, it will normally be added to the stabilized mixture, followed by further 
agitation and homogenization. 
Ointments are solutions of high melting point and lower melting point 
hydrocarbons, usually mineral oil and petroleum jelly. The active drug and 
other excipients are incorporated in much the same way as with creams with the 
semi-solid matrix being heated to assist dispersion of these additives. 
An advantage of ointments over creams is that, when used as a base for 
sterile products such as ophthalmics, being solutions they can be sterilized by 
filtration after the addition of a soluble active or prior to the final addition of an 
insoluble sterile active ingredient. Cream bases would break down under 
microfiltration conditions.
Modern ointments based on polyethylene glycols (PEGs), which are available 
in a range of viscosities, have the advantages of typical ointments but are 
water miscible. 
Pastes are similar to ointments except that they contain much higher 
insoluble solids content. They are prepared in a similar fashion, with the 
semi-solid base being added to the solids gradually with mixing until the 
required concentration is achieved and the dispersion is uniform. Pastes are 
used where a particularly high concentration of the medicinal compound is 
needed in contact with the patient's skin (such as for burns, prevention of 
sunburn or the treatment of nappy rash). 
Gels are semisolid systems in which a liquid phase is held within a threedimensional 
polymeric matrix consisting of natural or synthetic gums, with 
which a high degree of physical or chemical cross-linking has been introduced. 
Polymers used to prepare pharmaceutical gels include natural gums such as 
tragacanth, pectin, carrageen, agar and alginic acid and synthetic materials such 
as methylcellulose, hydroxyethylcellulose, carboxymethylcellulose and the 
carbopols (synthetic vinyl polymers with ionizable carboxyl groups). 
6.1.12 Suppositories 
The original suppositories were hand-formed pellets based upon white paraffin 
wax and containing active material and relevant excipients dissolved or 
dispersed in the melted matrix. Eventually the need for standardization resulted 
in the development of pre-formed moulds into which the cold product mass was 
forced by means of a piston and cylinder arrangement. 
This slow process was later superseded for volume production by warming 
the mass to its melting point and pouring the liquefied material into split 
moulds, which were then solidified by cooling. 
The early types were wrapped in greaseproof paper packaging and were 
successful except that any rise in ambient temperature would result in melting, 
with subsequent leakage and product spoilage; hence the introduction of plastic 
disposable mould materials which were closed with adhesive or heat-sealed 
cover strips. Initially the moulds were sold as pre-formed strips containing 
typically five moulds. Machinery was developed which filled these strips in 
rows, followed by cooling/solidification and the application of seal tapes. 
These machines have relatively low output, but are suitable for the production 
rates often associated with this dosage form. Later form-fill-seal machines 
provide capacity for larger product sales, and involve the forming of moulds 
automatically on-line, followed by filling, cooling and heat sealing using a 
single packaging material. A feature of all fill-seal suppository machines is the 
need to allow for the shrinkage coincident with the cooling/solidification
process. This requires that the filled moulds are cooled to allow solidification of 
the contents prior to sealing, and the machines are often quite long in size to 
accommodate the length of the cooling section. 
6.1.13 Oral, nasal, aural drops and sprays 
Oral medicines applied in drop form are usually neonatal versions of paediatric 
syrups and suspensions. They are filled into small bottles, often of a flexible 
plastic that allows the container to be squeezed so that the requisite number of 
drops of liquid can be exuded through the plastic dropper insert. 
Nasal solutions are similar except that the formulation will usually be 
isotonic with nasal secretions to preserve normal ciliary action. The drugs used 
in such formulations include ephedrine, for reducing nasal congestion, antibiotics, 
antihistamines and drugs for the control of asthma. 
Products formulated as aural drops, usually referred to as otic preparations, 
include analgesics, antibiotics and anti-inflammatory agents. They are usually 
based on glycerin and water, since glycerin allows the product to remain in the 
ear for long periods. In the anhydrous form, glycerin has the added benefit of 
reducing inflammation by removing water from adjacent tissue. 
Sprays used orally or nasally, are similar in formulation to their equivalent 
drops, being simple solutions and suspensions traditionally applied to the 
mouth, throat or nose by bulb type spray devices. Modern formulations make 
use of plastic pump sprayers or simple flexible bottle/nozzle combinations to 
produce the required spray pattern. 
6.1.14 Ophthalmic preparations 
Two formulation types are generally used in ophthalmic treatment; ointments and 
liquid drops, which together provide for both water soluble and oil soluble active 
principals. They are produced in the same way as oral formulations in terms of 
the equipment and processes, although a higher level of cleanliness is required. 
Products for the treatment of eye disorders have traditionally been manufactured 
under clean conditions, not least to avoid complications arising from 
the introduction of foreign particles to the eye (such as corneal ulcers or loss of 
eyesight). The need for medicines used topically on the eye surface to be aseptic 
was not originally thought necessary, owing to the fact that under normal 
conditions the eye's surface is in direct contact with the external environment, 
which contains many infective agents. Thus, like the alimentary system, the eye 
was thought able to cope with such challenges without additional protection. 
More recently however, it has become accepted that under many circumstances 
requiring medicinal treatment, the eye has an increased liability to infection by 
organisms such as Staphylococci or Pseudomonas aeruginosa, and should
therefore not be exposed to any substance likely to give rise to such infection. It 
is now an internationally recognized pharmacopoeial requirement that ophthalmic 
preparations be prepared aseptically. 
6.1.15 Injections 
A potentially unwanted feature of orally dosed medicines is their introduction 
to the body's system via the route designed for digestion, a process more 
effective in decomposition of chemical entities than in their intact delivery to 
the remotest regions of human or animal physiology! 
The mouth, throat, stomach and intestines contain a complex mixture of 
enzymes and acids, which will usually ensure that any orally-ingested medicine 
is, at the very least, altered before it can be absorbed into the bloodstream. It is 
the bloodstream that distributes the absorbed material and until the said 
material enters the bloodstream it is unable to create any effect beyond areas 
of immediate contact within the alimentary system. 
Hence, if a medicinal substance has poor stability in acid solution or is easily 
broken down by digestive enzymes, it is of very little use in disease control as it 
will probably not reach those parts of the body's systems requiring treatment. 
A method of avoiding this effect and delivering the substance closer to the site 
of the illness or infection is via a transcutaneous injection. Although some 
drugs are unstable in body fluids including blood, the injectable route very 
much enhances the possibilities for overcoming instability problems. 
The two most common forms of injection are intramuscular, where the 
substance is injected into tissue containing small blood vessels and therefore 
remains most effective local to the injection site; and intravenous, involving 
direct injection into a larger blood vessel, thus ensuring rapid transit around the 
body. A further procedure involves sub-cutaneous injection, used for the 
deposition of controlled-release formulations. 
Whether for intramuscular or intravenous use, these products are liquids or 
suspensions, which are produced as a pre-sterilized material contained in 
ampoules or vials. The medicinal product may be based on aqueous or oil 
formulation depending on the relative solubility of the drug substance and/or 
the required release rate into the surrounding body tissue. Most injectable 
products are made as single-dose containers, although multi-dose systems are 
available for use in vaccination and in veterinary practice. 
Additionally, drugs requiring sustained application via intravenous infusion 
over long periods are produced as large volume systems (typically 500 or 
1000 ml). 
Liquid products in solution can be filled under sterile conditions within 
suitable clean areas, the solution being itself sterilized by filtration using
0.2 micron porosity filters. However, the preferred manufacturing procedure is 
to ensure sterility by terminal sterilization of the filled ampoules or vials, by 
autoclaving or gamma irradiation. Only where such terminal sterilization 
techniques are likely to cause decomposition of the drug substance is it 
considered acceptable to rely only upon manufacture under sterile conditions 
to achieve the required standard. In such cases the extent of sampling for 
sterility testing of the final product will be increased. 
Although sometimes desirable for the terminal sterilization of heat-sensitive 
suspensions, it should be noted that irradiation is not without problems. Apart 
from the obvious safety considerations, the effect of gamma radiation on the 
type of glass used for ampoule and vial manufacture is to cause brown 
discolouration, thus adversely affecting subsequent inspection operations. 
The generation of free radicals within product solutions is also a possibility, 
with consequent chemical deterioration. 
Where the active drug is unstable in solution (such as for certain antibiotics) 
the product is filled into vials, under sterile environmental conditions, as a dry 
powder. Such materials are often very moisture-sensitive, and special arrangements 
need to be made to ensure a low-humidity environment in areas of 
product exposure. A key consideration here is that the products are themselves 
required to be sterile before the filling operation, which implies preliminary 
processing under sterile conditions. 
The filling of powders into vials involves considerations not customary for 
liquid filling, such as the mechanism used for dosage weight control. Similar 
techniques to those used for capsule filling have been tried, but most suffer from 
excessive particulate contamination generation. Modern high-speed sterile 
powder filling machines utilize a vacuum/pressure technique which forms 
a temporary solid compact from the product powder prior to its ejection into the 
vial. 
Although some powder products can be sterilized by gamma 
irradiation or heat sterilization, most cannot be treated this way. Methods 
adopted to manufacture bulk sterile products include spray drying, bulk freeze 
drying, and crystallization under sterile conditions. 
An alternative technique for the manufacture of products exhibiting 
instability in solution is to prepare such solutions using non-sterile product 
material and sterilize them by filtration, fill them within a controlled time-span 
into vials in small batches, and freeze dry. This method ensures that a solution 
can be produced, sterile filtered and filled under aseptic conditions, then recrystallized 
by sublimation within the vial. 
Equipment for this process relies on the use of special vial seals or plugs 
which, when partially inserted into the vials, allow evaporation of the solvent
during the drying phase. The drying is followed by the automatic full insertion 
of the plugs within the dryer chamber, under aseptic conditions. In this way the 
finished filled vials can be demonstrated to be equivalent to vials filled with 
liquid under aseptic conditions. 
6.1.16 Sterilization techniques 
Products intended for parenteral administration must not contain viable 
microbial organisms and their manufacture will inevitably involve one or 
more sterilization stages. Such stages may be used for the drug substance, 
the filling container or the finished product itself. 
Even where materials are processed under conditions of strict asepsis, it is 
now required that the finished product should be subjected to a terminal 
sterilization process wherever possible. 
A number of possible methods exist for the sterilization of products and 
materials, and the most appropriate method will be selected after careful 
consideration of the effects that the various alternative systems might have 
on those materials. Each method has particular benefits when applied to specific 
requirements. 
The commonly used systems for sterilization include moist heat (autoclaving), 
dry-heat, chemical treatment, irradiation, high-intensity light and solution 
filtration. With the exception of the last one, all the methods rely on a combination 
of intensity and time to achieve the required reduction in microbial 
content. 
Another factor to be considered is the possibility for pyrogens to be present 
in the sterilized material or component. Pyrogens are substances that cause a 
rise in the patient's body temperature following administration of the injectable 
pharmaceutical. They are in fact complex polysaccharides arising from the 
breakdown of bacterial cells, and are most likely to be present following moist 
heat sterilization or other lower-temperature sterilization techniques (such as 
irradiation). 
Autoclaving 
The most useful and longest-standing batch sterilization technique is autoclaving, 
which exposes the subject materials to saturated steam at a temperature/ 
time combination appropriate to the stability of those materials. 
Established effective sterilization conditions range from 30 minutes at 115°C, 
to 3 minutes at 134°C. Commercially available autoclaves are supplied with 
standard cycles that provide time/temperature combinations falling within this 
range. These standard cycles include specific time/temperature combinations
and also the facility for cooling large-volume product solutions in containers at 
the end of the sterilization phase, by means of deionized or purified water sprays. 
The latter process includes the simultaneous application of cooling water and 
sterile compressed air to the autoclave chamber, in order to prevent highpressure 
drops across the container walls and consequent breakages. 
Provided that the steam in the autoclave is saturated and free from air, the 
different cycle temperatures may be attained by developing various specified 
pressures in the autoclave. It is preferable however to control the process by the 
temperature attained rather than by the pressure, as the presence of air in the 
autoclave results in a lower temperature than that expected under the correct 
conditions from the indicated pressure. In the case of porous materials, the air 
must be abstracted or displaced from the interstices in order to achieve 
sterilizing conditions, as the presence of residual pockets of air within the 
material may prevent contact between the steam and parts of the load. 
The period of heating must be sufficiently long to ensure that the whole of 
the material is maintained at the selected temperature for the appropriate 
recommended holding time. The time taken for the material to attain the 
sterilizing temperature or to cool at the end of the holding time can vary 
considerably and depends on a number of factors, including the size of the 
container or object and the thickness of its walls, and the design, loading, and 
operation of the autoclave. It is necessary, therefore, that adequate tests are 
conducted to ensure that the procedure adopted is capable of sterilizing the 
material and that the material can withstand the treatment. Chemical indicators 
can be included in the autoclave load, which change colour after the specified 
temperature has been maintained for a given time. Reliance should not be 
placed, however, on chemical indicators except when they suggest failure to 
attain sterilizing conditions. 
The process can be monitored by temperature-sensitive elements (thermocouples) 
at different positions within the load. Some indication that the heat 
treatment has been adequate can be gained by placing indicators at positions 
within the load where the required conditions are least likely to be attained 
(such as the chamber drain). 
For the purposes of validating the sterilization conditions, the bactericidal 
efficiency of the process may be assessed by enclosing in different parts of the 
load small packets of material containing suitable heat-resistant spores, such as 
those of a suitable strain of Bacillus stearothermophilus. These are checked 
subsequently for the absence of viable test organisms. 
It is common practice for autoclaves to be double-ended with access doors 
opening into a clean preparation area on the infeed side and an aseptic filling area 
on the outfeed, although single-door autoclaves are used in some applications.
Dry heat 
Dry heat sterilization, often referred to as depyrogenation, uses high temperature 
conditions in the absence of moisture to destroy contaminating organisms 
and eliminate pyrogenic material. It is particularly useful for sterilizing glass 
containers (such as vials) or any other product-contacting material that will 
tolerate the required temperature. Typical conditions for this process are 2000C 
or more with a residence time at that temperature of 15 minutes, although 
sterilization alone is achievable at lower temperature/time combinations. The 
process can be operated on a batch basis using double-door machines (built into 
barrier walls in a similar manner to autoclaves), which accept clean containers 
on the non-sterile side and deliver them sterilized on the aseptic side. 
Modern high-output filling lines use continuous tunnel-type sterilizers, 
which include complex air-handling systems and deliver the cooled, sterilized 
containers into the aseptic filling machine located within the aseptic area. The 
validation of high-temperature sterilization techniques requires similar considerations 
to those applicable to autoclaving. 
Heating with a bactericide 
This process can be used for sterilizing aqueous solutions and suspensions of 
medicaments that are unstable at the higher temperatures attained in the 
autoclaving process. 
In this process, a bactericide is included in the preparation at the recommended 
concentration and the solution or suspension, in the final sealed 
container, is maintained at 98° to 1000C for 30 minutes to sterilize the product. 
The bactericide chosen must not interfere with the therapeutic efficacy of the 
medicament nor be the cause of any physical or chemical incompatibility in the 
preparation. 
Ambient chemical methods 
Formaldehyde was once used extensively as a means of sterilizing spaces such 
as aseptic production rooms and surgical operating theatres, but is now rarely 
used owing to its high toxicity and relative corrosiveness. It is only an effective 
sterilant in the presence of moisture; the process involves raising the ambient 
room humidity by water spraying, followed by the sublimation on an electric 
hot plate of paraformaldehyde pellets. 
Peracetic acid has been used as an alternative to formaldehyde for the 
sterilization of small spaces, such as filling machine enclosures, isolators, 
together with their contents. Like formaldehyde, it is corrosive and toxic and, 
therefore, is of limited application. It has been used in admixture with hydrogen
peroxide for the sterilization of isolators. Peracetic acid has the advantage that 
the sterilizing effect is (as with all chemical sterilants) dependent on concentration, 
which can be easily measured with suitable detection equipment. 
Hydrogen peroxide has now largely supplanted peracetic acid for smallspace 
sterilization, as this agent is far less likely to cause corrosion of 
equipment items. It is also used for sterilizing syringes, ampoules and other 
packaging materials. 
Hydrogen peroxide is used at concentrations of lOOOppm in air and is 
regarded as product-safe due to its decomposition products being water and 
oxygen. It has a melting point of 00C, and its commonly used 30% aqueous 
solution has a boiling point of 1060C. 
It is, however, toxic, having a time-weighted exposure limit of 1 ppm and an 
acute toxicity limit of 75 ppm. Another disadvantage has been the difficulty in 
monitoring accurately the concentration of hydrogen peroxide vapour under 
sterilization conditions, although in recent times suitable sensors have been 
developed. These sensors have relatively slow response times, making real-time 
analysis of hydrogen peroxide difficult, but it is now possible to reliably 
validate the sterilization process. 
Various alcohols (ethanol, iso-propanol) can be used to decontaminate the 
surfaces of containers or equipment items, usually by swabbing. However, this 
activity cannot be relied upon to provide sterility in its own right and must be 
preceded by a validated sterilization process. 
Ethylene oxide sterilization 
Certain materials cannot be sterilized by dry heat or autoclaving for reasons of 
instability, but they may be sterilized by exposure to gaseous ethylene oxide. 
This process can be carried out at ambient temperatures and is less likely to 
damage heat-sensitive materials. It does, however, present difficulties in control 
of the process and in safety, and is currently only considered where it offers the 
only solution to a problematic sterilization requirement. It must be performed 
under the supervision of experienced personnel and there must be adequate 
facilities for bacteriological testing available. The most frequent use of the 
technique in the pharmaceutical area is for the sterilization of medical devices 
(such as plastic syringes). 
Compared to other methods of sterilization, the bactericidal efficiency of 
ethylene oxide is low and consequently particular attention should be paid to 
keeping microbial contamination of subject materials to a minimum. 
Ethylene oxide is a gas at room temperature and pressure. It is highly 
flammable (at levels as low as 3% in air) and can polymerize, under which 
conditions it forms explosive mixtures with air. This disadvantage can be
overcome by using mixtures containing 10% of ethylene oxide in carbon 
dioxide or halogenated hydrocarbons, removing at least 95% of the air from the 
apparatus before admitting either ethylene oxide or a mixture of 90% ethylene 
oxide in carbon dioxide. It is also very toxic to humans (time-weighted average 
exposure limit 1 ppm) and has been demonstrated to be carcinogenic. For these 
reasons ethylene oxide sterilization is no longer frequently used as an industrial 
process. 
There are two processes used for ethylene oxide sterilization, one at normal 
and the other at high pressure. The low-pressure process uses a 10% v/v 
concentration, a temperature of 200C and a cycle time of around 16 hours. A 
suitable apparatus consists of a sterilizing chamber capable of withstanding the 
necessary changes of pressure, fitted with an efficient vacuum pump and with a 
control system to regulate the introduction of the gas mixture, maintain the 
desired gas pressure, adjust the humidity within the chamber to the desired 
level and, if required, a heating element with temperature controls. 
The high-pressure process was developed to enhance output by reducing cycle 
times. It uses a more-substantial chamber design, suitable for the lObarg operating 
pressure. The temperature is typically >50°C and the cycle time 3 hours. 
As with any chemical sterilization process, the combination of time and 
sterilant concentration is the key factor. The sterilizing efficiency of the process 
depends upon: 
the partial pressure of ethylene oxide within the load; 
the temperature of the load; 
the state of hydration of the microorganisms on the surfaces to be sterilized; 
the time of exposure to the gas. 
All these factors must be closely controlled for successful sterilization. The 
sensitivity of microorganisms to ethylene oxide is dependent on their state of 
hydration. Organisms that have been dried are not only resistant to the process 
but are also slow to rehydrate. Due to this, it is not sufficient to rely solely 
on humidification of the atmosphere within the chamber during the sterilizing 
cycle. 
It has been found in practice that hydration and heating of the load can be 
more reliably achieved by conditioning it in a suitable atmosphere prior to 
commencing the sterilization. 
Some materials absorb ethylene oxide and, because of its toxic nature, great 
care must be taken to remove all traces of it after the sterilization is finished; 
this is achieved by flushing the load with sterile air.
Irradiation 
Sterilization may be effected by exposure to high-energy electrons from a 
particle accelerator or to gamma radiation from a source such as cobalt-60. 
These types of radiation in a dosage of 2.5 mega-rads have been shown to be 
satisfactory for sterilizing certain surgical materials and equipment, provided 
that precautions are taken to keep microbial contamination of the articles to a 
minimum. This method is not, however, widely regarded as a safe means of 
product sterilization, due to the possibility of chemical decomposition of many 
pharmacologically active substances. 
This method can also be used for some materials that will not withstand the 
other sterilization methods. It has the advantage over other 'cold' methods of 
sterilization in that bacteriological testing is not an essential part of the routine 
control procedure, as the process may be accurately monitored by physical and 
chemical methods. It also allows the use of a wider range of packaging 
materials. 
Control of the process depends upon exposure time and radiation level. It is 
important to ensure that all faces of the load are exposed to the required 
radiation dose. 
Ultraviolet light 
Ultraviolet light has long been known as a form of energy with bactericidal 
properties. It has particular uses in the maintenance of sterility in operating 
theatres and animal houses, and for the attenuation of microbial growth in water 
systems. Ultraviolet light exists over a broad wavelength spectrum (0.1 to 
400 nm) with the bactericidal (UVC) component falling in the range 200 to 
300nm with a peak at 253.7nm. 
It is particularly useful for maintaining sterility in pre-sterilized materials 
and is used widely in isolator pass-through chambers to protect the internal 
environment of the isolator. It can also be used for continuous production 
sterilization of pre-sterilized components feeding into such isolators. 
It can be used to sterilize clean materials in a continuous cycle provided that 
they are fully exposed to the radiation, but this is a relatively slow process 
requiring an exposure time of up to 60 seconds to achieve a 5-log reduction in 
viable organisms. 
High-intensity pulsed light 
A recently developed method of sterilization uses very short pulses of broadspectrum 
white light to sterilize packaging, medical devices, pharmaceuticals, 
parenterals, water and air. It has been demonstrated that this process kills high 
levels of all micro-organisms. Each light flash lasts for a few hundred millionths
of a second but is very intense, being around 20,000 times brighter than 
sunlight. The light is broad-spectrum, covering wavelengths from 200 to 
lOOOnm, with approximately 25% in the UV band. The latter component 
provides the sterilizing effect in short-duration high-power pulses, although the 
total energy required is quite low — an economic advantage. 
High kill rates equivalent to 7-9 log reductions in spore counts have been 
demonstrated using a few pulses of light at an intensity of 4-6 joules cm"2. 
Although the UV component provides the effectiveness of this method, it is 
considerably more rapid than conventional UV systems. Continuous in-line 
sterilization is, therefore, practical with this technology. 
Pulsed light sterilization is applicable to situations and products where light 
can access all the important surfaces and also penetrate the volume. It will not 
penetrate opaque materials, but is efficiently transmitted through most plastics 
and may be used to sterilize many liquid products. 
Filtration (liquids) 
Liquids may be sterilized by passage through a bacteria-proof filter. This 
process has the advantage that the use of heat is avoided, but there is always a 
risk that there may be an undetected fault in the apparatus or technique used, 
and because of this each batch of liquid sterilized by filtration must be tested for 
sterility compliance. 
Sterilizing filters can be made of cellulose derivatives or other suitable 
plastics, porous ceramics, or sintered glass. The maximum pore size consistent 
with effective filtration varies with the material of which the filter is made and 
ranges from about 2 urn for ceramic filters to about 0.2 um for plastic 
membrane filters. 
Particles to be removed in the sterilizing process range in size from 1 to 5 fjm 
diameter, down to viruses of 0.01 jam. It appears at first sight that filters cannot 
remove particles smaller than the largest pore size of the filter. However, 
filtration occurs in a wide variety of mechanisms, including impaction, 
adsorption, adhesion and electrostatic effects, so that in practice particles 
much smaller than the interstitial channels may be effectively filtered out. 
Filters for liquid sterilization have pore sizes of 0.2 urn, usually preceded by 
coarser pre-filters to remove larger particles. These filters are all fabricated as 
cartridges that are installed in leak-tight housings. For the filtration of liquids, 
hydrophilic forms of the filter material are used. 
All standard filter types must comply with bacterial challenge tests 
performed by the manufacturer, which can be correlated with other integrity 
tests carried out routinely by the end-user.
Non-disposable filters must be tested periodically before use to ensure that 
their efficiency has not become impaired, using one or more of the following 
integrity test methods. Filters should be integrity tested after each sterilization 
and after each filtration. All integrity testing is performed on wetted filters. The 
tests depend on the principle that airflow through the wetted porous membrane 
is diffusive up to a certain pressure (the bubble point) and is a function of pore 
size and pressure. Above the bubble point, liquid is displaced from the 
membrane and bulk flow of gas occurs. 
Bubble point test: In this test, air pressure upstream of a wetted filter is 
slowly increased. The pressure at which a stream of bubbles occurs downstream 
of the filter is the bubble point pressure. If a filter has a damaged membrane or 
an insecure housing seal, the test pressure will be below that specified by the 
manufacturer. 
Forward flow test: A test pressure below the bubble point pressure 
is applied to a wetted filter. The diffusive airflow rate through the filter is 
measured. If it exceeds a specified value the filter is judged to be insecure. 
Pressure hold test: A section of pipework upstream of the wetted filter 
is pressurized (below the bubble point). The rate of pressure decrease is 
measured. For a filter to be judged intact, this must occur below a specified rate. 
Filtration is best carried out with the aid of positive pressure, as this reduces 
the possibility of airborne contamination of the sterile filtered solution through 
leaks in the system. If the filtration is likely to take a long time and the 
preparation is susceptible to oxidation, nitrogen or other inert gas under 
pressure should be used rather than compressed air. 
Filtration (gases) 
The uses of sterile air or inert gas in pharmaceutical sterile processing include 
the aseptic transfer of liquids using pressure, and blowing equipment dry after 
sterilization. In addition to these positive applications, air or gas also enters 
aseptic equipment during fluid transfers or cooling operations, and in all cases 
the air and gas must be completely free of micro-organisms. Air sterilization 
can be achieved by filtration with the required filter porosity being 0.2 /im as for 
liquids. Integrity testing also needs to be carried out in the same manner as with 
liquid filters. 
6.1.17 Aerosols 
The use of pressurized systems for the application of Pharmaceuticals became 
common after World War II, when such methods were used for the topical 
administration of anti-infective agents, dermatological preparations and 
materials used for the treatment of burns. A logical development of spray
technology, the aerosol relies on the propulsive power of a compressed or 
liquefied gas. The latter type have been of greater benefit, based on gases 
boiling at below room temperature (200C) and at pressures ranging from zero to 
120psi above ambient. 
Initial applications utilized flammable hydrocarbon gases, which were then 
largely replaced for pharmaceutical use by chlorofluorocarbons, notably for use 
in inhalation products. Recent developments have worked towards the replacement 
of the suspected ozone-depleting chlorofluorocarbons with hydrofluoroalkanes 
for environmental reasons. 
A further method of avoiding the oral route for internal administration is to 
introduce the drug substance by inhalation. 
Aerosol products for inhalation use first appeared in the mid-1950s and were 
used for treatment of respiratory tract disorders, based on the establishment of 
several key benefits: 
rapid delivery to the affected region; 
avoidance of degradation due to oral or injectable administration; 
reduced dosage levels; 
ease of adjustment to patient-specific dosage levels; 
avoidance of possible interactions with concurrently-administered oral or 
parenteral drugs; 
ease of patient self-administration. 
The typical modern pharmaceutical aerosol consists of an aluminium 
container, a product (in powder, solution or suspension form), a propellant 
and a cap/seal incorporating a metering valve. The propellant provides 
pressurization of the container at normal temperatures, and expels the product 
when the valve is opened. The dose is controlled by the valve orifice 
configuration, which allows the release of a single shot of product liquid 
together with sufficient propellant gas to ensure production of an aerosol. 
Continuous aerosol sprays for topical application use slightly different valve 
types that do not limit the dose size. Such products also sometimes utilize 
compressed gases to provide propulsion, including carbon dioxide, nitrogen 
and nitrous oxide. 
The manufacture of pharmaceutical aerosols is complicated by the need to 
maintain a pressurized environment for the propellants during storage, mixing 
and filling. This includes the systems used for transporting the propellants from 
the storage location to the point of use, and is made more complex where 
flammable materials are involved. 
The relatively complex nature of gaseous aerosol manufacture has led to the 
consideration of other methods for the delivery of drug substances by inhala-
tion, including the creation of fine particles suspended in an air stream 
generated by the patient himself. Such powder inhalations utilize micronized 
powders delivered in unit-dose quantities, held in a device that simultaneously 
releases the fine material into air flowing through the device at the same time as 
that airflow is initialized by the user. By careful design using a multi-dose 
approach, a metered dose system providing relief of patient symptoms over a 
convenient time period is possible. Several such systems are currently available 
or under development. 
6.1.18 Delayed and sustained release systems 
The objective of any drug delivery system is to provide a specified quantity of 
the therapeutic agent to the appropriate location within the body, and to sustain 
the level of that agent so that a cure or symptom relief is achieved. In practice 
drugs are delivered in a broad-brush manner, which ensures arrival of sufficient 
drug to the body location needing it, but simultaneously provides the drug to 
parts not requiring treatment. This approach may ensure coverage but is 
somewhat wasteful and may engender unwanted reactions. 
A targeted approach is therefore potentially valuable and there are a number of 
ways in which this can be achieved. The possible advantages of this approach are: 
improved patient compliance; 
reduced drug substance usage; 
reduced side effects; 
reduced drug accumulation; 
improved speed of treatment; 
improved bioavailability; 
specific delay effects possible; 
cost saving. 
The objective stated above has two parts, namely the creation of a suitable 
drug level at the required site, and the maintenance of that ideal level for a 
period suited to the completion of treatment. 
The first objective can be achieved by delayed release of the drug when 
taken orally, by localized application by injection, or by topical application 
local to the required site in the case of shallow-tissue disorders. Methods used 
for ensuring adequate levels of the therapeutic agent include sustained-release 
coatings for tablets and capsules, and formulations of injectable or topical 
drugs that allow controlled release of the active principal. 
The combination of delayed and sustained release properties for orally dosed 
material can ensure, for example, that the drug is released, at a controlled rate, 
in the duodenum rather than the stomach. Such controlled-release is achieved
with oral dosage by the formulation or coating of tablets and capsules so that 
the excipients (either internally or as part of the coating material) have a 
physical action on the drug dispersion or dissolution rate. 
Injectable drugs in a suitable formulation can offer delayed or sustained 
release when delivered intramuscularly, as a 'depot'. Dissolving or dispersing 
the drug in a liquid medium that is not readily miscible with body fluids can 
reduce the rate of absorption. Oil solutions or suspensions are often 
employed for this effect, while aqueous suspensions can be used with 
insoluble drugs. 
An alternative injectable route is the use of solid material injected subcutaneously, 
the 'depot' thus being formulated to ensure suitable release rates. 
The surgical implantation of drugs can be even more targeted, albeit at 
increased patient risk. 
Topical drug application has a number of benefits, especially the opportunity 
to remove the material from the skin by washing, so reducing and ultimately 
stopping the rate of application. The absorption of drugs via the skin e.g. 
transdermal products, including intra-ocular routes involves the formulation of 
the actives in such a way that they can be released from the carrier material at 
the rates required. Such formulation can involve the use of microporous 
materials to which or within which the drug is applied or mixed, applied 
directly or attached to a substrate (such as adhesive plasters). 
6.1.19 Microencapsulation 
The process of microencapsulation involves the deposition of very thin coatings 
onto small solid particles or liquid droplets and differs from the technique 
of, for example, tablet coating in that the particles involved are much 
smaller — typically 1 to 2000 /im in diameter. 
The benefits to pharmaceutical product development relate to the very small 
and controlled size of the particles involved. The technique alters the physical 
characteristics of the materials concerned to the extent that: 
liquid droplets can exhibit solid particle characteristics; 
surface properties are changed; 
colloidal properties are changed; 
pharmacological effects are enhanced or reduced by changing release 
patterns; 
the surrounding environment is separated from the active drug substance. 
Although some similar effects can be achieved by alternative methods, the 
microcapsule can, due to its small size, be used in many product applications 
which would not otherwise be technically practical.
Methods available for manufacturing microcapsules include spray drying, 
pan coating and air suspension coating. The former is of particular value in 
the production of very small microcapsules (typically 1 to 100 /mi in diameter), 
and has been used in protein-based product manufacture in which a protein 
solution is sprayed into a co-current air stream to form microcapsules. The codrying 
of such materials with pharmaceutically-active substances is capable of 
producing particles of such substances coated with a protective or carrier layer. 
6.1.20 Ingredient dispensing 
All pharmaceutical manufacturing operations involve the use of one or more 
chemical materials in pre-defined quantities on a batch or campaign basis. Such 
materials are most often held in a storage location, in containers providing 
sufficient quantities of the material to enable the manufacture of more than one 
batch. These containers will be of such design as to afford the required level of 
protection of the material during the storage period and facilitate allocation to 
the dispensary. 
The activity involved in the weighing of materials on a batch-by-batch basis 
is known as dispensing, and may be considered as the first step in the 
manufacturing process. 
The sub-division of a bulk material into smaller batch lots inevitably 
involves the removal of that material from its original container. The environment 
in which this process is conducted must, therefore, be of a quality suitable 
for the intended use of the manufactured pharmaceutical product. For example, 
the dispensing of ingredients for the manufacture of oral-dose products will 
usually be conducted under class 100,000 conditions (to US Federal Standard 
209e). The same operation for handling sterile ingredients for injectable 
products will usually be conducted under class 10 or 100 conditions, possibly 
using a glove-box. 
Another key feature of dispensing is the need for assurance that the 
operation has been carried out correctly. This need will often be met by the 
checking of each weight by a second operator. With modern computercontrolled 
dispensing systems, the latter situation is most common, as the 
reliability of the dispensing process itself is such that only the potential for 
errors in transit to the production area need to be checked. 
Containers 
As indicated above, the 'input' container will be of such design as to protect the 
integrity of the material, and so too must the container used for transferring the 
dispensed ingredient to the manufacturing location. Where high-potency
ingredients are involved, the latter must also ensure that subsequent handling 
can be performed without risk to operating personnel. Thus, a contained 
transfer system might be employed for this purpose (see Section 6.4 on page 
176). 
Incoming materials are likely to be contained in polyethylene-lined kegs 
(solids) or steel drums (liquids). These containers may hold as much as 200 kg 
of material and be transported on clean pallets. Space for the staging of such 
pallets adjacent to the dispensing zone is therefore required, together with 
handling devices suitable for positioning them conveniently for the removal of 
the required weights or volumes of ingredients. 
Dispensed materials may be placed in similar containers to those used for 
incoming items. However, it is more usual for these aliquots to be transferred to 
manufacturing using dedicated sealable dispensed-material containers, often 
reserved for particular substances, and carrying provision for secure identification 
of the contents. 
Weighing systems 
As pharmaceutical ingredients are usually dispensed by weight (rather than 
volume), a suitable set of weighing scales is required. Scale sensitivity and 
accuracy usually diminish as capacity increases, so a two or three-scale 
arrangement is not uncommon. Thus, the active ingredients, which are likely 
to be of lower batch weights than the non-active or excipient materials, will 
usually be weighed-out on scales of higher accuracy. The three scales might, 
typically, have capacities of lkg, 10 kg and 100 kg respectively. The chosen 
scale capacities will depend on overall batch weights and on the weight of the 
active, or smallest, ingredient. 
Electronic weighing scales are common in modern dispensaries, and these 
can be linked to computer-controlled dispensary management systems and to 
automatic identification and weight-label printers. 
Operator protection and airflows 
The protection of operating personnel from exposure to high-potency drug 
substances is as important during dispensing operations as it is in the 
subsequent processing. Hence, the arrangement of modern dispensing areas 
utilizes individual booths in which the ingredients for one product batch at a 
time are weighed and packaged. The operator must wear suitable protective 
clothing, which should include hair covering, long-sleeved gloves, dust mask, 
footwear and close-woven fabric overalls. 
Modern pharmaceutical dispensing booths employ a ventilation scheme that 
seeks to separate the operator's breathing zone from the area in which product
or excipient powders or liquids are exposed during dispensing. The basic 
principle relies on a downward sweeping of the ventilation air, from the ceiling 
above and behind the operator, to the lower edge of the booth wall facing them. 
Thus, any dust generated during scooping of materials into receiving containers 
is entrained in the air stream and kept away from the operator's head. A typical 
dust entrainment velocity is 0.45 m s~ \ and proprietary dispensing booths are 
designed to provide an operating zone in which the air stream moves at or above 
this velocity. 
The air leaving the lower back wall of the booth may be filtered to remove 
entrained ingredient dust and recirculated, while supply air make-up and 
recirculated air will generally be filtered and conditioned to the environmental 
quality standard required by the product being dispensed, typically class 
100,000 for oral-dose products. Figure 6.12 illustrates a typical airflow 
arrangement in a downflow dispensing booth. 
Cooling coil 
Air filter 
Control panel 
Air 
flow Fan 
Air 
filter 
Bench Air 
filter 
Protection zone 
Figure 6.12 Sectional diagram of dispensing booth
Surplus materials 
The disposal of surplus material remaining at the end of a product campaign in 
ingredient input containers generally poses a dilemma for dispensary managers. 
The options are to return the part-used container to the main raw material 
warehouse or to retain it as part of dispensing stock for later use. There is no 
universal 'best alternative', the decision being affected by such factors as the 
availability of space for storage within the dispensary area, the proximity of the 
main warehouse, the ownership of material stocks within the dispensary and 
warehouse, the sophistication of the materials management system, the level of 
security of the storage location etc. These all need to be considered when this 
issue is decided, but the overriding factor must be the security and integrity of 
the material itself. 
Cross-contamination risks 
In multi-product pharmaceutical manufacturing plants it is inevitable that the 
dispensary will be required to handle two or more products, probably at the 
same time. Thus, individual dispensing booths must operate in such a way as to 
ensure that there is no risk of materials from one product contaminating 
another. This is achieved by ventilation air pressure regimes that combine 
recirculatory air flow with slight positive pressure relative to adjacent access 
corridors and storage areas. By this means, dusts generated during the 
dispensing activity will be entrained and intercepted by the booth's extract 
filters, thus avoiding dispersion to the external environment. Meanwhile, any 
contaminant present in the adjacent spaces will be prevented from entering the 
booth by the positive pressurization. 
Cleaning arrangements 
One potential source of cross-contamination is the equipment and surfaces used 
during ingredient material handling. It is, therefore, important that all contaminated 
containers and utensils are removed from the dispensing booth for 
disposal or cleaning at the end of the operation, and that all working surfaces, 
including the fabric of the booth itself, are subjected to a validated cleaning 
procedure. Utensil and container washing is most effectively carried out in 
automatic washing machines, which should also incorporate a drying cycle. 
Open-sink washing of such items is unlikely to provide a validatable process, 
and should generally be avoided. 
Operator clothing is a further source of contamination, and operators must 
change their outer garments when product changes are made, and in all cases 
should change their gloves between sequential batches.
Labelling 
It is essential that all dispensed ingredients are reliably identified — including 
the batch number and name of the product batch that is to contain the 
ingredient, the item weight and material name. It may also include the identity 
of the dispensing operator and the time and date of dispensing. Although the 
manual generation of labels can be acceptable (assuming suitable checking 
systems exist and are in use), it is now considered worthwhile to arrange for 
these to be produced automatically by the dispensary management system. 
Thus, at the end of each weighing operation, the acceptance by the operator of 
the correctness of the weight and identity will initiate a bar-coded or alphanumeric 
label being printed by a printer located adjacent to his workstation. 
Such labels, usually of the self-adhesive variety, will then be applied to the 
dispensed material's container. 
Materials management systems 
Modern dispensary management systems are computer-driven, with fullyvalidated 
batch recipe information held electronically. They are most often 
linked to business management systems such as MRP2, warehouse management 
systems, and intermediate specialist control suites which organize the 
flow of material throughout the production process and seek to prevent errors in 
material usage. The latter, which usually incorporates the dispensary management 
element, must comply with the principles of cGMP and must, therefore, 
be driven by fully-validated software — this makes such systems very specialized 
and potentially costly. 
Materials management systems automatically update stock levels at each 
stage in the material pathway, including transfers of ownership between 
different departments (between warehousing and production, for example). 
They ensure that only approved material can be allocated for use, or indeed 
used, and that materials are consumed in accordance with normal stock rotation 
principles (such as first in, first out). 
The specific role of the dispensary management system is to ensure that 
ingredients are weighed out in accordance with pre-programmed recipe 
information and in the correct sequence. Instructions to dispensing operators 
may be provided via a printed batch sheet or visually by VDU screen. 
The systems often also include provision for printing of ingredient labels 
that provide identity, weight and batch code information, in either bar-code 
or alphanumeric form. Various add-on facilities may also be incorporated, such 
as programmed weigh-scale calibration routines, and authorized-operator 
identification. 
Next Page
6.2 Principles of layout and building design 
6.2.1 Introduction 
It has been said that the layout of a building can be designed in at least six 
different but equally acceptable ways. This may well be the case, although the 
degree of acceptability will vary depending upon the criteria applied by the 
accepting authority. 
The criteria that give rise to the differences in pharmaceutical secondary 
production building layouts include, but are not limited to: 
safety/means of escape; 
complexity of the enclosed processes or activities; 
personnel level, type of occupancy and movement; 
ease of materials movement; 
specialized environmental classifications; 
type of partition construction; 
the structural design of the building. 
6.2.2 Personnel safety 
The primary safety consideration for all buildings is means of escape in the 
event of fire or other emergency. The issues are complex and covered by 
legislation and fire engineering principles, and will not be discussed here. 
However, the pharmaceutical engineer is well advised to take account of the 
basic considerations when planning the process-led layout of a building, and in 
doing so should seek the advice of a qualified architect at the earliest practicable 
point. Although failure to do so may not result in a potentially dangerous 
building, it will almost certainly involve time-consuming and costly reconsideration 
of the building layout during its architectural design phase. 
Another important safety consideration relates to the product itself. In some 
cases the active materials involved in pharmaceutical manufacture are toxic in a 
high-exposure situation, and special precautions will then become necessary. 
These may involve modifications to the layout to accommodate specialist 
machines or environmental control equipment. The need for the use of 
flammable materials, although less common nowadays, may also arise and in 
such circumstances the design of the building may have to include the results of 
area zoning. This can be onerous, as construction materials may need different 
selections from those made elsewhere in the building, while the need for 
separate ventilation systems is also possible. 
Previous Page
6.2.3 Process and activity complexity 
Even simple pharmaceutical manufacturing and packaging processes must be 
carried out in areas with controlled environments. It is common practice to 
group final packaging operations, which usually involve the handling of 
products in a partially enclosed condition (such as filled and capped bottles, 
tablet blister packs) in a single room with limited spatial separation between 
linked groups of machinery, but with a common ventilation system. This is 
possible if the environment provides the required temperature (and sometimes 
humidity) to ensure product stability and that product cross-contamination is 
negated by the primary enclosure. 
In the case of manufacturing operations, even where products require similar 
levels of product protection, separate environmental and spatial arrangements 
are usually necessary to prevent cross-contamination. It is, therefore, usual for 
manufacturing and primary packaging processes to be conducted in productspecific 
common environments and is essential for such processes where any 
degree of cross-contamination is hazardous to the product or patient to be 
separated physically, as a minimum. 
It is possible for different products to share a common ventilation system, 
but only if that system allows for a single pass of the air supply, or if the 
recirculated air is passed through suitable filters. These filters must be of such 
porosity that it is possible to provide demonstrable evidence that any product 
dust passing through is of such low level as to ensure that products cannot 
become contaminated with one another at levels which pose measurable risk to 
patients. 
Production processes involving specially clean conditions for product 
exposure (such as for parenteral, ophthalmic or inhalation products) add further 
complexity to the environmental and space planning activity. The transfer of 
materials between areas of differing cleanliness classification often involves a 
process such as sterilization by autoclaving or other means. Hence, the relative 
size, shape and position of the rooms on either side of the transfer process are 
important. In any event it is often considered necessary to separate such 
different areas by the insertion of air locks, in which decontamination of 
materials and equipment can be performed prior to transfer. This procedure will 
also be required during active product manufacture to prevent the possible 
spread of such material to adjacent areas. 
In reality, the product mix in any production facility may be such that the 
above considerations demand dedicated spaces for different products. This 
demand inevitably impacts on the layout of the building, and it is for this reason 
that those responsible for facility space planning should understand the many 
and varying considerations.
6.2.4 Personnel occupancy level, type and movement 
Although the use of machinery for manufacturing and packaging operations is 
widespread and increasing, pharmaceutical production requires the employment 
of people for the control of material selection, movement, processing and 
inspection, and it is unlikely that such employment will be eliminated in future. 
A further consideration is that growing sexual equality coupled with a 
decreasing incidence of heavy lifting and movement means that both male and 
female production operators are equally likely to be employed on a process. 
However, the numbers of personnel likely to be engaged on a given 
operation is relatively low by general industrial standards, so that this feature 
does not generally pose great difficulties in facility planning. 
There is, nevertheless, a feature of pharmaceutical (and especially cleanroom) 
operations that needs careful consideration. Just as air locks are a 
characteristic of material transfers, operator clothes-changing rooms are a 
common necessity in the protection of products from people. Clean areas 
generally need to avoid people-generated particles, while aseptic areas additionally 
need to be protected where possible from microbial contamination. 
A careful selection of clean-room clothing, in terms of body coverage and 
particle-shedding ability, will significantly reduce both particle and microbial 
levels within the occupied clean room. Synthetic fibres generally shed lower 
particulate levels than cotton, and ceramic-coated synthetic materials are 
extensively used for clean-room clothing manufacture. However, fully covering 
synthetic-fibre clothing may enhance perspiration and thus microbial release, so 
high-specification clean areas should be operated at slightly lower temperatures 
to compensate. 
The frequency of personnel movement within secondary production areas is 
not generally problematic, bearing in mind the relatively small numbers and the 
confined nature of the operations. However, the increased load on changing 
facilities at break times should not be overlooked, and neither should the ease of 
movement during any emergency condition. This is particularly important with 
clean areas, where many restrictions on movement occur such as the use of 
multiple doors, changing room step-over barriers etc., and where over-ambitious 
attempts to seal emergency doors to prevent dirt ingress have been known 
to result in blocked escape routes. 
6.2.5 Materials movement 
It is often the case that, along with personnel movement, material movement 
within pharmaceutical manufacturing facilities dominates the consideration of 
building layout. The separation of material and personnel pathways and the 
avoidance of cross-overs can consume a large amount of time during initial
planning. While such considerations are important, the extent of such importance 
must be first established by the performance of a movement review, which 
in turn requires a full understanding of the operation of the facility and the type, 
size and frequency of movements. It will often be found that the problems 
envisaged are imaginary, and bearing in mind the low-volume nature of most 
pharmaceutical products this should not be surprising. 
The pharmaceutical engineer must, therefore, play a key role in establishing 
the realities of building layout design and ensure that he/she provides advice to 
those with whom he/she is working. 
Once the understanding of material movement is established, consideration 
can be given to key factors such as corridor width, door width and type, and the 
adjacency of related operational areas. 
One key item sometimes overlooked in preliminary planning is equipment, 
both fixed and mobile. Care must be taken in layout design and equipment 
selection to ensure that larger equipment can be moved through the facility to 
its final operational position, and that routinely-mobile items have transport 
routes which have been planned with their movement in mind. 
6.2.6 Specialized environments 
Where products demand special environmental conditions, the building structure 
and layout should include separate spaces for their manufacture and/or 
storage. 
In general, these special environments either have increased cleanliness, 
unusual temperature or humidity, or provide extra levels of separation from 
surrounding areas by virtue of high potency or other risks. For example, aseptic 
conditions are required for the manufacture of injectable forms, demanding 
higher standards of surface cleanability and ventilation air filtration efficiency. 
These features must be used in conjunction with stricter operator clothing 
regimes and closely defined operating/handling procedures. Layout considerations 
must include provision for separation from lower-grade areas by means of 
air locks. Positive pressurization of the processing areas is of course necessary 
to prevent ingress of dirt and microorganisms. 
An important feature of aseptic processing areas is the selection of structure 
and finishes. It is not uncommon in modern facilities to employ modular 
partitioning systems with close-tolerance self-finished panels. These have the 
advantage of providing crevice-free stable walls and ceilings which do not move 
or crack, even when the main building structure surrounding the area is liable to 
move due to thermal expansion/contraction. In conjunction with heavy-duty 
clean-area grade welded vinyl flooring systems, these modular clean rooms 
provide reliable and easily maintained surfaces ideal for aseptic operations.
They are, however, relatively expensive, and a lower-cost alternative is the 
use of steel-frame and plasterboard systems for walls and ceilings, coupled with 
vinyl flooring. This approach also provides a good-quality environment, but 
requires higher levels of maintenance attention due to joint-cracking potential 
and less-durable surface finishes. 
Where products are especially temperature or moisture-sensitive, the rooms 
in which they are exposed to the operating environment need to be supplied 
with ventilation air which has been conditioned to the required levels. This 
requirement may not affect the layout of the area concerned, although air locks 
coupled with positive room pressurization may be included to ensure greater 
control of the special environment. However, it will demand changes to the air 
handling system, and this is typically achieved by localized heating, cooling or 
dehumidification of the supply air. 
Care must be taken when humidity levels are unusually low (below 
20% RH), as operating staff may suffer dehydrating effects such as sore throats 
and cracked lips, which may be avoided by reduced individual working periods 
in the areas concerned. 
Where product materials of an active nature are exposed in-process, operating 
personnel may be protected by personal protective equipment, provided the 
exposure is of short duration (for example, during maintenance or product 
transfers). Alternatively, isolation/barrier methods should be employed to 
prevent such exposure. However, under either scenario it is possible that 
product dust may be emitted, and the rooms involved should be designed to 
take account of this possibility by the use of negative pressurization and the 
inclusion of air locks. It also requires consideration of room exhaust air filtration 
to protect the external environment, preferably sited at the room wall or ceiling 
interface. Such filtration systems should include a method by which the exhaust 
filters can be changed from within the room in a safe manner, personnel 
involved being protected by temporary personal protective equipment. 
An additional desirable feature of active product processing areas is easyclean 
surfaces for walls, floors and ceilings. This is essential to ensure 
containment. 
6.2.7 Internal structure 
Certain products and processes demand special consideration of construction 
and finishes. However, it is a general requirement for pharmaceutical production 
and storage areas that they should be easily maintainable in a clean 
condition, and walls, floors and ceilings, together with pipework, ductwork and 
electrical features should be designed with this in mind.
It is first necessary to consider the degree of product exposure at each stage 
of the storage, dispensing or production process, and to consider the risks to the 
product from such exposure. This analysis will provide a framework for the 
selection of surface finishes in each area. Thus, the movement and storage of 
materials that are always enclosed in sealed containers requires a very different 
selection of surfaces from that needed where sterile materials are filled into 
ampoules under aseptic conditions. 
Provided that the need for cleanability in all areas of pharmaceutical 
manufacturing plants is ensured, a variety of surface finishes are available for 
selection. These range from painted blockwork walls, sealed concrete floors 
and insulated and plastic-faced liners to ceilings in warehouses, to fully 
sealed and crevice-free clean-room systems with coved interface joints in 
sterile areas. 
In production areas it is generally wise to avoid the use of painted 
blockwork, in favour of a plastered and painted finish. It is also best to 
avoid suspended ceilings with lay-in tiles, as these do not provide effective 
barriers between processing areas and the technical/services areas above 
and may allow dirt ingress. In such areas it is also preferable to provide 
access to services distribution and plantroom areas, which does not 
involve direct penetration of the walls or ceilings of the operating spaces 
themselves. 
6.2.8 Building structure 
The choice of structural materials can affect the internal environmental 
conditions. An example of this is the effect of external environmental conditions 
on natural expansion/contraction of the building's structural material. 
Steel framed buildings will naturally provide greater potential for such movement 
than those fabricated from concrete or similar materials. In any case, these 
natural movements must be taken into account in the structural design of the 
building, and the presence of expansion joints in walls, floors and ceilings may 
be the consequence. Wherever possible, such joints should be avoided in 
manufacturing areas, except in the case of aseptic processing rooms where they 
must be avoided. 
6.3 The operating environment 
6.3.1 Introduction 
As a consequence of the increasing regulatory pressures being exerted on the 
industry, the environment in which secondary production is undertaken has
become progressively subject to greater inspection by the authorities. The 
'environment' covers a number of issues, each of which is covered in the 
following sections: 
the avoidance of cross-contamination; 
product segregation; 
cleaning; 
environmental classification; 
ventilation systems; 
surface finishes; 
lighting selection. 
The art of providing the correct operating environment lies in the selection 
of the systems that provide, as a minimum, no greater risk of contamination of 
the end product than has been accepted by the authorities during the drug 
approval process. This requires the engineer to select systems that meet this 
standard and are: 
economically justified; 
operable; 
maintainable (to the 'as-new' conditions); 
to cGMP standards. 
6.3.2 Avoiding cross-contamination 
All customers wish to receive exactly what they have ordered (or been 
prescribed, in the case of patients). Failure to do so can have unacceptable, 
even fatal, results. However, even if the product is correctly delivered, without 
proper controls it can be contaminated with another material. This potentially 
can have severe side effects, particularly if the patient suffers from a reaction to 
the contaminating material. Clearly, if the potential contaminant is another 
pharmaceutically active material or a viable organism, cross-contamination 
must be rigorously controlled. 
The most likely sources of cross-contamination are: 
the operator; 
the previous batch of material; 
other materials in the working environment (such as paint, dust, microorganisms, 
implements). 
Sources of contamination can be identified and the level of risk determined 
for each product. However the industry has established a number of 
standard practices to reduce the contamination risks at all times. These 
'standards' are commonly described as part of 'current' Good Manufacturing
Practice (cGMP) and are not always available in written form, although many 
guides have been published. The following paragraphs identify the main 
sources of cross-contamination in secondary pharmaceutical manufacture. 
The operator brings to work several sources of contamination. External 
contaminants, such as soil, clothing fibres, etc. can be removed by the use of 
personal hygiene techniques on arrival at work and the wearing of non particle 
shedding clothing for production duties. Personal contaminants, such as dead 
skin scales and living organisms on the skin surface or in exhaled air, cannot be 
eliminated but risks from them should be reduced when these are known to be 
hazardous to products. 
It is normal practice for all operators to change into clothing suitable for 
their duties on arrival at the manufacturing plant. Except in the lowest classes of 
operating environment, operators will change all their external clothing for 
'coverall' man-made fibre working clothes, wear dedicated shoes and cover 
their hair and ears with a fine mesh hair cover. Those with beards/moustaches 
may be required to use a beard 'snood'. 
As the quality of the environment increases, the standard of clothing and 
other protective coverings will increase. It is becoming standard practice, 
therefore, for manufacturing plants to have a series of change requirements 
to match the operating conditions. In the extreme — aseptic production — 
areas, operators have only their eyes exposed to the environment. Operators 
with infections are not permitted to work in these aseptic conditions, as the 
risk to the product is too high, even when protected by further containment 
methods. 
Training is the principle method by which operators can learn to avoid the 
risk of creating cross-contamination. It is essential that they fully understand 
the need for absolute adherence to the Standard Operating Procedures (SOP), 
which have been developed to reduce the risk to the product in manufacture. 
Strict compliance with the clothing disciplines is required to avoid bringing 
contamination from one product to another on their clothes/skin. Learning to 
work at a pace that does not create excessive particulate disturbance requires 
skill and practice, particularly over exposed product. 
It should never be forgotten that the human body loses particles of skin 
throughout the working day (see Table 6.1). These particles can become the 
chief source of contamination in a clean working environment. 
The previous batch of material will always be a source of cross-contamination. 
Only when the previous batch is made from exactly the same 
components does this create no risk. 
Segregation of products and the cleaning procedures required to avoid 
contamination are discussed in Section 6.3.3 (see page 163).
Unless a process is undertaken in a totally contained manner, it can be 
assumed that the materials utilized in the manufacture of a product will be in the 
manufacturing environment. This is caused by many sources, but normally 
from particulate escapes, aerosols of liquids and from operators' clothing. 
Methods of handling these materials can significantly reduce their discharge 
to the environment and the training of the operator is essential in the reduction 
of contamination risk from these sources. 
Cleaning of the equipment and the surrounding areas can clearly reduce the 
level of contamination risk, but the need for excessive cleaning regimes should 
be avoided. Careful planning of production batches can reduce, or even 
eliminate, the need for cleaning between batches. Excipients (non-active 
ingredients) may be used in many formulations and, therefore, cleaning 
between batches of different products using the same active ingredient may 
be reduced in scope. 
In summary, the risk of cross-contamination from a previous batch must be 
understood and reduced to an acceptable level. 
Other materials can be present in the environment and not be caused 
directly by the operator or the previous batch of materials. A main source of 
such contaminants is the poor design of the premises in which the operations 
are undertaken. Information is given on surface finishes later in this chapter, but 
the particle shedding properties of all surfaces can be a source of contamination 
when the process materials are exposed to the environment. 
Of more importance is the elimination of any surface on which contaminants 
can collect and later fall into the process. Flat surfaces should be replaced by 
sloping faces of easily cleaned materials; fixed equipment should be enclosed 
and ideally sealed to the ceiling; doors and windows do not require architraves; 
controls should be built into the walls or equipment; lights should be sealed to 
their surrounds; service outlets should be designed with minimum exposed 
surfaces. 
Table 6.1 Release of human skin flakes to the environment 
Activity Flakes released per minute 
Sitting still 100,000 
Moving limbs gently 500,000 
Moving limbs actively 1,000,000 
Standing up/sitting down 2,500,000 
Walking/climbing stairs 10,000,000
Two further important sources of contamination should be considered in the 
design of all facilities with risks reduced to a minimum: 
the movement of air; 
the movement of process materials. 
Most modern pharmaceutical premises are provided with air handling plants 
that supply a controlled volume of air to each process area. Correct specification 
and installation of the air system is essential to ensure an acceptable level 
of contamination of the air supply into a process area. Additionally, air 
movements between process areas can carry contaminating particles. This 
risk has to be considered for each process area and solutions found, usually by 
air locks, to prevent particulate movement between areas. 
All process materials have to come from outside the process area at some 
stage. Liquids can be piped directly to a process without external contact, but 
dry materials have to be transported. If this transport involves movement 
between areas, the facility design and process operations have to assume that 
other spilt materials can contaminate the materials. Cleaning regimes on entry 
to a process area will need to be agreed at an early stage in the process design. 
6.3.3 Product segregation 
Product segregation is needed to avoid contamination by another product. This 
would ideally be by installing separate facilities for each product, but this can 
rarely be achieved due to prohibitive capital costs. The industry has, therefore, 
adopted a number of universally applied segregation techniques: 
Do not produce high-risk products in the same facilities as low-risk products. 
Antibiotics are always manufactured in facilities designed to produce only 
this type of product, as historically, patients have suffered reactions from 
cross-contamination of low-risk products by antibiotics. Hormonal products 
are normally manufactured in dedicated facilities for this potentially highly 
active material. Segregation allows specific cleaning and materials handling 
technology to be used in a dedicated manner as well as specific operator 
protection and training. 
Manufacture products requiring the same environmental standards in one 
area. Products at high risk of contamination, such as sterile products, require 
far higher quality environments and the cleaning regimes are more stringent. 
These areas should be kept to a minimum. Operators need special purpose 
clothing (to protect the product) and training to work in these areas.
Dedicate an area to the production of one product at a time and ensure that 
the area and equipment are thoroughly cleaned before commencing the 
manufacture of a new product in the same area. 
Contain the production process, ideally within the manufacturing equipment. 
Where this is not possible, use airflow (laminar airflow or local extract) and 
enclosures to retain product spillage within the smallest possible area. 
Establish fully validated cleaning regimes for each product in each area/ 
equipment item. It is essential to know, and be able to demonstrate, that the 
production area and equipment is clean at the end of a production run. 
'Clean', in this context, means that trace elements of the previous product 
left behind after the cleaning process are below acceptable limits. 
Product segregation is therefore the practice of 'avoidance'. By avoiding the 
factors that cause cross-contamination between products, the risks are reduced 
to an acceptable level. For example: 
keep different products in separate locations; 
ensure that labelling clearly identifies the product and its components; 
never manufacture one product in the presence of another; 
prepare standard operating procedures that do not create a risk of cross 
product contamination; 
use clean equipment at the start of a new production run; 
identify the risks of cross product contamination (e.g. operator's clothing) 
and reduce these risks; 
train operators in the use of equipment and production processes; 
audit the production processes to ensure conformity. 
6.3.4 Cleaning 
Equipment 
Emphasis has been placed on the need to avoid cross-contamination between 
products. The major source of such contamination, if not removed by cleaning, 
is the equipment in which the product is prepared, closely followed by sources 
outside the equipment. 
It is not sufficient just to clean the equipment and assume that any risk of 
contamination has been removed. Every individual operator would use their 
own method of cleaning if they were not trained. Their individual methods will 
vary from time to time and there is no guarantee that any of the operators' 
methods will provide cleaning to the standards required to reduce the risk of 
contamination to an acceptable minimum.
It is critical to establish cleaning procedures that can be repeated consistently. 
Different procedures may have to be established for each product and all 
the cleaning procedures have to be validated for effectiveness. 
Manual methods of cleaning cannot be guaranteed to be one hundred percent 
effective unless by 'overkill'. Mechanical means of cleaning, however, can be 
accurately reproduced on demand. For this reason, modern pharmaceutical 
plants are normally designed with 'in-built' Clean In Place (CIP) capability. 
CIP technology, established in the brewing industry, is based on the 
combination of chemical/detergent action and mechanical action (from the 
effect of direct impact on, or flow of water over, surfaces). The sequence 
normally utilized consists of: 
initial hot or cold rinse to remove gross contamination; 
caustic detergent rinse to remove adhering materials; 
hot or cold water rinse; 
neutralizing acid rinse (if required); 
hot or cold water rinse; 
final water rinse of a quality equivalent to that used in the process. 
Water quality is a critical factor in CIP systems and any possibility of 
contaminants being introduced by water from the cleaning process must be 
eliminated. For this reason, de-ionized water to USP23 or BP is normally used 
throughout the CIP sequence with a final rinse of Purified Water or Water for 
Injection quality if the process demands this standard of cleanliness. 
CIP systems are normally controlled by automatic sequence rather than 
manual operation. 
Large surfaces to be cleaned by CIP systems require the use of mechanical 
devices, such as spray heads, and an understanding of the 'shadow' effects 
created by internal fittings. Specialist companies supply both the equipment 
and 'know-how' for this technology. 
Although cleaning by direct impact using spray heads can be designed into 
process equipment, the interconnecting pipework can only be cleaned by the 
flow of water and chemicals over the surfaces. Experience indicates that 
turbulent flow is required to provide maximum cleaning effect. This turbulent 
flow is normally created by flow rates at or above 1.5 ms~l and the design of a 
CIP system should ensure that all process pipework is subject to this minimum 
flow rate. 
The duration of flow of CIP fluids is determined by examination of the 
effects of the CIP process on the system. Access is, therefore, required to all 
cleaned surfaces during the validation of the cleaning process.
For this reason, most process pipework installations subject to a CIP system 
are designed to be taken apart on an agreed schedule to enable the cleaning 
procedures to be re-validated. 
Materials of construction are frequently fabricated to a higher standard than 
is required by the process, to enable the cleaning procedure to be fully effective. 
Contamination sources outside the equipment can be eliminated by the 
total containment of the process. For many reasons, the design of pharmaceutical 
processes cannot always permit this ideal arrangement and, in practice, many 
sources of contamination will exist that have to be controlled during production. 
The following brief paragraphs aim to give an indication of some of the chief 
contamination sources that are created by normal operation of a process, and 
the techniques for avoiding these are outlined. 
(a) Materials received into the facility from outside sources 
These are expected to be contaminated by any material normally present during 
transport and materials handling operations. Normally all such materials are 
double wrapped (plastic linings inside outer containers) and are frequently 
over-wrapped by stretch film. Cleaning, other than gross contamination, will be 
left until the material is to be used. 
(b) Sampling 
This is undertaken of all incoming materials and requires the breach of the 
materials containment system. For this reason, sampling is undertaken within a 
sampling booth and the material containers will be cleaned externally before 
entry into the booth. The inner and outer containers will be resealed before 
return to storage. 
(c) Storage and internal transport 
These will not normally provide a severe risk of contamination, but all inner 
and outer containers must be kept sealed. Again, the outer containers will be 
cleaned before entry into the production area. 
(d) Dispensing operations 
This is naturally a dusty operation when dealing with dry materials. Contamination 
of other materials from this dust must be reduced to a minimum by 
cleaning the dispensary area. It should be noted, however, that cleaning between 
the dispensing of different materials for the same product is normally only on a 
limited housekeeping basis.
(e) Charging/discharging operations 
Transfer into and from process equipment is normally dust free for operator 
safety reasons. Where, however, this operation is not dust free, the resultant 
dust spillage can be expected to contaminate all surfaces in the operating room 
as well as the operator. The operating area must be thoroughly cleaned on the 
completion of a production run, or at least once a week. 
(f) The operator 
The operator has freedom of choice in where to go and what to do. This 
freedom has to be strictly controlled, with high quality training provided and 
absolute discipline exercised to prevent the transfer of contaminating products 
between different process operations. Current practice indicates use of specific 
clothing for each production room and personal cleaning regimes on leaving 
the room. These cleaning regimes may be as limited as an external clothing 
change or as severe as air showers or water deluges, depending on the nature of 
the product and the company's policies. 
(g) Processing equipment 
This is normally selected to be non-particle shedding and, therefore, is not 
considered to provide a contamination risk. Care should be taken over new or 
maintained equipment that can be delivered with surface contamination 
invisible to the naked eye. 
(h) Room fabric 
This includes walls, floors, ceilings, doors, service entries, lights, etc. All have 
to be carefully chosen to avoid particle shedding characteristics and have easily 
cleaned surfaces. Ledges should be designed out of the room, but where 
unavoidable, should be sloped to prevent dust traps. 
(i) Air handling systems 
These bring a continuous source of replacement air to the operating environment. 
Care must be taken in the design of the air handling plant, equipment and, 
particularly, filters to prevent external contaminants being carried into the 
operations. The following sub-section provides information on the environments 
that have been found to be acceptable for pharmaceutical production. 
6.3.5 Environmental classification 
Pharmaceutical environments are classified by the number of particles of 
specific sizes contained in a measured volume of air, together with requirements 
for temperature and humidity. The information in this section is on 
European and United States requirements.
The most easily understood classification comes from US Federal Standard 
209D (Table 6.2) and, although theoretically superseded, is still in extensive 
use. It is based on imperial measurements. 
This Federal Standard has been updated to version 209E by conversion to SI 
units of measurement (see Table 6.3). 
FS 209E permits the continuing use of 'English' terminology although SI 
units are preferred. Of particular importance in the Federal Standard is the need 
to specify and measure particle counts as either 'as-built' (no operators or 
equipment present), 'at rest' (equipment installed, but no operators present) or 
'in operation' (equipment in use and operators present). 
Table 6.3 United States Federal Standard 209E — air classifications 
Class name 
SI 
Ml 
M1.5 
M2 
M2.5 
M3 
M3.5 
M4 
M4.5 
M5 
M5.5 
M6 
M6.5 
M7 
English
1 
10 
100 
1000 
10,000 
100,000 
Class limits (volume units) 
0.1mm 
(m3) 
350 
1240 
3500 
12,400 
35,000 
(ft3) 
9.91 
35.0 
99.1 
350 
991 
0.2 mm 
(m3) 
75.7 
265 
757 
2650 
7570 
26,500 
75,700 
(ft3) 
2.14 
7.5 
21.4 
75 
214 
750 
2140 
0.3 mm 
(m3) 
30.9 
106 
309 
1060 
3090 
10,600 
30,900 
(ft3) 
0.875 
3.00 
8.75 
30.0 
87.5 
300 
875 
0.5 mm 
(m3) 
10.0 
35.3 
100 
353 
1000 
3530 
10,000 
35,300 
100,000 
353,000 
1,000,000 
3,530,000 
10,000,000 
(ft3) 
0.283 
1.00 
2.83 
10.0 
28.3 
100 
283 
1000 
2830 
10,000 
28,300 
100,000 
283,000 
5.0 mm 
(m3) 
247 
618 
2470 
6180 
24,700 
61,800 
(ft3) 
7.00 
17.5 
70.0 
175 
700 
1750 
Table 6.2 United States Federal Standard 209D — air classifications 
Class 
1 
10 
100 
1000 
10,000 
100,000 
Class limits in particles per cubic foot of size/ 
particle sizes shown (micrometers) 
0.1mm 
35 
350 
NA 
NA 
NA 
NA 
0.2 mm 
7.5 
75 
750 
NA 
NA 
NA 
0.3 mm 
3 
30 
300 
NA 
NA 
NA 
0.5 mm
1 
10 
100 
1000 
10,000 
100,000 
5.0 mm 
NA 
NA 
NA 
7 
70 
700
In all cases, services must be functional. 
There are a number of European Standards available based on national 
standards. The European Directives that created The Rules Governing Medicinal 
Products in the European Community' cover air classification systems for 
the manufacture of sterile products (see Table 6.4). These classifications are 
now considered as the established European standard and, for members of the 
EEC, are legal requirements. 
In the 'Rules and Guidance for Pharmaceutical Manufacturers 1997' 
prepared by the MCA, a similar table is published for sterile production that 
gives further guidance between the 'at rest' and 'in operation' conditions 
(see Table 6.5). 
For these airborne particulate classifications, the MCA also publish a table 
giving recommended limits for microbiological monitoring of clean areas 'in 
operation', (see Table 6.6, page 170). 
Table 6.4 Air classification system for manufacture of sterile products 
Grade 
A Laminar air flow 
work station 
B
CD 
Max permitted number 
of particles per m3 
equal to or above 
0.5 mm 
3500 
3500 
350,000 
3,500,000 
5.0 mm 
None 
None 
2000 
20,000 
Max permitted number 
of viable micro-organisms 
per m3 
Less than 1 
5 
100 
500 
Extract from The Rules Governing Medicinal Products in The European Community. 
Note that class A refers to the air classification around the exposed product, whilst 
class B refers to the background environment. 
Table 6.5 Airborne particulate classifications — MCA guidelines 
Grade 
AB
C
D 
Maximum permitted number of particles per m3 equal to or above 
At rest 
0.5 mm 
3500 
3500 
350,000 
3,500,000 
5.0 mm 
0
0 
2000 
20,000 
In operation 
0.5 mm 
3500 
350,000 
3,500,000 
Not defined 
5.0 mm
0 
2000 
20,000 
Not defined
The participate classifications in use are normally referenced by either the 
FS 209D system (100, 10,000, etc.) or by the EEC rules (A, B, etc.). These two 
classifications correspond approximately and both are accepted by the 
regulatory authorities. In summary, Table 6.7 provides a brief check for 
the user. 
It is recommended that the designer specify the particulate levels in the 'at 
rest' condition. In addition to the particulate levels, room operating conditions 
of temperature, humidity and pressure must be specified. 
Humidity creates contamination risk to the product from condensation, 
absorption and human perspiration. It is, therefore, normal practice to maintain 
the operating conditions at 45% to 55% relative humidity. 
Where a product is expected to absorb water from the environment, such 
as effervescent tablets, hard gelatine capsules, etc., the humidity has to 
be reduced. The humidity has to be controlled at a level that is acceptable 
Table 6.7 Approximate equivalent international standards 
MCA 
guidelines 
1997 
AB 
CD 
FS 209D 
1988 
1 
10 
100 
100 
1000 
10,000 
100,000 
FS 209E 
1992 
M1.5 
M2.5 
M3.5 
M3.5 
M4.5 
M5.5 
M6.5 
ECC 
rules 
1992 
AB 
CD 
Germany 
VDI 2083 
1990 
1
23
3
4
5
6 
UK 
BS 5295 
1989 
CD
E or F 
E or F 
G or H 
JK 
ISO 14644 
Parti 
Draft 
3
4
5
5
67
8 
Table 6.6 Recommended limits for microbial contamination (average values) — MCA 
guidelines 
Grade 
AB
CD 
Air sample 
cfu m3 
<1 
10 
100 
200 
Settle plates 
(diam. 90 mm), 
cfu/4 hours 
<1
5 
50 
100 
Contact plates 
(diam. 55 mm), 
cfu/plate 
<1
5 
25 
50 
Glove print 
5 fingers, 
cfu/glove 
<1
5 
Note that all the above tables are published with comprehensive notes. It is important that 
these notes are fully understood before proceeding with the design of the environment.
to the operator as well as avoiding risk to the product. In extreme cases, it will 
be necessary to provide the operator with a breathing air supply. 
Temperature should normally be maintained at a level that permits the 
operator to work in comfort. The air supply temperature should allow for heat 
gains from all sources within the operating area. Many alternative methods 
of temperature control are available and the designer should seek expert advice. It 
is essential, however, to maintain the room temperature within the specified — 
and validated — limits over the full range of operational conditions. 
Where production has to be undertaken at temperatures normally unacceptable 
to the operator, e.g., cold rooms, then protective clothing should be provided. 
Pressure differentials are an essential part of the design of a clean room 
facility. To protect a product from contamination from outside sources, it is 
normal practice to pressurize the rooms in which the product is exposed to the 
environment. Where a sequence of operating rooms is installed, pressure 
'cascades' are frequently used so that the most sensitive areas are at the highest 
pressure and the least sensitive at, or just above, atmospheric pressure. This 
situation is most frequently present in aseptic operations. 
Where the product concerned is of high potency, negative pressure is used to 
contain the hazard to within the operating area. The risk from external 
contamination is usually reduced by surrounding the negative pressure room 
with other areas (e.g. changing rooms) at positive pressure. 
The most commonly used pressure differential is 15 Pa. 
6.3.6 Ventilation systems 
Ventilation systems designed into any secondary pharmaceutical facility need 
to be carefully designed, installed, controlled and operated. The designer should 
consult with experts in this field to achieve the desired conditions within the 
process areas, but the following paragraphs give some general guidance. 
The environmental standards specified within any operating area must be 
maintained to those standards at all times when process operations are active. 
At no time should the product, or the surfaces with which the product comes 
into contact, be exposed to environmental conditions that may cause unacceptable 
contamination. In practice, this means that ventilation systems will be 
fully operational for the majority of the time and only revert to night/weekend 
operation when all risks of contamination have been contained. 
Assuming that the ventilation system has been correctly designed and 
installed, the system should not provide any significant source of contamination. 
This is achieved by both filtration of the air supply and monitoring and 
control of the pressure, temperature and humidity in each operating area.
Each area will have been commissioned against a specification that meets 
the environmental classification for the product being made and the area will be 
monitored on a regular basis for maintenance of this classification. Any 
deviation has to be reported and action taken. Significant deviation from 
acceptable limits will result in cessation of production. 
To prevent this extreme situation, ventilation systems are normally designed 
to meet the following criteria: 
Class A: Laminar airflow through terminal HEPA (High Efficiency Particulate 
Air) filters at a velocity of 0.45 m s~1 ± 20% at the working position 
(MCA guidance) with low-level extract. In all cases, operations at Class A 
should be contained within a purpose-designed workstation with no operator 
access other than gloved hands. 
Class B: Downward airflow through terminal HEPA filters with low-level 
extract. The operator will be working and creating high particle counts in this 
area. Air volumes should be sized to ensure that particulate conditions for the 
'at rest' state will be achieved in the unmanned state after a short 'clean-up' 
period of 15-20 minutes. 
To ensure that the air movement is able to clean up the working area, current 
designs now utilize turbulent air movement delivered by purpose designed 
diffusers. 
Class C: Airflow provided through (normally terminal) HEPA filters with air 
movement of sufficient volume to maintain the classification of the area. There 
is considerable debate on the use of low-level extract for Class C areas, but 
there is no specific requirement. Air volumes should be sized to ensure that 
particulate conditions for the 'at rest' state would be achieved in the unmanned 
state after a short 'clean-up' period of 15-20 minutes. 
The higher cost of installing low-level extracts needs to be considered 
against the risks created by moving particles in the air stream over the entire 
working area when high-level extracts are used. 
Class D: Airflow provided through filters (normally HEPA) with air movement 
of sufficient volume to maintain the classification of the area. High-level 
extract is the usual installation for this classification. Air volumes should be 
sized to ensure that particulate conditions for the 'at rest' state would be 
achieved in the unmanned state after a short 'clean-up' period of 15-20 
minutes. 
Where possible, air movement should be designed to flow downward over 
any exposed product to avoid particulate entrainment being carried over the 
product. 
In areas where the majority of operations only require a minimal environmental 
classification, it is acceptable to provide higher local environmental
conditions by use of air curtains. A good example of this method of protection 
can be found in many packing halls, where the general area will be to Class D, 
but local conditions around the product at the filling head will be to Class C. 
HEPA filters are normally used to achieve the stated environmental classifications. 
Within Europe, the grades of HEPA filter are distinguished by the use of 
EU classifications, each of which has a known retention efficiency at 0.3 mm (see 
Table 6.8). 
In the USA, HEPA filters are required to have efficiencies of 99.97% 
(EU12 and greater). 
Not only is it essential that the filter specifications meet the requirements of 
the environment, but also that the installation does not compromise the filter 
integrity. This can be caused through damage to the filter medium, or through 
passage of unfiltered air between the medium and its frame, or between the 
frame and the air supply system. Assurance of the integrity of an installed filter 
system must be subject to an 'in-situ' integrity test. 
6.3.7 Surface finishes 
Throughout this section, emphasis has been placed on the avoidance of possible 
contamination of the product. Consideration has been given to sources of 
contamination from outside the operating environment but it is equally 
important to appreciate that the fabric of the area and the equipment in 
which the product is produced, can itself contaminate the product. 
All materials of construction should be non-particle shedding. Traditional 
building materials must, therefore, be sealed by the application of a surface 
coating. Current practice is to use a two part epoxy coating (or equivalent) that 
provides both an abrasion resistant surface and a sufficient degree of elasticity 
to avoid minor wall movements opening up hair line cracks, thus permitting 
particulate escape. 
The use of partition systems has become widespread and several alternative 
systems are available. These systems, although more expensive, eliminate the 
Table 6.8 Classification of retention efficiencies of HEPA 
filters 
Eurovent classification 
EUlO 
EUIl 
EU12 
EUl 3 
EU14 
Efficiency at 0.3 mm (%) 
>95-<99.9 
>99.9-< 99.97 
>99.97-< 99.99 
>99.99-< 99.999 
>99.999
wet building trades and provide an acceptable pharmaceutical finish with no 
further surface treatment. 
Joints in wall and ceiling construction are normally filled with a silicone 
sealant that permits some building movement without any crevices forming. 
For ease of cleaning, joints between walls and floors are always coved in any 
area in which product is exposed. Current practice is to cove at wall to wall 
joints in Class C areas and also wall to ceiling joints in Class A/B areas. 
Floors present a more difficult choice, as they have to accept movement of 
heavy loads, building settlement and movement as well as possible damage 
from containers, etc. Currently, epoxy floor coatings up to 6 mm thick are 
proving successful, but their expense limits their use to the more severely 
loaded areas. Vinyl floor, wall and ceiling coverings are an acceptable 
solution — reserved for lightly loaded areas and are the material of choice in 
Class A/B areas for many manufacturers. 
In selecting materials of construction for the building elements, thought 
must also be given to the damage that may be caused by the normal daily 
operations, such as trucks and pallets hitting walls. Where such damage would 
expose particulates, wall protection is usually provided. 
The cleaning regimes in the production environment normally involve 
wetting the surfaces of the area. In the controlled environment, these conditions 
provide excellent sources for microbial growth and it is, therefore, important to 
ensure that surface finishes do not support microbial growth. 
Process equipment comes into intimate contact with the product and, 
therefore, the materials of construction are of most significance. Non-corroding 
materials are essential, not only to prevent contamination of the product, but 
also to stop any damage to the surface finish of the equipment. 
A poor surface finish harbours crevices that can support microbial growth 
and traces of previous products and cleaning agents. For this reason, emphasis 
is placed on the specification of surface finishes and the methods by which they 
are prepared. 
The great majority of pharmaceutical process equipment is fabricated from 
316 or 316L stainless steel because of the non-corrosive nature of the material 
for most products and the ease with which it can be given a high quality surface 
finish. The surface finishes are normally specified (as Ra — average roughness) 
in either micro inches or microns. The polishing medium grit size should 
not be used as an indication of the surface finish. 
Individual producers of stainless steel equipment will use both mechanical 
and electro polishing methods. Electro polishing gives a higher quality look to 
the surface and provides a more rounded edge to the microscopic grooves in the
polished steel. This more desirable finish is, however, more expensive than 
mechanical polishing. 
The selection of the surface finish is determined by: 
existing standards within a facility; 
end user preference; 
the need for a reduction in crevice size to reduce microbial growth; 
cost. 
Table 6.9 lists surface finishes specified for stainless steel equipment. 
6.3.8 Lighting selection 
Apart from the need to ensure a safe working environment, the regulatory 
authorities are interested in the lighting levels in a facility to ensure the 
manufacturing operations are undertaken without error. 
Although many operations in the modern pharmaceutical production facility 
are now automatically controlled, the operator still needs to oversee these 
operations. Frequently his work requires him to read Standard Operating 
Procedures and the slightest risk of error caused by misreading the instructions, 
instrumentation and alarms is not acceptable. 
Lighting selection must, therefore, ensure that the level of illumination is 
sufficient to read documentation, displays and instrumentation and that this 
does not cause operational difficulties from glare, reflection or too high an 
intensity. 
Designers of pharmaceutical facilities are recommended to take expert 
advice in the illumination specifications to ensure that all working areas are 
well lit throughout. 
Table 6.9 Polished finished on stainless steel sheet — Sillavan metal services 
Description 
Coarse grade 80 grit 
Coarse grade 180 grit 
Silk 
Supersilk 
Brush 
Bright buff 
Bright polish 
Mirror 
BS1449 No 
3A 
3B 
3B 
3B 
3B 
No 7 
No 7 
No 7 
Approx. jim 
Ra value 
2.5 
1.0 
0.4 
0.35 
0.2 
0.05 
0.05 
0.05 
Reflectivity % 
10 
10 
30 
30 
30 
48/55 
53/60 
58/63
6.4 Containment issues 
6.4.1 Operator protection 
Pharmaceutical manufacturing operations involve the handling of sophisticated 
chemical compounds, many of which can exhibit toxic effects on personnel 
handling them in concentrated quantities. Additionally, and often at the same 
time, pharmaceutical materials and products can suffer if exposed to the 
operating environment (for example, sterile products for injection). 
Operator protection can be provided by means of personal equipment 
(gloves, overalls, masks), while the creation of suitable macro-environments 
can provide aseptic facilities for injectable manufacture. However, the validity 
of these methods is questionable, and the use of techniques which enclose the 
product materials in a smaller space and provide means of remote operator 
access have become commonplace. These techniques are known as isolator or 
containment technology. Although the application of these methods differs 
between operator and product protection requirements, there are similarities in 
the equipment involved. 
6.4.2 Product protection 
A second application of containment technology is its use for the protection of 
products from environmental contamination. This application applies particularly 
to the aseptic manufacture of injectable or infusion products, which has 
traditionally been performed in high-quality environments conforming to 
Class 100 or better (to US Federal Standard 209E). The accepted approach is 
for the equipment and operations involved to be sited in Class 100 clean rooms, 
with localized enhanced protection to Class 10 being provided by fixed or 
mobile air supply units. The latter are designed to provide airflow of minimum 
turbulence (effectively 'laminar' flow when the units are unoccupied) so as to 
minimize particulate pick-up by the air steam in areas where sterile product or 
product-contacting components are present. 
This arrangement has been demonstrated over a period of twenty or more 
years to provide minimal validated risk of contamination, and this proven 
assurance has given rise to its use in the majority of modern pharmaceutical 
aseptic processing facilities. 
However, two undesirable features remain: 
the construction and operation of facilities reaching Class 100 conditions is 
expensive; 
there remains the possibility of human operator contact with product 
materials, with consequent risk of contamination.
Hence, recent developments of isolator technology have concentrated on the 
use of such equipment to provide a reliable localized barrier between the 
product and the operator, with the isolator forming a separate sealed environment 
of Class 100 or better, within which aseptic manipulations can be 
performed, either by hand using glove ports or automatically. 
Apart from the increased potential for reliable sterility, the use of isolators 
having a sealed high-grade internal environment has meant that the surrounding 
room space need not be to the same high standard. Current opinions 
differ on the desirable room environment quality, the regulatory view being 
based on Class 10,000, while some authorities among users and equipment 
manufacturers claim reliable validated operation at Class 100,000. Clearly, the 
capital and operating cost of such environments is lower than that of a Class 
100 suite. 
The isolator equipment commonly used for aseptic processing is sophisticated 
and by no means low cost, but it does allow lower cost surroundings while 
supplanting the need for localized laminar flow units and often filling machine 
guards. 
It is possible to link several machines for washing, sterilizing, filling, 
capping and sealing of injectable product containers within a set of linked 
barrier isolators or use a form-fill-seal technique 
6.5 Packaging operations 
6.5.1 Introduction 
The early days of pharmaceutical product packaging saw predominantly manual 
systems involving, for example, the hand counting of pills or tablets which were 
dispensed to the patient in a suitable container, often merely a paper bag! 
As demand and availability increased, the risk of mistakes became greater 
due to the wider range of products available and the frequency of dispensing. 
The same factors applied to the production of medicines, where centralization of 
manufacture led to multiple pack despatches. Increasing standardization led to: 
automated counting; 
pre-printed standard labelling; 
specific tested containers; 
secure capping/sealing; 
pre-printed cartons. 
Next Page
Hence, recent developments of isolator technology have concentrated on the 
use of such equipment to provide a reliable localized barrier between the 
product and the operator, with the isolator forming a separate sealed environment 
of Class 100 or better, within which aseptic manipulations can be 
performed, either by hand using glove ports or automatically. 
Apart from the increased potential for reliable sterility, the use of isolators 
having a sealed high-grade internal environment has meant that the surrounding 
room space need not be to the same high standard. Current opinions 
differ on the desirable room environment quality, the regulatory view being 
based on Class 10,000, while some authorities among users and equipment 
manufacturers claim reliable validated operation at Class 100,000. Clearly, the 
capital and operating cost of such environments is lower than that of a Class 
100 suite. 
The isolator equipment commonly used for aseptic processing is sophisticated 
and by no means low cost, but it does allow lower cost surroundings while 
supplanting the need for localized laminar flow units and often filling machine 
guards. 
It is possible to link several machines for washing, sterilizing, filling, 
capping and sealing of injectable product containers within a set of linked 
barrier isolators or use a form-fill-seal technique 
6.5 Packaging operations 
6.5.1 Introduction 
The early days of pharmaceutical product packaging saw predominantly manual 
systems involving, for example, the hand counting of pills or tablets which were 
dispensed to the patient in a suitable container, often merely a paper bag! 
As demand and availability increased, the risk of mistakes became greater 
due to the wider range of products available and the frequency of dispensing. 
The same factors applied to the production of medicines, where centralization of 
manufacture led to multiple pack despatches. Increasing standardization led to: 
automated counting; 
pre-printed standard labelling; 
specific tested containers; 
secure capping/sealing; 
pre-printed cartons. 
Previous Page
Much of this paralleled the growth of other consumer products, but the 
special security and safety requirements of medicines have extended pack 
features, which now include: 
tamper-evident closures; 
child-resistant closures; 
special protection against hostile shipping environments; 
security coding systems. 
The early manual assembly of packaged products has given way to 
progressively more-automated methods. Machines for counting unit dose 
products (such as tablets) and discharging the correct number into manuallypresented 
containers soon gave way to in-line counting, filling, capping, 
labelling and cartoning units linked by conveyors. These transport systems 
had gateing, accumulation and flow control elements built-in. Thus, the modern 
packaging line incorporates sophisticated handling and sensing equipment 
designed to minimize human intervention and eliminate human error. 
As seen later in this chapter, the structure of healthcare management 
arrangements is leading to increasingly sophisticated and patient-dedicated 
packaging, which curiously is taking developments full-circle and returning the 
objective back to the days of direct patient-specific dispensing. 
6.5.2 Tablets and capsules 
The packaging of solid unit-dose items is generally carried out in one of two 
ways. These utilize multiple-item containers (typically glass or plastic bottles) 
and blister packs. 
Bottle packs 
This packaging type utilizes containers with screw or press-on caps, containing 
either a single course of treatment, or larger types intended to be used for 
dispensing from, in order to produce such single courses. 
Methods of tablet/capsule counting range from photo-electronic sensing 
types to pre-formed discs or slats having a fixed number of cavities. 
All counting methods have potential inaccuracy due to the non-symmetrical 
shape of tablets and capsules and the possibility of broken tablets giving false 
counts. Individual tablets or capsules have low weight in comparison with the 
container, so that container weight variation can be greater than the weight of an 
individual item. Thus, post-filling check weighing methods cannot be relied 
upon to detect missing tablets/capsules in a container. 
As a result, modern counting machines are equipped with missing-item 
detection systems, utilizing infrared sensing or matrix camera technology.
Containers may be of either glass or plastic, but are increasingly of the 
latter as plastic materials with improved moisture-resistance have been developed. 
Capping systems have been designed which prevent non-evident pilferage 
or which are resistant to the attentions of young children. These benefits do, 
however, become disadvantageous when used for arthritic patients, who may 
have difficulty in opening the packs. 
Bottle packs have other disadvantages, namely: 
they offer no record of the dose having been taken; 
multiple-product treatment regimes mean the patient coping with several 
different containers; 
frequent pack opening may lead to product spoilage and risk of spillage; 
paper labels may become soiled, with risk of lost product identity. 
However, they have two significant advantages, being generally cheaper to 
produce and of smaller size than the equivalent blister pack. 
Blister packs 
These are produced by a form-fill-seal process using PVC or similar thermoplastic 
material in reel form as the blister material. For products having 
enhanced moisture sensitivity, plastics such as polyvinylidine chloride may 
be used. The blister cavities are formed from the thermoplastic film using 
heated die plates or drums, with plug or vacuum assistance. 
Tempered aluminium lidding foil with laminated plastic or an adhesive 
coating allows the two parts of the pack to be heat-sealed together. 
An alternative to plastic films for blister pack formation is the use of coldformed 
aluminium foil, which can offer improved product protection from 
moisture ingress. 
Blister forming methods include the use of continuous-motion cylindrical 
formers with blister cavities machined into them, or flat-platen types which 
cycle in a manner which matches the horizontal speed of the blister web, giving 
higher potential outputs. 
The sealing together of the filled blister and lidding foil is achieved by the 
concurrent flow of the two material streams followed by the application of heat 
and pressure using heated rollers or platens. 
Modern machines can operate at speeds of typically 400 blisters per minute, 
giving an equivalent tablet/capsule output of 4000 per minute for a ten-item 
blister.
A critical factor influencing machine output is the mechanism used 
for feeding the tablets/capsules into the formed blister cavities. Similar 
methods for detecting missing items to those used for bottle packs are 
employed. 
It is not uncommon for finished blister packs to contain more than one 
blister strip. This packaging method requires the blister form-fill-seal 
machine to incorporate a stacking/counting unit for the blisters, prior to 
carton insertion. 
6.5.3 Liquids 
Liquid Pharmaceuticals are packaged using either bottles or sachets, the latter 
being used for unit-dose applications. 
Bottles 
Early production systems for bottle filling were based upon manual dispensing 
from a bulk supply using a measuring container. As precision-moulded bottles 
became available and demand rose, methods of filling to a fixed level were 
established. Initially manual in operation, this approach was followed by a 
semi-automatic method in which the bottle was presented to a machine, which 
created a partial vacuum inside the bottle thereby encouraging the flow of liquid 
from a bulk tank or hopper. The liquid level rose in the bottle until it reached the 
height of the vacuum nozzle, when flow ceased. This vacuum method was 
developed for beverage production and is still used in some small companies. 
Manually presented level-fill systems led on to automated bottle movement 
and presentation, with consequent increases in output. Indeed, the basic 
technology is still used in high-speed beverage production. 
However, the fill-to-level method suffers from the disadvantage that the 
filled volume varies according to the accuracy of bottle moulding, making it 
relatively unsuitable for pharmaceutical product use. 
In consequence, modern pharmaceutical liquid packaging systems utilize 
volumetric measurement, either by means of adjustable-stroke piston pumps, or 
by positive-displacement rotary lobe-type pumps controlled by rotation sensors. 
High-speed dosing machines utilize 'diving nozzle' systems in order to 
reduce air entrainment and foaming problems (see Section 6.5.5 on page 182). 
Sachets 
Sachet packaging is mostly used for powders, which are then reconstituted with 
water or another suitable diluent by the end-user. However, a small number of 
examples exist of liquid-filled sachets. The pack is an ideal single-dose 
provision system. Sachets are formed from laminated foils, usually including
a plastic inner layer with aluminium foil centre laminate and an outer layer of 
paper that provides a printable surface. 
The sachets are formed as three-side sealed units prior to filling, and the final 
top seal is then applied, together with a batch/expiry date code. 
Sachet packaging is more common for non-pharmaceutical products, where 
outputs can be as high as 100 sachets per minute. 
The assembly/collation and cartoning methods of sachets are basically 
similar to those for tablet blister packs. 
6.5.4 Powders 
The powder is not a common finished dosage form for Pharmaceuticals, 
but it is frequently used for granule or powder formulation products that 
have low stability in solution (such as antibiotic syrups/suspensions for 
paediatric use). 
Products manufactured are typically in bottle or sachet form, the latter used 
for single-dose applications. 
Powder filling systems can be either volumetric or gravimetric. The former 
is most often typified by auger filling machines, in which a carefully designed 
screw rotates in a funnel-shaped hopper containing the product powder. As the 
auger rotates, the number of rotations determines the volume of powder 
delivered at the bottom outlet of the funnel and into the container. Rotation 
sensors are used to control this number so that the volume and hence weight 
dose is also controlled. 
A second volumetric system is the 'cup' type, in which a two-part telescopic 
cylindrical chamber is opened to the powder in a hopper and thus filled. The 
volume of this chamber is adjustable by varying its height telescopically. By 
rotating the position of the chamber between the powder hopper and a 
discharge chute, a controlled volume/weight of powder is discharged via the 
chute into the bottle or sachet. Automation of bottle or sachet feed allows 
relatively high output to be achieved. 
A key feature of all volumetric systems is the control of powder level in the 
hopper, as the height of product powder above the infeed to the dosage control 
system affects the bulk density of the powder and hence the weight dosed. 
A weight-dosing system can also be used for bottle filling. This method 
involves the automatic pre-weighing of the empty bottle followed by approximate 
dosing of typically 95% of the required fill weight (using an auger or cup 
filler). The partially filled bottle is then re-weighed and the weight compared 
with that of the empty bottle so as to allow calculation of the required top-up 
weight. The bottle finally passes under a top-up filler which delivers a 
calculated final amount to achieve the target weight.
The advantage of this approach is that the overall dosage accuracy can be 
greater, due to the finer control capability of the lower weight second/top-up 
dose. 
6.5.5 Creams and ointments 
These products are mostly filled into collapsible tubes, but occasionally into 
jars. The latter are filled and packed in much the same way as liquids. These 
semi-solids are also applied to impregnated tulles, although they are generally 
for burns treatment, where aseptically-produced versions apply. 
Tubes 
Tubes used for pharmaceutical preparations are either of the fully collapsible 
aluminium or aluminium/plastic laminate type, or are non-collapsible plastic. 
They are filled with product from the seal end before closing — the aluminium 
types being closed after filling by flattening and folding, while the plastic types 
are sealed by heat/impulse methods. 
Filling machines are usually of the rotary plate type, with empty tubes 
inserted into holders fixed into this plate from a magazine by means of an 
automatic system. On low-output machines, tube insertion may be performed 
by hand. 
The product is filled from a hopper via piston type dosing pumps through 
nozzles and into the tubes. These nozzles are often arranged so that they 'dive' 
into the empty tube and are withdrawn as the product is filled, a technique used 
to minimize air entrainment. The bulk product hopper is often stirred and 
heated, typically using a hot water filled jacket, in order to enhance product flow 
and uniformity. 
Empty tubes are usually pre-printed with product information. This print 
includes a registration mark which allows the filling machine to sense the 
orientation of the tube, and rotate it prior to sealing so that the product name or 
details are conveniently positioned for user-reading. 
Modern machines can also be equipped with code scanners that check a preprinted 
bar-code, comparing this code with microprocessor-held recipe information, 
and reject or produce an alarm on any false codes. 
6.5.6 Sterile products 
It can be assumed that products manufactured aseptically arrive at the packaging 
stage in sealed containers that assure the integrity of the product. 
The exceptions to this are items manufactured using integrated form-fill-seal 
systems, and impregnated dressings, where specific handling arrangements 
apply.
Ampoules and vials 
Although some unit sterile products (both liquid and powder) are filled into preprinted 
ampoules or vials, it is not uncommon that these components are 
effectively unidentified prior to labelling. It is, therefore, essential that filling 
controls are such as to ensure that the containers are held in identifiable lots, 
and that these lots are labelled with minimum delay or handling. It is thus usual 
for ampoules and vials to be labelled immediately following aseptic filling or 
terminal sterilization. In the latter case, they will be held in sterilizercompatible 
trays that are used as loading cassettes for the labelling machine. 
Wherever possible, manufacturers will arrange for unlabelled injectable 
product containers to have a form of product-specific machine-readable code. 
In such cases, the first task of the labelling machine will be to read this code and 
compare it with recipe information held in its control system. 
As with oral-dose products, modern labelling systems use self-adhesive preprinted/
coded labels in reel form. It is common for these labels to use a 
transparent substrate such as polyester film to facilitate product visual inspection 
after labelling. 
Pre-printed code checking is also included in modern labelling machine 
technology, and is again linked to control-system recipe information. 
Syringes 
Similar procedures apply to syringe packaging as for ampoules and vials, but 
the inconvenient shape of pre-filled syringes means that specifically engineered 
handling systems are required. 
Form-fill-seal 
High-volume production of single-dose and large volume infusion solutions is 
frequently performed using integrated-system technology. This approach is 
based on the use of high-quality thermoplastic materials (such as polypropylene) 
in granule form being heat-moulded in an enclosed system within a 
controlled-environment machine enclosure, to produce sterile empty containers, 
which are immediately filled in-situ with the sterile-filtered product 
solution. The filled containers, which may be single or multiple-moulded units, 
are immediately heat-sealed prior to emerging from the controlled enclosure. 
This form of production requires sophisticated and expensive machinery but 
has high throughput and the possibility of locating the forming-filling unit in an 
area of lesser environmental quality. It is also possible to emboss product and 
batch code information onto the containers at the point of manufacture, thus 
enhancing identification integrity.
Creams and ointments 
Such products are often filled aseptically into collapsible tubes using techniques 
similar to those employed for non-sterile products. These procedures are 
most often used for the manufacture of ophthalmic ointments. 
Another application for semi-solid products is in the preparation of 
impregnated dressings. Although it is not a common product type, it has 
particular importance in the manufacture of material for the treatment of severe 
skin conditions, including burns. The technology involves the dosing of the 
medicated product onto a suitable substrate (usually tulle) in reel form, in a 
continuous or semi-continuous automated process carried out under aseptic 
conditions. The impregnated tulle is then cut into unit-treatment sections, 
which are packed into sachets, using a form-seal process. The sachet-forming 
material would consist of paper/foil/plastic laminates in reel form, presterilized 
by irradiation. 
6.5.7 Container capping and sealing 
Solid or liquid products packed in glass or plastic bottles, jars or tubs require 
some form of lid closure to protect the contents. A typical bottle closure would 
be a pre-moulded screw-on plastic cap with a composite paper wad to provide a 
seal. 
Such caps were originally hand-applied and tightened, but this action gave 
rise to unreliable seals and leakage, so mechanized systems were developed 
which provided a constant application torque, although the bottles were still 
hand-presented. As outputs increased the arrangement was changed to one of 
automatic presentation, application and tightening. 
A small number of incidents of product pilferage occurred, so the consequent 
requirement for tamper-evidence led to various attempts to provide a 
'pilfer-proof feature. One such, for jars, involved the application of a 
plastic /aluminium foil laminate, heat-sealed onto the jar by means of 
heat/impulse sealer (similar to the system used for instant coffee jars). This 
solution provided the added benefit of enhanced product protection from 
moisture ingress. 
Alternative tamper evident methods included the use of roll-on aluminium 
type caps, where the bottle thread is followed by spinning rollers that form the 
cap thread. These have also been utilized without the tamper-evident feature. 
Plastics have been used successfully for many years as a material for both 
container and cap manufacture. These include both screw and press-on flexible 
plastic caps, the latter also being employed for glass bottles. Such flexible 
materials have the added possibility of including a press-on tamper-evident cap, 
which combines adequate product protection with ease of application.
6.5.8 Container labelling and coding 
Early labelling systems used vegetable or animal-derived semi-solid glues 
manually applied to paper labels, which were then applied to the container. 
This approach had many failings, notably: 
there were no reliable checks on label identity or batch code; 
the position of the label on the container was not fixed; 
there was no automatic batch coding. 
Later systems, still used in many non-pharmaceutical applications, retain the 
use of wet glues but employ machine-application. Early versions of such 
machines employed automatic batch code printing, although the resulting print 
quality was not good. 
Most modern pharmaceutical labels are of the self-adhesive type, which 
allows cleaner operation and reliable appearance. Automatic machines usually 
include product bar-code scanning and automatic batch coding, with alarm 
systems for integrity failings. 
A long-standing feature of pharmaceutical packaging has been the use of 
market-specific labelling. This requirement gives rise to a potentially wide 
range of label alternatives, with stock holding and cost consequences. 
A modern system has been developed to overcome this problem, utilizing 
plain self-adhesive label stock onto which all product, batch and expiry details 
are automatically printed in multiple colours using microprocessor controls. 
Recipe information held by the microprocessor system is fully validated to 
ensure correct output. 
6.5.9 Cartoning 
The placement of filled containers of liquid or unit solid pharmaceutical 
products into cartons was initiated for a number of reasons, including the 
need to insert leaflets providing patient usage instructions and, in the case of 
liquid products, the addition of a standard dose-measuring spoon. Such 
placements were initially performed manually. 
As demand and output increased, automatic machines were introduced. This 
automation created a number of challenges to consider, for example: 
• the importance of detail design, accuracy of cutting and assembly of blank 
cartons to ensure efficient mechanical erection and closing; 
> the importance of humidity control during carton storage due to the effect of 
moisture on carton board making it less pliable and increasing friction — 
very significant for higher speed machines;
• the engineering design of cartoning machinery to allow smooth and reliable 
high-speed operation. 
Modern cartoners may be fitted with automatic leaflet insertion, using preprinted 
plain sheets, folded prior to insertion, or reel-fed leaflet stock. They 
may also be fitted with automatic batch and date coding and code scanning to 
determine correctness of carton type and overprinted information. 
Automatic and semi-automatic cartoners are generally of four alternative 
types, which are characterized by method of motion indexing (intermittent 
or continuous) and by direction of container insertion (horizontal or 
vertical). Intermittent motion vertical (IMV) machines are used frequently in 
pharmaceutical packaging, not least because they can be operated in a 
manner that permits manual insertion of bottles/leaflet spoons at one or 
more operator stations. For high throughput, however, continuous motion 
horizontal (CMH) machines are favoured. 
6.5.10 Collation, over-sealing, case packing and palletizing 
The automation of 'end of line' operations within the pharmaceutical industry 
is not a universal practice, although it is becoming more commonplace for 
higher-output packing lines. Owing to the fact that, at this late stage in the 
production cycle, the product is fully sealed, protected and identified, the 
equipment required for final packaging does not generally need to be specialized. 
It is, thus, acceptable for it to be of the same type and source as that used 
for consumer goods packaging. 
Collation of filled cartons and over-sealing with cellulose or polymer film is 
common for many medium-selling products. On low-output packing lines the 
collation is performed by hand and the over-sealing is performed using a semiautomatic 
heat-sealing unit with manual operation. 
For higher-speed lines, typically over 20 cartons per minute, the collation 
of cartons and feeding into a wrapper/sealer is often performed automatically. 
Automatic case packing and palletizing is not universally used, due to the 
relatively low outputs typical of many pharmaceutical products. However, it is 
not unknown, and once again, consumer goods equipment is employed. 
One advantage of automatic final packaging is that it facilitates the 
automatic application and checking of outer carton labels. 
6.5.11 Inspection systems 
Modern pharmaceutical packaging systems rely heavily on inspection systems 
to verify the correctness of critical product parameters, including: 
• fill volumes or unit counts;
absence of contamination; 
container seal integrity; 
container label identity; 
label position and orientation; 
carton identity; 
outer container label identity; 
batch number; 
manufacturing date; 
expiry date. 
In common with other consumer product industries, the pharmaceutical 
industry originally relied on human visual inspection to detect contamination 
and pack faults. Examples included the use of visual checking for particulate 
contamination in ampoules and liquid vials, container, label, cartons identity 
checking, and the monitoring of fill levels. 
These procedures were known to be of limited reliability due to operator 
fatigue and attention-span limitations, and also suffered from slow and variable 
output rates, especially if inspection speeds were operator-controlled. 
Initial mechanized systems, in which the containers were automatically 
presented to the operator's line of sight in an economically efficient 
manner, were introduced. These still relied on operator visual acuity and 
attention, with benefits to output and reliability, but these were not significantly 
faster than a competent human operator, and remained less than 100% 
reliable. 
A considerable amount of survey work was carried out in the 1960s and 
1970s, especially in connection with injectable product inspection, and the data 
generated was used to compare performance with mechanical methods. 
Camera-based systems were introduced during the 1970s by a small number 
of European and Japanese companies, and these provided benefits in terms of 
improved output rates to match similarly improved filling machine performance. 
Detection rates were improved and became more consistent, but the 
machines were limited in capability to a set number of reject types, largely due 
to limitations in the camera technology. These rejects were based upon 
physically measurable parameters (including volumes, counts, contaminants). 
The introduction of digital matrix camera technology during the 1980s 
gave rise to an expansion in automatic inspection capabilities. These 
microprocessor-driven systems can be programmed to recognize deviations 
from standard shapes, the presence of contaminants, and even the correctness 
of components codes and batch and expiry-date numbers.
As with many advances in production technology, the improvements in 
inspection systems have arisen from the quality and output-led demands of the 
pharmaceutical and other high-volume product industries. These challenges 
have been met by the machinery and equipment manufacturing industry and the 
reader is recommended to approach these manufacturers for information on the 
latest advances in this fast-moving area of technology. 
6.6 Warehousing and materials handling 
6.6.1 Introduction 
The storage of materials for pharmaceutical manufacture and the products 
themselves utilizes systems and procedures much like those employed in any 
high-volume consumer products operation. However, there are some special 
considerations applicable to pharmaceuticals resulting from the critical need to 
ensure the integrity of raw materials and products, and these affect the selection 
of storage systems, materials management systems and material transportation 
arrangements. The ultimate choice of system available in each of these aspects 
will be influenced by many 'normal' considerations, but ultimate pharmaceutical 
product security and integrity are the overriding factors. 
6.6.2 Conventional storage 
The extent of raw materials and finished product holding typical of pharmaceutical 
industry operations is not normally considered large. Hence, automated 
high-capacity storage systems are not always required or cost-effective. In these 
situations, 'conventional' warehousing, consisting of racking systems having, 
typically, no more than five pallets in the vertical direction and aisle widths 
between rack faces of around 2.5 to 3 metres, are common, assuming standard 
1.0 x 1.2 metre size pallets. 
The advantage of this arrangement is that the racking can be fully freestanding 
with no top-end fixing, and regular ride-on counterbalance fork lift 
trucks (which can also be used in a variety of non-warehouse duties) are 
suitable for stacking and de-stacking movements. 
Although such arrangements are relatively low-cost, they do have certain 
disadvantages, notably that the pallet density per unit floor area is low, so that 
the area utilization is poor where site space is limited. A further specific 
disadvantage for pharmaceutical warehousing is that, being basically flexible 
and operator-controlled, the extent of automatic cGMP compliance in relation 
to material segregation is effectively zero, and adherence to procedures
becomes the only method of avoiding mistakes in the selection of materials for 
production. 
A solution to these deficiencies is the employment of automated systems 
(see Section 6.6.4). 
6.6.3 High bay options 
Where material volumes are high, in terms of total inventory and frequency of 
movements, conventional warehousing is inefficient, both in storage density 
and in speed of pallet insertion and removal. Where site space is limited, the 
storage density is especially significant. 
High bay warehouses, having vertical pallet stacks of between 5 and 20 
units, provide a solution to high-density storage requirements. They typically 
have narrower trucking aisles and special trucks which cannot be utilized for 
non-warehouse duties. The trucks can be of two alternative types — operatorcontrolled 
ride-on, or automatic crane. The former has many similarities with 
conventional systems, whereas the latter has no direct operator involvement and 
is controlled by a computerized materials management system. Many permutations 
are possible, and the selection will depend on material selection 
frequency, total capacity, number of alternative materials, etc. 
Computer-controlled systems have considerable benefit in pharmaceutical 
warehousing duties, as quality assurance is enhanced by the automated nature 
of material selection and location (see Section 6.7). 
These high racking configurations usually require structural bracing at the 
top in order to provide stability. Indeed, it is not unusual for very high 
warehouses to utilize the racking system as part of the building structure, 
with exterior cladding and roofing supported off the rack framework. 
6.6.4 Automated warehousing 
Some of the major international pharmaceutical companies have invested in 
automated production systems, including warehousing. The latter, based on 
high bay arrangements, utilize materials management systems for the control of 
material movement and usage, interfaced to warehouse control systems that 
handle the insertion, removal and security of raw materials and finished 
product. Such warehouses are typically un-manned and employ stacker cranes. 
As there is no physical operator involvement in materials selection, it is 
possible for automated warehouses to be employed for the storage, in a single 
warehouse, of raw materials and finished products having 'quarantine' as well 
as 'approved' status. The selection of materials is controlled by the materials 
management system, which carries material status information and transmits 
simple location-only instructions to the warehouse crane.
This type of warehouse and management system may integrate with 
automatic production systems, where material movement within the manufacturing 
area is also mechanized, and where the production materials are always 
enclosed within the processing equipment or transfer containers. 
'Islands of automation' arrangements are ideally suited for single-product 
manufacturing facilities, but have also been employed for multiple generic 
product manufacture. Their most significant challenges relate to the specification 
of control systems and their validation, and to the design of mechanisms 
for enclosed material transfer. 
6.7 Automated production systems 
6.7.1 Introduction 
Earlier sections of this chapter refer to the application of automatic manufacture 
systems. 
The adoption of automation in pharmaceutical manufacturing is driven by 
the need to minimize costs, and the desire to avoid the effects of human error. 
As labour costs increase, the reduction of direct manpower requirements makes 
economic sense. At the same time, the cost of pharmaceutical machinery is 
escalating as a result of enhanced technical sophistication and cost inflation, so 
that increased daily running times are necessary to meet return on investment 
criteria. 
Although automated materials handling has been and continues to be 
utilized in pharmaceutical manufacture and warehousing, its application has 
generally been restricted to operations which basically involve a single-product 
type (such as tablets), or those where high-potency product containment has led 
to the development of enclosed systems. 
The additional costs of fully automatic, or 'lights out' operation, are largely 
related to the inclusion of microprocessor-based monitoring and control 
systems, the hardware costs of which are steadily reducing in real terms. 
Hence the cost/benefit relationship is moving in favour of the adoption of 
automation. 
In addition to these manpower and capital cost savings, automation can 
bring other advantages, including: 
improved product consistency and quality; 
enhanced adherence to validated systems; 
reduced services usage per unit output.
6.7.2 Process automation 
Automatic semi-continuous operation of individual process units where bulk 
material input and product output systems are possible (including tablet 
presses, capsule fillers, inspection units) is achieving greater acceptance. 
Such units utilize automated sampling for off-line QC analysis, as well as 
automated measurement and feedback control of fill/compression weight, 
hardness and thickness. Self-diagnosis of electronic systems coupled with 
automatic switching of backup systems can also be expected to become 
common in the medium term. 
Other less continuous processes can more easily be automated (such as 
granulation, drying, blending), as the number and range of control parameters 
are limited. However, automation of the product transfer arrangements linking 
these individual steps is perceived to be more difficult to achieve due to the 
greater separation distances involved and the need for connection and 
disconnection. 
This perception can be answered by amalgamating unit operations within 
single areas, having 'permanent' connections between process steps, and using 
validated Clean In Place systems for inter-batch decontamination. This 
approach allows complete sets of linked operations to be run as 'continuous' 
processes. Applications of this nature are common in certain other industries 
and technology transfer is clearly a major opportunity. 
Additionally, where scale of operation and product mix permit, Automated 
Guided Vehicle (AGV) systems for IBC movement with automatic docking 
facilities can be utilized. This is particularly attractive where bin movements 
and docking operations can take place within technical (non-GMP) areas. 
6.7.3 Packaging automation 
There is considerably wider scope for automation in pharmaceutical packaging 
operations, where higher unit volumes and repetitive tasks traditionally require 
the employment of large labour forces. Cost reduction and quality improvements 
have been achieved throughout the industry over the past 40 years by the use of 
automated operations and higher-speed machinery. There remains considerable 
scope for further automation of these activities, but factors determined by market 
and regulatory pressures are of great current interest (i.e., the movement towards 
original-pack dispensing and patient-specific production). 
The following section of this chapter describes a pioneering approach to 
meeting these challenges, and provides useful information on the engineering 
aspects associated with packaging automation. The authors are grateful to 
Richard Archer of The Automation Partnership for agreeing to the inclusion of 
this section.
6.8 Advanced packaging technologies 
6.8.1 Introduction 
Compared to most other manufacturing sectors, the pharmaceutical industry 
occupies a unique position where the direct manufacturing cost of many of its 
products is a small proportion of the end user price. The major costs in 
pharmaceutical companies are the indirect ones in R&D, marketing and 
distribution, not manufacturing. In simplistic terms, it could be said that 
pharmaceutical manufacturing costs were not really important. If this statement 
seems contentious (which it deliberately is), consider the impact on respective 
company profitability of halving the production cost of a car compared with 
that of a tablet and how such a proposition would be viewed. For a car company, 
manufacturing costs are of paramount importance in achieving competitiveness, 
with the whole product design and development process geared to 
manufacrurability and provision of maximum product features and choice at 
minimum cost. For the pharmaceutical companies, the primary emphasis is on 
discovering and launching increasingly effective molecules and therapies. 
Provided there is a method of manufacture that can be well controlled and 
monitored, the actual direct production cost is comparatively unimportant. 
This unique situation has changed, however, as pharmaceutical prices have 
come under greater scrutiny from governments, healthcare providers, insurance 
companies and the challenge of changes in the selling and distribution of 
prescription drugs. Both the direct and indirect costs of production and 
distribution are under pressure, while the market is demanding greater 
choice, improved service, faster response and lower prices. 
In many respects, therefore, the pharmaceutical industry is now having to 
face the same issues of cost and flexibility that most manufacturing sectors had 
to address decades ago. The industry is, however, unfamiliar with the key 
principles of truly flexible manufacturing and much of the available processing 
equipment is unsuited to rapid changeover and responsiveness. Too few 
pharmaceutical companies today recognize that the ultimate objectives of 
advanced flexible manufacturing are reducing indirect costs and generating 
new business opportunities, not direct cost reduction. 
Packaging of solid dosage products is indicative of these aspects. The 
current equipment is comparatively high speed and is geared to long, efficient 
production runs in one pack format. Increasing pack variants and inventory 
reduction pressures have led to smaller batch sizes, but this then results in lines 
where changeover time often exceeds running time.
This section describes how a radically different approach to tablet packaging 
has been developed which seeks to address these new market issues. The 
objective, as with modern car manufacturing, is to reduce the viable batch 
quantity to a single product unit. 
6.8.2 Conventional pharmaceutical packaging and distribution 
Conventional drug packaging lines are geared to large batch quantities of single 
products, which are subsequently distributed through a complex internal and 
external chain of warehouses and distributors. (It has been suggested that it 
typically takes six months from packaging for a prescription drug to be received 
by the patient). It could be said that the inflexibility of the conventional 
packaging process is the cause of the current multi-stage distribution route 
rather than a consequence of it. Remember that the end user ultimately purchases 
one pack at a time; in other words, large batch quantities are a consequence of the 
existing packaging/distribution process not a customer requirement. 
Traditional bottle filling systems are mechanically tooled and controlled, 
using tablet specific slats or pocketed disks to provide a pre-determined fill 
quantity. Tablet inspection, if used, is usually provided by eye. Changeover can 
take up to a shift to achieve and is primarily a mechanical technician task. 
Market data suggests that purchases of this type of filling system are declining 
markedly and that electronically controlled vibratory fillers are now selling in 
increasing numbers. These newer technology fillers, while theoretically slower, 
have fewer, if any, tablet specific components, use electronic counting methods 
and incorporate some basic automatic inspection of tablet area. Product 
changeover can be achieved in perhaps 1-2 hours. The trend to these new 
types of machine indicates that the industry is beginning to recognize that 
equipment flexibility is more important than absolute speed. 
Aside from filling, the other areas of inflexibility in packing is the production 
and control of printed material, most particularly labels. Off-line printing 
techniques are used and the resulting materials are handled and released using 
control methods not dissimilar to those needed for producing banknotes. 
Nevertheless, labelling errors still cause around 50% of product recalls, with 
significant costs both financially and to product/company image. 
While many pharmaceutical companies recognize the limitations imposed 
by their packaging equipment, it has been an area of relatively slow technology 
change. There are two related causes for this. Much of the pharmaceutical 
packaging equipment is produced by companies who, with few exceptions, are 
small relative to their customers. Not unreasonably, the equipment companies 
do not have the financial or technical resources to undertake major new product 
development programmes involving radically different technology and tend to
concentrate on enhancing their existing products. In contrast, the pharmaceutical 
companies have the size and financial resources to develop new equipment 
but traditionally have not sought to develop their own packaging equipment and 
have sat back awaiting new offerings, preferably from well known vendors. It is 
not difficult to see how these two effects can lead to technology stagnation. 
A further restricting factor is the relationship in pharmaceutical companies 
between marketing and engineering. Again taking car manufacture as a 
comparison, the linkage between these two departments in pharmaceutical 
companies is relatively small. Marketing would not naturally look first to areas 
such as packaging engineering for significant new business opportunities. It is 
typical to find internal 'new production technology' groups with no formal 
marketing involvement or, indeed, 'new market development teams' with no 
engineering input. Innovative in-house process technology developments have, 
therefore, to be justified against relatively small efficiency gains in direct labour 
reduction and material usage, rather than the substantial returns associated with 
new business generation. The end result is that where internal process 
innovation is pursued, it is often under-funded, has a low commercial priority 
and lacks a clear business objective and focus. 
6.8.3 What does the market want? 
In the last ten years the distribution channels for pharmaceuticals in the United 
States have undergone some dramatic changes and continue to do so. Pressure 
from corporate health programmes, medical insurance providers and government 
to reduce healthcare costs has resulted in new purchasing and distribution 
routes emerging. A key example is the explosive growth of companies who 
manage the purchase of pharmaceuticals on behalf of health plan providers. 
These companies act on behalf of the healthcare provider and negotiate 
substantial volume discounts with the drug producers against a restricted list 
of recommended drugs. These companies handle patient prescriptions at 
centralized semi-automated facilities and the packaged drug is shipped direct 
by mail to the patient. The conventional manufacturer/wholesaler/pharmacy 
distribution route is completely bypassed. A substantial proportion of the US 
population now receives many of its prescription pharmaceuticals in this way. 
Other organizations, such as hospitals and nursing homes, are now pursuing 
similar methods to obtain price benefits through centralized pharmacies. Whilst 
these are primarily US phenomena today, it would be naive to assume that 
similar developments will not appear in Europe in due course once the financial 
impact of these programmes become apparent to government-funded health 
services.
There are a number of other market-related issues, all of which mitigate 
against conventional drug packaging methods. These include: 
'globalization' of production by companies such that a single site may now 
produce all country and pack variants of a drug, requiring multiple 
label/language formats in the same facility with frequent changeovers; 
the requirement of the large supermarket-based pharmacy chains to have 
product identification and expiry date incorporated in a label bar-code to 
allow automated stock control. Conventional label production methods do 
not handle this need easily. Many chains are seeking their own branding on 
the label in addition to, or instead of, the manufacturer's name; 
label and insert data change frequently in response to new drug indications 
and side effects. Obtaining pre-printed material can delay the launch of a new 
or revised product by several weeks; 
direct management of retail shelf space by the supplier. 
In summary therefore the market is demanding: 
increasing pack complexity, variety and customization; 
order delivery in a day with no intermediate handling and inventory costs; 
frequent pack design changes; 
single pack unit batch quantities; 
lower end user pricing. 
The implication is that a make to order strategy is needed rather than make to 
stock. It is apparent that better management of, or enhancements to, conventional 
drug packaging lines will not address these new market needs, and that 
radically different equipment will be needed whose technology origins may be 
from outside the pharmaceutical industry. 
6.8.4 New technologies 
Other industries had to address the responsiveness/flexibility issues many years 
ago in order to survive. These manufacturers have had to take the initiative in 
stimulating the development and implementation of new manufacturing 
process equipment. Many of the principles and technologies that have resulted 
from this are equally applicable to pharmaceutical packaging. 
Technologies that are relevant to an advanced tablet packaging system 
include: 
'robotic' equipment design: Whilst not necessarily using anthropomorphic 
arms, the underlying technology of electronically controlled actuation can 
give rise to machines that can switch instantaneously, under computer control,
between different tasks and make intelligent decisions at high speed. That 
these machines may be both slower and more expensive than their less flexible 
predecessors should be neither surprising nor a problem, when the bigger 
commercial issues described earlier are taken into account. 
image processing: Machine vision is increasingly used for identification and 
inspection functions. The exponential growth in cheap computing power 
means that complex inspection and counting functions can be implemented 
in practical systems. 
product identification: A wide range of identification methods is available 
which allows product to be located and tracked by remote methods. Radio 
Frequency (RF) tags are extensively used in car manufacture to locate and 
route cars and components through variable process paths. These feature a 
short-range (50 mm) radio receiver/transmitter, memory electronics (typically 
a few Kbyte), and a battery in a compact, low cost format. All relevant 
product/process option data can be written to these tags and a complete 
process history recorded. On completion of the process these tags are reset 
and returned to the process start. These 'active' tracking methods have 
benefits over passive techniques, such as bar-codes, because they eliminate 
much of the need for large centralized tracking computers. 
real-time computer control: The use of smart machines depends on direct 
high-speed computer control. Whilst computer control of chemical processes 
is well understood in pharmaceuticals, it is comparatively uncommon to find 
computers used in this way in secondary processes. In general, computers are 
used only for scheduling, supervisory machine control and paperwork 
generation. The uncertainty of computer validation only leads to further 
caution over using direct computer control. 
on-line printing: Printing technology has been revolutionized in the last 
decade as sophisticated, low cost, high quality equipment has appeared, 
mostly for the office market. It is perhaps ironic that a packaging manager 
probably has more sophisticated computer power and printing technology on 
the department secretary's desk than on the packaging lines. Developments 
in ink jet, laser and thermal printing allow single, unique, high quality 
images to be produced rapidly and on demand. Technology developments for 
other industries will soon allow near photographic image quality to be 
achieved at line speed. Real time generation of unique single labels is already 
a practical proposition in both monochrome and colour. 
Much of the necessary technology for an advanced, high flexibility, tablet 
packaging line already existed. The challenge was to select and configure it in 
an appropriate way.
6.8.5 Postscript technology 
In 1991 The Automation Partnership ('TAP') began collaboration on a number 
of developments of novel manufacturing processes with Merck and Co. TAP 
offered a skill set in robotics, machine vision and computer control, while 
Merck recognized the need to take the initiative in developing radically 
different, advanced secondary process technologies. A number of these projects 
were aimed at line changeover time reduction, particularly the areas of tablet 
filling and on-line label printing. These early projects resulted in prototype 
production equipment which demonstrated that much higher levels of flexibility 
could be achieved for small batch, single product packaging, under GMP. 
These were still aimed at make for stock production. 
These separate developments led subsequently to a concept, which became 
known as 'Postscript', for customer-specific packaging of tablets. With this, a 
customer order, down to a single bottle of tablets (such as a prescription), could 
be received electronically, counted, inspected, packed, uniquely labelled and 
despatched within a few minutes. Ideally, there would be little direct manual 
involvement in the process and a very high degree of integrity would be 
guaranteed by the system design. In principle, the line concept could receive, 
pack and directly despatch small end user orders within a day, eliminating all or 
most of the conventional distribution chain, large intermediate product inventories 
and the need for complex scheduling/forecasting systems. In other 
words, it would be closely aligned to the new market needs discussed earlier. 
Not surprisingly, the concept was received with a mixture of technical 
concerns and business interest. It was decided that Merck would jointly 
develop, construct and demonstrate a near full-scale pilot line which would 
include all the essential novel elements and allow the feasibility and practicality 
of the new process to be assessed. The key functions and technology are 
described below; however, the concept's modularity allows a range of alternative 
configuration and capacities to be created for other specific needs. 
6.8.6 Pilot plant configuration and equipment 
The pilot plant line uses a U-shaped configuration with a conventional process 
flow involving empty bottles entering at the line start then progressing through 
filling, capping, labelling, collation and packing into shippers at the end. 
For the purpose of demonstration, the pilot line was configured to receive 
small (hypothetical) electronic orders from customers, such as individual retail 
pharmacies, for a combination of differing product types. In this first case, up to 
four different tablet or capsule types were packed on the line simultaneously 
(although by adding a further four-channel filler modules this could be easily 
expanded to sixteen products or beyond). Orders comprised typically 20 bottles
for a single customer with unique labels on each bottle showing the product 
identification, manufacturer, tablet count and the retailer's address. The bottles 
were packed into an order shipper at the line end, together with a dispatch label 
and order manifest. The system was, however, equally capable of packing a 
single patient prescription. 
The key elements of the line were as follows: 
(a) Tuck' 
The line had about two hundred identical 'pucks', which were used to carry 
individual bottles through the system. The base of the puck contained a 
proprietary RF tag, which allowed all relevant details of the order to be carried 
through the process with the bottle. The fingers on the upper part of the puck 
located the bottle while still allowing it to rotate for labelling. Specific finger 
designs allowed differing bottle sizes to be processed. 
(b) Puck Handling Station (THS') 
Four PHS's were used on the line to provide tracking and routing. The data on 
the puck could be erased, written or read at the PHS and the puck plus bottle 
could then be sent in alternative directions or rejected if faulty. 
(c) Flexible filler 
The filler was a novel patented design that used a vibratory feed, conveyor belt, 
imaging system and diverter to feed, inspect, count and divert tablets to the 
bottle. The filler consisted of four separate identical channel modules, each of 
which processed one single tablet type. Each channel could process between 
500 and 1000 tablets per minute (dependent on tablet/ capsule size) and every 
tablet was automatically inspected for size, shape and colour. Damaged or 
rogue tablets were automatically diverted out of the stream and eliminated from 
the count. Tablet count was verified by two independent systems and any count 
discrepancies resulted in bottle rejection. The tablet count in a given channel 
could be varied for each successive bottle. 
(d) Labeller 
The labelling station used a conventional labelling machine but with a 
customized high-speed thermal printer. A specific label was printed on blank 
feedstock, in response to the bottle's puck data, and then applied. The label 
could also be verified by on-line print quality and character verification 
systems. The label incorporated a unique bar-coded serial number, giving 
each bottle a unique identity.
(e) Collation system 
The order collator used multiple tracks and gates to assemble complete order 
sets. The puck determined the order routing. On completion of the order, the set 
was released and the bottles transferred from the pucks to a tote and then to the 
shipper carton. 
(f) Control system 
The system used multiple networked PC's to provide machine control, system 
monitoring and order tracking. System set up and running was through a touch 
screen. The system software was developed and tested under a structured 
environment suitable for validation. 
(g) Ancillary equipment 
The line used a conventional capper, and standard equipment, such as cotton 
and desiccant inserters, could be easily added as additional stations. The pucks 
were transferred on normal slat conveyors. The neck of the bottle, irrespective 
of its size, was always in the same position relative to the puck base. An 
overhead conveyor returned the empty pucks back to the line start. The pucks 
were reloaded with empty bottles using conventional unscramble/centrifugal 
feeder mechanisms. 
6.8.7 Packing flow 
The process flow is as follows: 
pucks are loaded with empty bottles fed from bulk and then queued on the 
conveyor; 
the first PHS erases all previous data on the puck and verifies a bottle is present; 
the filler receives a common train of empty bottles/pucks which feed the four 
channels as required; 
the filler receives data on the next bottle's fill requirement from the controller 
and then inspects and counts the correct number of tablets into that bottle. The 
puck receives all the data specific to the bottle while filling is in progress. Any 
errors in filling (such as a count error) give rise to an error flag in the puck data; 
the second PHS verifies the data on the puck and rejects any misfilled bottles. 
If appropriate, routing to alternate parallel cappers could occur at this point 
(e.g., choice of regular or tamper proof formats); 
the capper applies the cap; 
the third PHS verifies cap placement and reads the relevant data from the 
puck for label printing; 
the on-line printer produces a correct sequential stream of labels, which are 
then applied by the labeller;
the final PHS reads the unique bar-coded bottle serial number on the label 
and correlates this with the serial number held on the puck. This ensures that 
the label is always correctly assigned to the right bottle; 
the collator uses the puck data to assemble completed orders. Note that 
several orders are processed in parallel — consecutive bottles on the line do 
not necessarily belong to the same order; 
successful completion of an order is reported back to the line controller. 
Parallel new orders are continually being initiated automatically. 
6.8.8 System features 
Particular features are: 
each filling channel operates asynchronously, i.e. the tablet fill speed and 
bottle rate through each channel will be different and may be zero at times 
depending on the content of individual orders; 
depending on tablet count per bottle, the throughput limit for each channel is 
determined by either the 500-1000 tablet/minute rate or the 20 bottle/ 
minute rate. For example, typical limits for a four channel filler module 
would be 80 bottles/minute at 30 tablets/bottle or 40 bottles/minutes at 100 
tablets/bottle; 
capsules and tablets can be packed simultaneously using identical channel 
equipment; 
the channels are physically isolated from each other and contained, with 
vacuum extraction to reduce dust generation and prevent cross-contamination; 
the product contact parts in a channel can be replaced within about ten 
minutes without the use of tools. There are no tablet-specific parts; 
the system can 'learn' the size/shape/colour profile of a new tablet design in 
about two minutes; 
labels can be designed off-line using standard software and then electronically 
downloaded into the system; 
on completing a run, the line automatically empties itself of orders; 
the system generates a separate computer batch record for every bottle 
processed, giving unparalleled traceability; 
an order can be filled, labelled, packed and ready for despatch within five 
minutes of receipt. 
Overall the pilot system has demonstrated all the specified functions and 
performance, and has shown that the concept is valid and achievable. It has 
been subjected to an extensive validation programme.
6.8.9 Future developments 
TAP is exploiting the technology more widely and is currently evaluating 
various applications in pharmaceutical packaging and distribution that might 
use a rapid pack to order approach. These include: 
direct supply to retailers; 
mail order pharmacy; 
clinical trial packing; 
product repackaging; 
hospital supplies. 
Each of these would use the same core technology but in different line 
configurations. TAP is also exploring the opportunities for a similar concept for 
blister pack products for the European market. On-line, on demand, printing of 
blister foil has already been demonstrated at a prototype level by TAP and 
similar systems are becoming available from other suppliers. 
6.8.10 Conclusions 
The Postscript system has demonstrated that the concept of automatically 
packing a batch quantity is both feasible and reliable for solid dosage forms in 
bottles. Changing to a true make to order strategy from make to stock methods 
is, therefore, becoming a viable proposition. Whilst the system has unique 
elements, many of the principles and technologies have been successfully 
transferred from related applications in other industries. Perhaps the most 
fundamental conclusion, however, is that pharmaceutical product packaging 
can change from what some perceive today as a non-value adding process, to 
being an important strategic manufacturing technique that generates significant 
new business opportunities.
7.1 I n t r o d u c t i o n 
This chapter briefly explains how risks to safety, health and environment (SHE) 
are managed in the pharmaceutical industry and how effective process design 
can eliminate or control them. The principles and practice of 'Inherent SHE', 
systems thinking, risk assessment, and compliance with legislation, are 
explained for the benefit of process designers and pharmaceutical engineers. 
Since this topic is too large to cover in a single chapter (see Figure 7.1), a useful 
bibliography is provided at the end for further reading. Specific pharmaceutical 
industry hazards that can be controlled by suitable process design are also 
reviewed. 
Effective process design is an essential requirement for controlling risks to 
safety, health and environment (SHE) in pharmaceutical production facilities. 
Process design that results in robust, inherently safe, healthy and environmentally 
friendly processes, simplifies the management of SHE through the 
complete life-cycle of a pharmaceutical facility. 
Fortunately, the considerable process design knowledge about SHE gained 
in the petrochemical, fine chemical, nuclear and other industries can be adapted 
and applied effectively in the pharmaceutical industry. Although, the pharmaceutical 
industry was slow to apply this knowledge initially, it has since 
expanded its use from primary to secondary production and other areas. 
7.2 SHE management 
The over-riding impact on SHE management over the last decades has come 
from societal pressure and legislation. Several major industrial accidents 
generated public concern and led to stricter legislation. Single-issue pressure 
groups raised public awareness, particularly concerning the protection of the 
environment, which led again to stricter legislation. As a result, the emergent 
requirement of recent SHE legislation worldwide is for auditable risk management 
based on effective risk assessment. 
7 
S a f e t y , h e a l t h a n d 
e n v i r o n m e n t ( S H E ) 
JOHN GILLEn
Exposure limits 
Occ. H. assessment 
Personal protection 
SHE criteria 
Auditing 
Change control 
Hazard identification 
Hazard assessment 
Area classification 
Fire safety 
Environmental 
impact assessment 
Dust explosion 
Layout design 
Civil law 
Statute law 
Occupational health 
Behaviour 
Insurance 
Cost benefits 
Documentation 
Standards 
Monitoring 
Validation 
Human factors 
Inherent SHE 
Risk assessment 
Business risk asst. 
Environmental 
impact 
Hazardous properties 
Other properties 
Equipment 
Buildings 
Land 
Legislation 
Values 
Exposure 
Motivation 
Education 
Straining 
Economics 
Procedures 
Control 
Computer systems 
Technology 
Business assets 
Materials 
Capital assets 
Society 
Individuals 
^Management] 
Quality 
Software 
Hardware 
People 
Environment 
Process 
systems, 
SHE 
Figure 7.1 The safety, health and environment domain map
7.2.1 Integrated SHE management 
Most pharmaceutical businesses adopt an integrated approach to managing 
SHE. In the past, safety, occupational health and environmental protection were 
usually managed as separate functions. The recognition that SHE was a line 
management responsibility that must be driven from the top to be effective 
converted the roles of SHE professionals from policemen to facilitators and 
enabled more effective use of SHE technical resources. It is well recognized 
that effective SHE management significantly reduces risks to product security 
and business as well as enhancing quality assurance. 
As explained previously, SHE management has been driven by societal 
pressure and legislation to manage and assess risks effectively. However, 
the sheer urgency of business survival requires effective risk management 
— accidents cost money. Successful businesses give SHE management 
high priority from economic necessity. High quality and effective SHE 
management are also seen to go hand in hand. In successful enterprises, 
SHE is managed from the top to the bottom of the business organization with 
accountabilities and responsibilities clearly stated. 
An effective SHE management system that is used in many successful 
businesses is shown diagrammatically in Figure 7.2. 
The SHE management system described in Figure 7.2 consists of a cycle 
of activities with feedback to ensure continuous improvement of SHE 
SHE policy 
Review Standards 
Guidelines Reports 
Audits Procedures 
Figure 7.2 The safety, health and environment management cycle
performance. The cycle starts with a clearly stated SHE policy for all staff. This 
policy, together with more detailed SHE performance standards, is mandatory 
for all business areas. It is important to note that international business SHE 
standards must be written so that they can be applied to different cultures and 
legislative systems. The quality feedback loop is closed by compliance reports 
and SHE monitoring that provides the substance for a board level annual review 
of the SHE management system and performance achieved. In the example of 
Figure 7.2, the standards will define acceptable risk criteria and procedures for 
performing risk assessment in an effective and auditable manner. 
This SHE management cycle is well suited to the pharmaceutical industry 
where similar quality assurance systems are well known and accepted. Most 
pharmaceutical businesses already have similar SHE management systems to 
that described. It is important that these systems include suitable hazard 
identification and risk assessment procedures and criteria so that SHE management 
is performed effectively. 
7.2.2 Safety culture 
Since the Industrial Revolution, attitudes to safety have changed considerably 
for the better. At the outset, injury and loss of human life were largely ignored 
in the drive for profit. However, several philanthropic industrialists and 
individual campaigners eventually persuaded the government of the day to 
pass legislation that required employers to provide reasonably safe working 
conditions for their employees and to record and report accidents. 
The gradual improvement in industrial accident rates that followed was in 
four stages (see Figure 7.3a, page 206). The first stage was driven by 
legislation. During this stage, when there were numerous accidents, it was 
relatively easy to make simple improvements in procedures and protection to 
comply with the law. The second stage reduction in accident rates was driven 
by loss prevention and was largely due to improvements in process design and 
equipment based on quantitative risk assessment. The third stage was driven 
by effective SHE management and by recognizing the importance of human 
factors. During this stage, several major accidents due to poor management 
occurred and legislation became stricter. Some pharmaceutical businesses 
may still be at this stage of safety management, but others have already 
identified a fourth stage of improvement. The fourth stage improvement 
depends on the behaviour of the people in the business organization and a 
potent 'Safety Culture'. This is a topic that is outside the normal province of 
process designers, but must be borne in mind during risk assessments 
involving human factors.
Figure 7.3 a) An accident rate reduction model, b) Life-cycle of a typical pharmaceutical 
product 
7.2.3 Change control 
Change is a natural phenomenon that occurs everywhere and is unavoidable. 
Change can be initiated deliberately to gain improvements or can occur 
unexpectedly. Whenever there is a change in a system, risks will be increased 
if there is no method of change control. Changes must, therefore, be controlled 
to eliminate or minimize risks. 
Accident 
rate 
Driving forces for improvement 
Legislation 
Loss prevention 
Effective risk management 
Behaviour 
Safety culture 
Time 7.3 a) 
Project phase: 
Research 
Process development 
Process and project definition 
Project design 
Procurement & construction 
Commissioning 
Operation 
Modifications 
De-commissioning 
Demolition 
Capital authorisation 
Time: 7.3 b)
There are two basic types of change. The most obvious type is change to 
hardware. Less obvious is the software change. Hardware or engineering 
changes are usually controlled on the basis of cost, although it is important 
to recognize that some inexpensive changes can, nevertheless, be very 
hazardous. Software changes are usually very easy to make and are often the 
most hazardous. (Software in this context includes not only computer software, 
but also procedural, organizational and people). It is extremely important that 
any system for managing change can identify whether risks are acceptable, 
regardless of the type or cost of the change. 
7.2.4 Performance management 
'You cannot manage what you cannot measure' is a well-known adage. 
Unfortunately, SHE performance is rather difficult to measure, particularly 
when it has been improved significantly. After the Industrial Revolution, the 
number of fatalities provided an easily recognizable and practical safety 
performance measure. As safety improved and fatalities became more rare, 
there were not enough to be able to determine trends easily, so major injuries 
were included to increase the event frequency. Eventually, as there were further 
safety improvements, minor accidents were included. The pharmaceutical 
industry has a good safety record, and even minor injuries are becoming too 
infrequent to be a reliable measurement of management control. Many 
organizations now record 'Near Miss' events as a more responsive performance 
measure. The measurement of SHE inputs such as training, auditing, documentation 
and human behaviour, are also used to provide more responsive and 
precise measures of performance. 
7.3 Systems approach to SHE 
'Systems thinking' is an extremely valuable tool in the pharmaceutical industry. 
This is because the industry involves a complex interplay between different 
people, organizations, cultures, processes, equipment, and materials. It is, thus, 
essential to consider the whole picture to take effective decisions. 'Systems 
thinking' must be at the heart of process design and management to control 
both SHE and business risks. The lateral thinking needed to obtain 'Inherent 
SHE' (discussed in Section 7.4) often stems from 'Systems thinking'. 
7.3.1 Basic principles 
'Systems thinking' or 'Holistic thinking' has been used widely by many 
disciplines to provide new and improved understanding of complex problems.
There are many definitions of the word 'System'. In the context of this book, a 
system is 'a whole' or 'a combination of many parts that work together towards 
a common goal'. The parts may be tangible or intangible, objective or abstract. 
Systems can be explained as a hierarchy. Every system exists inside a higher 
system called its environment. A system can also be divided into subsystems 
that can be similarly divided into sub-sub-systems. For example, an international 
pharmaceutical business will operate in many countries, and include 
research, development, commercial and manufacturing organizations. Each 
organization will have people, processes and equipment at different locations. 
At any one location there will be processes that contain equipment items. An 
equipment item will be made of several parts and each part will be made of 
several elements. 'Systems thinking' involves the whole system from the top of 
the business down to the last bolt connecting one of the equipment parts into the 
whole. Determining the correct balance between the depth of detail and the case 
of understanding a system is very important in process design and risk 
assessment. 
7.3.2 System definition 
It is not always possible to define a system with sufficient clarity to resolve a 
particular problem. This is usually because there is insufficient knowledge 
about the system elements or their interactions, or because the system is too 
complex to understand in its entirety. Systems that involve human activities are 
particularly difficult to model. Nevertheless, system models, even imprecise 
ones, can be constructed to improve understanding of the problem and, thus, 
guide improvements. 
In general, the better the system definition, the easier it is to identify 
problems within the system. When systems definition is poor, problem solving 
depends on the investigative methods used to probe the system and a balance 
must always be struck between the effort spent on systems definition and that 
spent on system investigation. For example, hazard identification techniques 
need to be more powerful or time-consuming when studying ill-defined 
systems. This aspect of systems thinking is very important when performing 
risk assessments, as will be explained later. 
7.3.3 Life-cycle considerations 
Pharmaceutical manufacturing systems exist in time as well as in a complex and 
international environment. It is, thus, very important to consider the changes 
that could occur to such systems over their normal life-cycle. This is 
particularly true when performing risk assessments. A snap-shot in time may 
not identify hazards that could occur later.
A typical pharmaceutical manufacturing project life-cycle will last for 
several years and consist of at least ten distinct stages (see Figure 7.3(b) on 
page 206). The research stage precedes the development stage to determine the 
product and processes. A series of commercial and therapeutic assessments of 
the project feasibility leads to the process design stage. Engineering procurement 
and construction stages follow this, and then the commissioning and 
validation stages are completed prior to beneficial production. The life-cycle 
continues for several years, usually involving many modifications and system 
changes until the product or process becomes obsolete. The facility may then be 
decommissioned, and finally demolished. Each of these stages present different 
hazards that must be assessed at the project outset. 
7.3.4 Business and commercial considerations 
In the past, SHE was usually maintained as a separate function in many 
organizations. The realization that SHE had a significant impact on business 
performance arose from holistic approaches to business management. Insurance 
systems, quality systems and manufacturing systems interact with 
SHE in a complex manner and systems models have been used to indicate 
the SHE contribution. Such studies have resulted in considerable crossfertilization 
of ideas and practices. Risk assessment is a particular activity 
that has been transformed from a basic engineering tool into a powerful 
business decision-making tool. 
7.4 Inherent SHE 
In practice, 'Inherent SHE' is the elimination of hazards by suitable process 
design so that processes are, by their very nature, safe, healthy, environmentally 
friendly, unaffected by change and stable. The more a process is 'Inherently 
safe', the less protective measures are needed, and the final result is then 
usually less expensive. 
7.4.1 Basic principle 
The basic principle of 'Inherent SHE' is to avoid hazards by suitable process 
design. Although the principle is simple it is, nevertheless, often overlooked, or 
used too late to implement. To apply the principle, it is essential to have 
sufficient time and flexibility to derive and assess the potential solutions that 
'Inherent SHE' can suggest. This means that 'Inherent SHE' thinking must be 
started early in the project life-cycle. It is best employed during the research and 
development stages when fundamental opportunities for change are possible.
However, 'Inherent SHE' thinking needs to be continued throughout the project 
life-cycle, particularly when changes are being evaluated. 
An ability to think holistically and laterally is very important when seeking 
an inherently safe solution to a problem. Several useful guide-words for 
'Inherent SHE' are given in Table 7.1. 
7.4.2 Inherent SHE examples in the pharmaceutical industry 
'Inherent SHE' has been used effectively in the pharmaceutical industry both in 
primary and secondary production. Inventories have always been much smaller 
than those in the heavy chemical industry due to the relatively high activity and 
low volume of the compounds used. Cleanliness and aseptic or sterile 
operations have also driven pharmaceutical engineers to reduce capital and 
operating costs using 'Inherent SHE' principles. 
In primary production, many of the crude production processes use 
hazardous chemicals. The production of hazardous chemicals such as phosgene 
in-situ is one example of inventory reduction. Other examples include the use 
of direct steam injection, direct nitrogen injection, lObarg milling, microwave 
Table 7.1 'Inherent SHE' guidewords 
Guideword 
ELIMINATE 
SUBSTITUTE 
INTENSIFY 
ATTENUATE 
SEPARATE 
Principles 
Avoid using 
hazardous processes 
or materials 
Use less 
hazardous materials 
or processes 
Reduce inventory, 
intensify 
or combine processes 
Dilute, reduce, 
simplify 
Separate chemicals 
from people and 
the environment 
What to consider 
Process chemistry, 
heat transfer fluids, 
refrigerants, processing aids, 
location 
Process chemistry, 
processing aids, location 
Other unit operations or 
equipment, continuous 
rather than batch, 
faster reactions, hazard density 
Keep it simple. Moderate 
the operating conditions. 
Consider process dynamics: 
• high inertia hazards develop 
slowly 
• low inertia deviations 
can be connected quickly 
Containment. Layout. Drains. 
Services. Remote control robotics
drying, solutions rather than isolation as dusty powder, and spray drying to 
obtain free-flowing particles. 
In secondary production, film coating was originally performed using 
flammable or environmentally unacceptable solvents. To overcome the problems 
that such solvents caused, aqueous coating processes were developed. To reduce 
operator exposure, multi-stage granulation processes to make fine active drugs 
free flowing for tabletting have been simplified, integrated, replaced by fluid-bed 
granulation, spray granulation, and occasionally by direct compression. 
7.4.3 Inherent quality and product security 
In the pharmaceutical industry, the principle of 'Inherent SHE' can also be 
applied to quality assurance and product security. This is particularly applicable 
to purification, formulation and packaging processes, discussed in the 
previous chapters. The aim is for robust processes that can be easily validated. 
All the guidewords described previously can be applied to achieve 'Inherent 
Quality'. 
7.5 Risk assessment 
The understanding of the word 'risk' varies considerably throughout society 
and has caused many communication problems. To avoid this problem, this 
chapter will use the Engineering Council (BS 4778) definition of risk as 
follows: 
'RISK is the combination of the probability, or frequency of occurrence of a 
defined hazard and the magnitude of the consequences of the occurrence. It is, 
therefore, a measure of the likelihood of a specific undesired event and its 
unwanted consequences! 
Risk assessment is an essential activity in pharmaceutical process design 
and management. The risk assessment of therapeutic versus toxic effects of 
Pharmaceuticals, research and development activities, clinical trials and business 
risks is not discussed here, although the same principles and methods can 
be applied. 
Risk assessment is performed at several stages in the life-cycle and is 
exemplified by the 'six-stage hazard study' methodology that has been adapted 
and used in various different forms in the chemical and pharmaceutical industry 
(see Figure 7.4 on page 212). 
The six-stage hazard study consists of Hazard Study 1 (HSl) to get the 
facts and define the system, Hazard Study 2 (HS2) to identify significant
Figure 7.4 The six-stage hazard study methodology for a typical pharmaceutical 
product 
hazards, Hazard Study 3 (HS3) to perform a hazard and operability study 
of the final design, Hazard Study 4 (HS4) and Hazard Study 5 (HS5) to 
check that the hazards identified have been controlled to acceptable 
standards, and Hazard Study 6 (HS6) to review the project and lessons 
learned. HS2 may be performed by several methods, including Preliminary 
Hazard Analysis (PHA). HS3 may also be performed in several ways, the 
most well known and powerful being Hazard and Operability Study 
(HAZOP) described later in Section 7.5.3. 
7.5.1 Risk assessment principles and process 
Risk assessment has been a human activity since men first walked on earth. 
People frequently perform risk assessment intuitively in their daily lives 
without realizing it. However, to present a logical and consistent approach to 
risk assessment, it is convenient to describe the risk assessment process as a 
series of separate activities. The risk assessment process is described in 
Figure 7.5 on page 213. The first activity is to perceive and define the 
system to be assessed. The second activity is to study the system to identify 
the hazards that it may contain. Each hazard identified is then studied further to 
estimate the consequences and likelihood of its occurrence. The combination of 
consequences and likelihood is then compared with a risk criterion to decide 
whether the risk is tolerable or not. These activities are described in more detail 
in the following sections. 
Time: 
Capital authorization 
Hazard study: HS1 HS2 HS3 HS4HS5 HS6 
Project phase: 
Research 
Process development 
Process and project definition 
Project design 
Procurement & construction 
Commissioning 
Operation 
Modifications 
Decommissioning 
Demolition
Figure 7.5 The risk assessment process 
7.5.2 System definition 
The first step in risk assessment is to define the system where the hazards exist. 
This step is crucial to the effectiveness of hazard identification. As explained 
previously, hazard identification in an ill-defined system will require more 
effort than in a well-defined system. It is, thus, important to try to model the 
system being assessed with as much detail and accuracy as possible. 
In pharmaceutical manufacturing systems, it is important to define the 
software as well as the hardware. The software includes all the human systems, 
process and maintenance organization, controls, procedures, information, 
computer software and all the intangibles involved in manufacturing. The 
hardware consists of the tangible items involved in manufacturing such as the 
process materials, equipment, buildings, services and products. 
It is advisable to start risk assessment by listing all the materials in the 
system to be studied. The materials' hazardous properties are then assessed, 
including their potentially hazardous interactions with each other. It is important 
to assess all the materials, including those that are used for services, 
cleaning, maintenance and activities supporting manufacture. 
Having assessed the hazardous properties of the materials in the system, it is 
then possible to assess the manufacturing activities and production processes. 
Process flowsheets, piping and instrument drawings, engineering line drawings, 
activity diagrams, pictures, batch sheets, standard operating procedures and 
computer logic diagrams are typical pharmaceutical industry process system 
models that are used. The most powerful system models, however, often reside 
Modify 
system 
Risk 
acceptance 
? 
Risk 
determination 
Assessment 
criteria 
Likelihood 
estimation 
Consequences 
estimation 
Threat 
identification 
System 
description
in the minds of the people who work within the system, so the selection of the 
risk assessment team is important. 
7.5.3 Hazard identification 
Effective hazard identification is best done by a carefully selected team of 
people and depends on two key factors — the accuracy of system definition 
and the method used to seek the hazards in the system. As explained previously, 
the better the system definition the easier it will be to identify the hazards 
within. A balance of effort must be struck between systems definition and 
hazard seeking. Hazards in a system that is defined completely and accurately 
in all its real or potential states may be obvious to the trained observer, but 
unfortunately this eventuality is rare. Since system definition in sufficient detail 
may not be possible, it is then essential to use hazard identification methods of 
increasing power, to generate deviations and ideas from the available system 
model and identify the hazards. 
There are many hazard identification methods available to suit all types of 
system and system definition. In the pharmaceutical industry, the most used 
hazard identification methods are check-lists, 'What If?', Preliminary Hazard 
Assessment (PHA) and Hazard and Operability Study (HAZOP). These are 
briefly described in the following paragraphs. 
Checklists 
Checklists require little explanation as they are widely used as reminders in 
daily life for shopping, travel and household chores. The problem is that if an 
item is not listed, it will not be thought about! Checklists should be constructed 
and tested by the people with the most experience and knowledge of the 
systems that they are to cover. Regular revision of checklists is essential to 
maintain their effectiveness, although this often leads to the lists becoming 
longer and longer. Checklists are most powerful when used creatively to 
stimulate the imagination and raise questions. A slavish, mechanical application 
of ticks to a long checklist will rarely produce very effective hazard 
identification but can be combined with 'What if?' to overcome this problem. 
Checklists are often used to identify hazards in plant modifications, 
proprietary equipment or laboratory activities. 
'What if?' 
'What if?' is a hazard identification method that uses the knowledge and 
experience of people familiar with the system to ask searching questions about 
its design and functions. Effective 'What if?' requires an experienced leader, 
since it is a brainstorming method and, therefore, not tightly structured.
When dealing with a large system, 'What if?' is best tackled by subdividing 
the system beforehand into specific subsystems. The study team performs a 
step-by-step examination of the best available system model from input to final 
output. Team members are encouraged to raise potential problems and concerns 
as they think of them. For each step, a scribe lists problems and concerns on a 
flip chart or notepad. These are then grouped into specific issues. Each issue is 
then considered by asking questions that begin with the words 'What if?' For 
example, 'What if the wrong material is added?' 'What if the next step is 
omitted?' and 'What if it gets too hot?' 
The questions and answers are recorded and then sorted into specific areas 
for further study. 'What if?' is usually run in short sessions of about an hour per 
subsystem with a team of two or three people. Although the results of 'What 
if?' can be severely limited by insufficient team knowledge and experience, this 
method and its many variations have been used with apparent success for many 
years. There are now several computer software packages commercially 
available for assisting and recording 'What if?' studies. 
'What if?' is often used at the research and development or feasibility study 
stages of the product life-cycle. It is also used for identifying hazards in plant 
modifications, proprietary equipment and laboratory or pilot plant activities. 
Preliminary hazard assessment 
Preliminary hazard assessment (PHA) was specifically developed to identify 
significant hazards during process development and feasibility studies. PHA is 
a variation of the checklist method that is enhanced by the creativity and 
judgment of a team of experts along the lines of a 'What if?' A list of specific 
subsystems is examined against a list of specific hazards to identify likely 
causes, consequences and preventive measures. Each hazard or hazardous 
situation identified is ranked in order of criticality to allocate priority for safety 
improvements. PHA is not a very searching hazard identification method, but is 
very useful for obtaining a structured overview of the hazards before resorting 
to more sophisticated and time-consuming methods later. PHA is a 'top down' 
method as it usually identifies the top events, such as loss of containment, 
which can then be investigated further down the chain of events until the prime 
causes are identified. It is a useful precursor to HAZOP. 
Hazard and operability study of continuous processes (HAZOP) 
HAZOP is one of the most powerful hazard identification methods available 
and has been well described in the literature. The imagination of a selected team 
is used to perturb a model of the system being studied by using a methodical 
process to identify potential accidents. The system is studied one element at a
time, and is a 'top down' method. The design intention of each element is 
defined and then questioned using 'guide words' to produce deviations from the 
intention. The causes, consequences, and safeguards for each deviation are then 
discussed and recorded. Any hazards that require further action or information 
are listed for follow-up later. 
HAZOP was originally developed for large-scale continuous petrochemical 
processes, but has been adapted and applied successfully to pharmaceutical 
batch processes. HAZOP of batch systems can be very time-consuming and 
requires an experienced hazard study leader to be completed effectively. The 
procedure for HAZOP of a continuous process is well described and many 
people have been trained in its use. Since the procedure for continuous systems 
is simpler than that for batch systems, it is described first (see Figure 7.6): 
• study the system model and sub-divide it into its key elements (Nodes). If a 
Piping and Instrument Drawing (P&ID) is used as the model, look at the 
arrangement of the lines and decide how to divide the drawing into study 
areas; 
• identify each element to be studied (Node) with a reference number. If a 
P&ID is used, number all the junctions that define the elements (Nodes) to be 
studied; 
• select an element (Node) for study; 
• state the design intention of the element (Node). This is an important step in 
the method and must be done carefully and precisely. The design intention 
Obtain a Piping and Instrument drawing (P&ID) of the system 
1. Study the system P&ID and subdivide it into nodes (discrete parts) 
2. Identify each node with a reference number 
3. Select a node for study 
4. State the design intention of the node 
5. Select a parameter in the design intention for study 
6. Apply the first guideword to the parameter 
7. Identify all deviations that could occur with causes, consequences and controls 
8. Record all deviations that require corrective action 
9. Allocate responsibility for completing the corrective actions 
10. Apply the next guideword. Repeat 7-9 until all guidewords have been applied 
11. Select the next parameter 
12. Repeat steps 6-11 until all relevant parameters have been studied 
13. Mark the node on the system P&ID to show it has been studied 
14. Select the next node and repeat steps 4-13 
15. Continue this process until all of the system has been studied 
Figure 7.6 HAZOP of a continuous process
defines the processes or activities involved in the element and the boundary 
for examination. The intention will include details of the process parameters 
that can be changed in the element. Typical parameters stated in the intention 
are flow, temperature, pressure, level and time; 
select a parameter for study; 
apply the guidewords to the intention relating to the parameter selected and 
identify any deviations from the intent. The guidewords are listed with brief 
examples of typical deviations in Table 7.2; 
for each deviation identified, study the causes, the effects and the safeguards 
provided; 
decide whether the deviation requires a design change or corrective action; 
record the decision and allocate the action to a team member for completion 
by an agreed review date. 
When using a computerized recording package, all the deviations are 
recorded and it is also possible to risk rank each deviation. This is useful for 
subsequent auditing of the study and for generating a project risk profile. When 
the study is recorded manually, it has been common practice to record only the 
actioned deviations, but this makes auditing difficult. It is recommended that all 
deviations studied be noted with suitable comments to explain actions taken or 
reasons for acceptance. A typical HAZOP Proforma for recording the study is 
shown in Figure 7.7 on page 218. 
once all the guidewords have been applied to the parameter selected, select 
the next parameter; 
repeat steps 6 to 10 for the second parameter; 
repeat steps 5 to 11 until all the parameters have been studied for the selected 
system element. Mark the element (Node) studied on the model (or drawing) 
with a crayon or highlighter to indicate that it has been studied; 
Table 7.2 Hazard and operability study guidewords 
Guideword 
NO (NOT or NONE) 
MOREOF 
LESS OF 
MORE THAN (or AS WELL AS) 
LESS THAN (or PART OF) 
REVERSE (the complete 
opposite of the intent) 
OTHER THAN (a different intent) 
SOONER/LATER THAN 
Example of a typical deviation 
No flow in pipe. No reactant in vessel 
Higher temperature. Higher level 
Lower velocity. Lower bulk density 
Two phase flow. Contamination 
Reduced concentration. Missing component 
Valve closes instead of opening. Heat rather 
than cool 
Non-routine operations 
maintenance, cleaning, sampling 
More/less time. Operation out of sequence
Figure 7.7 Hazard and operability study report form 
select the next element (Node) for study and repeat steps 4 to 12; 
continue this process until all the system elements (Nodes) have been studied; 
record all actions and file all associated documents in the project SHE dossier; 
the Hazard Study Leader (HSL) then reviews the study overall to prioritize 
the hazards identified. Depending on this overview, the HSL may then 
perform further studies such as a CHAZOP of the computer systems, or a 
Failure Modes and Effects Analysis (FMEA) of critical items; 
the project manager plans HAZOP action review meetings to ensure that the 
actions are implemented satisfactorily. The HSL appends remarks to the 
HAZOP report to check whether further hazard study of the changes made is 
required at these reviews. 
HAZOP procedure for batch processes 
Batch processes are more difficult to define and study than continuous 
processes because they are time-dependent, flexible, subject to changes of 
product and process and frequently involve multiple-use equipment. A batch 
process element can exist in any one of several different states depending on the 
batch process sequence. At a given time, a batch process element is either active 
or inactive. An active or inactive batch process element can also exist in several 
different conditions. An active element can be waiting for a previous batch step 
to complete, or for a subsequent step to be prepared. Active elements are also 
subject to sampling, inspection, batch changeover and other activities that are 
Hazard Study 3: Report Form Project: Session: Drawings: 
HSL: Team: 
Node: Parameter: Intention: 
Sheet of 
Date 
Guideword Deviation Causes of Deviation Consequences Safeguards Actions to be taken Ref. No. By Remarks Date 
Completed
governed by external factors. An inactive element may be undergoing cleaning, 
maintenance, product changeover or merely waiting for the next planned 
production campaign. 
Another factor that complicates batch processes is human intervention. Most 
batch processes have stages that are controlled manually. Human reliability 
assessment of key operations may sometimes be essential to maintain quality 
and production efficiency. The use of computer control may alleviate some of 
the human reliability problems, but then generates additional complexity of a 
different nature. A hazard study of batch process computer systems will be 
required as an additional exercise. 
The hazard study of batch processes is very demanding. The hazard 
study team needs to work very intensely and creatively to link all the 
diverse elements of the batch system together without missing interactions 
or deviations. It is always very difficult at the end of a hazard study to be 
absolutely sure that all the hazards in a batch process have been 
identified. 
Effective HAZOP of a batch process depends on the HSL and the study 
team. HSLs experienced in the hazard study of batch processes all adopt similar 
approaches to the HAZOP methodology, but each will have different ways of 
running a particular study. There is no right or wrong way of doing HAZOP on 
a batch process. The method used must be tailored to suit the study. The 
following approach may be helpful: 
The team members discuss the batch system in general terms to get an 
overview. They use the available documents and drawings to get a clear 
understanding of the key problem areas and to agree on the level of detail 
required for the study. 
The team identify the main sub-systems in order to plan the study. A 
maximum of six or seven is a practical guide. These can then be sub-divided 
to provide the full detail when each is studied individually. There may be 
some duplication and overlaps, but this should not be a cause for concern. It 
is useful to identify a single key element to anchor the attention of the hazard 
study team. For example this might be a reactor with several sub-systems 
such as a heating/cooling system, a charging system, a services supply 
system, an effluent system, and so on. 
The team then construct an activity diagram for the batch process. This step 
ensures that the team understand all the batch process sequences and 
activities. Alternatively the team may decide to use the operating instructions 
for the same purpose.
At this point in the study the HSL has to decide on the level of detail. The 
level of detail will be decided by the preliminary discussions, the results of 
PHA and the complexity of the process. It is worthwhile to perform a firstpass 
hazard study to identify specific areas for deeper study later. A useful 
first-pass hazard study method is as follows: 
o Select the first activity on the activity diagram, or the first step in the 
operating sequence. 
o State the intention of the activity. This must identify the materials, 
equipment, process parameters, and controls. The connections and interactions 
with the total system including the operator and operating 
sequence must also be identified by reference to engineering line drawings, 
the batch sheet and the operating procedures. 
o Apply the HAZOP guidewords to the activity selected. For the first-pass 
study, these are applied to the activity transformation verb, object and 
subject alone. For example, apply the guidewords to 'Fill vessel'; 'Dry the 
batch'; 'Load clean ampoules'; 'React A with B'; 'Operator starts pump'; 
'Computer regulates flow', etc. Use the guidewords in the widest sense to 
generate deviations from the intention. The stated intention relates the 
causes and effects to the drawings and procedures. Several of the 
deviations generated at the start will be re-generated many times over 
when applying guidewords to activities later in the study. The first activity 
studied always generates the most deviations, and, as the study of other 
activities proceeds, fewer new deviations are generated, as most will have 
been identified already. 
o For each guideword, the HSL controls the discussion and recording of 
causes, consequences and safeguards for each deviation to suit the 
creativity and enthusiasm of the team. When ideas are flowing freely it 
is best to record only the deviations and their causes. The effects, 
safeguards and actions can then be discussed when the idea flow ebbs. 
The discussion of the effects and safeguards will then usually set the ideas 
flowing again, and so on. 
o Repeat the above steps for the rest of the activities on the activity diagram. 
o Once all the activities have been studied, make a final overview of the 
whole system. It is useful to use the PHA checklist for this purpose, 
particularly to identify any conditions that could have an effect on the 
whole system. 
o The team decide whether to study any activities or equipment items in 
more detail using the detailed HAZOP batch process method described as 
follows.
The detailed hazard study examines every step of the batch process 
sequence. For each step, each item of equipment used is studied elementby-
element for each equipment state ('Active', 'Inactive', and any other 
state in which it may exist). The parameters for each equipment state are 
then studied using the guidewords. A simplified logic diagram of the 
process is shown in Figure 7.8. 
To perform a study of the whole batch process as thoroughly as this 
would be excessively time-consuming, so it is important to restrict this 
degree of detail to the process steps that have been identified from the firstpass 
study. The Pareto principle that about 80% of the risk lies in 20% of 
the system can be used as a guide to deciding what to include. The 
decisions on how to perform HAZOP of a batch process will be governed 
by the experienced judgment of the HSL. 
7.5.4 Consequences estimation 
A single hazardous event may have many consequences, some of which may 
develop over a significant time period. The final outcomes are, thus, difficult to 
predict with confidence. The Sandoz warehouse fire is a good example of this 
phenomenon. A fire started in a warehouse containing chemicals that were 
potential pollutants. The fire developed extremely rapidly and the local 
population was alerted to close windows and stay indoors to avoid breathing 
the resultant heavy and foul-smelling smoke. The firemen applied large 
volumes of water to control the fire as foam alone proved ineffective. The 
Obtain system operating procedure or activity diagram and all relevant drawings 
1. Select the first step in the procedure or activity diagram 
2. Relate this step to the rest of the system (e.g. P&ID, layout, etc.) 
•3. Select a system element in the step (e.g. an equipment item) 
4. Select a node in the system element (e.g. a pipe or valve) 
•5. Select a state for the node (e.g. active, inactive, other) 
6. Select a parameter for the node in the state chosen 
7. State the design intention of the node for the state and parameter chosen 
8. Apply the first guideword to the parameter 
9. Identify any deviations that could occur and their effects in the system 
10. Record deviations that require corrective action 
11. Allocate responsibility for completing corrective action 
12. Select the next guideword. Repeat 8-12 until all guidewords have been applied 
13. Select next parameter Repeat 7-13 until all relevant parameters have been studied 
14. Select the next state of the node. Repeat 6-14 until all states have been studied 
15. Mark the element (node) on the system P&ID to show it has been studied 
• 16. Select the next node and repeat 5-16 until all nodes have been studied 
•17. Select the next system element and repeat 4-17 until all elements have been studied 
18. Select the next process step and repeat 2-18 
19. Continue this process until all of the system has been studied 
Figure 7.8 HAZOP of a batch process
firewater dissolved the stored chemicals and eventually flowed off the site and 
into the nearby Rhine. The Rhine was polluted and suffered severe ecological 
damage over a length of 250 km. The reparation and litigation costs were 
enormous. As a result of this incident, legislation was passed to ensure that all 
warehouses containing potential pollutants were provided with firewater 
containment to reduce the likelihood of such an event happening again. 
The overall consequences of a hazardous event evolve over time in a chain of 
events triggered by the first event. Although the cause of the event may be 
determined, the consequences are probabilistic. A typical chain is initiated by 
an event that causes a loss of containment of energy or hazardous material. 
Depending on the size of the leak, the efflux will then act as a source for further 
dispersion in the local atmosphere. The resultant explosion, toxic cloud, fire or 
combinations of all three may then affect the local population, depending on the 
weather conditions at the time and the local population distribution. A useful 
method for evaluating potential outcomes of a hazardous occurrence is to draw 
an event tree. An example of the event tree for a solvent leak inside a building is 
shown in Figure 7.9. 
The potential consequences arising from many major industrial hazards have 
been modelled along such chains of events to estimate the effects quantitatively. 
There are, thus, a great many methods and tools available for estimating the 
potential consequences of hazardous events that have been developed in the 
heavy chemical and nuclear industries. 
Vapour detector 
sounds alarm 
Operator stops 
overflow and 
activates foam 
deluge to prevent 
vapour cloud 
Ignition 
prevented 
in building 
Post-accident 
outcome 
YES 
YES YES 
NO 
NO 
Aqueous solvent in sump 
Potential fire elsewhere. 
Solvent in drains until the 
alarm is dealt with 
Solvent catches fire 
(see next Event Tree) 
Large spilage of 
flammable solvent in 
processing building 
YES 
NO YES 
NO 
NO 
Solvent catches fire 
(see next Event Tree) 
Aqueous solvent in sump 
Fire likely elsewhere 
Large solvent spilage 
Figure 7.9 Event tree for a solvent leak inside a building
In the pharmaceutical industry, where the inventories of hazardous materials 
and energy are usually much less than those categorized as major hazards, the 
immediate consequences of fire, explosion and toxic releases are potentially 
less severe than in the heavier industries. Nevertheless, the available consequence 
models can still be used. In addition, there are many pharmaceutical 
chemicals and intermediates that can present environmental hazards as great as 
those from the major hazards industries. The consequences of these hazards are 
best estimated by the models developed and proved for the heavier industries. 
Since most pharmaceutical processes are performed inside buildings, even 
small leaks can generate enclosed flammable atmospheres, which can explode 
with potentially serious consequences. Suitable models are not yet available for 
such indoor situations so expert technical advice will usually be required to 
estimate the consequences of indoor situations. The knock-on effects on 
adjacent facilities must also be considered. 
It is important not to under-estimate the ultimate consequences of fire and 
explosion in the pharmaceuticals industry. The very high value of pharmaceutical 
materials, laboratories and markets can cause potentially very large 
consequential losses in the event of a fire. The chain of consequences that 
can result is usually quite different from those experienced in the heavy 
chemical industries as the effects on markets are often greater than on people. 
The consequential business loss of a pharmaceutical business can be several 
orders of magnitude higher than that of the low margin high volume industries. 
The consequences of hazardous events in the pharmaceutical manufacturing 
industry can usually be estimated to the nearest order of magnitude by 
experienced judgment to make a preliminary estimate of severity. The preliminary 
estimate can then be used to decide whether to use the more powerful 
consequence models. 
The simplest approach to consequence estimation is to consider the 'Worst 
Case' that can be imagined for each hazardous event identified. The extent of 
the worst case and the events that must occur to contribute to it can then be 
determined. Ideas for other scenarios can then be developed by brainstorming 
around the 'Worst Case'. It is also useful to consider a 'typical' consequence of 
lower severity as another reference point in the scale of potential consequences. 
As there are usually several possible outcomes, an event tree approach may be 
helpful to explore the possibilities, otherwise experienced judgment and risk 
ranking can be used to select the possible outcomes for the final risk 
assessment. 
When estimating the consequences in this way, it is practical to consider the 
effect of each identified hazardous event on five key targets:
people; 
the environment; 
process plant, equipment and buildings; 
the product; 
the business. 
By considering separately the potential effects on people, society, the 
environment, material assets, the product and the business, the severity of the 
consequences can be estimated fairly consistently. Various yardsticks such as 
the number of injuries, fatalities, emissions, fires, explosions, or nominal costs 
in monetary terms can be used to build up a reasonably accurate and 
quantitative estimate of the overall consequences. 
The severity of the consequences can then be ranked in a simple scale of 
consequences using verbal descriptions such as 'Very Severe", 'Severe', 
'Moderate', 'Slighf and 'Very Slighf in decreasing order of overall loss to fit 
a risk ranking matrix, described in Figure 7.12 (see page 232). The consequences 
ranked as ' Very Severe' and 'Severe' may then require quantified risk 
assessment using more sophisticated models depending on the likelihood of 
occurrence. 
7.5.5 Likelihood estimation 
Having identified all the hazardous situations and their consequences, the next 
step in the risk assessment process is to estimate the likelihood of occurrence. 
This is very difficult to do consistently without using a logical method and 
some form of quantification because people are notoriously unreliable at 
estimating the likelihood of hazardous events. Any human judgments must 
be explained and recorded so that they can be justified on a logical basis. 
The likelihood of occurrence is usually expressed as a frequency (events/unit 
time) or as a probability (a dimensionless number between 0 and 1). In some 
situations the likelihood may be expressed as a probability over a specified time 
interval and for a particular event or individual. Probability theory and the 
various probability distributions and methods used for reliability estimation are 
described fully elsewhere and are not covered in this guide. 
There are essentially two ways to estimate the likelihood of a hazardous 
event. The first and most reliable way is to use historical data that matches the 
event as exactly as possible. The second way is to calculate the likelihood from 
generic data or from relevant data obtained locally using mathematical models. 
It is important not to use 'off-the-cuff opinions to estimate likelihood since 
these will invariably be misleading.
Estimating the likelihood of hazardous events from historical data 
Historical data should always be carefully checked to ensure that it fits the event 
being studied as closely as possible. Very old data may not be representative of 
current conditions. The accuracy of the data and the conditions under which it 
was obtained must also be carefully checked and validated. If possible, confidence 
limits for the data should be derived using suitable statistical methods. 
The stage in the life-cycle of equipment can also affect the validity of the 
data collected. Typical equipment failure rates follow a 'bath tub' curve through 
the equipment life-cycle shown in Figure 7.10. The curve predicts high failure 
rates at start-up, which decrease steeply during the early life, then level out to a 
constant failure rate for the main life, eventually increasing linearly in the final 
wear-out stages. 
Sparse data should be analyzed using statistical methods to estimate the 
expected mean and deviation. The negative exponential probability distribution 
and the Poisson distribution have been used successfully for system or 
component failure rate estimation in the pharmaceutical industry. 
Historical data that matches the event exactly is often very difficult to obtain. 
This is a particular problem for the events of interest to the pharmaceutical 
industry. Although there are many databanks containing data of major hazards 
incidents, fires, explosions, toxic gas releases, etc., there is currently little data 
that has been derived from the pharmaceutical industry. 
The problem of using data that is not exactly applicable when no other data 
is available is best resolved by adopting a conservative (high) value for the 
Mean 
Failure 
Rate 
Start-up Useful working life Wear-out 
Time 
Figure 7.10 'Bath-tub' curve for equipment failure rate
initial likelihood calculation. Once a conservative estimate has been obtained, 
lower values can then be inserted to assess the sensitivity of the estimate to the 
data. In many cases, particularly in pharmaceutical manufacturing processes, 
the equipment data may not have such an impact on the estimate as the human 
error estimates. 
Estimating the likelihood of hazardous events using mathematical models 
There are many mathematical modelling methods available for estimating the 
likelihood of occurrence of hazardous events. Some of the methods suited to 
the pharmaceutical industry are listed in Table 7.3 and explained briefly in the 
following paragraphs. 
Order of magnitude frequency ranking 
A preliminary estimate of likelihood is always useful in deciding whether to 
use the more time-consuming techniques available. Order-of-magnitude 
frequency ranking is one of the most effective methods for this purpose. 
The method uses a combination of verbal and quantitative data to define a 
frequency band for the event studied. A range of five frequencies can be used 
as a guideline, stepping up in orders of magnitude to fit the five-by-five risk 
ranking matrix described later in Figure 7.12 (see page 232). For example, the 
lowest frequency would typically be one event per ten thousand years 
(l/10,000yrs). The highest would then be once a year (1/year) with intermediate 
steps of 1/1000yrs, 1/100yrs, and l/10yrs. 
These could then be described in increasing frequency as 'Very Unlikely', 
'Unlikely', 'Average'; 'Likely'; and 'Very Likely'. Finer or coarser frequency 
bands can be used to suit individual system requirements. 
Using these broad frequency bands for risk ranking still requires a realistic 
estimate of the frequency for each identified hazardous event. Realistic, if very 
approximate, frequency estimates can be based on local records and knowledge 
or on generic data from the sources previously mentioned. 
Table 7.3 Likelihood modelling examples 
Basis of modelling method 
Real events and statistics 
Expert judgment 
Logical algorithms 
Simulation 
Description of method 
Constant failure rates 
Order-of magnitude frequency 
ranking 
Fault tree analysis, 
human reliability analysis 
Monte Carlo method
Fault tree analysis 
A fault tree is a logically constructed diagram used to model the way that 
combinations of failures cause the event of interest (the top event) to occur. The 
construction of a fault tree provides valuable insights into the way that 
hazardous events interact even if no data is inserted for calculations. However, 
the main use of fault trees is to calculate hazardous event frequencies or 
probabilities. 
The logical arrangement of the 'And' and 'Or' gates of the fault tree is more 
critical to the overall calculation of the likelihood of the top event than the 
accuracy of the data inserted. If the logic is incorrect or key elements are 
omitted, the results will be misleading. It is important to have an independent 
check of the fault tree logic before accepting the results. 
It is advisable to keep the logic as simple as possible. A rule of thumb is that 
if there are more than twenty elements in the tree then subdivision is 
worthwhile. In the pharmaceutical industry, if a problem requires a fault tree 
more complex than this, then a way of avoiding the problem altogether by 
changing the system is usually sought (Inherent SHE). If a better system cannot 
be identified and the fault tree cannot be simplified, then experienced safety and 
reliability engineers should be consulted. 
Human reliability estimation 
Pharmaceutical production processes rely heavily on human operators in nearly 
all aspects, ranging from direct intervention in process operation to business 
decision-making. This can cause problems when attempting to quantify risks 
accurately as human factors are hard to define precisely. 
Although it is relatively straightforward to estimate equipment reliability 
consistently, human reliability estimation, in spite of many years of research, is 
still something of an art. It is important to realize that, when estimating the 
likelihood of a hazardous event, the probability of beneficial action by an 
operator should not be a critical factor to achieve the target criterion. There 
should always be adequate protection in place to ensure that the operator action 
is not critical to the safe operation of the system. 
Human tasks can be classified as 'Skill based', 'Rule based' or 'Knowledge 
based'. Skill based tasks that depend on physical skill and manual dexterity are 
fairly well understood and can be estimated with some confidence. Tasks where 
rules or procedures are important are not so well understood. Some guidance is 
available for formulating clear instructions, but ensuring compliance with rules 
is governed by human behaviour. It is difficult to estimate the effectiveness of 
training and management on behaviour. Knowledge based tasks that depend on
the knowledge and mental models of the operator cannot be modelled with any 
confidence at present. 
The most effective approach is to make a preliminary estimate of the effects 
of human reliability to help decide whether a more detailed analysis is 
warranted. For the best possible circumstances, when an operator is not stressed 
by the situation or his local environment, is well trained and healthy, a failure 
probability of 1 in 1000 (0.001) may be assumed. For the worst possible 
circumstances when the operator is highly stressed, in poor health, in a noisy 
and uncomfortable environment, and is not trained, it is almost certain that 
failure will occur (probability of failure 1.0). Values of failure probability 0.1 
and 0.01 can be selected between these two extremes to fit the local conditions. 
For most activities by well-trained staff in the clean and comfortable environments 
in the pharmaceutical industry, a human failure probability of 0.01 may 
be assumed as a first estimate. For primary production areas, where the 
environment is less comfortable and the processes more difficult to operate, a 
probability of 0.1 may be assumed. 
If a more rigorous treatment is indicated then there are several techniques 
that can be used in consultation with human factors specialists. The 
'Technique for Human Error Rate Prediction' (THERP) considers the task 
in separate stages linked by a fault tree and estimates the probability of 
failure for each stage. The probabilities are calculated from the likelihood of 
detection, the chance of recovery or correction, the consequences of failure if 
it is not corrected, and the 'Performance Shaping Factors' (PSF) governing 
the task. THERP requires considerable time and specialist expertise to derive 
the best estimates of human failure probability. Task analysis can be used 
when a particular task is critical to the business, and the preliminary estimate 
indicates that more precision is required. Task analysis must be performed by 
an expert practitioner to be effective and can prove very costly and time 
consuming. 
Monte Carlo method 
The Monte Carlo method uses numerical simulation to generate an estimate of 
event probabilities for complex systems. Although the method is very powerful, 
it can be very time-consuming if the system failure rate is low. Fortunately there 
are several computer software packages available to ease this burden and the 
method has become widely used throughout the industry. 
7.5.6 Risk assessment criteria 
Risk acceptability criteria govern the management of SHE, quality and business 
performance. If the criteria are set too high, the costs become exorbitant, but if
set too low, the consequential losses become excessive. Risk criteria must be set 
to give the correct balance between the cost of prevention and protection and 
the cost of a potential loss. Since obtaining this balance is hampered by 
uncertainty, risk criteria definition is usually an iterative process with frequent 
reviews and adjustments. In the pharmaceutical industry, risk acceptability 
criteria are usually expressed qualitatively to comply with legislation, codes of 
practice or approved standards. The use of quantitative criteria is still evolving 
in the industry to meet the requirements of tighter budgets and stricter 
legislation. 
Acceptability 
A particular problem that is often encountered is how to decide whether risk 
criteria are acceptable. Acceptable to whom? Risk acceptability criteria can 
only be acceptable to the people who will be affected. Sometimes, when the 
benefits seem to outweigh the perceived risk, people will tolerate a risk until it 
can be made acceptable. In the pharmaceutical industry, risk acceptability 
criteria are dominated by product security and quality as these govern the 
potential consequences to the people who use the industry products. The risks 
from pharmaceutical manufacturing operations, however, are subject to the 
same acceptability criteria as the rest of industry. Risks must be managed in 
such a way that they are tolerable to employees and to the general public. 
Risk acceptability criteria range and precision 
The range of risk acceptability criteria is very large. Many people seek 'Zero 
Risk' at the unattainable bottom end of the range. The concept of 'Zero Risk' is 
often mentioned when the potential consequences of a particular risk are 
extremely severe yet extremely unlikely. There are some risks that could harm 
future generations to such an extent that society would never agree to take them. 
This is the basis of the 'precautionary principle', which is often quoted to stop 
particular risks from being taken. 
There are many practical and achievable risk criteria that society will 
accept. The industrial regulators have used upper and lower boundaries of risk 
with risks in between these levels controlled to be 'As low as reasonably 
practicable' (ALARP). The ALARP principle has been widely and effectively 
interpreted over many years in the law courts as a practical criterion of risk 
acceptability. 
Recent environmental legislation uses the phrase 'Best available technology 
not entailing excessive cost' (BATNEEC) in a similar manner. There are many 
other qualitative definitions of risk acceptability criteria such as these. Unfortunately, 
qualitative risk criteria, which are not very precise, may be interpreted
in many different ways. Comparative risk criteria such as 'Better than' or 'Not 
worse than' some clearly specified example, are more precise and simpler to 
interpret. 
Approved codes of practice and standards set by bodies such as the 
American Society of Mechanical Engineers (ASME) and the British Standards 
Institution (BSI) provide another way of defining risk acceptability criteria. The 
relevant ASME or BS codes can be specified for particular systems to define an 
acceptable level of safety assurance. For example, a specified requirement that a 
pressure vessel is designed to BS 5500 or ASME VIII; Div. 1 is a well-known 
criterion of acceptability. 
Simple risk acceptability criteria 
A simple and very useful method for setting risk acceptability criteria, which is 
easy to explain and apply within the pharmaceutical industry, is 'risk ranking'. 
Risk ranking is based on the intuitive idea that the events with the worst 
consequences should have the least chance of occurrence to have an acceptable 
risk. 
By plotting consequence severity against event likelihood, a borderline of 
acceptability may be drawn between areas of acceptable and unacceptable risks 
as shown in Figure 7.11 (see page 231). This principle was first described and 
used in the nuclear power industry. If the curve is represented as a matrix, semiquantitative 
risk ranking becomes possible as shown in Figure 7.12 (see page 
232). A range of consequence severities is designated along the vertical axis 
and a range of likelihoods along the horizontal axis. The number of subdivisions 
on each axis can be decided to suit individual requirements for 
precision. A three by three matrix is often used for coarse screening risks, but a 
five by five matrix is more discriminating. The risk of a specific hazardous 
event can then be located in the matrix by its severity and likelihood 
coordinates. 
Each square in the matrix is allocated a number to represent the level of 
risk. The convention used is that the higher the number in the matrix, the 
higher the risk. For a five by five matrix as shown in Figure 7.12 (see page 
232), the top right-hand square is numbered 9 and the bottom left-hand square 
numbered 1. A diagonal band of 5s might then be defined across the matrix to 
discriminate between 'Acceptable' and 'Unacceptable' risks. Hazardous events 
with coordinates above the diagonal band are unacceptable, while events with 
co-ordinates below the band are judged acceptable. Events with co-ordinates 
in the diagonal band need further study, as this is an area of uncertainty where 
the apparent clarity of the method should not be allowed to cloud experienced
Figure 7.11 Consequence severity versus likelihood curve 
judgment. Risk ranking is only a coarse filter of the unacceptable risks from 
the trivial. 
The Risk Ranking Matrix, thus, provides a coarse risk acceptability criterion 
that can be tailored to suit particular situations. The allocation of the numbers 
can be skewed to make the criterion as strict or as lenient as required. For example 
the 5s could be classed as unacceptable. Alternatively different numbers could 
be placed in the matrix. To reduce the amount of judgmental bias on likelihood, 
guide frequencies can also be provided along the horizontal axis. 
7.5.7 Quantitative risk assessment 
The most well defined risk criteria for process design and management are 
quantitative. Even so, absolute values for risk acceptability criteria are often 
difficult to justify because quantitative risk assessment (QRA) is not a precise 
tool and usually involves idealized assumptions and the use of unvalidated data. 
In addition, QRA calculations, although logical and mathematically exact, 
often depend on human judgment. This usually means that QRA is mostly used 
LIKELIHOOD 
Acceptable 
SEVERITY 
Unacceptable
for comparisons or for sensitivity analysis. (Sensitivity analysis is the process of 
testing the effects of different values of the data or assumptions made on the 
predictions from QRA models). Sensitivity studies are important for checking 
QRA models and for pinpointing key risk areas for improvement. The main 
advantage of QRA is that it enables the final risk decisions to be explained 
logically and quantitatively against quantified risk acceptability criteria. 
Acceptability criteria for risks to people and the environment from fire, 
explosion, toxic gases and pollution have been developed and agreed in many 
industrial areas. Some of the most widely used quantitative risk acceptability 
criteria in the chemical industry are those for fatalities, but there has been 
considerable debate about using them for regulation because the risks to the 
public attract much controversy. 
The resultant data, experience and techniques give useful guidance for 
setting risk criteria for potential fatalities or pollution in the pharmaceutical 
industry. Risk acceptability criteria for product quality and business risks are 
still under development and are the subject of considerable debate. 
Figure 7.12 Risk ranking matrix 
Likelihood 
Very low Low Normal High Very high 
Guide frequency: 1/100,000 yr.1/10,000 yr. 1/1,000 yr. 1/100 yr. 1/10yr. 
Very severe 
Severe 
Moderate 
Slight 
Very slight 
5 6 7 8 9
8
7
6
5 4 3 2 1
2 3 4 5
6 5 4 3
4 5 6 7 
Severity of 
Consequences
Risks to the public 
When the problem of controlling major industrial hazards was first being 
studied, the Advisory Committee on Major Hazards suggested that a 'serious 
accident' frequency of once in 10,000 years might just be regarded as the 
borderline of acceptability. This frequency was subsequently used as a basis for 
arguments about the acceptability of major risks from process plant in many 
countries. The estimated effects on process personnel and the public from such 
accidents was also used as a guide to the acceptability of risks to individuals. 
One practical acceptability criterion often used is that the risk to a member 
of the public from a major industrial accident should not be significantly worse 
than that from the pre-existing natural risks. Using this principle and an analysis 
of natural fatality statistics, this equates, on average, to a chance death of less 
than one in a million (1.0 x 10~6) per year per person exposed. Recent 
legislation in the Netherlands uses 1.0 x 10~6 per person per year as the 
maximum tolerable risk for new major hazard plants. For a specific industrial 
hazard that could kill a member of the public, a target value of 1.0 x 10~7 per 
person per year has been suggested. 
Although it is difficult to agree quantitative risk acceptability criteria, it is 
necessary to do so in order to be able to do QRA. On this basis, it is suggested 
that the risk acceptability criterion for pharmaceutical industry manufacturing 
plant accidents that could cause public fatalities should be less than 1x10~6 
per person per year shown in Table 7.4. 
Risks to process operators 
Quantitative risk acceptability criteria based on event frequencies have been 
widely used for ranking process risks in order of priority for action. A criterion 
that has often been used for assessing process hazards is that the risk of death 
for a plant operator should not exceed the risk of death for a fit adult staying at 
Table 7.4 Guidelines for QRA in the pharmaceutical industry 
Hazardous event 
Public fatality from a specific 
plant hazard 
Public fatality from 
all process hazards 
Process operator fatality from a 
specific plant hazard 
Process operator fatality 
from all process hazards 
Risk acceptability guideline 
<0.1 x 10~6 per person per year 
<1.0 x 10~6 per person per year 
<7.0 x 10~6 per person per year 
<35.0 x 10~6 per person per year
home. On this basis, the chemical industry for many years has aimed that the 
risk of death from all process hazards should have a probability of occurrence 
of less than 35.Ox 10~6 per year per person exposed. It was considered that the 
risk of death from a specific process hazard should be a fifth of the total and 
targeted at 7.0 x 10~6 per person per year. 
It has also been suggested that the risks to the public should be an order of 
magnitude less than that for process personnel. This suggestion, taken with the 
public risk guideline described previously, implies that the risks to plant 
operators should be less than 1 x 10~6 per person per year. This is of the 
same order of magnitude as the criterion derived by the chemical industry. Risk 
criteria for process operators in the pharmaceutical industry can be developed 
on a similar basis (see Table 7.4 on page 233). 
7.5.8 Risk assessment and validation 
Risk assessment by hazard study and process validation have had different 
histories during their evolution (see Figure 7.13). During the last decade, 
however, the two methodologies have drawn closer together in the pharmaceutical 
industry so that they overlap in several areas. Figure 7.14 shows these 
Method study 
HAZOP 
Continuous processes HAZOP 
Batch processes HAZOP 
Primary pharms. manufacture HAZOP 
pharmaceutical formulation Six-stage hazard study chemical process 
Six-stage hazard study pharmaceutical process 
Primary Pharmaceuticals manufacturing process 
Computer systems validation 
Non-aseptic pharmaceutical processes 
Water treatment processes 
Aseptic 
pharamceutical processes Sterilisation processes 
Analytical testing 
1960 1970 1980 1990 
Figure 7.13 A brief history of hazard study and process validation
Figure 7.14 The six-stage hazard study methodology and process validation for a 
typical pharmaceutical product 
areas of overlap diagrammatically. The diagram represents a six-stage hazard 
study applied to a typical pharmaceutical project life-cycle with the associated 
validation activities included. 
As mentioned earlier in this chapter, the six-stage hazard study consists of 
Hazard Study 1 (HSl) to get the facts, Hazard Study 2 (HS2) to identify 
significant hazards, Hazard Study 3 (HS3/HAZOP) to perform a hazard and 
operability study of the final design, Hazard Study 4 (HS4) and Hazard Study 5 
(HS5) to check that the hazards identified have been controlled to acceptable 
standards, and Hazard Study 6 (HS6) to review the project and lessons learned. 
Although Chapter 4 provided a full explanation of validation, it is useful to 
re-state the activities that overlap with the six-stage hazard study process. 
Process validation starts with the preparation of a User Requirements Specification 
(URS) followed by a Functional Specification (FS) for engineering 
design and procurement. Installation Qualification (IQ) and Operation Qualification 
(OQ) are performed to prove that the URS and FS have been met prior 
to the final process qualification or process validation. 
A quantitative analysis of several hazard studies showed that about 50% of 
the hazards identified by HAZOP were related to quality and validation issues. 
The use of the existing guidewords, thus, appeared to be effective from the 
Hazard study: HS1 HS2 HS3 HS4HS5 HS6 
Project phase: 
Research 
Process development 
Process and project definition 
Project design 
Procurement & construction 
Commissioning 
Operation 
Modifications 
De-commissioning 
Demolition 
Capital authorisation 
Process validation: URS FS IQOQ PVAL 
Time:
quality viewpoint. It was further improved by having validation experts in the 
hazard study teams. Unfortunately, any quality hazards identified as late as HS3 
by HAZOP could be costly in time and effort to prevent or protect against. The 
most important thing to do is to increase the emphasis on quality earlier in the 
life-cycle at HSl and 2. 
The hazard study of computers has always been difficult to perform with 
complete confidence that all the main hazards could be identified. The lack of 
confidence is due to the complexity and volume of the interactions between the 
hardware and the software. It is impossible to analyze all the computer codes in 
a reasonable time-scale, in even the simplest systems. Computer Hazard and 
Operability Study or CHAZOP was developed in an attempt to identify the 
significant hazards with reasonable confidence. CHAZOP has been successfully 
used with computer applications data flow and logic diagrams treating the 
computer operating systems and watchdogs as 'Black Boxes'. CHAZOP and 
similar techniques are still being improved to provide more confidence that the 
significant hazards can be identified. 
As explained in Chapter 4, the validation of computer and critical automated 
systems has advanced considerably over the last few years, building on the 
work of systems analysts, CHAZOP and process validation methods. Computer 
validation has concentrated on a life-cycle approach, building quality into 
computer systems from their conception. Computer validation is currently the 
most effective means of ensuring that computer systems hazards are controlled 
acceptably. 
The synergy between hazard study and computer validation in the pharmaceutical 
industry is now well established. Hazard study and computer validation 
operate together and share techniques and information produced by the 
function that is the most effective. 
7.6 Pharmaceutical industry SHE hazards 
The pharmaceutical industry has similar SHE hazards to those of the chemical 
industry, but to different degrees of severity. Chemical reaction, fire, explosion, 
toxic, environmental, occupational health, mechanical energy and radiation 
hazards are well described in the literature together with methods of assessing 
and controlling them. The chapters on primary and secondary production, 
process utilities and services, laboratory design, and process development and 
pilot plants also cover these hazards where relevant. This chapter will only 
briefly consider the particular aspects of these hazards that apply to the 
pharmaceutical industry. The hazards arising in specific pharmaceutical 
processes, which are not encountered elsewhere, will also be discussed briefly. 
Next Page
quality viewpoint. It was further improved by having validation experts in the 
hazard study teams. Unfortunately, any quality hazards identified as late as HS3 
by HAZOP could be costly in time and effort to prevent or protect against. The 
most important thing to do is to increase the emphasis on quality earlier in the 
life-cycle at HSl and 2. 
The hazard study of computers has always been difficult to perform with 
complete confidence that all the main hazards could be identified. The lack of 
confidence is due to the complexity and volume of the interactions between the 
hardware and the software. It is impossible to analyze all the computer codes in 
a reasonable time-scale, in even the simplest systems. Computer Hazard and 
Operability Study or CHAZOP was developed in an attempt to identify the 
significant hazards with reasonable confidence. CHAZOP has been successfully 
used with computer applications data flow and logic diagrams treating the 
computer operating systems and watchdogs as 'Black Boxes'. CHAZOP and 
similar techniques are still being improved to provide more confidence that the 
significant hazards can be identified. 
As explained in Chapter 4, the validation of computer and critical automated 
systems has advanced considerably over the last few years, building on the 
work of systems analysts, CHAZOP and process validation methods. Computer 
validation has concentrated on a life-cycle approach, building quality into 
computer systems from their conception. Computer validation is currently the 
most effective means of ensuring that computer systems hazards are controlled 
acceptably. 
The synergy between hazard study and computer validation in the pharmaceutical 
industry is now well established. Hazard study and computer validation 
operate together and share techniques and information produced by the 
function that is the most effective. 
7.6 Pharmaceutical industry SHE hazards 
The pharmaceutical industry has similar SHE hazards to those of the chemical 
industry, but to different degrees of severity. Chemical reaction, fire, explosion, 
toxic, environmental, occupational health, mechanical energy and radiation 
hazards are well described in the literature together with methods of assessing 
and controlling them. The chapters on primary and secondary production, 
process utilities and services, laboratory design, and process development and 
pilot plants also cover these hazards where relevant. This chapter will only 
briefly consider the particular aspects of these hazards that apply to the 
pharmaceutical industry. The hazards arising in specific pharmaceutical 
processes, which are not encountered elsewhere, will also be discussed briefly. 
Previous Page
7.6.1 Chemical reaction hazards 
Chemical reaction hazards assessment 
As explained in Chapter 5, the primary production processes to produce active 
drugs involve a wide variety of complex reactions and reaction sequences. 
Many of these reactions may be exothermic or may evolve gases at high rates, 
and could cause reactor over-pressure. It is, thus, essential to establish the basis 
for safe operation in the laboratory before scaling up such reactions. It is good 
practice to perform a methodical assessment (described by Barton and Rogers 
in the bibliography) summarized as follows: 
define the process chemistry and operating conditions and the process 
equipment to be used; 
evaluate the chemical reaction hazards of the process, including potential 
maloperation; 
select and specify safety measures; 
implement and maintain the selected safety measures. 
There are many published procedures for evaluating chemical reaction 
hazards. Whatever procedure is used, it is essential that tests are performed 
and interpreted by qualified people. This is because there are many factors that 
may affect the test data such as sample size, container material, heating rate, 
thermal inertia and endothermic effects. 
Control of runaway reactions 
Runaway reactions are thermally unstable reactions where the heat of reaction 
can raise the temperature of the reactants sufficiently to accelerate the reaction 
rate out of control. The temperature at which the runaway starts is often termed 
the onset temperature. Such reactions are normally controlled by cooling the 
reactor, or by controlling the addition of the reactants. Loss of reactor cooling or 
agitation during the course of an exothermic reaction are two of the commonest 
causes of runaway reactions. A runaway reaction can cause the reactor contents 
to boil, generate vapour or explode, and over-pressurize the reactor. 
There are several protective measures that can be used to mitigate the effects 
of a runaway reaction. The most common protection is emergency venting, but 
containment, crash cooling, drown-out and reaction inhibition provide other 
options. 
Reactor venting 
Reactor over-pressurization can occur by overcharging with compressed gases 
or liquids, by excessive vapour generation due to overheating, or by a runaway 
reaction. Such events are normally avoided by adopting suitable operating
procedures and control systems. When control is lost, the most effective way to 
prevent damage to the reactor is to relieve the pressure through an emergency 
relief system. The design of reactor pressure relief systems is well described in 
the literature and will not be explained here. However, some key questions to 
ask are as follows: 
what is the maximum pressure that the vessel can contain? 
what pressure will activate the relief system? 
will the relieved material be a liquid, a vapour or a two-phase mixture? 
what is the maximum expected relief rate to avoid over-pressurization? 
is the area of the relief device sufficient to handle the maximum expected 
relief rate? 
is the pressure drop in the relief system low enough to prevent overpressurization 
during venting? 
will the relief device survive in normal reactor operations (for example, 
bursting disk under vacuum)? 
will the relief device re-seal after depressurization? 
is the material ejected from the reactor toxic or environmentally harmful? 
does the relief system exhaust to atmosphere in a safe place? 
7.6.2 Fire and explosion haiards 
In the pharmaceutical industry, fire and explosion hazards arise most frequently 
when handling flammable solvents or finely divided organic powders. Flammable 
materials or mixtures are frequently used for the reactions such as 
hydrogenation, nitration, Grignard reaction, and oxidation in primary production 
processes. Occasionally chemical intermediates or by-products in primary 
production processes may be pyrophoric or explosive. Flammable solvents and 
finely divided solids are also encountered in purification and secondary 
production processes. It is, thus, essential to obtain information about the fire 
and explosion properties of all materials that occur in the manufacturing 
processes in order to establish a basis for safe operation. 
Material fire and explosion properties 
All materials used must be tested for fire and explosion properties. In the 
pharmaceutical industry it is very important to test dusts and finely divided 
powder, as almost all of these can form explosive mixtures with air. The test 
methods and procedures are well described in the literature and will not be 
described here. It is essential to obtain specialist advice to interpret the test 
results to achieve a safe process design, although the key parameters that 
influence safe process design are as follows:
gases and vapours: 
lower explosive limit in air; 
upper explosive limit in air; 
critical oxygen content; 
density; 
minimum ignition energy; 
auto-ignition temperature; 
minimum flame diameter, 
flammable and highly flammable liquids: 
flash point; 
boiling point; 
lower explosive limit in air; 
upper explosive limit in air; 
auto-ignition temperature; 
vapour density, 
finely divided powders and dusts: 
dust classification; 
maximum dust explosion pressure; 
critical oxygen content; 
St rating (maximum rate of pressure rise during explosion); 
minimum ignition energy; 
train firing. 
Area classification of plants handling flammable gases and liquids 
The handling of flammable gases in the pharmaceutical industry is usually 
restricted to hydrogenation processes and to fuel gases supplied for process 
utilities and services. The inventories are usually small and leaks can be well 
controlled, so that the probability of an uncontained gas cloud explosion in the 
open air is very low. The main hazards occur inside buildings, where even small 
leaks of flammable gas can form explosive mixtures in air. Risk management of 
flammable gases in buildings relies on leak prevention, containment, ventilation, 
and control of ignition sources. 
The inventories of flammable liquids in pharmaceutical processes can often 
be substantial, so fire and vapour cloud explosions are significant hazards. 
These hazards are exacerbated inside buildings, particularly when solvents are 
handled at temperatures above their flash point. Risk management relies on 
similar controls to those used for flammable gases with the additional 
possibility of vapour knock-down and foam systems to control leaks or 
spillages.
The hazards of handling flammable gases and liquids in plant areas are 
identified and risks assessed by a team of suitably qualified people to provide 
suitable controls. This activity is called Area Classification (British Standard 
5345) and is performed as follows: 
list all flammable and combustible materials used in the area to be studied, 
with quantities; 
obtain all relevant fire and explosion properties for the materials listed; 
obtain an engineering drawing of the area to be studied and identify and list 
the possible sources of flammable atmospheres; 
study the area using the 'Source of Hazard' method described in BS 5345; 
estimate the extent of the following zones around each source using standard 
procedures: 
o zone 0: A zone in which a flammable atmosphere is continuously present 
for long periods; 
o zone 1: A zone in which a flammable atmosphere is likely to occur in 
normal operation; 
o zone 2: A zone in which a flammable atmosphere is not likely to occur in 
normal operation and, if it occurs, will only exist for a short time; 
o non-hazardous: A zone in which a flammable atmosphere is not likely to 
occur at all. 
record the decisions on an Area Classification drawing; 
decide the review frequency. 
Dust explosion hazards 
It is worth re-iterating that most finely divided powders handled in pharmaceutical 
production processes can form explosive mixtures in air. Dust explosion 
properties are determined in specialized laboratories by qualified staff that 
use approved test equipment and procedures. The tests and their interpretation 
are well described in the literature, but are beyond the scope of this chapter. 
However, a few rules-of-thumb may be useful for preliminary process design 
and risk assessments as follows: 
most organic materials with a particle size less than 75 microns will form 
explosive mixtures in air; 
the lower explosive limit in air for most organic dust clouds is between 
15-60 gm m~3 depending on the temperature but independent of ignition 
energy. {These are very dense clouds that would obscure a 100 W light at 
about two metres); 
the upper explosive limit is generally very high at 2-6 kg m~3. Most dust 
explosions will generate a final contained pressure that is about ten times the
start pressure. {This means that atmospheric pressure systems designed to 
withstand 10 Barg should contain a typical dust explosion)) 
most explosive dusts can be inerted by limiting the atmospheric oxygen 
concentration to less than 8% v/v; 
the most severe consequences arise from secondary dust explosions that are 
caused by the ignition of very large dust clouds generated by the primary 
explosion dislodging dust held on ledges, etc. in the vicinity. 
There are several methods of protection against dust explosion hazards. The 
first step is to eliminate potential ignition sources. The possibility for incendive 
electrostatic sparks must be removed by adequate earthing of metal conductors 
and electrostatic charges. The next step is to provide protection against dust 
explosion. The most well known methods are explosion venting, inerting, 
suppression and containment. The protection most frequently used for dryers, 
storage vessels and conveying systems is to vent the explosion to the atmosphere 
via rupture disks or panels. Venting must be to a safe place and must not 
cause environmental hazards. Inerting is often used when venting to a safe area 
is not possible or if the vented material can cause environmental hazards. 
Containment is frequently used for milling and dust separation processes where 
the equipment can be made to withstand the dust explosion pressure with 
reasonable economy. Suppression can present quality problems and is usually 
only used for systems where there are hybrid mixtures of dusts and flammable 
vapours or gases. 
7.6.3 Occupational health haiards 
Occupational health hazards arise in the workplace when uncontrolled harmful 
substances or conditions exist that can adversely affect the health of the workers. 
The exposure of staff to external hazards from the environment and from their 
life outside work is also important as it can affect their response to exposure at 
work. This chapter will only consider the effects of workplace hazards. 
To achieve good occupational health in the workplace, hazard identification, 
risk assessment and the selection of suitable controls against hazardous 
exposure are essential. Engineering and procedural controls must also take 
account of the additional controls provided by occupational hygiene. For 
example, in certain circumstances, it may be necessary to monitor workplace 
emissions or to provide health surveillance of the operating staff. 
Occupational exposure limits 
Toxicologists, epidemiologists, physicians, occupational hygienists and 
research workers provide the essential information for defining the Occupa-
tional Exposure Limits (OELs) that are used to define and maintain healthy 
working conditions. The information for setting these criteria is either obtained 
by direct experiment or by modelling data from experiments performed in 
similar systems. The complexity of some of these issues is outside the scope of 
this brief review. 
Occupational health risks arise from operator exposure to materials and 
physical conditions that occur in the working environment. The materials can 
be chemicals, biologically active substances or ancillary materials used in the 
workplace. Exposure to these materials can affect the health of the person 
exposed by inhalation, skin contact and absorption, or ingestion. The immediate 
effects are termed acute effects. If exposure is over a long period of time and 
the effects persist, these are termed chronic effects. 
(a) Materials 
The OELs for materials that cause chronic effects are usually based on an 
8-hour time weighted average exposure. Highly active materials are allocated 
shorter times such as the 15-minute time weighted average exposure. Some 
materials may be allocated both long and short-term exposure limits. 
The dose-response relationship for a toxic substance is the relationship 
between the concentration at the site of ingress and the intensity of the effect on 
the recipient. It is difficult to interpret dose-response relationships for a 
particular individual, so the assessment of occupational health risks from 
toxic materials requires considerable knowledge and experience. 
Pharmaceutical research and development of biologically active compounds 
generates occupational health hazards for which the exposure limits are often 
unknown. New compounds are thus tested for toxic effects as well as therapeutic 
efficacy as a key part of the research and development programme. In the early 
research and development stages it is essential to assess substances for occupational 
health risks even though reliable data may not be available. This is done by 
defining in-house OELs on the basis of experience and available models 
assuming that there is a threshold below which there are no adverse effects. 
These in-house OELs or preliminary standards may then be altered to match 
the experimental data obtained as research progresses. From the process design 
viewpoint, the in-house exposure limits are used as the best information 
available, but it is important to record the fact in the process documentation. 
Subsequent changes to OELs will require a re-examination of those system 
elements that are affected. 
A particular problem encountered in pharmaceutical research involving 
animals or biotechnology is allergic reactions. Allergy depends very much on 
the individual exposed. Susceptible individuals may respond to minute
amounts of allergen that are too small to measure. In these cases it is impossible 
to define a reliable OEL for control purposes because a threshold cannot be 
determined. In these circumstances, it is normal to work to approved codes of 
practice for known allergens, to provide personal protection, and to perform 
health surveillance of operators exposed. 
Great care is needed to interpret occupational health data. As a simple 
example, the OEL for a nuisance dust is often loosely quoted as lOmgm"3; 
8 hr TWA (Time Weighted Average). However, this value is strictly for total 
inhalable dust concentration: the OEL for the respirable fraction is 5mgm~3; 
8hr TWA. Table 7.5 provides some idea of the range of OELs that may be 
encountered in the pharmaceutical industry for inhaled substances. These 
simple examples are only intended to be used for discussing process 
design issues with occupational health practitioners and are not provided as 
standards. 
(b) Physical conditions 
The assessment of the effects of physical conditions such as temperature, 
humidity, noise, vibration, and electromagnetic radiation is more straightforward 
than for materials because they have been well researched and the dosage 
and effects can be monitored more reliably. Physical effects that are not doserelated 
such as the stresses and strains arising from manual operations are more 
difficult to assess. Back problems, repetitive strain injury and eye strain are 
usually controlled by ergonomic workplace and equipment design backed up 
by education and training based on the findings of medical research and 
ergonomics. Recent legislation requires that such risks should be assessed at the 
design stage of new manual systems. 
Most of the physical hazards that can occur in the workplace can be 
controlled by following recognized codes of practice to control dose-related 
exposure. The number and change rate of physical hazards is much less than for 
Table 7.5 A typical range of occupational exposure limits encountered in the 
pharmaceutical industry 
Description of 
inhaled substance 
'Nuisance dusts' 
Toxic substances 
Highly toxic substances 
Extremely toxic substances 
Range of occupational 
exposure limits 
l-10mgm"3 
0.1-1 mgm~3 
0.01-0.1 mgm~3 
<0.01mgm"3 
Typical example 
Starch dust 
Solvents, Common 
chemicals 
Cytotoxins 
Carcinogens
chemical and biological hazards which makes physical hazards relatively 
simpler to study. The main physical hazards to consider are heat, humidity, 
air quality, noise, vibration, ionizing radiation, non-ionizing radiation, and 
electricity. Typical occupational health criteria for physical hazards are given in 
Table 7.6. These values are solely for discussion purposes with the relevant 
experts. A qualified occupational hygienist should always decide the relevant 
criteria for a pharmaceutical project. 
Occupational health legislation 
The regulations governing occupational health management now established 
throughout the western world all require risk assessment of occupational health 
Table 7.6 Typical physical hazard occupational health criteria 
Workplace physical 
hazard 
Temperature 
Humidity 
Air change rate 
Noise 
Vibration 
Non-ionizing 
radiation 
Ionizing radiation 
Typical occupational health 
criteria 
30.0 deg Centigrade 
(Wet bulb) 
26.7 
25.0 
40%-60% R.H. 
> 10 changes of air/hour 
>90 dB(A) (LEP,d) 
Magnitude: 2.8ms"2rms 
Frequency: 
Whole body: 0.5-4.0Bz. 
Hand — arm: 8-1000 Hz. 
<50mW/cm2 @5cms 
< 10 mW/cm2 in workplace 
Depends on laser classification 
50 mSv (5 rem/year) 
5 mSv (0.5 rem/year) 
Comments 
Continuous 
light work 
Continuous 
moderate work 
Continuous 
heavy work 
Guidance for 
comfort only 
Rule-of-thumb 
guide only 
Ear protection 
required at or 
above this level 
for 8 hr TWA 
exposure 
8 hr TWA level 
for taking 
preventive action 
Microwaves 
(2450MHz) 
Microwaves 
Laser light 
Total exposure 
to radiation (ICRP) 
for workers 
whole body 
Any other person; 
whole body
hazards. In the UK, the Control Of Substances Hazardous to Health Regulations 
1994 (COSHH) requires the employer to assess the workplace risks from 
handling substances hazardous to health, to identify any control or personal 
protection measures needed, to maintain these measures and where necessary 
monitor workplace exposure and/or provide health surveillance. COSHH also 
requires the employer to provide information, instruction and training about the 
hazards, the risks and the controls required and also to keep auditable records. 
Legislation will often define specific occupational exposure limits for 
substances or physical conditions that are known to present health risks. The 
limits for toxic substances under the COSHH legislation, for example, are 
expressed as Maximum Exposure Limits (MELs) and Occupational Exposure 
Standards (OESs). MELs are allocated to substances such as carcinogens that 
have known serious health effects but for which no threshold of effect can be 
identified. OESs are allocated to substances that could cause serious health 
effects above a specific and clearly definable threshold exposure. 
Occupational health systems description 
Occupational health hazard identification and risk assessment can only be 
performed effectively with a clearly defined system model. The minimum 
requirement is for a simple process block diagram and a brief description of the 
activities that can give rise to occupational health hazards. A list of process 
operations and operator tasks is essential to determine the extent of exposure. 
The list can be used to prepare an activity diagram of the operator actions and 
movements suitable for hazard study. The activity diagram information can 
then be used to plot operator movements on the workplace layout drawing. The 
model can be improved considerably by indicating the harmful emissions on 
the same drawing to identify the interactions between the operator, process and 
emissions. 
Occupational health controls 
Occupational health hazards are identified by a team of knowledgeable people 
studying the system model and activity diagram. It is helpful to include an 
occupational hygienist in the team to interpret the applicable exposure limits 
and advise on the best controls for emissions that cannot be eliminated. Typical 
controls are based on containment, ventilated enclosures, local exhaust ventilation, 
dilution ventilation, personal protection or combinations of these main 
types. If possible, personal protection should be avoided as it hampers operator 
activities and is costly to implement and maintain.
Occupational health impact assessment 
For a typical pharmaceutical project, it is important to write a formal 'occupational 
health impact statement' that describes the occupational health hazards 
identified and the principles of the control regime needed to comply with 
legislation and in-house standards. In the six-stage hazard study methodology 
this is done as part of hazard studies 1 and 2. The activities necessary to 
complete this assessment are as follows: 
identify the occupational health hazards present and list them. For chemical 
and biological materials identify the amounts used in the process and other 
hazards that they may present (Materials Hazard Checklist); 
obtain the Material Safety Data Sheets (MSDS) for each hazardous 
substance identified. If a MSDS is not available, consult an occupational 
health specialist for guidance, particularly if there is no information about 
OELs or hazard categories for specific materials; 
for each hazard, identify the potential routes of entry into the bodies of the 
operators or staff exposed to the hazards; 
state the control principles to be used to meet the OELs or other occupational 
health criteria for each hazard. The control principles for maintenance, 
cleaning activities, emergencies and abnormal operation are particularly 
important; 
specify the control measures that will be used and state the test and 
maintenance procedures to ensure that they remain effective. The exact 
details may not be known at the early stages, so the aim here is to provide 
engineering guidance; 
state whether health surveillance or exposure monitoring will be required; 
specify any personal protective equipment (PPE) that may be required; 
state whether any specific training will be necessary for hazard awareness, 
use of PPE, etc.; 
define the actions and responsibilities for further occupational health 
assessments that may be required, such as COSHH assessments that will 
be needed during construction, commissioning and start-up; 
record all the findings and necessary actions in a formal report. 
7.6.4 Environmental hazards 
The protection of the environment is a major concern of modern society, but 
opinions about the best way forward vary considerably. In the context of the 
environmental risks to a pharmaceutical project, the whole life-cycle must be 
assessed as far into the future as can be reasonably predicted. The following
paragraphs provide a brief overview of environmental risk assessment and 
current environmental legislation. 
Environmental hazards in the pharmaceutical industry 
In the pharmaceutical industry the main environmental hazards associated with 
routine operations are solvent emissions to air and emissions to the aquatic 
environment. Releases due to loss of containment in an accident or during a fire 
or other emergency can also cause pollution of the aquatic and ground 
environments. 
(a) Routine solvent emissions to air 
The pharmaceutical industry emits relatively small amounts of volatile organic 
compounds (VOCs) but is, nevertheless, under pressure to reduce existing 
releases. The abatement of routine batch process releases at source is difficult as 
VOC emissions are usually of short duration and high concentration. The best 
available technology not entailing excessive costs (BATNEEC) for such 
emissions is usually 'end of pipe abatement' technologies such as adsorption, 
absorption, condensation, etc. Unfortunately such measures increasingly 
require the use of manifolds and catchpots that can cause additional problems 
from cross-contamination of the product or fire and explosion hazards. 
The prevention of cross-contamination is a particular problem in purification 
and formulation processes where systems to remove solvent vapours are needed 
to protect the environment. In such systems, the containment of potentially 
explosive atmospheres may generate an explosion hazard that will require 
additional protection measures. One solution to this problem is to use inert 
atmospheres to minimize the explosion risks, but this then adds the risk of 
asphyxiation of operators and will require suitable controls in enclosed areas. 
(b) Routine emissions to the aquatic environment 
Aqueous discharges from pharmaceutical processes are usually collected and 
pretreated to reduce the environmental impact before release off-site. The 
relatively small volumes involved rarely make biological treatment on-site 
economical and so this is usually performed at the local sewage works. Solvent 
discharges are recovered if possible either on-site or off-site. If recovery is not 
possible it may be possible to use waste solvents as a fuel source during 
incineration. 
It is important to be able to monitor routine discharges to drain from 
processes that involve polluting chemicals. Process drains should not be buried 
and should have suitable access for regular inspection. Surface water and 
process drains should be segregated and studied to identify any potential
interconnections during storms or emergencies. Any bunds, catchment basins 
or effluent pits should be leak proof and regularly checked for integrity to 
prevent accidental leakages. 
(c) Loss of containment 
Emergency relief discharges of volatile materials or dusts can contaminate both 
the aquatic and ground environments. This is a major concern in primary 
production as the chemicals and intermediates used to prepare crude bulk drugs 
are all potential pollutants and some may be severe pollutants. The release of 
such chemicals to atmosphere as a result of a runaway reaction or major 
spillage, for example, could be potentially damaging to the environment. 
Catchment or 'dump' systems to collect any emergency emissions may be 
essential to comply with legislation. Unfortunately, if manifolds or interconnections 
are used for this purpose they may cause explosion, over-pressure, 
or fire hazards that must be controlled by additional protective measures. 
Solids handling and particulates can cause risk to the environment at all 
stages of pharmaceutical production. As previously explained, most of the dry 
solids handled in pharmaceutical processes can cause a dust explosion hazard. 
Dust explosions can be contained in equipment designed to withstand 
>10Barg, pressure, and vented, inerted, or suppressed in weaker equipment. 
If dust explosion venting is used, it may cause serious pollution and more costly 
alternatives of containment and suppression will be needed to protect the 
environment. The cost of cleaning up soil contamination from emergency 
releases of biologically active dusts or solids can be prohibitive. 
A large fire on a primary production process or warehouse can lead to 
environmental pollution. Apart from the environmental damage arising from 
smoke and soot, fire-fighting water containing dissolved chemicals can cause 
pollution of local watercourses and damage to water treatment works. Firewater 
retention systems may be needed to prevent the contamination of local watercourses 
or ground waters. Fortunately, formulated products present fewer 
pollution problems as they are usually hermetically contained in small 
quantities. 
(d) Early identification of environmental hazards 
The environmental, safety and health risks must always be considered together 
rather than individually, as there is considerable interaction between them. 
Environmental protection is usually very costly, so it is important to attempt 
to avoid environmental hazards by eliminating them at the outset. Since 
pharmaceutical processes are usually registered before a capital project is 
started, it is thus important to consider environmental hazards at the research
and development stages. At the very least, researchers should perform a 
rudimentary 'What If?' or Preliminary Hazard Analysis (PHA) to assess 
chemical routes or process alternatives for environmental hazards. 
Environmental legislation 
In Europe the Directive 85/337/EEC 'The assessment of the effects of certain 
public and private projects on the environment' came into effect in 1988. The 
Directive requires an environmental impact assessment for all projects that 
could have significant environmental impact before consent to proceed is given. 
It has been incorporated into the legislation throughout the European Union, 
and in the UK by The Environmental Protection Act 1990 that is now 
implemented by the Environment Agency. Established under the Environment 
Act 1995, the Environment Agency took over the functions of Her Majesty's 
Inspectorate of Pollution, the National Rivers Authority, Waste Regulatory 
Authorities, and some parts of the Department of the Environment (internet 
web-site: http://www.environment-agency.gov.uk). 
The UK Environmental Protection Act 1990 requires that certain prescribed 
processes may only be operated with an authorization. The Act defines two 
systems of pollution control, Integrated Pollution Control (IPC) and Local 
Authority Air Pollution Control (LAAPC). The Environment Agency regulates 
IPC and also authorizes prescribed processes. The local authorities and 
metropolitan boroughs enforce and authorize LAAPC, which covers air 
pollution only. The local authorities also administer the Town and Country 
Planning (Assessment of Environmental Effects) Regulations 1988 for which 
there is a guide to performing environmental assessment procedures (HMSO 
1992). Pharmaceutical production processes require environmental assessment 
under Schedule 2 of these regulations only if they have significant effects on the 
environment. The industry also has a 'Duty of Care' under Part 2 of the UK 
Environmental Protection Act 1990 for assessing and disposing of its wastes, 
even when they are handled by contractors. To decide the level of compliance 
required by the regulations it is necessary to assess the environmental hazards 
for all projects. 
Environmental protection systems description 
Environmental protection systems are usually an integral part of pharmaceutical 
process systems and appear on the same engineering drawings as other 
systems. To clarify the interactions of environmental protection and process 
systems it is advisable to prepare a separate block diagram that shows all the 
environmental contact points with the process systems. All gaseous, liquid and 
solid emissions should be clearly identified together with estimates of the
emission rates. The procedures for normal operation, cleaning and maintenance 
should also be studied to identify how process interactions could generate 
normal and abnormal emissions. Any emergency procedures or provisions such 
as explosion relief must also be included in the systems description. 
Environmental hazards identification 
There is much quantitative information available to identify how substances can 
pollute water. Regulations make use of this information by categorizing 
substances for their pollution effects. The European Directive 76/464/EEC 
defined the 'Black' and 'Grey' lists to categorize substances for control 
purposes. Substances on the 'Black' list are considered to be the most harmful 
and pollution from these must be eliminated. Substances on the 'Grey' list are 
considered to be less harmful and pollution levels are controlled at national 
level. The German Chemical Industries Association (VCI) has developed a 
system for rating substances for their water endangering potential, and have 
published tables for a wide range of materials. 
Environmental risk assessment 
An environmental risk assessment is required internationally by law for most 
projects that could have significant effects on the environment. The format of 
the environmental risk assessment may be prescribed by some regulations. The 
reader is recommended to read 'A Guide to Risk Assessment and Risk 
Management for Environmental Protection' (HMSO 1995) for an informative 
description of environmental risk assessment. Although simple risk ranking can 
be used within a project to make decisions about alternative courses of action, 
formal approval from the relevant authority may require more quantitative 
assessment to prove compliance with their criteria. 
The aim of most assessments is to ensure that the project management 
consider the environmental issues at the earliest possible stages of the project. 
Suitable action can then be taken to prevent environmental damage if necessary. 
Environmental risk acceptability criteria 
Environmental risk acceptability criteria have become more stringent due to 
research on the environment that has revealed many previously unsuspected 
sources of damage, and that has raised levels of public concern for the 
environment. General principles such as the 'Precautionary Principle', 'As 
Low as Reasonably Practical' (ALARP), 'Best Available Techniques Not 
Entailing Excessive Cost' (BATNEEC), and 'Best Practicable Environmental 
Option' (BPEO) have been discussed as bases for setting criteria, and some 
have been developed within legal frameworks.
Environmental risk acceptability criteria are defined separately for gaseous, 
aqueous and solids emissions to atmosphere, water courses, ground water and 
soil. The limits imposed by the authority that governs a project will vary 
considerably and it is essential to define these at the project outset. An 
environmental impact assessment must be made so that the project design 
complies with these limits. 
Environmental impact assessment 
Although some pharmaceutical projects may not require a formal environmental 
impact assessment by law, it is essential to perform the assessment for 
project design purposes and to meet SHE management criteria. A typical 
environmental impact assessment should include the following headings: 
site selection; 
visual impact; 
building and construction; 
normal emissions; 
abnormal emissions; 
site remediation. 
7.6.5 Specific pharmaceutical process hazards 
Laboratories and pilot plants 
(a) Laboratories 
As explained in Chapter 9, research, development, production, analytical and 
quality control laboratories are designed and engineered to high standards, and 
are typically operated under Good Laboratory Practice (GLP) by experienced 
and well trained staff. Laboratory risk assessments are performed to comply 
with legislation, such as the UK COSHH regulations, during the design and 
engineering of new laboratory projects. Laboratories are extremely important 
business assets. 
The main risks in laboratories arise from uncontrolled changes to the original 
design and operating systems. For example, when new equipment is installed it 
will usually contain integrated circuits and computer controls. The ease with 
which the software can be modified may allow in-built safeguards to be 
inactivated or to generate unexpected hazards. The new owner of such equipment 
may lack the knowledge to assess its hazards and inadvertently cause an accident. 
The use of automated equipment or robotics to perform potentially violent 
chemical reactions can also lead to accidents in laboratories. It is essential in
these circumstances to perform a rigorous HAZOP and CHAZOP to define safe 
operating procedures, to enable validation, and to implement adequate change 
controls to avoid unacceptable risks. 
Some laboratory equipment may incorporate hazardous materials in a way 
that the purchaser may not be aware of. An example of this is the use of 
Nuclear Magnetic Resonance (NMR) equipment. NMR equipment uses superconducting 
magnets that are cooled by liquid nitrogen and helium. The cooling 
systems are provided with emergency pressure relief to prevent hazardous overpressurization 
in the event of overheating. Unless suitable ducting to atmosphere 
is provided, the pressure relief may discharge gases directly into the 
working area where anyone present could be asphyxiated. 
Scaling up the use of liquid nitrogen for storing tissues, etc. in closed 
laboratories or confined spaces is another hazard that may not be recognized 
without a hazard study. Laboratory workers can become very accustomed to 
using small quantities of liquid nitrogen but may forget the asphyxiation hazard 
if the scale of use increases. Whenever significant amounts of liquid nitrogen 
are to be used it is essential to perform a risk assessment beforehand to design 
safe handling and control systems. 
The hazards of using fume cupboards on a temporary basis without suitable 
fire and explosion protection are well known. This problem can be encountered 
in laboratories where there is a high rate of change and fume cupboard space is 
limited and can be exacerbated when potentially exothermic reactions, or 
reactions involving flammable liquids, are run automatically outside normal 
working hours. It is essential to implement a strict change control system for 
such circumstances. 
(b) Pilot plants 
The design of pilot plants is described in Chapter 10. However, effective risk 
assessment of new pilot plants is often difficult because it is not possible to 
define exactly what the plant will be used for. This problem is usually addressed 
by specifying 'Worst Case' and 'Typical' process conditions and materials to 
define a reasonably realistic model suitable for risk assessment. 
The main hazard in pilot plants is uncontrolled change. Once a pilot plant 
has been built and is in operation, strict change control procedures must be 
enforced. Comparison of proposed changes with the original system design can 
help to decide whether further risk assessment is necessary. 
A six-stage hazard study and risk assessment for new pilot plant projects 
will ensure that the users and engineers can agree the user requirements. 
The added advantage is that the methodology may generate new ideas and 
eliminate significant hazards before any capital is spent.
Crude bulk drug production 
The production of pharmaceutical intermediates and crude bulk drugs involving 
fine chemical or biotechnological batch processes involves many hazards 
such as fire, explosion, toxicity, pollution, product contamination, health 
hazards and energy release that are well known both inside and outside the 
industry. Most of the processes that contain such hazards are designed using 
codes of practice, hazard study and risk assessment to minimize the risks. 
The following list of problems that have been encountered and successfully 
resolved by using hazard study and risk assessment indicates the range of 
application: 
the design, operation and maintenance of safe systems for handling toxic 
materials; 
control of potentially exothermic reactions; 
effluent control and environmental hazard control systems design, operation 
and maintenance; 
nitrogen inerting systems design, operation and maintenance; 
safe systems of operation using batch process control computers; 
dust explosion prevention and control systems design, operation and 
maintenance; 
electrical earthing systems design, operation and maintenance; 
fire protection and prevention systems design, operation and maintenance; 
sampling systems design, operation and maintenance; 
fermenter 'Off gas' filtration; 
fermenter downstream processing; 
cleaning and maintenance systems and procedures; 
designing process systems to cope with the increasing activity and cost of 
bulk drugs. 
Purification 
Bulk drug purification is the final stage of primary production and produces the 
purest material in the product supply chain (see Chapter 5). For many years 
effective hazard study and risk assessment of the production processes has 
enabled this purity to be achieved safely, securely and with minimal environmental 
impact. 
Purification processes involve mainly physical changes to the crude drug. 
The process hazards involved may be less severe than those encountered in 
crudes production and the main concern is product quality. The typical 
purification operations of dissolution, carbon adsorption, filtration, chromatographic 
processes, ion exchange, drying, milling and so on, are all amenable to
conventional hazard study and risk assessment. The list of known hazards 
would include dust explosions, solvent fires, environmental pollution and many 
of the process hazards associated with cleaning, sampling and maintenance that 
were listed for the crudes processes. However, it is the hazards to product 
quality that require particular attention. Hazard study, particularly HAZOP, can 
contribute to improved operability and quality of purification processes. Risk 
assessment may also be used to balance quality criteria and SHE criteria. 
Quality assurance may sometimes compete with SHE criteria. One example 
is the routine testing of fire-fighting systems in bulk crude and drug purification 
facilities. Reliable fire prevention and protection is essential to protect the 
business from potentially serious interruption. The problem of testing sprinklers, 
water deluge systems and foam pourers, without causing product quality 
problems has raised many arguments between the quality assurance staff and 
the fire engineers in the past. 
Secondary production 
The design of second production processes has been described in Chapter 6, so 
only specific hazards and risk assessment topics are considered here. 
(a) Formulation 
The cleanliness and product security of formulation processes is obtained by 
removing ancillary equipment from the processing area to 'Plant Rooms'. The 
design of the plant rooms is often less demanding than for processing areas. 
Designers sometimes regard plant rooms as peripheral and only give design 
priority to such rooms when they are critical to GMP, such as for the provision 
of demineralized water or water for injection. Even then, the room layout is 
rarely optimized. Plant rooms are often congested, difficult to access, and 
difficult to work in. Valves and controls are often badly positioned for manual 
operation or maintenance. Plant rooms located in the process area ceiling space 
or in basements may have low headroom and rarely have natural lighting, so 
require reliable emergency lighting during electrical power cuts or fires. Safe 
systems of work for lone working in plant rooms are essential. In addition to 
these hazards, plant rooms may sometimes be used for unauthorized storage of 
equipment and materials. Plant rooms are essential targets for hazard study and 
any pharmaceutical project for a new facility should include the hazard study of 
plant rooms in a six-stage hazard study programme. 
The major problem of granulation and tabletting processes is the control of 
biologically active and combustible dust clouds. As was the case with bulk drug 
purification processes, a key requirement of the process design is the control of 
such dusts by containment to minimize operator exposure and to comply with
GMPs. Containment may generate potential harm to the operators and to the 
environment from dust explosions in equipment such as granulators, dryers, 
mills and conveying systems. The balance of risk between toxic and combustible 
dust hazards will govern the basic process design and is best achieved as 
part of a six-stage hazard study. (If flammable solvents are used, the risks are 
increased considerably). 
Alternatively, for a new formulation project, an inherently dust free process 
may be sought. Direct compression, microwave drying, mixer-granulators, and 
other such developments aimed at eliminating dust exposure and explosion 
problems may bring their own particular hazards. The selection of the process 
must be done as early in the project as possible to allow time to evaluate such 
options satisfactorily. 
Tablet or spheroid film coating with solutions in flammable solvents 
involves the hazards of fire and environmental pollution. These hazards can 
be eliminated if aqueous coating can be used instead, although very powerful 
incentives may be needed to develop aqueous film coating for existing solventcoated 
products because of re-registration problems. A comprehensive hazard 
study together with a combination of QRA and cost benefit analysis can help to 
decide the most effective alternative. 
A typical formulation project will include many items of equipment that are 
purchased and installed as modular packages 'off the shelf such as autoclaves, 
sterilizers, freeze-dryers, chillers, Water for Injection (WFI) units, demineralized 
water units, centrifugation units, fluid bed dryers. The hazards that can 
arise will vary depending on the materials processed and the type of process 
performed. It is very important to determine the level of hazard study and risk 
assessment that has been performed by the supplier and to check that it meets 
SHE and quality criteria. Many suppliers perform FMEA, HAZOP and risk 
assessments as part of their equipment design process, but integrating their 
equipment into a pharmaceutical project may generate unforeseen hazards. 
In many project situations, it may be necessary to perform a risk assessment of 
each module before it is installed in the pharmaceutical system. 
(b) Packaging 
New packaging facility design and operation can be improved considerably by 
six-stage hazard study. Although the safety, health and environmental hazards 
involved may not be as severe as in other pharmaceutical processing activities, 
the potential quality improvements, the minimization of minor accidents and 
the improvements in layout and operability that can be achieved are very 
worthwhile. Hazard study and risk assessment are particularly beneficial if the 
project is to accommodate aseptic filling or new packaging technology. The
increasing use of computerized control systems for packing lines may require 
FMEA and CHAZOP to complement HAZOP during a six-stage hazard study 
and as part of the validation exercise. 
(c) Warehousing and distribution 
Warehouses containing expensive pharmaceuticals are always scrutinized 
closely by accountants as major centres of working capital. However, the 
high stock value may not be as important as the potential business interruption 
arising if it were lost. The hazard study and risk assessment of warehouses and 
their contents is thus very important to pharmaceutical business activity. 
Fire is the main warehouse hazard, so risk assessment is essential to decide 
the best combination of fire prevention and protection to be provided. As 
prevention is better than protection, the 'Inherent SHE' principle suggests that 
the fire load and potential business loss should be minimized by suitable 
compartmentation or stock separation. However, this principle may conflict 
with productivity improvements such as high-rise automated warehousing. If 
fire prevention is not possible, passive or active fire protection must be used. 
The quantitative risk assessment of fire protection systems, however, may prove 
to be difficult as reliability data is often unavailable. The consequences of a fire 
may also be difficult to estimate. Insurers often use the 'worst case' complete 
destruction scenario, but a very small fire can still generate enough smoke to 
contaminate all the stock held. Depending on the type of stock held, firewater 
retention may also be required to comply with environmental regulations. 
In countries where earthquakes occur, the location and construction of 
warehouses require specialized risk assessment and design. Similarly the risks 
of flooding in some locations require risk assessment. 
Archives 
The value of pharmaceutical archives in business terms is generally very 
high — a fact which is often overlooked when designing new facilities. The 
archived documents, samples of product, new chemical entities, tissues and 
other materials must be stored securely to meet legislative requirements. A 
hazard study of existing archives and sample stores will often reveal that 
significant risks have been taken inadvertently; for example, it would not be 
unusual to find documents stored in basement areas with no special fire 
precautions or that storage is under fragile pipes or service drains. Archive 
areas may be visited infrequently and rarely audited for fire safety. 
Most pharmaceutical projects will review archive requirements during HSl 
and HS2 study of business risks and Quality Assurance. The PHA guideword
'Other Threats', interpreted by an experienced hazard study team, may also 
prompt a study of archiving. 
7.7 Safety, health and environment legislation 
The pharmaceutical industry must comply with both SHE legislation and the 
pharmaceutical product regulations explained in Chapter 2. This section only 
considers the SHE legislation. 
7.7.1 Overview of SHE legislation worldwide 
All engineers and designers need to have an understanding of the law and its 
relevance to risk issues in their sphere of operations. In most pharmaceutical 
companies, it is recognized that the legal SHE requirements provide a minimum 
standard for risk management and assessment. Most organizations operate to 
more stringent standards in the interest of product security and business risk 
management. Since SHE legislation is being updated and augmented continuously, 
it is essential to keep abreast of changes in the law by using commercially 
available legal databases, preferably electronic and accessible through e-mail, 
such as those by OSHA and EPA in the USA. 
7.7.2 Overview of UK SHE legislation 
In the UK, most SHE legislation has been, and still is being, updated and 
amended to comply with the requirements of recent EU Directives. The Health 
and Safety Executive (HSE) have powers and duties under the Health and 
Safety at Work etc. Act 1974 to ensure that UK industry complies with the 
regulations passed under this and subsequent acts and regulations. The HSE 
provides useful guidance booklets that are published for all the safety and 
health regulations in force in the UK. Environmental legislation is implemented 
by the Environment Agency, established by the Environment Act 1995. A list of 
some of the main UK regulations that govern SHE in the pharmaceutical 
industry is given below as an overview, although readers should always check 
with HSE and the Environment Agency for up-to-date legislative requirements: 
Health and Safety at Work Etc. Act 1974; 
o Management of Health and Safety at Work Regulations 1992; 
o Manual Handling Operations Regulations 1992; 
o Provision and Use of Work Equipment Regulations 1992; 
o Workplace (Health, Safety and Welfare) Regulations 1992; 
o Personal Protective Equipment at Work (PPE) Regulations 1992; 
o Health and Safety Display Screen Equipment Regulations 1992;
o Control of Substances Hazardous to Health Regulations 1994 (COSHH); 
o Genetic Manipulation Regulations 1989; 
o Genetically Modified Organisms (Contained Use) Regulations 1992; 
o Control of Asbestos at Work Regulations 1987; 
o Supply of Machinery (Safety) Regulations 1992; 
o The Ionizing Radiation Regulations 1985; 
o Noise at Work Regulations 1989; 
o Pressure Systems and Transportable Gas Containers Regulations 1989; 
o Electricity at Work Regulations 1989; 
o Chemicals (Hazard Information and Packaging for Supply) Regulations 
1996 (CHIPS); 
o Carriage of Dangerous Goods by Road and Rail (Classification, Packaging 
and Labelling) Regulations 1994; 
o Carriage of Dangerous Goods by Road Regulations 1984; 
o Control of Industrial Major Accident Hazard Regulations 1984, 1988, 
1990 (CIMAH); 
o Control of Major Accident Hazards (COMAH) 1998; 
o Construction (Design and Management) Regulations 1994 (CDM); 
o The Construction (Health, Safety and Welfare) Regulations 1996; 
o Health and Safety (Safety Signs) Regulations 1996; 
o Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 
1995 (RIDDOR); 
o The Health and Safety (Consultation with Employees) Regulations 1996; 
Fire Precautions Act 1971; 
o Fire Safety and Safety of Places of Sport Act 1987; 
o Fire Precautions (Workplace) Regulations 1997; 
Building Act 1984; 
o Buildings Regulations 1991; 
Environmental Protection Act 1990; 
Factories Act 1961; 
o Highly Flammable Liquids and Liquefied Petroleum Gas Regulations 
1972. 
7.7.3 Litigation 
The foregoing legislation in the UK comes under Criminal Law. However, 
individuals can seek redress through the Civil Law by the process of litigation. 
Lawyers, particularly in the USA, have been actively increasing their business 
in this area. Several successful lawsuits against large organizations have led to 
extremely large financial awards and it is now very common for individuals to 
sue for redress.
Engineers, process designers, managers, and risk assessors may often be 
exposed to litigation, or have to act as expert witnesses on behalf of their 
organizations. It is essential in these cases to have the best legal representation 
and advice available. The process of the law is complex and upheld by the 
lawyers. Technical or moral quality is of no use without a thorough knowledge 
and understanding of the law. 
B i b l i o g r a p h y 
Gillett, J.E., 1997, Hazard Study and Risk Assessment in the Pharmaceutical Industry, 
ISBN 1-57491-029-9, Interpharm Press Inc. 
Barton J. and Rogers R., 1993, Chemical Reaction Hazards, ISBN 0-85295284-8, 
Institution of Chemical Engineers. 
Pitblado R. and Turney R., 1996, Risk Assessment in the Process Industries, 2nd Edition, 
ISBN 0-85295-323-2, Institution of Chemical Engineers. 
Kletz T.A., Chung P., Broomfield E. and Shen-Orr C, 1995, Computer Control and 
Human Error, ISBN 0-85295-362-3, Institution of Chemical Engineers. 
Waring A., 1996, Practical Systems Thinking, ISBN 0-412-71750-6, International 
Thomson Business Press. 
HS(G)51, 1990, The Storage of Flammable Liquids in Containers, ISBN 0-11-885533- 
6, HMSO. 
HS(G)50, 1990, The Storage of Flammable Liquids in Fixed Tanks (Up to 10,000m3 
total capacity), ISBN 0-11-88-55-32-8, HMSO. 
Dept. of the Environment, 1995, A Guide to Risk Assessment and Risk Management for 
Environmental Protection, ISBN 0-11-753091-3, HMSO.
8.1 Introduction 
The design engineer may ask why this book covers the design of utilities and 
services and their maintenance, as these are common throughout industry. 
However, these systems have become important parts of asset management and 
should no longer be an afterthought following the completion of the 'pharmaceutical' 
part of the design. Consideration throughout the design makes the 
validation stage so much easier. 
The impact on the design of engineering workshops for maintenance and 
servicing of production and the utilities is outlined in this chapter and aspects 
particularly relevant to the pharmaceutical industry are emphasized. 
Regulatory inspectors spend a lot of time looking at the design of water 
supplies, air conditioning systems, their operation and cleaning, and how they 
impact on pharmaceutical processes. They also want to know how the business 
is run and organized and who is responsible. 
The ideal pharmaceutical facility (using the USA convention to mean 
the entire building, services plant, services distribution and production equipment) 
is: 
simple; 
has accessible plant and services; 
reliable; 
does not breakdown, go out of adjustment or wear out; 
fully documented. 
The engineer wants the information on the plant in a form that his people can 
understand, to enable them to maintain it easily and find a quick solution to a 
problem. The production department wants a flexible plant available at all times, 
while the quality assurance department wants a plant which performs to design, 
with written procedures that are always followed and documented and where all 
changes are recorded and validated. The company wants all this at minimal cost. 
8 
D e s i g n o f u t i l i t i e s 
a n d s e r v i c e s 
JACKIE MORAN, NICK JARDINE and CHRIS DAVIES
Engineering has moved from being a service to becoming an essential part 
of overall profitability and is now spoken of in terms of asset management. 
Asset management is the consideration of the activity as the ownership of a 
major company resource, i.e. the plant and equipment rather than as the 'fixer' 
connotation normally appended to maintenance. 
Maintenance is now using fault analysis, more sophisticated monitoring of 
the equipment and methods to assess performance to concentrate effort where it 
is needed. Less maintenance, correctly performed, can be shown to give 
increased up time. 
There are two aspects to achieving trouble free operations: 
• management and organization; 
• engineering design and specification. 
Management requires a clear understanding of the objectives of the 
engineering function to enable organization and planning and to ensure 
people are available when required. Clear objectives enable the choices to be 
evaluated and selected. 
Organization requires a clear statement of responsibilities and functional 
relationships of staff and contractors, selection and training of people, setting 
up external contracts, followed by a system to measure the performance of the 
engineering department and make improvements. 
Planning ensures the information is fed into the design at the right time, the 
facility is designed, built and tested to the design, the people and systems are in 
place when the plant is in use and the facility is maintained. 
The design of the engineering space and content of workshops and offices is 
a subsidiary design exercise based on all the above. 
Engineering design requires use of all available engineering knowledge, 
analytical skills and design experience, by a systematic questioning of the 
design for operability and maintenance throughout the design and construction. 
8.2 Objectives 
The engineering function in a pharmaceutical facility is a cost centre (it has 
a direct impact on the costs and profitability of the company) and, therefore, 
must be justified. Engineering costs, as with all costs in the industry, are 
constantly being reviewed. 'Downsizing', 'internal customers', 'process 
re-engineering', 'delighting the customer' and 'core activities' are terms in 
common use in the pharmaceutical industry. The emphasis is on trouble free
operation for the customers and they expect no breakdowns, the lowest cost 
and to be able to plan production without concern over availability of the 
equipment. 
No longer is a new facility designed with a clean slate to set up the 
maintenance department. A greenfleld site does not automatically have an 
engineering complex with all the essential functions of machine shop, welding 
and fabrication, instruments, design office and a full set of satellite workshops 
throughout the production areas. The objectives and measures for the engineering 
function, therefore, need to be determined with the customers. These will 
depend on company policy, the location and the type of operation in the facility. 
For example, the following may have different objectives: 
• 'Over the Counter' (OTC) facility; 
• 24-hour freeze drying operation; 
• handling cytotoxic products; 
• a local packaging operation. 
8.3 Current good manufacturing practice 
The attention of the inspecting bodies is moving from the process and 
production operations to the research and development activities before 
production and to the services plant and maintenance during production. 
Increased importance is being placed on validation of the plant and equipment 
and maintenance of the validated state. For a new product, the pre-approval 
inspection will require a fully validated plant. Subsequent inspections will 
examine production records and follow these through the maintenance routines. 
FDA guidelines cover these principles (see Chapters 2, 3 and 4). They 
require: 
• appropriate design of facilities; 
• equipment history and records or database; 
• written procedures and evidence that procedures are followed; 
• a maintenance programme. 
This enables engineers to set up systems to ensure control of their activities. 
It places a requirement to know the plant and equipment and to be able to show 
that it is receiving the correct maintenance. It requires method statements of the 
maintenance and the description of the tasks. 
To do this requires planning, systems and records. There is only one good 
time to start this — at the inception of the project.
8.4 Design 
During design, decisions are made which affect the maintenance and operation 
of plant and equipment. Maintenance considerations are as important as the 
process, the production capacity of the facility or the tests to be performed by a 
quality assurance laboratory. Access and routes for maintenance are as 
important as those for production and quality assurance staff and for supplying 
materials to the facility. 
Maintenance requirements must be considered during the design stage, as 
the cost at this stage is minimal compared to the costs after completion and the 
consequential costs of poor performance. 
The maintenance strategy should be part of the initial design study and will 
determine action during design and installation. The maintenance staff should 
be part of the project team. The engineer responsible for maintenance should be 
appointed and be responsible for design decisions and acceptance of the plant 
and facility. The validation master plan will have been formulated and an 
essential part of validation is the clear trail from design intent to finished 
facility 
Co-ordination of the services and the structure are critical. The question to 
ask of every service line and connection is 'Why do I need access and how do I 
check it?' 
8.4.1 Building 
The materials of construction and general size and shape of the building are 
important. Heights of floors and size and location of plant rooms are part of the 
design process. The service loads should be calculated in the front-end design 
to size the main elements of the plant and an allowance made for the inevitable 
increase in these during design development. This determines the area for 
services and the location of main plant areas. 
Separate engineering floors can be justified. Floor to floor heights should be 
generous. The increase is in structural cost of floors and envelope. The floors 
are needed to support the plant, so are not extra and the increase in envelope 
cost is minimal. The cost of services plant and its controls can represent up to 
60% of the total project and the civil structure up to 10%. 
All ducting requiring inspection should be on the plant floor and not hidden 
in false ceilings. This leads to structural slab ceilings in parenteral areas. Where 
services are run above a false ceiling (such as an office suite) there are beams 
supporting the floor and, if it is a reinforced concrete structure, there are caps on 
the columns, which will reduce the space locally.
The structure must allow for access for services. The increasing electrical 
power and controls require co-ordination and affect the structural design. A 
reinforced concrete structure can become complex when many conduits pass 
through an area. 
Thought must be given to future service requirements, for example, in an 
analytical laboratory additional services may be required or the bench layout 
may be changed to suit new methods. The floor must be designed to permit 
these changes without affecting the strength and a grid of soft spots may be 
required. 
The trend is to locate the drives and services of production plant in service 
areas. These should be designed with good access and enough space for 
maintenance. 
Inlet and exhaust should be located to suit the prevailing wind and may 
require a special study on a multi-building site. 
Details such as the design of windows or atria for cleaning are important. A 
glass stair tower may look good but will be costly to clean. It may need 
specialist contract equipment, which will require steel reinforced concrete pads. 
Building expansion joints should not run through critical areas and should be 
kept away from heavy traffic routes. Parenteral production areas should be on a 
good slab to minimize floor cracks. 
Wet services should not run over critical areas. If this is unavoidable there 
should be no joints and all items requiring maintenance should be located away 
from the area. Inspection points and clean-outs for drains should be located in 
service or plant areas. Particular attention should be paid to the design and 
construction of service penetrations to process areas. 
8.4.2 Maintenance access routes 
Movement of engineering personnel should be part of the overall people and 
materials movement study. Separate engineering floors allow separate access 
routes for staff, which reduces contamination of the production space; fire 
escape routes or separate external entrances can provide access, for example. A 
WFI plant may require a specific changing area and decontamination for parts 
to be fitted to the plant. 
8.4.3 Plant access 
The structure and the openings to the plant areas must be designed to allow 
removal of the largest maintainable item without affecting the integrity of the 
production facility. 
Adequate access for maintenance of the plant and services should be 
provided. Any valves requiring maintenance must be accessible even if this
means locating them away from the heater batteries. Test points should be 
accessible. With conventional design tools basic decisions such as location of 
pumps, motors, valves, traps, filters, etc. can be made. 
The mechanical, electrical and control services in a modern pharmaceutical 
plant area need co-ordination to ensure that there are no clashes and that a 
normal-sized person can reach all areas of the plant requiring access. Drawings 
to 1:20 scale in plan and elevation of plant areas are required to check this. 
Alternatively, modern design software using 3D could be used. 
8.4.4 Storage of consumables 
As part of the strategy, a decision is required on the storage of spare filters and 
other consumables that are used infrequently. If they are held on-site then dry, 
safe storage is needed. 
Solvents 
In this context solvents are considered as organic liquids that provide a vehicle 
for bringing reactants together, moderating reaction conditions, preferentially 
extracting one component from another, or cleaning equipment, but are not 
themselves reactants. 
Most solvents are flammable, often with low flash points, usually of low 
reactivity and generally non-corrosive. 
Bulk, drum or IBC storage and distribution 
In any design, one of the early decisions must be the choice between bulk, drum 
or IBC storage and distribution. In the absence of other factors, bulk storage is 
the preferred option since it usually provides advantages in terms of economics, 
minimized labour involvement and more effective integration in automatically 
controlled processes. Despite the greater inventory, bulk storage also has the 
better safety record. 
In practice, this early decision will be made primarily on the basis of the 
individual batch quantities combined with estimated campaign or annual 
consumptions. The choice may be influenced to a lesser extent by the existence 
or otherwise of a tank farm, site space considerations, capital versus operating 
costs, or occasionally the package availability of the solvent involved. 
Bulk storage is the method of choice for much primary production but for 
pilot plants with reactors of say 0.2 to 1 m3 capacity, drummed solvents often 
provide an appropriate solution where one-off batches or very short runs are 
common.
In secondary production, solvents are frequently needed only for equipment 
cleaning and in such relatively small quantities that supplies in drums or even 
smaller containers often suffice. 
Bulk storage siting 
One of the initial decisions relates to the location of storage tanks; whether in a 
dedicated tank farm serving a number of buildings or by placing alongside the 
production units they supply. 
In laying out a greenfield site, space could be set aside for a tank farm 
specifically able to meet the initial site needs but capable of expansion to cater for 
bulk solvent demands as the site develops. However, the benefits of centralized 
control, minimized space and facilities for tanker unloading and sampling and 
reduced vehicle movements on-site must be weighed against higher first costs for 
set-up and piping distribution to production buildings. Once established the 
marginal costs of adding further solvents or destinations are likely to be small in 
comparison with the alternative approach of siting storage tanks adjacent and 
dedicated to individual production units as and when the need arises. 
The majority of new designs will, however, be applied to existing plants 
where choices between centralized versus local storage are not applicable and 
location will be dictated by the site philosophy and space availability. 
When decisions are taken to locate tanks adjacent to the buildings they 
serve, conscious recognition must be given to the additional restrictions 
imposed, particularly in ensuring the safety of the facilities in the event of 
fire. Such limitations can affect the total quantities stored against the proximity 
of building walls, the nature and fire resistance of their structure, location of 
doorways, windows and fire escape routes. 
Wherever tanks are located a prime requirement is good road tanker access, 
not only to allow safe docking for unloading but also to facilitate rapid vehicle 
exit if required by a serious incident. Siting should minimize or avoid 
obstructing site roads during unloading which, with quality control checks 
can occupy several hours per visit. 
For the most part storage tanks should be located above ground. Although 
below ground installations provide some advantage in terms of fire protection, 
environmental concerns and the costs of providing satisfactory protection and 
leakage detection often prove prohibitive. 
General 
Having located the storage area and associated tanker bay, facilities are needed 
to allow safe sampling of the cargo before discharge, often in the form of a 
height adjustable overhead gantry. Occasionally, weather protection is provided
by canopies, usually without sidewalls, which inhibit ventilation and dispersion 
of flammable vapours. 
Over recent years pumping has become the preferred method of emptying road 
vehicles, in contrast to the increasingly rare use of compressed air discharge with 
its attendant drawbacks of formation of flammable atmospheres, potential for 
solvent contamination and unnecessary emission of vapours. Though some road 
tankers are equipped with pumps, one forming part of the storage facility itself 
will give greater assurance of cleanliness especially if dedicated to one material. 
Provision of static earthing, safety showers, self-sealing hose couplings and 
vapour balance connections (between tank and tanker head spaces) are safety or 
environmental features of an almost mandatory nature. 
In sizing storage vessels the main factors will be the annual consumption 
together with individual batch and campaign requirements. Consideration 
should, however, also be given to ensuring that tanks are sized to contain a full 
tanker load plus a margin to allow for order lead times as well as unexpected late 
deliveries (caused by inclement weather, for example). Typically 10% ullage is 
applied once the actual storage volume has been determined. 
The normal practice is for tanks to be installed within bunds, mainly to 
protect the environment against leakage. More than one tank can be located in a 
single bund provided its capacity is adequate to accommodate the capacity of 
the largest one plus 10%. Good bund design should allow adequate access 
between bund and tank walls for maintenance and to ensure ease of escape in an 
emergency. For similar reasons, wall heights need to be limited. Low walling 
has the additional benefit of promoting vapour dispersion. Since bunds collect 
rainwater, arrangements are needed for its periodical removal. 
Where several tanks are located together or single tanks are located close to 
buildings, drench systems can provide cooling in the event of fire in adjacent 
vessels or buildings. The need or otherwise of such protective devices is 
determined in conjunction with insurers, the Health and Safety Executive or 
similar authorities. 
Most storage tanks for highly flammable solvents (flash point below 32°C) 
are blanketed with an inert gas as a safety precaution. For some materials, 
excluding oxygen and moisture helps to maintain solvent quality and for this 
reason it is also applied to less flammable situations. Nitrogen is the most 
common inerting gas, but carbon dioxide is an occasional alternative with 
typical blanketing pressures of 10 to 20mBar. 
Distribution 
From the storage tanks, solvent will be distributed via a system comprising 
pumps, distribution pipework and usually metering devices and filters.
An appropriate control will be built into the scheme. Distribution pumps are 
generally located outside the tank bunds on plinths arranged to drain away any 
leaked fluid. The pumps (duty plus standby for critical situations) and 
distribution main are sized to support the number of vessels that need to be 
filled simultaneously. Branches to individual users will be based on the desired 
filling time for that vessel. With non-water miscible and hydrocarbon solvents, 
particular emphasis should be placed on reducing fluid velocities to minimize 
static build up. This applies especially where the presence of moisture may 
create two phases. 
Pump differential heads are determined using standard calculations accommodating 
pipeline, filter and instrument losses and static heads including 
pressure in receiving equipment. Calculations should cover the full operating 
envelope of the system. Centrifugal pumps provide low cost, reliable service 
with packed glands or single mechanical seals suitable for the majority of 
installations. Magnetic drive pumps eliminate the leak potential of rotating 
seals. 
To minimize leakage and avoid establishing unnecessary zoned areas, 
welded lines are preferred with the minimum of joints for maintenance 
purposes. Solvents do not usually need to be distributed through ring 
mains — continuous circulation wastes energy and can cause unwanted heat 
and static. Long pipe runs especially where subject to temperature variations 
(for example exposure to direct sunlight) require protection against hydrostatic 
overpressures, most commonly in the form of a small relief valve returning to 
the source vessel. 
Protection of pumps with inlet strainers is good practice as is end-of-line 
filtration, largely to remove rust scale and similar particulates. Where higher 
standards are demanded, micron filters can be fitted usually alongside a 
downstream piping specification change to stainless steel to avoid potential 
recontamination from lower grade materials. 
Batching meters are a common and economical means of metering solvents 
into receiving vessels with satisfactory levels of volumetric accuracy for most 
purposes. Versions are available with output signals suitable for integration into 
computer and other control arrangements. Load cells, level gauges and 
transmitters and other devices on either source or receiving vessels provide 
alternative means of metering with varying degrees of applicability, accuracy 
and cost. 
Most bulk distribution systems are connected to many destination vessels so 
that the final shut-off valve will be exposed to its internal conditions. This final 
valve must, therefore, provide positive shut-off to ensure no back contamination; 
where circumstances are more critical, the final solvent valve can be
mounted on a manifold with other services, with double protection being 
provided by another valve between the manifold and vessel. 
To eliminate risks from static, it is vital that all metal components of 
flammable solvent systems are checked for earth continuity both before 
bringing into use and after modification. Arrangements to allow entering 
solvents to run down vessel walls helps to eliminate 'free fall' static generation. 
Materials of construction 
Most solvents are produced in plant fabricated from carbon steel. 
Hence, this material is adequate for many pharmaceutical grade solvent 
storage and distribution systems. For certain solvents or where absence 
of colour is important, stainless steel is an alternative constructional 
material. 
Carbon steel is similarly suitable for the majority of distribution 
pipework although it may well be upgraded to stainless steel downstream 
of final filtration. Such upgrade avoids pick-up of particulate downstream 
of the filtration, minimizes internal corrosion where the tail end of the 
solvent pipe may be exposed to reactor contents and provides cleaner, 
maintenance-free piping within the process areas. The latter point is 
particularly important for ensuring GMP in secondary manufacturing 
plant. 
Plastic and glass-fibre are rarely, if at all, employed for flammable solvent 
handling. Difficulties of static elimination, potentially inadequate chemical 
resistance and above all their lack of fire resistance make them unsuitable. 
Solvent suppliers are always willing to offer advice on suitable materials of 
construction and their advice should be sought if there is any doubt over the 
suitability of one material over another. 
Recovery 
Most plants have some form of solvent recovery plant to reduce the costs of 
purchasing new solvent and disposing of contaminated solvent waste. Steam 
stripping is usually used for this application, so non-polar solvents with low 
boiling points are preferred. 
Recovered solvent is usually stored separately from new solvent, with 
the facility to top up with new solvent as required. It is common to 
use the recovered solvent in the initial stages of production with the 
new solvent being used for the final filter washes. New solvent is always 
used for cleaning.
8.5 Utility and service system design 
There is a temptation to specify spare capacity and duplicates of plant for run 
and standby. Care should be taken with this approach, as over-sizing fans and 
pumps can lead to control problems. 
Run and standby may require more control. For example, do you alternate 
between the two or have 'run' installed and 'standby' unbelted or as a noninstalled 
spare? On WFI a simple system with no dead leg is required. 
Duplicate pumps require more valves and give dead areas unless complex 
controls are provided. Standardized flange spacing and a non-installed spare 
can replace duplicate steam reduction sets. 
A risk analysis or Failure Mode Effect Analysis (FMEA) may need to be 
carried out to decide the strategy. 
Shut off valves should only be used sparingly. Breaking a complex service 
into many sub-sections with shut off, in the hope of being able to carry out 
selective shutdown, is expensive and you will have to prove that the flows in the 
part plant are still within design limits. It may be better to shut down the whole 
system for repair work. 
Multiple-use HVAC plants should be avoided. They are difficult to keep in 
balance and prevent cross-contamination. 
Table 8.1 shows the type of system categories that may be required and the 
areas of utilization. 
Table 8.1 Utility system categories 
Utility category 
Compressed 
Gas and Vacuum 
Water 
Steam 
Type of system 
Service comp air 
Process/instrument 
comp air 
Breathing air 
Special gases 
Vacuum-cleaning 
Vacuum-service 
Vacuum-process 
Domestic H&C 
Purified 
WFI 
LTHW 
Condensate 
Chilled water 
Cooling water 
Service 
Clean 
Possible area of utilization 
Plantroom Packing
Some examples are discussed in this section. The intention is not to provide 
prescriptive solutions, but to indicate factors that will influence the design and to 
suggest sources of information that may be useful to the designer. It is important 
to achieve a clear understanding of the requirements of the system under 
consideration, in terms of quantity and quality, at the outset of the design 
process, as this will allow a proper assessment to be made of the best methods 
available for meeting the requirements of the system. Of particular importance 
when specifying the quality will be the likelihood of contact with the product, i.e.: 
• part of final product, e.g. water; 
• direct contact, e.g. solvents; 
• indirect contact, e.g. Clean In Place; 
• no contact, e.g. thermal fluids. 
For fluids with no contact with the final product there are many similarities 
with standard chemical manufacturing facilities, but these areas will also be 
discussed for the sake of completeness. This chapter will also discuss the 
effects of forthcoming regulatory requirements, allowing for any future 
expansion and systems to prevent cross-contamination of utilities with process 
uses. Table 8.1 gives a checklist for determining possible requirements for 
utility systems in various types of pharmaceutical facilities. 
Possible area of utilization 
Laboratory Creams 
liquids 
Tablets oral Aerosols Sterile bio
8.5.1 HVAC 
The types of HVAC systems commonly found in secondary pharmaceutical 
facilities are extremely diverse and are selected mainly on the basis of the 
required environmental conditions and the specified level of product contain- 
Table 8.2 HVAC system designs 
Description of 
HVAC system 
objectives 
1. Natural ventilation only 
2. Mechanical ventilation 
3. Mechanical ventilation with 
heating and/or cooling 
4. Air conditioning i.e. heating and 
cooling and humidity control to 
meet a specified band of 
temperature and humidity 
5. Full air conditioning i.e.: heating 
and cooling and humidity control 
to meet a specified condition of 
temperature and humidity 
6. As 4 or 5 below but including 
a low humidity set point 
(i.e. below approx. 50% RH) 
7. As 4, 5 or 6 below with specified 
clean room conditions 
8. As 7 below but with Class 100 
laminar flow distribution 
9. As above but recirculation in 
lieu of Total Loss 
10. Separate systems for each work 
centre and total loss systems to 
minimize risk of cross-contamination. 
Terminal HEPA filters on supply 
and extract. Sterile (positive) or 
containment (negative) pressure 
cascades. Low humidity. Dust 
extract. Specified classification 
of clean room 
Possible 
applicable 
areas 
Plant rooms, warehouse 
Plant rooms, warehouse, changing 
Warehouses, receipt and despatch, 
changing, bin floor, dry products, 
creams/ointments, packing hall, 
corridors, offices. 
Warehouses, receipt and despatch, 
changing, bin floor, computer rooms, 
dry products, creams/ointments, 
packing hall, corridors, offices 
Offices, stability rooms special stores, 
computer rooms, dry products, 
creams/ointments, packing hall, 
offices. 
Dry filling, capsule manufacturing, 
aerosols, dry products. 
Creams, dry products, aerosols, steriles. 
Steriles, dry products (for dust control) 
As above 
Sterile, dry products, aerosols, cytotoxics, 
vaccines, clinical trials, bio pharms.
ment. As the degree of control associated with these factors increases, the 
complexity, and therefore, cost of the HVAC system increases proportionately. 
Table 8.2 details the main types of HVAC systems commonly used in 
secondary pharmaceutical facilities. 
Associated 
plant 
As 6 below + terminal 
HEPA Filters + Dust 
extract. Note: Total loss 
demands the highest 
possible plant loads 
As above but reduced 
plant loads 
As 7 below 
As 4, 5 or 6 and HEPA 
filtration 
As 5 below and 
dehumidification 
Input/extract fans, heating 
and cooling, filters and 
humidification. Note: 
Requires greater plant 
capacity than 4 below 
Input/extract fans, heating 
and cooling + filters and 
humidification 
Input/extract fans + 
heating and/or 
cooling + filters 
Input/extract fans + filters 
High and low level louvres 
Temp and 
humidity 
control 
As 6 below 
As 6 below 
As 4, 5 or 6 below 
As 4, 5 or 6 below 
Any manufacturing 
conditions with low RH 
i.e.: 19°C30%RHor 
18°-22°C30% RH Max. 
Specified manufacturing 
conditions (not lower than 
50% RH) i.e.: 21°C 50% 
RH summer and winter 
Comfort conditions usually 
a specified band for summer 
and winter i.e.: 20°C-24°C 
30%-60% RH 
Max or min temperature 
control only i.e.: Max 25°C 
50C-IO0C above external 
temperature in summer 
10°C-20°C above external 
temperature in summer 
Filter 
standard 
HEPA 
HEPA 
HEPA 
HEPA 
EU3 to 
EU9 
EU3to 
EU9 
EU3to 
EU9 
EU3to 
EU9 
EU3 
NIL 
Clean 
room 
class 
100-100,000 
100-100,000 
100-100,000 
100-100,000 
NIL 
NIL 
NIL 
NIL 
NIL 
NIL
These systems are shown in schematic form in Figures 8.1 to 8.10. 
8.5.2 Air 
Compressed air is used in pharmaceutical applications for driving pumps and 
back flushing bag filters. Atmospheric air is passed through a 50 urn or smaller 
aperture filter to remove insects, dust and pollen before it enters the compressor. 
Care should be taken to ensure that the air intake is not immediately adjacent to 
sources of solvent vapour or combustion fumes. 
The air is compressed to an appropriate pressure for the system, taking into 
account the maximum required design pressure and distribution system 
pressure drop. 
The air is then filtered again using a 0.1-0.5 urn filter and dried to remove 
any compressor oil and condensed water. The pipework is usually carbon steel 
or galvanized carbon steel. 
A general specification for air for these duties is: 
• particulate filtration to 0.1 micron; 
• pressure dew point at 7Barg H- 30C; 
• oil filtration to 0.01 ppm; 
• normal operating pressure 7Barg. 
Instrument air is used for actuating valves. Compressed air is filtered to remove 
dirt and oil mist, which can clog the actuator. The pipework is usually carbon 
steel or galvanized carbon steel. The specification of the air varies according to 
Figure 8.2 Mechanical ventilation 
Extract 
fan 
Panel filter 
Input fan . 
Figure 8.1 Natural ventilation 
Air outlet at 
high level 
Air inlet at 
low level
Figure 8.4 Air conditioning 
Humidifier Panel filter 
Min. 
fresh 
air 
Supply fan 
Bag filter 
Heating coil 
Cooling coil 
user requirements and guidance should be sought from valve suppliers. A 
general specification for instrument air is: 
particulate filtration to 0.01 micron; 
pressure dew point at 7Barg — 400C; 
oil filtration to 0.003 ppm; 
normal operating pressure 7Barg. 
Figure 8.3 Heating and ventilation 
Heating and/or 
cooling coils 
Supply fan 
Panel filter and 
bag filter 
Min. fresh 
air
Figure 8.5 Air conditioning with zone reheat 
Breathing air is used to protect personnel from dust and toxic fumes by 
supplying air to hoods or full suits. British Standard BS4275 covers the design 
of distribution systems for breathing air. 
The breathing air system is usually supplied from the compressed air 
system. The air is then filtered, purified and dried before distribution to the 
end users. The use of compressed air for breathing means that the location of 
the compressor air inlet is especially important to prevent toxic fumes from 
entering the breathing air system. 
Min. 
fresh 
air 
Zonal reheat 
for close 
temperature 
control 
IO 
other 
areas 
Min. 
fresh 
air 
Humidifier 
To 
other 
areas 
Zone 
reheat 
Figure 8.6 Low humidity air conditioning
Figure 8.8 Laminar flow clean room 
HEPA 
filter 
wall Exhaust 
plenum 
Panel . 
filter 
Bag 
filter 
Plant 
mounted 
HEPA 
Min. 
fresh 
air 
Cooling coil 
in bypass 
Figure 8.7 Low humidity clean room air conditioning 
Low level 
extracts 
Terminal 
supply 
HEPA 
filters 
Zonal reheat 
for close 
temperature 
control 
To 
other 
areas 
Mm. 
fresh 
air 
Dehumidifier
Plant 
mounted 
HEPA 
Terminal 
HEPAs 
Dehumidifier 
Min, fresh 
air to 
replace 
dust 
extract 
Safe change 
HEPA filter 
Dust extract 
from equipment 
Low level exhaust 
HEPA filters 
Dust extract 
fan 
Safe change 
filter 
Figure 8.9 Low humidity containment clean room 
Self-cleaning 
dust filter 
Total 
exhaust 
Full 
fresh 
air 
Extract 
fan Pre-heating 
Pre-cooling 
Dehumidifier Re-heating 
Re-cooling 
Supply 
fan 
Plant 
mounted 
HEPA 
Terminal 
HEPAs Safe change 
HEPA filters 
Dust extract . 
fan 
Safe change 
filter 
Self-cleaning 
dust filter 
Figure 8.10 Low humidity total loss containment clean room
Air purification units may contain the following equipment: 
0.01 micron pre-filter to remove solids; 
activated carbon adsorption bed to remove hydrocarbons; 
desiccant drier to remove water; 
catalytic element to remove carbon monoxide; 
final filter; 
carbon monoxide monitor alarm; 
flow meter; 
low pressure alarm. 
BS4275 states that provision must be made to warn operators if the system 
fails. An emergency supply facility is usually provided in the form of a storage 
tank or cylinder. 
There should be a minimum number of manual isolation valves in the 
distribution system due to the possibility of these valves being mistakenly 
closed whilst the system is in use. The materials of construction for pipework 
can be galvanized carbon steel or degreased copper. The distribution system 
ends in self-locking fittings that feed directly into the PE air hoods or suits. 
Process air is used for feeding to fermenters or for processing equipment for 
parenterals. Process air is sterile, i.e. filtered to 0.2 micron. For fermenters, the 
air may have other gases added such as carbon dioxide; the gas used being 
dependant upon the cell culture being grown. Materials of construction are 
usually stainless steel and the pipework and fittings must be suitable for 
occasional steam sterilization. As a guideline, the general specification for 
instrument air (see page 275) is also applicable as it is the basic source of air 
for this purpose. 
8.5.3 Vacuum 
General vacuum systems are normally connected to a number of process 
vessels through a common pipeline and are used for evacuating process 
equipment prior to nitrogen blanketing, filling head tanks from drums and 
transferring from one vessel to another. The actual vacuum achieved is not 
critical, but is of the order of 200mBarg. 
For filtration, a vacuum pump is normally connected to a single filter via a 
receiver. The vacuum is connected to the liquid outlet of the filter and used for 
transferring filtrate from the filter to the receiver. The vacuum is applied to the 
receiver and the receiver is usually fitted with a vent condenser to prevent the 
vapours from reaching the vacuum pump. The pipework is commonly stainless 
steel as a minimum, as the filtrate is often reused either directly or after 
distillation.
For drying, the vacuum may be used to dry the solid on the filter by applying 
to the top of the filter or dryer. There will be a vent filter on the dryer to prevent 
the solids from entering the vacuum system. The solvent vapours will be 
condensed using a condenser supplied with refrigerant and collected in a 
receiver. The vacuum used for drying will depend upon the maximum 
temperature which can be applied to the product balanced against the likelihood 
of pulling solids into the vent filter causing a blockage. 
The use of vacuum in distillation systems on pharmaceutical facilities is 
common, in order to depress the boiling point of distillation mixtures where 
some component of the mixture is sensitive to heat. Since depression of boiling 
point is inversely proportional to the system pressure, this duty gives the 
greatest demand for high vacuum with requirements for system pressures of 1- 
2mbarg being commonplace. 
There are two main types of vacuum pump: 
• liquid seal; 
• dry running. 
Liquid seal pumps use fluid to provide a liquid seal between the pump casing 
and the central impellor. As the maximum achievable absolute vacuum is the 
vapour pressure of the seal fluid at the operating temperature, the choice of 
sealing fluid is important. 
The seal fluid can be run on a single pass or on recirculation. A single pass 
type is the most appropriate choice for vapour streams containing solids, 
condensed solvent vapours or corrosive gases. This is due to the flushing action 
of the sealing fluid preventing the build-up of contaminants to corrosive 
concentrations leading to pump damage. The downside to this, however, is 
the increased amount of effluent produced, which is costly in terms of sealing 
fluid. Recirculating seal fluid systems require additional equipment such as a 
cooler (to remove heat from the condensing process vapours and the power of 
the pump motor) and a pot that can be topped up with fresh sealing fluid and 
which has an overflow to drain. The recirculating system produces less effluent 
but if not correctly maintained or cleaned can become blocked with solids or 
the seal fluid can be completely displaced by solvent. A further downside is that 
if the cooler is not effective, the exhaust gases may also contain a greater 
amount of solvent and the pump may produce a poor vacuum due to the 
increase in vapour pressure of the seal fluid at the higher operating temperature. 
Dry running pumps are similar in operation to liquid ring pumps but use oil 
for the lubricating fluid. The tolerances within the pump are much smaller and, 
therefore, much less oil is required. The choice of lubricating oil is important as 
this can react with the process vapours and choke the pump.
Dry running pumps are also intolerant to some corrosive gases but, unlike 
single pass liquid ring pumps, they do not have the protection of the flushing 
action. These pumps are capable of very high vacuums and in clean process 
conditions, are superior to liquid ring pumps, with less effluent produced. 
Multistage units can produce very high vacuums required for purification of 
primary product from close isomers by distillation. 
Vacuum pumps are usually fitted with an inlet condenser or small vessel to 
receive any liquid carryover or condensate. All pipework should fall towards 
the catch pot to prevent back flow of condensed vapour to the equipment item. 
If the vacuum pump is used for more than one vessel, care should be taken 
that vapours and condensate cannot reach the other vessels. The pipework 
should be arranged to minimize pressure drops and pipelines should have 
long radius elbows or pulled bends to prevent erosion due to solids carryover. 
There should be the lowest possible number of in-line devices to avoid 
blockages. 
A condenser that uses a refrigerant can be used, but care should be taken if water 
is being removed from the vapour and gas stream. The discharge of the pump is 
fitted with a device to remove entrained liquid prior to discharge to atmosphere. 
Care should be taken to ensure that discharge pipework has a low pressure drop as 
this will control the absolute vacuum the pump is capable of achieving. 
The pipework is suitable for the process but care should be taken, in the case 
of a reduced specification at the receiving vessel, that no dirt or corrosion 
products could back flow to the vessel. 
Care must be taken when cleaning, especially in the case of filter failure on a 
dry vacuum line, that any change in pipework specification occurs after the high 
point, in order to ensure that no corrosion products can back flow in the 
condensed vapour. If the condensed solvent is to be recycled, the use of 
stainless steel pipework throughout is recommended to ensure cleanliness. 
8.5.4 Clean steam 
Clean steam is used in pharmaceutical applications where steam or its 
condensate is in direct contact with the product. The end use of steam demands 
that it is supplied dry, saturated and free of entrained air. The requirement for 
chemical purity is primarily what differentiates clean steam from plant steam. 
The prohibition of corrosion inhibitors and anti-scaling additives influences 
generator design and materials of construction. Clean steam and plant steam 
systems should be completely separate. 
The requirement to use clean or pure steam is governed by the cGMP to 
avoid contamination of the product.
The major use for clean steam is in the sterilization of process and specialist 
water systems. Clean steam is also used in autoclaves and sometimes for the 
humidification of clean rooms. Pure steam is used in processes producing 
parenterals, which demand the use of WFI and here the steam must not be 
contaminated with micro-organisms or endotoxins (pyrogens). The steam must 
be of the same specification as the WFI (to BP or USP standards for WFI) and is 
also used for the sterilization of WFI systems. 
The uses of clean steam in pharmaceutical plant are fundamentally different 
from the uses of pure grades of water, as steam is rarely used as part of the product 
and only traces come into contact with the final product. It could be argued that 
the steam need not be to such a high specification, but it is generally used in the 
final stages of production where precautions against contamination are 
most stringent. 
Clean steam and pure steam are usually produced in a dedicated steam 
generator. The generator is heated using plant steam. The heat exchanger is 
double tubesheet with an air gap between plant and clean sides which prevents 
contamination. 
The generator is fitted with a device to remove entrained liquid droplets that 
may contain bacteria or endotoxins from the vapour stream. This may take the 
form of a demister pad or some sort of baffle arrangement. 
The generator is usually manufactured in stainless steel 316 L or possibly 
titanium due to the corrosive nature of pure water. It is important not to let too 
much non-condensable gas (0.5% by volume) into the steam distribution 
system, as this will form a coating on the vessel surface and prevent efficient 
heat transfer. There is normally an aseptic sampling device before and after the 
generator to allow for sampling for endotoxins. The feed water to the generator 
is purified and free of volatile additives such as amines or hydrazines. As 
generators will only usually reduce the endotoxin concentration by a factor of 
1000 whatever the quality of the feed material, it is important to control 
endotoxins in the inlet water to minimize the chance of spikes of high 
endotoxins in the pure steam system. 
Steam is a sterilizing agent so although the materials of construction are 
required to be 316 or 304 stainless steel for reasons of corrosion resistance, the 
pipelines do not require special internal finishes and can be connected using 
flanges. The main consideration for distribution systems is their ability to 
remove condensate. Condensate poses the risk of micro-organism growth and 
reduces the effectiveness of sterilization. To ensure effective removal of 
condensate there should be steam traps at all low points and at 30 m intervals 
of pipework. The pipelines should incline towards the point of use by 1:100 and 
be properly supported to prevent sagging. Any in-line fittings should be
designed to prevent condensate collection. Any lines not used continuously 
should be fitted with their own steam trap arrangement to prevent the build up 
of condensate above the isolation valve. 
Condensate should not be recovered for use as clean steam. It could be 
returned to the plant steam boiler if not heavily contaminated, although the 
small quantities of condensate involved make this impractical and it is therefore 
usually sent to drain. There should be an air break between the condensate lines 
and the drains to prevent back flow of condensate. The drains should be suitable 
for dealing with hot corrosive water. The steam traps should be 316 stainless 
steel, free draining, with the minimum number of internal crevices i.e. 
thermostatic type. Condensate quality for clean steam systems should 
comply with the USP or BP specification for WFI. 
For fermenter systems growing recombinant or pathogenic organisms, 
where there is a possibility of contamination, the condensate should be fed 
to the kill tanks (see Section 8.9). 
8.5.5 Inert gases 
Nitrogen is used to blanket vessels, for liquid transfers, filtration, cleaning bag 
filters, and for blowing process lines clear. It is also used for inerting explosive 
atmospheres in solids handling equipment and for pressure testing vessels. 
Nitrogen can be produced in pressure swing absorption systems from air, by 
other means from air, or from liquefied nitrogen in storage tanks and cylinders. 
Pressure swing absorption can produce nitrogen at a reduced specification if the 
unit is undersized and, therefore, should not be used for critical applications 
such as inerting of mills. Liquid nitrogen can be produced in many different 
grades and, therefore, it is important to select the correct grade for the 
application. It must be remembered that the grade must be for the highest 
requirement if the system is for site wide nitrogen supply. Some grades of 
nitrogen contain hydrocarbons (dependant upon the manufacturing route) and 
these would be unacceptable for flammable environments. cGMP requirements 
normally specify nitrogen to be filtered to 0.1 micron when in contact with 
the primary product, i.e. once the bulk pharmaceutical chemical has been 
produced. 
The material of construction for pipework is usually carbon steel. The 
highest pressure required and the maximum line pressure drops set the pressure 
of the main. The back flushing of filters is usually the highest pressure and is of 
the order of 6 Bar g. Normal maximum operating pressures for systems of this 
type are of the order of lOBarg. 
Hydro fluoro alkanes (HFAs) are a group of gases that have been developed 
to take the place of the old CFC refrigerant gases. They are used as propellant
for pharmaceutical aerosols and their main property is their degreasing effect, 
which means that diaphragm pumps are usually used for transfer. They are 
expensive and, therefore, leakages in the system should be kept to a minimum. 
In the interest of cleanliness the materials of construction are stainless steel for 
HFA systems. 
8.5.6 Specialist water supplies 
This section offers an overview of the main aspects of water and steam 
production and use in pharmaceutical facilities. This area is covered in far more 
detail in the ISPE 'Baseline Pharmaceutical Engineering Guide Volume 4: Water 
and Steam Guide'. 
There are many types of water to be found in pharmaceutical facilities. A 
few of the main types are as follows: 
towns water is usually straight from the mains and may vary in quality 
throughout the year. The specification can be obtained from the local water 
company and is usually given as a yearly average. There may be two or more 
water sources for a given plant, and the characteristics of water from these 
different sources may vary widely; 
process water is normally towns water that has passed through a site break 
tank; 
de-ionized/demineralized and softened water has passed through some form 
of water softening process to remove calcium and magnesium ions that can 
cause scale on heat exchanger surfaces and in reactors; 
purified water has usually been softened and passed through a UV source to 
remove bacteria. There are various specifications for this as discussed later in 
the section. The most suitable of these depends upon the market for the final 
product but generally the water is soft and contains a reduced number of 
bacteria; 
water for injection/pyrogen-free water has been softened and has a low 
bacterial count and a reduced endotoxin loading. There are a number of 
different specifications for this type of water. The USP and BP specifications 
are the most commonly used for WFI. 
Towns and process water is treated to give all the other types of water by 
using the following processes (amongst others): 
organic scavenger — removes organics (may be naturally occurring); 
duplex water softeners — removes calcium and magnesium salts on a 
continuous basis; 
coarse filtration — removes dirt and debris;
break tank — protects water supply and protects against short-term failure 
of supply. Often a mandatory requirement under water bye-laws; 
reverse osmosis unit — removes solids, salts and bacteria; 
electrical deionization — removes the ions present, effectively softening the 
water; 
UV sterilization — kills a significant number of the remaining live bacteria. 
Potable water is used widely in the pharmaceutical industry as a solvent, a 
reagent and a cleaning medium. 
Purified water is used in the preparation of compendial dosages. While 
Water for Injection is generally used for sterile products, it is also used for 
cleaning equipment used to make such products. 
Specifications for specialist waters are laid down by British Pharmacopoeia 
(BP), European Pharmacopoeia (Ph.Eur.) and United States Pharmacopoeia 
USP. These documents also describe the tests that must be carried out to prove 
the water is to specification. 
Historically these specifications were much the same. Recently however, 
there have been moves to harmonize the BP and Ph.Eur. specifications but the 
USP specification has changed. This change has lead to a drastic reduction in 
the number of tests required and specifies only Total Organic Carbon (TOC) 
and conductivity, both of which can be measured continuously using online 
monitoring equipment. It would also appear that the specification of the water 
has been tightened by change. 
At present there is some confusion about the specification of WFI and 
purified water mainly because of the wide differences in requirements between 
the USP and BP/Ph.Eur. water specifications. The main problem is that WFI 
must be produced by distillation in the BP and Ph.Eur. specifications, but can be 
produced by reverse osmosis in the USP specification. Although it would 
appear that the BP and Ph.Eur. will probably follow the USP at some point in 
the future, it has left manufacturers who market their products in both the 
Europe and America with something of a dilemma. With this in mind, it is 
important to be clear of the desired final product specification when initially 
specifying a new water system. 
After treatment to produce purified water or WFI, the water is collected in a 
receiver, which is either jacketed or has an in-line heater. The vessel is normally 
a cylindrical dished end vessel designed to withstand the vacuum that may 
occur during steam sterilization. The vessel is 316 stainless steel to prevent 
corrosive attack by the hot purified water. 
The tank is fitted with a relief device and possibly some sort of relief 
monitoring device. The vent is fitted with a HEPA filter to prevent the ingress of
microorganisms and is normally heated, to prevent blockage of the hydrophilic 
filter packing with water. The vent is fitted with a drain via a steam trap to allow 
any condensate in the vent line to be drained off. 
The water is pumped from the vessel through a heat exchanger and then to 
the distribution system. The heat exchanger can be a shell and tube or plate 
variety but must be of the double wall type. The water is maintained at 80-900C 
by heating with steam. The distribution system can be a ring main or closed 
line. Ring mains are favoured as the hot purified water continuously cleans 
them, but single lines are acceptable to the Food and Drug Administration 
(FDA) as long as they are regularly cleaned and validated. The pump must be 
fitted with a casing drain to allow drainage after sterilization. 
8.5.7 Heat transfer fluids 
Hot oil is used for reaction temperatures greater than about 1800C and is 
dedicated to a small number of reactors. 
The system consists of an electrically heated element, pumped loop, 
distribution pipework and expansion tank. The tank may be vented to atmosphere 
or nitrogen blanketed. The latter increases the life span of the oil by 
reducing oxidization of the hot oil at the surface. The system will need periodic 
draining and cleaning to prevent build up of carbon on the heat transfer surfaces. 
The type of oil specified is dependant upon the desired operating range, but 
the oils are normally silicone based and, therefore, have high boiling points and 
are highly stable at sustained high temperatures. 
Heat transfer oils may also be used where it is critical to prevent water 
reaching the reagents, for example, if this produces an explosive reaction. The 
vessel will then be heated using a pumped loop with the normal services 
(steam, cooling water, refrigerant) on a heat exchanger in the loop. This system 
will also need an expansion tank. 
8.5.8 Refrigeration systems 
'Fridge' systems are used to cool reactors, in batch crystallization or as vent 
condensers on volatile solvent tanks. Glycol is usually used as the heat transfer 
medium with ethylene glycol being used for nonfood use and propylene glycol 
for food use. 
There are usually two tanks, with one to hold the chilled glycol supply and 
the other to receive the refrigerated glycol return. The glycol in the return tank 
is then passed to the supply tank via the chiller or may overflow to the supply 
side via a weir system. 
The concentration of glycol is specified by the desired minimum operating 
temperature of the process vessels, so care must be taken to ensure that the
glycol concentration remains at the required level. Low glycol concentrations 
may cause freezing of the line's contents, whilst excessive concentrations of 
glycol may cause problems in the pump due to its viscosity exceeding the pump 
specification. 
The heat removed from the glycol in the chiller is either discharged to the 
cooling water or to the air via forced draft coolers. 
8.6 Sizing of systems for batch production 
The sizing of utilities requires a good knowledge of all the operations in the 
plant including the other utility operations and HVAC requirements. A large 
amount of information is required and the processing part of the plant needs to 
be designed before the utilities are designed. 
Information required includes: 
mass balance; 
energy balance; 
batch times; 
mode of operation i.e. 24 hr, 5 day etc. 
The first step is to produce lists of users for each utility with some 
assessment of the mode of operation, i.e. continuous/intermittent. The next 
stage is to attempt to assign a quantity to the users for each operation. Some 
trivial requirements can be ignored. 
Electricity 
A motor list is usually made which details power requirements and whether the 
power requirement is intermittent or continuous. Depending upon the electrical 
zoning of the plant, it may be necessary to construct a switch room for housing 
the MCC panels and control equipment. 
Cooling water 
Using the mass balance and batch times it is possible to calculate the cooling 
requirements of the process. The summertime cooling water temperature should 
be used to give a worst case. The cooling requirements of utility systems, for 
example HVAC and refrigeration equipment, need to be included here. If the 
process has more than one stage running concurrently a Gantt chart needs to be 
constructed and the heat loads for a day/week should be considered. From the 
data, a graph of duty versus time can be produced from which the peak 
requirements can be ascertained. The designer should also look at the worst 
possible case and at situations which are not part of standard operation i.e. startup 
and shutdown. Future expansion requirements should also be considered. It
should be noted that cooling towers come in a limited range of sizes, which vary 
between suppliers. The final choice of actual size is, therefore, constrained by the 
supplier chosen. As with all design sizing there is a balance between capital cost 
and flexibility of operation. 
Steam 
The method for sizing steam-raising systems is as described above, but an 
additional consideration is the required pressure. This can either be standard 
site steam pressures or an individual consideration of the desired final 
temperature within the process vessel. The flow rate of steam at the desired 
pressure can be calculated for all the duties and from the above the overall heat 
duty for the system can be ascertained. Allowance should be made for heat 
losses in the distribution system and for future expansion of the system. 
Nitrogen 
The flow rate for purging can be calculated but care must be taken in designing 
these systems for plant including filtration operations, as these are batch 
operations. Using the batch cycle time (or an estimate of this), the volumetric 
flow rate for this duty can be found. Some flow rates will be specified by 
suppliers, such as backflushing of bag filters. From the volumetric flow rate at 
the user pressure, the volumetric flow rate at the distribution pressure can be 
calculated. Again using a graph of duty versus time for the process the overall 
flow rate at the supply pressure can be found. The supply pressure will depend 
upon the users' maximum requirements. 
The system will normally have an accumulator depending upon the critical 
nature of the uses to which it is put and the method of producing nitrogen. The 
supply pressure will be reduced within the plant to give the variety of pressures 
required. There is usually a relief valve after the pressure-reducing valve to 
protect downstream equipment from an overpressure within the nitrogen 
system. The main criteria are: 
pressure required at end user and supply; 
quality; 
quantity; 
temperature at user; 
application e.g. tank blanketing, reaction control. 
Compressed air 
The ratio of the maximum to minimum capacity of the utility is known as the 
turn down ratio. All systems should have the capacity to be turned down if part
of the plant is under maintenance or if the process is changed for any reason. To 
allow for future expansion, new systems should not be designed to be operating 
at their peak loading for 24 hours a day. 
If the ratio of maximum to minimum load is greater than about 10, 
consideration should be given to the use of two or more smaller units, which 
increases the flexibility of the utility. This would increase initial capital cost but 
would, if properly controlled, reduce the running costs of the plant. Multiple 
units may also reduce down time, as the plant may be able to operate on a single 
unit when not under peak loading. 
Duty/standby 
Critical systems should have a duty standby facility such that some of the 
equipment is not run continuously. This allows time for maintenance without 
the necessity for shutdown periods. 
If there is a single duty of short duration with high flow rate, capital costs can 
be reduced by having some sort of accumulation system to allow a smaller unit 
to be installed. 
8.7 Solids transfer 
For charging biologically active solid materials into reactors, it is important to 
determine: 
the quantity to be added; 
the sizes of kegs to be used; 
the Occupational Exposure Limit (OEL); 
whether the material is explosive; 
whether contact with air is acceptable; 
whether waste bags and filters can be removed safely? 
Glove-boxes are used for solids input and kegging of primary product. The 
requirements in primary production are usually controlled by the characteristics 
of the product, i.e. the particle size range, the explosive characteristic of the 
material and whether it is necessary to exclude air or moisture. 
8.8 Cleaning systems 
All reactor systems require cleaning if a batch has failed or for period 
maintenance. Some items of plant are also used for different processes and 
cleaning between these is required, and often this must be validated.
In batch reactor systems, cleaning can be carried out by boiling either water 
or solvent in the vessel to give the degree of cleaning required. Validation of the 
cleaning procedure will be necessary. 
8.8.1 Clean in Place (CIP) 
The first thing to consider for CIP is what is to be achieved by this process and 
what is to be removed. The systems themselves are very simple, consisting of a 
tank filled with the correct concentration of cleaning medium, heated by 
recirculation to the required cleaning temperature and then introduced in the 
pipework or vessel. This is pumped through the lines and back to the tank or to 
a drain. The lines are then flushed with water and may be blown with nitrogen 
before the system goes back to production. The important consideration here is 
the superficial velocity of the cleaning medium. 
8.8.2 Steam in Place/Sterilize in Place (SIP) 
Cleaning of lines and vessels using steam can be broken into two main types — 
Steam in Place or Sterilize in Place, with the main difference being that 
Steam in Place does not have a quantitative check on the microbial content 
of the lines after cleaning and that the procedure is not validated. If the 
requirement is to minimize the biological loading of the system without 
the total removal of the biological population then Steam in Place is the 
most appropriate choice. Sterilize in Place is used in biotechnological processes 
to clean the vessel between batches and for periodic cleaning of Water 
for Injection (or purified water) storage and distribution systems. This 
process requires validation to ensure that the cleaning process can be repeated 
with confidence. 
Steam to be used for cleaning must be pure steam (see Section 8.1.4) 
and is usually reduced down to 1.2 Bar g at the point of use, corresponding 
to the usual sterilization temperature of 1210C, which is the temperature 
at which Bacillus Stereothermophilis spores are destroyed. The vessel 
is normally cleaned by CIP first, as the steam will only sterilize the 
surface, and the vessel internals are checked to ensure cleanliness. Steam 
is injected into the highest point and collected at the lowest. The time 
taken for sterilization is determined by the initial bacterial loading and 
the final bacterial loading required and is governed by the exponential 
equation: 
N = Noe~Kt 
where: N is the number of colony forming units (cfu/ml) at the end of the 
sterilization;
N0 is the number of colony forming units (cfu/ml) at the start of the 
sterilization; 
k is an empirical constant for the organism in question at the sterilizing 
temperature; 
t is time in seconds. 
Clearly, the same percentage level of reduction in biological loading can be 
achieved by sterilizing for longer at a lower temperature. 
The actual time is usually determined during commissioning by covering the 
vessel or pipework with thermocouples, and timing from when the coldest spot 
reaches the required sterilization temperature and then relating back to the 
vessel temperature probe reading. 
Vessel requirements 
The vessel must be capable of withstanding any vacuum produced by the 
sudden condensation of the steam. Care must be taken in the design of any 
vessel that is to be cleaned in this manner to minimize crevices in the vessel and 
any connecting pipework. The vessels are normally dished end design. 
Care must be taken that the condensate produced by the cleaning 
process can drain away as pockets of warm condensate will not adequately 
be sterilized. The process is validated by swabbing or by strips impregnated 
with a substance that changes colour when exposed to a given time/ 
temperature combination. 
8.9 Effluent treatment and waste minimization 
The following section is a brief overview of a broad area of knowledge. More 
detail can be found in the standard texts on the subject. All chemical 
manufacturing processes produce waste streams and, as all treatment and 
disposal costs money, it is sensible to reduce waste wherever possible. Waste 
minimization can save money but all effluent treatments have costs. The chosen 
waste disposal strategy is based on economics, regulatory compliance and 
commercial secrecy. Health and Safety has a part to play in any decision, as 
legislation requires pollution control to follow an integrated approach. It is 
unacceptable simply to move pollution from one form to another, for example, 
air stripping of ammonia from a liquid effluent to produce a gaseous discharge. 
8.9.1 Types of effluent produced by process 
Pharmaceutical processes do not tend to produce large amounts of solids but 
produce large amounts of waste water contaminated with solvent, reaction
products and inorganic salts, some waste solvents, tars from solvent recovery, 
scrubber liquors, and contaminated gaseous waste streams. 
This tends to produce small amounts of high Chemical Oxygen Demand 
(COD) waste broth, large amounts of wash waters and some gaseous effluents, 
all of which may be contaminated with microorganisms. There may be 
commercial reasons as well as environmental to prevent the organisms leaving 
the site, such as if the organism is novel or genetically engineered. 
In general, most waste streams pass to a jacketed vessel known as a kill tank. 
Periodically the vessel contents are heated to the temperature required to kill the 
organism. Here the costs of any treatment process (capital, operating, maintenance, 
disposal) must be weighed against the present cost of disposal. 
8.9.2 Options for effluent treatment (in order of expense) 
direct recycling; 
sell to waste processor, for example, waste IPA is used in car screen washes 
and waste aluminium hydroxide (from Friedel Craft's reactions) is used in 
antacid tablets; 
recovery and reuse with some form of clean up, such as solvent recovery; 
to the foul sewer with simple gravity separation and pH modification; 
incineration, although some materials such as iodine based contaminants 
cannot be incinerated because they form acid flue gases which corrode the 
incinerator; 
landfill is becoming increasingly expensive due to the reduced number of 
suitable sites, pressure by local populations and the substantially increased 
Landfill Tax. 
8.9.3 Regulatory requirements 
There are a number of regulations that relate to waste, including the following: 
Control of Substances Hazardous to Health (COSHH) (1994); 
Environmental Protection Act (1990) — the main aspects being that a 
producer of waste is responsible for knowing where that waste ends up; 
Water Industries Act (1991) — controls operation of water treatment 
companies, as well as companies delivering waste to them; 
Trade Effluent Prescribed Substances Regulations (1991) — Red List — 
this determines which chemicals cannot be released to air or atmosphere.
8.9.4 Licensing and regulatory bodies 
Water company 
The local water company grants consents for discharge of chemical waste to the 
foul sewer. Here industrial effluent is mixed with sewage and eventually ends 
up at the sewage treatment works where it is treated by various physical means, 
before being fed to bacteria and other organisms. If it is an existing site, a 
consent limit will already be set detailing flowrates and levels of contaminants. 
The water company may require information on the toxicity of the effluent to 
bacteria that break down the sewage and can ask for further information until 
they are satisfied that the effluent is not a danger to the works. 
For discharges from a process the amount, concentration of major contaminants 
and likely disposal method for each stream are required. The COD load of 
the process can be calculated and any Red List chemicals identified. 
Environmental Protection Agency (EPA) 
The Environmental Protection Agency (EPA) grants consents for discharge to 
the river system. The limits for discharge to rivers are much stricter than to the 
sewage treatment works, but it is very unusual for a pharmaceutical plant of any 
appreciable size to be discharging to rivers and not to the sewage treatment 
works. 
EPA regulates Integrated Pollution Control reports for all notifiable 
processes. A report must be submitted to the EPA which details equipment, 
process, effluent produced, control strategies. 
The EPA has also taken over the duties of the old Her Majesty's Inspectorate 
of Air Pollution (HMAIP) and consequently grants consents and regulates 
releases to atmosphere. 
8.9.5 Gaseous effluents 
The release of gaseous effluents is always controlled by regulation. There are no 
cost savings other than a reduction in raw materials costs to be offset against the 
cost of installing and operating abatement equipment. 
Characterizing gaseous waste streams 
contaminant characteristics; 
gas stream characteristics; 
design and performance characteristics.
Commonly used treatment processes 
participate: 
o hydrocyclone; 
o fabric filters; 
vapours: 
o wet scrubbing; 
o biological scrubbing; 
o absorption, adsorption; 
o combustion; 
o condensers. 
8.9.6 Liquid effluents 
For discharges to the foul sewer, the local water company usually asks for the 
following information on any aqueous effluent: 
Chemical Oxygen Demand (COD); 
Biochemical Oxygen Demand (BOD); 
Suspended Solids (SS); 
flow rate; 
PH; 
heavy metals; 
contaminants such as cumulative or persistent materials, which will not be 
broken down at the works, may build up in the water supply system. Phenols 
are also a problem as they may taint the taste of the final drinking water if 
water for potable use is abstracted downstream of the sewage works outfall. 
Many phenols also have a bactericidal effect, and may therefore compromise 
the operation of biological treatment plants. 
The water company treats each effluent on a case by case basis but will give 
a consent limit for the whole site. 
Pre treatment 
(1) Equalization 
For batch processes, a useful method of reducing loading on the pre-treatment 
system is to allow streams to mix to a more standard effluent. This optimizes the 
treatment process and reduces the amount of chemicals added, as some 
neutralization takes place within the buffer storage. This is normally achieved 
by a system of sumps or receiving tanks to smooth out the differing streams 
from a batch process. The pH is then modified to neutrality and suspended 
solids removed.
Primary treatment processes 
(1) Removal of suspended solids 
This may be achieved by a number of techniques, including flocculation and 
skimming or addition of aluminium/iron salts and gravity separation. This 
process may also remove colour and polar molecules. Turbidity, pH and flow 
are usually measured at the exit to the foul sewer and there should be some 
means of sampling the waste stream. 
(2) Removal of liquids 
Many effluents are contaminated with organic solvents, greases, and the like. 
These may be removed by means of a simple interceptor, where liquids are 
separated by means of one floating on the other, or by one of the more 
complicated systems for enhancing liquid/liquid separation. Lamella plates 
may be introduced into the interceptor, as in the American Petroleum Institute 
separator; fine bubbles may carry lighter substances to the surface for skimming, 
as in Dissolved Air Flotation (DAF); or hydrocyclones may be used to 
enhance gravity separation. All these techniques tend to decrease the required 
plan area of plant at additional capital and/or running cost. 
Secondary treatment processes 
(1) Biological treatment 
This uses a number of processes, which are conventionally split into two main 
groups, based upon whether they are carried out in the presence or absence of 
air.
Anaerobic processes are carried out in the absence of air — the organisms 
carrying out the process are actually poisoned by oxygen. These processes 
carry an advantage over aerobic processes, in that the end products of 
fermentation include hydrogen, methane, and other flammable substances. 
These substances can be burned to produce heat, or used in modified diesel 
engines to generate electricity. The process can, therefore, be a net energy 
producer if carried out at sufficient scale. The plant required for conventional 
anaerobic treatment can be very large, but newer techniques are reducing the 
size of unit operations. The higher the COD of the effluent, the more likely it 
will be that anaerobic treatment will prove suitable. Far stronger effluents can 
be treated anaerobically than aerobically, and the total containment of the 
system that is required to exclude air means that highly odorous effluents can be 
treated without causing a public nuisance.
Aerobic processes may use passive air, active air, passive pure oxygen or 
active pure oxygen to provide suitable living conditions for bacteria that 
degrade organic (and some inorganic) substances, mostly to carbon dioxide, 
water, and oxidized inorganic salts. There are a great number of techniques for 
aerobic treatment, differing in how the oxygen is brought into contact with the 
organisms, whether the organisms are free in suspension, or attached to some 
media, and whether the process is continuous or batch. There are many other 
small differences between the generic and proprietary systems on offer, but 
those preceding have the greatest effect on the important system characteristics, 
such as resistance to shock loading, running costs, capital costs and unit sizes. 
(2) Sludge treatment 
All flocculative and biological treatment processes produce quantities of 
sludge, irrespective of what some manufacturers may claim. Biological treatment 
sludge is produced in quantities proportional to the total COD put to 
treatment. There are two main problems with these sludges: their 'instability' 
(their likelihood to rapidly commence to rot, releasing noxious gases) and their 
bulk (since most biological sludges are greater than 95% water). 
Sludges may be stabilized by means of an additional biological treatment 
stage, for example aerobic digestion, or by chemical means, such as lime 
addition. This is another area with a wide range of competing solutions. Having 
consulted with specialists and decided upon the stabilization strategy, some 
means of reducing the volume of sludge is usually found desirable, especially if 
it is to be transported off-site. 
The main strategies for volume reduction are analogous to standard 
dewatering and drying techniques. Not only do they often start in a non- 
Newtonian state, their characteristics may change with feed conditions to the 
treatment process, and as a result of continuing biological activity. 
The resultant stabilized, concentrated sludges may be in the form of slurries, 
cakes, pellets, etc. These may be incinerated, landfllled, or sold for soil 
treatment. 
Physical/chemical treatments 
As well as conventional biological secondary treatment systems, there are 
several physical and chemical treatments, removing either specific contaminants, 
or groups of contaminants with similar properties. 
Ozone, peroxides, pure oxygen, air, and a number of other agents may be 
used. Although these processes tend to take up less space than biological 
methods, they can be very expensive in terms of running costs, especially with 
respect to the power costs of ozone systems.
Tertiary treatment 
In order to allow recycle or reuse of effluent treated by means of the preceding 
processes, or in the case of discharge direct to watercourse, it may be necessary 
to give the cleaned effluent a final polish or moderate its properties in some 
other way. There are again a number of different techniques for this, with 
ultrafiltration being common as a good final barrier method to prevent 
recirculation of undesirable substances. 
8.9.7 Solid effluents 
The solid effluent such as bags, filter cartridges, etc., are incinerated or 
landfilled and sludges from primary and secondary treatment processes are 
treated as previously described. There may be additional constraints on some 
solid waste, for example laboratory sharps, clinical waste, or waste contaminated 
with specific biological or chemical agents. These often require separation, 
marking of containers, and final disposal route. 
8.10 General engineering practice requirements 
8.10.1 Production area workshops 
Space is required in the production areas for: 
storage of change parts for product changes. These should be in purpose 
built units with clear identification; 
tools for changeover adjacent to the equipment. In the pharmaceutical 
industry, there are many short runs on packaging equipment and change 
over time can be lengthy particularly with blister packs; 
diagnostic equipment for fault finding; 
measuring equipment to check the environment and calibrate instruments on 
the production equipment; 
manuals and records of maintenance, although the latter can consist of a 
computer terminal. This promotes cleanliness and ensures a single central 
record is maintained; 
minor repairs and modifications; 
overhauls of equipment. 
This can be a combination of local storage units and area workshops and is 
determined by the working methods agreed in the design brief.
8.10.2 Records 
A master plant record — a logical, comprehensive set of information on the 
facility should be assembled starting at day one of design with the design brief 
and following design through all stages. Any changes in design intent and 
design decisions made should be recorded. Engineering change control will 
ensure this happens and is necessary to show the trail from design concept to 
completion. It is also a good project cost control tool. 
The framework for the record system should be established early. A 
numbering system for drawings and plant should be agreed. The finance 
department will want to record the asset value and ideally the same system of 
numbers should be used. This system will ensure that the required information: 
is available; 
can be found; 
can be updated; 
can be put into systems to monitor, calibrate, and record repairs and use of 
spares; 
can show that the plant is maintained and performing to design. 
8.10.3 Plant numbering 
All plant systems, will generally be numbered sequentially from 001. There is a 
P&I diagram for each system. A system list gives the locations and the areas 
served, which are shown simply in the P&I diagram. All items on the P&I are 
numbered sequentially with functions indicated by the symbol and prefix letters 
e.g., MDMOO1245 is a motorized damper modulating in system 1 and is item 
245 on the P&I diagram. Building management control system outstations 
sometimes control more than one plant system and this will need to be covered 
in the BMS system documentation. 
A similar method can be devised for electrical panels and distribution 
boards. 
8.10.4 Measurement and calibration 
All product significant controls or measuring elements must be calibrated and 
the calibration traceable to a National Standard. It should be possible to identify 
the procedures associated with the processes that would detect an instrument 
problem and, if there are problems, whether the procedures would detect them 
every time and soon enough. The most serious implications will be associated 
with critical instruments or instruments in safety-related applications, so a 
greater margin for error should be used in these cases.
The implications of instrument malfunction are frequently so serious that a 
cautious estimate of the calibration interval is justified — if an interval is overcautious, 
it will soon be revealed as such. 
This is an activity that may be desirable to keep in-house. List the types and 
numbers of instruments to be calibrated and the frequency of calibration to 
determine the staff and space required. It may be possible to draw on the 
experiences of calibration from other sites; the same instrument in a similar 
application may exist, with several years worth of calibration history (e.g. 
magnehelic gauges) and an optimized calibration interval. This will improve 
the level of confidence in an estimated calibration interval but must not be used 
as a substitute for a thorough evaluation of each application; each will be 
unique in some respect. 
Bear in mind that the period between calibrations can be increased if 
successive calibrations show no deviation. For example, after three successive 
calibration checks without need for adjustment, it may be possible to double the 
calibration interval. 
8.10.5 Computer systems 
Software packages, such as Computer Aided Maintenance Management 
System (CAMMS) are available but these will only assist with handling the 
data rather than determining the system. The software package must be 
validated (see Chapter 4). 
The system should be chosen early and records added. The cost and 
problems of trying to enter the information after the plant has been handed 
over usually result in incomplete records. If staff who will ultimately use the 
system enter the data as the work progresses, they will learn the system and the 
plant. It is essential to manage the quality of this data not only at entry but also 
throughout its required life. Failure to do this effectively will render the 
CAMMS system useless and an expensive burden on the operation. 
A corresponding reference system for manuals should be set up. Backup 
copies of all software and records should be made and stored in a secure fire 
resistant area. 
8.11 Installation 
8.11.1 Staff duties 
The maintenance engineer should be part of the project team. 
The technicians should be on-site from the beginning and they should be 
sent on acceptance trials of major plant. There should be a budget for minor 
changes, to improve maintenance, and a rigid change control followed.
The technicians should be involved in the IQ/OQ and should ensure that all 
drawings represent 'as built' and are marked up as the installation progresses. 
8.11.2 Training 
The core team of technicians may have been selected for their knowledge and 
experience but they will need further training in analytical skills and fault 
finding. Involvement in the project is a good training activity and technicians 
can be trained at suppliers during construction. If it is planned to contract out 
maintenance, their designated staff should be trained on the equipment. They 
will also need training on cGMP practices and, if a CAMMs system is in place, 
they will need training on its use. 
8.12 In-house versus contractors 
Suppliers of large capital equipment such as refrigeration plant and specialist 
systems such as fire sprinklers have contract maintenance departments. 
In USA and Canada the trend is to contract out facilities management, and 
consultants and contractors are set up to carry out this service. Major 
contractors in the UK are now investigating the feasibility of offering this 
service, as they already have the organization to manage sub-contractors and 
have established working relationships with preferred suppliers. 
An Invitation to Tender or Request for Proposal will be needed, which will 
specify the requirements and the measures used to compare bids and monitor 
performance. An Invitation to Tender will typically have the following headings: 
background; 
objectives; 
present situation; 
proposed system; 
company needs; 
nominated staff and qualifications of staff; 
job functions; 
tasks for various job functions; 
reports required; 
confidentiality; 
vendors qualifications; 
timing of proposal; 
format of reply; 
contractors guarantees; 
evaluation of proposal.
The evaluation of proposal lists all the information that is needed to compare 
bids, such as rates, response times, references, safety record etc. 
Partnership is another concept, with agreed performance and profit sharing 
on improved efficiency. 
Contracting out the maintenance is not a simple option. There will still need 
to be sufficient in-house expertise to effectively control the relevance and 
quality of the external work. 
8.13 Planned and preventive maintenance 
8.13.1 Reasons for planned maintenance 
improved equipment reliability; 
reduce lost production time; 
cost avoidance; 
unscheduled repairs and downtime; 
cost control; 
more accurate budgets; 
satisfy FDA and local requirements; 
we deserve a good nights sleep! 
8.13.2 Planned maintenance 
Planned maintenance, in its simplest form, is applying the manufacturer's 
routines to the plant at the frequencies they recommend. If done conscientiously 
and properly, this will reduce breakdowns but it is labour intensive and 
can result in application with no thought to hours run, duty and environment. 
Many routines are invasive and can affect the plant if not done correctly. It can 
result in over-maintenance and rarely can be completed due to pressure to 
reduce downtime. 
Improvements have been made using hours run meters on the starters and 
BMS systems to log hours run. 
8.13.3 Preventive maintenance and reliability centred maintenance 
This requires a better understanding of the plant and its use. It involves more 
extensive examination and review of inspection reports and repair work; an 
assessment of the potential for failure; emphasis on methods of assessing 
failure and effort concentrated on those items likely to fail and whose failure 
has the most significant effect on the facility.
It uses techniques of condition monitoring: 
observation and use of analytical skills; 
analysis of oils; 
vibrations analysis; 
Sound/sonic testing; 
Infrared testing. 
All the above require a base line of the 'as installed', new condition as a 
reference. 
Good preventive maintenance requires: 
systems (manual or computerized) to track, schedule and record the 
preventative maintenance; 
system of identifying equipment uniquely; 
good equipment records; 
written procedures; 
following procedures; 
technically competent resources; 
safe working practices and training. 
8.14 The future? 
More companies will offer contract maintenance and facility management 
services. The engineering function will reduce in number but increase in 
engineering and management skill. Plant will be computer monitored and 
controlled. Confidence and knowledge of computer systems and software will 
increase and BMS will be used more, removing parallel monitoring 
and measurement systems. (This is dependent on the BMS software being 
validatable). 
Trouble free operation requires effort. It starts by clearly defining the 
engineering operating objectives at the beginning of a project and using 
these to determine the strategy and organization of the engineering department 
and to prepare a plan to bring this about. Then it requires a lot of detailed effort 
throughout design and construction on the design and organization. 
Then, once this is in place, performance should be measured, reviewed and 
improved.
Bibliography 
1. Haggstrom, M., New Developments in Aseptic Design Relating to CIP and SIP, 
Biotech Forum Europe 3 (92) 164-167. 
2. Latham, T., 1995, Clean steam systems, Pharmaceutical Engineering, 
March/April. 
3. Smith P.J., 1995, Design of clean steam distribution systems, Pharmaceutical 
Engineering, March/April. 
4. FDA Guide to Inspection of High Purity Water Systems, July 1993. 
5. Honeyman, T., et al, 1998, Pharmaceutical water: In over our heads? European 
Pharmaceutical Review, Aug. 
6. Pharmeuropa, 1997, (9) 3 Sept. 
7. Clean Steam, booklet published by Spirax Sarco. 
8. US Pharmacopoeia 23 Fifth Supplement, Water for Pharmaceutical Purposes 
General Information, pp. 3547-3555. 
9. US Pharmacopoeia 23 Fifth Supplement, Purified Water, pg 3443, Water for 
Injection, pp. 3442. 
10. Metcalf and Eddy, 1991, Wastewater Engineering Treatment Disposal and Reuse, 
3rd ed (McGraw Hill, USA). 
11. Baseline Pharmaceutical Engineering Guide VoI 4\ 'Water and Steam Guide, ISPE, 
1999. 
Acknowledgements 
The following persons are thanked for their invaluable help with the writing of 
this chapter: Roger Freestone, Ken Gutman, Trevor Honeyman and Sean 
Moran.
9.1.1 A need for quality control 
Safety has escalated to number one on the agenda of pharmaceutical companies 
worldwide. Quality Control is the mechanism by which safety is achieved and 
measured and the Quality Control (QC) laboratory provides a crucial and 
integral role in achieving the safety objective. 
There is a wide spectrum of laboratory types, from schools through to 
genetic research, undertaking tasks which may take only a few minutes or 
literally years to complete. The focus of this chapter is on QC laboratories and 
their purpose, operational requirements and design features, many of which are 
common to other laboratory types. The QC laboratory has an important place in 
pharmaceutical production. The activities undertaken in the laboratory rarely 
contribute directly to the pharmaceutical manufacturing process, but the 
function of the laboratory remains essential to the final product. 
9.1.2 Complex issues require clear procedural guidance 
Earlier chapters provide a basic understanding of the complexity of pharmaceutical 
production. To appreciate how important it is to have a structured and 
quantifiable approach to any production process it is necessary to examine the 
process beyond chemistry and biotechnology. Dividing the process into 
functional categories reveals opportunity for failure error in each. In developing 
an understanding of how, even within the most highly automated facilities, 
there is infinite scope for something to go wrong, it is clear that due attention 
should be paid to the preparation and implementation of safe operating 
procedures. Consider the implications for controlled functionality in each 
category: 
• facility: 
o construction and materials; 
o maintenance and cleaning; 
L a b o r a t o r y d e s i g n 
DUNCAN LISLE-FENWICK 
9.1 Introduction 9
o age — wear, corrosion, deformity; 
o control and measure — accuracy, calibration; 
o warning systems — detection and alarm. 
operatives: 
o skill level — training, experience; 
o awareness — familiarity, tiredness; 
o attitude — positive, safety conscious, composed. 
environment: 
o temperature, humidity, air flow — direction, velocity; 
o contamination — to the product, to the environment; 
o hazardous — explosive, flammable, toxic. 
raw materials: 
o quality — composition, constitution; 
o storage and transportation — stability, containment, shelf life; 
o dispensing, handling, containment. 
Design, construction and operating codes and standards exist to ensure 
all factors are given due consideration and a consistent approach. The 
designer, constructor and operator use knowledge and experience in the 
endeavour to provide a facility which functions safely and correctly time 
after time. 
In pharmaceutical production scientific accuracy is the major factor 
contributing to repeatability. Accuracy is the degree to which measurement 
can be recorded. The principle measurements are: weight, volume, velocity, 
duration (hence flow rate), temperature and pressure. Measurements 
apply to solid, liquid or gas states or any combination producing slurries, 
solutions, vapours etc. Precise measurement is essential to avoid potentially 
catastrophic reactions and of course it is crucial to the effective product 
formulation. The process itself introduces stringent specifications for 
equipment and machinery to attain the high tolerances imperative to the 
uncompromising quality demanded. 
9.1.3 What is the purpose of quality control laboratories? 
The pharmaceutical process involves design, material selection, product 
manufacture and finishing. Each process conforms to codes of practice, 
regulatory standards and statutory legislation in an effort to produce consistent 
product quality. A clearly defined, structured and regulated process 
is the quality assurance demanded by the market for any product to 
succeed. Quality control establishes the measure of confidence that the 
market has in any product. All products rely upon consumer confidence.
Manufacturers build their reputations on the quality of their products, 
reputations that are established by years of faultless products. Reliability 
can only be achieved through strict quality control. Pharmaceutical 
production demands strict quality control maintained by thorough checking 
and inspection, constant monitoring and rigorous testing performed 
scientifically against exacting specification criteria — enter the quality control 
laboratory. 
9.1.4 What purpose do quality control laboratories serve in 
pharmaceutical production? 
To appreciate how important quality control is to pharmaceutical production, 
the analogy of a familiar, tangible product, similar, albeit simpler, in process to 
that operated in pharmaceutical manufacture will be used. Consider the humble 
cornflake, we know exactly what to expect, a consistent product time after time. 
Quality control procedures guarantee to deliver the same quality product 
virtually every time. Confidence that quality is maintained, the product is 
purchased without hesitation or doubt, yet the level of quality control that 
produces cornflakes to satisfy the publics'discerning palette is not high enough 
to meet the demands of pharmaceutical production. On the rare occasion that a 
burnt cornflake is encountered in the breakfast bowl, it is simply removed 
without thought. Subconsciously, a quality control inspection has been 
conducted, as happens every day before anything is bought or consumed. 
This ultimate quality control inspection is an impossible task when applied 
to a pharmaceutical product. Typically the product is artificially coloured, 
artificially flavoured and has an artificial aroma. To further confound the senses, 
the active ingredient is a fractional component of the dosage form. Consequently, 
human senses and judgment cannot be relied upon to verify the quality 
of pharmaceuticals. Fortunately, manufacturers can be relied upon to supply the 
precise dosage of active drug every time. 
9.1.5 How does the quality control laboratory benefit pharmaceutical 
production? 
Regulatory compliance is the subject of Chapter 2 and reference should be 
made to that chapter for a detailed understanding of regulatory aspects. With 
regard to QC laboratories, regulatory compliance is concerned with the 
continuance of the product licence. Laboratory samples and test results must 
be strictly maintained and catalogued for easy access. Product traceability is 
essential as it is the essence of validation. Should the burnt cornflake scenario 
ever occur in a pharmaceutical product, the consequences could be fatal and
widespread, and it is vital that the root cause is quickly identified and isolated. 
The priorities for traceability are: 
prevention of further unnecessary victims arising; 
evaluation and quantification of the problem; 
possible development of an antidote; 
identification of other affected products; 
rectification of the root cause. 
Validation is an all-encompassing process; it begins at the design stage and 
continues through into operation. Each step must satisfy regulatory guidelines 
and be precisely documented. This approach to pharmaceutical production is 
known as Good Manufacturing Practice (GMP) (see Chapter 3). 
Quality control is one thing, but care must be taken not to confuse 
validation with quality control. Quality control is an integral part of 
validation. The onerous procedures pursued in securing a validated product 
must only be applied to the appropriate steps of the process to avoid 
unnecessary expense administering the procedures and exhausting effort 
maintaining the high standards that are a prerequisite of the regulatory 
authorities. 
Perhaps surprisingly, regulatory compliance is complementary to commercial 
viability. Commercial viability of pharmaceutical products relies on 
consumer confidence in the product. This confidence is based on the 
manufacturing companies reputation. The company's reputation is built on 
their ability to demonstrate repeatability and reliability. Independent regulation 
provides an ideal vehicle for marketing that ability. There are other 
commercial benefits to GMP. Pharmaceutical manufacture is an expensive 
business, whether batch or continuous process. Rigorously structured and 
controlled production improves efficiency, reduces waste and manages plant 
shutdown. Large pharmaceutical companies lead in the field in development 
and improvement of production facilities. Whether inadvertently or planned, 
developments in manufacturing technique and improvements in equipment and 
control systems have led regulatory authorities to raise the standards of 
acceptability. 
9.2 Planning a laboratory 
9.2.1 Design concept 
The most important factor in designing a laboratory is safety.
Aspects of safety that should be considered when evaluating laboratory 
design should fall into two categories: 
• physical space; 
• air flow control. 
Physical space 
The definition of physical space is controlled by a number of criteria, often 
conflicting and always challenging the skill of the designer to harmonize 
between regulatory compliance, functional requirement and available space. 
Function and operation 
Establish the activities undertaken in the laboratory. Determine the space 
requirements for each activity and any special features associated with the 
function. 
• bench space: Typically determined by the laboratory equipment size and any 
peripheral equipment such as PCs and printers. Depth should be considered 
as well as length; 
• bench height: 900 mm is standard for activities undertaken from a standing 
position with normally transient attendance by the operator. Stools are 
usually provided for occasional use. 750 mm is standard for activities 
conducted in a seated position, usually where the operation duration is 
extensive; 
• bench frame construction: A variety of frame options are available, 
generally of steel construction. Each provides a combination of features: 
o underbench unit: floor standing or suspended; 
o frame visibility: exposed or concealed; 
o structural support: floor or wall and/or spine; 
Selection of a frame type will depend on a number of criteria: 
o flexibility: ease of repositioning/replacing units; 
o cleaning: access to floor space below and behind unit; 
o integrity: load capacity depends upon combined structural integrity of 
structural supports, the frame and the under bench units; 
o appearance: exposed frames can dominate the overall appearance; carefully 
considered, they can add feature interest to the laboratory design. 
Concealed frames reduce the amount of dirt traps providing a more 
hygienic aesthetic; 
o cost: flexibility, cleanliness, strength and aesthetics each come with a price 
tag — specify appropriately to the task duty and responsibly to respect the 
budget;
o special: special heavy-duty frames with anti-vibration mounts are available 
where vibration sensitive equipment is to be used such as finely 
calibrated balances. 
o storage space: well-planned and ample storage is essential to safe 
laboratory operation. Every instrument, container, reagent etc. should 
have a dedicated and purpose designed home to promote efficiency and 
safety in the laboratory. For this reason, a diverse range of storage unit 
types are available; from a simple, eye level, glass reagent shelf to special 
ventilated cabinets in fireproof construction with automatic door closers. 
The range of storage unit styles is too extensive to list here. Each 
manufacturer has a large selection of modular units to complement their 
laboratory bench systems. A popular solution to storage problems is storage 
wall systems, integrating a variety of unit types within a modular frame over the 
entire length and height of a wall. The generic requirements for each type of 
storage unit are discussed in this chapter. 
Typical types of storage to be considered include: 
• under bench: Cupboard, drawer or combined units. Internally cupboards 
may be provided with shelves or may house equipment such as vacuum 
pumps, waste disposal units, etc. Drawers may be supplied with an array of 
guides specifically designed to hold equipment, glassware, etc. in an 
efficient, tidy and safe manner; 
• safety cabinet: 
o personnel emergency safety equipment; 
o breathing apparatus — gas masks etc.; 
o first aid — medical kit and instruction; 
o fire fighting — hand held extinguishers; 
o hazard spillage — absorbent sand. 
• pull-out storage: Each with entire pull-out units or individual pull-out 
shelves. Each designed to provide easier access to otherwise deep storage 
space where there is a risk of upsetting objects stored close to the front. It is 
particularly useful for glassware and chemical storage; 
• solvent/flammable storage: Provided with a system of mechanical extract 
ventilation discharging to atmosphere, designed to prevent the build up of 
flammable vapours within the cupboard. Enclosed in a fire resistant casing to 
contain any fire for a specified period. Fitted with an automatic door closer 
that is activated on detection of fire. This type of cupboard may be fitted with 
carousel, rotary shelving to reduce the risk of accidental spillage whilst 
containing any vapours within the cupboard. Shelves are lipped and a
removable collection tray is housed in the bottom of the cupboard to contain 
any spillage; 
• acid/alkali/chemical storage: Provided with a system of mechanical 
extract ventilation discharging to atmosphere, designed to prevent the 
build up of toxic fumes within the cupboard. This type of cupboard may 
be fitted with pull out shelves and is usually lipped to contain spillage. A 
removable collection tray in the bottom of the cupboard is provided to 
contain any excess spillage; 
Construction materials used for the storage of chemicals, solvent, acid 
and alkali must be considered carefully, particularly where spillage is likely 
to occur. All materials have some inherent weakness that causes it to react 
with the chemical resulting in corrosion, softening/dissolving, ignition/fire, 
toxic emission or simple mechanical failure. Common materials used include 
fiberglass, galvanized steel, stainless steel, polypropylene and glass — each 
selected for chemical compatibility and physical suitability. 
• controlled temperature: Often in laboratory operations, it is necessary to 
store materials at low temperatures. This may be in refrigerator units with 
storage temperatures a few degrees above zero or freeze units providing subzero 
storage or at the extreme, cryogenic storage systems achieving —830C. 
Each of these units may require floor space within the laboratory. Typically 
they are freestanding vertical units with a single door, internally divided into 
compartments with individual pullout trays. Temperature controls and 
displays are clearly visible on front of the units. Usually units are designed 
to suit a 600 mm module. 
Operational considerations 
(a) Laboratory equipment 
The requirements for laboratory equipment will depend upon the procedures to 
be conducted within the laboratory. Laboratory procedures are generally 
analytical. The laboratory operator prepares a schedule of equipment with 
approximate sizes, which will indicate the safety considerations for each piece 
of equipment, specifying where fume hoods or fume cupboards are required to 
control emissions. The schedule may include useful information on service 
utilities for equipment, power, gas, water, air, etc., complete with loads, 
flowrates, and diversity figures. Typically the equipment includes: 
• gas chromatographs (GCs); 
• high pressure liquid chromatographs (HPLCs); 
• rotary evaporators;
• ovens; 
• furnaces; 
• ultrasonic baths; 
• balances. 
Armed with this information it is possible to evaluate the basic quantity of 
furniture items required to satisfy the demands of the laboratory operation. 
(b) Ancillary equipment 
A host of equipment and storage facilities is required to support any laboratory 
operation. Guidance is required from the laboratory operator as to the most 
appropriate and essential items, but generally these will include: 
o glassware washers and driers; 
o refrigerators; 
o freezers; 
o safety station — eyewash and safety shower; 
o water purifiers; 
o vacuum pumps; 
o gas generators or cylinders; 
o all types of storage. 
(c) Personnel and ancillary space 
Laboratory operators undertake a number of functions within the laboratory 
and, whilst they may spend a lot of time at the workbench, they also need an 
area for report writing and filing. Outside the laboratory, facilities are required 
for personnel washing and changing, rest and recreation and archive storage of 
records and samples. 
(d) Workflow 
The definition of space requirements discussed above provides a quantitative 
analysis of space requirements for the laboratory. To begin to plan a laboratory 
into a useful layout requires an understanding of workflow. 
The laboratory operator has the best understanding of workflow and work 
patterns within the laboratory. A simple flow chart or bubble diagram by the 
laboratory operator will ensure the laboratory design satisfies the demands of 
the busy schedule of activities in the contemporary laboratory. 
Workflow should aim to be in one direction with necessary support facilities 
provided at each step. Back tracking and cross-over should be avoided as these 
dramatically increase the risk of accident. 
Timing is important — analytical processes may take minutes or hours to 
complete. The slowest process dictates the throughput of the laboratory.
Workflow is improved by increasing the numbers of critical equipment items 
(subject to budget). 'Bottlenecks' should be identified and recorded. 
Whilst it may be practical to provide utility services to all bench areas, costs 
aside, it is not always practical to provide additional space for process and 
utility activities in sufficient number to meet the demand; any limitation must 
be accepted by the laboratory operator. 
Storage space is essential. Storage must be well distributed around the 
laboratory. Glassware and other implements should be readily available from a 
number of local storage units. Chemicals should generally be dispensed from a 
central safe storage location. Trolleys may be used to transport chemicals safely 
and as a mobile workbench. The laboratory layout must make adequate 
provision for safe parking of the trolley whilst it is in use as an extension to 
the work area. 
(e) Material flow 
Sample receipt, handling and storage feature highly in the work flow requirements 
for the laboratory. 
Once a sample is received into the laboratory it is catalogued before being 
processed further. The sample is then dispensed into a number of units for 
different analytical procedures, each catalogued according to the batch requirements. 
All handling operations must be undertaken with due regard to safety, 
requiring the use of safe working practice and safety procedure. The use of 
adequate protective clothing and specialist equipment are essential. The 
laboratory design must make provision for storage of safety equipment, clean 
and dirty protective clothing. Changing facilities with showers may be required 
for some facilities. Clearly identifiable disposal units, segregated according to 
hazard are as important to safety as safe handling of materials. The laboratory 
operators must have reasonable access to a safety shower and eyewash 
facility. 
The route for analytical procedures should be planned to be in one direction 
only, with no crossing of paths or doubling back. Consideration must also be 
given to the segregation of the different operations — for example, wet 
chemistry areas are designed to contain spillages and splashing whilst balances 
are often placed in separate rooms to minimize the effects of adverse room air 
turbulence and moisture. 
(f) Work scope 
In large laboratory buildings, different functions are undertaken in separate 
laboratory rooms, each with appropriate facilities and finishes. In major 
research complexes, individual laboratory buildings may be designed for
different research areas including chemistry, biology, microbiology, biotechnology 
and animal research (which owing to its political sensitivity is more 
often referred to as Central Research Support Facility or Biology Support Unit). 
There are many support functions which may be undertaken within 
laboratories such as small-scale production (for clinical trials), kilo labs, 
instrument and equipment calibration, dispensing and preparation of chemical 
additives (subject to regulatory restrictions), physical testing. 
(g) Personnel flow 
Laboratories are hazardous places. The high level of manual handling of 
dangerous materials, including flammable, toxic, corrosive, radioactive, carcinogenic, 
bacterial, viral and pathogen, place operators into potentially lethal 
environments. Whilst laboratories are generally restricted to small quantities of 
such materials, the consequences of an accident may not be confined to the 
laboratory, placing the environment and local communities at risk. 
Whatever the risk or consequence, strict manual handling policies must be 
adopted. The laboratory designer must consider the philosophy when establishing 
the basic design and layout. Personnel need to be able to move around the 
laboratory freely without cause to disturb colleagues who may be undertaking 
hazardous operations (albeit with controlled conditions). The operator may also 
be required to manoeuvre a trolley or cart, carrying hazardous materials, around 
the laboratory. To ensure these functions are undertaken safely, adequate space 
must be provided between benches. Fume cupboards need to be positioned 
where operators have room to manoeuvre freely without being cramped by 
walls or other fixtures and clear from potential collision with other operators 
and mobile equipment. The diagrams in Figures 9.1-9.3 (pages 314 to 320) 
illustrate the general principles of spacing within a laboratory. 
(h) Fume cupboards 
The use of fume cupboards within a laboratory varies considerably depending 
upon the nature, frequency and duration of activities undertaken which are 
either hazardous or are susceptible to contamination. When considering what 
operations are undertaken within a fume cupboard, it is important to evaluate 
the viability of multi-function use. Where apparatus can be set up and 
dismantled in a relatively short time and frequency of use is low, fume 
cupboards may be utilized for a number of different activities. Keeping the 
number of fume cupboards low not only saves space and capital costs, it also 
aids HVAC design. Fume cupboards extract enormous volumes of air from the 
room. By the nature of a fume cupboard operation, this air must be exhausted to
(a) Separation of undisturbed zone 
from traffic routes 
(b) Spacing where same operator 
uses fume cupboard and bench top, 
or where occasional traffic only is 
anticipated 
(c) Spacing determined by airflow 
requirements 
(d) Spacing determined by airflow 
requirements 
Bench top 
Figure 9.1 Minimum distances for avoiding disturbances to the fume cupboard and its 
operator
atmosphere. Detail on the design of air systems is discussed later in this chapter 
in Section 9.5. 
There are a number of different types of fume cupboards available depending 
on operational requirements. The construction details of each are described 
(f) Spacing that avoids undue 
disturbance of airflow. 
Face of column not in front of plane 
of sash 
(g) Spacing that avoids undue 
disturbance of airflow. 
Face of column in front of plane 
of sash 
(h) Spacing that avoids undue 
disturbance of airflow. 
Except where door includes 
air transfer grilles 
(j) Spacing that avoids undue 
disturbance of airflow. 
Except where door includes 
air transfer grilles 
Figure 9.1 {Continued) 
300 
1500 
Fume cupboard Fume cupboard Fume cupboard 
1000 
Fume cupboard
(a) A bench at right angles to cupboard 
face may keep traffic away from 
undisturbed zone but work at bench 
will cause disturbance to air flow 
(b) Projecting bench will help to keep 
traffic clear of undisturbed zone and 
work at bench will have little effect on 
air flow if sufficient distance between 
cupboard and projecting bench is 
allowed 
(c) Projecting walls and the positioning 
of doors can be effective in defining 
traffic routes 
(d) Columns can assist the 
definitions of traffic routes 
Figure 9.2 Planning arrangements for avoiding disturbances to the fume cupboard and 
its operator from other personnel 
300
1000 
Bench top 
Kime cupb<5ard 
Bench top 
Fume cupboard Fume cupboard Fume cupboard
(e) In a small laboratory, the 
fume cupboard should be clear 
of personnel entering through 
doors 
(f) Too much movement 
in front of fume cupboards 
should be avoided by 
providing more than the 
minimum distances between 
faces of fume cupboards 
and bench tops 
(g) Too much movement 
in front of fume cupboards 
should be avoided by 
providing more than the 
minimum distances between 
faces of fume cupboards 
and bench tops 
Figure 9.2 (Continued) 
Bench top 
Fume cupboard 
Zone for doors 
1000 
Bench top 
Bench top 
Bench top 
Bench top 
Fume cupboard Fume cupboard 
Bench top 
Bench top 
Fume cupboard I Fume cupboard 
Bench top
Figure 9.3 Escape routes 
later in this chapter in Section 9.4. The principle selection criteria are 
summarized below: 
o size: Generally available in modular widths to complement laboratory 
benches: 1200 mm, 1500 mm, 1800 mm, 2000 mm, 2100 mm are typical; 
o sash: Sashes come in a variety of configurations, with vertical sliding 
being the most common. Horizontal sliding is restrictive but when 
(a) Escape routes should not 
cross a hazard area where 
there is no alternative escape 
route 
(b) Escape routes should not 
cross a hazard area where 
there is no alternative escape 
route 
Bench top 
Fume cupboard Bench top 
Bench top
Bench top Fume cupboard
(d) Alternative escape routes should supplement an 
escape route that crosses a hazard 
Figure 9.3 {Continued) 
Bench top 
Traffic will pass close to front of cupboard 
(c) Principle escape routes should not cross hazard areas 
Bench top 
Alternative escape routes 
Bench top 
Bench top 
Fume cupboard 
300 
Bench top 
jBench top 
eench top 
BeVh top 
Fume cupboardFume cupboardFume cupboardFom^cupboardl
Alternative escape routes 
Bench top 
Bench top 
Bench top 
Bench top 
Bench top 
Fume cupboardFume cupboardFume cupboardFume cupboard 
(e) Alternative escape routes should be provided from all hazard areas 
in laboratories with more than one fume cupboard 
Figure 9.3 {Continued) 
combined with vertical sliding it provides a more versatile arrangement. 
Large sashes are often split horizontally to limit travel and headroom 
requirements. Sashes normally start at bench top level. If large equipment 
is envisaged then a lower level is appropriate. Some fume cupboards are 
'walk in' to accommodate large or heavy apparatus. 
o safety: Most fume cupboards designs are intended to protect the operators 
from the hazardous materials being handled. This is achieved by creating a 
negative air pressure across the open sash face. The velocity across the 
face is usually measured at 0.5ms"1 (termed face velocity). Maintaining 
the face velocity for a variety of sash sizes and opening heights is the 
fundamental design principle for fume cupboards. 
Other safety considerations are that the fume cupboard must offer 
protection to the operator from fire and explosion, both of which demand 
careful consideration in the selection of suitable construction materials. 
Details of construction are discussed later in Section 9.4. 
o facilities: The function of the fume cupboard determines the nature and 
number of facilities. These will include utility and laboratory services such 
as power, water, air, gas, etc.; equipment frames; sinks and troughs etc. 
Details of all available services are included later in Section 9.6. 
o air systems: Although air systems will be covered in depth further on in 
this chapter, it is worth mentioning that there are two fundamentally
different types of fume cupboard — those which extract all air and those 
which recirculate air. Recirculatory fume cupboards rely upon local filters 
to ensure a safe working environment is maintained. This type must only 
be used for low risk operations. Total extract type remove all air to 
atmosphere, thus, providing a safe working environment within the 
laboratory. 
9.3 Furniture design 
9.3.1 Bench construction systems 
Figure 9.4 on page 322 illustrates some of the common bench construction 
systems. 
Pedestal furniture 
The pedestal system of benching provides a rigid bench construction by directly 
supporting the work surface on the underbench units of furniture. 
The system is highly cost-effective and commonly features a wide range of 
modular size units to suit most installations. This pedestal system is an ideal 
solution in those applications where there is an infrequent requirement for 
underbench furniture to be interchanged, although should any changes be 
needed, they can easily be carried out utilizing the services of a maintenance 
department. 
1C frame bench construction system 
This type of bench construction system is ideal for applications where 
flexibility in the choice of units is a requirement, together with a clear floor 
space for cleaning. 
The system provides a rigid bench construction capable of taking heavy 
loads. It does not require any floor or wall fixings. 
This type of system accepts both suspended and movable types of furniture. 
Also with both types of unit, the framing allows the units to be placed adjacent 
to one another without gaps. 
Cantilever bench framing 
Cantilever bench framing is ideal in installations where flexibility and ease of 
floor cleaning is required, as there is no horizontal floor leg to cause any 
obstruction. The design allows for suspended, movable and removable underbench 
units to be placed anywhere along the length of the benching and 
repositioned at any time without interference.
U ^ J K J ^ I ^ L J 
Under bench unit on plinth Under bench unit on leg frame Under bench unit 
on "C" frame 
L J Gl \ ^ \ ^r 
Movable under bench unit Heavy duty cantilever framing Heavy duty cantilever framing 
on "C" frame suspended under bench unit movable under bench unit 
H H N P 
Table frame bench construction Table frame bench construction 
suspended under bench unit movable under bench unit 
Figure 9.4 Laboratory bench framing and under bench units 
Available as both standard and heavy-duty cantilever supports, each requires 
a degree of wall-support for any perimeter benching. Both types are also 
suitable for island and peninsular benches. 
• standard cantilever framing: This system is designed solely for use with 
removable under bench units; the units themselves provide the necessary 
additional support to the worktop; 
• heavy-duty cantilever framing: The alternative heavy-duty cantilever support 
system is manufactured from heavier steel sections and, although needing 
more robust wall and floor fixings, is suitable for movable and suspended 
underbench units.
Table frame bench construction system 
The construction of the table frame is designed to offer both flexibility and 
economy. It is rigid and can accept heavy loads with minimum deflection. The 
design will accommodate either suspended or movable furniture units. 
When used against a service spine accommodating the mechanical and 
electrical outlets, further flexibility may be achieved by using modular table 
units, which obviate the need for long runs of benching. Table frames are 
generally fitted with adjustable feet for levelling. 
Tall storage cupboards 
A wide range of tall storage cupboards is available: 
acid/alkali cupboard; 
solvent storage cupboard; 
safety cabinet; 
storage cupboard with pull-out shelves. 
Accessories 
A large range of integrated accessories is available, such as a comprehensive 
range of drawer dividers. 
9.3.2 Bench top materials 
There is a varied range of bench top material available to suit any application. 
Materials are extensively tested. The most popular materials are detailed below, 
but this is by no means an exhaustive list. Sizes quoted are typical for the 
material. The suitability of each material for use with a range of chemicals is 
summarized in Table 9.1. 
Laminate 
These bench tops have a thickness of 30 mm. They are covered with laminate 
with a rolled front edge and bonded to a high-density particle board base. All 
ends are sealed with a 4 mm thick edging strip of polypropylene. 
Epoxy resin 
Epoxy resin tops are manufactured from solid epoxy resin and are selfsupporting. 
The tops generally have a thickness of 15 mm, with a dished 
edging strip 10 mm high, giving a 25 mm thick edge; an alternative is available 
with a thickness of 19 mm with a 6 mm raised edge.
Table 9.1 Chemical Resistance Chart 
Kambala Iroko 
Melamine Laminate 
Solid Grade Laminate 
Stoneware 
Epoxy Resin 
Stainless Steel 
Tiles 
PVDF 
PP 
PVC 
Glass 
Slate 
Linoleum 
ACIDS 
Sulphuric 
Hydrochloric 
Fuming Nitric 
Perchloric 
Nitric 
Chlorine 
Hydrofluoric 
REAGENTS 
Ammonia 
Sodium Hydroxide 
Silver Nitrate 
Potassium Permanganate 
Iodine (in 15% Potassium Iodide SoIn.) 
Bromine 
STAINS 
Malachite Green 
Crystal Violet 
Carboxy Fuchsin 
SOLVENTS 
Acetone 
Toluene 
Methyl Alcohol 
Carbon Tetrachloride 
Diethyl Ether 
: No effect 
: Slight staining after wiping surface clean 
: Severe staining and potential corrosion after prolonged use 
Not suitable
Solid grade laminate 
These tops are normally fabricated from 20 mm thick boards with the edges cut 
square and polished. Alternatively the front edge can be radiused and polished. 
Solid wood 
These tops are generally available in Iroko, Kambala or Beech with a thickness 
of 25 mm or 30 mm. They are constructed from narrow boards jointed with 
special 's' joint and waterproof glue. They can be linseed oil finished or 
varnished. 
Stoneware 
The tops are of solid acid-resistant, glazed stoneware. All tops have a thickness 
of 30 mm, with a raised front edge 7 mm high. End edging strips of PoIybutylene 
Teraphthalate (PBTB) are available for protection and dishing of ends. 
Stainless steel 
Two types of stainless steel top are normally available either bonded onto a 
wood core or self-supporting with suitable reinforcing on the under side. Tops 
are usually manufactured from Type 316 acid resistant stainless steel. The 
standard construction is either flat with an overall thickness of 25 or 30 mm, or 
with a raised edge with a thickness of 32 or 37 mm. 
Tiles 
Tiled worktops are manufactured utilizing a laminated board base, with all 
surfaces double-sealed with epoxy resin. First grade chemical-resistant tiles are 
bonded to the base and jointed with chemically resistant cement to an epoxy 
grout. 
These tops are available as flat worktops or with a raised edge. Flat tops have 
a thickness of 30 mm. Raised edge tops have a thickness of 37 mm. 
Plastic veneered 
Plastic veneered tops are available in three types of veneer: 
polyvinylidene fluoride (PVDF); 
polypropylene (PP); 
polyvinyl chloride (PVC). 
All tops are covered with one of these materials bonded onto a high density 
chip board core, with either a flat edge, or with a raised edge all round. In the 
latter case, the special raised front edge section is welded to the work surface 
and taken down from the front edge.
Flat tops have a thickness of 30 mm. The raised front edge has a thickness of 
37 mm. 
Slate 
Slate bench tops are used almost exclusively for balance benches. High quality 
tops are of Welsh Blue Slate with polished edges and thickness of 25 mm or 
30 mm. 
Glass 
These tops are manufactured from a core of block board, covered on both sides 
with white melamine laminate and veneered with 6 mm thick glass. The 
toughened glass top surface may be acid etched to give a matt finish. These 
tops are usually available either flat with a front plastic edging strip in a 
cumulus green colour or dished with a plastic edging profile. All joints are 
sealed with silicon rubber sealant. 
Linoleum 
These tops are manufactured from a core of block board, edged with an insert of 
heavy-duty linoleum. The tops have a thickness of 25 mm or 30 mm. 
Chemical resistance chart 
Table 9.1 (see page 324) shows the chemical resistance, at specific concentrations, 
of the materials used for bench top surfaces and fume cupboard liners. 
Note that it is only intended to indicate the possible effect of the more 
commonly used acids, reagents, stains and solvent. It is not intended as a 
fully comprehensive guide. 
9.3.3 Service spine systems 
A wide variety of service spines are available, ranging from conventional box 
spines through to different types of flexible multi-service spines and modules to 
suit specific applications. 
Bench mounted box service spines 
Commonly manufactured from melamine faced board with all exposed edges 
veneered in polypropylene or similar. All electrical outlets are mounted onto the 
vertical front fascia while mechanical services and drip wastes (if required) are 
positioned on the top fascia. Where necessary, reagent shelves can also be fitted 
to these spines.
Floor mounted box service spines 
Floor mounted service spines offer the advantage of flexibility; loose benching 
may be positioned up to them and not necessarily attached. Also, all services 
can be installed and tested prior to final bench installation. 
All spines are supported from an angle iron framework, which accommodates 
the mechanical service pipework, electrical conduit and cladding panels. 
Flexible, multi-service spine system 
This is a pre-fabricated self-supporting spine. Consisting of a metal section, 
with adjustable feet, it can accommodate and support a number of different 
mechanical services and waste lines. 
For maximum flexibility, a capping strip may be fitted at bench level. 
Alternatively, where flexibility is of minimal concern, the work surface can be 
taken flush to the spine. 
Situated above the work surface is the mechanical service strip, generally 
made of solid grade laminate, which can either be of a closed type — the strip 
being taken down to worktop level — or of an open module design which 
allows a gap above the worktop. 
Trunking for electrical outlets is usually above the mechanical services. 
Over this, trunking may be height adjustable reagent shelves specified in glass, 
melamine laminate or solid grade laminate. The reagent shelf support may also 
incorporate scaffold supports suitable for small diameter rods. 
Designed in modular lengths to suit most applications, all services are preinstalled 
in the factory enabling pressure testing to be undertaken before 
dispatch. Thus, on-site installation time is minimized because it is only 
necessary to make the joints at the module ends. 
Compact, multi-service module 
A multi-service module allows for a high-density distribution of mechanical 
and electrical service outlets. 
This type of module is suitable for use as a service bollard with either table 
frames or mobile trolleys placed against it (an ideal situation for analytical 
instrumentation) or, alternatively, mounted above a wall bench to provide a high 
density of outlets in a limited space. 
A further use of this module is to site it between two fume cupboards. This 
enables services to be supplied to both cupboards from a single source and 
obviates the need for service outlets to be sited in the fume cupboard itself. 
Typically, the module is fabricated from moulded sections with a lower 
section accommodating the mechanical services and the top section housing the 
electrical outlets. Intermediate sections can be added to accommodate
additional mechanical services or outlets for clean instrument gases. The 
service feed pipes for these modules can either be sited overhead or below. 
Suitable cladding panels may be used to conceal these service pipes. 
Overhead service boom 
The use of the overhead service boom, in conjunction with mobile tables, 
ensures that maximum flexibility is achieved in laboratory benching layout. 
When used with standard benching, all services are supplied from the boom, 
leaving the work surface completely free for apparatus and instrumentation. 
Booms are available single-sided for wall benches and double-sided for 
island/peninsular benches. 
Boom frames are constructed from metal sections to accommodate the 
mechanical service outlets with electrical trunking above for 13 amp electrical 
outlets. Solids grade laminate panels are fitted as a closure to the bottom of the 
boom. Double-sided booms may be fitted with guardrails at the bottom. 
The units are suspended from the soffit on uprights fitted with mounting 
plates. Services are supplied to the boom from overhead and may be enclosed in 
a dropper box. 
9.3.4 Balance and instrument benches 
Balance benches 
These benches are specially designed to support analytical balances and other 
sensitive instruments. 
Benches are usually constructed from heavy-gauge steel sections and fitted 
with adjustable feet. The framing supports, via anti-vibration pads, an antivibration 
work surface consisting of a heavy, thick terrazzo plate. The whole 
metal structure is often clad in a separate melamine veneered enclosure to give 
additional protection. 
Instrument benches 
These are compact benches specifically designed to house analytical instruments 
together with associated computer and printer equipment. 
Benches are based on mobile trolleys fitted with two fixed and two lockable 
castors. Uprights are fitted to the back of the bench to accommodate the cable 
store, removable cladding panels, electrical and mechanical services, shelf and 
swivel monitor stand. 
Typically, a melamine laminate worktop is included, under which may be 
housed additional units, fitted with either cupboards or drawers or a pull-out 
writing flap or pull-out shelves.
Electrical and mechanical services (such as instrument gases) are connected 
to the bench from socket and service outlets on adjacent benches via flexible 
cables and service pipes. 
9.3.5 Tables and trolleys 
Tables 
Two types of table frame are available: the ' C frame support and the 'H' frame 
support. Both types are available in various lengths, depths and heights, or in 
continuous runs to suit specific applications. 
The 1C frame table support 
This is normally manufactured from rectangular steel with connecting rails. 
The cantilever support is fitted with adjustable feet for levelling. Tables are 
usually fitted with melamine laminate worktops or with other materials. 
According to availability 'C frame support tables are designed to carry a 
limited load. 
The 'H' frame table support 
The leg frames are usually manufactured from rectangular steel sections, are of 
welded construction and are fitted with levelling feet. Longitudinal rails are also 
steel section. Tables are commonly supplied with melamine laminate tops but 
any other materials may be specified. 
Trolleys 
These trolleys are typically manufactured as for 'H' frame tables but are fitted 
with double-wheel castors equipped with rubber tyres, one diagonally opposed 
pair of castors being lockable. These trolleys are fitted with melamine laminate 
worktops and shelf, and have a good load carrying capacity. Other worktop 
materials are always an option. 
9.4 Fume cupboards 
When considering the layout of a laboratory, the design and positioning of fume 
cupboards is of critical importance. Poor design or bad positioning of a fume 
cupboard is not only a safety hazard, but it can detract from the working 
environment (see Section 9.2 on planning a laboratory).
9.4.1 Typical fume cupboard construction 
Support system 
Fume cupboards can be supported on pedestal unit furniture, cantilever 'C 
frames or table frames with suspended or movable units of furniture. Frames are 
usually of epoxy powder coated rolled hollow section (RHS) mild steel. 
Carcass materials 
Mild steel frame sections are commonly used to support external panels 
of epoxy powder coated steel or compensated laminate-faced mediumdensity 
fibre board. (In compensate laminate a balancing laminate is applied 
to the hidden inside face to prevent exposed facing laminate distorting the 
board). 
Top cover access panels 
Designed to be easily demountable, top cover panels may be either epoxycoated 
steel or laminate finished board to match fume cupboard outer panels. 
Basic internal construction 
The back panel is constructed from solid grade laminate, whereas the side and 
top panels are melamine-veneered boards. Generally the top panel has a cut-out 
fitted with laminated safety glass, complete with a removable light cowl and 
light tube. Explosion flaps may also be fitted in the top panel. 
A back baffle of solid grade laminate is specifically designed to give an even 
face velocity. It should include slots to ensure good scavenging at the sides and 
at the back corners of the cupboard. Scaffold points may be fitted to the back 
baffle. 
Sash design 
The vertically sliding sash is commonly made of toughened or laminated safety 
glass in a metal frame with profiles finger pull to improve airflow characteristics 
at the lower edge. Suspension is usually by stainless steel cables and lead 
counter balance weight, the cables running over ball raced nylon pulleys, all 
arranged on a fail-safe principle in the event of cable failure. Sashes may 
include horizontal sliding side sashes within the vertical sash frame or 
horizontally split sashes used where a limited room height restricts normal 
sash operation.
Airflow 
Either a by-pass is fitted above the sash to reduce the face velocity at the lower 
sash openings and to give a constant extract volume, or a microswitch is fitted 
to signal the extract system to reduce the extract volume by way of an actuated 
damper or variable speed fan motor. A profiled metal sill fitted at the front of the 
work surface ensures good low-level extraction. 
The top of the cupboard should be fitted with an aerodynamically designed 
take-off manifold of fire resistant polypropylene, or similar, ready for connection 
to the extract system. The manifold should include a condensate collar and, if 
necessary, a condensate drain. 
Utilities 
Service outlets are fitted on the centre back wall or the side-walls of the 
cupboard with control valves fitted into a front fascia rail which also 
accommodates the electrical outlets. Alternatively, controls may be located 
on each side of the fume cupboard. Refer to Section 9.6 for details of the 
services available and distribution systems. 
9.4.2 Fume cupboard liner and baffle materials 
There is no single, practical construction material for fume cupboard liners that 
is suitable for all reagents. A comprehensive range of construction materials is 
available, with each suited to the specific use to which the cupboard is to be put. 
See Table 9.1 on page 324 for material selection guide. 
Liners and back baffle materials 
• melamine: veneered high-density board; 
• duraline: modified resin and fibreglass filled sheet; 
• solid grade laminate; 
• polypropylene; 
• PVC; 
• stainless steel — Grade 316, natural finish; 
• toughened glass with backing. 
Melamine veneered high-density board 
This is highly suitable for use as a construction material for side and top panel. 
Careful consideration must be given to the detail design and construction of the 
cupboard to ensure that exposed sides or ends do not come into contact with 
fumes.
This material is only suitable for general-purpose fume cupboards. It is not 
suitable for use with perchloric acid, radio-isotopes or cupboards which have 
heavy duty acid use, i.e., metallurgical digestion cupboards or those fitted with 
a water wash facility. 
Duraline 
A cost-effective, modified resin and fibreglass filled sheet designed to have 
good flame retardance, mechanical strength and chemical resistance. 
Solid grade laminate 
This can be used either for the construction panels of the cupboard, utilizing a 
thick board, or for the lining panels and back baffles, requiring a reduced 
thickness board. 
This material is very suitable for general-purpose fume cupboards and for 
cupboards used in low-level radio-isotope applications. It is not suitable for 
perchloric or heavy acid use. 
Plastic 
Polypropylene or PVC liners and back baffles are typically fabricated from 
16 mm thick material. The plastic liners are excellent for fume cupboards used 
predominantly for heavy acid applications. Some solvents will cause the plastic 
to soften. However, once the solvent has evaporated, the plastic will usually 
appear unaffected. The disadvantage of these liners is their relatively low 
temperature tolerance. PVC softens at 600C and polypropylene at 900C. 
If electric hot plates are used in fume cupboards with these types of liner, the 
power supply should only be energized once the extract fan is switched on. If 
gas hot plates are used, a solenoid should be fitted in the supply line to inhibit 
the use of these hot plates when the extract fan is switched off. The fan 
minimizes the effect of radiant heat on the plastic liners. 
Stainless steel 
Stainless steel liners are manufactured from acid resistant (Grade 316) stainless 
steel. They are normally available as either fabricated sectional liners with 
joints sealed with silicon rubber or one-piece liners and worktop with all 
corners radiused for ease of cleaning. Care should be taken in selecting this 
material for specific applications as stainless steel is, to some degree, affected 
by acids (see Table 9.1 on page 324). 
When used for acid applications i.e. perchloric acid including Kjedahl 
digestion, these fume cupboards should be fitted with water washing jets to 
enable washing away of any condensed acids after a series of experiments.
Stainless steel fabricated liners are suitable for use in low-level radio-isotope 
applications. For higher-level use, one-piece liners should be specified. 
Epoxy resin liners 
Solid epoxy resin liners are generally fabricated from 6 mm thick epoxy resin 
sheets. All joints are sealed using epoxy resin grout. These liners are suitable 
for general-purpose fume cupboard use and for high acid use. Some staining 
may occur when they are used for concentrated acid applications, although the 
base material normally remains unaffected. Some solvents may also affect this 
material (see Table 9.1 on page 324). 
9.4.3 Fume cupboard work surface materials 
Fume cupboard work surfaces may be selected from the higher specification 
range of bench top materials where chemical resistance and the ability to 
provide an integral raised rim are important selection criteria. 
Work surfaces 
• solid epoxy resin; 
• solid grade laminate; 
• stainless steel — either heavy gauge with reinforcing on underside or light 
gauge with all edges turned over and under and bonded to a WBP plywood 
base; 
• quarry tiles — on WBP plywood base, bedded and pointed with acid 
resistant cement; 
• polypropylene — bonded to a WBP plywood base. 
It is advisable to incorporate raised edges to work surfaces to contain spillage. 
9.4.4 Fume cupboards for specific purposes 
Fume cupboards for use with some specific reagents or for certain types of 
analysis require special consideration. Detailed below are cupboards designed 
to meet some of the more common of these applications. 
Fume cupboards used for Kjeldahl digestion 
Due to the problems of both heat and condensed acid, either stainless steel or 
polypropylene liners should be used. Ideally, the necks of the Kjedahl digestion 
flasks should be manifolded together to enable the majority of the acid fumes to 
be extracted via a water vacuum pump — the fume cupboard only being used 
as a secondary containment device. Alternatively, a proprietary digestion 
apparatus, incorporating its own heater and local extraction may be used.
Polypropylene liners give the best chemical resistance and are quite acceptable 
if electric heating mantles are used. However, if Bunsen burners are used, care 
must be taken not to overheat or burn these liners. It is good practice to have a 
solenoid valve in the gas supply line energized by the extract fan motor. This 
inhibits the use of the gas burners without the extractor fan switched on. 
For both liner materials, it is desirable to fit a water wash device in the 
cupboard to facilitate washing down after a series of digestions. 
Fume cupboards for use with perchloric acid 
Fume cupboards designed for this use should be fitted with either stainless steel 
or polypropylene liners. When stainless steel liners are used, there can be a 
certain amount of acid attack on this material; however, the by-products of this 
corrosion are safe and their presence can be minimized by the frequent use of 
the water wash system. The disadvantage of propylene liners is that when 
perchloric acid is used, it is normally heated and the heat generated can cause 
distortion of the plastic liner. Therefore, care must be taken to ensure that the 
heat source is not placed too near the sidewalls or back baffle. 
Due to the possibility of explosive perchlorates being formed by the 
condensed acids, the fume cupboards and associated duct work should be 
fitted with water wash jets to enable the system to be washed down after a series 
of experiments. 
Consideration should also be given to fitting a fume scrubber immediately 
adjacent to the cupboards before the main fume extraction ductwork so that any 
condensed acid can be washed out. If this is done, then the ductwork after the 
scrubber will not need to be fitted with the water wash jets. 
Fume cupboards for use with hydrofluoric acid 
If significant quantities of hydrofluoric acid are to be used (and evaporated), the 
fume cupboard should be fitted with polypropylene liners. The cupboard 
should also be fitted with either a water wash system to enable washing 
down of any condensed acids after a set of experiments or with easily 
removable baffles to enable manual washing of the inside of the cupboard. 
Additionally, the extract system should be fitted with a fume scrubber, either 
adjacent to the cupboard or on the roof before the extract fan, to inhibit 
fluorides being emitted into the atmosphere. It should also be remembered that 
because of etching, the sash should be made of plastic i.e., clear PVC or 
polycarbonate, rather than glass.
Fume cupboards for use with radio-isotopes 
When considering fume cupboards for radio-isotope work, several factors 
which affect design need to be taken into account. These include the isotope's 
level of activity, its half-life, the need for filtration and the suitability of the 
cupboard's face velocity. 
If the cupboard is only to be used for tracer work, standard solid grade liners 
with a face velocity of 0.5ms"1 may be suitable. For dilution work or high 
levels of activity, the fume cupboard may need a one piece welded liner of 
stainless steel together with an extract system fitted with high efficiency 
particulate air (HEPA) filters. Carbon filters may be required for some work. 
9.4.5 Special design fume cupboards 
Low-level fume cupboards 
Low-level (distillation) fume cupboards allow work requiring tall items of 
equipment to be carried out. The sash opens to the full height of around 
1800 mm. 
Normally two proportionally opening sashes are fitted. Both are interconnected 
and operate on a fail-safe principle. Services are supplied to the 
cupboard from either an adjacent multi-service module or from a service 
fascia strip built into the underbench unit. 
Walk-in fume cupboards 
Walk-in fume cupboards provide an especially large workspace with a clear 
inside height of around 2100 mm and cupboards that are usually fitted with two 
independently movable front sashes. Sashes are steel-framed with the upper one 
often being fitted with two horizontal sliding sashes. 
Frequently the standard cupboard sides are fitted with access ports with top 
hung flaps to allow cables and hoses to be passed through from adjacent multiservice 
modules. Alternatively, front fascia panels are fitted to house the 
mechanical and electrical controls with the mechanical outlets fitted to the 
sidewalls of the cupboard. 
Special application fume cupboards 
This fume cupboard is specifically designed for heavy duty, aggressive 
chemical use, such as for acid digestions where the significant amounts of 
condensed acids produced could affect the life of conventional cupboards. 
Ideally, the cupboard is fitted with a two-piece, angled back baffle that is 
easily removable to allow decontamination and cleaning of the whole interior of 
the cupboard. The baffle is designed to give one third of the total extract volume
extracted from the lower baffle opening and two thirds of the extract volume 
extracted through the top baffle opening. The internal configuration of the 
cupboard combined with the baffle openings ensures that fumes generated 
within the cupboard are first directed towards the lower baffle opening, then the 
fumes migrate up and adjacent to the back baffle and are extracted via the top 
baffle opening. The baffle is fitted with a condensate trough at the bottom with 
connection to drain. 
While the carcass and sash construction of this cupboard is generally the 
same as basic models, the special application cupboard and its back baffle 
should be lined with approximately 5 mm thick ceramic, the top panel of solid 
grade laminate and the sash of laminated safety glass. For hydrofluoric acid use, 
the cupboards need to be lined with polypropylene and fitted with an 
aluminium back baffle that is polyamide-coated. Polycarbonate is recommended 
for the sash. 
The special application fume cupboard should be specifically designed to 
accommodate a scrubber/demister unit for the removal of contamination from 
the extract air system before discharge into the atmosphere, especially important 
where perchloric or hydrofluoric acids are used. 
9.5 Extraction hoods 
Local bench extraction hoods 
For many types of operation, where only small amounts of noxious fumes 
(smoke, vapour, gases) or occasional high temperatures are generated, local 
extraction at source is ideal. 
A local bench extraction hood uses laboratory supply air to produce a coneshaped 
vortex within its confines to capture any noxious substances and extract 
them efficiently and quickly. 
Hoods may be fabricated from PVC or epoxy coated steel or stainless steel. 
It is important that the maximum height of a hood, above the source of 
emissions, should not be greater than its diameter. 
A variable speed axial flow fan for supplying air from the laboratory may be 
mounted at the back of the casing, or the hood may be ducted to a central extract 
system. 
Drop front steel extract hoods 
Hoods are generally fabricated from steel and finished in epoxy powder coated 
paint. They normally have a vertically adjustable front cowl and are suitable for 
extracting radiant heat from ovens, muffle furnaces etc.
Chromatography spray hoods 
These hoods are usually fabricated in PVC and are specifically designed for the 
spraying of chromatography plates. They are fitted with a louvred back baffle to 
give good extraction, and also with chromatography plate holders. 
The hood is suitable for wall mounting or can be fitted at the rear corner of a 
fume cupboard. It is advisable for the extract duct to be flexible enabling the 
hood to be lowered during use or pushed up out of the working area of the fume 
cupboard when not in use. 
Fume hoods 
Fume hoods are available fabricated from epoxy powder coated steel, aluminium, 
PVC or polypropylene. They are available in a wide variety of styles to 
suit individual requirements. These hoods may be fitted with internal baffles to 
produce a high velocity peripheral extraction in order to improve containment. 
They may also be fitted with side and back panels. 
9.6 Utility services 
Services may range from simple installations, requiring just hot and cold water, 
drainage and possibly natural gas, to more sophisticated installations which use 
high-quality instrument gases. 
Service pipework may be carried out both in-factory, using pre-plumbed 
service spines or by traditional plumbing methods with the pipework being 
battened to the wall or clipped to the furniture units. 
9.7 Fume extraction 
One of the most important areas of laboratory design is in the design and 
engineering of fume cupboard extract systems. No matter how good the design 
of the fume cupboard itself, safe containment remains critically reliant on the 
performance of the extract system. Not only must the system achieve the 
correct volume flow required for a particular cupboard or cupboards, consideration 
must also be given to noise, condensate drainage and to ensuring that 
ductwork does not contravene fire regulations. 
Design criteria 
Extract systems should be designed to provide a maximum duct velocity of 
5-6HiS"1. This velocity is sufficient to ensure good scavenging of the duct in 
order to inhibit any build up of contamination within the duct, whilst not being
high enough to generate undue air noise within the ducting system. Generous 
radius moulded bends are recommended in all systems up to 600 mm diameter. 
Rectangular ductwork, and circular ductwork above 600 mm diameter may 
have fabricated bends. 
Careful consideration should be given to the routing of all ductwork, so that 
it is taken outside the building, or to a firebreak service void, by the most direct 
route. Horizontal ductwork is to be minimized; where long runs are necessary, 
they are to be laid to a fall with a condensate drain at the lowest point. All 
extract systems, whether they serve a single cupboard or several cupboards (in 
which case a manifolded system may be used subject to safety criteria), require 
volume control dampers (butterfly type) to be fitted for system balancing. 
Normally all joints in ductwork are solvent welded socket and spigot type. If 
required, flanged ductwork, with Neoprene gaskets may also be specified for 
particular applications. 
Materials of construction 
The most commonly used ductwork material is UPVC, which is suitable for 
most applications. Where necessary, due to fire regulations, this ductwork can 
be GRP-coated to give 30-60 minutes' fire resistance, negating the requirement 
for fire dampers that introduce an additional safety hazard. For very specific 
applications stainless steel or galvanized steel ductwork is available. 
Fume extraction fans 
Fume extraction fans are fabricated from either UPVC or polypropylene. Fans 
should be generously sized to enable the impeller speed to be kept to a 
minimum for quiet operation. Flexible sleeves are recommended to isolate the 
fan for connection to ductwork. 
Multi-vane forward curved blade type impellers provide maximum efficiency. 
They may be either directly driven or with indirect drive via 'V belts 
and pulleys. 
Motors with either single or three phase supply are available depending 
mainly on the load. Motors should be suitable for external use, as most 
installations find fans mounted on the roof. 
Typically a fan unit is mounted on a galvanized steel angle frame complete 
with anti-vibration mounts.
9.7.1 Specialized ancillary equipment 
Fire dampers 
In those situations where it is sometimes necessary for ductwork to pass 
through firebreak walls or into general purpose building service ducts, it may be 
necessary to fit fire dampers. Fire dampers must provide the same corrosion 
resistance as the ductwork. Consequently fire dampers are usually fabricated 
from a stainless steel outer casing fitted with a stainless steel folding curtain 
shutter. 
The shutter is fitted with stainless steel constant tension closure springs and 
is held open by a fusible link which releases the shutter in the case of fire. The 
fire damper is fitted into the partition wall and access hatches are provided in the 
ductwork for maintenance and testing. Owing to complex routing requirements 
or simply the sheer quantity of individual ducts, the configuration of fire 
dampers often makes it impossible to provide accessible access hatches. In 
these cases, motorized dampers provide an acceptable alternative. 
Water wash systems 
For some applications, such as extract systems handling perchloric acid, it is 
necessary to fit a water wash system. Spray jets are fitted into the ductwork, 
spaced approximately at a 1.5 metre pitch on vertical ductwork and at a 1 metre 
pitch for horizontal runs. Jets may be manifolded by a plastic supply pipe and 
controlled from a valve on the fume cupboard. 
It is important to note that a water wash system should only be used for 
washing the ductwork and removing any condensed acids after a series of 
experiments. It should not be used during a series of experiments as the spray 
will contaminate the experimental work. A booster pump may be required if the 
head of water is not sufficient for the higher jets to operate satisfactorily. 
Fume scrubbers 
For those extract systems handling perchloric or hydrofluoric acid, fume 
scrubbers may well be needed. Two types of fume scrubber are generally 
available — the compact scrubber/demister unit and the tower scrubber. 
In the case of perchloric acid, the compact scrubber/demister unit is ideal as 
it can be fitted adjacent to the fume cupboard. Therefore, all ductwork from the 
outlet of the scrubber will remain uncontaminated and water washing will not 
be required. This scrubber can also be used for hydrofluoric acid applications. 
The compact scrubber/demister unit is only suitable for connection to single 
fume cupboards. Where larger volumes of extract air are to be handled, from 
several cupboards, then a tower scrubber must be used.
Fume scrubbers are normally fabricated from UPVC and, in the case of 
tower scrubbers, feature GRP reinforcement. All scrubbers comprise three 
sections: 
the holding or capacity tank for the scrubbing media; 
a packed scrubbing section fitted with wash jets; 
a demist section to remove the washing media before discharge into the 
extract system. 
The installation requires a circulating pump provided with a water supply 
with a ball valve fitted, together with drain connection. 
9.7.2 Air input systems 
A factor sometimes overlooked in fitting out a new laboratory is that fume 
cupboards extract a considerable volume of air from the laboratory area. In nonventilated 
laboratories without sealed windows, it may not be necessary to 
install an air input system if there are only a small number of fume cupboards, 
as approximately six to eight air changes per hour can be achieved within the 
laboratory by natural leakage. In modern laboratory blocks with well sealed 
windows or in those where there are large number of fume cupboards, 
consideration must be given to the installation of an air input system. In this 
instance, care must be taken in the siting of the actual input grilles so that 
turbulence at the fume cupboard face is minimized. As a general principle, no 
input grille should be within 1.5 m of the face of the fume cupboard. 
Ideally, the input grilles or slot diffusers should be on the opposite side of the 
laboratory to the cupboards in order to 'wash' the laboratory with clean air. The 
use of grilles or slot diffusers is suitable to achieve room air change rates of up 
to 20 per hour. If the air change rate is above this, then a perforated ceiling grid 
should be used. 
9.8 Air flow systems 
9.8.1 Air-handling for the laboratory 
Air management control systems, which when considered at the planning stage 
of a laboratory, provide economies in both capital investment and operational 
costs. Variable airflow reduces the entire air requirements which as a result 
enables the building ventilation system to be designed smaller, thereby 
reducing investment costs. Operational costs are minimized through continual 
adjustment of the air flow to meet the current working situation. The rate of all
supply and extract air may be computer controlled to optimize plant operation 
providing lower energy consumption and the opportunity to introduce diversity 
factors to reduce capital and operational cost. 
Construction and components 
(a) Airflow controller 
The airflow controller is a processor which monitors and regulates the volume 
of extract air depending on the position of the sash. Upper and lower nominal 
limits are established for the open and closed sash positions. For all other sash 
positions the air flow rate is determined as a linear function. 
A sensor constantly measures the air volume and adjusts a damper when 
variations occur until the present value is achieved. The sensor is placed in a 
bypass system to protect it from aggressive fumes. 
Most airflow controllers can be switched to different operational modes: 
normal operation, night operation (lower amount of air) as well as emergency 
operation (maximum amount of air with fully opened damper). They can also 
be provided with volt-free contacts for connection to a building management 
system (BMS). 
(b) Sash controller 
The sash controller is a processor responsible for closing the sash when no-one 
is standing in front of the fume cupboard. Continual controlling of the sash 
opening ensures an optimum working condition with maximum safety. Typically, 
a passive infrared detector senses the movement of a person in front of the 
fume cupboard. When the person moves away, out of range of the detector and 
following a pre-set time delay, the automatic sash closing function is initiated. 
(c) Manual volume control damper 
The manual damper maintains a constant pre-set air volume even under varying 
pressure conditions. Such regulation is found in permanent vented units with 
constant air volume (cabinets, vented underbench units). The required volume 
of air for these installations is a burden which has been taken into consideration 
in balancing the room air. 
Temporarily vented units, canopies or local extraction hoods etc. which are 
either switched on and operated at the full pre-set air volume or off, incorporate 
a damper which sends a signal to indicate its operational condition enabling this 
to be taken into account in the process of adjusting the air volume levels.
(d) Group controller 
The controller constantly receives on-line data on the current individual air 
requirements from all variable extracting units in the laboratory (fume 
cupboards, temporary running extracting units). It processes this data and 
sends a control signal in the form of a nominal electrical signal to the supply air 
damper which adjusts the volume of air. In this way the group controller acts as 
a link between the extract air dampers of the individual units and the supply air 
damper of the laboratory. 
Where available, the BMS may undertake the function of the group 
controller. 
(e) Supply air dampers 
The damper receives the control signal from the group controller and adjusts 
the supply for compliance with the applicable specifications for air volume and 
room pressure. 
(f) Supply air grilles 
Sufficient air grilles should be allowed for supply air to the laboratory without 
draughts. 
(g) Supply air and extract air ducting 
These ensure optimal guiding of air in the room. 
9.8.2 Air handling efficiency 
Within the modern laboratory the emphasis on safety has led to an increase in 
the number of fume cupboards, local extract hoods and ventilated cabinets. The 
resultant demand on air flow creates unrealistically high air change rates. The 
consequences of not addressing the problem could lead to: 
large air handling equipment; 
large ductwork; 
high energy costs; 
complex control systems. 
The first three are a product of the air volumes; safety requirements do not 
permit air recirculation, therefore, all treated air supplied to the laboratory is 
dumped. Complex controls are necessary to manage the diversity on air volume 
demand. Depending upon the number of extract units and the operational 
requirements, systems may incorporate multiple fan and damper arrangements
for both supply and extract air, monitored by probes and sensors. There are a 
number of methods which may be adopted to improve the efficiency of 
laboratory air flow. 
Fume cupboard face velocity control system and laboratory air input controls 
Significant savings can be achieved in running costs to heated or cooled air 
input to the laboratory. 
A number of fume cupboards may be served by a single extraction fan. 
Make up air supply is introduced to the laboratory via a standard Air Handling 
Unit (AHU) with heating/cooling coils. 
The important feature of this type of system is that the extract fan runs at full 
volume at all times. This ensures that the discharge velocity remains constant 
and thus the contaminated air is dumped. As the fume cupboard sashes are 
closed, the total volume extracted through the fume cupboards is reduced to 
only 15%. Hence, a fresh air bleed damper is built into the system which allows 
air to be taken from outside the building through the extract fan to make up the 
85% reduction and, thus, maintain the discharge velocity. The fresh air bleed 
damper can be operated by an adjustable weighted arm, or by an actuator 
controlled by the extract duct pressure. 
The air input system is required to provide make up air for the fume 
cupboards. The air is taken from outside the building and heated or cooled as 
required. For laboratories where the building fabric is not well sealed, in order 
to control the air input, a duct probe is used to produce a signal proportional to 
the extract volume. 
For laboratories where the building fabric is well sealed, it is possible to 
measure the differential pressure between the laboratory and an adjacent area 
with a stable pressure regime. This measurement can be used to produce an 
output signal to control the air input to balance the variable extract volume. 
A significant feature of this system is the possibility of applying a diversity 
factor to the air input and extract units. Typically installations may operate 
where the air input unit and the extract fan are sized to 50% of the maximum 
design volume of the total fume cupboards. 
This means that 50% of the fume cupboards can be open with the other 50% 
closed or all the fume cupboards can be half open, i.e., any combination of sash 
openings up to a total for all fume cupboards of 50% opening. Controls and 
alarms operate by measuring the face velocity on the fume cupboards. An alarm 
would be activated centrally and/or on individual fume cupboards if the total 
50% opening is exceeded. Application of a diversity factor with an integrated 
control and alarm system can result in very substantial cost savings.
Secutromb auxiliary air fume cupboards 
Conventional fume cupboards achieve their containment by extracting large 
volumes of heated laboratory air to provide a sufficiently high face velocity to 
contain fumes. This process can result in a high rate of room air change and 
heating or conditioning of this air is often expensive. Furthermore, should 
additional fume cupboards be required, it may not be possible to supply 
sufficient air necessary for efficient extraction to these cupboards. 
The Secutromb fume cupboard works on a completely new and novel 
principle which involves auxiliary air from outside the laboratory area being 
supplied to the cupboard. As a result of the configuration of the cupboard's air 
input plenum ducts and the positioning of the extract take-off ducts, two contra 
rotating vortices are formed. As the air in the centre of the vortex is moving 
faster than the air on the outside, a negative pressure is formed in the centre of 
the vortex and any fumes generated within the fume cupboard migrate to and 
into the vortex and are then extracted via the extract take-off ducts. 
In practice, this means that there is a vertical extract column at each side of 
the cupboard over its whole height. This ensures good scavenging of the 
cupboard and, very importantly, the concentration of fumes within the cupboard 
is very much lower than that found in conventional cupboards. Up to 70% of 
the total extract volume can be supplied as auxiliary air with only 30% needing 
to be extracted from the laboratory itself. 
The auxiliary air must be heated and be within 80C of the laboratory air. In 
air-conditioned laboratories, no cooling is necessary. Auxiliary air should, 
however, be filtered to ensure plenum gauzes do not become blocked as a result 
of atmospheric contamination. 
As an additional safety feature, an airflow controller may be incorporated. 
This microprocessor-controlled system monitors the rate of flow of supply/ 
auxiliary and extract air and controls their flow rates within pre-set limits. 
Airflow controllers have audible and visual alarms to warn in the event of either 
auxiliary air or extract system failure. 
9.9 Safety and containment 
Filters 
Fume cupboards used for radio-isotope applications normally require the 
extract system to be fitted with HEPA filters and, in some instances (for 
example, isotopes of iodine), carbon filters may also be required.
With minor modification and the addition of a pneumatically operated 
volume control damper, the unit can be used as a constant face velocity module 
(i.e. total extract volume is variable dependent on sash position). 
Maximum permitted leak concentration of test gas in accordance with DIN 12 924 
Front sash closed 0.2 ppm 
One-third open 0.5 ppm 
Fully open 0.8 ppm 
Sash lock/airflow failure alarm module 
A combined sash lock/airflow failure alarm module should be designed to 
satisfy statutory safety standards. Generally the unit would comprise three 
separate parts: 
• alarm airflow sensor; 
• Printed Circuit Board (PCB) assembly in an enclosure; 
• annuciator front fascia plate. 
The alarm airflow sensor is typically a hot wire anemometer device. A 
sensor uses two signal diodes, one of which is heated. The diode is cooled by 
ambient air passing over it, its signal then being compared with the second 
unheated diode, which acts as a comparator for variations in ambient air 
temperature. Velocity sensors are usually installed in the top panel of the fume 
cupboard and produce a stable signal, which represents the face velocity. 
Volt-free contacts may be included for remote monitoring and fume 
cupboard status and for fan stop-start relay. 
Typically the annuciator face-plate incorporates an analogue meter showing 
'Safe-Unsafe' face velocity with green and flashing red indicator lights. It may 
also feature an audible alarm with a mute button to show low face velocity. The 
face-plate can incorporate fan 'stop-start' buttons. A sash 'high' release button 
with red and green indicator lights is used where the sash is raised above its 
working height for setting up experimental apparatus in the cupboard.
10.1 Introduction 
Process development facilities and pilot plants are an integral part of research 
and development operations for all major pharmaceutical companies seeking to 
provide new products for the future. Their design, construction, commissioning 
and validation have their own special problems arising out of the individual 
company's traditional research methods, the class of compounds to be developed 
and the regulatory requirements. 
These facilities are frequently multi-purpose and/or multi-product and the 
processes used are constantly under development. The design requires a degree 
of 'crystal ball gazing' because future requirements usually need to be included 
in the specification. 
The full range of pharmaceutical processing needs to be covered by process 
development facilities and pilot plants from chemical synthesis to production of 
the active pharmaceutical ingredient, through physical manipulation to formulation, 
production of the final dosage form, filling and packaging. 
The problems for chemical synthesis facilities are often different to those for 
other facilities. In the case of chemical synthesis facilities, the large number 
of chemicals used exacerbates the difficulties. In the other facilities it is often 
the problems of cross-contamination and the variety of machines required that 
dominate. 
Pilot facilities for primary and secondary manufacture require a greater 
degree of flexibility for the reconfiguration of equipment compared with 
general production operations. This is easier for secondary operation than for 
primary as the reactors and other items of chemical apparatus are more difficult 
to reposition and link into each other. 
For secondary operation, although it is necessary to have a dedicated sterile 
unit, all other operations are usually self-contained. 
This chapter summarizes the main design requirements that are necessary in 
these facilities for development and small scale manufacture. The detailed 
P r o c e s s d e v e l o p m e n t 
f a c i l i t i e s a n d p i l o t 
p l a n t s 
ROY KENNEDY and KEITH PLUMB 1 0
requirements follow the principles for primary and secondary operation in 
earlier chapters. 
10.2 Primary and secondary processing 
The division between primary and secondary processing is to some extent 
arbitrary and different manufacturers place the dividing line at different places 
in the total manufacturing process. In general all the chemical stages up to and 
including the manufacture and purification of the active pharmaceutical 
ingredient are part of primary processing. In some cases, physical manipulation 
processes such as milling are also included. All the steps after purification 
(except in some cases milling) are usually included in secondary processing. 
The decision of where to place the dividing line is often based on: 
• the type of purification and physical manipulation processes that are 
required; 
• the chemicals used within the purification and physical manipulation 
processes; 
• the need for the primary process to stop at a point where meaningful samples 
can be taken; 
• the type of building and facilities available to the manufacturer. 
10.3 Process development 
There are a number of stages in the development of pharmaceutical products. 
These stages are driven by the regulatory process, which is summarized in 
Chapter 2. The initial research that involves searching for new chemical entities 
is usually carried out at the laboratory scale and is not discussed further in this 
chapter. 
Once a promising new chemical entity (NCE) has been discovered, tests will 
begin on the compound to confirm that it has the required activity, stability, and 
low toxicity. It will also be necessary at this pre-clinical stage to identify that the 
compound can be synthesized by a practical route and that it can be purified and 
formulated. 
Much of this pre-clinical trials activity will be carried in the laboratory but it 
may be necessary to carry out some work in a small-scale pilot facility. Once 
this work has been completed, the clinical trials themselves start. These are 
carried out in three stages with increasingly large quantities of material. For the 
first stage, a small-scale pilot facility (in the order of one-hundredth of
production scale) will usually be sufficient, whereas normal scale pilot facilities 
(one-tenth production scale) are usually required for stage three clinical trials. 
During process development the whole manufacturing process will need to 
be both scaled up and optimized. Initially, this will be carried out in the smallscale 
pilot plant, followed by the normal scale pilot plant. The final scale up 
work will be carried out using production scale equipment and this may take 
place some time after the product launch since it is often possible to produce 
launch quantities of material using the pilot plant. 
Process optimization needs to take place early in development because the 
regulatory authorities require the stage three clinical trials to be carried out 
using material produced by the same manufacturing process as is used for the 
full scale. 
Stage three clinical trials are usually carried out at one-tenth of production 
scale because the regulatory authorities expect the scale up from the development 
scale to the product scale to be no more than ten fold. 
10.3.1 Good manufacturing practice 
Once the clinical trials start it is necessary to produce all the material required 
based on the GMP requirements detailed in chapter 3. Although the regulations 
are directed primarily at the stages after chemical synthesis, the principles 
should be applied throughout the whole manufacturing process. 
Contamination by operating staff 
The operating staff in a pharmaceutical facility is likely to be the main source of 
product contamination. Body particles are continually shed as people move 
around. Microbiological contamination is always a problem and all stages of 
production, apart from the early stages of chemical synthesis, will require a 
hygiene regime. The operating staff can cause cross-contamination between 
different products and/or different intermediates by material spilt on their 
clothing. Clean clothing needs to be regularly supplied. 
Small-scale facilities involve a large number of manual operations, so 
contamination by the operating staff may be greater than on a production 
scale. For particularly sensitive products, this may require a high level of 
protective clothing for the operators and/or the use of laminar flow booths or 
glove-box isolators. 
Cross-contamination 
Cross-contamination between different products and/or different chemical 
intermediates is a major source of drug adulteration. Since small facilities 
can be used to make a large variety of products and/or intermediates, the
possibility of cross-contamination needs to be addressed at the design stage. 
Issues to be considered include: 
(a) Easy to clean equipment 
Small-scale chemical synthesis equipment is cleaned either manually, by using 
mobile cleaning rigs or by refluxing with solvents. Equipment will need to be 
accessible and may need to be disassembled to allow access. This is a design 
requirement. 
Mobile cleaning rigs using high-pressure hot water jets with or without the 
use of detergents can be useful for cleaning. However, these rigs can only be 
effective if all contaminated surfaces can be accessed. It can be used to clean 
large bore pipework. Careful consideration will need to be given to the possible 
safety hazards. Hot water can easily lead to scalding of operating personnel and 
some strong detergent solutions are particularly corrosive. 
Refluxing with solvents can be useful for reaction equipment fitted with a 
condenser. 
Other small-scale equipment will almost certainly need to be stripped down 
for cleaning. 
In all cases, it is important for the equipment to be constructed without 
crevices that hold chemical, particulate or microbiological contamination and 
to have a surface finish that is inherently easy to clean. Glass, electropolished 
stainless steel and PTFE are common easy to clean finishes. 
(b) Primary containment 
Primary containment is based on the actual equipment used to do the 
processing, for example, reaction vessels should have a closed top and a seal 
on the agitator. Full primary containment can be achieved when solids are 
charged via glove-box isolators and liquid connections are made via hygienic 
dry break connections. 
Primary containment minimizes the need for secondary containment and 
reduces the building standards required. 
(c) Secondary containment 
Secondary containment involves placing the manufacturing equipment in some 
form of ventilated enclosure resulting in a number of conflicting issues: 
segregating flameproof areas from safe areas; 
protecting the product from contamination by the operating staff and the 
environment; 
protecting the operating staff from highly active materials.
These three requirements can be resolved by the use of air locks, the correct 
pressurization routines and correct extraction and ventilation regimes. For more 
details see Chapters 6 and 8. 
Some enclosures are large enough for the operating staff to enter via air 
locks. In other cases the equipment can be enclosed in a down flow booth with 
an open front that allows the equipment to be operated with sufficient air 
velocity to protect both the operator and the material being produced. 
With the smallest scale equipment, it is possible to place the equipment 
inside an isolator with equipment being operated via gloves or a half suit. Such 
a system offers a high level of protection and is frequently used when the 
material being produced is highly potent or highly active. 
(d) Ventilation 
Ventilation can be used in laminar down flow or cross flow booths to protect the 
product from cross-contamination and the operating staff. Also an increase or 
decrease in pressure in different areas prevents the flow of air into or out of the 
room. 
Ventilation systems require careful design because they can be the cause of 
cross-contamination themselves, particularly if one ventilation system is used 
to serve more than one area. 
(e) Cross-contamination by the operating staff 
This is minimized by making efficient use of the containment system, by 
ventilation, providing clean clothing at regular intervals and appropriate 
changing facilities. 
Materials of construction 
The materials of construction for pharmaceutical equipment are covered by 
both European and American guidelines and regulations. 
Paragraph 3.39 of the European Guide to Good Manufacturing Practice 
states: 'Production equipment should not present any hazard to the products. 
The parts of the equipment in contact with the product must not be reactive, 
additive or absorptive to such an extent that it will affect the quality of the 
product and thus present any hazard.' 
Section 211.65 of the Code of Federal Regulations title 21 states: 'Equipment 
shall be constructed so that surfaces that contact components, in-process 
materials, or drug products shall not be reactive, additive, or absorptive so as to 
alter the safety, identity, strength, quality, or purity of the drug product beyond 
the official or other established requirements.'
To meet these regulations, it is necessary to specify materials of construction 
that are corrosion resistant, easy to clean, do not release material into the 
process by leaching of the material or absorb any of the process materials. The 
most commonly used materials are glass and stainless steel. These are corrosion 
resistant and easy to clean if constructed correctly. However, it is possible for 
glass to absorb ions from some chemicals, which can lead to cross-contamination, 
and materials can be leached out of stainless steel unless the surface is correctly 
treated. 
Polymeric and elastomeric materials need to be chosen to have the widest 
range of chemical resistance as well as being able to withstand the range of 
temperatures likely to be encountered. These materials are particularly problematical 
with respect to chemicals being leached out of them because they 
generally include a range of plasticizers to improve their stability or flexibility. 
Whatever materials are used, documentation will be required to demonstrate 
that the specified material has been installed. 
Surface finishes 
To ensure that equipment is easy to clean, liquids drain easily and solids do not 
adhere to walls, it is necessary to consider the surface finish of both the inside 
and outside of equipment. 
Metals normally need some form of treatment such as polishing. Mechanical 
polishing of metals requires the use of grits that are held together by soaps and 
grease and these can become embedded in the surface and lead to product 
contamination. At the small scale this problem is best overcome by having the 
metal surface electro-polished. Mechanical polishing is suitable for the external 
surface of metal. 
Non-metallic materials such as glass and PTFE have an inherent smooth and 
easy to clean finish both on the outside and the inside. However, other nonmetallic 
materials are much less smooth. Lining these with PTFE may be 
required. An external finish will need to take into account the likelihood of 
damage due to manhandling the equipment. 
Material storage and handling 
Systems to ensure that intermediates and products are not confused are of 
fundamental importance to good manufacturing practice. At the small scale 
there is likely to be a large number of materials to be stored, made by different 
processes. The storage handling system must be able to prevent different 
materials from being incorrectly identified and must prevent the same material 
made by differing processes from being mixed.
10.4 Small-scale pilot facilities 
10.4.1 Chemical synthesis - primary manufacture 
Reaction equipment 
Small-scale pilot facilities with capacities ranging from 20-100 litres are 
generally required for the chemical synthesis stage of the manufacturing 
process. To provide a high level of corrosion resistance such facilities usually 
use glass equipment that can be configured for the particular processes taking 
place. In some cases the whole rig is built from scratch and then dissembled 
when it is not required. This type of rig is often called a 'kilo-lab'. 
Solids handling equipment 
Simple filters, centrifuges and dryers will be required since most pharmaceutical 
intermediates are solids. The solids handling equipment will be corrosion 
resistant and mobile so that it can be connected to the reaction equipment. 
Depending on the quantities of material used, stage three clinical trials 
material may be produced by this equipment, which will replicate the type of 
equipment that will be used on a production scale and have a similar modus 
operandi. This solids handling equipment can be hired from equipment 
vendors. 
Small-scale solids handling equipment suffers from the problem that much 
of the product may be held up in the equipment and consequently the yield is 
very low. 
Multi-purpose equipment 
Some specialist vendors provide multi-purpose equipment that can be used for 
reactions, filtration and drying. These units have the advantage that they reduce 
handling and, thus, reduce the exposure of the operators to the chemicals. 
However, as this equipment is complex it is usually a compromise and the result 
is not cost effective and less than optimal for each unit operation. 
Solvents 
Most chemical syntheses use flammable solvents which means that the smallscale 
facility will need to be a flameproof area. Since these facilities are often 
located within laboratory complexes it is necessary to separate flameproof areas 
from safe areas. This can be achieved by the use of pressurized air locks and in 
some cases the pressurization of the safe areas.
Small-scale facilities make use of a large number of solvents usually handled 
in drums requiring a flameproof drum handling and storage area, outside the 
building to reduce ventilation needs. A method of safely transferring the 
solvents from the drums to the manufacturing equipment is required. One 
method involves moving the drums from the drum store to a dispensary area, 
where the required quantity is decanted into a safe solvent container that is used 
to transfer the solvent to the reaction area. In other cases intermediate 
containers may be used to transfer the liquid from the drum store to the 
dispensary. 
Since a large number of solvents are used in small-scale facilities it is 
unusual to find solvent recovery facilities included in the area, unless one or 
more solvents used in larger quantities can be recovered using the equipment 
used for the chemical processes. The recovery of solvents prevents crosscontamination 
and enables them to be disposed of safely. 
Toxicity 
Although the final drug product manufactured may have a low potency, the 
chemical intermediates that are made during the synthesis of the active 
pharmaceutical ingredient are often highly potent. The design of the facility 
must ensure that the operating personnel are protected. This in part may be 
covered by the building design, but also it will require the use of fume 
cupboards, local extract ventilation, glove-boxes, rapid transfer ports, 
contained transfer couplings and air suits. 
Environmental considerations 
The chemical synthesis route of many pharmaceuticals is highly complex (see 
Figure 1.1, page 3). In many cases more than 20 intermediates are made before 
the active pharmaceutical ingredient is prepared. Even if every stage has a high 
yield the overall yield can be very low. This means that facilities must be 
provided for all the waste streams to be handled. 
The large variety of chemicals produced in low volumes usually precludes 
the use of an on-site effluent treatment plant for handling all the waste streams. 
Liquids and solids must be put into groups that can be mixed together for 
disposal; for example, halogenated solvents will need to be separated from nonhalogenated 
solvents. 
Depending on the quantities involved and their toxicity, vapour and gaseous 
emissions will be treated. Vapours can often be condensed using a low 
temperature system — the use of a liquid nitrogen cooling system is economical 
at the small scale. Solvent, acid or alkali scrubbing systems may be
required for the gaseous emissions. The choice of equipment will depend on the 
chemicals used and the flexibility required. 
10.4.2 Physical manipulation 
Physical manipulation is a process not involving a chemical reaction that 
changes the purity of the material. It usually involves crystallization, filtration, 
chromatography, milling, drying or blending for example. This type of process 
is frequently required to achieve one or more of the following requirements: 
• crystal morphology; 
• moisture content; 
• specific particle size; 
• particle surface physico-chemistry. 
Depending on the product, the equipment for crystallization, filtration and 
centrifugation may be the same equipment as is used in the chemical synthetic 
process, and so most of the comments made in Section 10.4.1 are relevant. 
However, other equipment is used to carry out a particular operation, such as 
milling, micronization or granulation. 
To achieve maximum flexibility this equipment needs to be mobile. In some 
cases developers may hire this equipment from the vendor when it is required. 
Many organic solids are explosive when finely divided and require explosion 
protection and it is likely that the most appropriate method will be to use inert 
gas blanketing. 
10.4.3 Manufacturing the final dosage form - secondary manufacture 
The first stage of the manufacturing process is formulation. This is the process 
of adding the drug(s) to one or more excipients (see Chapter 6 for more details 
of excipients) to provide the correct mixture for the final dosage form. These 
may be solids or liquids depending on the final dosage form. 
Liquids, gels, creams and syrups 
If the final dosage form is a liquid, gel, cream or syrup then the equipment used 
for chemical synthesis may be suitable for the required blending operation. 
However, some formulations such as those required for aerosols require 
specialist formulation equipment because the propellants used are pressurized 
liquids with vapour pressures in the region of 3 to 4 bar g. 
Conversely it may be advantageous to use equipment located close to the 
filling equipment, which may require dedicated formulation equipment, so that 
it is possible to run the formulated product directly to the filling machine. If this 
is not then the product would be transferred into one or more intermediate
vessels and moved to the filling area. Rapid transfer between formulation and 
filling is a particular requirement with terminally sterilized products, as these 
must be formulated, filled and sterilized within 24 hours. 
The choice of whether to use equipment directly connected to the filling 
equipment is determined by the nature of the product and overall facilities 
available to a company. 
Solids 
When the final dosage form is a tablet or pellet a solids mixing system is 
required. Small specialist solids mixing equipment is usually provided for 
formulation. At the smallest scale this equipment may be hand operated and 
similar to a modern version of a pestle and mortar. 
Solid dosage forms, such as tablets, capsules, suppositories and solid dose 
inhalers require a second manufacturing stage beyond formulation for their 
production. 
The machinery required is highly specialized and designed to carry out a 
particular task. Whilst hand operated bench scale equipment exists, this is 
usually only used to test the formulation and demonstrate that the required final 
dosage form can be produced. Such equipment is suitable for use at the preclinical 
stage. 
Once material is produced for clinical trials, small-scale automatic machines 
is required. Since these materials are designed to make specific final dosage 
forms, pharmaceutical companies often specialize in a small range of dosage 
forms. This reduces the number of machines required. 
10.4.4 Filling 
Filling is the process of putting the finished pharmaceutical product into its 
primary container, which may be a bottle, vial, ampoule, tube, aerosol can, or 
blister pack. 
In the early stages of clinical trials, automated filling machines may not be 
used for tablets and capsules as these can be filled and packed by hand. 
Suppositories are filled by machine as they are easily damaged, and solid 
dose inhalers will almost certainly be filled by machine due to their complexity. 
However, they will only be simple semi-automatic machines. 
For liquid products, the filling operation produces the final dosage form. 
When only small quantities of these are required hand operated bench scale 
machines may be used, larger quantities will require automatic machines. These 
are specialist machines and, as with solid dose machines, companies tend to 
specialize in a few dosage forms.
With liquid filling it is usual to connect the formulation equipment to the 
filling machine so that the liquid can be transferred directly. In production 
facilities it is common to have completely integrated filling lines with filling, 
check weighing and washing connected together. At the smaller scale flexibility 
can be increased by keeping the individual machines separate and manually 
moving the filled packs from one unit to another. 
10.4.5 Packing 
Most pharmaceutical products are sold in some form of secondary packaging. 
This gives protection to the primary packaging and allows detailed instructions 
to be included with the product. Packing is the process of putting the product 
already in its primary packaging into its secondary packaging. 
At the early stages of clinical trials, this can be carried out by hand. 
However, once the required quantities increase to more than a 1000 containers, 
a semi-automatic packing machine is usually necessary. If it is expected that the 
product will be packed by machine during the production process, then the 
chosen pack(s) will need to be tested on the packing machine during clinical 
trials to prevent a delay to the product launch. 
There are several stages to packing: 
• labelling the primary packaging; 
• putting the primary packaging and instructions into the secondary 
packaging; 
• printing lot specific information on the secondary packaging; 
• fastening a tamper evident label to the secondary packaging; 
• over-wrapping the secondary packages into collated parcels; 
• packing the over-wrapped parcels into cases. 
Maximum flexibility can be achieved by using semi-automatic operations 
with each machine separated and fed by hand. The placing of the primary 
package and the instructions in the secondary packaging is usually a manual 
operation. The machine then folds and closes the secondary packaging, carries 
out any external printing and attaches the tamper evident label. Case packing is 
usually carried out by hand at this scale. Hiring the machines from the vendor or 
using contract packing-companies may be an option. 
10.4.6 Building design 
To handle the large number of processes reaching the pre-clinical trials stage, 
the building layout must be flexible and allow the use of mobile equipment. 
Often the buildings for small-scale facilities consist of a number of processing 
rooms on the ground floor with a service floor above providing all the required
services such as air conditioning. The process rooms may have technical spaces 
for other general purpose equipment, such as hydraulic power packs, vacuum 
equipment, or condensers. The rooms can also be used for access to some of the 
pipework as it enters the process space. 
The rooms are fitted out with a minimum amount of furniture and 
process equipment so that mobile equipment can be moved around and 
equipment set up. 
In some instances, one part of a specialized fixed equipment item is designed 
to be placed in a clean environment while other parts are designed to be 
installed in a technical space. Examples of this are horizontal dryers and 
centrifuges. The materials being handled are fed into the machine in the clean 
area and discharged in the clean area whereas the mechanical parts of the 
machine and the solvent handling equipment are located in the technical space. 
This is achieved by siting the equipment in the wall of the room. 
Depending on the level of instrumentation and control, it may also be 
appropriate to have separate control rooms away from the processing rooms. It 
is usually advantageous to have the control room adjacent to the processing 
rooms to be able to observe the operations. 
Changing rooms 
Changing rooms are an integral part of any pharmaceutical facility. For smallscale 
facilities these will need to be designed to ensure that the operators can be 
dressed in suitable clothing, that cross-contamination does not occur and that 
any highly active materials are not carried out of the building on clothing. 
A number of different changing rooms might be required to allow access to 
different parts of the building. 
Equipment store 
With small-scale facilities making use of mobile equipment, consideration must 
be given to the clean equipment store. It must ensure that the equipment is not 
damaged during storage. 
Equipment may need to be stored on GMP pallets so that it can be moved 
easily and so that multilevel staging can be used to save space. 
Each unit should be numbered and have a log book which clearly 
identifies its status (clean/dirty) and the processes for which it has been 
used. There should be an appropriate place for signatures of the operators 
and supervisors.
Access for potable equipment 
To be able to move equipment around a building safely, sufficient access for 
movement should be designed. Consideration should be given to: 
• the width of corridors; 
• turn areas; 
• size of doors; 
• size of lifts; 
• size of transfer hatches. 
Office/write up areas 
Experimental work generates large quantities of data and reports. Some writing 
areas will be required within the development areas adjacent to the equipment. 
In other cases it is necessary to have an office and write up area out of the main 
development areas but within the same building. This is because some 
processes run for a considerable time and only need to be visited for short 
times but at regular intervals. Often it is necessary to go through several 
different change areas, one after another, in order to arrive at an area of a higher 
or lower status within a building and this can take some time. Offices between 
the changing areas allow this time to be reduced. 
Environmental control 
Pharmaceutical products need to be handled in controlled environments to 
prevent contamination. With small-scale, flexible, frequently manually operated 
equipment, it may be difficult to provide primary containment and, 
therefore, high quality secondary containment is required. (See Chapter 8 for 
more details of room environments). 
To achieve the required flexibility, it may be necessary to provide the 
equipment to supply many of the rooms with high quality air. To prevent crosscontamination 
it may be necessary to provide each room with its own standalone 
system. 
Fume extraction and the use of flammable solvents will have an impact on 
the choice of equipment to be used for environmental control. 
Laboratory 
Since development requires many experimental tests to be carried out and 
adjustments are made to the process on the results of these tests, an in-house 
laboratory is necessary. In some cases this may be close to the process and, to 
reduce testing time, may be inside the area controlled by the innermost 
changing area.
Airlocks/pressure regimes 
The pressure regime within a building must ensure that air flows in the desired 
direction. The pressure regime along with the air locks between each area must 
be designed to prevent the following arising: 
• product contamination; 
• cross-contamination; 
• flammable vapour/dust contacting a non-flameproof and non-explosion 
proof electrical equipment; 
• highly active compounds contacting unprotected operators or the outside 
environment. 
Engineering workshop 
Small-scale equipment is often built into test rigs and modified frequently as the 
process develops. With equipment in controlled environments and operators 
having passed through a number of change areas, it is often appropriate to have 
a small engineering workshop close to the process rooms. This area must be 
carefully designed to ensure that tools are not lost and that the area does not 
become a source of contamination. 
Movable walls 
Processing areas can be made more flexible if movable walls are used. To 
achieve this, the services need to come through the ceiling where possible. With 
the correct choice of materials it is possible to have movable walls even when a 
very high quality environment is required. 
Communication between areas 
With the need for operators to be dressed in appropriate clothing for different 
areas and with need to protect the product, it is not possible to walk around a 
pharmaceutical facility with ease. This means that communication between 
areas can be difficult. 
Consideration should be given to speech panels, intercom systems, transfer 
hatches, visual panels and CCTV system to improve communication. Consideration 
should also be given to the safety of personnel working in areas that may 
be 'remote' from other areas within the building. This is particularly relevant 
where hazards exist.
Equipment cleaning 
Dedicated equipment cleaning areas will be required. In some cases solvents 
are used for cleaning and this will require explosion proof electrical 
equipment. 
Automated washing machines can be used and these have the advantage of 
reducing the labour requirements, producing reproducible results and keeping 
all the liquids handling equipment in the technical spaces. 
Building services 
For a flexible small-scale facility it will be necessary to provide a wide range of 
services to some or all of the process areas. The services will depend on the 
processes being carried out, but are likely to include: 
• water for injection (not usually required for the early stages of chemical 
syntheses); 
purified water; 
potable water; 
compressed air; 
breathing air; 
nitrogen; 
vacuum; 
air conditioning with temperature and humidity control; 
fume extraction (usually only required for chemical syntheses or where 
solvents are used); 
steam; 
cooling water; 
single fluid heat transfer fluid; 
services for solvents used in high volumes (e.g. recovery for safe disposal). 
10.4.7 Controls and instrumentation 
The control and instrumentation requirements for a small-scale facility will 
depend on the range of products being made and the equipment being used. The 
following considerations will need to be taken into account: 
the equipment selected will have to be compatible with the environment in 
which it will be used; 
the instrumentation should be suitable for in-house calibration so that it is not 
affected by the many processes used; 
control systems loops should be short, simple and flexible.
10.5 Chemical synthesis pilot plants 
10.5.1 Introduction 
According to a senior executive from one of the pharmaceutical industry's 
major multinationals, the future of the pharmaceutical industry will be 
'moulded by science, shaped by technology and powered by knowledge'. His 
views would no doubt be shared by the bosses of the other top nine 
pharmaceutical companies who, in the previous 12 months spent between 
them over .10 billion on research and development. 
Pilot plants are an essential component of the R&D operations of all major 
pharmaceutical companies seeking to provide new products for the future. The 
particular requirements for the design of each individual pilot plant will depend 
very much on the company's traditional research methods, the class of 
compounds likely to be developed and the regulatory requirements. However, 
there are some features that must be considered in every case. 
A typical pilot plant for primary chemical manufacture will normally be 
used to transform chemical processes from the original laboratory bench 
procedure towards practical industrial scale manufacturing facilities. Alternative 
process routes will be compared and evaluated until the optimum mix of 
process safety and operability, product quality and manufacturing cost are 
achieved. 
The pilot plant will also be used for the synthesis of samples and supplies to 
be used for formulation development, clinical trials, safety assessment and 
stability testing. It will normally comprise facilities and equipment for dispensing, 
reaction, separation, filtration and drying and finishing and will, therefore, 
normally include downflow booths, reactors, filtration equipment, a range of 
different types of dryers, and sieving, milling or micronizing equipment. 
When the engineer is asked to produce a design for a new chemical pilot 
plant, the main challenges will include: 
scope definition; 
multi-product and multi-process capability; 
flexibility; 
GMP operation; 
layout; 
regulatory requirements; 
political aspects. 
The following sections look at each of these areas in more detail.
10.5.2 Scope definition 
Each pharmaceutical manufacturer has their own ideas on the best pilot plant to 
suit their needs. For example, when asked their opinion following a tour of a 
competitor's highly complex fully automated plant, the pilot plant manager 
from a major pharmaceutical company replied: 'I would be much happier with 
a glass bucket and a thermometer!' 
The point is that it is extremely important to adopt a team approach when 
working on scope definition. The team must include the ultimate user(s), 
bearing in mind that these people are normally chemists or pharmacists and are 
not always aware of the impact of seemingly small changes on the overall 
engineering design. 
When plant facilities to handle novel processes are being designed, it is 
unrealistic to expect that the user's needs would be fully specified from the start. 
The process parameters are generally unknown, so the only way to proceed is to 
develop a capacity model by considering sample processes. 
The capacity model can then be reviewed against previous pilot plant 
activity and the perceived business needs. 
The useful life of a pilot plant should be at least ten years, so it pays to spend 
time at the front-end of the project speaking to the business managers and 
considering how the company's future products may evolve. 
10.5.3 Multi-product capability 
The plant must have the capability to permit the handling of future unknown 
compounds. This may be obtained by: 
• using simple (manual) material handling systems; 
• using materials of construction for the equipment and pipework that have a 
high resistance to corrosion; 
• providing a high degree of product segregation to prevent cross-contamination; 
• providing a high level of containment to protect the operators and the 
environment; 
• providing cleaning systems that allow rigorous decontamination between 
different product runs. 
• using materials of construction that do not react with the product contact 
parts. 
10.5.4 Multi-process capability 
In order to provide this capability, the pilot plant will need:
• a speculative range of vessel sizes (typically 50 to 2000 litres) in a suitable 
mix of materials of construction, based on the capacity model developed 
earlier; 
• vessels with variable volume capability, e.g. double jacket reactors; 
• variable temperature capabilities for the reactors, possibly via the use of a 
single heat transfer fluid system. A typical plant provides heating/cooling in 
the range of 1500C to -300C; 
• portable/mobile equipment, which allows equipment to be brought closer 
together avoiding complex piping runs and provides better utilization of 
available space; 
• services such as water, air, steam, nitrogen, heat transfer fluid and perhaps 
solvents, should be piped to all areas where it is remotely possible that 
processing will take place, including areas set aside for future expansion. 
This will allow maximum flexibility and provide a hedge against changes of 
function due to market forces; 
• a high quality de-mineralized water system providing a supply to purified 
water requirements; 
• equipment that is suitable for Cleaning In Place (CIP), in order to reduce 
downtime between processes. 
10.5.5 Uncharacterized products/processes 
The very purpose of a chemical pilot plant, i.e. to synthesize New Chemical 
Entities (NCEs), means that the potential hazards of the processes and 
compounds involved are not normally known at the time the facility is being 
designed. It is, therefore, necessary to provide high levels of primary and 
secondary containment. 
The dispensary design will have to allow for raw materials with widely 
differing hazard potential, which are received in a wide variety of packaging 
sizes and shapes. 
Most pilot plants have down-flow booths for operator protection during 
dispensing and a local extract ventilation system provided across all other areas. 
Other containment options, depending on the severity of the hazard, include 
glove boxes and full air suits. 
It is a key part of the design function to classify the types of compound that 
will be entering the facility and adjust containment levels accordingly. 
Another aspect of containment is the need to restrict atmospheric or other 
emissions of harmful substances to levels that are acceptable to the Environment 
Agency. For example, releases of Volatile Organic Compounds (VOCs) 
such as solvents must be prevented and will require the installation of a 
scrubbing or recovery system.
Good operating procedures in compliance with the legislation require that 
the volume of all waste materials is kept to a minimum and that all hazardous 
waste is disposed of in a safe, legal and traceable way. 
The multi-function basis and the lack of a defined process, may mean that 
novel methodology will be required to allow meaningful Safety, Health and 
Environmental (SHE) reviews to be carried out. Typically, this would involve 
the development of system envelopes (including control systems), which would 
be reviewed against guidewords to ensure that the design is sound. Such 
reviews would be expected to highlight those issues that are chemistry specific. 
These areas would have to be noted, and then developed in more detail prior to 
the introduction of each new process into the plant. 
10.5.6 Operation 
The way in which a chemical pilot plant is operated depends very largely on its 
designated purpose, but also on the traditions of the client/owner. However, 
because of the unknown and potentially hazardous nature of the compounds 
and processes to be employed, many major companies prefer to have the 
reaction areas of their pilot plant normally unmanned. 
This is of course contrary to the chemist's preference for reaction visibility. 
Typically they like to observe changes of colour or state as the reaction 
proceeds. 
On a smaller capacity plant, which is operated at medium temperatures and 
pressures, this requirement may be satisfied by using borosilicate glass 
equipment allowing the operators to observe the reaction areas through 
windows. 
On larger plants where the processes involve more onerous conditions, the 
use of glass is not tenable. In this situation, some companies have provided the 
chemists with the possibility to make real time observations of the reactor 
contents by using closed circuit television cameras. 
10.5.7 Layout 
As with most of the other topics discussed in this section, the type of layout 
adopted by the design team will very much depend on the owner's past 
experience and culture. 
Free access is highly desirable to allow easy maintenance and enable the 
inevitable plant modifications. 
Many modern pilot plants have adopted a vertical modular arrangement (see 
Figure 10.1) which allows gravity feed to be used in processing and is well 
suited to moving products between the modules via flow stations. However, this 
type of arrangement is by no means universal. A large number of manufacturers
still prefer the traditional 'reactor hall' arrangement with separate areas for the 
finishing steps including filtration, drying and particle size reduction. 
One point worth mentioning is the high level of HVAC that chemistry pilot 
plants will require in order to provide the required level of air filtration, pressure 
differentials and clean environments. This means that the routing of process 
R Reactor 
T Feed tank 
C Receiver 
Filter 
Other 
modules 
Filter 
evaporate 
extract 
Effluent 
Filter dryer 
centrifuge 
Effluent 
Filter 
Figure 10.1 Typical module schematic chemistry pilot plant 
T5 
R3 
T4 
C 
T6 
R2 
Filter 
R1 
T1 T2 T3
pipework and building services ductwork will be a critical task. It is wise to 
decide at an early stage in the project to separate these two major services to 
avoid possible clashes. 
10.5.8 Controls 
If you ask a typical pilot plant user what type of instrumentation and control 
system they prefer, they will invariable reply 'simple!' This is fine when you are 
working with passive substances and reactions, but totally unsuited to the needs 
of the modern pharmaceutical research establishment. 
The main factors affecting the choice of control system are: 
• data acquisition and storage; 
• operational safety; 
• multi-functional requirements; 
• environmental aspects; 
• regulatory compliance. 
The raison d'etre of the pilot plant is to research and develop alternative 
process routes for the preparation and scale-up of NCE's for pharmaceutical 
products. In order to achieve this mission, it must have a system for the 
recording, storage, retrieval and collation of the critical parameters observed 
during each process run. 
Many pharmaceutical products are themselves highly active, or are manufactured 
from highly active materials. This requires high levels of containment 
to protect the pilot plant operators. If containment fails, the control system must 
stop the process and activate a fail-safe alarm procedure to direct uninvolved 
personnel away from the area of risk. 
In addition to highly active substances, the controls will be required to alert 
the operators to runaway exothermic reactions and possibly detect leakage of 
flammable compounds. 
Some pharmaceutical products have a hydrogenation step in their manufacture. 
Hydrogen has very wide explosive limits and very low minimum 
ignition energy. A suitable control package in this case would, include at least 
hydrogen detectors and a trip system. 
The multi-purpose capabilities required of most modern pilot plants can also 
have a major impact on the choice of control system. If the plant is 
reconfigurable the control system must allow for these changes. A pilot plant 
recently completed for a major pharmaceutical manufacturer has around 250 
valid equipment configurations. In order to ensure that the configuration set-up 
is correct, a system of electronic tagging is scanned and checked by the control 
system for the required 'recipe'. If the arrangement is correct, the system
reveals a password that must be manually entered into the process control 
computer before process operations can begin. 
During the design of the above plant, it was found that one of the most 
economic ways of providing flexibility whilst still meeting processing and 
containment requirements was to use mobile equipment that could be installed 
at various locations throughout the plant. Each item of equipment has its own 
instrumentation and control requirements, which are identified, powered, 
controlled and recorded by the control system. When the equipment is correctly 
located, an umbilical cable is connected using a plug and socket, which 
provides the necessary signal and control for that item of equipment. At the 
same time the control system re-assigns the internal address of that equipment 
item to suit the new location. 
It is not always easy to find instrumentation that will operate across the full 
temperature range of the pilot plant whilst still meeting GMP requirements. 
Often detailed studies must be undertaken to identify and select the most 
appropriate type of sensors to be installed. 
The control system must monitor and control equipment that is installed to 
ensure that the emission limits laid down by the Environment Agency are not 
exceeded. It must take GMP into account and be suitable for validation to meet 
the requirements of the regulatory authorities. 
10.5.9 Legal and regulatory requirements 
Chemical synthesis pilot plants for the pharmaceutical industry must be 
designed to be safe and not pollute the environment. The multi-function 
basis and the lack of defined process will probably mean that novel methodology 
will have to be developed and agreed with the legislative authorities prior 
to Safety, Health and Environmental (SHE) reviews being carried out. 
Due to the multi-process nature of the plant, safety reviews need to take 
place throughout the life of the facility. The initial reviews take place during the 
engineering design phase, then during commissioning and following that, 
whenever a new process configuration is required during operation. 
As the plant will normally be used to manufacture small quantities of 
product for clinical trials and potentially subsequent marketing purposes, it 
must be designed to meet current Good Manufacturing Practice (cGMP) and be 
suitable for validation by the appropriate regulatory authority. 
The pilot plant may also be used to demonstrate the suitability of the selected 
manufacturing process for industrialization. The normal scale-up factor permitted/
accepted by the regulators is 10:1.
10.5.10 Cost 
There is no precise guidance on the relative costs of pilot plants when compared 
to typical manufacturing facilities, other than that the unit cost of the pilot plant 
will always be higher. 
The reasons for the higher costs are simply put down to the wide range of 
features previously described, which are employed to obtain maximum flexibility 
and benefit from what normally represents a major investment without 
guaranteed returns. 
It should be expected that the ratio of engineering costs to overall costs 
would also be higher than for conventional manufacturing units. 
The complexity of the pilot plant design to increase as engineering 
progresses should be expected and allowed for. This will be brought about as 
solutions are evolved to problems, and by new technology coming available 
which improve the general usefulness of the plant. 
Validation cost is very significant and must be considered from the outset. 
10.5.11 Political aspects 
A new chemical pilot plant will often be of major strategic importance to the 
owner, not only because it provides the vital link in developing promising, newly 
discovered products to market, but also because it demonstrates to investors that 
this is very much a research-led organization, planning for future growth. 
The new facility may often be the only facility of its kind within the 
company, so the design, layout and its worldwide location may be subject to 
thorough vulnerability analysis to ensure its security and availability. 
10.6 Physical manipulation pilot plants 
The equipment used for physical manipulation includes: 
crystallizers; 
filters; 
filter/dryers; 
centrifuges; 
dryers; 
mills; 
micronizers. 
This equipment makes use of solvents and gravity flow and is used within 
the same facility as a chemical synthesis pilot plant. For equipment that falls 
into this category, most of the detail given in Section 10.5 will be appropriate. 
However, it should be remembered that physical manipulation is being applied
to an active pharmaceutical ingredient (API) and that the equipment will need 
to be compatible with the GMP requirements. 
In a few cases the physical manipulation equipment does not make use of 
solvents and gravity flow is of no particular advantage. In these cases this 
equipment may be included in a final formulation facility and the information 
contained in Section 10.7 will be appropriate. 
10.7 Final formulation, filling and packing pilot plants 
The equipment used in this type of pilot plant is a smaller version of the 
production scale equipment. Facilities are usually built to cope only with 
certain types of products. For example, a facility to manufacture tablets is likely 
to be able to cope with a large variety of different products because the 
processes involved in making tablets are similar even if the active ingredient is 
completely different. However, this facility would be completely different to 
one making inhalation products even if the tablet and the inhalation product 
contained the same active ingredient. 
The design of these types of pilot plants is discussed in detail. 
10.7.1 Cross-contamination 
With the potential to use a large number of products within a pilot plant, crosscontamination 
is a problem, which means that containment is important. With 
automatic equipment dedicated to specific purposes it is possible to make use of 
primary containment to some extent, but with the need to make frequent 
changes and modifications it is probably wise to provide secondary containment. 
The secondary containment may be in the form of isolators around the 
equipment but it may be appropriate to have each piece of equipment in its own 
room. 
To maintain flexibility it will be necessary to have easily cleaned 
equipment. Some use may be made of Clean In Place techniques, but it is 
inevitable that equipment will have to be disassembled. This can be one of 
the greatest sources of airborne particulates, which can lead to cross-contamination, 
so this need must be considered at the design stage. Rooms and 
isolation cabinets will need to be designed with easy cleaning in mind. 
10.7.2 Material flow and storage 
Due to the potential to use many different products and with processes being 
under development, it is easily possible to mix up materials. Materials flows
need to be simple and prevent incompatible materials coming into contact with 
one another. 
Good housekeeping is a major priority. Storage facilities must have 
sufficient space for easy access and materials must be readily identifiable. 
Separate areas for raw materials, quarantine materials and passed finished 
products are required. 
10.7.3 Flexibility 
At the production scale, equipment for the final formulation, filling and 
packaging is often connected directly together. This is good for the high 
production levels required at the full scale, and it allows a high level of 
automation and minimizes labour requirements. However, such systems are not 
flexible. 
Flexibility can be increased by having stand-alone machines and moving the 
output from one machine to the next by hand. This requires a number of 
suitable mobile containers to be included in the design. 
It is also possible that some of the smaller machines can be made mobile, 
which allows the facility to have a reduced number of processing areas with 
equipment not in use stored in an appropriate place. 
10.7.4 Automation 
With filling and packing it is necessary to automate the machines at one-tenth 
the production scale. However, to enable easy change between different 
products, the automation should be kept as simple as possible. Changes to 
the system must be possible without reconfiguring the computer software which 
would require a high level of documentation to validate software changes. 
10.7.5 Building requirements 
The building requirements for final formulation, filling and packing pilot plants 
is similar to that required for small-scale facilities with the following 
differences: 
fewer rooms are required but the rooms will be larger; 
if mobile equipment is used it will be larger. It may only be possible to move 
the equipment by having very large doorways or by having removable walls; 
fewer building services will be required and it is likely that each room will only 
be supplied with the services appropriate to the equipment used in that room.
10.8 Safety, health and environmental reviews 
The requirement to carry out a number of different processes makes a SHE 
review difficult at the design stage. It is necessary to carry out some form of 
generic review and to examine those processes that are currently known. 
The introduction of each new process will require further SHE audits to 
ensure no new problems have been introduced. 
10.9 Dispensaries 
Dispensaries are an important part of pharmaceutical processing and are 
described in Chapter 6. Since small-scale facilities and pilot plants use a 
large number of products, dispensaries are a major area of risk from crosscontamination. 
Dispensaries need to be considered at the design stage and integrated with 
the operation of the facility. Sufficient space must be allowed to ensure 
operations are safe and efficient. 
10.10 Optimization 
Processes carried out within small-scale facilities and pilot plants are not 
usually optimized, because the facility is multi-functional. It is usually 
necessary to sacrifice speed of processing and product recovery in order to 
achieve flexibility. 
Equipment should be chosen to ensure that it is: 
quick and easy to change between products; 
easy to clean; 
retains the minimum amount of product; 
simple to operate; 
conforms to GMP requirement. 
10.11 Commissioning and validation managemeni 
The User Requirement Specification is always difficult to define for these 
facilities, however once the Design Qualification has been agreed and signed 
off Installation Qualification is similar to that for production scale commissioning 
except that it is an ongoing operation as new processes are being 
continually introduced. 
Performance Qualification is more of a problem because data will need to be 
added to the validation files each time a new process comes on-line.
11.1 Introduction 
Biotechnology, 'the application of biological systems and organisms, to 
technical and industrial processes and products' is not a new discipline. The 
fermentation of grain using yeast to produce alcohol has been taking place for 
centuries in most cultures throughout the world. However, advances over the 
past 20 or so years in the field of molecular biology and hybridoma technology 
have provided us with many new opportunities for improved processes and 
products. Human healthcare in particular is now beginning to benefit from these 
rapid advances in modern biotechnology, proving that it offers much more than 
just the promise of new drugs to solve many of the serious health issues facing 
mankind. The first bio-pharmaceuticals reached the market nearly a decade ago 
and are making a significant contribution not only to health care around the 
world, but also to the finances of the companies manufacturing them. 
Bio-pharmaceuticals, which generally include vaccines, blood and blood 
products, allergenic extracts, and biological therapeutics, are regulated under a 
whole range of guidelines from a variety of regulatory authorities. These 
authorities require that bio-pharmaceuticals be manufactured and prepared at a 
facility holding an unsuspended and unrevoked licence. Lack of clarity about 
licensing requirements can lead one to make major investments in large-scale 
manufacturing facilities before initiating the clinical trial(s) necessary to 
demonstrate the safety and effectiveness of the products. Such investments 
can result in significant financial loss if the product is not ultimately brought to 
market. This chapter will attempt to clarify the regulatory requirements for the 
use of small-scale and pilot facilities. For details of regulatory aspects see 
Chapters 2, 3 and 4. 
P i l o t m a n u f a c t u r i n g 
f a c i l i t i e s f o r t h e 
d e v e l o p m e n t a n d 
m a n u f a c t u r e o f 
b i o - p h a r m a c e u t i c a l 
p r o d u c t s 
TINA NARENDRA-NATHAN 
1 1
The principals that apply to small-scale and pilot plant facilities equally 
apply to manufacturing facilities. 
11.2 Regulatory, design and operating considerations 
11.2.1 Regulatory considerations 
The development of important new biological products is expensive and timeconsuming 
and companies must be able to forecast and evaluate their 
expenditures for this process. Constructing a new large-scale facility to 
manufacture a product that has not been fully tested in clinical trials could 
result in a major financial loss, with the company being unable to recover a 
major capital expenditure if the product is not ultimately brought to market. For 
some companies the best financial option may be the use of a pilot facility 
where the product may be manufactured at a smaller scale than would be for an 
approved product. While regulatory authorities do not object to the use of pilot 
production facilities for the manufacture of clinical material, provided such 
manufacture is in compliance with the requirements applicable to investigational 
drugs, many companies are concerned that these facilities and the 
products manufactured in them would not be eligible for establishment 
licensure. 
Although the advances in the technology have been staggering, it must 
be recognized that the same basic regulations and requirements are still 
applicable to the manufacture and control of bio-pharmaceuticals as for 
'conventional' Pharmaceuticals. The regulatory requirements for taking a 
conventional pharmaceutical through clinical trials to the market, however, 
emphasize the physico-chemical analysis of the 'final dosage form', which is 
then correlated with a suitable bio-assay to provide assurances of product 
uniformity. On the other hand, with a bio-pharmaceutical which cannot be 
totally defined by simple analyses of its physico-chemical characteristics and 
biological activity, most of the complexities occur during the bulk manufacturing 
process, while the preparation of the final dosage form for most part is 
rather 'uncomplicated'. 
For this reason, the bio-pharmaceutical industry, together with the 
regulatory authorities, decided to focus upon the entire manufacturing 
process and not simply on the monitoring and analysis of the final dosage 
form. This is important as the quality, safety, and efficacy attributes of a biopharmaceutical 
for which end-product controls alone are inadequate, can only 
be assured by having comprehensive controls over the entire manufacturing
process. Therefore, as well as validating the consistency of manufacture and 
characterizing the final product, constant monitoring throughout processing 
is also stressed. This results in much work needing to be completed even 
before the clinical trials could commence. For example, over 750 different 
separate 4in-process tests' are carried out in the manufacture of a recombinant 
human growth hormone, whereas only about 60 tests are required in the 
chemical synthesis of a conventional peptide hormone such as the thyroid 
hormone. 
In order to further streamline the approval process, the regulatory authorities 
have recently changed their procedures to eliminate the requirement for a 
separate establishment licence for certain 'well-defined' classes of biological 
products. Recent scientific advances, both in methods of manufacture and 
analysis, means that some products developed through biotechnology can be 
characterized in ways not historically considered possible, thereby enabling the 
authorities to allow well-characterized biological products to be regulated under 
a single application. 
The guiding principle is that an application for establishment licensure can 
be made for any facility (regardless of the scale of manufacture) which has been 
fully qualified, validated, operates in accordance with current good manufacturing 
practices (cGMPs) and which also complies with applicable local laws 
and regulations. These facilities should be distinguished from facilities used in 
research and development that may not operate under appropriate current good 
manufacturing practices (cGMPs). When manufacture of a product is transferred 
from a pilot to a different facility, a demonstration of product consistency, 
as well as data comparing the two products, together with the relevant 
process validation data should be submitted to the regulatory authorities. This 
should include a description of the manufacturing changes that have occurred, a 
protocol for comparing the products made in each facility, and the data 
generated using this protocol, as well as documentation on process validation 
and all stability data for the product manufactured in the new facility. It would 
be expected that the methods of cell expansion, harvest, and product purification 
would be identical except for the scale of production. For each manufacturing 
location, a floor diagram should be included that indicates the general 
production facility layout, as well as information on product, personnel, 
equipment, waste and air flow for production areas; an illustration or indication 
of which areas are served by each air handling unit; and air pressure 
differentials between adjacent areas. 
It is, therefore, quite obvious how important it is that the manufacturer 
discusses with the regulatory authorities what data are necessary to compare 
products, as such data may range from simple analytical testing to full clinical
trials, and could well be required even before the product made using the new 
facility or process is allowed to be included in any further clinical trials. 
11.2.2 Design considerations 
The cost of building facilities that are fully validated and in compliance with 
cGMP can be overwhelming to biotechnology companies with limited finances. 
The basic design and construction costs are driven higher by the various 
regulatory, containment, process utilities and waste treatment requirements. In 
addition, companies also demand increased value from their clinical production 
facilities. The facility design must, therefore, allow for flexibility of operations, 
diverse process utility requirements, as well as for campaigning different 
products in the same facilities. 
It is possible to build such facilities in a cost effective, flexible manner, while 
satisfying the regulatory requirements as well as ensuring that the completed 
facility will provide all the functions intended. The most effective techniques 
used to manage such a project would be the use of the concept of 'total project 
management'. 'Total project management' means integrating regulatory 
requirements, design and engineering, validation, as well as construction 
requirements on one single schedule, to determine the critical path (least 
time to completion). This leads to more effective management, permitting 
'what if scenarios that can result in substantial savings in time and cost, 
especially if cost estimation is implemented early in the design phase. 
The key element is to begin with the careful analysis of manufacturing 
process needs and to define the facility requirements specifically. Careful site 
selection is important to eliminate any costly surprises. It is also important to 
avoid over-specifying very expensive process utilities. This would be followed 
by the implementation of modular facilities design and construction. A well 
thought-out facility design using pre-engineered, self-contained elements can in 
many cases be the most cost effective, flexible solution to clinical production. 
The application of modular clean rooms to create the cGMP facilities for 
different products can therefore be achieved. 
Buildings and facilities used in the manufacture, processing, packing, or 
holding of bio-pharmaceuticals should be of suitable design, size, construction 
and location to facilitate cleaning, maintenance and proper operations. 
Adequate space should be provided for the orderly placement of equipment 
and materials, to prevent mix-ups and contamination among different raw 
materials, intermediates, or the final product. The flow of raw materials, 
intermediates and the product through the building or buildings, should be 
designed to prevent mix-ups and contamination. To prevent mix-ups and 
contamination, there should be defined areas and/or other control systems
for all the important activities. Also, facility design must be integrated in 
support of the process in order to comply with cGMP and other regulatory 
requirements such as: 
flow of personnel, materials, product, equipment or glassware, and waste 
flows; 
product separation and/or segregation; 
aseptic and/or sterile processing; 
sanitary design — cleaning and decontamination and spill containment; 
bio-hazard containment and/or isolation; 
special clean utilities; 
solvent recovery, handling, and storage; 
HVAC zoning, pressurization, and filtration; 
drain and exhaust systems. 
11.2.5 Operating considerations 
Implementing cGMP 
The current Good Manufacturing Practices (cGMPs) mentioned above are 
those practices designed to demonstrate that the control over the process, the 
facility, and the procedures used in the manufacture, maintains the desired 
quality of the product, be it a conventional drug or a bio-pharmaceutical, and 
consequently protects the product's integrity and purity. The implementation of 
cGMP is now a legal requirement and certainly makes for better quality 
products and sound economic sense. 
As technology and scientific knowledge evolve, so does understanding of 
critical material, equipment and process variables that must be defined and 
controlled to ensure end product homogeneity and conformity with appropriate 
specifications. The cGMP regulations would not achieve their statutory 
mandated purposes if they were not periodically reassessed to identify and 
eliminate obsolete provisions or to modify provisions that no longer reflect the 
level of quality control that current technology dictates and that the majority of 
manufacturers have adopted. cGMP regulations are based on the fundamental 
concepts of quality assurance: 
• quality, safety, and effectiveness must be designed and built into a product; 
• quality cannot be inspected or tested into a finished product; 
• each step of the manufacturing process must be controlled to maximize the 
likelihood that the finished product will be acceptable.
Even though cGMPs have been known and have been evolving for over 20 
years, many pharmaceutical and biotechnology companies (both established 
companies and those just starting operations) still need to achieve a sound basic 
understanding and implementation of the fundamental rationale and requirements 
of cGMP. There is still a persistent lack of understanding among a 
limited number of manufacturers with respect to certain of the cGMP regulations. 
Some pharmaceutical firms have not subjected their procedures to 
sufficient scrutiny, while others have failed to update such procedures to 
accommodate changes or advances in the manufacturing process. In some 
cases, manufacturers may be relying on methods and procedures that were 
acceptable at some time in the past, but that are not acceptable in light of current 
standards. The regulatory authorities have also encountered serious deficiencies 
particularly with validation procedures designed to ensure the quality of the 
manufacturing process. 
Those implementing cGMPs in the design of bio-pharmaceutical facilities 
must recognize the inherent variability in the manufacturing processes. 
A distinction can be drawn between the application of cGMPs to wellcharacterized 
operations, such as filling and finishing, and the nature of the 
early stages of biotech product manufacturing typified by the attributes below: 
raw material variances; 
product yields; 
non-linear process flow, reprocessing; 
process complexity. 
The role of process validation 
cGMP regulations specify the nature and extent of validation that is necessary 
to ensure that the resulting products have the identity, strength, quality and 
purity characteristics that they purport to possess. The term validation is used 
for those elements of the manufacturing process under the control of the 
manufacturer, while the term qualification is used for those items produced by a 
person other than the manufacturer, or otherwise not under the control of the 
manufacturer. Process validation is the establishment of documentary evidence 
to provide a high degree of assurance that a specifically defined process, using 
specified equipment and systems, which when in control, will consistently and 
reliably yield a product meeting its pre-determined specifications and quality 
attributes or characteristics. 
So what does validation actually mean to the ordinary scientist responsible 
for putting together a process for the manufacture of a bio-pharmaceutical. 
Validation is simply the formal process of establishing with a high degree of
assurance, and demonstrating to the relevant authorities, through a programme 
of documented tests, challenges, and results, that an item of equipment, system, 
or process actually and consistently does what it claims to do. Because it 
guarantees the ability to achieve and routinely maintain a product of a quality 
which meets all its pre-determined specification, it provides for a better 
understanding of how the equipment, system, or the process works, as it 
highlights potential weaknesses and enables corrective action to be taken. Also, 
by demonstrating reliable and consistent performance, validation also ensures 
profitability, because a validated process should be under control to such an 
extent that any deviation could be detected and enable corrective action to be 
taken. 
So how and when do the regulatory authorities recommend that process 
validation be carried out? The validation programme should begin with the raw 
material in the warehouse or stores, and finish when the final product is fully 
packaged and ready for use. When any new manufacturing formula or method 
of preparation is adopted, steps should be taken to demonstrate its suitability for 
routine processing. However, validation is required not just when a totally new 
and untried item of equipment or system is adopted, but on every occasion that 
any of the above is substantially amended, as product quality and/or the 
reproducibility of the process may be affected. Also, processes and procedures 
should undergo periodic critical re-validation to ensure that they remain 
capable of achieving the intended results. 
Experience has shown that a simple, logical, well-planned approach is the 
key to achieving success with process validation. Not only will this minimize 
the mountain of documentation required, but will also provide the training for 
process, plant and maintenance personnel, as well as providing the basis for any 
calibration and preventative or routine engineering maintenance programmes 
required. Also, if validation is planned, interfaced and integrated with the 
design and construction phase of the operation, then user requirements can be 
addressed, enabling the overall timelines to completion to be shortened. This in 
turn will minimize expensive duplication of effort, by identifying and enabling 
correction of potential design mistakes or omissions. 
Validation strategy 
The validation programme should embrace steps in the process that are critical 
to the quality and purity of the final product and should include all associated 
facilities, operating utilities and equipment. All critical process operations and 
facilities are required to be systematically investigated to ensure that the 
product can be manufactured reliably and reproducibly using all the predefined 
production and control methods. It is important to remember, however,
that the level of validation should be appropriate to the end use of the product. 
The requirements become less stringent, but no less important, further away 
from the final process step. A final dosage filling facility for a parenteral will 
require a much higher degree of validation than an intermediate bulk production 
facility. 
Validation begins with the development of the Master Validation Plan. It is 
important to combine the MVP with the construction schedule to ensure that 
validation is a focus of the total effort and that validation documentation is 
available as necessary and prepared concurrently with construction, and to 
ensure that the overall time to complete validation is minimized. The VMP 
should include and cover the following: 
• a summary of the validation philosophy, its approach and rationale; 
• a definition of the product in terms of its critical quality attributes, including 
purity, qualitative and quantitative impurity profiles, physical characteristics 
such as particle size, density, polymorphic forms, moisture and solvent 
content, if appropriate, homogeneity, and whether the product is susceptible 
to microbial contamination; 
• a summary of the methodologies and techniques to be used; 
• identification of process steps and parameters that could affect the critical 
quality attributes of the product, and the range for each critical process 
parameter expected to be used during routine manufacturing and process 
control. These should be determined by scientific judgment, and typically be 
based on knowledge derived from research and scale-up batches, unless a 
specific parameter can only be determined from manufacturing experiences 
gained from a production-scale batch; 
• validation planning worksheet identifying individual tasks; 
• list of available resources — both internal and external; and resource 
levelling to establish the time required for the project based on the available 
resources. 
The documentation related to the validation programme is as important as 
the execution of the programme itself, if not more so. The design and 
implementation of the documentation system involves the preparation, 
review (audit), and authorization of all required validation protocols for the 
standard operating procedures (SOPs), and manufacturing instructions, including 
calibration methods (metrology programmes), acceptance and certification 
criteria, as well as the assignment of responsibility. The validation protocol is 
the blueprint of the validation process for a particular drug product. It is the 
written plan describing the process to be validated, including the equipment 
used, and how validation will be conducted. The protocol should specify a
sufficient number of replicate process runs to demonstrate reproducibility, and 
provide an accurate measure of variability among successive runs. 
Execution of validation field activities 
This begins with installation qualification (IQ), followed by operational 
qualification (OQ) and finishes with performance qualification (PQ), which 
covers both equipment (or system) validation, and process qualification, 
including establishing critical circumstances for re-validation. 
Installation qualification (IQ) is the formal process of verifying and establishing 
confidence that an item of equipment or system was received and 
installed, meets the specification as ordered and intended, that the proper 
utilities are available and supplied, that it is installed as recommended by the 
manufacturer, any local or state codes, standards and cGMP, and is capable of 
consistently operating within established limits and tolerances. 
It is clear, therefore, that the 'as-built' drawings and other documents 
supplied by the manufacturer are essential to successfully carry out installation 
qualification (IQ). 
Operational qualification (OQ) is the formal process of verifying and 
establishing that such an item of equipment or system, once installed, is 
capable of satisfactory operation as specified and intended, over the entire 
range of operational parameters such as pressures, temperatures, etc. It involves 
water commissioning to check the various ancillaries such as motors and 
valves, and usually follows installation qualification (IQ), but can also be 
carried out concurrently. 
Performance qualification (PQ) is the formal process of verifying and 
demonstrating confidence by rigorous challenges and testing, that this item 
of equipment or system, once installed and operationally qualified, is capable of 
operating effectively and reproducibly in the process step for which it is 
intended. This is normally carried out in two parts: 
equipment (or system) validation; 
process qualification. 
Equipment (or system) validation involves the following as appropriate: 
sterilization validation by using temperature mapping techniques, followed 
by the verification of asepsis, or sterility testing; 
containment validation, using the host organism or another 'safe' organism; 
calibration of instruments and certification; 
validation of computer hardware and software used in the process; 
cleaning validation, particularly important in multi-product facilities.
Next comes process qualification. Process qualification is the major component 
of the whole validation effort, as it relates directly to the changes the raw 
material undergoes during its transformation to the final product. Process 
qualification is where each critical process step in the manufacture is defined 
with sufficient specificity and each such step is suitably challenged and tested to 
determine its adequacy and capability. It is essential that the validation runs are 
as representative as possible to routine manufacturing steps in terms of 
activities, conditions and characteristics, to ensure that the results obtained 
are relevant to routine production. The performance of the various challenges 
and the compilation of the results must confirm conclusively that the equipment 
or system involved in the process step is capable of providing the pre-described 
confidence levels. Manufacturers are also expected to have validation reports 
for the various key process steps. For example, if an ion-exchange column is 
used to remove endotoxins, there should be data documenting that this process 
is consistently effective. By determining endotoxin levels before and after 
processing, a manufacturer should be able to demonstrate the validity of this 
process. It is important to monitor the process before, during and after to 
determine the efficiency of each key purification step. Spiking the preparation 
with a known amount of a contaminant to demonstrate its removal is a useful 
method to validate such a procedure. 
Prospective, concurrent and retrospective validation 
Prospective validation covers activities that should be conducted prior to the 
commercial distribution of the product manufactured by either a new or 
substantially modified process. When carrying out prospective validation, 
data from laboratory and/or pilot-scale batches should identify critical quality 
attributes and specifications, critical process steps, control ranges, and inprocess 
tests. Scale-up batches can be used to generate data to confirm or refine 
earlier work, however production-scale batches are needed to provide data 
showing consistency of the process, using validated analytical methods. The 
number of consistent process runs would depend on the complexity of the 
process or the magnitude of the process change being considered. Although 
three consecutive, successful production batches should be used as a guide, 
there may be situations where additional process runs are warranted to prove 
consistency of the process, for example, for products with complex processes 
such as a recombinant cell fermentation, or for processes with prolonged 
completion times, such as with an animal cell culture. 
Regulatory authorities consider concurrent validation to be a sub-set of 
prospective validation. They recognize that in a limited number of cases it may 
not be possible to complete validation of a process in a timely manner before
distribution of the product, when data from replicate production runs are 
unavailable, possibly because only a limited number of batches intended for 
clinical or orphan drug products have been produced. In such cases, the 
manufacturer should do all the following: 
• perform all the elements of prospective validation, exclusive of replicate 
production run testing, before releasing any batch for distribution; 
• document the reasons for not completing process validation; 
• batch production records, in-process controls, and analytical data from each 
process run should be evaluated thoroughly to determine whether or not each 
batch should be released. 
This approach should not be viewed as a viable alternative if the number and 
frequency of production batches permit timely completion of process validation 
prior to product distribution. Also, if analysis of the data shows that the process 
used to manufacture the distributed batches was not, in fact, validated, no 
additional batches should be distributed until corrections have been implemented 
and the process is deemed to be validated. 
Retrospective validation may be conducted for a well-established process 
that has been used without significant changes, such as changes in raw 
materials, equipment, systems, facilities, or in the production process, that 
affect the critical quality attributes of the product. This validation approach 
should only be used when there is sufficient history on past production batches 
to demonstrate that the process consistently produces acceptable products, and 
where: 
• critical quality attributes and critical process parameters have been identified 
and documented; 
• appropriate in-process specifications and controls have been established and 
documented; 
• there have not been excessive process or product failures attributable to 
causes other than operator error or equipment failure unrelated to equipment 
suitability; 
• impurity profiles have been established for the existing product. 
The number of batches to review will depend on the process, but, in general, 
data from 10 to 30 consecutive batches should be examined to assess process 
consistency. All batches within the selected review period should have been 
manufactured by the same process and have the same documented history of 
controls and tests as the current products.
Cost of validation 
So why does validation cost so much, take so long, and what can be done about 
it? Validation of a bio-pharmaceutical facility is based on the time-consuming 
accumulation of details and sometimes the cost of validation can exceed the 
total cost of a project's architecture and engineering fees. Precious validation 
time could be spent trying to obtain information from designers, engineers, 
contractors and manufacturers, which could have been specified and provided if 
it were considered an integral part of the project. Additionally, most project 
managers are more concerned with completing the facility than with completing 
validation. The key is, therefore, to make validation an integral part of the 
project and include the validation master plan, preparation of protocols and 
SOPs, and their execution, as a series of tasks on the critical path in the total 
project schedule. 
In conclusion 
It is clear that process validation represents a sizeable investment in time and 
resources, usually taking place during a time period when the scientist and plant 
personnel are already heavily involved in start-up related activities. The 
resulting time constraints can often affect the quality of the work needed, so 
it is important to identify the pitfalls normally encountered during the process 
of validation so that they can be avoided. 
Under-estimating or under-resourcing the amount of work required is the 
most common problem; a simple, well planned, and logical approach to 
validation is the key to overcoming this problem. 
Surprisingly, too much validation can also be a problem; however, by 
identifying the critical conditions for each step in the process, it should be 
possible to avoid this pitfall and save valuable resource and effort. 
Re-validation and change control 
Once the validation and certification procedure is completed, the equipment, 
system or process is considered acceptable for use, but only under those 
conditions and functions specified in the validation protocol. To preserve the 
validated status of a process, measures must, therefore, be taken that will allow 
any significant process changes to be recognized and addressed promptly. For 
example, a slight change in the physical characteristics of an ingredient, or in 
the order of adding ingredients, may alter the specification of a product. 
Because of such effects, re-validation is necessary after any change in process 
or product characteristics or control procedures. Such a change control 
programme should provide for a classification procedure to evaluate changes 
in raw materials, manufacturing sites, scale of manufacturing, manufacturing
equipment and production processes. Regulatory authorities categorize 
changes to an approved application as major, moderate, or minor, depending 
on the nature and extent of the changes, and on their potential to have an 
adverse effect on the identity, strength or concentration, quality, purity, or the 
potency of the product, and on the process, as they may relate to the safety or 
effectiveness of the product. 
A major change is defined as one that could significantly affect the critical 
quality attributes of the product. Such changes that have a substantial potential 
to have an adverse effect on the product and require submission of a supplement 
for approval by the regulatory authorities prior to the distribution of the product 
made using the change, should be justified by additional testing and if 
appropriate, re-validation. Some examples include: 
• process-related changes, such as the extension of culture growth time leading 
to a significant increase in the number of cell doublings beyond validated 
parameters; new or revised recovery procedures; new or revised purification 
process, including a change in a column; a change in the chemistry or 
formulation of solutions used in processing; a change in the sequence of 
processing steps, or addition, deletion, or substitution of a process step; 
reprocessing of a product without a previously approved reprocessing 
protocol; 
• changes relating to the manufacturing processes or analytical methods that 
results in changes of specification limits or modifications in potency, 
sensitivity, specificity, or purity; establishes a new analytical method; deletes 
a specification or an analytical method; eliminates tests from the stability 
protocol; or alters the acceptance criteria of the stability protocol; 
• scale-up requiring a larger fermenter, bioreactor or purification equipment 
(applies to production stages up to the final purified bulk); 
• changes in the composition or the final dosage form of the biological product 
or even of ancillary components, such as new or different excipients, carriers, 
or buffers; 
• new or different lot of, or source for, in-house reference standard or reference 
panel, resulting in the modification of reference specifications and/or an 
alternative test method; 
• extension of the expiration dating period and/or a change in storage 
temperature, container/closure composition, or other conditions, other 
than changes based on real time data in accordance with a stability protocol 
in the approved licence application; 
• installation of a new Water for Injection (WFI) system, or modifications to an 
existing WFI system that would have a significant potential to stress or
challenge the system, such as lengthy or complicated distribution system 
extensions to service new or remote production areas, use of components of 
lesser quality or function, expansions of ambient temperature water distribution 
loops, or conversion from hot loop to ambient loop; 
• change of the sites at which manufacturing, other than testing, is performed; 
addition of a new location; contracting of a manufacturing step in the 
approved licence to be performed at a separate facility; 
• conversion of production and related areas from single into multiple product 
manufacturing areas, especially as there may be changes to the approved and 
validated cleaning procedures as well as additional containment requirements; 
• changes in the location (room, building, etc.) of steps in the production 
process, which could affect contamination or cross-contamination precautions; 
• major construction, or changes in location, involving or affecting environmentally 
controlled manufacturing or related support areas such as new 
buildings; new production areas or rooms in existing build-in-support 
systems with significant potential to affect air, water, or steam quality; 
installation of a new HVAC system involving or affecting environmentally 
controlled manufacturing or related support areas; modifications to an 
existing HVAC system that supplies aseptic processing areas. 
Moderate changes have a moderate potential to adversely affect the product 
and require a supplementary submission to the regulatory authorities at least 30 
days prior to distribution of the product made using the change. Some examples 
include: 
• automation of one or more process steps without a change in process 
methodology; 
• addition of duplicated process chain or unit process, such as a fermentation 
process or duplicated purification columns, with no changes to the in-process 
parameters; 
• addition or reduction in number of pieces of equipment (e.g., centrifuges, 
filtration devices, blending vessels, columns) to achieve a change in 
purification scale not associated with a process change; 
• change in the fill volume (per vial or syringe) from an approved production 
batch size and/or scale, excluding those that involve going from a single 
dose to a multi-dose vial, or changes in product concentration, both of which 
should be submitted as a supplement requiring prior approval; 
• changes in responsible individuals specified in the approved application, 
including manufacturers' representatives, responsible experts and other 
individuals designated to communicate with the authorities;
• modification of an approved manufacturing facility or room that is not likely 
to have an adverse effect on safety, sterility assurance, purity or potency of 
product, such as adding new interior partitions or walls to increase control 
over the environment; 
• manufacture of an additional product in a previously approved multipleproduct 
manufacturing area using the same equipment and/or personnel, if 
there have been no changes to the approved and validated cleaning 
procedures and there are no additional containment requirements; 
• change in the site of testing from one facility to another, such as from a contract 
laboratory to the licence holder, from an existing contract laboratory to a new 
contract laboratory, or from the licence holder to a new contract laboratory; 
• change in the structure of a legal entity that would require issuance of new 
licences, or a change in name of the legal entity or location; 
• addition of release tests and/or specifications, or tightening of specifications 
for intermediates; 
• minor changes in fermentation batch size using the specifications of the bulk 
or final product; 
• modifications to an existing HVAC system involving or affecting environmentally 
controlled manufacturing or related support areas, but not aseptic 
processing areas, with no change in air quality. 
Minor changes are those that are unlikely to have a detectable impact on the 
critical attributes of the product. Such changes would not shift the process in 
any discernible manner and might be implemented with minimal testing and 
revalidation. For example, like-for-like equipment replacements where identical 
or similar equipment is introduced into the process, is unlikely to affect the 
process if adequately installed and qualified. Such changes should be described 
and reported by the manufacturer on an annual basis. Examples would include: 
• addition of equipment for manufacturing processes which is identical to the 
primary system and serves as an alternate resource within an approved 
production room or area; 
• upgrade or minor corrective change to production air handling, water, or 
steam supply systems using equipment of the same or similar materials of 
construction, design and operating parameters, and not affecting established 
specifications; such as the removal of dead legs in the WFI system. This, 
however, does not include replacement of parts or routine repair and 
maintenance, which would not be changes to an approved application and 
would not need to be reported; 
• relocation of analytical testing laboratories between areas specified in the 
licence;
• room upgrades, such as the installation of improved finishes on floors/walls; 
• installation of non-process-related equipment or rooms to improve the 
facility, such as warehousing refrigerators or freezers; 
• modifications in analytical procedures with no change in the basic test 
methodology or existing release specifications provided the change is 
supported by validation data; 
• change in harvesting and/or pooling procedures, which does not affect the 
method of manufacture, recovery, storage conditions, sensitivity of detection 
of adventitious agents or production scale; 
• replacement of an in-house reference standard or reference panel (or panel 
member) according to SOPs and specifications in an approved licence 
application; 
• tightening of specifications for existing reference standards to provide 
greater assurance of product purity, identity and potency; 
• establishment of an alternative test method for reference standards, release 
panels or product intermediates, except for release testing of intermediates 
licensed for further manufacture; 
• establishment of a new Working Cell Bank (WCB) derived from a previously 
approved Master Cell Bank (MCB) according to a SOP on file in the 
approved licence application; 
• change in the storage conditions of in-process intermediates based on data 
from a stability protocol in an approved licence application, which does not 
affect labelling, except for changes in storage conditions, which are specified 
by regulation; 
• change in shipping conditions, such as temperature, packaging or custody, 
based on data derived from studies following a protocol in the approved 
licence application; 
• a change in the stability test protocol to include more stringent parameters, 
such as additional assays or tightened specifications; 
• addition of time points to the stability protocol; 
• replacement of equipment with that of identical design and operating 
principle involving no change in process parameters; 
• upgrade in air quality, material, or personnel flow where product specifications 
remain unchanged. Involves no change in equipment or physical 
structure of production rooms; 
• relocation of equipment within an approved operating room, rearrangement 
of the operating area or rooms where production is performed or relocation 
of equipment to another approved area to improve product/personnel/raw 
material flow and improve segregation of materials with no change in room 
air classification;
• modifications to the pre-treatment stages of a WFI system, including purified 
water systems used solely for pre-treatment in WFI production; 
• change in the simple floor plan that does not affect production process or 
contamination precautions; 
• trend analyses of release specification testing results for bulk drug substances 
and drug products obtained since the last annual report. 
Change control procedures 
No change that could affect performance in any way should be allowed without 
the written approval of at least the production, QA and engineering departments. 
Such changes should only be handled through a change control 
procedure with protocols for initiating and proving the change, together with 
procedures for re-validation. Such change control measures may apply to 
equipment, SOPs, manufacturing instructions, environmental conditions, or 
any other aspect of the process or system that has an effect on its state of control 
and, therefore, on the state of validation and should include procedures to: 
• prevent unauthorized modifications to a validated system; 
• evaluate proposed changes against development and technology transfer 
documents; 
• identify and evaluate all proposed changes to assess their potential effects on 
the process and determine if, and to what extent, re-validation is needed; 
• ensure that all documents affected by changes are promptly revised; 
• determine the impact of changes on the critical chemical and physical 
attributes of the product, such as its impurity profiles, stability, etc. 
Changes implemented to improve process yields should be evaluated carefully 
to determine if they result in new or higher levels of impurities; impurity 
profiles of resulting batches should be comparable to the batches used in drug 
safety and clinical testing, and evaluated to ensure that these do not have an 
adverse effect on analytical methods, due to increased interference caused by 
new or higher levels of impurities and by-products; and analytical methods 
should be modified as necessary to ensure that they are capable of detecting and 
quantifying impurities. 
11.3 Primary production 
The manufacture of bio-pharmaceuticals involves certain specific considerations 
arising from the nature of the products and the processes. Unlike conventional 
Pharmaceuticals, which can be manufactured, analysed and characterized using 
Next Page
• modifications to the pre-treatment stages of a WFI system, including purified 
water systems used solely for pre-treatment in WFI production; 
• change in the simple floor plan that does not affect production process or 
contamination precautions; 
• trend analyses of release specification testing results for bulk drug substances 
and drug products obtained since the last annual report. 
Change control procedures 
No change that could affect performance in any way should be allowed without 
the written approval of at least the production, QA and engineering departments. 
Such changes should only be handled through a change control 
procedure with protocols for initiating and proving the change, together with 
procedures for re-validation. Such change control measures may apply to 
equipment, SOPs, manufacturing instructions, environmental conditions, or 
any other aspect of the process or system that has an effect on its state of control 
and, therefore, on the state of validation and should include procedures to: 
• prevent unauthorized modifications to a validated system; 
• evaluate proposed changes against development and technology transfer 
documents; 
• identify and evaluate all proposed changes to assess their potential effects on 
the process and determine if, and to what extent, re-validation is needed; 
• ensure that all documents affected by changes are promptly revised; 
• determine the impact of changes on the critical chemical and physical 
attributes of the product, such as its impurity profiles, stability, etc. 
Changes implemented to improve process yields should be evaluated carefully 
to determine if they result in new or higher levels of impurities; impurity 
profiles of resulting batches should be comparable to the batches used in drug 
safety and clinical testing, and evaluated to ensure that these do not have an 
adverse effect on analytical methods, due to increased interference caused by 
new or higher levels of impurities and by-products; and analytical methods 
should be modified as necessary to ensure that they are capable of detecting and 
quantifying impurities. 
11.3 Primary production 
The manufacture of bio-pharmaceuticals involves certain specific considerations 
arising from the nature of the products and the processes. Unlike conventional 
Pharmaceuticals, which can be manufactured, analysed and characterized using 
Previous Page
chemical and physical techniques capable of a high degree of consistency, the 
production of bio-pharmaceuticals involves processes and materials that display 
an inherent variability, resulting in variability in the range and nature of the 
by-products. Moreover, the control and characterization of bio-pharmaceuticals 
usually involves bio-analytical techniques, which have a greater variability than 
the usual physico-chemical determinations. In addition, genetically modified 
cells, although providing special opportunities for producing novel protein 
sequences that exhibit improved activity compared to that of the natural 
molecule, necessitate special considerations of process design and operation. 
Therefore, the methods used in the manufacture, control, and use of these 
bio-pharmaceuticals make certain precautions necessary, and are a critical 
factor in shaping the appropriate regulatory control. Bio-pharmaceuticals 
manufactured by such methods include vaccines, immune-sera, antigens, 
hormones, cytokines, enzymes and other products of fermentation, including 
monoclonal antibodies and products derived from r-DNA, and can be largely 
defined by reference to their method of manufacture: 
• microbial cultures, excluding those resulting from r-DNA techniques; 
• microbial and cell cultures, including those resulting from recombinant DNA 
or hybridoma techniques; 
• extraction from biological tissues; 
• the propagation of live agents in embryos or animals. 
This chapter applies to the production, extraction, purification and control of 
such bio-pharmaceuticals manufactured for use in clinical trials or for marketing, 
as human or veterinary medicines, and applies to the point where the 
product is rendered sterile — i.e. the bulk active substance. 
11.3.1 Starting materials 
The source, origin and suitability of starting materials should be clearly defined. 
In instances, where the necessary tests take a long time, it may be permissible to 
process starting materials before the results of the tests are available. In such 
cases, release of a finished product is conditional on satisfactory results of these 
tests. Where sterilization of starting materials is required, it should be carried 
out where possible by heat, although other appropriate methods may also be 
used for inactivation of biological materials, such as gamma-irradiation for 
serum supplements used in the culture of animal cells. 
Control of raw materials 
Many of the raw materials used in fermentation processes can have significant 
impact on the subsequent recovery. As they are usually derived from animal
sources (such as serum, transferrin, etc.), they represent potentially variable 
sources of contaminants such as viruses, mycoplasma, or even hydrolytic 
enzymes. Pre-treatment of these raw materials by heating, acidification or 
sterile filtration is often necessary to avoid contaminating the production cells 
as well as the product. For example, contamination of the seed train by serum 
borne mycoplasma or virus may irreversibly repress cell growth and product 
titre; and once the cells are contaminated, they will produce poorly, and the 
harvest fluid will most likely contain degradative enzymes that decrease the 
quality of the purified product. 
A monoclonal antibody (mAb) may also be a raw material when used for 
purification of the product. In such cases, the standards for their production 
should be at least as rigorous as those for the product it is used to purify. The 
manufacturer must fully characterize the mAb-producing cell line, establish 
that it is free from adventitious agents, assess the purity of the mAb and validate 
its purification process for the removal of nucleic acids and viruses, as well as 
minimize residual levels of the mAb in the product of interest. 
Raw materials should be handled and stored in a manner to prevent 
contamination and cross-contamination. Identifying labels should remain 
legible, and containers should be appropriately cleaned before opening to 
prevent contamination. Written procedures should be established describing the 
purchase, receipt, identification, quarantine, storage, handling, sampling, testing 
and approval or rejection of raw materials, and such procedures should be 
followed. Bagged and boxed raw materials should be stored off the floor and 
suitably spaced to allow cleaning and inspection, and those stored outdoors 
should be in suitable containers. For solvents or reagents delivered in bulk 
vessels, such as in tanker trucks, a procedural or physical system, such as valve 
locking or unique couplings, should be used to prevent accidental discharge of 
the solvent into the wrong storage tank. Each container or grouping of 
containers of raw materials should be assigned and identified with a distinctive 
code, lot or receipt number with a system in place to identify each lot's status. 
Large containers, such as tanks or silos, which are used for storing raw 
materials, including their attendant manifolds, filling and discharge lines, 
should also be appropriately identified. 
Receipt, sampling, testing, and approval of raw materials 
Upon receipt and before acceptance, each container or grouping of containers 
of raw materials should be examined visually for appropriate labelling, 
container damage, seal integrity (where appropriate) and contamination. Raw 
materials should be held under quarantine until they have been sampled, tested 
or examined as appropriate and released for use. Representative samples of
each shipment of each lot should be collected for testing or examination in 
accordance with an established procedure. The number of containers to sample 
and the sample size should be based upon appropriate criteria, such as the 
quantity needed for analysis, sample variability, degree of precision desired, 
and past quality history of the supplier, and the sample containers properly 
identified. 
At least one test should be conducted to verify the identity of each raw 
material. A supplier's certificate of analysis may be used in lieu of performing 
other testing, provided the manufacturer has a system in place to evaluate 
vendors (vendor audits) and establishes the reliability of the supplier's test 
results at appropriately regular intervals. For hazardous or highly toxic raw 
materials, where on-site testing may be impractical, suppliers' certificates of 
analysis should be obtained showing that the raw materials conform to 
specifications. However, the identity of these raw materials must be confirmed 
by examination of containers and labels, and the lack of on-site testing for these 
hazardous raw materials should be documented. 
Use and re-evaluation of approved raw materials 
Approved raw materials should be stored under suitable conditions and, where 
appropriate, rotated so that the oldest stock is used first. Raw materials should 
be re-evaluated, as necessary, to determine their suitability for use, for example, 
after prolonged storage, or after exposure to heat or high humidity. 
Rejected raw materials 
Rejected raw materials should be identified and controlled under a quarantine 
system designed to prevent their use in manufacturing or processing operations 
for which they are unsuitable, and if necessary discarded by appropriate 
methods. 
11.3.2 Cell culture, fermentation and process control 
Cell bank system and cell culture 
The starting material for manufacturing a bio-pharmaceutical includes bacterial, 
yeast, insect or mammalian cell culture which expresses the protein 
product or monoclonal antibody (mAb) of interest. In order to prevent the 
unwanted drift of characteristics which might ensue from the repeated subcultures 
or multiple generations, the production of biological medicinal products 
obtained by microbial or animal cell culture should be based on a system of 
master and working cell banks (MCB, WCB) consisting of aliquots of a single
culture. Also known as seed lots, such cell bank systems are used by 
manufacturers to assure the identity and purity of the starting raw material. 
The MCB is derived from a single colony of prokaryotic (bacteria, yeast), or 
a single eukaryotic (mammalian, insect) cell stored cryogenically, and is 
composed of sufficient ampoules of culture to provide source material for the 
WCB. The WCB is defined as a quantity of cells derived from one or more 
ampoules of the MCB, stored cryogenically, and used to initiate a single 
production batch. Both the MCB and the WCB must be stored in conditions 
that assure genetic stability. Generally, cells stored in liquid nitrogen or its 
vapour phase are stable longer than cells stored at — 700C. 
Establishment of cell banks should be performed in a suitably controlled 
environment to protect the cells and, where applicable, the personnel handling 
them. During the establishment of the cell banks, no other living or infectious 
material such as viruses, other cell lines or cell strains, should be handled 
simultaneously in the same area or by the same persons. Only authorized 
personnel should be allowed to handle the material, and this handling should be 
done under the supervision of a responsible person. It is desirable to split the 
cell banks and to store the parts in more than one location so as to minimize the 
risks of total loss. All ampoules containing the cell banks should be treated 
identically during storage. 
Cell banks should be established, stored and used in such a way as to 
minimize the risks of contamination or alteration. They should be adequately 
characterized and tested for contaminants and shown to be free of adventitious 
agents such as fungi, bacteria, mycoplasma, and exogenous viruses; tested for 
tumourigenicity; and probed for the expression of any endogenous retroviral 
sequences by using conditions known to cause their induction; and their 
suitability for use demonstrated by the consistency of the characteristics, and 
quality of the successive batches of product. The number of generations (or 
doublings or passages) between the cell bank and the finished product should 
be as low as is practicable. 
Inoculation and aseptic transfer 
Inoculation of the seed culture into the fermenter or bioreactor, as well as all 
transfer and harvesting operations must be done using validated aseptic 
techniques. Additions or withdrawals from fermenter or bioreactors are 
generally done through steam sterilized lines and steam-lock assemblies. 
Steam may be left on in situations where the heating of the line or the 
vessel wall would not be harmful to the culture. If possible, the media 
used should be sterilized in-situ, using a Sterilization in Place (SIP) or a 
continuous sterilization system (CSS), and any nutrients or chemical added
beyond this point must be sterile. Additions of materials or cultures, and the 
taking of samples, should be carried out under carefully controlled conditions 
to ensure that the absence of contamination is maintained. Care should be taken 
to ensure that vessels are correctly connected when additions or samplings take 
place. In-line sterilizing filters should be used where possible for the routine 
addition of air and other gases, media, acids or alkalis, and defoaming agents, to 
the fermenter or bioreactor. 
Process monitoring and control 
It is important for a fermenter or bioreactor to be closely monitored and tightly 
controlled to achieve the proper and efficient expression of the desired product. 
The parameters for the fermentation process, including information on growth 
rate, pH, waste by-product levels, addition of chemicals, viscosity, density, 
mixing, aeration, and foaming, must, therefore, be specified and monitored. 
Other factors that may affect the finished product, such as shear forces, processgenerated 
heat, should also be considered. Many growth parameters can 
influence protein production. Although nutrient-deficient media are used as a 
selection mechanism in certain cases, media deficient in certain amino acids 
may cause substitutions. The presence of such closely related products may 
cause difficulties later on during the separation and purification stages, and may 
have implications both for the application of release specifications and the 
effectiveness of the product purification process. 
Containment considerations 
Bioreactor systems designed for recombinant microorganisms require not only 
that a pure culture is maintained, but also that the culture be contained within 
that system. Such containment can be achieved by the proper choice of a hostvector 
system that is less capable of surviving outside a laboratory environment, 
as well as by physical means, when this is considered necessary. For the 
cultivation of recombinant cell lines, there are defined and established physical 
containment levels. Good Large-Scale Practice (GLSP) level of physical 
containment is recommended for large-scale production involving viable, 
non-pathogenic and non-potent recombinant strains derived from host organisms 
that have an extended history of safe large-scale use, and for organisms 
that have built-in environmental limitations that, although allowing optimum 
growth in the fermenter, have limited survival outside in the environment. 
Biosafety level 1 (BLl) level of physical containment is recommended for 
large-scale production of viable recombinant organisms that require BLl 
containment at the laboratory scale. Similar recommendations exist for BL2 
and BL3. No provisions are made for the large-scale research or production of
viable recombinant organisms that require BL4 containment at the laboratory 
scale. 
Personnel considerations 
The immunological status of personnel should be taken into consideration for 
product safety. All personnel engaged in the production, maintenance and 
testing should be vaccinated where necessary with appropriate specific 
vaccines and have regular health checks. Apart from the obvious problem of 
staff exposure to infectious agents, potent toxins, or allergens, it is necessary to 
avoid the risk of contaminating a production batch with infectious agents. 
Therefore, visitors are generally excluded from production areas. Furthermore, 
in the course of a working day, personnel should not pass from areas where 
exposure to live organisms or animals is possible to areas where other products 
or different organisms are handled. If such passage is unavoidable, clearly 
defined decontamination measures including change of clothing and shoes and, 
where necessary, showering should be followed by staff involved in any such 
production. 
11.3.3 Product recovery and purification 
Once the fermentation process is completed, the desired product is extracted, 
isolated, separated and, if necessary, refolded to restore conflgurational integrity, 
and then purified. Whether the product is intra-cellular or extra-cellular, 
soluble, insoluble or membrane bound or located in a subcellular organelle will 
influence the choice of extraction method and buffer components used. 
Typically, manufacturers develop downstream processes on a small scale and 
determine the effectiveness and limitations of each particular processing step. 
Allowances must, therefore, be made for several differences when the process is 
scaled-up. Longer processing times can adversely affect product quality since 
the product is exposed to various reaction conditions, such as pH and 
temperature, for longer periods. Product stability under such varying purification 
conditions must, therefore, be carefully defined. 
Product recovery 
Determining the optimal time of harvest is an important area of interaction 
between fermentation and recovery. Often, allowing a culture to run longer 
results in an increase in titre, but with a concomitant increase in cellular debris 
and degraded forms of the product. Although it may be simple to overcome the 
effect of increased cell debris by increasing the capacity of the downstream 
equipment, it is much more difficult to purify away the slightly altered or 
degraded forms of the product.
With extra-cellular products, it is possible to achieve a high degree of 
purification by simply removing the cells. For the recovery of extra-cellular 
proteins, the primary separation of product from producing organisms is 
accomplished by centrifugation or membrane filtration. Ultra filtration is 
commonly used to remove the desired product from the cell debris. The 
porosity of the membrane filter is calibrated to a specific molecular weight, 
allowing molecules below that weight to pass through while retaining molecules 
above that weight. Centrifugation can be open or closed, although the 
adequacy of the environment must be evaluated for open centrifugation. 
Following centrifugation, other separation methods, such as ammonium 
sulphate precipitation and aqueous two-phase separation, can also be employed 
to concentrate the product. 
With extra-cellular products, cell breakage is unnecessary and undesirable. 
Cell disintegration not only releases membrane fragments that can foul process 
equipment, but also undesirable impurities derived from the cell cytoplasm, 
particularly host cell proteins and DNA. The harvest/cell separation operation 
is more difficult with mammalian and other animal cells, as they are much more 
fragile than bacterial or yeast cells. Consequently, high-speed centrifuges may 
not be appropriate and these cells must be harvested with special low shear, low 
centrifugal field centrifuges. Harvesting can also be carried out effectively and 
efficiently using depth or tangential flow filtration. The advantage of filtration is 
its ability to achieve quantitative increases in product yield by washing 
(diafiltration) the cells. 
Intra-cellular or membrane-bound products will require detergents or 
organic solvents to solubilize them. For the recovery of completely intracellular 
products, the cells must be disrupted after fermentation, which can be 
achieved by chemical, enzymatic or physical methods. Following disruption, 
the cellular debris is removed either by centrifugation or filtration. 
Purification 
Further purification steps primarily involve a variety of chromatographic 
methods to remove impurities and to bring the product closer to final 
specifications. One or more of the following column chromatography techniques 
usually achieves this: 
affinity chromatography; 
ion-exchange chromatography (IEC); 
gel filtration or size-exclusion chromatography (SEC); 
hydrophobic interaction chromatography (HIC); 
reverse-phase HPLC (RP-HPLC).
A prior knowledge of the protein stability and its sensitivity to temperature, 
extremes of pH, proteases, air and metal ions will also aid the design of a 
purification procedure. If the product to be purified is an enzyme or receptor it 
may be possible to exploit its activity by affinity purification on a substrate or 
ligand, or an analogue. Knowledge of the size and pH of the protein will 
indicate suitable matrices and conditions for gel filtration and ion-exchange 
chromatography. The final use of the product will define how much of the 
purified protein is required, whether loss of activity can be tolerated, how pure it 
should be, and the time and cost of purifying it. If it is for research use, the 
quantities required are reasonably small, whilst in terms of purity the removal 
of interfering activities becomes essential. In contrast, for therapeutic applications, 
purity is of the utmost importance and quantities required are relatively 
small. 
Selection and sequence of the downstream processing steps 
Each protein has a unique combination of properties that can be exploited for 
purification. Thus by combining a series of steps that exploit several of these 
properties, the protein can be purified from a mixture. Each technique should be 
evaluated for its capacity, resolving power, probable product yield and cost, and 
would use a different property of the product, such as charge or hydrophobicity, 
to effect adsorption and separation. These factors must be balanced against one 
another and the requirement for each stage of the purification. Moreover, the 
number of steps in a purification process should be limited by ensuring that the 
product from one technique can be applied directly onto the next step without 
further manipulations. 
The capacity of the technique is defined as the amount of sample (in terms of 
volume and protein concentration) that can be handled. A key requirement 
early in the purification is often to reduce the volume when high capacity 
techniques such as precipitation methods, which can handle the large initial 
volumes and protein concentrations, are often used first. Of the chromatography 
steps, those involving adsorption have the highest capacity. Gel filtration or 
size exclusion chromatography has a low capacity and is, therefore, usually 
inappropriate for early stages and is mostly used as a final clean up. 
The resolution of a technique determines how efficiently it separates proteins 
from one another. Precipitation steps have low resolution, whilst chromatography 
steps are more highly resolving. Affinity chromatography often shows 
extremely high resolution and it is possible to frequently achieve purification 
factors of greater than 1000 fold. 
Due to the nature of the various interactions and the conditions used, each 
technique will show a range of average yields. Precipitation with ammonium
sulphate and aqueous two-phase extraction usually gives yield of more than 
80%, whilst affinity methods often result in lower yield (~60%) due to the 
harsh conditions required for the elution of the product. 
With respect to cost, affinity techniques are usually expensive and so not 
often used as an initial purification step. A cheaper technique such as ion 
exchange chromatography is usually used first to remove the bulk of the 
contaminants such as particulate matter, lipids and DNA. 
Integration with upstream operations 
In the narrowest definition, downstream processing is the purification of 
proteins from conditioned media or broths. However, many controllable factors 
that influence purification occur early in the production process. The integration 
of downstream processing with upstream operations such as molecular biology 
and fermentation can, therefore, provide significant downstream opportunities. 
The interaction between molecular biology and recovery can take several 
forms. With recombinant DNA products, purification can be influenced before 
the starting material is even available. Given the gene sequence, it is possible to 
predict how the product will behave on size separation media and ion exchange 
resins, although the actual ionic properties of the protein may be influenced by 
its tertiary structure. Leading or tail sequences can be added to impart properties 
that will make the protein easier to purify. It is common practice in bacterial 
systems to employ fusion proteins to enhance expression, secretion or the 
subsequent recovery. In mammalian and animal cell systems, the tools of 
molecular biology are used to enhance expression levels and to alter the 
biological properties of the final product. Higher titres provide a direct benefit 
to the recovery process by increasing the ratio of product to contaminant, 
thereby reducing the fold purification that is ultimately required and also by 
enabling reductions in the operation volumes of early steps. 
Perhaps the most important examples of process integration occur in the 
interaction between recovery and fermentation. One of the primary areas of 
interaction between these disciplines is the development of suitable media for 
cell growth. In cases where the expression system uses an amplified selectable 
marker, it may be necessary to maintain selective pressure during some or all 
stages of cell culture. The use of media supplements such as serum may release 
this selection pressure, resulting in a decrease in expression level, as well as 
adversely affecting the overall recoverability by leading to complex formation 
and product degradation. This problem can be overcome by the use of low 
serum, fractionated serum, or even serum-free medium.
11.3.4 Primary production facilities 
The risk of cross-contamination between biological medicinal products, especially 
during those stages of the manufacturing process in which live organisms 
are used, may require additional precautions with respect to facilities and 
equipment, such as the use of dedicated facilities and equipment, production on 
a campaign basis and the use of closed systems, until the inactivation process is 
accomplished. The degree of environmental control of particulate and microbial 
contamination of production premises should, therefore, be adapted to the 
product and the production step, bearing in mind the level of contamination of 
the starting materials and the risk to the finished product. 
Production on a campaign basis may be acceptable for spore-forming 
organisms provided that the facilities are dedicated to this group of products, 
and not more than one product is processed at any one time. Simultaneous 
production in the same area using closed systems such as fermenters may be 
acceptable for products such as monoclonal antibodies and products prepared 
by recombinant DNA techniques. Processing steps after harvesting may be 
carried out simultaneously in the same production area provided that adequate 
precautions are taken to prevent cross-contamination. For killed vaccines and 
toxoids, such parallel processing should only be performed after inactivation of 
the culture or after detoxification. Equipment used during the handling of live 
organisms should be designed to maintain cultures in a pure state and 
uncontaminated by external sources during processing. 
Positive pressure areas should always be used to process sterile products, but 
negative pressure in specific areas at point of exposure of pathogens is 
acceptable for containment reasons. Where negative pressure areas or safety 
cabinets are used for aseptic processing of pathogens, they should be 
surrounded by a positive pressure sterile zone. Air filtration HVAC units 
should be specific to the processing area concerned and recirculation of air 
should not occur from areas handling live pathogenic organisms. The layout 
and design of production areas and equipment should allow effective cleaning 
and decontamination. The adequacy of cleaning and decontamination procedures 
should be validated. Pipework systems, valves and vent filters should be 
properly designed to facilitate cleaning and sterilization. The use of CIP and 
SIP systems should be encouraged. Primary containment should be designed 
and tested to demonstrate freedom from leakage risk. Effluents that may 
contain pathogenic microorganisms should be effectively decontaminated. 
Genetically engineered organisms 
When handling genetically engineered materials, the biosafety controls 
required should include testing facilities that adequately provide a controlled
environment and separation of test systems, as well as adequate and appropriate 
areas for receipt and storage of both the host organism and test 
substance, as well as for any other materials, such as any stocks of plants, 
feed and soils used in the study, as well as facilities for waste disposal. Both the 
laboratory facilities and any separate outdoor testing facilities, such as greenhouses 
and field sites, that are used for testing the genetically engineered 
substance should be of sufficient design (layout, size and location) to provide 
the necessary containment of appropriate biosafety level to protect personnel 
and the environment. They should be designed to provide a barrier to the 
unintended release of any organisms if a spill or application accident were to 
occur, and the decontamination facilities should be separated from the other 
areas of the facility. The laboratory should have decontamination procedures 
for containing or killing genetically engineered organisms and host organisms. 
Moreover, the facility should have proper ventilation, so that air flows from 
areas of low contamination to areas of higher contamination, and complete air 
containment and decontamination should be provided. Environmental conditions 
such as temperature, humidity and ventilation should be monitored using 
appropriate instruments, and recorded and specified in the protocol for the 
ongoing study. 
Animal quarters and care 
Animals are used for the manufacture of a number of biological products, for 
example polio vaccine (monkeys), snake anti-venoms (horses and goats), rabies 
vaccine (rabbits, mice and hamsters) and serum gonadotropin (horses). 
Animals may also be used in the quality control of most sera and vaccines, 
such as for pertussis vaccine (mice), pyrogenicity (rabbits), BCG vaccine 
(guinea-pigs). Quarters for animals used in the production and control of 
biological products should be separated from the production and control areas. 
The health status of animals from which some starting materials are derived, 
and of those used for quality control and safety testing, should be routinely 
monitored and recorded. Staff employed in such areas must be provided with 
special clothing and changing facilities. 
11.3.5 Safety issues 
The presence of process-related contaminants in a bio-pharmaceutical is chiefly 
a safety issue. The sources of contaminants are primarily the cell substrate 
(DNA, host cell proteins and other cellular constituents, viruses), the media 
(proteins, sera and additives) and the purification process (process-related 
chemicals and product-related impurities).
Residual host cells 
In the early days, there were concerns about the safety of immortal transformed 
cell lines since, by definition, they were thought to contain oncogenic 
DNA or proteins. In addition to the issues arising from the transformed nature 
of these cells, there were also concerns regarding the contamination of these 
cell lines by adventitious agents such as viruses, fungi and mycoplasma. 
Furthermore, there were also concerns about the immunogenicity resulting 
from residual host cell proteins, in patients who received drugs that were 
purified from recombinant sources. There are various regulatory guidelines for 
the characterization of the cells used in the manufacture of bio-pharmaceuticals. 
The exhaustive characterization of the cell banks by diverse methods 
provides at least an initial degree of confidence that the resultant products can 
be safely injected into humans. Concerns over the presence of residual host 
cell proteins have largely been put to rest by relying on well-established 
techniques for sterile filtration, as well as in advances in analytical method 
development. 
Residual contaminating proteins 
Due to the concern about the safety of proteins from non-human sources with 
respect to the generation of immune responses, recombinant proteins are 
generally being brought to unprecedented levels of purity. It can be as difficult 
to quantitate and prove the levels of purity as it is to achieve them. For example, 
whereas the purity of albumin preparations is commonly about 95-99%, the 
purity of recombinant products such as human growth hormone, human insulin 
or even hepatitis B vaccine is greater than 99.99% with respect to host proteins. 
In order to measure impurities at this level, two major analytical strategies have 
been developed. The first method, which is uniquely applicable to all 
recombinant products, is the use of a blank run. This involves fermentation 
and recovery using a host cell containing the selectable marker but lacking the 
gene for the product, thereby enabling the manufacturer to specifically prepare 
and quantitate the host cell derived impurities. The second approach is the 
direct measurement of the impurities. The most general method uses an 
immuno-assay based on antibodies to the host cell proteins. Although this 
type of assay is complex both in its development and composition, it provides 
an extremely sensitive way to quantitate protein impurities in each batch of 
product.
Residual nucleic acids 
When immortalized mammalian cells were first considered as host systems for 
recombinant protein, there was substantial theoretical concern about the 
possibility of DNA from recombinant immortal cell lines causing oncogenic 
events in patients receiving products from these cell substrates. However, 
various scientists have shown that DNA does not induce any oncogenic events 
when injected into immuno-suppressed rodents, even at levels at least eight 
orders of magnitude greater than that expected in a dose of human therapeutic 
protein such as t-PA. It is most likely the naked DNA is degraded very quickly 
to inactive fragments and nucleotides by circulating nucleases. With current 
technology it is possible to directly measure the DNA content of clarified cell 
culture fluid and the early processing steps with a DNA dot blot assay using 32P 
labelled DNA derived from the host cell line. For some products, especially 
those that are administered in multi-milligram quantities, it is necessary to 
demonstrate a reduction to assure a level of DNA of less than 10 pg per human 
dose. This can be further validated by spiking 32P labelled DNA into aliquots 
of process fluid and then purifying the samples on representative scaled-down 
versions of the recovery process operations. 
Viruses 
The presence of retro-viruses in continuous mammalian and other animal cell 
lines has received a great deal of attention because of concern that these 
particles can potentially cause oncogenic events in man. However, the approach 
of demonstration of freedom from functional retro-viruses in the culture is not 
usually sufficient to answer regulatory concerns, because it is always possible 
that there might be levels of retro-virus just below the sensitivity limit of these 
assays or that the specificity of the retro-virus assays might not be broad enough 
to pick up some unusual potential contaminant. To address this issue, the 
authorities require the testing of the harvested culture fluid directly for the 
presence of retro-viruses, or following concentration by ultracentrifugation 
before analysis, to increase the sensitivity of electron microscopy past the 
estimated detection limit of 106 particles per ml. These direct measurements 
can be supplemented by validated process procedures for removal and/or 
inactivation of putative retro-viruses. Only with steps that are truly independent 
is it legitimate to determine the total clearance as a result of the clearances from 
the individual steps. Moreover, the use of more than one model virus and the 
assay of the virus by more than one technique would also serve to strengthen the 
believability and validity of this approach.
Pyrogen and endotoxins 
In contrast to bacterial fermentations, especially of gram-negative bacteria such 
as E.coli, mammalian and other animal cell fermentations should contain little 
or no pyrogen, and the recovery process should not need to incorporate steps to 
remove pyrogens. The process strategy thus becomes oriented more towards 
keeping pyrogens out rather reducing their levels, and it is much more 
important to keep raw materials and equipment pyrogen-free. 
11.4 Secondary production 
One of the more difficult processes to regulate, and one which has presented 
considerable problems over the years, is that of the manufacture of sterile biopharmaceuticals. 
During the past few years, a number of sterile batches from 
different manufacturers have been reported to have exhibited microbiological 
contamination. One manufacturer had approximately 100 batches contaminated 
in a six month time period, whilst another had approximately 25 batches 
contaminated in a similar period; other manufacturers have had recalls due to 
the lack of assurance of sterility. Not surprisingly, the manufacture of sterile 
bio-pharmaceuticals is subjected to special requirements relating to the minimizing 
of risks of microbiological, as well as of particulate and pyrogen 
contamination. 
The manufacture of a sterile pharmaceutical must be performed in closed 
systems with minimal operator handling, although much of this depends on the 
skills, training and attitudes of the personnel involved. Quality assurance is 
particularly important and this type of manufacture must strictly follow carefully 
established and validated methods of preparation and procedure. Most 
bio-pharmaceuticals cannot be terminally sterilized and must, therefore, be 
manufactured by aseptic processing. Thus, it is important to recognize that as 
there is no further processing to remove contaminants or impurities such as 
particulates, endotoxins and degradants, sole reliance for sterility or other quality 
aspects, must not be placed on any terminal process or finished product test. 
11.4.1 Starting materials 
The manufacture of a sterile bio-pharmaceutical should be performed and 
supervised by competent people. The purchase of starting materials is an 
important operation, which should involve staff who have a thorough 
knowledge of the suppliers and who should only purchase from approved 
suppliers named in the relevant specification. The source, origin and suitability 
of starting materials should be clearly defined; the various components, contain-
ers and closures that are received, identified, stored, handled, sampled, tested and 
approved or rejected should be regularly inspected, and the system should be 
challenged to test if it is functioning correctly. There must be written procedures 
describing how these operations are done and if the handling and storage of 
components are computer controlled, the programme must be validated. 
Control of raw materials 
Written procedures should be established describing the purchase, receipt, 
identification, quarantine, storage, handling, sampling, testing and approval or 
rejection of raw materials, and such procedures should be followed. In fact, it is 
beneficial for all aspects of the manufacture and control of the starting material 
in question, including handling, labelling and packaging requirements, as well 
as complaints and rejection procedures to be discussed with the supplier. All 
materials and products should be handled and stored under the appropriate 
conditions established by the manufacturer, in an orderly fashion to permit 
batch segregation and stock rotation, as well prevent contamination or crosscontamination. 
The manufacturer must be able to show that the containers and 
closures are compatible with the product, will provide adequate protection for 
the product against deterioration or contamination, are not additive or absorptive, 
and are suitable for use. 
Receipt, sampling, testing and approval of raw materials 
Incoming materials should be physically or administratively quarantined 
immediately on receipt, until they have been sampled, tested or examined as 
appropriate and released for use or distribution. They should be checked to 
ensure that the consignment corresponds to the order, and examined visually 
for integrity of package and seal, for correspondence between the delivery note 
and the supplier's labels, for damage to containers and any other problem that 
might adversely affect the quality of a material. The receiving records must be 
traceable to the component manufacturer and supplier and should contain the 
name of the component, manufacturer, manufacturer's lot number, supplier if 
different from the manufacturer, and carrier. All handling of starting materials, 
such as receipt and quarantine, sampling, storage, labelling, dispensing, 
processing, packaging and distribution should be done in accordance with 
written procedures or instructions and, where necessary, recorded. 
The number of containers to sample and the sample size should be based 
upon appropriate criteria, such as the quantity needed for analysis, sample 
variability, degree of precision desired and past quality history of the supplier, 
and the sample containers properly identified. At least one test should be 
conducted to verify the identity of each raw material. A supplier's certificate of
analysis may be used instead of performing other testing, provided the 
manufacturer has a system in place to evaluate vendors (vendor audits) and 
establishes the reliability of the supplier's test results at appropriately regular 
intervals. For hazardous or highly toxic raw materials, where on-site testing 
may be impractical, suppliers' certificates of analysis should be obtained 
showing that the raw materials conform to specifications. However, the identity 
of these raw materials must be confirmed by examination of containers and 
labels, and the lack of on-site testing for these hazardous raw materials should 
be documented. Intermediate and bulk products purchased as such should also 
be handled as though they were starting materials. 
Starting materials in the storage area should be appropriately labelled and 
should only be dispensed by designated persons, following a written procedure, 
to ensure that the correct materials are accurately weighed or measured into 
clean and properly labelled containers. Materials dispensed for each batch 
should be kept together and conspicuously labelled as such. Information on the 
labels should provide traceability from the component manufacturer to its use 
in the finished product, and should bear at least the following information: 
• the designated name of the product and the internal code reference where 
applicable; 
• a batch number given at receipt; 
• the status of the contents (e.g. in quarantine, on test, released, rejected) where 
applicable; 
• an expiry date or a date beyond which re-testing is necessary, if appropriate. 
When fully computerized storage systems are used, all the above information 
need not necessarily be in a legible form on the label. 
Use and re-evaluation of approved raw materials 
Approved raw materials should be stored under suitable conditions and, where 
appropriate, rotated so that the oldest stock is used first. Raw materials should 
be re-evaluated as necessary to determine their suitability for use, for example, 
after prolonged storage or after exposure to heat or high humidity. Sanitary 
conditions in the storage area, stock rotation practices, re-test dates and special 
storage conditions, such as protection from light, moisture, temperature and air, 
should be checked regularly. 
Rejected raw materials 
Rejected raw materials should be identified and controlled under a quarantine 
system designed to prevent their use in manufacturing or processing operations 
for which they are unsuitable.
11.4.2 Final processing operations 
Sterile products are usually produced by dissolving the non-sterile bulk active 
substance in a solvent and then filtering the solution through a sterilizing filter. 
After filtration, the sterile bulk material is separated from the solvent by 
crystallization, precipitation and spray-drying or lyophilization. During these 
final processing operations, all necessary in-process controls and environmental 
controls should be carried out and recorded, and any significant deviation 
from the expected yield should be recorded and investigated. 
Critical manufacturing steps 
Each critical step in the manufacturing process should be done by a responsible 
individual and checked by a second responsible individual. If such steps in the 
processing are controlled by automatic mechanical or electronic equipment, its 
performance should be verified. Critical manufacturing steps not only include 
the selection, weighing, measuring and identifying of components, and addition 
of components during processing, but also the recording of deviations in 
the manufacturing record, testing of in-process material and the determination 
of actual yield and percent of theoretical yield. These critical manufacturing 
steps should be fully validated and documented when done. At all times during 
processing, all materials, bulk containers, major items of equipment and, where 
appropriate, the rooms used, should be labelled or otherwise identified with an 
indication of the product or material being processed, its strength (where 
applicable), batch number and the stage of production. Labels applied to 
containers, equipment or premises should be clear, unambiguous and in the 
company's agreed format. It is often helpful in addition to the wording on the 
labels to use colours to indicate status, such as quarantined, accepted, rejected 
and clean. 
Preparation 
Before any processing operation is started, steps should be taken to ensure that 
the work area and equipment are clean and free from any starting materials, 
products, product residues or documents that are not required for the operation 
being planned. Intermediate and bulk products, and all starting materials should 
be kept under appropriate conditions. Checks should be carried out to ensure 
that pipelines and other pieces of equipment used for the transportation of 
products from one area to another are connected in a correct manner. Noncombustible 
gases, and all solutions, in particular large volume infusion fluids, 
should be passed through a microorganism retaining filter if possible, immediately 
prior to filling. Any components, containers, equipment and any other 
article required in the clean area where aseptic work takes place should be
sterilized and passed into the area through double-ended sterilizers sealed into 
the wall, or by a procedure which achieves the same objective of not 
introducing contamination. Bioburden and contamination levels should be 
monitored before sterilization and where appropriate, the absence of pyrogens 
should also be monitored. The interval between the washing, drying and the 
sterilization of components, containers and equipment, as well as between their 
sterilization and use should be minimized and subject to a time-limit appropriate 
to the storage conditions. 
Batching 
Many of these bio-pharmaceutical products lack preservatives, inherent bacteriostatic 
or fungistatic activity. Obviously, the batching or compounding of bulk 
solutions should, therefore, be controlled in order to prevent any potential 
increase in microbiological levels that may occur up to the time that the bulk 
solutions are filter sterilized. One concern with any microbiological level is the 
possible increase in endotoxins that may develop. Good practice would, 
therefore, include working in a controlled environment, and in sealed tanks 
to control accessibility, particularly if the non-sterile product solutions are to be 
stored for any period prior to sterilization. 
Filling 
The filling of bio-pharmaceuticals into ampoules or vials presents many of the 
same problems as the processing of conventional pharmaceuticals. The batch 
size of a bio-pharmaceutical is likely to be small and the validation of aseptic 
processes presents special problems when the batch size is small. In these cases, 
the number of units filled may be the maximum number filled in production and 
because of the small batch size, filling lines may not be as automated as for 
other products typically filled in larger quantities. Moreover, filling and sealing 
will often be a hand operation, presenting great challenges to sterility; and with 
more involvement of people filling these products, more attention should be 
given to environmental monitoring. Typically, vials to be lyophilized are 
partially stoppered by machine. However, some filling lines have even been 
observed using an operator to place each stopper on top of the vial by hand. The 
immediate concern in this case is the avenue of contamination offered by the 
operator. Due to the active involvement of people in filling and aseptic 
manipulations, the number of persons involved in these operations should be 
kept to a minimum, and the environmental programme should include an 
evaluation of microbiological samples taken from people working in such 
aseptic processing areas. Some of the problems that are routinely identified
during filling include inadequate attire, deficient environmental monitoring 
programmes and failure to validate some of the basic sterilization processes. 
One major concern is the use of inert gas to displace oxygen during both the 
processing and filling of the solution, and therefore, limits for dissolved oxygen 
levels for the solution must be established for products that may be sensitive to 
oxidation, and parameters such as line speed and the location of the filling 
syringes with respect to their closures should be defined. In the absence of inert 
gas displacement, the manufacturer should be able to demonstrate that the 
product is not affected by oxygen. Another major concern with the filling 
operation of a lyophilized product is the assurance of fill volumes. Obviously, a 
low-fill would represent a sub-potency in the vial. Unlike a powder or large 
volume liquid fill, a low-fill would not be readily apparent after lyophilization, 
particularly for a product where the active ingredient may be only a milligram. 
Due to its clinical significance, sub-potency in a vial can potentially be very 
serious. 
Lyophilization (freeze drying) or spray drying 
Many bio-pharmaceuticals are lyophilized because of stability concerns. 
Unfortunately, the cGMP aspects of the design of lyophilizers have lagged 
behind the sterilization and control technology employed for other processing 
equipment. It is not surprising that many problems with the lyophilization 
process have been identified. These problems are not limited to biopharmaceuticals, 
but generally pertain to lyophilization of all products including 
bio-pharmaceuticals. With regard to bulk lyophilization, concerns include 
air classification, aseptic barriers for loading and unloading the unit, partial 
meltback, uneven freezing and heat transfer throughout the powder bed, and the 
additional aseptic manipulations required to break up the large cake. For bulk 
lyophilization, unlike other sterile bulk operations, media challenges can be 
performed, and hence suitable validation studies must be carried out. 
There are also concerns over the spray drying of sterile bio-pharmaceuticals, 
including the sterilization of the spray dryer, the source of air and its quality, the 
chamber temperatures, and the particle residence or contact time. In some 
cases, charring and product degradation have been found for small portions of a 
batch. These should all be assessed during process validation. 
Sterile filtration of products which cannot be sterilized in their final container 
If the product cannot be sterilized in the final container, then solutions or liquids 
must be filtered through a sterile filter of nominal pore size of 0.22 micron 
(or less), or with at least equivalent microorganism retaining properties, into a 
previously sterilized container. Such filters can remove most bacteria and
moulds, but not all viruses or mycoplasmas, so consideration should be given to 
complementing the filtration process with some degree of heat treatment. 
Moreover, if other means of sterilization in the final container were possible, 
then final sterile filtration alone is not considered sufficient. The specification 
for the filters should include information such as its fibre shedding characteristics, 
the criteria used for the selection of the filter, as well as the procedures 
used for integrity testing of the filters. The integrity of the sterilized filter should 
be verified before use, and should be confirmed immediately after use by an 
appropriate method such as a bubble point, diffusive flow, or the pressure hold 
test. The time taken to filter a known volume of bulk solution, the maximum 
filtration pressures and the pressure differential across the filter should also be 
determined during validation, and any significant differences from this should 
be noted and investigated. The same filter should never be used for more than 
one working day unless such use has been validated. If filters were not changed 
after each batch is sterilized, there should be data to justify the integrity of the 
filters for the time periods utilized and prove that grow-through has not 
occurred. 
Terminally sterilized products 
Steam sterilization is the preferred method of those currently available. 
However, before any sterilization process is adopted, its suitability for the 
product and its efficacy in achieving the desired sterilizing conditions in all 
parts of each type of load to be processed should be demonstrated by physical 
measurements and by the use of biological indicators where appropriate. There 
should also be a clear means of differentiating products which have not been 
sterilized from those which have, with each basket, tray or other carrier of 
products or components clearly labelled with the name of the product, its batch 
number, and an indication of whether or not it has been sterilized. Typically, a 
sterile pharmaceutical contains no viable micro-organisms and is nonpyrogenic. 
Parenteral drugs in particular must be non-pyrogenic because the 
presence of pyrogens can cause a febrile reaction in human beings. Pyrogens 
are the products of the growth of microorganisms, so any condition that allows 
microbial growth should be avoided in the manufacturing process. Pyrogens 
may develop in water located in storage tanks, dead legs and pipework, or from 
surface contamination of containers, closures or other equipment, and may also 
contain chemical contaminants that could produce a pyretic response in humans 
or animals even though there may be no pyrogens present. 
Therefore, the procedures used to minimize the hazard of contamination 
with microorganisms and particulates of sterile bio-pharmaceuticals become 
extremely important. Preparation of components and other materials should be
done in at least a grade D environment in order to give low risk of microbial and 
participate contamination, suitable for filtration and sterilization. Where the 
bio-pharmaceutical is at a higher than usual or an unusual risk of microbial 
contamination; for example, because the product actively supports microbial 
growth, or must be held for a long period before sterilization, or needs to be 
processed in other than closed vessels, then all the preparation should be carried 
out in a grade C environment. Filling of a bio-pharmaceutical for terminal 
sterilization should be carried out in at least a grade C environment. Where the 
product is at an unusual risk of contamination from the environment because, 
for example, the filling operation is slow or the containers are wide-necked or is 
necessarily exposed for more than a few seconds before sealing, the filling 
should be done in a grade A zone, with at least a grade C background. 
Finishing of sterile products 
Filled containers of bio-pharmaceuticals should be closed by appropriately 
validated methods. Containers closed by fusion, for example, glass or plastic 
ampoules, should be subject to 100% integrity testing, while those closed by 
other means should be checked for integrity according to appropriate procedures. 
Containers sealed under vacuum should be tested for maintenance of that 
vacuum after an appropriate, pre-determined period. Filled containers should 
be inspected individually for extraneous contamination or other defects, and if 
inspection is done visually, it should be done under suitable and controlled 
conditions of illumination and background. Where other methods of inspection 
are used, the process should be validated and the performance of the equipment 
checked at intervals with the results recorded. 
Some sterile bio-pharmaceuticals may be filled into different types of 
containers, such as sterile plastic bags. For sterile bags, sterilization by 
irradiation is the method of choice because it leaves no residues, although 
some manufacturers use formaldehyde. A major disadvantage is that formaldehyde 
residues may, and frequently do appear in the sterile product. If multiple 
sterile bags are used, operations should be performed in an aseptic processing 
area. Since all the inner bags have to be sterile, outer bags should also be 
applied over the primary bag containing the sterile product in the aseptic 
processing area. One manufacturer was found to apply only the primary bag in 
the aseptic processing area, resulting in the outer portion of this primary bag 
being contaminated when the other bags were applied over this bag in nonsterile 
processing areas! Important validation aspects of the sterile bag system 
include measurement of residues, testing for pinholes, foreign matter (particulates), 
as well as for sterility and endotoxins.

11.4.3 Secondary (sterile) production facility 
Manufacturing operations are divided into two categories — those where the 
product is terminally sterilized and those which are conducted aseptically at 
some or all stages. The design, validation and effective operation of clean 
rooms for the manufacture and testing of pharmaceuticals, biotechnology and 
medical device products is among the most exacting and challenging activities. 
Patient's lives, product integrity, company profitability and regulatory compliance 
all factor into the risks inherent if the clean room is not built right and does 
not function right. The manufacture of sterile products should be carried out in 
clean areas, entry to which should be through airlocks for personnel and/or for 
equipment and materials, and maintained to an appropriate standard of 
cleanliness, and supplied with air that has passed through filters of an 
appropriate efficiency. Adequate space must be provided for the placement 
of equipment and materials to prevent mix-ups for operations such as the 
receiving, sampling, and storage of raw materials; manufacturing, processing, 
packaging and labelling; storage for containers, packaging materials, labelling 
and finished products; as well as for production and control laboratories. 
Facility design features for the aseptic processing of sterile bulk active products 
should include temperature, humidity and pressure control, and there must be 
adequate lighting, ventilation, screening and proper physical barriers for all 
operations including dust, temperature, humidity and microbiological controls, 
with the various operations of component preparation, product preparation and 
filling carried out in separate areas within the clean area. 
Area classification and monitoring of controlled environments 
Clean areas for the manufacture of sterile products are classified according to 
the required characteristics of the environment. Each manufacturing operation 
requires an appropriate level of cleanliness in the operational state, in order to 
minimize the risks of particulate or microbial contamination of the product or 
materials being handled. In order to meet in-operation conditions, these areas 
should be designed to reach certain specified air-cleanliness levels in the at-rest 
occupancy state. The at-rest state is the condition where the installation is 
installed and operating, and is complete with production equipment, but has no 
operating personnel present. The in-operation state is the condition where the 
installation is functioning in the defined operating mode with the specified 
number of personnel. 
For the manufacture of sterile medicinal products there are normally four 
grades of clean areas. The requirement and limit for these areas depend on the 
nature of the operations carried out. Grade A is for the aseptic preparation and 
filling of products, and the local zone for high risk operations such as the filling
zone, stopper bowls, open ampoules and vials, making aseptic connections. 
Normally such conditions are provided by a laminar airflow workstation, which 
should provide a homogeneous air speed of 0.45 m s~! ± 20% (guidance 
value) at the working position. Grade B is for aseptic preparation and filling, 
and the background environment for grade A zone. Grade C is for the 
preparation of solutions to be filtered and the filling of products that are at 
high risk. Grade D is a clean area for carrying out less critical stages in 
the manufacture of sterile products, for the handling of components after 
washing, and for the preparation of solutions and components for subsequent 
filling. 
11.4.4 Safety issues 
Contamination control 
Manufacturing on a campaign basis is typical in the bio-pharmaceuticals 
industry. Whilst this may be efficient with regard to system usage, it can 
present problems when it is discovered in the middle of a campaign that a batch 
is contaminated. Frequently, all the batches processed in a campaign in which a 
contaminated batch is identified are suspect. Such failures should be investigated 
and reported, and the release of any other batches in the campaign should 
be justified. Some of the more significant product recalls have occurred because 
of the failure of a manufacturer to conclusively identify and isolate the source 
of a contaminant. 
When working with dry materials and products, special precautions should 
be taken to prevent the generation and dissemination of dust. This could result 
in the risk of accidental cross-contamination arising from such uncontrolled 
release of dust, gases, vapours, sprays or organisms from materials and 
products in process, from residues on equipment and from operators' clothing. 
The significance of this risk varies with the type of contaminant, and the 
product being contaminated. Amongst the most hazardous contaminants are 
highly sensitizing materials, biological preparations containing living organisms, 
certain hormones, cytotoxics, and other highly active materials. Products 
in which contamination is likely to be most significant are those administered 
by injection and those given in large doses and/or over a long time. 
Environmental control 
Containers and materials liable to generate fibres should be minimized in clean 
areas. All components, containers and equipment should be handled after the 
final cleaning process in such a way that they are not re-contaminated. After 
washing, all components should be handled in at least a grade D environment.
The handling of sterile starting materials and components, unless subjected to 
sterilization or filtration through a micro-organism-retaining filter later in the 
process, should be done in a grade A environment with grade B background. 
However, the handling and filling of aseptically prepared products should be 
done in a grade A environment with a grade B background. The preparation of 
solutions that are to be sterile filtered during the process should be done in a 
grade C environment; however, if not filtered, the preparation of materials and 
products should be done in a grade A environment with a grade B background. 
The preparation and filling of sterile suspensions should be done in a grade A 
environment with a grade B background if the product is exposed and is not 
subsequently filtered. Prior to the completion of stoppering, the transfer of 
partially closed containers, as used in lyophilization (freeze drying) should be 
carried out either in a grade A environment with grade B background, or in 
sealed transfer trays in a grade B environment. 
Prevention of cross-contamination 
In clean areas, and especially when aseptic operations are in progress, all 
activities should be kept to a minimum, and the movement of personnel should 
be controlled and methodical to avoid excessive shedding of particles and 
organisms due to over-vigorous activity. The production of non-medicinal 
products should not be carried out in areas or with equipment destined for the 
final processing of bio-pharmaceuticals. Certainly, operations on different 
products should not be carried out simultaneously, or consecutively in the 
same room, unless there is no risk of mix-up or cross-contamination, and 
preparations of microbiological origin should not be made or filled in areas 
used for the processing of other sterile medicinal products; however, vaccines 
of dead organisms or of bacterial extracts may be filled, after inactivation, 
in the same premises as other sterile medicinal products. Manufacture in 
segregated areas is required for products such as penicillins, live vaccines, 
live bacterial preparations and certain other specified biologicals, or 
manufacture by campaign (separation in time) followed by appropriate 
cleaning. Precautions to minimize contamination should be taken 
during all processing stages including the stages before sterilization. 
These include using closed systems of manufacture, as well as appropriate 
air-locks and air extraction; using cleaning and decontamination procedures 
of known effectiveness, as ineffective cleaning of equipment is 
a common source of cross-contamination; as well as keeping protective 
clothing inside areas where products with special risk of cross-contamination 
are processed.
Control of sterility 
Manufacturers are expected to validate all critical aseptic processing steps in the 
manufacture of bio-pharmaceuticals with at least three consecutive validation 
runs. Such validation must encompass all parts, phases, steps and activities of 
any process where components, fluid pathways or in-process fluids are 
expected to remain sterile. Furthermore, such validation must include all 
probable potentials for loss of sterility as a result of processing and account 
for all potential avenues of microbial ingress associated with the routine use of 
the process. 
Sterility testing 
The sterility test applied to the finished product should only be regarded as the 
last in a series of control measures by which sterility is assured. The test should 
be fully validated for the product(s) concerned with any examples of initial 
sterility test failures thoroughly investigated. In those cases where parametric 
release has been authorized, special attention should be paid to the validation 
and the monitoring of the entire manufacturing process. Samples taken for 
sterility testing should be representative of the whole of the batch, but should in 
particular include samples taken from parts of the batch considered to be most 
at risk of contamination. For example, for products that have been filled 
aseptically, samples should include containers filled at the beginning and at the 
end of the batch, and after any significant intervention. For products that have 
been heat sterilized in their final containers, consideration should be given to 
taking samples from the potentially coolest part of the load. 
Media fill validation 
Validation of aseptic processing should include simulating the process using a 
nutrient medium, the form of which is equivalent to the dosage form of the 
product, although suitable microbiologically-inert non-media alternatives 
would also be acceptable. This process simulation test should imitate as closely 
as possible the routine aseptic manufacturing process and include all the critical 
subsequent manufacturing steps, and should be repeated at defined intervals 
and after any significant modification to the equipment and process. The 
number of containers used for a medium fill should be sufficient to enable a 
valid evaluation. For small batches, the number of containers for the medium 
fill should at least equal the size of the product batch. The contamination rate 
should be less than 0.1% with 95% confidence level, and care should be taken 
that any validation does not compromise the processes, although the limitations 
of 0.1% media fill contamination rate should be recognized for the validation of 
aseptic processing of a non-preserved single dose bio-pharmaceutical, stored at
room temperature as a solution. Any alternative proposals for the validation of 
the aseptic processing of bio-pharmaceuticals may be considered by the 
regulatory authorities, but only on a case-by-case basis. For example, it may 
be acceptable to exclude from the aseptic processing validation procedure 
certain stages of the post-sterilization bulk process that take place in a totally 
closed system. Such closed systems should, however, be Sterilized in Place by a 
validated procedure, integrity tested for each lot, and should not be subject to 
any intrusions whereby there may be the likelihood of microbial ingress. 
Suitable continuous system pressurization would be considered an appropriate 
means for ensuring system integrity. 
Control of pyrogens and endotoxins 
Typically, a sterile pharmaceutical contains no viable microorganisms and is 
non-pyrogenic. Parenteral drugs must be non-pyrogenic because the presence 
of pyrogens can cause a febrile reaction in human beings. As pyrogens are the 
products of the growth of microorganisms, any condition that allows microbial 
growth should be avoided. Parenterals may also contain chemical contaminants 
that could produce a pyretic response in humans or animals, even if there are no 
pyrogens present. Moreover, in addition to pyrogens, microorganisms could 
contaminate the process stream with by-products such as glycosidases and 
proteases, which irreversibly alter or inactivate the product and as a result could 
adversely affect product stability. 
The manufacturing process strategy, therefore, should be oriented more 
towards keeping endotoxins and pyrogens out as well as trying to reduce their 
levels. In some instances, where pipework systems for aqueous solutions have 
been shown to be the source of endotoxin contamination in sterile products, the 
manufacturer should be able to give assurance that there are no 'dead legs' in 
the system. In addition, water sources, water treatment equipment and treated 
water should be monitored regularly for such chemical and biological contamination 
and, as appropriate, for endotoxins. 
Some manufacturers have argued that if an organic solvent is used in the 
manufacture of a sterile product, then the endotoxins levels are reduced at this 
stage. As with any operation, this may or may not be correct, and should be 
proven. For example, one manufacturer who conducted extensive studies using 
organic solvents for the crystallization of a non-sterile pharmaceutical to the 
sterile product observed no change from the initial endotoxin levels. In the 
validating the reduction or removal of endotoxins, challenge studies can be 
carried out on a laboratory or pilot scale to determine the efficiency of the step. 
However, since endotoxins may not be uniformly distributed, it is also important 
to monitor the bioburden of the non-sterile product(s) being sterilized. For
example, gram negative contaminates in a non-sterile bulk drug product prior to 
sterilization are of concern, particularly if the sterilization (nitration) and 
crystallization steps do not reduce the endotoxins to acceptable levels. 
11.4.5 Out of specification 
Regulatory authorities require that suitable process controls be established 
using scientifically sound and appropriate specifications, standards, sampling 
and re-sampling, testing and re-testing. These should be designed to ensure that 
all materials relating to the bio-pharmaceutical manufacture, such as components, 
containers, closures, in-process materials, labelling, including the 
product conform to appropriate standards of identity, strength, quality and 
purity. These controls should be used for the determination of conformity to 
applicable specification, for the acceptance of each batch (or lot) of material 
relating to manufacture, processing, packing, or the holding of the pharmaceutical. 
'Out of specification' is defined as an examination, measurement, or 
test result that does not comply with such pre-established criteria. cGMP 
guidelines require written procedures to be in place to determine the cause 
of any apparent failure, discrepancy, or out of specification result. Out of 
specification results can be caused by laboratory error, non-process or operator 
error, or by process-related error, such as personnel or equipment failures. If, 
however, the result could not be clearly attributed to sampling or laboratory 
error, then there should be scientifically sound procedures and criteria for the 
exclusion of any test data found to be invalid and, if necessary, for any 
additional sampling and testing. 
Re-testing 
Although re-testing may be an appropriate part of the investigation, an 
investigation consisting solely of repeated re-testing is clearly inadequate. If 
quality is not built into a product, re-testing cannot make it conform to 
specifications. The number of re-tests performed before it can be concluded 
that an unexplained out of specification laboratory result is invalid, or that a 
product is unacceptable, is a matter of scientific judgment. There are no 
regulations on specific re-testing procedures, although manufacturers are 
expected to have written investigation and re-testing procedures, applying 
scientifically sound criteria. A variety of written and unwritten practices and 
procedures have been observed, under which manufacturers have disregarded 
out of specification laboratory results after minimal re-testing, re-sampling, 
inappropriate averaging of results or inappropriate testing. Some manufacturers 
then proceeded to release a product without a thorough investigation or an 
adequate justification for disregarding an out of specification result. Regulatory
authorities recognize the distinction between the limited investigation that may 
be necessary to identify a laboratory error and the more extensive investigation 
and testing necessary when out of specification results may be attributed to 
another cause. The manufacturer may impose additional criteria beyond those 
required to ensure identity, strength, quality and purity under cGMP regulations 
or as required for licensure. Although such internal controls are encouraged, 
under some circumstances it is possible to have test results that violate the 
internal standards, without being out of specification, as defined by regulations. 
The investigation should extend to other batches of the same product, and 
other products that may have been associated with the specific failure or 
discrepancy. 
Re-testing for pyrogens and endotoxins 
As with sterility, re-testing for pyrogens or endotoxins can be performed and is 
only acceptable if it is known that the test system was compromized and the 
cause of the initial failure is known, thereby invalidating the original results. It 
cannot be assumed that the initial failure is a false positive without sufficient 
documented justification. Again, any pyrogen or endotoxin test failures, the 
incidence, procedure for handling, and final disposition of the batches involved, 
should be investigated thoroughly, and the reasons for re-testing fully justified. 
Sterility re-testing 
The release of a batch, particularly of a sterile bio-pharmaceutical, which fails 
an initial sterility test and passes a re-test is very difficult to justify. Sterility retesting 
is only acceptable if the cause of the initial non-sterility is known, and 
thereby invalidates the original results. It cannot be assumed that the initial 
sterility test failure is a false positive. This conclusion must be justified by 
sufficient documented investigation, and repeated sampling and testing may not 
identify any low level contamination. Sterility test failures, the incidence, 
procedures for handling, and final disposition of the batches involved should be 
routinely reviewed. 
Reprocessing 
The term reprocessing describes steps taken to ensure that the reprocessed 
batches will conform to all established standards, specifications and characteristics, 
and relates to steps in the manufacturing process that are out of the 
normal manufacturing processing sequence or that are not specifically provided 
for in the manufacturing process. As with the principal manufacturing process, 
reprocessing procedures should be validated. All the data pertaining to the 
reprocessed batches, as well as the data used to validate the process, should be
reviewed and detailed investigation reports, including the description, cause 
and corrective action taken, should be available for the batch. The number and 
frequency of process changes made to a specific process or step can be an 
indicator of a problem experienced in a number of batches. For example, a 
number of changes in a short period of time can be an indicator that that 
particular process step is experiencing problems. 
Rejection 
The demonstration of the adequacy of the process to control other physicochemical 
aspects is an important aspect of validation. Depending upon the 
particular bio-pharmaceutical, these include potency, impurities, particulate 
matter, particle size, solvent residues, moisture content and blend uniformity. 
For example, if the product is a blend of two active products or an active 
product and an excipient, then there should be some discussion and evaluation 
of the process for assuring uniformity. The process validation report for such a 
blend should include documentation for the evaluation and assurance of 
uniformity. Manufacturers occasionally reject the product following the purification 
process or after final processing. As with all pharmaceutical products, it 
is expected that any batch failing specifications is investigated thoroughly, and 
reports of these investigations are complete. For example, during one production 
campaign it was noted that approximately six batches of a bio-pharmaceutical 
product were rejected because of low potency and high levels of 
impurities. The problem was finally attributed to a defective column and, as a 
result, all the batches processed on that particular column were rejected. 
11.5 Design of facilities and equipment 
11.5.1 Facility design 
When designing facilities for bio-pharmaceutical manufacture, the following 
activities should be considered as areas to control contamination: 
the receipt, identification, storage and withholding from use of raw materials 
or process intermediates, pending release for use in manufacturing; as well as 
the quarantine storage of intermediates and final products pending release 
for distribution; 
the holding of rejected raw materials, intermediates and final products before 
final disposition; 
the storage of released raw materials, intermediates and final products; 
manufacturing and processing operations; 
Next Page
reviewed and detailed investigation reports, including the description, cause 
and corrective action taken, should be available for the batch. The number and 
frequency of process changes made to a specific process or step can be an 
indicator of a problem experienced in a number of batches. For example, a 
number of changes in a short period of time can be an indicator that that 
particular process step is experiencing problems. 
Rejection 
The demonstration of the adequacy of the process to control other physicochemical 
aspects is an important aspect of validation. Depending upon the 
particular bio-pharmaceutical, these include potency, impurities, particulate 
matter, particle size, solvent residues, moisture content and blend uniformity. 
For example, if the product is a blend of two active products or an active 
product and an excipient, then there should be some discussion and evaluation 
of the process for assuring uniformity. The process validation report for such a 
blend should include documentation for the evaluation and assurance of 
uniformity. Manufacturers occasionally reject the product following the purification 
process or after final processing. As with all pharmaceutical products, it 
is expected that any batch failing specifications is investigated thoroughly, and 
reports of these investigations are complete. For example, during one production 
campaign it was noted that approximately six batches of a bio-pharmaceutical 
product were rejected because of low potency and high levels of 
impurities. The problem was finally attributed to a defective column and, as a 
result, all the batches processed on that particular column were rejected. 
11.5 Design of facilities and equipment 
11.5.1 Facility design 
When designing facilities for bio-pharmaceutical manufacture, the following 
activities should be considered as areas to control contamination: 
the receipt, identification, storage and withholding from use of raw materials 
or process intermediates, pending release for use in manufacturing; as well as 
the quarantine storage of intermediates and final products pending release 
for distribution; 
the holding of rejected raw materials, intermediates and final products before 
final disposition; 
the storage of released raw materials, intermediates and final products; 
manufacturing and processing operations; 
Previous Page
• packaging and labelling operations; 
• all laboratory operations. 
Control of microbiological, physical, and chemical contamination 
The regulatory authorities require the establishment of, and adherence to, 
written procedures designed to prevent microbiological contamination of 
Pharmaceuticals purporting to be sterile. These requirements also cover such 
procedures as the validation of any sterilization process, and are intended to 
reflect the fact that whether aseptic processing techniques or terminal sterilization 
methods are used, either technique must be validated. Where microbiological 
specifications have been established for the product, then facilities should 
also be designed to limit objectionable microbiological contamination, especially 
if different bio-pharmaceuticals are handled in the same premises and at 
the same time. For the production of the same products, campaign working may 
be acceptable in place of dedicated and self-contained facilities. 
Products can become contaminated with physical or chemical contaminants 
in a variety of ways. For example, ineffective cleaning procedures may leave 
residues of the product or cleaning agents in the equipment; production workers 
may fail to take proper precautions while transporting a substance from one 
area to another thereby introducing a contaminant to the second production 
area; or particles may become airborne and travel to production areas 
throughout the facility. A number of substances such as dust, dirt, debris, 
toxic products, infectious agents, or residue of other drugs or drug components 
can also contaminate products. 
Experience indicates that the potential dangers of contamination are more 
extensive and varied than once believed. For example, adulteration of the 
sterile product with sensitizing substances (such as penicillin, cephalosporins), 
substances having high pharmacological activity or potency (such as steroids, 
cytotoxic anti-cancer agents), infectious agents (such as spore-bearing organisms), 
and products that require viral inactivation or reduction (such as live 
viruses, products from animal cells), may pose health risks to humans or 
animals, even at minimal levels of exposure. Preventing cross-contamination 
of such potentially active substances is the goal and manufacturers are 
expected to identify any such substances posing a serious threat of contamination 
and to control it through dedicated production processes. Moreover, 
because the identity or even the presence of some of these contaminants may 
not be known, health care professionals providing care to a patient suffering 
from such an adverse effect may be unable to provide appropriate medical 
intervention.
Most contamination, however, can be controlled to an acceptable level 
through measures such as proper planning and implementation of cleaning 
and sanitation processes, employee training, gowning, and air filtration. cGMP 
guidelines require that manufacturers set contamination limits on a substanceby-
substance basis, according to both the potency of the substance and the 
overall level of sensitivity to that substance, and prohibit the release of the 
product for distribution if these limits were exceeded. Depending on the product, 
a variety of measures may be acceptable to eliminate cross-contamination; there 
may, however, be situations where nothing short of dedicated facilities, air 
handling and process equipment would be sufficient, especially if there are no 
reasonable methods for the cleaning and removal of a substance or compound 
residues from buildings, facilities and equipment. For example, a manufacturer 
might develop a hypothetical product of high therapeutic potential that also 
poses a high risk of contamination and if it posed a special danger to human 
health, dedicated facilities would be required. If, however, experience demonstrated 
that the product did not pose such a risk, or if changes in manufacturing 
technology greatly reduced the risk, then dedicated facilities might no longer be 
required. 
Sanitation 
The sanitation of clean areas is particularly important. Any building used in the 
manufacture, processing, packing or holding of bio-pharmaceuticals and their 
intermediates should be maintained in a clean and sanitary condition. Sanitation 
procedures should apply to work performed by contractors or temporary 
employees as well as work performed by full-time employees during the 
ordinary course of operations. Written procedures should, therefore, be 
established, assigning responsibility for sanitation, and describing the cleaning 
schedules, methods, equipment and materials to be used in cleaning buildings 
and facilities, and for the use of suitable rodenticides, insecticides, fungicides, 
fumigating agents, or other cleaning and sanitizing agents to prevent the 
contamination of equipment, raw materials, packaging and labelling materials, 
as well as the final product. Where disinfectants are used, more than one type 
should be employed, and monitoring should be undertaken regularly to detect 
the development of resistant strains. Disinfectants and detergents should be 
monitored for microbial contamination, and those used in grades A and B areas 
especially should be sterile prior to use. 
Monitoring programmes in controlled environments 
It is the responsibility of the manufacturer to develop, initiate and implement an 
environmental monitoring programme tailored to specific facilities and condi-
tions and capable of detecting any adverse drift in microbiological conditions in 
a timely manner, allowing meaningful and effective corrective action. Such 
microbiological monitoring programmes should be utilized to assess the 
effectiveness of cleaning and sanitization practices and of personnel that 
could have an impact on the bioburden of the controlled environment. Routine 
microbial monitoring, regardless of how sophisticated the system may be, will 
not and need not identify and quantify all microbial contaminants present in the 
controlled environment. It can only provide information to demonstrate that the 
environmental control systems are operating as intended. The objective of 
microbial monitoring is, therefore, to obtain representative estimates of 
bioburden in the environment. 
The environmental monitoring programme for the manufacture of sterile 
bio-pharmaceuticals should include the daily use of surface plates and the 
monitoring of personnel, with alert or action limits established, and appropriate 
follow-up corrective action taken when they are reached. Where critical 
aseptic operations are performed, monitoring should be frequent using 
methods such as settle plates, volumetric air and surface sampling (such as 
swabs and contact plates). Additional microbiological monitoring is also 
required outside production operations, for example, after validation of 
systems, cleaning and sanitization. The particulate conditions for the at-rest 
state should be achieved in the unmanned state after a short clean-up period of 
about 15-20 minutes (guidance value) after completion of operations. The 
particulate conditions for grade A in operation should be maintained in the 
zone immediately surrounding the product, whenever the product or open 
container is exposed to the environment. It may not always be possible to 
demonstrate conformity with particulate standards at the point of fill when 
filling is in progress due to the generation of particles or droplets from the 
product itself. 
Some manufacturers utilize UV lights in operating areas. Such lights are of 
limited value as they may mask a contaminant on a settle or aerobic plate or 
may even contribute to the generation of a resistant (flora) organism. Therefore, 
the use of surface contact plates is preferred, as they will provide more 
information on levels of contamination. There are some manufacturers that 
set alert/action levels on averages of plates. For the sampling of critical 
surfaces, such as operators' gloves, the average of results on plates is 
unacceptable. The primary concern is any incidence of objectionable levels 
of contamination that may result in a non-sterile product. Since processing is 
commonly carried out around the clock, monitoring of surfaces and personnel 
during the second and third shifts should also be routine.
In the management of a sterile operation, periodic (weekly/monthly/ 
quarterly) summary reports of environmental monitoring should be generated. 
Trained personnel should evaluate any trends when data are compiled and 
analysed. While it is important to review environmental results on a daily basis, 
it is also critical to review results over extended periods to determine whether 
trends are present, as they may be related to decontamination procedures, 
housekeeping practices, personnel training, cross-contamination and the potential 
for microbial build up during production. A full investigation should, 
therefore, include a review of area maintenance documentation, sanitization 
documentation, the inherent physical or operational parameters, and the 
training status of personnel involved, while a limited investigation triggered 
by an isolated, small excursion might include only some of these areas. Based 
on the review of the investigation and testing results, the significance of the 
event and the acceptability of the operations or products processed under that 
condition can be ascertained. Any investigation and the rationale for the course 
of action should be documented and included as part of the overall quality 
management system. 
11.5.2 Laboratory design 
The design of a laboratory that handles any bio-pharmaceutical, which may 
include infectious agents, should provide secondary containment to protect the 
people as well as the environment outside the laboratory from exposure to any 
infectious materials. Laboratory design should take into account the nature of 
the material being handled, the process step or study being planned for 
investigation, and the degree of biosafety necessary. They must be sufficient 
to enable the proper conduct of the study and must provide appropriate space, 
environmental conditions, containment, decontamination areas and support 
systems, such as air and water, for the study being conducted. 
There are three types of laboratory designs that provide four different levels 
of containment. They all consist of three elements: laboratory practices and 
techniques, safety equipment, and laboratory facilities. The first two elements 
are considered primary containment, since they provide protection within the 
laboratory to personnel and the immediate environment. The third element, the 
design of the laboratory itself, is considered secondary containment since it 
protects persons and the environment outside of the facility. Changes in vendor 
and/or the specifications of major equipment and reagents would require 
re-validation. Each laboratory should have documentation and schedules for 
the maintenance, calibration and monitoring of all laboratory equipment 
involved in the measurement, testing and storage of raw materials, product, 
samples, and reference reagents, and more importantly the laboratory personnel
should be adequately trained for the jobs they are performing. Important 
characteristics of each of the biosafety levels are summarized below. 
Basic laboratory 
These are appropriate for Biosafety levels 1 and 2. They are used for studies 
where there is a minimum level of hazard, the personnel are able to achieve 
sufficient protection from the implementation of standard laboratory practices, 
and the organisms used in the study are not associated with any diseases in 
healthy adults. 
Biosafety level 1 
The organisms involved are defined and characterized strains, which are of 
minimal hazard and are not known to cause disease in healthy human adults. 
Although access to the laboratory may be restricted, the facility is generally not 
closed off from the rest of the building. The laboratory is designed to facilitate 
cleaning, with space between equipment and cabinets, and bench tops that are 
impervious to water and resistant to solutions. Personnel should be knowledgeable 
in all laboratory procedures and supervised by a scientist trained in 
microbiology or a related science. Most work is conducted on open bench tops, 
with procedures performed in a manner that limits the creation of aerosols, and 
special containment equipment is not usually needed. Decontamination of work 
surfaces should be done daily and after spills, and all contaminated wastes 
should be decontaminated before disposal. Each laboratory has a hand-washing 
sink. Personal safety equipment, such as laboratory coats or uniforms, should 
be worn and hands washed before and after handling viable materials. Any 
contaminated materials that will be decontaminated at another location should 
be transported in a durable leak proof container that is sealed before removal 
from the area. 
Biosafety level 2 
Work done under Biosafety level 2 involves organisms of moderate potential 
hazard. Many of the characteristics of this level are the same as those for 
Biosafety level 1. However, for Biosafety level 2, laboratory access is limited 
while work is being conducted, and only persons informed of the potential 
hazards of the environment and who meet any other entry restrictions developed 
by the organization should be allowed entry. Biological safety cabinets 
(Class I or II) should be used for containment when procedures with a high 
potential for creating infectious aerosols such as centrifugation or blending are 
conducted or when high concentrations or large volumes of infectious agents 
are used. An autoclave should be available for use in decontaminating
infectious wastes. Personnel should be trained in handling pathogenic agents 
and be under the direction of skilled scientists. Before leaving the area, 
personnel should either remove any protective clothing and leave it in the 
laboratory, or cover it with a clean coat. Skin contamination with infectious 
materials should be avoided and gloves worn when such contact is unavoidable. 
Spills and accidents causing overt exposure to infectious materials should be 
reported promptly with appropriate treatment provided and records of the 
incident maintained. If warranted by the organisms at use in the laboratory, 
baseline serum samples for all at-risk personnel should be collected and stored. 
Containment laboratory 
Containment laboratories qualify as Biosafety level 3 facilities and are designed 
with protective features to allow for the handling of hazardous materials in a 
way that prevents harm to the laboratory personnel, as well as the surrounding 
persons and environment. These may be freestanding buildings or segregated 
portions of larger buildings, as long as they are separated from public areas by a 
controlled access zone. Containment laboratories also have a specialized 
ventilation system to regulate airflow. 
Biosafety level 3 
Work done under Biosafety level 3 conditions can occur in clinical, diagnostic, 
teaching, research or production facilities, and involves organisms that may 
cause serious or potentially lethal disease following exposure through inhalation. 
The laboratory is, therefore, segregated from general access areas of the 
building, and two sets of self-closing doors must be passed through to enter the 
laboratory from access hallways. Access should be limited to persons who must 
be present for programme or support functions, and the doors remain closed 
during experiments. Protective clothing should be worn in the laboratory and 
removed before exiting the facility, and all such clothing should be decontaminated 
before laundering. All work with infectious materials should be 
conducted in a biosafety cabinet (Class I, II or III) or other physical containment 
device, or by personnel wearing the necessary personal protection 
clothing. Upon completing work with infectious materials, all work surfaces 
should be decontaminated. Walls, ceilings and floors should be water-resistant 
to facilitate cleaning, and windows should be closed and sealed. The laboratory 
sinks should be operable by foot, elbow or automation, and be located near the 
exit of each laboratory area. Vacuum lines should be protected with high 
efficiency particulate air (HEPA) filters and liquid disinfectant traps. The 
HEPA-filtered exhaust air from Class I or II biosafety cabinets may be 
discharged directly to the outside, or through the building exhaust system, or
be recirculated within the laboratory if the cabinet is appropriately certified and 
tested. 
Maximum containment laboratory 
These laboratories are Biosafety level 4 facilities. Maximum containment 
laboratories are designed to provide a safe environment for carrying out studies 
involving infectious agents that pose an extreme hazard to laboratory personnel, 
or may cause serious epidemic disease. These facilities have secondary 
barriers, including sealed openings into the laboratory, air locks, a double door 
autoclave, a separate ventilation system, a biowaste treatment system, and a 
room for clothing change and showers that adjoins the laboratory. 
Biosafety level 4 
This safety level is necessary for work with organisms that present a high 
individual risk of life-threatening disease. These facilities are usually located in 
an independent building, or in a separate, isolated, completely segregated, 
controlled area of a larger building. Access to the facility should be controlled 
by the use of locked doors. All personnel entering should sign a logbook, must 
enter and leave the facility through the clothing change and shower rooms, and 
must shower before exiting. Any supplies or materials that do not enter through 
the shower and change rooms must enter through a double door autoclave, 
fumigation chamber, or airlock that is decontaminated between each use. All 
organisms classified as Biosafety level 4 should be handled in Class III 
biosafety cabinets, or in Class I or II biosafety cabinets used in conjunction 
with one-piece positive pressure personnel suits ventilated by a life support 
system. All biological materials removed from a Class III cabinet, or the 
maximum containment laboratory in a viable condition, should be placed in a 
non-breakable, sealed primary container and enclosed in a secondary container 
that is removed through a disinfectant dunk tank, fumigation chamber, or 
airlock. All other materials must be autoclaved or decontaminated before 
removal from the facility. Walls, floors and ceilings of the facility together 
should form a sealed internal shell, with any windows resistant to breakage. 
Most importantly, the facility should be available for the quarantine isolation 
and treatment of personnel with potential or known laboratory-related illnesses. 
11.5.3 Equipment design 
The types of equipment commonly used in a bio-pharmaceutical facility will 
vary based not only on the types of processes and organisms used, but also on 
whether the equipment is used during development, during testing, or during 
manufacture of material for clinical trials and marketing. Types of equipment
commonly used include bioreactors, air compressors, sterilization equipment, 
product recovery systems such as centrifuges and cell disrupters, waste 
recovery and decontamination equipment, sampling and analysis instruments, 
safety equipment such as biosafety cabinets and protective clothing, equipment 
for transporting biological materials such as sealed containers, and environmental 
control equipment. 
Equipment capacity and location 
As always, the equipment used in the manufacture, processing, packing or 
holding of the bio-pharmaceutical product or any of the process intermediates 
should be of appropriate design, adequate size and construction, and suitably 
located to facilitate operations for its intended use and for its cleaning and 
maintenance. Closed equipment should be used when feasible to provide 
adequate protection of the bulk-active and any intermediates, and always in 
the case of sterile products. When equipment is opened or open equipment is 
used, appropriate precautions should be taken to prevent contamination or 
cross-contamination of bulk active substance and intermediates. New equipment 
must be properly installed and operate as designed, and must be cleaned 
before use according to written procedures, with the cleaning procedures 
documented and validated. 
Equipment construction and installation 
Equipment should be constructed and installed, to enable easy cleaning, 
adjustments and maintenance. Equipment should be constructed so that 
surfaces that come into contact with raw materials, intermediates, bulk active 
substances or sterile products, are not reactive, additive, or absorptive, so as to 
alter the quality, purity, identity, or strength of the product beyond the 
established specifications. Similarly, any substances required for the operation 
of equipment, such as lubricants, heating fluids or coolants, should not contact 
raw materials, packaging materials, intermediates, or the bulk active, so as to 
alter its quality and purity beyond established specifications. If the equipment 
requires calibration, there must written procedures for calibrating the equipment 
and documenting the calibration. With filters, the type of filter, its 
purpose, how it is assembled, cleaned, and inspected for damage, and if a 
microbial retentive filter, methods used for integrity testing, should be specified. 
Qualification of equipment should ensure that it is installed according 
to approved design specifications, regulatory codes, and the equipment 
manufacturers' recommendations, and that it operates within the limits and 
tolerances established for the process.
Biosafety cabinets 
These are common primary containment devices for work involving infectious 
organisms. Their primary function is to protect the laboratory worker and the 
immediate environment by containing any infectious aerosols produced during 
the manipulation of organisms within the cabinet. Biosafety cabinets are 
classified into three types (I, II and III) based on their performance characteristics. 
Class I and II cabinets are appropriate for use with moderate and highrisk 
micro-organisms. They have an inward face velocity of 75 linear feet per 
minute and their exhaust air is filtered by HEPA filters. They can be used with a 
full width open front, an installed front closure panel, or an installed front 
closure panel equipped with arm-length rubber gloves. The Class II cabinet is a 
vertical laminar-flow cabinet with an open front. In addition to the protection 
provided by the Class I cabinet, these cabinets also protect materials inside the 
cabinet from extraneous airborne contaminants since the HEPA filtered air is 
recirculated within the workspace. The Class III cabinet is a totally enclosed, 
ventilated, gas tight cabinet used for work with infectious organisms. Work in a 
Class III cabinet is conducted through connected rubber gloves. The cabinet is 
maintained under negative pressure with supply air drawn in through HEPA 
filters, and exhaust air filtered by two HEPA filters and discharged to outside the 
facility using an exhaust fan that is generally separate from the facility's overall 
exhaust fan. However, it is important to remember that each of the cabinet types 
is only protective if it is operated and maintained properly by trained personnel. 
Organism preparation 
Other commonly used laboratory equipment in a biotechnology laboratory or 
facility includes culture plates, roller bottles, shake flasks, and a seed fermenter. 
These are used to bring the organism or the cell line from its origination in the 
master cell bank through its preparation for growth and/or propagation. 
Bioreactors or fermenters 
Fermenters or bioreactors play a central role in biotechnological processes, with 
their main purpose being to grow and/or propagate a microorganism or a cell 
line in a controlled, aseptic environment. The most popular type is the 
mechanical fermenter, which uses mechanical stirrers to agitate the culture, 
and one of the most commonly used mechanical fermenters is the stirred tank 
reactor. In order to satisfy the metabolic requirements of the microorganism or 
the cell line, aeration must be adequate to provide sufficient oxygen, and those 
using agitation need to be designed to maintain a uniform environment within 
the bioreactor. Major attributes of a good bioreactor are that it should be 
economical, robust, of simple mechanical design, easy to operate under aseptic
conditions, of reasonably flexible design with respect to the various process 
requirements, with no dead zones giving good control to bulk flow, and have 
good heat and mass transfer. 
The level of sophistication involved in the design of a fermenter is largely a 
function of the requirements of the process. Stainless steel is commonly chosen 
as the material of construction for the fermenter, as it can withstand repeated 
cycles of sterilization (1210C for at least 30min) without breakage and has 
better heat transfer than glass. Other sterility considerations include smooth and 
crevice free welded joints; short, straight pipework with appropriate slopes to 
avoid accumulation of pockets of liquid during operation; all wetted internals 
polished to 180-200 grit finish, and all other materials used amenable to steam 
sterilization. 
There should be adequate monitoring and control equipment to control the 
metabolic processes, by monitoring parameters such as pH, temperature, 
agitation, and aeration rates within the bioreactor. For off-line systems, a 
sample is taken from the bioreactor at specified intervals and chemically 
analysed using automated laboratory instruments — these can have a lengthy 
turnaround time for analytical results and do not provide a high level of 
containment. For on-line systems, sampling and analysis are done continuously, 
often requiring additional secondary containment. In-line or at-line systems, 
however, provide a continuous, non-invasive indication of bioreactor conditions, 
through the use of probes, sensors, and sampling devices that directly 
contact the material. 
Temperature within the fermenter is maintained by circulating water at a 
controlled temperature through the jacket of the fermenter, which envelops the 
complete level of liquid in the shell. Baffle plates are provided inside the jacket 
for effective circulation of the cooling or heating medium in the jacket, with a 
drain port provided at the bottom for efficient removal of condensate at the end 
of sterilization, and a vent at the top of the jacket. Bioreactor aeration system is 
designed for supplying sterile moisture-free air rate at 0-3 vvm (volume of air 
per volume of liquid per minute), although an aeration rate of 0.2-0.3 vvm is 
commonly used. Medical air (compressed air) at 1.5 bar g, from which moisture 
and oil vapours are stripped, is supplied from an air compressor, passed through 
a pressure regulator, flowmeter and a steam sterilizable air filter to remove 
undesirable organisms and particles from the air. This sterile filtered air is 
sparged into the fermenter through the sparger, which usually consists of an 
open-ended stainless steel pipe discharging directly under the agitator. The 
fermenter requires a versatile agitation system to ensure optimal mixing at low 
shear. The agitator port is sealed, either with a double mechanical seal with a 
sterile condensate lubrication system, or a magnetically coupled seal system.
The seal assembly is selected primarily with consideration of the cell line used, 
the heavy wear and tear and the repeated sterilization cycle the system 
undergoes. The main elements of the agitation system consist of the baffles 
on the shell wall for breaking vortex during peak agitation and impellers with 
adjustable height on the vertical shaft. 
Product recovery 
A product recovery or purification system is required to separate and concentrate 
the desired product from the contents of the bioreactor. Such systems 
include centrifugation, cell disruption, broth conditioning, filtration, extraction, 
chromatography, and drying and freezing techniques — the type of equipment 
depending on the type(s) of product handled. 
Centrifuges are used to separate viable cells from liquid culture broth 
and include batch-operated solid bowl machines, semi-continuous solidsdischarging 
disc separators, or continuous decanter centrifuges. Batch centrifuges 
include the solid-bowl disc centrifuge, one-chamber centrifuges (used for 
protein fractionation from blood plasma), zonal centrifuges (used to separate 
intracellular and extra-cellular products such as in virus purification or cell 
constituent isolation), and tubular centrifuges (used to separate liquid phases). 
Biosafety cabinets must be used during solids removal from batch centrifuges. 
Semi-continuous solids-discharging machines generally provide the best 
containment and are the most widely used type for biotechnology applications. 
Filtration units are also used to separate cellular, intra-cellular or extra-cellular, 
solids from broth. Types of filtration units include continuous rotary drums, 
continuous rotary vacuum filters or tangential flow filtration systems using 
either microporous or ultrafiltration membrane filters. The type of filtration unit 
used depends on the type of product being recovered. 
Cell disruption is used to recover intra-cellular products and can be 
performed using mechanical or non-mechanical methods. Mechanical methods 
include ball mills and high-speed homogenizers, whilst non-mechanical 
methods include chemical or enzymatic lysis, heat treatment, freeze-thaw or 
osmotic shock. Non-mechanical methods are easily contained and are most 
often used in biotechnology laboratories. Chromatography processes such as 
affinity or gel filtration are used to purify intra-cellular or extra-cellular 
products, using an eluting solvent in a packed column and collected in a 
fraction collector. If adequate containment is provided, such as a biological 
safety cabinet, product recovery using chromatography can be used to purify 
hazardous organisms. Other purification equipment includes centrifugal extractors 
(used for liquid-liquid extraction), spray packed, mechanically agitated, or 
pulsed columns. Either freezing or drying may be used to facilitate the handling
and storage of products. Organisms to be frozen are placed in vials and frozen. 
The most common types of dryers used are freeze dryers and vacuum tray 
dryers, and since freezing provides primary containment and produces less 
aerosols than dryers, it is more appropriate for product storage. If drying is 
performed, proper filtration and ventilation systems must be provided. 
Isolator technology 
The use of isolator technology to minimize human interventions in processing 
areas usually results in a significant decrease in the risk of microbiological 
contamination of aseptically manufactured products from the environment. 
There are many possible designs of isolators and transfer devices. The isolator 
and the background environment should be designed so that the required air 
quality for the respective zones can be realized. The air classification required 
for the background environment depends on the design of the isolator and its 
application and for aseptic processing it should be at least grade D. In general, 
the area inside the isolator is the local zone for high-risk manipulations, 
although it is recognized that laminar airflow may not exist in the working 
zone of all such devices. The transfer of materials into and out of the unit is one 
of the greatest potential sources of contamination. Such transfer devices may 
vary from a single door to double door designs to fully sealed systems 
incorporating sterilization mechanisms. Isolators should be introduced only 
after appropriate validation. Validation should take into account all critical 
factors of isolator technology, such as the quality of the air inside and outside 
(background) the isolator, sanitization of the isolator, the transfer process and 
isolator integrity. Isolators are constructed of various materials more or less 
prone to puncture and leakage. Monitoring should be carried out routinely and 
should include frequent leak testing of the isolator and glove/sleeve system. 
Computer and related automatic and electronic systems 
These are used in the control of critical manufacturing steps in bio-pharmaceutical 
manufacture. They should be appropriately qualified and validated to 
demonstrate the suitability of the hardware and software, to perform assigned 
tasks in a consistent and reproducible manner. The depth and scope of the 
validation programme would depend on the diversity, complexity and criticality 
of the system. All changes should be approved in advance and performed by 
authorized and competent personnel, and records kept of all changes, including 
modifications and enhancements to the hardware, software and any other 
critical components of the system, to demonstrate that the modified system is 
maintained in a validated state.
Appropriate controls over computer or related automatic and electronic 
systems should be exercised to ensure that only authorized personnel make 
changes in master production and control records. Procedures should be 
established to prevent unauthorized entries or changes to existing data. Systems 
should identify and document the persons entering or verifying critical data. 
Input to and output from the computer or related system should be checked for 
accuracy at appropriate intervals and where critical data are entered manually, 
there should be an additional check on the accuracy of the entry. This may be 
performed by a second operator, or by the system itself. 
A back-up system should be available to respond to system breakdowns or 
failures that result in permanent loss of critical records. Back-ups may consist of 
hard copies or other forms, such as tapes or microfilm, that ensure back-up data 
are exact, complete and secure from alteration, inadvertent erasure or loss. The 
current regulations also require that a 'back-up file of data entered into the 
computer or related system shall be maintained except where certain data, such as 
calculations performed in connection with laboratory analysis, are eliminated 
by computerization or other automated processes'. If computerization or 
another automated process has eliminated such calculations 'then a written 
record of the programme shall be maintained along with data establishing proper 
performance' emphasizing that the manufacturer must actually establish proper 
performance. 
Regulatory authorities require additional information to be available for 
pre-approval inspection. The information provided should include a brief 
description of procedures for changes to the computer system. For each of 
the systems, a list of the manufacturing steps that are computer-controlled 
should be provided, together with the identity of the system's developer 
(i.e. developed in-house or by an external contractor). The validation summary 
should include: 
• a narrative description of the validation process (or protocol), including 
acceptance criteria; 
• certification that IQ and OQ have been completed; 
• an explanation of the parameters monitored and tests performed; 
• a validation data summary; 
• an explanation of all excursions or failures; 
• deviation reports and results of investigations for all excursions or failures. 
11.5.4 Sterilization methods 
All the equipment used in the processing of bio-pharmaceuticals should be 
capable of being sterilized and maintaining sterility. Sanitization rather than
sterilization of critical equipment such as crystallizers, centrifuges, filters, spray 
and freeze dryers is totally unacceptable. All sterilization processes should be 
validated, with particular attention given when the adopted sterilization method 
is not described in the current edition of the Pharmacopoeia, or when it is used 
for a product that is not a simple aqueous or oily solution. Where possible, heat 
sterilization is the method of choice. 
Biological indicators 
If biological indicators are used, strict precautions should be taken to avoid 
transferring microbial contamination from them. In some cases, testing of 
biological indicators may become all or part of the sterility testing. Various 
types of indicators are used as an additional method for monitoring the 
sterilization and assuring sterility, including lag thermometers, peak controls, 
Steam Klox, test cultures and biological indicators. Biological indicators are of 
two forms, each of which incorporates a viable culture of a single species of 
microorganism. In one form, the culture is added to representative units of the 
lot to be sterilized, or to a simulated product that offers no less resistance to 
sterilization than the product to be sterilized. In the second form, the culture is 
added to disks or strips of filter paper, metal, glass or plastic beads, and used 
when the first form is not practical, as is the case with solids. If using indicators, 
there should be assurances that the organisms are handled so they do not 
contaminate the manufacturing area or the product, and they should be stored 
and used according to the manufacturer's instructions, and their quality checked 
by positive controls. 
Sterilization by moist heat 
The method of choice for the sterilization of equipment and transfer lines is 
saturated clean steam under pressure. In the validation of the sterilization of 
equipment and transfer systems, temperature sensors and biological indicators 
should be strategically located in cold spots where condensate may accumulate, 
such as the point of steam injection and steam discharge, and in low spots such 
as the exhaust line. Steam must expel all the air from the sterilizer chamber to 
eliminate cold spots, and from the drain lines connected to the sewer by means 
of an air break to prevent back siphoning. After the high temperature phase of a 
heat sterilization cycle, precautions should be taken against contamination of a 
sterilized load during cooling. There should be frequent leak tests on the 
chamber when a vacuum phase is part of the cycle. One manufacturer utilized a 
steam-in-place system, but only monitored the temperature at the point of 
discharge and not in low spots in the system where condensate accumulated and 
caused problems. Care should be taken to ensure that steam used for steriliza-
tion is of suitable quality and does not contain additives at a level that could 
cause contamination of product or equipment. Any cooling fluid or gas in 
contact with the product should be sterilized unless it can be shown that any 
leaking container would not be approved for use. 
Both temperature and pressure should be used to monitor the process. 
Control instrumentation should normally be independent of monitoring instrumentation 
and recording charts. Where automated control and monitoring 
systems are used, they should be validated to ensure that critical process 
requirements are met. Each heat sterilization cycle should be recorded on a 
time/temperature chart with a sufficiently large scale, or by other appropriate 
equipment with suitable accuracy and precision. The position of the temperature 
probes used for controlling and recording should be determined during the 
validation, and where applicable checked against a second independent 
temperature probe located at the same position. Chemical or biological 
indicators may also be used, but should not take the place of physical 
measurements. The time required to heat the centre of the largest container 
to the desired temperature must be known, and sufficient time must be allowed 
for the whole of the load to reach the required temperature before measurement 
of the sterilizing time-period is commenced. Charts of time, temperature and 
pressure should be filed for each sterilizer load. The items to be sterilized, other 
than products in sealed containers, should be wrapped in a material which 
allows removal of air and penetration of steam but which prevents recontamination 
after sterilization. 
Sterilization by dry heat 
There are some manufacturers who sterilize processed bulk bio-pharmaceutical 
powders by the use of dry heat. As a primary means of sterilization, its 
usefulness is questionable because of the lack of assurance of penetration into 
the crystal core of a sterile powder, although some sterile bulk powders can 
withstand the lengthy times and high temperatures necessary for dry heat 
sterilization. Process validation should cover aspects of heat penetration and 
heat distribution, times, temperatures, stability (in relation to the amount of heat 
received) and particulates. Any air admitted to maintain a positive pressure 
within the chamber should be passed through a HEPA filter. Where this process 
is also intended to remove pyrogens, challenge tests using endotoxins should be 
used as part of the validation. 
Sterilization by radiation 
Radiation sterilization is used mainly for the sterilization of heat sensitive 
materials and products, although ultra-violet irradiation is not normally an
acceptable method of sterilization. Many medicinal products and some packaging 
materials are radiation-sensitive, so this method is permissible only when 
the absence of deleterious effects on the product has been confirmed experimentally. 
Validation procedures should ensure that the effects of variations in 
density of the packages are considered, and biological indicators may be used as 
an additional control. Materials handling procedures such as the use of 
radiation sensitive colour disks should also be used on each package to 
differentiate between irradiated and non-irradiated materials and prevent mixups. 
During the sterilization procedure the radiation dose should be measured, 
and the total radiation dose should be administered within a predetermined time 
span. For this purpose, dosimetry indicators that are independent of dose rate 
should be used, giving a quantitative measurement of the dose received by the 
product itself. These should be inserted in the load in sufficient numbers and 
close enough together to ensure that there is always a dosimeter in the irradiator. 
Where plastic dosimeters are used they should be used within the time limit of 
their calibration, and dosimeter absorbances should be read within a short 
period after exposure to radiation. 
Sterilization with ethylene oxide 
There are some manufacturers who still use ethylene oxide for the surface 
sterilization of powders as a precaution against potential microbiological 
contamination during aseptic handling, even though a substantial part of the 
sterile pharmaceutical industry has discontinued its use as a sterilizing agent. Its 
use is now in decline because of residual ethylene oxide in the product and the 
inability to validate ethylene oxide sterilization, as well as employee safety 
considerations. As a primary means of sterilization, its use is questionable 
because of the lack of assurance of penetration into the crystal core of a sterile 
powder, and therefore, this method should only be used when no other method 
is practicable. Process validation should show that there is no damaging effect 
on the product and that the conditions and time allowed for degassing are such 
as to reduce any residual gas and reaction products to acceptable limits for the 
type of product or material. The nature and quantity of packaging materials can 
significantly affect the process, so materials should be pre-conditioned by being 
brought into equilibrium with the humidity and temperature required by the 
process before exposure to the gas. The time required for this should be 
balanced against the opposing need to minimize the time before sterilization. 
For each sterilization cycle, records should be made of the time taken to 
complete the cycle, of the pressure, temperature and humidity within the 
chamber during the process, the gas concentration, and the total amount of gas 
used. After sterilization, the load should be stored in a controlled manner under
ventilated conditions to allow residual gas and reaction products to reduce to 
the defined level. 
Sterilization with formaldehyde 
The use of formaldehyde is a much less desirable method of equipment 
sterilization. A major problem with formaldehyde is its removal from pipework 
and surfaces and it is rarely used primarily because of residue levels in both the 
environment and the product. Since formaldehyde contamination in a system or 
in a product is not going to be uniform, merely testing the product as a means of 
demonstrating and validating the absence of formaldehyde levels is not 
acceptable; there should be some direct measure, or determination of the 
absence of formaldehyde. Key surfaces should be sampled directly for residual 
formaldehyde. One large pharmaceutical manufacturer had to reject the initial 
batches coming through the system because of formaldehyde contamination. 
Unfortunately, they relied on end product testing of the product, and not on 
direct sampling to determine the absence of formaldehyde residues on equipment. 
Sterilization In Place (SIP) 
SIP systems require considerable maintenance, and their malfunction has 
directly led to considerable product contamination and recall. One potential 
problem with SIP systems is condensate removal from the environment. 
Condensate and excessive moisture can result in increased humidity, and 
increases in levels of microorganisms on surfaces of equipment. Therefore, 
environmental monitoring after sterilization of the system is particularly 
important. Another potential problem is the corrosive nature of the sterilant, 
whether it is clean steam, formaldehyde, peroxide or ethylene oxide. In two 
recent cases, inadequate operating procedures have led to weld failures. 
Therefore, particular attention should be given to equipment maintenance 
logs, especially to non-scheduled equipment maintenance, and the possible 
impact on product quality. Suspect batches manufactured and released prior to 
the repair of the equipment should be identified. 
11.5.5 Cleaning procedures and validation 
Regulatory authorities requiring that all equipment and facilities be clean prior 
to use and be maintained in a clean and orderly manner, are nothing new. Of 
course, the main rationale for requiring clean equipment and facility is to 
prevent contamination or adulteration of medicinal products. Historically, 
authorities have looked for gross insanitation due to inadequate cleaning and 
maintenance of equipment and/or poor dust control systems, and were more
concerned about the contamination of non-penicillin drug products with 
penicillins, or the cross-contamination of drug products with potent steroids 
or hormones. Certainly, a number of products have been recalled over the past 
decade due to actual or potential penicillin cross-contamination. 
Rationale and procedures 
Cleaning, and its validation, including facility disinfection, personnel control 
and equipment cleaning, has recently come under increasing scrutiny. Numerous 
regulatory actions and comments have been issued, resulting in many 
questions regarding the selection, use, testing, documentation and validation of 
cGMP sanitation programmes. Regulatory authorities now expect manufacturers 
to have written procedures detailing the cleaning processes used for 
various pieces of equipment. If manufacturers have only one cleaning process 
for cleaning between different batches of the same product, and use a different 
process for cleaning between product changes, then the written procedures 
should address these different scenarios. Similarly, if manufacturers have one 
process for removing water-soluble residues and another process for non-water 
soluble residues, the written procedure should address both scenarios and make 
it clear when a given procedure would be followed. Some manufacturers may 
decide to dedicate certain equipment for certain process steps that produce 
residues that are difficult to remove from the equipment. Any residues from the 
cleaning process itself, such as detergents and solvents, also have to be removed 
from the equipment. 
Equipment should be cleaned, held and, where necessary, sanitized at 
appropriate intervals to prevent contamination or cross-contamination that 
would alter the quality or purity of the product beyond the established 
specifications. Even dedicated equipment should be cleaned at appropriate 
intervals to prevent the build-up of objectionable material or microbial growth. 
As processing approaches the purified bulk active substance, it becomes 
important to ensure that incidental carry-over of contaminants or degradants 
between batches does not adversely impact the established impurity profile. 
However, this does not always apply to a bio-pharmaceutical, where many of 
the processing steps are accomplished aseptically, and where it is often 
necessary to clean and sterilize equipment between batches. Non-dedicated 
equipment should be thoroughly cleaned between different products and, if 
necessary, after each use. If cleaning a specific type of equipment is difficult, 
the equipment may need to be dedicated to a particular bulk active substance or 
intermediate. Moreover, because the potency of some of these materials may 
not be fully known, cleaning becomes particularly important.
The microbiological aspects of equipment cleaning consist largely of 
preventive measures rather than removal of contamination once it has occurred. 
There should be some evidence that routine cleaning and storage of equipment 
does not allow microbial proliferation. For example, equipment should be dried 
before storage, and under no circumstances should stagnant water be allowed to 
remain in equipment. Subsequent to the cleaning process, equipment should be 
sterilized or sanitized where such equipment is used for sterile processing, or for 
non-sterile processing where the products may support microbial growth. Thus, 
the control of the bioburden through adequate cleaning and storage of equipment 
is important to ensure that subsequent sterilization or sanitization procedures 
achieve the necessary assurance of sterility. This is also particularly important 
from the standpoint of the control of pyrogens in sterile processing, since 
equipment sterilization processes may not be adequate to achieve significant 
inactivation or removal of pyrogens. 
In sterile secondary production areas, all the equipment, fittings and 
services, as far as is practicable, should be designed and installed so that 
operations, maintenance and repairs can be carried out outside the clean area. If 
sterilization is required, it should be carried out after complete re-assembly 
wherever possible. The practice of re-sterilizing equipment if sterility has been 
compromised is important. When equipment maintenance has been carried out 
within the clean area, the area should be cleaned, disinfected and/or sterilized 
where appropriate before processing recommences if the required standards 
of cleanliness and/or asepsis have not been maintained during the work. 
A conveyor belt should not pass through a partition between a grade A or B 
area and a processing area of lower air cleanliness unless the belt itself is 
continually sterilized (for example, in a sterilizing tunnel). 
Equipment must be clearly identified as to its cleaning status and content. 
The cleaning and maintenance of the equipment should be documented in a 
logbook maintained in the immediate area. Establishing and controlling the 
maximum length of time between the completion of processing and each 
cleaning step is often critical in a cleaning process. This is especially important 
for operations where the drying of residues will directly affect the efficiency of 
a cleaning process. In all cases, the choice of cleaning methods, cleaning agents 
and levels of cleaning should be established and justified. When selecting 
cleaning agents, the following should be considered: 
• the cleaning agent's ability to remove residues of raw materials, precursors, 
by-products, intermediates, or even the bulk active substance; 
• whether the cleaning agent leaves a residue itself; 
• compatibility with equipment construction materials.
Validation of cleaning methods 
Validation of cleaning procedures has generated considerable discussion since 
the regulatory authorities started to address this issue. The first step is to focus 
on the objective of the validation process, and some manufacturers fail to 
develop such objectives. It is not unusual to see manufacturers use extensive 
sampling and testing programmes following the cleaning process without really 
evaluating the effectiveness of the steps used to clean the equipment. Several 
questions need to be addressed when evaluating the cleaning process. For 
example, at what point does a piece of equipment or system become clean? 
Does it have to be scrubbed by hand? What is accomplished by hand scrubbing 
rather than just a solvent wash? How variable are manual cleaning processes 
from batch to batch and product to product? What other methods for cleaning 
can be utilized — wipe clean, spray, fog, immersion, ultrasonic, re-circulating 
spray? Is the contamination viable or non-viable? Are there identifiable baseline 
bioburden and residue levels? The answers to these questions are obviously 
important to the inspection and evaluation of the cleaning process, and to 
determine the overall effectiveness of the process. They may also identify steps 
that can be eliminated for more effective measures and result in resource 
savings for the manufacturer. 
In general, cleaning validation efforts should be directed to situations or 
process step where contamination or incidental carry-over of degradants poses 
the greatest risk to the product's quality and safety. The manufacturer should 
have determined the degree of effectiveness of the cleaning procedure for each 
bio-pharmaceutical or intermediate used in that particular piece of equipment. 
In the early stages of the operation, it may be unnecessary to validate cleaning 
methods if it could be shown that subsequent purification steps can remove any 
remaining residues. It must be recognized that for cleaning, as with any other 
processes, there may be more than one way to validate the process. In the end, 
the test of any validation process is whether the scientific data shows that the 
system consistently does as expected and produces a result that consistently 
meets pre-determined specifications. Moreover, cleaning should also be shown 
to remove endotoxins, bacteria, active elements and contaminating proteins, 
while not adversely affecting the performance of the equipment. In cases where 
cleaning reagents are required for decontamination or inactivation, validation 
should also demonstrate the effectiveness of the decontamination/inactivating 
agent(s). 
Validation of cleaning methods should, therefore, reflect the actual 
equipment use patterns. For example, if various bulk actives or intermediates 
are manufactured using the same equipment, and if the same process is used to 
clean the equipment, a worst-case bulk active or intermediate can be selected
for the purposes of cleaning validation. The worst-case selection should be 
based on a combination of potency, activity, solubility, stability and difficulty of 
cleaning. In addition, such cleaning and sanitization studies should address 
microbiological and endotoxin contamination for those processes intended or 
purported to reduce bioburden or endotoxins in the bulk active substance or 
other processes where such contamination may be of concern, for example with 
non-sterile substances used to manufacture parenteral products. 
Documentation 
Depending upon the complexity of the system and the cleaning process, and the 
ability and training of operators, the amount of detail and specificity in the 
documentation necessary for executing various cleaning steps or procedures 
will vary. Some manufacturers use general SOPs, while others use a batch 
record or log sheet system that requires some type of specific documentation for 
performing each step. When more complex cleaning procedures are required, it 
is important to document the critical cleaning steps, including specific 
documentation on the equipment itself and information about who cleaned it 
and when. However, for relatively simple cleaning operations, the mere 
documentation that the overall cleaning process was performed might be 
sufficient. Other factors such as history of cleaning, residue levels found 
after cleaning and variability of test results may also dictate the amount of 
documentation required. For example, when variable residue levels are detected 
following cleaning, particularly for a process that is believed to be acceptable, 
the manufacturer must establish the effectiveness of the process and operator 
performance. 
Protocols 
Cleaning validation protocols should have general procedures on how cleaning 
processes will be validated. It must describe the equipment to be cleaned; 
methods, materials and extent of cleaning; parameters to be monitored and 
controlled; and validated analytical methods to be used. The protocol should 
also indicate the type of samples (rinse, swabs) to be obtained, and how they are 
collected, labelled and transported to the analysing laboratory. Validation 
procedures should address who is responsible for performing and approving 
the validation study, the acceptance criteria and when re-validation will be 
required. Validation studies should be conducted in accordance with the 
protocols, and the results of the studies documented. There should be a detailed 
written equipment cleaning procedure that provides details of what should be 
done and the materials to be utilized. Some manufacturers list the specific 
solvent for each bio-pharmaceutical and intermediate. For stationary vessels,
Clean In Place (CIP) apparatus is often encountered. Diagrams, along with 
identification of specific valves, will be necessary for evaluating these systems. 
Sampling 
After cleaning, there should be some routine testing to assure that the surface 
has been cleaned to the validated level, and to ensure these procedures remain 
effective when used during routine production. Where feasible, equipment 
should be examined visually for cleanliness. This may allow detection of gross 
contamination concentrated in small areas that could go undetected by 
analytical verification methods. Sampling should include swabbing, rinsing, 
or alternative methods such as direct extraction, as appropriate, to detect both 
insoluble and soluble residues. The sampling methods used should be capable 
of quantitatively measuring levels of residues remaining on the equipment 
surfaces after cleaning. There are two general types of sampling that have been 
found acceptable — the most desirable is the direct method of sampling the 
equipment surface, and the other is the use of rinse solutions. 
Direct surface sampling 
The advantages of direct sampling are that areas hardest to clean, but which are 
reasonably accessible, can be evaluated, leading to the establishment of a level 
of contamination or residue per given surface area. Additionally, residues that 
are dried out, or are insoluble, can be sampled by physical removal. Swab 
sampling may be impractical when product contact surfaces are not easily 
accessible due to equipment design and/or process limitations, such as the 
inner surfaces of hoses, transfer pipes, reactor tanks with small ports or 
handling active materials, and small intricate equipment such as micronizers 
and micro-fluidizers. One major concern is the type of sampling material used 
and its impact on the test data, since the sampling material may interfere with 
the test. For example, the adhesive used in swabs has been found to interfere 
with the analysis of samples. Therefore, it is important to assure early in the 
validation programme that the sampling medium and the solvent used for 
extraction from the medium are satisfactory and can be readily used. 
Rinse samples 
This is the analysis of the final rinse water or solvent for the presence of the 
cleaning agents last used in that piece of equipment. Two advantages of using 
rinse samples are that a larger surface area may be sampled, and inaccessible 
systems or ones that cannot be routinely disassembled can be sampled and 
evaluated. However, the disadvantage of rinse samples is that the residue or 
contaminant may not be soluble or may be physically occluded in the
equipment. An analogy that can be used is the dirty pot — in the evaluation of 
cleaning of a dirty pot, particularly with dried out residue, one does not look at 
the rinse water to see that it is clean; one looks at the pot. A direct measurement 
of the residue or contaminant should be made for the rinse water when it is used 
to validate the cleaning process. For example, it is not acceptable to simply test 
rinse water for water quality (does it meet the compendia tests?), rather than test 
it for potential contaminates. In addition, indirect monitoring such as conductivity 
testing may be of some value for routine monitoring once the cleaning 
process has been validated. This would be particularly true, where reactors and 
centrifuges and pipework between such large equipment can only be sampled 
using rinse solution samples. 
Analytical methods and establishment of limits 
How do you evaluate and select analytical methods to measure cleaning and 
disinfection effectiveness in order to implement basic cleaning validation and 
to establish routine in-use controls. Regulatory authorities do not set acceptance 
specifications or methods for determining whether a cleaning process is 
validated because it is impractical for them to do so due to the wide variation in 
equipment and products used throughout the industry. 
With advances in analytical technology, residues from the manufacturing 
and cleaning processes can be detected at very low levels. The sensitivity of 
some modern analytical apparatus has lowered some detection thresholds to 
below parts per million (ppm) down to parts per billion (ppb). Some limits that 
have been mentioned by industry representatives in literature or presentations, 
include analytical detection levels such as 10 ppm, biological activity 
levels such as 1/1000 of the normal therapeutic dose, and organoleptic levels 
such as no visible residue. The residue limits established for each piece of 
apparatus should, therefore, be practical, achievable and verifiable. If levels of 
contamination or residual are not detected, it does not mean that there is no 
residual contaminant present after cleaning; it only means that levels of 
contaminant greater than the sensitivity or detection limit of the analytical 
method are not present in the sample. The manufacturer's rationale for 
establishing specific residue limits should be logical, based on their knowledge 
of the materials involved, be practical, achievable and verifiable, have a 
scientifically sound basis, and be based on the most deleterious residue. 
Limits may, therefore, be established, based on the minimum known pharmacological 
or physiological activity of the product or its most deleterious 
component. 
Another factor to consider is the possible non-uniform distribution of the 
residue on a piece of equipment. The actual average residue concentration may
be more than the level detected. It may not be possible to remove absolutely 
every trace of material, even with a reasonable number of cleaning cycles. The 
permissible residue level, generally expressed in parts per million (ppm), 
should be justified by the manufacturer. The manufacturer should also 
challenge the analytical method in combination with the sampling method(s) 
used, to show that the contaminants can be recovered from the equipment 
surface, and at what levels, i.e. 50% or 90% recovery. This is necessary before 
any conclusions can be made based on the sample results. A negative test may 
also be the result of poor sampling technique. 
Clean In Place methods 
Where feasible, Clean In Place (CIP) methods should be used to clean 
process equipment and storage vessels. CIP methods might include fill and 
soak/agitate systems, solvent refluxing, high-impact spray cleaning, spray 
cleaning by sheeting action, or turbulent flow systems. CIP systems should be 
subjected to cleaning validation studies to ensure that they provide consistent 
and reproducible results, and once they are validated, appropriate documentation 
should be maintained to show that critical parameters, such as time, 
temperature, turbulence, cleaning agent concentration, rinse cycles, are 
achieved with each cleaning cycle. However, the design of the equipment, 
particularly in facilities that employ semi-automatic or fully automatic Clean 
In Place (CIP) systems, can represent a significant concern. For example, 
sanitary type pipework without ball valves should be used, since non-sanitary 
ball valves make the cleaning process more difficult. Such difficult to clean 
systems should be properly identified and validated, and it is important that 
operators performing these cleaning operations are aware of potential 
problems and are specially trained in cleaning these systems and valves. 
Furthermore, with systems that employ long transfer lines or pipework, 
clearly written procedures together with flow charts and pipework diagrams 
for the identification of valves should be in place. Pipework and valves 
should be tagged and easily identifiable by the operator performing the 
cleaning function. Sometimes, inadequately identified valves, both on 
diagrams and physically, have led to incorrect cleaning practices. Equipment 
in CIP systems should be disassembled during cleaning validation where 
practical to facilitate inspection and sampling of inner product surfaces for 
residues or contamination, even though the equipment is not normally 
disassembled during routine use. 
Test until clean 
Some manufacturers are known to test, re-sample and re-test equipment or 
systems until an 'acceptable' residue level is attained. For the system or
equipment with a validated cleaning process, this practice of re-sampling 
should not be utilized and is only acceptable in rare cases. Constant re-testing 
and re-sampling can show that the cleaning process is not validated, since these 
re-tests actually document the presence of unacceptable residue and contaminants 
from an ineffective cleaning process. The level of testing and the re-test 
results should, therefore, be routinely evaluated. 
Detergent 
The manufacturer must consider and determine the difficulty that may arise 
when attempting to test for residues if a detergent or soap is used for cleaning. 
A common problem associated with detergent use is its composition — many 
detergent suppliers will not provide specific composition, making it difficult for 
the user to evaluate residues. As with product residues, it is important that the 
manufacturer evaluate the efficiency of the cleaning process for the removal of 
residues from the detergents. However, unlike product residues, it is expected 
that no (or for ultra sensitive analytical test methods — very low) detergent 
remains after cleaning. Detergents are not part of the manufacturing process 
and are only added to facilitate cleaning during the cleaning process, so they 
should be easily removable or a different detergent should be selected. 
11.6 Process utilities and services 
11.6.1 Water systems 
Water is a very important component of bio-pharmaceutical processes. Water of 
suitable quality is required depending on the culture system used, the phase of 
manufacture and the intended use of the product. Tighter chemical and 
microbiological quality specifications are required during certain process 
steps such as cell culture, final crystallization and isolation, and during early 
process steps if impurities that affect product quality are present in the water 
and cannot be removed later. Where water is treated to achieve an established 
quality, the treatment process and associated distribution systems should be 
qualified, validated, maintained and routinely tested following established 
procedures to ensure water of the desired quality. The water used should 
meet the standards for potable water as a minimum for the production of biopharmaceuticals. 
The potable water supply, regardless of source, should be assessed for 
chemicals that may affect the process, and information should be periodically 
sought from local authorities about potential contamination by pesticides or 
other hazardous chemicals. For example, if water is used for a final wash of a
filter cake, or if the bulk active substance is crystallized from an aqueous 
system, then the water should be suitably treated, such as by de-ionization, 
ultrafiltration, reverse osmosis or distillation, and tested to ensure routine 
compliance with appropriate chemical and microbiological specifications. If 
the water is used for final rinses during equipment cleaning, then the water 
should be of the same quality as that used in the manufacturing process. Water 
used in the final isolation and purification steps of non-sterile bulk actives 
intended for use in the preparation of parenteral products should be tested and 
controlled for bioburden and endotoxins. 
The quality of water, therefore, depends on the intended use of the finished 
product. For example, only Water for Injection (WFI) quality water should be 
utilized as process water; this is because, even though water may not be a 
component of the final sterile product, water that comes in contact with 
the equipment or that enters into the bioreactor can be a source of impurities 
such as endotoxins. On the other hand, for in-vitro diagnostics purified 
water may suffice. For heat-sensitive products where processing such as 
formulation is carried out cold or at room temperature, only cold WFI will 
suffice, and the self-sanitization of a hot WFI system at 75° to 800C is lost. As 
with other WFI systems, if cold WFI water is needed, point-of-use heat 
exchangers can be used; however, these cold systems are still prone to 
contamination, and should be fully validated and routinely monitored both 
for endotoxins and microorganisms. 
Water treatment plants and distribution systems should be designed, 
constructed and maintained to ensure a reliable source of water of an appropriate 
quality. They should never be operated beyond their designed capacity. 
For economic reasons, some biotechnology companies manufacture WFI 
utilizing marginal systems, such as single pass reverse osmosis, rather than 
by distillation. Many such systems have been found to be contaminated, 
typically because they use plastic pipes and non-sealed storage tanks, which 
are difficult to sanitize. Although some of the systems employ a terminal 
sterilizing filter to minimize microbiological contamination, the primary 
concern is endotoxins which the terminal filter may merely serve to mask. 
Such systems are, therefore, totally unacceptable. Moreover, the limitations of 
relying on a 0.1 ml sample of WFI for endotoxins from a system should also be 
recognized. 
New water quality requirements were brought into effect in 1996. These 
updated requirements provide major cost savings to those manufacturers who 
needed to produce and maintain pure water systems, and allowed for the 
continuous monitoring of water systems with a reliance on instrumentation 
rather than laboratory work, thereby reducing labour and operating costs.
Previous standards required a battery of expensive and labour intensive 
chemical, physical, and microbiological testing, many of which only provided 
qualitative information. Advances in technology and instrumentation mean that 
simple, cost effective replacements have become available. However, before 
changing to the new testing standards, manufacturers should evaluate their 
existing water system in terms of compliance with existing operations, 
reliability, maintenance and improved monitoring. 
11.6.2 Medical air 
Medical air is a natural or synthetic mixture of gases consisting largely of 
nitrogen and oxygen, containing no less than 19.5 percent and not more than 
23.5 percent by volume of oxygen. Air supplied to a non-sterile preparation or 
formulation area, or for manufacturing solutions prior to sterilization, should be 
filtered at the point of use as necessary to control particulates. However, air 
supplied to product exposure areas, where sterile bio-pharmaceuticals are 
processed and handled, should be filtered under positive pressure through 
high efficiency particulate air (HEPA) filters. These HEPA filters should be 
certified and/or Dioctyl Phthalate tested. Tests for oil (none discernible by the 
mirror test), odour (no appreciable odour), carbon dioxide (not more than 
0.05%), carbon monoxide (not more than 0.001%), nitric oxide and nitrogen 
dioxide (not more than 2.5 ppm), and for sulphur dioxide (not more than 5 ppm) 
should also be carried out. Medical air is packaged in cylinders or in a low 
pressure collecting tank. Containers used should not be treated with any active, 
sleep-inducing, or narcosis-producing compounds, and should not be treated 
with any compound that would be irritating to the respiratory tract. Where it is 
piped directly from the collecting tank to the point of use, each outlet should be 
labelled Medical Air. 
11.6.3 Heating, ventilation and air conditioning (HVAC) systems 
A bio-pharmaceutical facility should have proper ventilation, air filtration, air 
heating and cooling. Therefore, adequate ventilation should be provided where 
necessary, and equipment for the control and monitoring of air pressure, 
microorganisms, dust, humidity and temperature should be provided when 
appropriate. This is especially important in areas where the product is exposed 
to the environment or handled in the final state. Air filtration, dust collection 
and exhaust systems should be used when appropriate, and if the air is 
recirculated, appropriate measures should be taken to control contamination 
and cross-contamination. For example, air from pre-viral inactivation areas 
should not be recirculated to other areas used for the manufacture of the sterile
bio-pharmaceuticals. Regulatory authorities require the following information 
to be available for pre-approval inspection: 
• A general description of the HVAC system(s) including the number and 
segregation of the air handling units, whether air is once-through or 
recirculated, containment features, and information on the number of air 
changes per hour; 
• Validation summary for the system with a narrative description of the 
validation process (or protocol), including the acceptance criteria; the 
certification that IQ, OQ, and certification of filters has been completed; 
the length of the validation period; validation data should include Performance 
Qualification data accumulated during actual processing; and an 
explanation of all excursions or failures, including deviation reports and 
results of investigations; 
• A narrative description of the routine monitoring programme including the 
tests performed and frequencies of testing for viable and non-viable 
particulate monitoring parameters; viable and non-viable particulate action 
and alert limits for production operations for each manufacturing area; and a 
summary of corrective actions taken when limits are exceeded. 
11.6.4 Decontamination techniques and waste recovery 
Air and gaseous waste streams 
Filtration 
The primary method of decontaminating exhaust gases mixed with liquid broth 
is through the use of filters. Before filtration, the mixture may be passed 
through a condenser, a coalescing filter and a heat exchanger. Filtration is 
accomplished either through pairs of high efficiency particulate air (HEPA) 
filters, or membrane filters used in series to decontaminate vent or exhaust 
gases. 
Incineration 
Another method of decontaminating air and gaseous waste streams is thermal 
destruction or incineration. Incineration may be used independently, or as a 
supplement to filtration, and is generally used for small volume gas streams. 
Automatic safety devices should be used with incinerators to protect against 
problems resulting from power failures and overheating.
Irradiation 
Irradiation involves exposing the waste materials to x-rays, ultraviolet rays or 
other ionizing radiation to decontaminate them. 
Liquid wastes 
Liquid wastes can be decontaminated through chemical or heat treatment. 
When liquid wastes are of limited volume, chemical treatment is often used, 
whilst for large volumes of liquid wastes, heat treatment is generally preferred. 
Also, since proteins present in liquid wastes can deactivate the sterilant used in 
chemical treatment, thermal sterilization may be more appropriate for wastes 
involving bioengineered microorganisms. 
Solid wastes 
Solid wastes such as microbial cultures, cell debris, glassware, and protective 
clothing, are generally decontaminated by autoclaving, followed by incineration 
if necessary. To decontaminate laboratory devices exposed to genetically 
engineered products, the most common practice is the use of pressurized steam 
that contains an appropriate chemical. For heat-sensitive equipment, such as 
electronic instruments, decontamination is generally achieved through chemical 
sterilization or irradiation. Gaseous sterilants are applied by a steam ejector 
that sprays down from overhead. If decontamination by steam, liquid, or gas 
sterilization is not possible, ionizing or ultraviolet radiation is used. However, 
since irradiation methods do not always inactivate all types of microbes, steam 
or gaseous chemical sterilization should be used for devices contaminated with 
genetically engineered organisms.