Adobe Flash Player.

  Adobe Flash Player

In Three Volumes VOLillvlE 1
EDITED BY Herbert A. Lieberman

Several years have passed since the first edition of Pharmaceutical Dosage 
Forms: Tablets was publlahed , During this time, considerable advances 
have been made in the science and technology of tablet formulation, manufacture, 
and t eating. These changes are reflected in this updated, revised 
and expanded second edition. 
The tablet dosage form continues to be the most widely used drug delivery 
system for both over-the-counter and prescription drugs. The term 
tablet encompasses: the usual compressed tablet j the compressed tablet that 
is sugar- or film-coated to provide dissolution in either the stomach or the 
intestine, or partially in the stomach and partially in the intestine j layered 
tablets for gastric and intestinal release; effervescent tablets; sustainedrelease 
tablets; compressed coated tablets; sublmgual and buccal tablets; 
chewable tablets; and medicated lozenges. These various dosage forms are 
described in depth in the three volumes of this series. 
In the first volume, the various types of tablet products are discussed; 
the second volume is concerned with the processes involved in producing 
tablets, their bioavailability and pharmacokinetics; and in the third volume, 
additional processes in tablet production are discussed, as well as sustained 
drug release, stability, kinetics. automation. pilot plant, and quality assurance. 
The first chapter in Volume 1 describes "Preformulation Testing. II This 
second edition of the chapter contains an extensive amount of new material 
on substance purity, dissolution, the concept of permeability. and some of 
the pharmaceutical properties of solids. In the second chapter, "Tablet 
Formulation and Design I II the plan for developing prototype formulas has 
been revised and an approach, using statistical design, is presented. 
There 1s consideration given to those elements in tablet formulation that are 
of importance to the operation of tablet presses with microprocessor controls . 
 There have been so many advances in the technology of wet granulation 
and direct compression methods since the first edition that what had 
previously been one chapter has now been expanded into two chapters. 
"Compressed Tablets by Wet Granulation" has been updated. and a new 
section on unit operations has been added. Information on the formulations 
of sustained-release tablets by wet granulation is included in the chapter. 
"Compressed Tablets by Direct Compression." a separate chapter new to 
this edition, contains: a table comparing all aspects of direct compression 
versus wet granulation; an extensive glossary of trade names and manufacturers 
of tableting excipients; a section on morphology of pharmaceutical 
excipients. including scanning electron photomicrographs; a discussion of 
direct compression of example active ingredients; and a considerably expanded 
section on prototype or guide formulations. 
The chapter entitled "Compression-Coated and Layered Tablets" describes 
the current technology for making these types of tablets. The 
chapter "Effervescent Tablets" has been expanded to include fluid-bed 
granulation techniques. updating on stability testing methods. new packaging 
materials. and methodologies for checking airtightness of sealed packets. 
The chapter on "Special Tablets" now contains information on long-acting 
and controlled-release buccal tablets as well as new sections On vaginal 
and rectal tablets. The chapter "Chewable Tablets" has increased its coverage 
to include microencapsulation and spray coating techniques. This 
chapter includes an update of the information concerned with excipients, 
colorants, direct-compression chewable tablets, and current manufacturing 
and product evaluation procedures related to these tablets. "Medicated 
Lozenges," the final chapter in Volume I, has increased its scope to include 
liquid-center medicated lozenges and chewy-based medicated tablets. 
Each of the tablet forms discussed requires special formulation procedures. 
Knowing how to make a particular type does not guarantee knowledge of how 
to make another. Since considerable expertise is required for the myriad 
tablet dosage forms, a multiauthored text seemed to be the only way to 
accomplish the editors' goals of providing knowledgeable and complete coverage 
of the subject. The editors chose authors to describe particular 
types of tablets on the basis of their experience, training. and high degree 
of knOWledge of their subjects, 
The authors were charged with the task of covering their technology in 
a way that would not be merely a review of the literature. Each chapter 
begins by assuming the reader is not very familiar with the subject. 
Gradually. as each chapter develops, the discussion becomes more advanced 
and specific. Following this format. we have intended the text to be a 
teaching source for undergraduate and graduate students as well as experienced 
and inexperienced industrial pharmaceutical scientists. The book 
can also act as a ready reference to all those interested in tablet technology. 
This includes students. product development pharmacists, hospital 
pharmacists, drug patent attorneys. governmental and regulatory scientists. 
quality control personnel, pharmaceutical production personnel. and those 
concerned with production equipment for making tablets. 
The authors are to be commended for the manner in which they cover 
their subjects ss well as for their patience with the editors' comments concerning 
their manuscripts. The editors wish to express their special 
thanks to the contributors for the excellence of their works, as well as for 
their continued forbearance with Our attempts to achieve our desired level 
of quality for this text. Although there has been a great deal written 
about various types of tablets, it is only in this multivolume treatment 
that this SUbject is completely described. The acceptability and usefulness

of these volumes is attributable to the efforts and skills of all of the contributing 
The topics. format. and choice of authors are the responsibilities of 
the editors. Any multiauthor book has problems of coordination and 
minimizing repetition. Some repetition was purposely retained because, 
in the editors' opinions. it helped the authors to develop their themes and 
because each individUal treatment is sufficiently different so as to be valuable 
as a teaching aid. The editors hope that the labors of the contributors 
and our mutual [udgments of subject matter have resulted in an up-todate 
expanded reference that will facilitate the work of the many people who 
use it. 
Preformulation Testing 
Deodatt A. Wadke, Abu T. M. Serajuddin, and Harold Jacobson 
E. R. Squibb & Sons. New Brunswick. New Jersey 
Preformulation testing is the first step in the rational development of dosage 
forms of a drug substance, It can be defined as an investigation of physical 
and chemical properties of a drug substance alone and when combined 
with excipients. The overall objective of preformulation testing is to generate 
information useful to the formulator in developing stable and bioavailable 
dosage forms that can be mass produced. Obviously. the type of 
information needed will depend on the dosage form to be developed. This 
chapter will describe a preformulation program needed to support the development 
of tablets and granulations as dosage forms. 
During the early development of a new drug substance, the synthetic 
chemist. alone or in cooperation with specialists in other disciplines (including 
preformulation), may record some data that can be appropriately considered 
as preformulation data. This early data collection may include such 
information as gross particle size. melting point, infrared analysis, chromatographic 
purity. and other such characterizations of different laboratory 
scale batches. These data are useful in guiding. and becoming part of, 
the main body of preformulation work. Interactions between the responsible 
preformulation scientist, medicinal chemist. and pharmacologist at the very 
early stages of drug development are to be encouraged and must also focus 
on the biological data. Review of such data for a series of compounds when 
available. and review of the physical chemical properties of the compounds 
with some additional probing studies if necessary. would help in the early 
selection of the correct physical and chemical form of the drug entity for 
further development. 
The formal preformulation study should start at the point after biological 
screening, when a decision is made for further development of the compound 
in clinical trials. Before embarking on a formal program. the preformulation 
scientist must consider the following: 
Available physicochemical data (including chemical structure, different 
salts available) 
Anticipated dose 
Supply situation and development schedule (i.e., time available) 
Availability of stability-indicating assay 
Nature of the information the formulator should have or would like 
to have 
The above considerations will offer the preformulation scientist some 
quidanee in deciding the types and the urgency of studies that need attention. 
Selectivity is very critical to the success of the preformulation 
program. Not all the preformulation parameters are determined for every 
new compound. Data, as they are generated, must be reviewed to decide 
what additional studies must be undertaken. For example, a detailed investigation 
of dissolution is not warranted for a very soluble compound. On 
the other hand, particle size, surface area, dissolution, and the means of 
enhancing rate of dissolution are important considerations in the preformuIation 
evaluation of a sparingly soluble drug. 
A typical preformulation program should begin with the description of the 
drug substance. The color, odor, and taste of the new drug must be recorded 
using descriptive terminology. It is important to establish a standard 
terminology to describe these properties in order to avoid confusion 
among scientists using different terms to describe the same property. A 
list of some descriptive terms to describe the most commonly encountered 
colors, tastes, and odors of pharmaceutical powders is provided in Table 1. 
The color of all the early batches of the new drug must be recorded 
using the descriptive terminology. A record of color of the early batches 
is very useful in establishing appropriate specifications for later production. 
When the color attributes are undesirable or variable, incorporation of a 
dye in the body or coating of the final product could be recommended. 
Table 1 Suggested Terminology to Describe 
Organoleptic Properties of Pharmaceutical 
Color Odor Taste 
Off-white Pungent Acidic 
Cream yellow Sulfurous Bitter 
Tan Fruity Bland 
Shiny Aromatic Intense 
Odorless Sweet 

Preformulation Testing 3 
Drug substances, in general, have characteristic odors and tastes. In 
tasting the new drug, due caution must be exerted. If taste is considered 
as unpalatable, consideration ought to be given to the use of a less soluble 
chemical form of the drug, if one is av8ilable-provided, of course, the 
bioav8ilability is not unacceptably compromised. The odor and taste may 
be suppressed by using appropriate flavors and excipients or by coating 
the final product. The flavors, dyes, and other excipients selected to 
alleviate the problems of unsightly or variable color, and unpleasant odor 
and taste must be screened for their influence on the stability and bioavailability 
of the active drug. 
Many drug substances are irritating to skin or sternutatory. Such 
information may already be available or developed during the course of preformulation 
studies. Where available , this information must be highlighted 
such that appropriate procedures for material handling and personnel protection 
can be developed. 
The preformulation sclentists must have some perception of the purity of a 
drug substance. It is not this individual's primary responsibility to rigorously 
establish and investigate the purity (Notwithstanding that this is an 
important subjeet ) . Such studies are most often performed in an analytical 
research and development group. But some early knowledge is necessary 
so that subsequent preformulation and/or early safety and clinical studies 
are not compromised as to their validity. This is not to mean necessarily 
that relatively inhomogeneous material or material showing some impurity be 
rejected for preformulation studies. It does mean that such properties be 
recognized and be acceptable. It is another control parameter that allows 
for comparison with subsequent batches. 
There are also more direct concerns. Occasionally an impurity can affect 
stability. Metel contamination at the level of a few parts per million 
is a relatively common example in which certain classes of compounds are 
deleteriously affected. Appearance is another area where a slight impurity 
can have a large effect. Off-color materials upon recystallization can become 
white in many instances. Further, some impurities require circumspection 
because they are potentially toxic. The presence of aromatic amines , 
suspected of being carclnogenic, is an example. In these instances, discussions 
must be initiated with the chemist preparing the material so that remedial 
action can be taken. Very often a problem batch can be made satisfactory 
by a simple recrystallization. 
Fortunately, the techniques used for characterizing the purity of a 
drug are the same as those used for other purposes in a preformulation 
study. Most of the techniques mentioned below are described in greater 
det8il elsewhere in the chapter and are used to characterize the solid state, 
or as an analytical tool in stability or solUbility studies. 
Thin-layer chromatography (TLC) and high-pressure liquid chromatography 
(HPLC) are of very wide-ranging applicability and are excellent tools 
for characterizing the chemical homogeneity of very many types of materials. 
Paper chromatography and gas chromatography are also useful in the determination 
of chemical homogeneity. 
All of these techniques can be designed to give a quantitative estimate 
of purity. Measures such as impurity index (II) and homogeneity index

(HI) are useful and easy to celculate , especially from the HPLC chromatographs. 
The II of a batch is defined as the ratio of all responses due to 
components other than the main one to the total response. Typically, the 
responses are obtained as area measurements in the chromatographic procedure. 
The obverse of the II is HI, which is the ratio of the response due 
to the main component to the total response. Figure 1, which shows an 
HPLC chromatograph for an experimental drug, illustrates determination 
of II and HI. The chromatograph was generated using a UV detector. In 
Figure 1, peak due to the main component occurs at a retention time of 
4.39 with an area response of 4620. The seven other minor peaks are due 
to the UV-absorbing impurities with total area response of 251. Thus, II 
in this case is 251/( 4620 + 251) ::: 0.0515, and the HI is 1 - 0.0515 ::: 
The United States Pharmacopeia (USP) has proposed a related procedure 
called ordinary impurities test that estimates impurities using TLC. In 
this test impurity index is defined as a ratio of responses due to impurities 
to that response due to a defined concentration of a standard of the main 
component. The USP is proposing a general limit of 2% impurities [1,2]. 
The II. HI, and the impurity index as proposed by the USP are not 
absolute measures of impurity since the specific response (i. e., molecular 
absorbances or extinction coefficients) due to each impurity is assumed to 
be the same as that of the main component. A more accurate analysis requires 
the identification of each individual impurity followed by preparation 
of standards for each one of them. Such information is almost always unavailable 
at the early stages of development. 
Other tools useful in the assessment of purity are differential and 
gravimetric thermal analyses. These techniques often provide a qualitative 
picture of homogeneity and also give direct evidence of the presence of solvates. 
Since these methods are simple and are used in characterizing the 
material, their use for purity information is incidental. The appearance 
of several peaks or the acuteness of an endotherm can often be indicative 
of the purity. Similar information may sometimes also be generated by

Preformulation Testing 5 
observing the melting point, especially with a hot-stage microscope. More 
quantitative information can be obtained by using quantitative differential 
scanning calorimetry or by phase-rule solubility analysis. 
As important to a compound's chemical characteristic are its physical 
ones. Crystalline form (including existence of solvates) is of fundamental 
importance, and for complete documentation of the compound X-ray powder 
diffraction patterns for each batch is desirable. This is simple to execute 
and provides useful information for later comparison and correlation to 
other properties. 
Various chemical and physical properties of drug substances are affected 
by their particle size distribution and shapes. The effect is not only on 
the physical properties of solid drugs but also, in some instances, on their 
biopharmaeeutieal behavior. For example, the bioavailability of griseofulvin 
and phenacetin is directly related to the particle size distributions of these 
drugs [3,4]. It is now generally recognized that poorly soluble drugs 
showing a dissolution rate-limiting step in the absorption process will be 
more readily bioavailable when administered in a finely subdivided state 
than as a coarse material. Very fine materials are difficult to handle [5]; 
but many difficulties can be overcome by creating solid solution of a material 
of interest in a carrier, such as a water-soluble polymer. This represents 
the ultimate in size reduction, since in a (solid) solution, the dispersed 
material of interest exists as discrete molecules or agglomerated molecular 
bundles of very small dimensions indeed. 
Size also plays a role in the homogeneity of the final tablet. When 
large differences in size exist between the active components and excipients, 
mutual sieving (demixing) effects can occur making thorough mixing difficult 
or, if attained, difficult to maintain during the subsequent processing 
steps. This effect is greatest when the diluents and active raw materials 
are of significantly different sizes. Other things being equal, reasonably 
fine materials interdisperse more readily and randomly. However, if materials 
become too fine, then undersirable properties such as electrostatic effects 
and other surface active properties causing undue stickiness and lack of 
flowability manifest. Not only size but shape too influences the flow and 
mixing efficiency of powders and granules. 
Size can also be a factor in stability; fine materials are relatively more 
open to attack from atmospheric oxygen, heat, light, humidity, and interacting 
exipients than coar-se materials. Weng and Parrott [6] investigated 
influence of particle size of sulfacetamide on its reaction with phthalic anhydride 
in 1: 2 molar compacts after 3 hr at 95C. Their data, presented 
in Table 2, clarly demonstrate greater reactivity of sulfacetamide with decreasing 
particle size. 
Because of these significant roles I it is important to decide on a desired 
size range, and thence to maintain and control it. It is probably safest to 
grind most new drugs having particles that are above approximately 100 urn 
in diameter. If the material consists of particles primarily 30 11m or less in 
diameter, then grinding is unnecessary, except if the material exists as 
needles-where grinding may improve flow and handling properties, or if 
the material is poorly water-soluble where grinding increases dissolution 
rate. Grinding should reduce coarse material to, preferably, the 10- to

Table 2 Influence of Particle Size 
on Conversion of Sulfacetamide 
Wadke, Serajuddin, and Jacobson 
Particle size of 
21.54  2.74 
19.43  3.25 
17.25  2.88 
15.69  7.90 
9.34  4.41 
Source: Modified from Weng, H., 
and Parrott, E. L., J. Pharm. Sci., 
73: 1059 (1984). Reproduced with 
the permission of copyright owner. 
40-11m range. Once this is accomplis hed, controlled testing can be performed 
both for subsequent in vivo studies and for in-depth preformulation 
studies. As the studies proceed, it may become apparent that grinding is 
not required and that coarser materials are acceptable. At that time, it 
is conceptually simpler to omit that step without jeopardizing the information 
already developed. The governing concept is to stage the material so that 
challenges are maximized. 
There are several drawbacks to grinding that may make it inadvisable. 
Some are of lesser importance. For example, there are material losses when 
grinding is done. Sometimes a static electricity buildup occurs, making the 
material difficult to handle. Often, however, this problem, if it exists, may 
be circumvented by mixing with excipients such as lactose prior to grinding. 
Reduction of the particle size to too small a dimension often leads to aggregation 
and an apparent increase in hydrophobicity, possibly lowering the 
dissolution rate and making handling more troublesome. When materials are 
ground, they should be monitored not only for changes in the particle size 
and surface area, but also for any inadvertent polymorphic or chemical 
transformations. Undue grinding can destory solvates and thereby change 
some of the important characteristics of a substance. Some materials can 
also undergo a chemical reaction. 
A. General Techniques for Determining Particle Size 
Several tcols are commonly employed to monitor the particle size. The most 
rapid technique allowing for a quick appraisal is microscopy. Microscopy, 
since it requires counting of a large number of particles when quantitative 
information is desired. is not suited for rapid. quantitative size determinations. 
However, it is very useful in estimating the range of sizes and the 
shapes. The preliminary data can then be used to determine if grinding is 
needed. A photomicrograph should be taken both before and after grinding. 
The range of sizes observable by microscopy is from about 1 um upward.

Preformulatton Testing 7 
For optical microscopy, the material is best observed by suspending it 
in a nondissolving fluid (often water or mineral oil) and using polarizing 
lenses to observe birefringence as an aid to detecting a change to an amorphous 
state after grinding. 
For a quantitative particle size distribution analysis of materials that 
range upward from about 50 um, sieving or screening is appropriate, although 
shape has a strong influence on the results. Most pharmaceutical 
powders, however, range in size from 1 to 120 um, To encompass these 
ranges, a variety of instrumentation has been developed. There are instruments 
based on lasers (Malvern), light scattering (Royco), light blockage 
(HIAC), and blockage of an electrical conductivity path (Coulter 
Counter). The instrument based on light blockade has been adopted by 
the USP to monitor the level of foreign particulates in parenteral products. 
The instrument will measure particle size distribution of any powder properly 
dispersed in a suspending medium. The concentration of sample suspension 
should be such that only a single particle is presented to the senSOl' 
in unit time, thus avoiding coincidence counting. 
other techniques based on centrifugation and air suspension are also 
available. Most of these instruments measure the numbers of particles, but 
the distributions are readily converted to weight and size distributions. 
The latter way of expressing the data is more meaningful. A number of 
classical techniques based on sedimentation methods, utilizing devices such 
as the Andreasen pipet or recording balances that continuously collect a 
settling suspension, are also known. However. these methods are now in 
general disfavor because of their tedious nature. Table 3 lists some of the 
common techniq ues useful for measurement of different size ranges (7). 
There are many mathematical expressions that can be used to characterize 
an average size. These refer to average volumes or weights, geometric 
mean diameters, and relationships reflecting shapes, such as the 
ratio of an area to a volume or Weight factor (8). 
Table 3 Common Techniques for 
Measuring Fine Particles of 
Various Sizes 
Particle size 
Technique ( lim) 
Microscopic 1-100 
Sieve >50 
Sedimentation >1 
Elutriation 1-50 
Centrifugal <50 
Permeability >1 
light scattering 0.5-50 
Source: Parrott, E. L., Pharm. Mfg., 
4: 31 (1985). Reproduced with the 
permission of copyright owner.

8 Wadke, Serajuddin, and Jacobson 
Cumulative WeiC)ht Percentage at the Indicated Sile 
80 I- 998 60 40 20 
60 lf... 
40 I- 
30 - 
20 - 
I&J 10 r-N 
en 6-
Figure 2 Log probability plot of the size distribution of a sample of triarncinolone 
A convenient way to characterize a particle size distribution is to construct 
a log probability plot. Log probability graph paper is commercially 
available, and particle size distributions resulting from a grinding operation 
with no cut being discarded will give a linear plot. An example is illustrated 
in Figure 2 for a powder sample of triamcinolone acetonide. The 
data used in the construction of Figure 2 are presented in Table 4. 
The numbers of particles in Table 4 are converted into weight fractions 
by assuming them to be spheres and multiplying by the volume of a single 
sphere (particle) calculated from the geometric relationship: 
where V is the volume and d the particle diameter (using the average 
value of the range given in the first column of Table 4). The result is 
the total volume occupied by particles in each of the size ranges and is 
given in the third column of the table. The volume is directly related to 
a mass term by the reciprocal of the density. However. since the density 
is constant for all particles of a single species and is rarely known accurately, 
it is sufficient to use the volume terms to calculate the weight percentages 
in each size range by dividing the total volume of all the particles 
into the volumes in each range (column 4 of Table 4). If densities were 
used, it is obvious that they would cancel out in this calculation. The 
cumulative weight percentage in each size range is shown in the last column. 
Statistical descriptions of distributions most often give a measure of 
central tendency. However, with powders the distributions are skewed in 
the direction of increasing size. This type of distribution can be described 
by the Hatch-Choate equation:

Preformulation Testing 9 
f = tn 
v'2n In 0g 
(tn d -2 In M) 2 ] 
2 In (Jg 
where f is the frequency with which a particle of diameter d occurs, and 
n is the total number of particles in a powder in which the geometric mean 
particle size is M and the geometric standard deviation is O"g' Equation (1) 
is succinctly discussed by Orr and Dalla Valle [9]. 
The two measures M and O"g uniquely characterize a distribution. and 
are readily obtained graphically from a log probability plot in which cumulative 
weight percentage is plotted against the particle size (Fig. 2). The 
geometric mean diameter corresponds to the 50% value of the abscissa, and 
the geometric standard deviation is given by the following ratios. the values 
for which are taken from the graph. 
84.13% size 0" = ~~..::....=.,~~ 
g 50% size 
50% size =...,....,...:...::...::.,-=--:..:..- 
15.87% size 
For the example, the values are 8.2 and 1. 5 um for the geometric mean 
particle size and its standard deviation, respectively. The latter is also a 
slope term. For particle size distributions resulting from a crystallization, 
a linear plot can often be obtained using linear probability paper. 
B. Determination of Surface Area 
The determination of the surface areas of powders has been getting increasing 
attention in recent years. The techniques employed are relatively 
simple and convenient to use. and the data obtained reflect the particle 
Table 4 Particle Size Distribution of a Ground Sample of Triamcinolone 
Volume of 
Weight Cumulative 
Size range No. of 
10- 3 (llm 3) percent weight 
(lJm) particles x in range percent 
22.5-26.5 5 38 0.2 100.0 
18.6-22.0 54 237 1.7 99.8 
14.9-18.6 488 1212 8.8 98.1 
11. 8-14. 9 2072 2552 18.5 89.3 
9.4-11.8 5376 3352 24.3 70.8 
7.4-9.4 9632 2989 21.7 46.5 
5.9-7.4 12,544 1888 13.7 24.8 
4.7-5.9 12,928 1008 7.3 11.1 
3.7-4.7 13,568 526 3.8 3.8

10 Wadke, Serajuddin, and Jacobson 
size. The relationship between the two parameters is an inverse one, in 
that a grinding operation that reduces the particle size leads to an increase 
in the surface area. 
The most common approach for determining the surface area is based 
on the Brunauer-Emmett-Teller (BET) theory of adsorption. An excellent 
discussion of the principles and techniques involved has been given by 
Gregg and Sing [10]. Briefly. the theory states that most substances will 
adsorb a monomolecular layer of a gas under certain conditions of partial 
pressure (of the gas) and temperature. Knowing the monolayer capacity 
of an adsorbent (L e  the quantity of adsorbate that can be accommodated 
as a monolayer on the surface of a solid. the adsorbent) and the area of 
the adsorbate molecule. the surface area can. in principle, be calculated. 
Most commonly. nitrogen is used as the adsorbate at a specific partial 
pressure established by mixing it with an inert gas. typically helium. The 
adsorption process is carried out at liquid nitrogen temperatures (-195C). 
It has been demonstrated that, at a partial pressure of nitrogen attainable 
when it is in a 30% mixture with an inert gas and at -195C, a monolayer 
is adsorbed onto most solids. Apparently. under these conditions the 
polarity of nitrogen is sufficient for van der Waals forces of attraction between 
the adsorbate and the adsorbents to be manifest. The kinetic energy 
present under these conditions overwhelms the intermolecular attraction between 
nitrogen atoms. However. it is not sufficient to break the bonding 
between the nitrogen and dissimilar atoms. The latter are most often more 
polar and prone to van der Waals forces of attraction. The nitrogen molecule 
does not readily enter into chemical combinations, and thus its binding 
is of a nonspecific nature (Le., it enters into a physical adsorption); consequently 
, the nitrogen molecule is well suited for this role. 
The BET equation is 
A(P O/P - 1) 
1 +-A 
where' A is the grams of adsorbate per gram of adsorbent, Am the value of 
that ratio for a monolayer, P the partial pressure of the absorbate gas, Po 
the vapor pressure of the pure adsorbate gas. and C a constant. The 
constant C is temperature-dependent, as are P and PO; consequently. measurements 
are made under isothermal conditions. The equation is that of 
a straight line. and the inverse of the sum of both the slope [(C - 1) I 
AmC] and the y intercept (llAmC) gives Am' In an experiment it is necessary 
to measure at various values of P; Po can be obtained from the literature. 
The other values are then readily calculated. Often the constant C 
is large and Equation (2) then simplifies to: 
A single-point determination (e. e.. using only one value of P) is then 
possible. Knowing the specific weight of adsorbate (Am) in a monolayer. 
it is possible to calculate the specific surface area (SSA) of the sample 
using the following equation:

Preformulation Testing 11 
SAA = 
m 2 
where N is the Avogadro number, AN2 the area of the adsorbate molecule 
(generally taken to be, for nitrogen, 16.2 x 10-20 m2 per molecule), and 
MN2 the molecular weight of the adsorbate. 
Several experimental approaches are available that enable rapidity and 
convenience as well as accuracy and precision. Volumetric techniques represent 
the classic approach, and the modern instrumentation available has 
made the procedure convenient. Gravimetric and dynamic methods are also 
available. The latter methods measure the adsorption process by monitoring 
the gas streams, using devices such as thermal conductivity detectors 
and transducers. 
An example using a dynamic method is illustrated in the data given in 
Table 5 for a sample of sodium epicillin. In addition to the instrument, 
the requirements are a supply of liquid nitrogen, several gas compositions, 
a barometer, and several gas-tight syringes. Briefly, the procedure entails 
passing the gas over an accurately weighed sample contained in an 
appropriate container immersed in liquid nitrogen, removing the liquid nitrogen 
when the adsorption is complete (as signaled by the instrument) , 
warming the sample to about room temperature, and measuring (via the 
instrument) the adsorbate gas released (column 3 of Table 5). Calibration 
is simply performed by injecting known amounts of adsorbate gas into the 
proper instrument port (columns 4 and 5 of Table 5). The other terms in 
the table are calculable; P is the product of the fraction of nitrogen in the 
gas mixture (column 1) and the ambient pressure. The weight of nitrogen 
adsorbed is calculated from the ideal gas law. Slight adjustments of these 
values are made in actual practice as outlined by the various instrument 
manufacturers. Plotting the second column against the last column gives 
Table 5 Specific Surface Area of a Sample of Sodium Epicillina 
%N2 Vol NZ 
Wt N
2 absorBed 
in Signal used Signal 
[A(PO/P - 1)]-1 He P/PO (area) (ml) (area) x 10 (g) 
4.9 0.0483 178 0.050 137 0.576 677.5 
9.7 0.0956 152 0.080 150 0.922 1130.9 
20.0 0.1972 248 0.100 238 1.153 2043.6 
29.9 0.2948 312 0.130 331 1.499 2957.5 
aSample Weight: 0.1244 g. Atmospheric pressure: 762.7 mmHg. Temperature: 
297 K. Results: Slope, 9210; y Intercept, 238.2; correlation coefficient, 
0.99995; specific surface area, 3.0 m2 g-1.

Table 6 Relationship Between 
Diameter of a Particle and 
Specific Surface Area 
Specific surface 
Diameter area 
ClJm) (m 2 g-l) 
0.25 24 
0.50 12 
1.0 6 
2.0 3 
4.0 1.5 
10.0 0.63 
15.0 0.4 
20.0 0.3 
40.0 0.15 
Wadke, Serajuddin, and Jacobson 
the necessary slope and intercept values for calculating the specific surface 
area. For the sample of sodium epicillin a single-point determination (using 
the gas containing 29.9% nitrogen) gives a value of 2.8 m2 g-l for the surface 
area, which agrees well with the value of 3.0 obtained using the 
multipoint procedure. 
It is of interest to note the relationship between a diameter and the 
surface area of a gram of material of hypothetical monosized particles 
shown in Table 6. At relatively large diameters, the specific surface area 
is insensitive to an increase in diameter, whereas at very small diameters 
the surface area is comparatively very sensitive. If there is little difference 
in the properties of pharmaceutical interest between particles of about 
1 urn to those of about 0.5 urn, measurement of surface area is of little 
value. In the contrary instance, where a pharmaceutical property changes 
significantly for small particle-size changes, such measurements would be 
Some further prudence is necessary when interpreting surface area 
data. Thus, although a relatively high surface area most often reflects a 
relatively small particle size, it is not always true. A porous or a strongly 
agglomerated mass would be exceptional. Also, as implied previously. small 
particles (thus of high surface area) agglomerate more readily, and often 
in such a manner as to render the inner pores and surfaces inaccessible 
to water (as in a dissolution experiment). Thus, they act as if they are 
of much larger diameter than they actually are. 
Solid drugs administered orally for systemic activity must dissolve in the 
gastrointestinal fluids prior to their absorption. Thus, the rate of

Preformulation Testing 13 
dissolution of drugs in gastrointestinal fluids could influence the rate and 
extent of their absorption. Inasmuch as the rate of dissolution of a solid 
is a function of its solubility in the dissolution medium, the latter could 
influence absorption of the relatively insoluble drugs. As a rule of thumb, 
compounds with an aqueous solubility of greater than 1% wIv are not expected 
to present dissolution-related absorption problems. In the application 
of this rule. however. one must consider the anticipated dose of the 
drug and its stability in the gastrointestinal fluids. A highly insoluble 
drug administered in small doses may exhibit good absorption. For a drug 
that is unstable in the highly acidic environment of the stomach, high solubility 
and consequent rapid dissolution could result in a decreased bioavailability. 
For these reasons. aqueous solubility is a useful biopharmaceutical 
parameter. The solubility of every new drug must be determined 
as a function of pH over the physiological pH range of 1 to 8. If the 
solubility is considered too low or too high. efforts to alter it may be 
A. Determination of Solubility 
A semiquantitative determination of the solubility can be made by adding 
the solute in small incremental amounts to a fixed volume of the solvent. 
After each addition, the system is vigorously shaken and examined visually 
for any undissolved solute particles. When some solute remains undissolved, 
the total amount added up to that point serves as a good and rapid estimate 
of solubility. When more quantitative data are needed, a suspension 
of the solute in the solvent is shaken at constant temperature. Samples 
are withdrawn periodically, filtered, and the concentration of the solute 
in the filtrates is determined by a suitable method. Sampling is continued 
until consecutive samples show the same concentration. In the clarification 
of the suspension samples, it should be borne in mind that many filter 
media have a tendency to adsorb solute molecules. It is therefore advisable 
to discard the first few milliliters of the filtrates. 
Solubility of an acidic or basic drug is pH -dependent and, as mentioned 
earlier, must be determined over the pH range 1 to 8. Since such compounds 
favor their own pH environment dictated by their pKa values, it 
becomes necessary to adjust the pH values of their saturated solution. 
There is no general method for this pH adjustment. In some reported 
studies. authors used buffers of appropriate pH values [11-13], whereas 
others used hydrochloric acid or sodium hydroxide solutions [14-17]. Since 
the pH of an equilibrated suspension of an ionizable compound in a buffered 
system may not be the same as that of the starting buffer. it is essential 
to determine pH of the system after equilibration. 
Solubility determinations of poorly soluble compounds present their own 
unique problems. Higuchi and coworkers [18] demonstrated that the solubilities 
of such compounds could be overestimated due to the presence of 
soluble impurities. The saturation solubility of a poorly soluble compound 
is not reached in a reasonable length of time unless the amount of solid 
used is greatly in excess of that needed to saturate a given volume of 
solvent. This is because the final rate of approach to saturation is almost 
exclusively dictated by the surface area of the dissolving solid. For example 
J equilibrium solubilities of benzoic acid and norethindrone are 3.4 
mg Im1 and 6 ug1m1 J respectively. In solubili.ty experiments initiated using 
twice the amount of each compound needed to saturate the medium, the

14 Wadke, Serajuddin. and Jacobson 
amount of benzoic acid remaining undissolved at near saturation would be 
approximately 500 times that for norethindrone. As a result. equilibrium 
solubility of benzoic acid would be reached faster. If one uses a disproportionately 
greater amount of norethindrone to compensate for its lower 
solubility, the contribution of soluble impurities present to the total mass 
dissolved would become significant. Suspending a SOO-fold excess of a 
solid with 1% soluble impurities would show a fivefold increase in the apparent 
solubility. To overcome this problem one must use a specific assay 
for the estimation of dissolved chemical of interest. Alternately. one could 
use the facilitated dissolution method as developed by Higuchi and coworkers 
[18]. Here, the drug is dissolved in a water-immiscible solvent and then 
partitioned into the aqueous phase which in turn is assayed. The method 
is rapid and provides a fairly good estimate of true solubility. 
Many compounds in solution degrade. thus making an accurate determination 
of the solubility difficult. For such compounds, Ohnishi and 
Tanabe (19) proposed a kinetic method. It consists of the determination 
of rate constants and orders of reactions for degradation of the solute in 
a solution and a suspension. If Vs is the velocity of the overall degradation 
of the solute from the suspension, then 
where i is the order of the reaction in solution. ki is the rate constant for 
the ith-order reaction, and [S] is the saturation concentration. The quantities 
Vs ' ki' and i are measurable kinetic parameters that lead to the determination 
of [S]. Ohnishi and Tanable used this approach to determine 
the solubility of benzyl chloride. In an aqueous solution, benzyl chloride 
hydrolyzes. At 20C, the authors found that Vs' the velocity of hydrolysis 
of benzyl chloride in suspension, was 1. 67 x 10-6 mole min-l. Analysis 
of degradation from solutions showed that benzyl chloride in solution 
degraded by first- and second-order reactions. At 20C, k 1 and k2' the 
first- and second-order rate constants, were determined to be 2.9 x 10-4/min- 1 
and 3.6 x 10- 1 M-l min- 1 respectively. The equation describing degradation 
of benzyl chloride from suspension at 20C would be 
which can be solved to yield a value of 3.9 x 10-3 M for [S], the solubility 
of benzyl chloride at 20C. The method presupposes that the rate of dissolution 
of the drug in suspension is much greater than its rate of degradation 
from solution. It is also essential to determine all the kinetic parameters 
under identical conditions of temperature. pH. etc. 
Difficulty is also encountered in the determination of solubility of 
metastable forms that transform to more stable forms when exposed to solvents. 
Here a method based on the determination of intrinsic dissolution 
rates is applicable [20]. For many compounds exhibiting polymorphism, 
the metastable forms, when exposed to solvents, are sufficiently stable to 
permit measurement of initial dissolution rates. These initial dissolution 
rates. according to the Noyes-Nernst equation, are proportional to the 
respective solubilities of the polymorphic forms. The proportionality constant 
for the stable and metastable forms of a given compound is the same. 
Thus, determination of the intrinsic dissolution rates of the stable and

Preiormutation Testing 
metastable forms and the solubility of the stable form permits calculation 
of the solubility of the metastable form. 
B. pH-Solubility Profile 
The degree of ionization and therefore the solubility of acidic and basic 
compounds depends on the pH of the medium. The saturation solubility 
for such compounds at a particular pH is the sum total of solubility of 
ionized and unionized forms. Kramer and Flynn [11] investigated relative 
contributions of the protonated and free basic forms of several drugs to 
their total solubilities under different pH conditions. For ionizable compounds 
a solution may be saturated with respect to one species or the 
other depending on pH. The pH at which the solution is saturated with 
respect to both the ionized and unionized forms is defined as pH - 
the pH of maximum solubility. For a base, the equation relating total solubility 
(ST) to solubilities of protonated (BH+) and free (BP forms is 
ST,pH < pH = [BH+J (1 + K
max S (H 0+] 
where the subscript pH < pH indicates that the equation is valid only 
at pH values below the pH , subscript s denotes saturation species, K 
max a 
is the apparent dissociation constant, and (H 30+] is the hydronium ion concentration. 
The equation applicable at pH values higher than the pH 
. max 
ST ,pH> pH = [B 1 (1 + _[H~3:-0_+_]) 
max s K 
The corresponding equations for an acidic compound are 
(1 + 
ST ,pH> pH = [A J 
max s 
where [A ] and [AH] denote concentrations of ionized and unionized forms. 
Since ionizable compounds may be available in free or salt forms, one 
could use either in solubility experiments. For example, Serajuddin and 
Jarowski [17] studied the solubility behavior of phenazopyrldine free base 
and its hydrochloride salt over the pH range 1 to 10. Their findings are 
presented in Figure 3. Phenazopyridine. a base with a pK a of 5.2, exhibits 
maximum solubility at pH 3.45 (pH ). It should be noted here that, demax 
pending on the starting material, in the region of pH experimentally 
determined solubilities are higher than the equilibrium solubilities. This

, 5 
:::i 4 
:::l ... g 3
F B 
Wadke, Serafuddin, and Jacobson 
Figure 3 pH-Solubility profiles of phenazopyridine hydrochloride (6) and 
phenazopyridine base (0) at 37C. Solubilities are expressed as hydrochloride 
salt equivalents. Lower pH values than that of a saturated solution 
of the salt in water (point B) were adjusted by stepwise addition of 
HCI solution (curve BA) i higher pH values were obtained by addition of 
NaOH solution (curve B'D). Similarly, pH values lower and higher than 
that of a saturated solution of base in water (point C) were also adjusted 
by the addition of HCI and NaOH solutions I respectively. Curve BE represents 
supersaturation of phenazopyridine base solution. Curve DF was 
:litted theoretically by using 0.037 mg ml-1 as base solubility and 5.20 as 
the pKa. Points B and Fare, respectively, apparent and theoretical pH  
[From Serajuddin, A. T. M., and Jarowski, C. I., J. Pharm. Sci.. max 
74: 142 (1985). Reproduced with the permission of copyright owner.] 
phenomenon is described as supersaturation. Supersaturated solutions are 
metastable and will precipitate excess solute in due course on standing. 
C. Solubility Product 
In a saturated solution of a salt with some undissolved solid, there exists 
an equilibrium between the excess solid and the ions resulting from the 
dissociation of the salt in solution. For a hydrochloride salt represented 
as BH+CI-, the equilibrium is 
where BH+ and Cl- represent the hydrated ions in solution. The corresponding 
equilibrtum constant K is given by

Preformulation Testing 17 
( 3) 
where each subscrtpted "a" denotes the appropriate activity. As a solid 
the activity of BH+Cl- is constant. Equation (3), therefore, reduces to 
( 4) 
The constant Ksp is known as the solubility product and determines the 
solubility of a salt. 
In practice Equation (4) can be modified substituting concentration for 
activity. For an ionizable drug as mentioned earlier total solUbility ST is 
the Sum total of [BH+J s and [B). Since [BH+]s ~ {B]. equation (4) for a 
hydrochloride salt reduces to 
K =ST{CI-] sp 
( 5) 
Equation (5) dictates that total solubility of a hydrochloride salt would decrease 
with an increase in the chloride ion concentration. This phenomenon 
is known as common ion effect. In Figure 3. the observed decrease in the 
total solubillty of phenazopyridine at pH values below pHmax was due to 
common ion effect. 
Since the gastric contents are high in chloride ion concentration. the 
common ion effect phenomenon suggests that one should use salts other than 
the hydrochloride to benefit fully from the enhanced solubility due to a salt 
form. Despite this. many drugs are used as hydrochloride salts. This is 
because solUbilities of most hydrochloride salts in the presence of chloride 
lon concentration normally encountered in vivo are sufficiently high. The 
suppression of solubility due to common ion effect under these conditions 
is not of sufficient magnitude to affect dissolution or bioavaiJability of 
these compounds. 
D. Solubilization 
When the drug substance under consideration is not an acidic or basic compound, 
or when the acidic or basic character of the compound is not amenable 
to the formation of a stable salt, other means of enhancing the solubility 
may be explored. The use of a more soluble metastable polymorph to 
enhance bioavailability of orally administered solids is one way to approach 
the problem. Other approaches to improve solubility or rate of dissolution 
include use of complexation and high-energy coprecipitates that are mixtures 
of solid solutions and dispersions. Riboflavin in solution complexes 
with xanthines , resulting in an increase in the apparent solubility of the 
vitamin [21,22]. The approach, however, has practical limitations. The 
primary requirement is that the complexing or solubilizing agent be physiologically 
inert. Thus, unless the solubilizer is an approved excipient, this 
approach is not recommended. In this regard. the use of water-soluble 
polymers to form high-energy coprecipitates is more acceptable. Griseofulvin 
is a water-insoluble, neutral polyethylene glycol antifungal antibiotic. Dispersions 
and solid solutions of griseofulvin in PEG 4000, 6000, and 20,000

18 Wadke. Serajuddin, and Jacobson 
dissolve significantly more rapidly than the wetted micronized drug. In 
the case of PEG 4000 and 20, 000, this treatment provided supersaturated 
solutions [23]. Subsequent studies with the PEG 6000 dispersion showed 
that, in humans, the dispersed drug was more than twice as available as 
from commercially available tablets containing the micronized drug [24]. 
In the majority of cases. efforts to alter solubilities of drugs are undertaken 
to improve the solubility. Occasionally, however, a less soluble form 
is desired. Thus, in the case of clindamycin , the less soluble pamoate 
salt is preferred over the soluble hydrochloride hydrate to circumvent the 
problem of the unpleasant taste of the drug [25]. Likewise, when a drug 
is inactivated by the acidity of gastric fluids, a less soluble form is preferred. 
Knowledge of solubilrty of a drug substance not only helps in making 
some judgment concerning its bioavailability but also is useful in the development 
of appropriate media for dissolution testing or for development of 
an injectable dosage form for certain pharmacological and comparative bioavailability 
studies. In the investigation of dissolution of drugs insoluble 
in a purely aqueous medium, a cosolvent may be used to provide sink conditions. 
In these situations, knowledge of solubility in water-miscible organic 
solvents such as lower molecular weight alcohols and glycols is useful. 
The latter data are also of use to a formulator in the development of 
dosage forms of drugs that are administered in very small doses. Here 
the drug is often dispersed among the excipients as a solution in an appropriate 
solvent. Solubility data are also useful to the formulator in 
choosing the right solvent for the purposes of granulation and coating. 
The use of a granulating solvent with a very high capacity to dissolve the 
active ingredient can lead to a phenomenon known as case hardening. Here 
the solute migrates and deposits on the periphery of granules during the 
drying operation. 
A good working knowledge of solubility is also essential at the preclinical 
stage for the proper interpretation of biological data. These data are 
invariably generated using extemporaneously prepared solutions/suspensions. 
An inappropriate choice of solvent in these studies could show an active 
drug to be inactive or a toxic one to be nontoxic because of inadequate 
solubilfty and consequent incomplete absorption, and result in wrong selection 
of a compound for further development. 
The absorption of solid drugs administered orally can be depicted by the 
following flowchart: 
Solid drug 
GI gluids 
dissolution .. Drug in 
solution in 
GI fluids 
absorption - Drug in 
where k 
and k a are rate constants for the dissolution and absorption 
processes, respectively. When dissolution is the significantly slower of the 
two processes (Le, , kd  k a) the absorption is described as dissolution 
rate-limited. Since dissolution precedes absorption in the overall scheme,

Preformulation Testing 19 
any change in the process of dissolution would influence the absorption. 
It is essential, therefore, to investigate the dissolution behavior of drug 
substances, especially those with moderate and poor solubility. Efforts 
are then undertaken to alter this process if deemed necessary. Also, a 
knowledge of comparative dissolution rates of different chemical (salt, ester, 
prodrug, etc .) and physical (polymorph, solvate, etc.) forms of a drug is 
necessary in selecting the optimum form for further development. 
A. Intrinsic Dissolution 
The dissolution rate of a solid in its own solution is adequately described 
by the Noyes-Nernst equation: 
AD(C - C) s ::: 
hV ( 6) 
dC / dt ::: dissolution rate 
A ::: surface area of the dissolving solid 
D ::: diffusion coefficient 
C ::: solute concentration in the bulk medium 
h ::: diffusion layer thickness 
V ::: volume of the dissolution medium 
Cs ::: solute concentration in the diffusion layer 
During the early phase of dissolution, Cs  C and is essentially equal to 
saturation solubility S. Surface area A and volume V can be held constant. 
Under these conditions and at constant temperature and agitation, Equation 
(6) reduces to 
dC::: KS 
K ::: AD/hV ::: constant. 
( 7) 
Dissolution rate as expressed in Equation (7) is termed the intrinsic dissolution 
rate and is characteristic of each solid compound in a given solvent 
under fixed hydrodynamic conditions. The intrinsic dissolution rate 
in a fixed volume of solvent is generally expressed as mg dissolved x 
(min-1 cm-Z). Knowledge of this value helps the preformulation scientist 
in predicting if absorption would be dissolution rate-limited. Kaplan [26] 
studied the dissolution of a number of compounds in 500 ml of medium 
ranging in pH from 1 to 8, at 37C, while stirring at 50 rpm. His experience 
suggests that compounds with intrinsic dissolution rates greater 
than 1 mg min-1 cm-2 are not likely to present dissolution rate-limited absorption 
problems. Those with rates below 0.1 mg min-1 cm-2 are suspect

20 Wadke, Serajuddin, and Jacobson 
and usually exhibit dissolution rate-limited absorption. For compounds 
with rates between 0.1 and 1. 0 mg min-1 cm-2, usually more information 
is needed before making any prediction. 
The determination of the intrinsic dissolution rate can be accomplished 
best using the rotating-disk method of Wood et 91. [27]. A schematic diagram 
of the Wood apparatus is shown in Figure 4. This method allows for 
the determination of dissolution from a constant surface. A constant surface 
is obtained by compressing the solid in a tablet die against a flat surface 
using a hydraulic press. The punch used is cut down in length; the 
punch is left in the die and secured in position using a rubber gasket. 
The assembly is then attached to the shaft of a constant-speed rotor. To 
study dissolution, the rotor assembly is lowered in the dissolution medium 
to a preset position, and the rotor is activated. The progress of the dissolution 
is followed by periodically sampling and assaying the dissolution 
medium for the dissolved solute. Alternately, the dissolution medium may 
be circulated through the cell of a spectrophotometer for continuous recording. 
When the temperature, the pressure used to prepare the constant 
surface, and the hydrodynamics of the system are properly controlled, the 
method provides very reproducible results. 
As expressed in Equation (7) dissolution rate of a compound is directly 
proportional to its solubility. However, solubilities of acidic and basic 
compounds, as mentioned earlier, are pH -dependent, and apparent deviations 
from the relationship expressed by Equation (7) have been reported 
for many ionizable drugs [28,29]. Thus, Serajuddin and Jarowski [17J 
noticed significantly different intrinsic dissolution behavior for phenazopyridine 
free base and its hydrochloride salt under apparently identical 
conditions. Their data are presented in Figures 5 and 6 and Table 7. It 
can be seen here that the rates of dissolution of the free base and its salt 
oStirrer Shaft 
Tablet Die 
1 I I 
o 1 2 3 4 5 
Scale ( ern) 
Figure 4 Schematic diagram of constant-surface assembly for the determination 
of intrinsic dissolution rates. [From Wood, J. H., Syarto , J. E., 
and Letterman, H., J. Pharm. Sci., 54: 1068 (1965). Reproduced with the 
permission of the copyright owner.]

Preformulation Testing 21 
CI 30 
~ 24 
0en en 
0 18 ~z
~ 12 <{ 
30 60 90 120 
Figure 5 Dissolution profiles of phenzopyridine hydrochloride from a 
surface area of 0.95 cm2 at 37C under pH-stat conditions. Key: (0) pH 
1.10: (0) pH 3.05: (D) pH 5.0: (.to) pH 7.0. Each point represents the 
mean  3D of three experimental values. 
Ol 500 E
~ 400 
300 ~z
30 60 90 
Figure 6 Dissolution profiles of phenazopyridine at 37e from a surface 
area of 0.95 cm2 under pH-stat conditions. Key: (0) pH 1.10: (e) 2.05: 
(0) pH 3.05: (D) pH 5. O. Each point represents the mean  3D of three experimental 
values. Data are expressed as hydrochloride salt equivalents.

22 Wadke. Serajuddin, and Jacobson 
Table 7 Intrinsic Dissolution Rates (J IA) and 
Solubilities in Bulk Media (Cs) of Phenazopyridine 
and Its Hydrochloric Salt at 37C 
J lA, mg cm-2 
pH of Cs' mg/ml 
medium Salt Base Salt or basea 
1.10 0.084 8.89 0.680 
2.05 0.94 3.200 
3.05 0.640 0.103 4.320 
5.0 0.638 0.013 0.090 
7.0 0.645 0.037 
~xpressed as hydrochloride salt equivalents. 
Source: Modified from Serajuddin , A. T. M., 
and Jarowski. C. I., J. Pharm. sct., 74: 142 
(1985). Reproduced with the permission of 
copyright owner. 
form are not directly proportional to their equilibrium solubilities. Also, the 
rates of dissolution for the two forms under identical medium pH conditions 
differ widely despite the constancy of equilibrium solubility. Such deviations 
from Equation (7) are explained on the basis of self-buffering action of dissolving 
species in the diffusion layer. The pH at the dissolving surface in 
these cases is different from the pH of the bulk medium. Under these conditions 
for the calculation of dissolution rate, one should use equilibrium 
solubility value at the pH of the dissolving surface. A good approximation 
of pH at the dissolving surface can be obtained by independently measuring 
pH of a suspension in the medium. 
B. Particulate Dissolution 
Particulate dissolution is another method of studying the dissolution of solids. 
Here no effort is made to maintain the surface area constant. A weighed 
amount of powder sample from a particular sieve fraction is introduced in the 
dissolution medium. Agitation is usually provided by a constant-speed propeller. 
Particulate dissolution is used to study the influence on dissolution 
of particle size, surface area, and mixing with excipients. Finholt [30] 
studied the dissolution of phenacetin granules prepared using different sieve 
fractions of the drug powder (Fig. 7). As expected. the rate of dissolution 
increased with a decrease in the particle size. Occasionally, however, one 
encounters an inverse relationship of particle size to dissolution, where particle 
size reduction decreases-or fails to improve-the dissolution. This 
may be explained on the basis of effective or available. rather than absolute, 
surface area; and it is caused by incomplete wetting of the powder. In such 
areas incorporation of a surfactant in the dissolution medium may provide the 
expected relationship.

Preformulation Testing 
0 125 
a' 100 E 
Source: From Schanker, L. S., J. Pharmacal. Exp. Ther., 
126: 283 (1959). Reproduced with the permission of the 
Williams & Wilkins Company. Baltimore. 
Although the definition of partition coefficient refers to distribution 
between two immiscible phases, in reality the lipid phase exhibits some 
finite solubility in the aqueous phase and vice versa. For this reason, in 
the determination of partition coeficient, both the aqueous and the organic 
phases are presaturated with respect to each other. The drug is then 
dissolved in either the aqueous or the organic phase, and the known volumes 
of the two phases are equilibrated by shaking. The phases are separated 
by standing or via centrifugation. For convenience, Leo et al. [38] 
suggested the use of centrifuge bottles fitted with glass stoppers for 
equilibration where centrifugation can be accomplished without further transfer 
of the liquid. The concentration of the solute is generally determined 
in one of the phases and the concentration in the other is obtained by 
difference. However, if there is a possibility that adsorption of the solute 
to glass may occur, both phases must be analyzed. Leo et al. [38] also 
caution against unnecessarily long shaking periods and vigorous shaking, 
which tends to produce emulsions. Some emulsions may not break even 
after centrifugation, thus giving incorrect partition coefficient values. The 
presence of electrolytes in the aqueous phase could also affect the partition 
coefficients of many solutes. This should be carefully investigated 
during preformulation study and must be taken into consideration while 
comparing values generated by different laboratories. 
Another method applicable for estimation of partition coefficients of a 
compound belonging to a family of structurally similar compounds utilizes 
a reverse phase HPLC system [39- 41]. Here, logarithmic value of partition 
coefficient (log P) is correlated linearly with the logarithmic value of HPLC 
capacity factor (log k) according to the following relationship:

Pretormuunion Testing 27 
~ 0 u. 
u. .. W0
0 i= 
i= -1 a: 
0. ... 
.2 Q. 
b.p.40-60 C ~ 
05 04 -03 -02 -01 0 01 
02 03 
Figure 8 Influence of the nature of organic phase on relation between 
partition coefficient and rate of intestinal absorption of some barbiturates. 
The aqueous phase used was pH 5.5 buffer. (From Kurz, H., Principles 
of drug absorption, in International Encyclopedia of Pharmacology and 
Therapeutics, Section 39B, Vol. 1, Pergamon Press, 1975.)

log P = a log k + b 
Wadke, Serajuddin, and Jacobson 
where a and b are constants. The capacity factor k is defined as 
t - t 
k = r 0 
where to is the column dead time and t r is the retention time of the solute. 
The advantages of the chromatographic method are that it (a) is fast, 
(b) uses micro samples, and (c) is suitable for substances containing impurities 
and for mixtures. However, this method requires a reference log 
P versus log k graph for structurally similar compounds. 
B. Ionization Constant 
Many drugs are either weakly acidic or basic compounds and, in solution, 
depending on the pH value, exist as ionized or un-ionized species. The 
un-ionized species are more lipid-soluble and hence more readily absorbed. 
The gastrointestinal absorption of weakly acidic or basic drugs is thus 
related to the fraction of the drug in solution that is un-ionized. The 
conditions that suppress ionization favor absorption. The factors that 
are important in the absorption of weakly acidic and basic compounds are 
the pH at the site of absorption, the ionization constant, and the lipid 
solubility of the un-ionized species. These factors together constitute the 
widely accepted pH partition theory [42- 46] . 
The relative concentrations of un-ionized and ionized forms of a weakly 
acidic or basic drug in a solution at a given pH can be readily calculated 
using the Henderson-Hasselbalch equations: 
[un-ionized form] 
pH ::: pKa + log [ionized form] for bases 
[ionized form] 
pH = PKa + log [un-ionized form] for acids 
( 8) 
Although Equations (8) and (9) tend to fail outside the pH limits of 4 
to 10, or when the solutions are very dilute (where the hydronium ion concentration 
is about equal to or greater than 5% of the total solute concentration). 
a useful estimate can still be made. To use these equations, however, 
it is necessary to know the pK a (the negative logarithm of the acidic 
ionization constant). The ionization constant refers to the following general 
The most prevalent acid and conjugate base types are HB, B- (e.g., acetic 
acid, acetate); HB-, B2- (e.g., bicarbonate, carbonate); and HB+, B (e.g., 
glycinium, glycine), respectively. 
Several methods are available for the determination of the ionization 
constant, and they are concisely described by Albert and Serjeant [47] 
and others [48]. For compounds with a reasonable solubility (about 0.01 M),

Preformulation Testing 29 
acid-base potentiometric titrations can be performed on 100-ml portions 
using titrants of about 0.1 molarity. The procedure entails the measurement 
of the pH as a fucntion of the amount of titrant added. Automatic 
titrimeters are well suited to this purpose. Calculations of the dissociation 
constant can then be made from these data; and, often, an accurate value 
can be obtained by measuring the pH at the half-neutralization point where 
the pH equals the pK a. If un-ionized and ionized forms of a drug in solution 
exhibit significantly different ultraviolet or visible absorption spectra I 
the absorbance data can be used for the determination of the ionization 
constant. Other methods for determining ionization constants include those 
based on the determination of solubility or partition coefficient as a function 
of pH of the aqueous phase and on conductimetric techniques. 
It is apparent from the Henderson-Hasselbalch equations that for acidic 
compounds the relative concentration of the un-ionized form would increase 
with a decrease in the pH of a solution. whereas the converse would hold 
for basic compounds. This fact is graphically illustrated in Figure 9. 
The stomach contents are acidic. ranging in pH from 1 to 3, whereas the 
pH of intestinal fluids ranges from 5 to 8. Hence, weakly acidic but not 
basic drugs would be preferentially absorbed from the stomach, whereas 
the intestine is the primary site for the absorption of bases. The dependency 
of absorption of weakly acidic and basic drugs on the pH of intestinal 
solution is illustrated by the data of Hogben and coworkers [46], shown 
in Table 9. Schanker .36]. who studied the absorption of a number of 
acidic and basic compounds from the rat colon, observed that weakly acidic 
compounds (pKa < 4.3) were absorbed relatively rapidly; those with pKa 
values ranging between 2.0 and 4.3 were absorbed more slowly; and strong 
acids (pKa > 2.4) were hardly absorbed. For bases, those with pKa values 
smaller than 8.5 were absorbed relatively rapidly; those with a pK a between 
9 and 12 were absorbed more slowly; and completely ionized quaternary 
ammonium compounds were not absorbed. Knowledge of the pKa of a drug 
is thus very useful in determining the most likely site of absorption of 
acidic and basic drugs. 
Z 50 
o 3.5 7.0 
10.5 14 
Figure 9 Correlation between pH, pKa and extent of ionization for acids 
(solid line) and conjugate acids of bases (dotted line) having pKa values 
of 3.5, 7.0, and 10.5. (From Kurz, H., Principles of drug absorption, in 
International Encyclopedia of Pharmacology and Therapeutics, Section 39B, 
Vol. 1, Pergamon Press, 1975.)

30 Wadke. Serajuddin, and Jacobson 
Table 9 Intestinal Absorption of Drugs from Solutions of Various pH 
Percent absorbed 
(pH range of intestinal solution) 
Drug pKa 3.6-4.3 4.7-5.0 7.2-7.1 6.0-7.8 
Aniline 4.6 40  7 48  5 58  5 61  8 
Aminopyrine 5.0 21  1 35  1 48  2 52  2 
p-Toluidlne 5.3 30  3 42  3 65  4 64  4 
Quinine 8.4 9  3 11  2 41  1 54  5 
5-Nitrosalicylic 2.3 40  0 27  2 <2 <2 
Salicylic 3.0 64  4 35  4 30  4 10  3 
Acetylsalicylic 3.5 41  3 27  1 
Benzoic 4.2 62  4 36  3 35  4 5  1 
p-Hydroxypropiophenone 7.8 61  5 52  2 67  6 60  5 
Source; Modified from Hogben , C. A. M., Tocco, D. J., Brodie, B. B., 
and Schanker, L. S., J. Pharmacol. Exp. Ther. , 125:275 (1959). Reproduced 
with the permission of The Williams 81 Wilkins Company, Baltimore. 
C. Permeation Across Biological Membranes 
In the assessment of absorption potential of drugs, in addition to the determination 
of the physical parameters discussed above, in vitro experiments 
using biological membranes are gaining increasing acceptance among the 
preformulation scientists. These techniques measure the rate of permeation 
of drugs in solution across the intestine of mouse or rat and provide very 
useful information pertaining to the absorption characteristics of drugs. 
Many of these techniques are adequately reviewed by Bates and Gibaldi 
The method first described by Crane and Wilson [50] and modified by 
Kaplan and Cotler [51] is very simple and reproducible. The apparatus 
of Crane and Wilson is shown in Figure 10. The technique utilizes an isolated 
segment of intestine of laboratory animal such as a rat or a mouse. 
The animal is fasted overnight but is allowed access to drinking water. It 
is then anesthetized using ether or chloroform, and the sm all intestine 
is removed via a midline incision of the abdomen. The intestine is rinsed 
in cold normal saline. After discarding approximately a 10 to 15-cm section 
from the pyloric end, the entire intestine is everted, using a bluntheaded 
steel or glass rod. The everted gut is stretched under a weight 
of 10 g and cut into two lO-em segments. A segment prepared in this 
manner is ligated at the distal end and attached at the proximal end to the

Preformulation Testing 
~ Tubing (E) 
Stopp-:'"er---....- I II 
Test Tube---"-I Polyethylene 
Figure 10 Test tube apparatus of Crane and Wilson. [From Crane, 
R. K., and Wilson, T. H., J. Appl. Physiol., 12:145 (1958). Reproduced 
with the permission of the copyright owner.] 
canulated end of tube E (see Fig. 10). A weight of 10 g is attached to the 
ligated end to keep the sac in a vertical position. The segment is suspended 
in about 80 ml of drug solution in a physiologically acceptable buffer, 
such as Krebs bicarbonate buffer. The drug-containing solution is preequilibrated 
at 37C and is maintained at this temperature during the experiment. 
The drug-containing solution is referred to as mucosal solution. 
A 2-ml aliquot of drug-free buffer, also preequilibrated at 37C and referred 
to as the serosal solution, is introduced into the sac via tube E. A 95: 5 
mixture of 02!C02 is continuously bubbled through the mucosal solution at 
a constant rate. The serosal solution is withdrawn at predetermined intervals 
and replaced with fresh, drug-free buffer. The concentration of the 
drug in the serosal fluid samples is determined using a suitable assay. 
Usually these experiments are carried out at different mucosal concentrations 
of drug. Constancy of amount transferred per unit time per unit 
concentration over a wide range of mucosal solution concentrations is indicative 
of passive transfer of drug. Passive transfer refers to a free diffusion 
across a barrier composed of channels of various sizes; no biologically 
active or electrochemical processes are involved. As the concentration 
gradient across the barrier is increased, the flux across the barrier also 
increases in direct proportion (Pick's first law). Kaplan and Cotler [51] 
studied a number of compounds using this technique and compared the

Preformulation Testing 33 
results to those obtained during in vivo experiments in dogs. Their results 
are shown in Figure 11. Of the 16 compounds studied, those that 
exhibited lag times of 15 min or less, and clearance values between 0.01 
and 0.04 ml min-1, showed no permeability-related problem when tested 
in vivo and administered in solution. Others with lag times of 50 to 60 
min and essentially unmeasurable clearance values showed poor in vivo 
absorption despite good dissolution characteristics. These data demonstrate 
the predictive value of the technique. The everted rat gut technique 
is also useful in the investigation of site of absorption in the intestine and 
in the determination of transport mechanisms [52]. 
Notwithstanding the usefulness of the everted rat gut technique. due 
caution must be exerted in the interpretation of the data so derived. Thus, 
Taylor and Grundy [53] report a very poor correlation between in vitro 
clearance values and in vivo absorption in rat and man for practolol and 
propranolol. The techniques based on use of isolated gut segments in 
vitro also tend to underestimate absorption potential since such segments 
lack a blood supply. In this regard, the in situ technique as described 
by Doluisio and coworkers [54] is a more reliable method for the calculation 
of absorption rates. In this technique an anesthetized male Sprague-Dawley 
rat is surgically prep ared such that the small intestine is exposed. Two 
syringes are connected using L-shaped glass canulae and secured using 
silk suture at the duodenol and ileal ends. After clearing the gut using 
perfusion fluid, the drug solution is introduced into the intestine. Aliquots 
of the lumen solution are then collected periodically at either the ileal or 
duodenal end. Chow and coworkers [55] used this technique to assess absorption 
potential of a series of ACE inhibitor prodrugs . Their data [56] 
presented in Table 10 showed good rank-order correlation between the firstorder 
absorption rate constant as determined in situ and percent of oral 
dose absorbed in vivo. 
Amidon [57] proposed calculation of a dimensionless parameter termed 
intestinal permeability from the data obtained using in situ rat gut technique. 
In this method, a solution of known concentration is perfused through a 
segment of rat intestine. After the gut wall is equilibrated with the 
Table 10 Relationship Between Absorption Rate Constants 
for a Series of ACE Inhibitors as Determined in vitro Using 
Doluisio Method and Their in vivo Absorption 
%Absorption Absorption rate constant 

Many drug substances can exist in more than one crystalline form with different 
space lattice arrangements. This property is known as polymorphism. 
The different crystal forms are called polymorphs. Occasionally, a solid 
crystallizes, entrapping solvent molecules in a specific lattice position and 
in a fixed stoichiometry, resulting in a solvate or pseudopolymorph. Many 
solids may be prepared in a particular polymorphic form via appropriate 
manipulation of conditions of crystallization. These conditions include 
nature of the solvent, temperature, rate of cooling, and other factors. 
Many times a solute precipitates out of solution so that the molecules in 
the resulting solid are not ordered in a regular array but in a more or 
less random arrangement. This state is known as the amorphous form. 
Usually shock cooling, a sudden change in the composition of the solvent 
of crystallization, or lyophilization results in an amorphous form. 
Different polymorphic forms of a given solid differ from each other 
with respect to many physical properties, such as solubility and dissolution, 
true density, crystal shape, compaction behavior, flow properties, and solidstate stability. It is essential. therefore, to define and monitor the solid 
state of a drug substance. Occasionally, it may be deemed necessary to 
actively search for a different polymorphic form to circumvent a stability, 
bioavailabilfty , or processing problem. The subject of polymorphism has

Preformulation Testing 35 
attracted considerable attention from preformulation scientists, and excellent 
reviews have appeared in the pharmaceutical literature [62- 66]  
A. Crystal Characteristics and Bioavailability 
Differences in the dissolution rates and solubilities of different polymorphic 
forms of a given drug are well documented in the pharmaceutical literature 
[67,68]. When the absorption of a drug is dissolution rate-limited, a more 
soluble and faster dissolving form may be utilized to improve the rate and 
extent of bioavailability. The work of Aguiar and others [69,70] on polymorphs 
of chloramphenicol palmitate and that of Miyazaki et al . [71] on 
cWortetracycline hydrochloride illustrate this point. 
Figure 12 shows comparative blood level data obtained in humans following 
oral administration of 1. 5 g of pure A and pure B forms of chloramphenicol 
palmitate and their mixtures [69]. These data show that the pure, 
more soluble form B was most bioavailable, whereas the pure, less soluble 
form A was least bioavailable , The bioavailability of the mixtures fell between 
these two extremes and was directly proportional to the concentration 
of B. 
Figure 13 shows intrinsic dissolution profiles for a and S forms of 
chlortetracycline hydrochloride. The in vivo data illustrated in Figure 14 
show that the more soluble S form is also more bioavailable. 
Wadke, Serajuddin, and Jacobson 
Figure 13 Dissolution curves of the a and S forms of chlortetracycline 
hydrochloride from compressed disks in water at 37C. [From Miyazaki, S., 
Arit, T., Hori, R., and Ito, K., Chem. Pharm. Bul1., 11:638 (1974). Reproduced 
with the permission of the Pharmaceutical Society of Japan.J 
earlier, the effect of polymorphism on bioavailabllity is mediated via enhanced 
dissolution. Hence, a deliberate attempt to uncover polymorphism 
with the intention of improving bioavailabllity should be undertaken only 
when there is reason to believe that the absorption is likely to be dissolution 
rate-limited. Obviously, for relatively soluble compounds this approach 
may not be warranted. 

8. Crystal Characteristics and Chemical Stability 
For drugs prone to degradation in the solid state, the physical form of 
the drug influences the rate of degradation. For example. aztreonam, a 
monobactam antibiotic. exists in needlelike u- and dense spherical s-crystalline 
forms. In the presence of high humidity (37C/75% RH), the a form 
undergoes B-Iactam hydrolysis more readily with a half-life of about 6 
months whereas the B form under identical conditions is stable for several 
years [72]. Inasmuch as two crystal forms of a labile drug could exhibit 
widely different solid-state stabilities, a preformuation scientist might consider 
changing the crystal form to alleviate and possibly eliminate a stability 
problem. This approach is demonstrated by the data presented in Figure 
15 for an experimental drug. Under stress conditions, the anhydrous crystalline 
form of this experimental drug degraded rapidly with a half-life of 
about 18 weeks. A solvate form of the drug under the same conditions 
was essentially stable. Desolvation of the solvate caused by excessive 
heat resulted in a new crystal form distinct from the anhydrous and solvate 
forms. The desolvated form under the test conditions degraded most 
rapidly. This case history illustrates not only the possible use of a polymorphic 
form to solve a stability problem but also the importance of controlling 
processing variables so that the integrity of the selected form is 
C. Crystal Characteristics and Tableting Behavior 
In a typical tableting operation, flow and compaction behaviors of the powder 
mass to be tableted are important considerations. These properties, 
among others, are related to the morphology, tensile strength, and density 
of the powder bed. As mentioned earlier, two polymorphic forms of the

38 Wadke, Seroiuiidin, and Jacobson 
same drug could differ significantly with respect to these properties. The 
morphology of a crystal also depends on crystal habit. The latter is a 
description of the outer appearance of a crystal. When the environment 
in which crystals grow changes the external shape of the crystals without 
altering their internal structure j then a different habit results. Crystal 
habit is influenced by the presence of an impurity, concentration, rate of 
crystallization, and hydrodynamics in the crystallizer. 
cole et al , [73] describe compaction processes as "packing of particles 
by diffusion into void spaces, elastic and plastic deformation, fracture and 
cold working and. finally. compression of the solid material." One or more 
of these subprocesses may be affected by crystal form and habit. Some 
investigation of polymorphism and crystal habit of a drug substance as it 
relates to pharmaceutical processing is desirable during its preformulation 
evaluation. especially when the active ingredient is expected to constitute 
the bulk of the tablet mass. Shell [74] studied the crystal habit and the 
tableting behavior of nine different lots of an experimental drug. Using 
single-crystal X-ray data and X-ray powder diffraction patterns, he found 
that the ratio of intensities at diffraction angles of 12. 09 and 8. 72 correlated 
well with the tableting behavior of the nine lots as [udged by an experienced 
operator. Summers et al . [75] showed that different polymorphs 
of sulfathiazole, barbitone. and asprin differed significantly in their compression 
characteristics. Likewise, 1maizumi and coworkers {76] observed 
that the crystalline form of indomethacin yielded tablets with better hardness 
characteristics than the amorphous form. 
D. Crystal Characteristics and Physical Stability 
Although a drug substance may exist in two or more polymorphic forms, 
only one form is thermodynamically stable at a given temperature and pressure. 
The other forms would convert to the stable form with time. This 
transformation may be rapid or slow. When the transformation is not rapid, 
the thermodynamically unstable form is referred to as a metastable form. 
In general, the stable polymorph exhibits the highest melting point, the 
lowest solubility, and the maximum chemical stability. A metastable form 
nevertheless may exhibit sufficient chemical and physical stability under 
shelf conditions to justify its use for reasons of better dissolution or ease 
of tableting. When use of a metastable form is recommended, for whatever 
reason, a preformulation scientist must assure its integrity under a variety 
of processing conditions so that appropriate handling conditions may be 
Polymorphic transformations can occur during grinding, granulating, 
drying, and compressing operations. Digoxin, spironolactone, and estradiol 
are reported to undergo polymorphic transformations during the comminution 
process [77]. Phenylbutazone undergoes polymorphic transformation as a 
result of grinding and compression r78]. Granulation, since it entails the 
use of a solvent, can lead to a solvate formation. On the other hand, if 
the molecule is initially a solvate, the drying step in the process may cause 
transformation to an anhydrous crystalline or amorphous form {79]. 
Good knowledge of polymorphism and polymorphic stability is also 
needed to predict long-term physical stability of dosage forms. Yamaoka

Preformulation Testing 39 
et al , [80] observed cappinglike cracking in tablets of anydrous crystalline 
carbochromen hydrochloride upon storage under high-humidity conditions. 
This was determined to be due to transformation of the anhydrous form into 
a dihydrate. 
Even when the stable form is the form of choice, it is advisable to monitor 
the crystal form of each lot of raw material. In the case of calcium 
pantothenate, the preferred form is the crystalline form. In the preparation 
of multivitamin tablets, calcium pantothenate is granulated with a few 
other vitamins and appropriate excipients. An amorphous form of calcium 
pantothenate is known which readily reverts to the stable form when wetted 
with a variety of solvents used as granulating solvents. Use of the amorphous 
form in multivitamin tablets prepared by a granulation process is, 
however, not desirable because the polymorphic transformation renders 
the granulating mass sticky, making further granulation virtually impossible. 
E. Techniques for StUdying Crystal Properties 
Various techniques are available for the investigation of the solid state. 
These include microscopy (including hot-stage microscopy), infrared spectrophotometry, 
single-crystal X-ray and X-ray powder diffraction, thermal 
analysis. and dilatometry. Single-crystal X-ray provides the most complete 
information about the solid state. It is, however, tedious, time consuming, 
and, hence, unsuitable for routine use. 
Powder X-ray diffraction is both rapid and relatively simple, and is the 
method of choice. The powder X-ray diffraction pattern is unique to each 
polymorphic form: amorphous materials do not show any patterns or show 
one or two broad peaks attributable to the presence of shortrange ordering. 
Powder X-ray diffraction does not always indicate if the crystalline material 
is a true polymorph or a solvate. In Figure 16 are shown typical powder 
X-ray diffraction patterns for anhydrous amorphous, anhydrous crystalline, 
and crystalline trihydrate forms of the antibiotic epicillin [81,82]. 
Differential thermal analysis and differential scanning calorimetry are 
particularly useful in the investigation of polymorphism and in obtaining pertinent 
thermodynamic data. Figure 17 shows differential thermal analysis 
patterns for two polymer-phs and a dioxane solvate form of SQ 10,996 [83]. 
Curve (1) is the differential thermogram ~Jr form A of SQ 10,996. It shows 
a melting endotherm at approximately 195C, followed by a decomposition 
endotherm at 250 to 300C. Curve (2) represents the differential thermogram 
for form B. It shows a melting endotherm at 180C, followed by a 
small exotherm characterizing transition to form A, which then melts and 
decomposes at 190C and 250 to 300C, respectively. Curve (3) is a thermogram 
for the dioxane solvate. It is similar to that of form B with the 
exception that it has an extra endotherm at 140C. This is a de solvation 
endotherm; upon desolvation, form B is generated. Other events on the 
thermogram of the solvate are identical to those seen for form B. 
Desolvation endotherms are not always as distinct as shown in this example. 
In these situations thermogravimetric analysis is very useful. The 
thermogravimetric analysis pattern for the dioxane solvate showed a loss in 
weight that began at 105C and was complete at about 140C. The loss 
represented 13% of the total weight, which corresponded to a 1: 1 solvate.

Limits of acceptability and, therefore, compromises must 
be reasonably defined. Because the measurements of these aspects of stability 
as well as determination of the shelf life (or expiration date) for the final 
dosage form require long-term stability studies for confirmation, they can be 
expensive and time consuming. Consequently, the preformulation scientist 
must try to define those study designs and conditions that show the greatest 
probability of success [84]. The objective, therefore, of a preformulation 
stability program is to identify-and help avoid or control-situations where 
the stability of the active ingredient may be compromised. For a drug substance 
to be developed into a tablet dosage form, this objective may be 
achieved by investigating the stability of the drug under the following three 
categories: (1) solid-state stability of drug alone; (2) compatibility studies 
(stability in the presence of excipients); (3) solution phase stability (including 
stability in gastrointestinal fluids and granulating solvents). 
The basic requisite for the execution of these studies is the availability 
of a reliable stability-indicating analytical method. For the most part, in the 
case of a new drug the preformulation scientist will not have a fully validated 
analytical method available. However, a reasonably reliable HPLC procedure 
can usually be developed very quickly. Also, and often as a precursor to 
adopting an HPLC method, TLC is very useful. TLC analysis can be quickly 
performed and several systems primarily using different solvents for development 
can easily be examined. The purpose of this type of approach is to 
increase the probability of the detection of degradation and lor impurities and 
to prevent being surprised later in the program, when such findings can 
have a devastating effect on schedules. 
The preformulation scientist must also be aware of changes adopted in 
the synthesis of the drug substance. Although the molecule may be identical 
no matter what the synthetic route, its manner of presentation to the environment 
(as mediated by particle size, porosity, solvation, and lor crystalline 
form) can have profound effects on stability. This is not a rare occurrence. 
Inasmuch as the tablet formulations are multicomponent systems, the 
physical state of excipients could influence the stability of the active. The 
state of hydration of excipient materials can have strong effects on an active. 
Using aspirin as a model drug, Patel et al , [85] showed how excipients that 
are either hydrated or contain adsorbed moisture can effect drug stability. 
In their study these researchers identified the ratio of drug to excipient content, 
equilibrium to ambient humidities, and the trapping of moisture in 
closed containers, thereby changing the internal environment of the package 
as the important factor affecting product stability. 
A. Solid-State Stability 
Solid-state stability refers to physical as well as chemical stability. In this 
section only chemical stability will be discussed. Physical changes caused by 
polymorphic transitions and hygroscopicity are discussed in Sections VIII and 
X, respectively. 
In general, pharmaceutical solids degrade as a result of solvolysis, oxidation, 
photolysis, and pyrolysis. Any investigation of stability must begin 
with an examination of the Chemical structure, which provides some indication

Preformulation Testing 43 
of the chemical reactivity [86]. For example. esters. Iactams , and, to a lesser 
extent, amides are susceptible to solvolytic breakdown. The presence of unsaturation 
or of electron-rich centers makes the molecule susceptible to freeradical-
mediated or photocatalyzed oxidation. Strained rings are more prone 
to pyrolysis. With a number of possibilities suggested, it is possible to design 
the proper stress conditions to challenge the suspected weaknesses. 
The physical properties of the drug. such as its solubility. pK a, melting 
point, crystal form, and equilibrium moisture content, also influence its stability. 
As a rule. amorphous materials are less stable than their crystalline 
counterparts. For structurally related compounds. the melting point may 
indicate relative stabilities. For example, in a series of vitamin A esters. 
Guillory and Higuchi [87] observed that the zeroth -order rate constant for 
the degradation of the esters was inversely related to their fusion temperatures. 
The nature of thermal analysis curves may also help in a stability 
prognosis. Broad. shallow endotherms are suggestive of less stable, less 
homogeneous species. A relatively dense material may better withstand ambient 
stresses. For example, aminobenzylpenicillin trihydrate is denser [68) 
and more stable [88) than its anhydrous crystalline counterpart. 
The mechanisms of solid-state degradation are complex and difficult to 
elucidate [89-92). A knowledge of the exact mechanism. while always useful, 
is most often not the first objective. The stability study should be designed 
to identify the faetos that cause degradation of the drug. As indicated earlier, 
the most common factors that cause solid -state reactions are heat, light, 
oxygen. and, most importantly, moisture. Clearly, there can be, and most 
often there is. considerable interplay among these factors. Heat and moisture 
can cause a material with a propensity to react with oxygen to do so more 
rapidly; conversely, the presence of moisture can render a substance more 
heat-labile. In the conduct of stability studies, where stability is influenced 
by more than one factor. it is advisable to study one factor at a time, holding 
others constant. 
Solid-state reactions, in general. are slow, and it is customary to use 
stress conditions in the investigation of stability. The data obtained under 
stress conditions are then extrapolated to make a predicition of stability under 
appropriate storage conditions. This approach is not always straightforward, 
and due care must be exerted in the interpretation of the data. 
High temperatures can drive moisture out of a sample and render a material 
apparently stable that would otherwise be prone to hydrolysis. Degradative 
pathways observed at elevated temperatures may not be operant at lower 
temperatures. Some ergot alkaloids [93] degrade completely within a year 
when stored at temperatures above 45C; however, the rate is less than 1% 
per year below 35C. Above 65% relative humidity the 13 form of chlortetracycline 
hydrochloride transforms into the a form. the rate of transformation 
increasing with the increased aqueous tension. At or below 65% relative humidity, 
however, no transformation is observed [94). Despite these shortcomings. 
accelerated stability studies are extremely useful in providing an 
early and a rapid prognosis of stability. Such studies are also used to force 
formation of degradants in amounts sufficient for isolation and characterization. 
This information can then be used not only in the understanding of 
reaction kinetics but, if necessary, to set limits on amounts of degradants. 
Elevated Temperature Studies 
The elevated temperatures most commonly used are 30, 40, 50, and 60aC-in 
conjunction with the ambient humidity. Occasionally, higher temperatures 
are used. The samples stored at the highest temperature should be

44 Wadke, Serajuddin, and Jacobson 
examined for physical and chemical changes at frequent intervals, and any 
change. when compared to an appropriate control (usually a smaple stored 
at 5 or -20C), should be noted. If a substantial change is seen. samples 
stored at lower temperatures are examined. If no change is seen after 30 
days at 60C. the stability prognosis is excellent. Corroborative evidence 
must be obtained by monitoring the samples stored at lower temperatures 
for longer durations. Samples stored at room temperature and at 5C may 
be followed for as long as 6 months. The data obtained at elevated temperatures 
may be extrapolated using the Arrhenius treatment to determine 
the degradation rate at a lower temperature. Figure 19 shows the degradation 
of vitamin C at 50, 60, and 70C (95]. 
Figure 20 shows the elevated-temperature degradation data plotted in 
the Arrhenius fashion, where the logarithm of the apparent rate constant 
is plotted as a function of the reciprocal of absolute temperature. The 
plot is linear and can be extrapolated to obtain the rate constant at other 
temperatures. Most solid-state reactions are not amenable to the Arrhenius 
treatment. Their heterogeneous nature makes elucidation of the kinetic 
order and prediction difficult. Long-term lower temperature studies are, 
therefore, an essential part of a good stability program. As indicated by 
Woolfe and Worthington (93], even a small loss seen at lower temperatures 
has greater predictive value when the assay variation is less than 2% and 
the experimental design includes adequate replication. These authors suggest 
a 3- to 6-month study at 33C with three replications. 
Stability Under High-Humidity Conditions 
In the presence of moisture, many drug substances hydrolyze, react with 
other excipients, or oxidize. These reactions can be accelerated by exposing 
the solid drug to different relative humidity conditions. Controlled 
humidity environments can be readily obtained using laboratory desiccators 
containing saturated solutions of various salts [96]. 
in turn are placed in an oven to provide a constant temperature. The data 
of Kornblum and Sciarrone [97] for the decarboxylation of p -aminosalicylic 
acid show a dependence on the ambient moisture (Fig. 21). These data 
reveal that the zeroth-order rate constant as well as the lag time depend 
on the aqueous tension. Preformulation data of this nature are useful in 
determining if the material should be protected and stored in a controlled 
low-humidity environment, or if the use of an aqueous-based granulation 
system should be avoided. They may also caution against the use of excipients 
that absorb moisture significantly. 
Photolytic Stability 
Many drug substances fade or darken on exposure to light. Usually the 
extent of degradation is small and limited to the exposed surface area. 
However, it presents an aesthetic problem-which can be readily controlled 
by using amber glass or an opaque container, or by incorporating a dye in 
the product to mask the discoloration. Obviously. the dye used for this 
purpose whould be sufficiently photostable. Exposure of the drug substance 
to 400 and 900 footcandles (fc) of illumination for 4- and 2-week 
periods. respectively. is adequate to provide some idea of photosensitivity. 
Over these periods, the samples should be examined frequently for change 
in appearance and for chemical loss  and they should be compared to samples 
stored under the same conditions but protected from light. The change 
in appearance may be recorded visually or quantitated by instruments specially 
designed for comparing colors or by diffuse reflectance spectroscopy. 
For example, a sample of cicloprofen became intensely yellow after 5 days 
under 900 fc of light. The progress of discoloration could be readily

Stability to Oxidation 
The sensitivity of each new drug entity to atmospheric oxygen must be 
evaluated to establish if the final product should be packaged under inert 
atmospheric conditions and if it should contain an antioxidant. Sensitivity 
to oxidation of a solid drug can be ascertained by investigating its stability 
in an atmosphere of high oxygen tension. Usually a 40% oxygen atmosphere 
allows for a rapid evaluation. Some consideration should be given as to 
how the sample is exposed to this atmosphere. As shallow a powder bed 
as is reasonable should be used with an adequate volume of head space to 
ensure that the system is not oxygen-limited. Results should be compared 
against those obtained under inert or ambient atmospheres. Desiccators 
equipped with three-way stopcocks are useful for these studies. Samples 
are placed in a desiccator that is alternately evacuated and flooded with 
the desired atmosphere. The process is repeated three to four times to 
assure essentially 100% of the desired atmosphere. The procedure is somewhat 
tedious in that it must be repeated following each sample withdrawal. 
While flooding the evacuated desiccator, the gas mixture should be brought 
in essentially at the atmospheric pressure. This study can often be combined 
with an elevated temperature study, in that the samples under a 40% 
oxygen atmosphere can also be heated.

Preformulation Testing 
B. Compatibility Studies: Stability in the Presence 
of Excipients 
In the tablet dosage form the drug is in intimate contact with one or more 
excipients; the latter could affect the stability of the drug. Knowledge of 
drug-excipient interactions is therefore very useful to the formulator in 
selecting appropriate excipients. This information may already be in existence 
for known drugs. For new drugs or new excipents, the preformulation 
scientist must generate the needed information. 
A typical tablet contains binders, disintegrants, lubricants, and fillers. 
Compatibility screening for a new drug must consider two or more excipients 
from each class. The ratio of drug to excipient used in these tests is very 
much subject to the discretion of the preformulation scientist. It should 
be consistent with the ratio most likely to be encountered in the final tablet, 
and will depend on the nature of the excipient and the size and potency 
of the tablet. Table 11 shows ratios suggested by Akers [98]. 
Carstensen et ala [99] recommended drug/excipient ratios of 20: 1 and 1: 5 
by weight for lubricants and other excipients, respectively. Often the 
interaction is accentuated for easier detection by compressing or granulating 
the drug-excipient mixture with water or another solvent. 
An illustration of importance of drug/excipient ratio on the drug stability 
is presented in Figure 22 [100]. These data show that the stability of 
captopril-a drug prone to oxidative degradation-in mixtures with lactose 
monohydrate was inversely proportional to its concentration. Similar observations 
were made for stability in mixtures with microcrystalline celluslose 
and starch [101]. 
The three techniques commonly employed in drug-excipient compatibility 
screening are chromatographic techniques using either HPLC or TLC, differential 
thermal analysis, and diffuse reflectance spectroscopy. 
Chromatography in Drug-Excipient Interaction Studies 
This involves storage of drug-excipient mixture both "as is" and granulated 
with water or solvents at elevated temperatures. The granulation may be 
carried out so that the mixture contains fixed amounts (e. g., 5-20%) of moisture. 
The mixtures can be sealed in ampules or vials to prevent any escape 
of moisture at elevated temperatures. If desired, the type of gas in the 
headspace can be controlled using either air, nitrogen, or oxygen. The 
samples are examined periodically for appearance and analyzed for any 
decomposition using HPLC or TLC. Unstressed samples are used as controls 
, Any change in the chromatograph, such as the appearance of a 
new spot or a change in the Rf values or retention times of the components, 
is indicative of an interaction. HPLC may be quantitated if deemed necessary. 
If significant interaction is noticed at elevated temperatures, corroborative 
evidence must be obtained by examining mixtures stored at lower 
temperatures for longer durations. If no interaction is observed at 50 
to 60C, especially in the presence of moisure and air, none can be expected 
at lower temperatures. Among the advantages of HPLC or TLC in this application 
are the following: 
Evidence of degradation is unequivocal. 
spots or peaks corresponding to degradation products can be isolated 
for possible identification. 
The technique can be quantitated to obtain kinetic data.

48 Wadke, Sera/uddin, and Jacobson 
Differential Thermal Analysis in Drug-Excipient 
Interaction Studies 
Thermal analysis is useful in the investigation of soUd-state interactions. 
Its main advantage is its rapidity. It is also useful in the detection of 
eutectics and other phase formations. Thermograms are generated for the 
pure components and their 1: 3, 1: I, and 3: 1 physical mixtures. In the 
absence of any interaction, the thermograms of mixtures show patterns 
corresponding to those of the individual components. In the event that 
interaction occurs, this is indicated in the thermogram of a mixture by the 
appearance of one or more new peaks or the disappearance of one or more 
peaks corresponding to those of the components. Figure 23 [102] shows 
separate thermograms of cephradlne , a broad-spectrum antibiotic, and tour 
excipients, namely, N-methylglucamine, tromethamine, anhydrous sodium 
Table 11 Suggested Excipient IDrug Ratio in CompatibU1ty Studies 
Weight excipient per unit weight drug 
(anticipated drug dose, mg) 
Excipient 1 5-10 25-50 75-150 )150 
Alginic acid 24 24 9 9 9 
Avice! 24 9 9 9 4 
Cornstarch 24 9 4 2 2 
Dicalcium phosphate 24 24 9 9 9 
Lactose 24 9 4 2 1 
Magnesium carbonate 24 24 9 9 4 
Magnesium stearate 1 1 1 1 1 
Mannitol 24 9 4 2 1 
Methocel 2 2 2 2 1 
PEG 4000 9 9 4 4 2 
PVP 4 4 2 1 1 
Starch 1500 1 1 1 1 1 
Stearic acid 1 1 1 1 1 
Talc 1 1 1 1 1 
Source: Modified from Akers, M. J., Can. J. Pharm. Sci., 11:1 (1976). 
Reproduced with the permission of the Canadian Pharmaceutical Association.

Preformulation Testing 
6 9 12 15 
Figure 22 Degradation ot captoprU in presence of lactose Fast-Flo at 700C 
and 75% RH. 
carbonate. and trisodium phosphate dodecahydrate. Figure 24 shows the 
thermograms for the corresponding four mixtures. Only the thermogram 
for the mixture with anhydrous sodium carbonate retains the significant 
cephradine exotherm at about 200C. An investigation of the stability at 
50C at cephradine in the presence of these excipients showed that all the 
excipients. with the exception of anhydrous sodium carbonate. had deleterious 
eftects on stability. 
Interpretation of thermal data is not always straightforward. When two 
substances are mixed. the purity of each is obliterated. Impure materials 
generally have lower melting points and exhibit less well-defined peaks in 
thermograrns. In the absence of an interaction, this effect is usually small , 
By using the ratios tor the mixtures suggested above, an insight can usually 
be obtained as to whether an interaction has occurred. The temperature 
causing thermal events to occur can be high. depending on the materials. 
If too high, the condition may be too stressful-forcing a reaction that 
might not occur at lower temperatures. Finally. if an interaction is indicated. 
it is not necessarily deleterious. The formation of eutectics, if not 
occuring at so low a temperature as to physically compromise the final product. 
is acceptable. The same may be true for compound or complex formation 
and solid solution or glass formation. Because thermal analysis involves 
heating. often it may be difficult to interpret the loss of features 
in the presence of. for example, polyvinylpyrrolidone. The latter melts at 
a relatively low temperature and. once Iiquld , may dissolve the drug.

50 Wadke, Serajuddin, and Jacobson 
tc} (d) 
50 100 150 200 250 50 100 150 200 250 
Figure 23 Thermograms of pure materials: (a) cephradine; (b) N-methylglucamine; 
(c) tromethamine; (d) anhydrous sodium carbonate; (e) trisodium 
phosphate dodecahydrate , [From Jacobson, H., and Gibbs, 1. S., 
J. Pharm. Sci 62: 1543 (1973). Reproduced with the permission of the 
copyright owner. J 
Diffuse Reflectance Spectroscopy in Drug-Excipient 
Interaction Studies 
Diffuse reflectance spectrophotometry is a tool that can detect and monitor 
drug-excipient interactions [103J. In this technique solid drugs, excipients, 
and their physical mixtures are exposed to incident radiation. A portion 
of the incident radiation is partly absorbed and partly reflected in a diffuse 
manner. The diffuse reflectance depends on the pacldng density of 
the solid, its particle size, and its crystal form, among other factors. 
When these factors are adequately controlled, diffuse reflectance spectroscopy 
can be used to investigate physical and chemical changes occurring 
on solid surfaces. A shift in the diffuse reflectance spectrum of the drug 
due to the presence of the excipient indicates physical adsorption, whereas 
the appearance of a new peak indicates chemisorption or formation of degradation 
product. The method of preparation of the drug-excipient mixture 
is very critical. Equilibration of samples prepared by dissolving the 
drug and the excipient in a suitable solvent, followed by removal of the 
solvent by evaporattion, provides samples that are more apt to show small

Preformulation Testing 51 
changes in the spectrum. The dried solid mixture must be sieved to provide 
controlled particle size. When a suitable solvent in which both the 
drug and excipient are soluble is not available, sample equilibration may be 
effected using suspensions. Changes in the diffuse reflectance spectra may 
be apparent in freshly prepared sample mixtures, indicating potential incompatibilities. 
In other instances, they may become apparent when samples 
are stressed. In the latter case, diffuse reflectance spectroscopy can be 
used to obtain kinetic information. Thus, Lach and coworkers [104] used 
the technique to follow interactions of isoniazid with magnesium oxide and 
with lactose in the solid state at elevated temperatures [104]. The data 
were used to approximate the time needed for the reactions to be perceptible 
when samples are sotred at 25C. In a like manner, Blaug and Huang 
[105] studied ethanol-mediated interaction between dextroamphetamine sulfate 
and spray-dried lactose in solid mixtures. 
50 100 150 
200 250 
Figure 24 Thermograms of mixtures of cephradine with (a) N-methylglucamine; 
(b) tromethamine; (c) trisodium phosphate dodecahydrate; (d) 
anhydrous sodium carbonate. [From Jacobson, H., and Gibbs, I., J. Pharm. 
Sci., 62: 1543 (1973). Reproduced with the permission of the copyright 

52 Wadke. Seraiutidin, and Jacobson 
C. Solution Phase Stability 
Even for a drug substance intended to be formulated into a solid dosage 
form such as a tablet, a limited solution phase stability study must be 
undertaken. Among other reasons. these studies are necessary to assure 
that the drug substance does not degrade intolerably when exposed to gastrointestinal 
fluids. Also, for labile drugs. the information is useful in 
selection of granulation solvent and drying conditions. ThUS, the stability 
of the dissolved drug in buffers ranging from pH 1 to 8 should be investigated. 
If the drug is observed to degrade rapidly in acidic solutions. al 
less soluble or less susceptible chemical form may show increased relative 
bioavailabtllty , Alternately, an enteric dosage form may be recommended 
for such a compound. Erythromycin is rapidly inactivated in the acidic environment 
of the stomach. Stevens et al , [106] recommend the use of 
 I \ x I  x 
\ I  
0 I 0 x \ I dD 
0" I <, .a 
10 3.0 5.0 7.0 9.0 
C!I o 
Figure 25 pH-Rate profile for the degradation of ampicillin in solution at 
35C. Apparent rate constants in buffers: ., HCI-KCI; 0, citric acidphosphate; 
x , H2B03-NaOH. Rate constants at zero buffer concentration: 
0, citric acid-potassium citrate; -, NaH2P04-Na2HP04; 0, citric acid-phosphate 
buffer. [Modified from Hou , J. P., and Poole, J. W., J. Pharm. ScL. 
54: 447 (1969). Reproduced with the permission of the copyright owner.]

Preformulation Testing 53 
relatively insoluble propionyl erythromycin lauryl sulfate (erythromycin estolate) 
to circumvent this problem. The work of Boggiano and Gleeson [107] 
shows that other salts, such as stearates and salts of carboxylic acids. are 
less satisfactory because hydrochloric acid of the gastric juice readily displaces 
the relatively weakly acidic anions and dissolves the antibiotic as the 
soluble hydrochloride salt. The estolate, being a salt of a very strong 
acid, lauryl sulfuric acid, is not affected by the hydrochloric acid. It remains 
undissolved and potent even after prolonged exposure to gastric acid. 
The availability of pH-rate profile data is sometimes useful in predicting 
the solid -state stability of salt forms or the stability of a drug in the presence 
of acidic and basic excipients , The pH -rate profile for ampicillin. a 
broad-spectrum s-Iactem antibiotic (Fig. 25), shows that the antibiotic is 
significantly less stable in both acidic and basic solutions [108]. Indeed, 
both the hydrochloride and the sodium salt of ampicillin are significantly 
less stable as solids, compared to free ampicillin, when exposed to moisture. 
Compounds containing sulfhydryl groups are susceptible to oxidation in the 
presence of moisture. These compounds are more stable under acidic conditions 
[109]. If a drug substance is judged to be physically or chemically 
unstable when exposed to moisture. a direct-compression or nonaqueous 
solvent granulation procedure is to be recommended for the preparation of 
tablets. Before using a nonaqueous solvent for this purpose, stability of 
the drug in the solvent must be ascertained-since many reactions that 
occur in aqueous solutions may take place in organic solvents. Reactions 
in solution proceed considerably more rapidly than the corresponding solidstate 
reactions. Degradation in solution thus offers a rapid method for the 
generation of degradation products. The latter are often needed for the 
purposes of identification and synthesis (to study their toxicity where appropriate) 
and the development of analytical methods. 
In addition to the physicochemical parameters described heretofore, information 
pertaining to certain other properties, such as density, hygroscopicit 
flowability, compactibility, compressibility, and wettability. is useful to the 
formulator. These properties influence the process of manufacture and are 
important considerations when the active drug constitutes the major portion 
of the final dosage form. 
A. Density 
Knowledge of the absolute and bulk densities of the drug SUbstance is very 
useful in forming some idea as to the size of the final dosage form. Obviously, 
this parameter is very critical for drugs of low potency, which 
may constitute the bulk of the final granulation of the tablet. The density 
of solids also affects their flow properties. In the case of a physical mixture 
of powders, significant difference in the absolute densities of the components 
could lead to segregation. 
B. Hygroscopicity 
Many drug substances exhibit a tendency to adsorb moisture. The amount 
of moisture adsorbed by a fixed weight of anhydrous sample in equilibrium

54 Wadke, Serajuddin, and Jacobson 
with the mosture in the air at a given temperature is referred to as equilibMum 
moisture content. The significance of adsorbed moisture to the stability 
of the solids has already been discussed. Additionally, the equilibrium 
moisture content may influence the flow and compression characteristics of 
powders and the hardness of final tablets and granulations. The knowledge 
of the rate and extent of moisture pickup of new drug substances permits 
the formulator to take appropriate corrective steps when problems are anticipated. 
In general, hygroscopic compounds should be stored in a well-closed 
container. preferably with a desiccant. 
The sorption isotherms showing the equilibrium moisture contents of a 
drug substance and excipients as a function of a relative vapor pressure 
may be determined by placing samples in desiccators having different humidity 
conditions. Zografi and his coworkers [110, 111] designed specialized 
equipments for more precise determination of the rate and extent of moisture 
sorption. Proper processing and storage conditions of drugs may be 
selected on the basis of sorption isotherms. A solid deliquesces or dissolves 
in the adsorbed layer of water when the relative humidity of atmosphere 
exceeds that of its saturated soltuton [112]. The latter condition is 
called critical relative humidity, or RHO' The dissolution of a crystalline 
solid into adsorbed water (surface dissolution) would not be expected to 
occur below RHO' However, Kontny et al. [111J recently showed that mechanical 
processing of solids such as grinding. milling. micronization, compaction. 
etc., can induce changes in their reactivity toward water vapor. 
As a result, the surface dissolution of drug may occur at a lower humidity. 
which may lead to chemical and physical instability problems during subsequent 
storage. Thus. whenever possible. preformulation study should 
be conducted with the form of material to be used in the final formulation. 
The moisture contents of excipients can also influence the physicochemical 
properties of solid dosage forms. The analysis of sorption isotherms 
of excipients such as cellulose and starch derivatives indicates that water 
may exist in at least two forms. "bound" ("solidIike") and "free" [113]. 
These two types of water may be differentiated by measuring heat of sorption 
and by DSC and nuclear magnetic resonance studies. It has been suggested 
that serious stability problems may be avoided by minimizing free 
water in the excipients. On the other hand, it has been observed that 
the removal of unbound water reduces the ability of microcrystalline cellulose 
[114] and compressible sugar [113] to act as direct-compaction materials. 
This is because free water is needed to provide plasticity to these 
systems. Free water on the external surface of powders can also affect 
powder flow [115]. 
C. Flowability 
The flow properties of powders are critical for an efficient tableting operation. 
A good flow of the powder or granulation to be compressed is necessary 
to assure efficient mixing and acceptable weight uniformity for the 
compressed tablets. If a drug is identified at the preformulation stage to 
be "poorly flowable , " the problem can be solved by selecting appropriate 
excipients. In some cases, drug powders may have to be precompressed 
or granulated to improve their flow properties. During the preformulation 
evaluation of the drug substance, therefore, its flowability characteristic 
should be studied, especially when the anticipated dose of the drug is large.

Preformulation Testing 
-J-_"""'- Sliding Shutter 
Figure 26 Schematic diagram of the apparatus for measuring angle of 
repose. [From Pilpe1, N., Chern. Process Eng., 46: 167 (1965). Reproduce 
with the permission of the publlsher , Morgan-Grampian, London.] 
Amidon and Houghton [116] discussed various methods of testing powde 
flow. Some of these methods are angle of repose, flow through an orifice, 
compressibility index, shear cell, etc. No single method, however, can 
assess all parameters affecting the flow. 
When a heap of powder is allowed to stand with only the gravitational 
force acting on it. the angle between the free surface of the static heap 811 
the horizontal plane can achieve a certain maximum value for a given powde 
This angle is defined as the static angle of repose and is a common way of 
expressing flow characteristics of powders and granulations. For most pha 
maceutical powders, the angle-of-repose values range from 25 to 45, with 
lower values indicating better flow characteristics. 
There are a number of ways to determine the angle of repose. The ex 
act value of the measured angle depends on the method used. The value 
of the angle of repose determined from methods where the powder is pourei 
to form a heap is often distorted by the impact of the falling particles. 
The method described by Pilpe1 [117] is particularly free of this distortion 
The apparatus used by Pilpel is shown in Figure 26. It consists of a container 
with a built-in platform. The container is first filled with the powder, 
which is then drained out from the bottom, leaving a cone on the 
platform. The angle of repose is then measured using a cathetometer. 
The angle-of-repose measurement has some drawbacks as a predictor of 
powder flow in that it lacks sensitivity. For example, in a study reported 
by Amidon et al , [116], sodium chloride, spray-dried lactose, and Fast-Flo 
lactose showed similar angles of repose, but their rates of flow through a 
6-mm orifice were quite different. Therefore, the use of more than one 
method may be necessary for the adequate characterization of powder flow. 
In general, acicular crystals (because of cross-bridging), materials 
with low density, and materials with a static charge exhibit poor flow. 
Grinding of acicular crystals generally results in an improvement in the 
flow. For other powders and granulations, incorporation of a lubricant

56 Wadke, Serajuddin, and Jacobson 
or glidant helps alleviate the problem. For powders with poor flow. usually 
a granulation step is suggested. 
D. Compaclibllity ICompressibllity 
Tablet formulations are multfcomponent systems. The ability of such a mixture 
to form a good compact is dictated by compressibility and compactibilfty 
characteristics of each component. 
Lueuenberger and Rohera [1181 defined "compressibility" of a powder 
as the ability to decrease in volume under pressure. and "compactibilltyll 
as the ability of the powdered material to be compressed into a tablet of 
specified tensile strength. Some indication of the compressibility and compactibility 
characteristics of a new drug substance alone and in combination 
with some of the common excipients should therefore be obtained as part 
of the preformulation evaluation. Use of a hydraulic press offers one of 
the simplest ways to generate such data. Powders that form hard compacts 
under applied pressure without exhibiting any tendency to cap or chip can 
be considered as readily compactible. 
The compa.ctibillty of pharmaceutical powders can be characterized by 
studying tensile strength. indentation hardness, etc., of compacts prepared 
under various pressures [118,119]. Hiestand and Smith [119) used tensile 
strength and indentation hardness to determine three dimensionless parametersstrain 
index, bonding index, and brittle fracture index-to characterize tableting 
performance of individual components and mixtures. For the determination 
of tensile strength. compacts are placed radially [120] or axially [121) between 
two platens, and forces required to fracture the compacts are measured. 
Values of tensile strength calculated from the forces required radially and axUdly 
are called. respectively, radial and axial tensile strengths. Jarosz and 
Parrott (122) suggested that a comparison of radial and axial tensile strengths 
of compacts may indicate bonding strengths of compacts In two directions and 
may be related to their tendency toward capping. They also used tensile 
strength to evaluate the type and concentration of binders necessary to tmprove 
the compactibiUty of powders. 
Hardness is defined as the resistance of a solid to deformation and is 
primarily related to its plasticity. It is commonly measured by the static 
impression method (Brinell test). The schematics of BrineU test apparatus 
are shown in Figure 27. In this method [118], a hard. spherical indenter 
of diameter D is pressed under a fixed normal load F onto the mooth surface 
of a compact. The resulting indentation diameter d is measured or 
calculated using the depth h. The Brinell hardness number (BHN) is then 
calculated by using the following equation: 
BHN =-=-:~~~==:;::;;1fD(
D - IDZ - d 2 
Compressibility of powders is characterized from the density-compression 
pressure relationship according to the Heckel plot [123,124]. The 
relevant equation is given below: 
1 KP 
log 1 - P
r el 
= 2.303 + A 
where Prel is the relative density. P is the compressional pressure. and K 
and A are constants. Information about the extent of compression. the

Preformulation Testing 
Figure 27 Schematics of apparatus for Brinell test. 
H., and Rohera , B. D., Pharm. Res., 3:12 (1986). 
permission of The copyright owner.] 
[From Leuenberger, 
Reproduced with the 
yield value or the rmmrnum pressure required to cause deformation of solid, 
and the nature of deformation (plastic deformation, brittle fracture) I etc., 
may be obtained from the Heckel plot. 
E. Wettability 
Wettability of a solid is an important property with regard to formulation 
of a solid dosage form [125]. It may influence granulation of solids, penetration 
of dissolution fluids into tablets and granules I and adhesion of 
coating materials to tablets. Wettability is often described in terms of a 
contact angle that can be measured by placing drops of liquids on compacts 
of materials. The more hydrophobic a material is, the higher is the contact 
angle, and a value above 900 (using water) implies little or no spontaneous 
wetting. Crystal structures can also influence the contact angle. 
For example, a and S forms of chloramphenicol palmitate have contact angles 
of 122 and 1080 , respectively. Changes in surface characteristics may also 
occur on milling. A second method of determining wettability uses the 
Washburn equation [126]. In this method, the distance a liquid penetrates 
into a bed of powder or a compact is measured [127,128]. Problems associated 
with wettability of powders, namely, poor dissolution rate, low adhesion 
of film coating, and the like, may be solved by intimate mixing with 
hydrophilic excipients or by incorporating a surfactant in the formulation. 
As indicated earlier, selectivity is very important to the success of any preformulation 
program. To achieve this it is suggested that the data as they 
become available be analyzed to decide which areas warrant further scrutiny 
The following examples of preformulation studies where certain parameters

58 Wadke, Serajuddin, and Jacobson 
were not studied will illustrate this approach. These examples also illustrate 
one format for organizing and presenting the data. 
A. Preformulation Example A 
1. Background 
1. Compound: SQ 10,996 
2. Chemical name: 7-Chloro-5, ll-dihydrodibenz(b ,e] [1, 4]oxazepine5-
3. Chemical structure 
Molecular wt: 272.71 
4. Lot numbers: RR001RB and NN006NB 
5. Solvents of recrystallization: Lot RROOlRB was crystallized from 
chloroform. Lot NN006NB was crystallized from a mixed solvent 
system consisting of ethyl acetate and ethyl alcohol. 
6. Purity: Lots RROOlRB and NN006NB were 99.5 and 99.4% pure 
as determined by thin-layer chromatography. Lot RROOffiB contained 
one impurity whereas NN006NB contained two impurities. 
7. Therapeutic category: Anticonvulsant. antidepressant 
8. Anticipated dose: About 400 mg single dose 
II. Organoleptic properties: SQ 10,996 is a white, odorless, and almost 
tasteless powder. 
II. Microscopic examination: Microscopic examination of the "as is" powder 
revealed that the material was anisotropic and birefringent. Crystals 
in lot RROOlRB were highly faceted with no specific shape predominating. 
Crystals in lot NNOOGNB were essentially rectangular and not as highly 
faceted as in lot RROOffiB. The crystals in both lots ranged in diameter 
from 20 to 40 11m. On micronization, crystals in both lots were reduced 
to less than 10 11m in diameter. 
V. Physical characteristics 
1. Density: Densities of the two lots are shown in Table 12. 
2. Particle size 
a. Lot RR001RB "as is" was examined using a light-scattering 
technique. The data are presented in Table 13.

Preformulation Testing 59 
b. Particle size distribution of lot RROOlRB, micronized, was 
measured by the Coulter Counter. The data are shown in 
Table 14. 
3. Surface area: Surface area of lot RR001RB "as is" increased from 
O.5 to 2. 7 m2 g-l on microni zation  
4. Static charge: Neither lot exhibited any apparent static charge. 
Table 12 Example A: Densities of 
Two Lots of SQ 10,996 
(g em ) 
Method RR001RB NN006NB 
Fluff 0.34 0.45 
Tap 0.58 0.55 
Table 13 Example A: Particle Size 
Distribution of SQ 10,996, Lot 
Size Percentage 
3-10 11m 63.25 
Less than 20 J.lm 93.13 
Less than 40 11m 99.77 
Over 40 11m 0.25 
Table 14 Example A: Particle Size 
Distribution of Micronized SQ 10.996 
Size Percentage 
Less than 2.58 11m 2.5 
Less than 4.09 J.lm 17.8 
Less than 5.15 lJm 36.8 
Less than 10.3 11m 93.8 
Less than 16.4 lJm 98.8 
Less than 20.6 11m 100.0

60 Wadke. Seraiuiuiin, and Jacobson 
Micronization (RR001RB) resulted in the development of a surface 
static which was considered as manageable. 
5. Flow properties: Both lots of SQ 10, 996 exhibited good flow 
characteristics. An angle of repose measurement was not made. 
6. Compressibility: SQ 10,996 compressed well into a hard disk 
which displayed some tendency to chip. Tableting of SQ 10,996 
may need the incorporation of some granulating agent. 
7. Hygroscopicity: SQ 10,996 powder (RROOlRB) previously determined 
to be anhydrous by thermogravimetric analysis was found 
to pick up no moisture over a period of 8 weeks when exposed 
at room temperature to relative humidities of up to 90%. 
8. Polymorphism: Because of the drug's low aqueous solubilrty a 
possible bioavailability problem existed. To circumvent this 
possibility. an intensive search to uncover a more soluble form 
of SQ 10,996 was made. These studies showed that lyophilization 
of a solution of SQ 10,996 in p -dioxane resulted in the formation 
of a dioxane solvate. Exhaustive drying eliminated the 
dioxane, leaving a polymorphic form (clearly demonstrable by 
powder X-ray diffraction and thermal analysis) and referred to 
as form II. The solutrllity of form II in aqueous solvent systems 
at room temperature was found to be twice that of for I. 
Under a variety of temperature and humidity conditions form II 
was found to be physically and chemically stable. 
V. Solution properties 
1. pH of 1% Suspension: Approximately 7 
2. pKa: Not determined 
3. Solubility: The solubility data for the two lots of SQ 10,996 
are presented in Table 15. 
4. Effect of solUbilizing agents: Because solubility in the aqueous 
systems was considered as very low, attempts were made to 
solubilize SQ 10,996 (RROOIRB) using different surfactants. 
The data in Table 16 illustrate the influence of different surfactants 
on the solubility of SQ 10,996. 
5. Partition coefficient: The n-octanol/water partition coefficient 
was very much in favor of the organic phase. The exact value 
was not determined. 
6. Dissolution rates 
a. Intrinsic: In 1 L of distilled water with stirring at 100 rpm 
and at 37C, the intrinsic dissolution rate of lot RR001RB 
was considerably lower than 0.1 mg min-1 cm-2. 
b. Paritculate: Particulate dissolution studies were performed 
on loosely filled capsules containing 50 rng of SQ 10,996 
(RR001RB). Because the dissolution of SQ 10,996 was considered 
as too slow, attempts to improve the dissolution rate 
were made. These included physically admixing with a surfactant 
, coprecipitating with a surfactant, micronizing, and 
granulating with sodium lauryl sulfate. In these instances 
mixtures containing the equivalent of 50 mg of SQ 10,996 
were encapsulated and the dissolution investigated. The 
dissolution medium was 500 ml of 0.1 N HCI at 37C. and it

Preformulation Testing 
Table 15 Example A: Solubilities in Various Solvents of SQ 10,996 
Solubility at 25C (mg ml-1 ) 
pH 7.2 Phosphate buffera 
0.1 N ncie 
Isopropyl alcoholb 
Methyl alcoholb 
Ethyl alcoholt> 
aConcentration of the saturated solutions determined spectrophotometrically 
by determining the absorbance at 290 nm. 
bDetermined gravimetrically by evaporating a known volume of the filtered 
saturated solution and determining the weight of the residue. 
Table 16 Example A: Solubilities in Water of SQ 10,996 in the Presence 
of Various Surfactants 
-1 Solubility (mg ml ) at 25C in the presence of 
Surfactant Idrug 
Note: All determinations were made using the spectrophotometric method.

Wadke, SeY'ajuddin, and Jacobson 
was stirred at 50 rpm using the rotating basket. The concentration 
of the dissolved drug was determined spectrephotometrically. 
Because of the limited aqueous solubility 
of SQ 10,996. the dissolution was performed under nonsink 
conditions. The data are shown in Table 17. 
Stability (solid) 
1. Heat: SQ 10,996 (RR001RB) was stable after 4 weeks at 60C 
when assayed by thin-layer chromatog-raphy, 
2. Humidity: SQ 10.996 (RROOIRB) was stable after 8 weeks of 
exposure to 50% relative humidity at 60C when assayed by thinlayer 
3. Light: SQ 10,996 (NN006NB) after 2 weeks of exposure to 900 
fc of illumination at 33C and ambient humidity did not show any 
visible discoloration or degradation when assayed by the thinlayer 
Drug-excipient compatibility studies: Potential drug-excipient interactions 
were investigated using differential thermal analysis. 
Thermograms were obtained for the drug alone and for its 1: 3  1: 1, 
and 3: 1 physical mixtures and aqueous granulations with magnesium 
stearate, Sta-Rx 1500 starch, lactose. dicalcium phosphate dihydrate. 
talc, Avicel, cornstarch. PEG 6000, Plasdone C, and stear-ic acid. 
These studies failed to provide any evidence of potential drugexcipient 
Solution stability: Because of excellent solid-state stability and very 
poor aqueous solubility. the solution phase stability of SQ 10,996 
was not investigated. 
Recommendations: The major potential problem associated with SQ 
10.996 is its poor aqueous solubility and consequent slow dissolution. 
This may result in incomplete and slow absorption. Use of the more 
soluble form II may alleviate this problem. The good solid-state 
physical stability of form II favors its usage. The method used in 
the present study for the preparation of form II is cumbersome. and 
an easier method which can be used in the manufacture of the bulk 
formulation should be developed. In the absence of the latter, the 
micronization of SQ 10,996 followed by granulation with an aqueous 
solution of sodium lauryl sulfate appeal's as a promising alternative. 
An in vivo study comparing the different alternatives must be undertaken 
at the earliest apportunity to select the right form of SQ 
10,996. The stability prognosis for SQ 10,996 tablet dosage form is 
Preformulation Example B 
1. Compound: SQ 20,009 
2. Chemical name: l-Ethyl- 4[ (l-methylethylidene)hydrazinol)_lHpyrazolo[ 
3, 4-bJpyridine-5-carboxylic acid, ethyl ester, hydrochloride 
(1: 1) 
3. Chemical structure

Table 17 Example J  DJ::'::"JlulLu ....1 1:J~ hI ::;90, Ii d.dLUI:l l'.lt:.dluJt:..ll::. 
-1 Amount dissolved (mg ml ) 
1: 1 Physical 1: 1 Physical 
mixture 1: 1 Co-ppt mixture 1: 1 Co-ppt 
Time SQ 10,996 with with with with 
(min) form I PEG 6000 PEG 6000 PVP PVP 
10 0.001 0.001 0.002 0.002 0.006 
20 0.001 0.010 0.010 0.012 0.015 
30 0.004 0.016 0.013 0.018 0.015 
40 0.007 0.018 0.016 0.019 0.020 
50 - 0.020 0.017 0.021 0.021 
60 0.007 0.022 0.019 0.022 0.024 
SQ 10,996 
with sodium 
lauryl 804 
1: 1 Co-ppt 
"l:l .... 
'Q> .... 
~..... -. ;::s 

64 Wadke. Serajuddin. and Jacobson 
C14H20CINS02 Molecular wt 325.80 
4. Lot number: RR004RA 
5. Solvent of recrystallization: Acetone and aqueous hydrochloric 
6. Purity: Batch RR004RA contained 0.15% impurities as determined 
by paper chromatography 
7. Therapeutic category: Psychotropic 
8. Anticipated dose: 25 to 50 mg single dose 
II. Organoleptic properties: SQ 20,009 is a white powder with a characteristic 
aromatic odor and a bitter taste. 
III. Microscopic examination: The crystals of SQ 20,009 are needlelike. 
IV. Physical characteristics 
1. Density: Fluff and tap densities of SQ 20,009 were determined 
to be 3.0 and 3.5 g cm-3, respectively. 
2. Particle size (microscopic): The needlelike crystals of SQ 20.009 
ranged in width from 2 to 10 um, On grinding in a small ballmill 
the average length was reduced to about 30 urn from about 
80 um, 
3. Surface area: Not determined 
4. Static change: SQ 20,009 "as is" material exhibited Some static 
change. Grinding of SQ 20,009 did not significantly alter this 
5. Flow properties: As would be expected with materials having 
needlelike crystals, SQ 20,009 was not very free flowing. Grinding 
of SQ 20,009 significantly improved its flow. 
6. Compressibility: SQ 20,009 compressed well into a hard disk 
which did not show any tendency to cap or chip. 
7. Hygroscopicity: When exposed to 80% relative humidity at room 
temperature, SQ 20,009 did not pick up any moisture over a 24hr 
8. Polymorphism: The potential problem associated with SQ 20,009 
is its instability in solutions. For this reason a less soluble 
material is desirable. However, high solubility of the material 
makes it very unlikely that a sufficiently less soluble form can 
be discovered. For this reason investigation of polymorphism of 
SQ 20,009 was not undertaken. The free base of SQ 20,009 is an 
oily liquid and is not considered suitable for development into a 
development into a solid dosage form. 
V. Solution properties 
1. pH of 1% Solution: 1. 9 
2. pKa: 2.04

Preformulation Testing 65 
3. Solubility: SQ 20,009 is exceedingly soluble in water and lower 
alcohols. In aqueous systems it dissolved in excess of 400 mg 
ml-1, and in lower alcohols it dissolved in excess of 100 mg ml-I. 
Because of the very high solubility, an exact solubility determination 
was not attempted. 
4. Partition coefficient: Not determined 
5. Dissolution (particulate): Capsules containing 50 mg of SQ 
20,009 showed 100% dissolution in 15 min. The dissolution was 
studied in 1 L of water at 37C at a stirring rate of 100 rpm 
using the rotating basket. 
VI. Stability (solid) 
1. Heat: SQ 20,009 was found to be stable after 12 months at 50C 
and ambient humidity. 
2. Humidity: Exposure of SQ 20,009 to a high humidity of 80% 
relative humidity showed a visible discoloration after 8 weeks. 
The samples were not assayed. 
3. Light: Upon exposure to 900 fc of illumination at 33C and 
ambient humidity, SQ 20,009 showed signs of yellowing after 2 
VII. Drug-excipient compatibility studies 
1. Differential thermal analysis: Using weight ratios of 1:3, 1: I, 
and 3: 1, mixtures of magnesium stearate, stearic acid, lactose, 
and Avicel with drug showed an interaction only with magnesium 
2. Thin-layer chromatography: Mixtures of SQ 20,009 and magnesium 
stearate, lactose, stearic acid, and Sta -Rx 1500 starch were 
stable after 8 months at 50C and 12 months at room temperature. 
VIII. Solution stability: Aqueous solutions of SQ 20,009 showed rapid timedependent 
changes in the ultraviolet spectrum. Analysis of the data 
and the degraded samples showed that the Schiff base moiety of SQ 
20.009 underwent reversible hydrolysis to the corresponding hydrazine 
compound and acetone. The hydrolysis was pH -dependent. 
The half-lives for hydrolysis at 37C in media of different pH are 
shown in Table 18. The ester function in SQ 20,009 is also susceptible 
to hydrolysis. Studies with a structural analog l-ethyl-4butylamino-
Jjl -pyrazolo[ 3, 4- b] pyridine- 5-carboxylic acid, ethyl ester 
showed that the ester function underwent significant hydrolysis only 
under alkaline conditions. 
IX. Recommendations: Under acidic conditions SQ 20,009 hydrolyzes 
rapidly. To prevent the inactivation of SQ 20,009 by gastric acidity, 
use of the less soluble pamoate salt should be considered. The use 
of a less soluble form should also be considered for overcoming the 
problem of bitter taste. Because of the hydrolytic susceptibility of 
SQ 20,009, the use of aqueous-based gr-anulating agent should be 
avoided. Because of the relatively low dose of SQ 20,009 it poor 
flow ability is not likely to present any significant problems. Nevertheless 
SQ 20,009 should be ground, to improve its flow and allow for 
better homogeneity.

Table 18 Example B: Half-Lives 
for the Hydrolysis of SQ 20.009 
under Various pH Conditions at 
pH Condition T 1/2(min) 
0.1 N HCl 5 
0.01 N HCI 50 
pH 3.0 70 
pH 4.0 150 
Wadke. Seraiuiuiin, and Jacobson 
C. Performulation Example C 
I. Background 
1. Compound: Cicloprofen (SQ 20.824) 
2. Chemical name: a-Methyl-~-fluorene-2-aceticacid 
3. Chemical structure 
Molecular wt 238.29 
4. Lot numbers: EE003EA 
5. Solvents of recrystallization: Lots EE003EA and EEOO7EA were 
recrystallized out of acetone-water. Lot EE009EC was precipitated 
from aqueous ammoniacal solution with acetic acid. 
6. Purity: Lots EE003EA. EEOO7EA, and EE009EC contained 3.4, 
4.0, and 0.7% impurities, respectively. when assayed by thinlayer 
7. Therapeutic category: Nonsteroidal anti-inflammatory 
8. Anticipated dose: 100 to 250 mg 
II. Organoleptic properties: Cicloprofen is a pale cream-eolor-ed powder. 
It is practically odorless and tasteless. 
III. Microscopic examination: Microscopic examination of the three lots 
of cicloprofen showed that the powders were anisotropic and bire

Preformulation Testing 67 
fringent , The crystals were platy and ranged in diameter form 
1 to 50 um, 
IV. Physical characteristics 
1. Density: Fluff, tap, and true densities of lot EE007EA were 
determined to be 0.22, 0.33, and 1. 28 g cm-3, respectively. 
2. Particle size: The particle size range of unmilled lot EE007EA 
was 1 to 50 um, About 30% of the particles counted were below 
10 um, with those below 5 urn accounting for about 50% 
of the particles. 
3. Surface area: Surface area of lot EE007EA was determined to 
to be 1. 05 m2 g-1. On milling the surface area increased to 
3.50 m2 g-1. 
4. Static charge: All three lots of cicloprofen exhibited significant 
static charge. The problem of static charge was accentuated 
on milling. Milling after mixing with an excipient such as lactose 
helped significantly in reducing the charge. 
5. Flow properties: An three lots of cic1oprofen exhibited poor 
flow characteristics. On milling the material balled up in aggregates 
and had extremely poor flow. Granulating the milled and 
unmilled materials with water significantly improved the flow behavior. 
6. Compressibility: Cicloprofen compressed well into hard shiny 
disks which showed no tendency to cap or chip. 
7. Hygroscopicity: Cicloprofen adsorbed less than 0.1% moisture 
after storage in an atmosphere of 88% relative humidity at 22C 
for 24 hr. 
8. Polymorphism: Cicloprofen was recrystallized from 23 different 
single solvents and 13 solvent-water mixtures, and from supercooled 
melts. No conclusive evidence of the existence of polymorphism 
was obtained. 
V. Solution properties 
1. pH of 1% suspension: 5.3 
2. pRa: A value of 4.1 was obtained using the solubility and the 
spectrophotometric methods. 
3. Solbility: The solubility data for cicloprofen are presented in 
Table 19. 
4. Partition coefficient: Partition coefficient (oil/water) of cicloprofen 
between amyl acetate and pH 7.3 McIlvaine citrate-phosphate 
buffer at 37C was determined to be 17.0. 
5. Dissolution rates 
a. Intrinsic: In 1 L of pH 7.2. 0.05 M phosphate buffer at 
37C and at 50 rpm. the intrinsic dissolution rate of lot 
EE007EA was 2.1 x 10-3 mg min-1 cm-2. 
b. Particulate: Dissolution studies were performed on loosefilled 
capsules containing 200 mg unmilled and milled cicloprofen 
(EEOO7EA) and 400 mg of aqueous granulations of 
1: 1 mixtures of milled and unmilled cicloprofen with anhydrous 
lactose. The dissolution medium was 1 L of pH 7.2, 
0.05 M phosphate buffer at 37C. stirred at 50 rpm. Under 
these conditions the 1: 1 granulations dissolved the fastest 
within 30 min. The DT50% (the time needed for 50% dissolu

Table 19 Example C: Solubility 
of Cicloprofen in Various Solvents 
at 25C 
Solvent (mg 011-1) 
Water 0.06 
0.1 N HCl 0.01 
pH 7.0 buffer 1.10 
Isopropyl alcohol tV 50 
Methyl alcohol tV 80 
Ethyl alcohol tV 75 
Methylene chloride >100 
Wadke, Serajuddin, and Jacobson 
tion) values for the milled and unmilled cicloprofen were 50 
and 40 min. respectively. The slower dissolution of the 
milled material is believed to be due to powder agglomeration, 
resulting in the reduction of effective surface area. 
VI. Stability (solid) 
1. Heat: After 6 months of storage at 50C and ambient humidity, 
cicloprofen (all three lots) showed approximately 4% degradation 
when examined by thin-layer chromatography. 
2. Humidity: After 1 month at 40C and 75% relative humidity, cicloprofen 
(all three lots) showed no detectable degradation. 
3. Light: On exposure to light cicloprofen became yellow. Samples 
of cicloprofen exposed to 900 fc of illumination were intensely 
yellow after 5 days. The exposed samples contained as many as 
five degradation products when assayed by thin-layer chromatography 
and accounted for less than 2% of degradation of ciclcprofen. 
VII. Drug-excipient compatibility studies: Mixtures in the ratios 1:1, 
1: 3, and 3: 1 of cicloprofen (EE007EA) with alginic acid, microcrystalline 
cellulose. calcium phosphate. gelatin, lactose, magnesium stearate, 
polyvinylpyrrolidone. sodium lauryl sulfate. cornstarch. stearic acid, 
and talc were examined using differential thermal analysis. This 
study failed to provide any evidence of potential interaction. Storage 
of these mixtures for 1 week at 70C at 75% relative humidity 
and up to 8 weeks at 40C at 75% relative humidity. followed by 
their examination by thin-layer chromatography, failed to provide 
any evidence of degradation. 
VIII. Recommendations: The major potential problem areas associated with 
cicloprofen are its low solubility, poor dissolution, poor flow. and

Preformulation Testing 69 
poor photolytic stability. Attempts to find a more soluble polymorph 
were not successful. Granulation of the powder is needed to improve 
both its flow and its dissolution. Any shearing of cicloprofen should 
be avoided to contain the problem of the static charge. Because of 
its photolytic instability cicloprofen should be protected from light 
as much as possible. Consideration also should be given to incorporation 
of a yellow dye in the tablets to mask any light-catalyzed 
1. PharmaeopeiaI Forum, 13: 2505 (1987). 
2. Pharmacopeial Forum, 13: 2683 (1987). 
3. L. F. Prescott, R. F. Steel, and W. R. Ferrier, Clin, Pharmacol. 
Ther., 11:496 (1970). 
4. R. M. Atkinson, C. Bedford, K. J. Child, and E. G., Tomich, Nature, 
193: 588 (1962). 
5. J. Piccolo and A. Sakr, Pharm. Ind., 46: 1277 (1984). 
6. H. Weng and E. L. Parrott, J. Pharm. Sci., 73:1059 (1984). 
7. E. L. Parrott, Pharm. Mfg., 4:31 (1985). 
8. A. N. Martin, J. Swarbrick, and A. Cammarata, Physical Pharmacy, 
Lea & Febiger, Philadelphia, 1969. 
9. C. Orr, Jr. and J. M. Dalla Valle, Fine Particle Measurement, MacMillan, 
New York, 1959. 
10. S. J. Gregg and K. S. W. Sing, Adsorption, Surface Area, and 
Porosity, Academic Press, New York, 1967. 
11. S. F. Kramer and G. L. Flynn, J. Pharm. Sci., 61: 1896 (1972). 
12. S Miyazaki, M. Oshiba , and T. Nadai , Int. J. Pharm., 6:77 (1980). 
13. C. D. Herzfeldt and R. Kimmel, Drug. Dev , Ind. Pharm., 9: 967 
(1983) . 
14. Z. T. Chowhan, J. Pharm. ScL, 67:1257 (1978). 
15. A. T. M. Serajuddin and D. Mufson, Pharm. Res . 2: 65 (1985). 
16. W. H. StrengandH. G. H. Tan, Int. J. Pharm., 25:135(1985). 
17. A. T. M. Serajuddin and C. I. Jarowski, J. Pharm. sa., 74: 142 
(1985) . 
18. T. Higuchi, F. L. Shih, T. Kimura, and J. H. Rytting, J. Pharm. 
Sci., 68:1267 (1979). 
19. R. Ohnishi and K. Tanabe, Bull. Chem. Soc. Jap., 41: 2647 (1971). 
20. G. Milosovich, J. Pharm. sa., 53: 484 (1964). 
21. D. E. Guttman and M. AthaIye, J. Am. Pharm. Assoc. [Sci. Ed.]. 
49:687 (1960). 
22. D. A. Wadke and D. E. Guttman, J. Pharm. sei., 54:1293 (1965). 
23. W. L. Chiou and S. Riegelman, J. Pharm. Sci., 58: 1505 (1969). 
24. W. L. Chiou and S. Riegelman, J. Pharm. Sci., 60: 1377 (1971). 
25. A. A. Sinkula, W. Morozowich, and E. L. Rowe, J. Pharm. Sci., 
62: 1106 (1973). 
26. s. A. Kaplan, Drug Metab. Rev., 1:15 (1972). 
27. J. H. Wood, J. E. Syarto, and H. Letterman, J. Ph.arm, Sci., 54: 
1068 (1965).

70 Wadke, Serajuddin, and Jacobson 
28. K. G. Mooney, M. A. Mintun, K. J. Himmelstein, and V. A. Stella, 
J. Pharm. Sci., 70: 22 (1981). 
29. V. K. Prasad, R. S. Rapaka , P. W. Knight, and B. E. Cabana, Int. 
J. Pharm., 11: 81 (1982). 
30. P. Finholt, Influence of formulation on dissolution rate, in Dissolution 
Technology, Industrial Pharmaceutical Technology Section, Acad , Pharm. 
Sci., Washington, D.C., 1974, p. 109. 
31. D. C. Monkhouse and J. L. Lach, J, Pharm. Sci., 61:1431 (1972). 
32. J. W. McGinity, D. D. Manes, and G. J. Yakatan, Drug Dev. Commun., 
1: 396 (1974-75). 
33. N. Shah, R. Pytelewski, H. Eisen, and C. Jorowski, J. Pharm. Sci., 
63:1339 (1974). 
34. A. Hussain, J. Pharm, ScL, 61:811 (1972). 
35. O. Kausch, Vergleichende Untersuchungen der intestinalen Resorption 
von Opium. Alkoloiden mit Einem Beitrag zur Beantwortung der Frage 
nach der Abhangigkeit der Resorption von der chernischen Konstitution: 
Dissertation, Universitat des Saarlndes, Germany, 1968. Cited in Ref. 
36. L. S. Schanker, J. Pharmacol. Exp. Ther., 126:283 (1959). 
37. H. Kurz, Principles of drug absorption, in International Encyclopedia 
of Pharmacology and Therapeutics, Vol. 1, Sec. 39B, Pharmacology of 
Intestinal Absorption -Gast.ro-Intestinal Absorption of Drugs, Pergamon 
Press, Elmsford, NY, 1975. 
38. A. Leo, C. Hanseh , and D. Elkins, Chem. Rev., 71:525 (1971). 
39. R. Kaliszan , Chromatography, 1987, p. 19. 
40. F. Gaspari and M. Bonati, J. Pharm. Pharmacol.. 39:252 (1987). 
41. T. Braumann, G. Weber, and L. J. Grimme, J. Chromatography, 261: 
329 (1983). 
42. P. A. Shore, B. B. Brodie, and C. A. M. Hogben, J. Pharmacol. 
Exp. Ther 119: 361 (1957). 
43. L. S. Schanker, P. A. Shore, B. B. Brodie, and C. A. M. Hogben , 
J. Pharmacol. Exp. Ther . 120: 528 (1957). 
44. L. S. Schanker, D. J. Tocco, B. B. Brodie, and C. A. M. Hogben, 
J. Pharmacol. Exp. Ther., 123:81 (1958). 
45. C. A. M. Hogben, L. S. Schanker, D. J. Tocco, and B. B. Brodie, 
J. Pharmacol. Exp. Ther., 120: 540 (1957). 
46. C. A. M. Hogben, D. J. Tocco, B. B. Brodie, and L. S. Schanker, 
J. Pharmacol. Exp. Ther., 125: 275 (1959). 
47. A. Albert and E. P. Serjeant, The Determination of Ionization Constamts. 
Chapman & Hall, London, 1971. 
48. 1. M. Kolthoff, S. Bruckenstein, and R. G. Bates, Acids and Bases 
in Analytical Chemistry, Interscience , New York, 1966. 
49. T. Bates, and M. Gibaldi, Gastro-intestinal absorption of drugs, Current 
Concepts in Pharmaceutical Sciences: Biopharmaceutics, Lea and 
Febiger, Philadelphia, 1970, p. 57. 
50. R. K. Crane and T. H. Wilson, J. Appl. Physiol.. 12:145 (1958). 
51. S. A. Kaplan and S. Cotler, J. Pharm. Sci., 61: 1361 (1972). 
52. S. C. Penzotti and J. W. Poole, J. Pharm . Sci., 63: 1803 (1974). 
53. D. C. Taylor and R. U. Grundy, J. ~harm. Pharmacol., 27 [Suppl.J 
54. J. T. Doluisio, N. F. Billups , L. W. Dittert, E. T. Sugita, and J. V. 
Swintosky, J. Pharm. sa., 58:1196 (1967).

Preformulation Testing 71 
55. W. S. Chow, S. A. Ranadive, S. A. Varia, and K. J. Kripalani, 
Abstr. Acad. Pharm. Sci., 14: 109 (1984). 
56. S. A. Ranadive, Private communication. 1987. 
57. G. L. Amidon. Determination of intestinal wall permeabilities, in 
Animal Models for Oral Delivery in Man: In Situ and In Vivo 
Approaches. Am. Pharrn, Assn., Acad , Pharm. Sci., Washington. 
D.C . 1983, p , 1. 
58. G. L. Amidon, M. Chang, D. Fleisher, and R. Allen, J. Pharm. Sci., 
71:1138 (1982). 
59. G. L. Amidon and P. J. Sinko, Proc ; In. Symp. Control. ReI. Bioact. 
Mater.. 14: 20 (1987). 
60. G. L. Amidon, G. D. Leesman, and R. L. Elliot. J. Pharm. sa., 69: 
1362 (1980). 
61. G. L. Amidon, B. H. Stewart, and S. Pognar, J. Controlled ReI., 2: 
13 (1985). 
62. J. Haleblian and W. McCrone, J. Pharm. su., 58: 911 (1969). 
63. S. Rosenstein and P. Lamy, Am. J. Hasp. Pharm., 26:598 (1969). 
64. G. S. Mewada, D. R. Parikh, and S. Somsekhar-a, Sci. Cult . 39: 
378 (1973). 
65. J. Haleblian, J. Pharm. Sci., 64: 1269 (1975). 
66. S. R. Byrn. Solid State Chemistry of Drugs. Academic Press, New 
York, 1982. 
67. E. Shefter and T. Higuchi, J. Pharm. Sci., 52:781 (1963). 
68. D. A. Wadke and G. E. Reier, J. Pharm, ScL, 61:868 (1972). 
69. A. J. Aguiar, J. Krc, A. W. Kinkel, and J. C. Symyn. J. Pharm. 
Sci., 56:847 (1967). 
70. A. J. Aguiar and J. E. Zelmer, J. Pharm. ScL, 58:983 (1969). 
71. S. Miyazaki, T. Artta, R. Hori, and K. Ito, Chern, Pharm. Bull., 22: 
638 (1974). 
72. D. M. Floyd, O. R. Kocy, D. C. Monkhouse, and J. D. Pipldn , U.S. 
Patent 282636 (1981). 
73. E. T. Cole, J. E. Rees, and J. A. Hersey, Pharm. Acta Helv., 50: 
28 (1975). 
74. J. W. Shell, J. Pharm. sa., 52:100 (1963). 
75. M. P. Summers, R. P. Enever, and J. E. Carless, J. Piuirtn, 
Pharmacal., 28: 89 (1976). 
76. H. Imaizumi, N. Nambu , and T. Nagai, Chem. Pharm. Bull., 28:2565 
(1980) . 
77. A. T. Florence and E. G. Salole , J. Pharm. Pharmacal. 28: 637 
(1976) . 
78. H. G. Ibrahim, F. Pisano, and A. J. Bruno, J. Pharm. Sci. 66: 
669 (1977). 
79. A. T. M. Serajuddin, J. Pharm. Pharmacal. 38: 93 (1986). 
80. T. Yamaoka, H. Nakamerchi, and K. Miyata, Chem. Pharm. Bull., 
30: 3695 (1982). 
81. J. P. Hou and A. Restivo, J. Pharm. ScL, 64:710 (1975). 
82. J. P. Hou, Private communication, 1976. 
83. I. Gibbs, A. Heald, H. Jacobson, D. A. Wadke, and 1. Weliky, 
J. Pharm. sa., 65: 1380 (1976). 
84. D. C. Monkhouse, Drug Dev. Ind. Pharm., 10:1373 (1984). 
85. N. K. Patel, S. J. Patel, A. J. Cutie, D. A. Wadke, D. C. Monkhouse, 
and G. E. Reier, Drug Dev , Ind. Pharm., 14:77 (1988).

72 Wadke, Serajuddin, and Jacobson 
86. C. Runti, Farmaco [Prat.], 28:451 (1973). 
87. K. Guillory and T. Higuchi, J. Pharm. Sci., 51: 104 (1962). 
88. D. A. Johnson and G. A. Hardcastle, U.S. Patent 3,157,640 (1963). 
89. J. T. Carstensen, Theory of Pharmaceutical Systems. Vol. 2: Heterogeneous 
Systems, Academic Press, New York, 1973, p. 294. 
90. J. T. Carstensen, J. Pharm. sa., 63:1 (1974). 
91. S. R. Byrn, J. Pharm. sa., 65:1 (1976). 
92. D. C. Monkhouse and L. Van Campen, Drug. Dev . Ind. Ptuirm, , 10: 
1276 (1984). 
93. A. J. Woolfe and H. E. C. Worthington, Drug Dev. Commun., 1:185 
94. S. Miyazaki, M. Nakano, and T. Arita, Chem. Pharm. Bull., 24: 
1832 (1976). 
95. R. Tardif, Pharm. sa., 54:281 (1965). 
96. L. Mertes (ed.), Handbook of Analytical Chemistry, McGraW-Hill, 
New York, 1963, p. 3-29. 
97. S. S. Kornblum and G. J. Sciarrone, J. Pharm. sa., 53: 935 (1964). 
98. M. J. Akers, Can. J. Pharm. sa., 11:1 (1976). 
99. J. Carstensen, J. Johnson, W. Valentine, and J. Vance, J. Pharm. 
sa., 53: 1050 (1964). 
100. N. B. Jain, Private communication, 1987. 
101. N. B. Jain, K. W. Garren, and M. R. Patel, Abstr. Acad. Pharm. 
Sci., 12: 146 (1982). 
102. H. Jacobson and I. Gibbs, J. Pharm. sa., 62:1543 (1973). 
103. D. G. Pope and J. L. Lach, Pharm. Acta Helv .. 50:165 (1975). 
104. W. Wu, T. Chin, and J. L. Lach , J. Pharm. sa., 59: 1234 (1970). 
105. S. M. Blaug and W. T. Huang, J. Pharm. Sci., 61:1770 (1972). 
106. J. W. Stevens, J. W. Conine, and H. W. Murphy, J. Am. Pharm. 
Assoc. [Sci. Ed.], 48: 620 (1959). 
107. B. G. Boggiano and M. Gleeson, J. Pharrn. sa., 65:497 (1976). 
108. J. P. Hou and J. W. Poole, J. Pharm. sa., 58:447 (1969). 
109. W. E. Godwin, The Oxidation of Cysteine and Cystine. Ph.D. thesis, 
Oklahoma State University, 1962. 
110. L. Van Campen, G. Zografi, and J. T. Carstensen, Int. J. Pharm . 
5:1 (1980). 
111. M. J. Kontny, G. P. Grandolfi, and G. ZografL Pharm. Res., 4:104 
(1987) . 
112. L. Van Campen, G. L. Amidon, and G. Zografi, J. Pharm. sci., 72: 
1381 (1983). 
113. G. Zografi and M. J. Kontny, Pharm. Res., 3:187 (1986). 
114. R. Huettcnrauch and J. Jacob, Die Pharmazie, 32:241 (1977). 
115. N. A. Armstrong and R. V. Griffiths, Pharm. Acta. Helv., 45: 692 
116. G. E. Amidon and M. E. Houghton, Ph.arm, Mfg., July 1985, p . 21. 
117. N. Pilpel, Chem. Process Eng., 46:167 (1965). 
118. H. Leuenberger and B. D. Rohera, Pharm. Res., 3: 12 (1986). 
119. E. N. Hiestand and D. P. Smith, Powder Technol., 38: 145 (1984). 
120. J. T. Fell and J. M. Newton, J. Pharm. sa., 59: 688 (1970). 
121. S. T. David and L. L. Augsburger, J. Pharm. Sci . 63:933 (1974). 
122. P. J. Jarosz and E. L. Parrott, J. Pharm. Sci., 71:607 (1982). 
123. R. W. Heckel, Trans. Metallo Soc. AIME. 221,671 (1961). 
124. R. W. Heckel, ibid., 221,10001 (1961).

Preformulation Testing 73 
125. P. York, Int. J. Ptiarm  14:1 (1983). 
126. E. D. Washburn, Phys. Rev., 17:374 (1921). 
127. D. T. Hansford. D. J. W. Grant, and J. M. Newton, Powder Technol., 
26: 119 (1980). 
128. G. Buckton and J. M. Newton, J. Pharm. PhaY'macol.. 38:329 (1986).

Tablet Formulation and Design 
Garnet E. Peck 
Purdue University 
West Lafayette, 
George J. Baley and 
Vincent E. McCurdy 
The Upjohn Company 
Kalamazoo, Michigan 
Gilbert S. Banker 
University of Minnesota 
Health Sciences Center 
Minneapolis. Minnesota 
The formulation of solid oral dosage forms, and tablets in particular, has 
undergone rapid change and development over the last several decades with 
the emergence of precompression, induced die feeding I high -speed and now 
ultrahigh-speed presses, automated weight-control systems, the availability 
of many new direct compression materials, and the microprocessor control 
of precompression, compression, ejection forces, as well as upper punch 
tightness on tablet presses. Some of the newer tablet presses have tablet 
rejection systems that are operated by a computer. Computer-controlled 
tablet presses only require an operator to set up the press at the proper 
tablet weight and thickness (or pressure). The computer can then assume 
complete control of the run. Still other tablet presses only require the operator 
to provide a product identification code to make tablets within specifications 
previously established and stored in the computer memory. 
Most recently, new concepts and federal regulations bearing on btoavailability 
and bioequivalence, and on validation, are impacting on tablet 
formulation, design, and manufacture. 
Once, lavish gold-plated pills were manufactured and marketed with 
little knowledge of their pharmacological activity. Appearance and later 
stability of the dosage form were the prime requirements of pharmaceutical 
preparations. The introduction of the friable pill denoted in part the realization 
that solid medicinals must-in some fashion-disintegrate within the 
body for the patient to benefit from the drug. We now realize that disintegration 
and dissolution alone do not insure therapeutic activity. As only 
one example of this point, Meyer et al [1] presented information on 14 
nitrofurantoin products, which were evaluated both in vitro and in vivo. 
All products tested met USP XVIII specifications for drug content, disintegration 
time, and dissolution rate; however, statistically significant differences 
in bioavailability were observed. 

76 Peck, Baley, McCurdy, and Banker 
The design of a tablet usually involves a series of compromises on the 
part of the formulator, since producing the desired properties (e. g., resistance 
to mechanical abrasion or friability, rapid disintegration and dissolution) 
frequently involves competing objectives. The correct selection and 
balance of excipient materials for each active ingredient or ingredient combination 
in a tablet formulation to achieve the desired response (i. e., production 
of a safe l effective, and highly reliable product) is not in practice 
a simple goal to achieve. Add to this fact the need today to develop tablet 
formulations and processing methods which may be (and must in the future 
be) validated, and the complexity of tablet product design is further increased 
in contemporary pharmaceutical development. Increased competition 
among manufacturers (brand versus generic, generic versus generic, and 
brand versus brand) has necessitated that products and processes be costefficient. 
Thus cost of a raw material or a particular processing step must 
be considered before a final tablet formulation or manufacturing process is 
Tablet formulation and design may be described as the process whereby 
the formulator insures that the correct amount of drug in the right form 
is delivered at or over the proper time at the proper rate and in the desired 
location, While having its chemical integrity protected to that point. 
Theoretically, a validated tablet formulation and production process is one 
in which the range in the variation of the component specifications and 
physical properties of the tablet product quality properties is known from 
a cause and effect basis. It is further known that raw materials specifications, 
at their limits, and when considered as interaction effects of the 
worst possible combinations l cannot produce a product that is out of specification 
from any standpoint. Likewise a validated tablet-manufacturing 
process is one which, when all the operating variables are considered, at 
any extremes which could ever be encountered in practice, and under the 
worst possible set of circumstances, will produce products that are within 
specifications. Total validation of a tablet product includes all combination 
effects involving formulation, raw materials variables, and processing variables, 
as well as their interaction effects, to assure that any system produced 
will be within total product specifications. 
The amount or quantity of a drug which is sufficient to elicit the required 
or desired therapeutic response can be affected by several factors. 
In the case of cornpendial or official drugs, the dosage levels have been 
predetermined. With certain drugs (e. g., griseofulvin), the efficiency of 
absorption has been shown to depend on the particle size and specific surface 
area of the drug. By reducing the particle size of such drugs, the 
dosage level may be reduced by one-half or more and still produce the same 
biological response. 
The form in which the drug is absorbed can affect its activity. Most 
drugs are normally absorbed in solution from the gut. Since the absorption 
process for most orally administered drugs is rapid, the rate of solution 
of the drug will be the rate-limiting step from the point of view of 
blood level and activity. 
Thus, we must consider the contribution and influence of the active 
components and nonactive components-both separately and together-to 
measure their impact on the pharmacological response of any tablet system. 
The timing of administration may affect when and how a drug will act (and 
to a certain extent where it acts) as will be discussed further in Section

Tablet Formulation and Design 77 
IV. A. Also. the timing of administration may be crucial in order to reduce 
gastric irritation (uncoated strong electrolytes are often given following 
food); to reduce drug interactions with food (formation of insoluble complexes 
between the calcium of milk and several antibiotics). reducing their 
bioavailability; or to enhance the solubility and bioavailability of certain 
drugs in foods (notably fats) by their administration with foods (e.g . 
griseofulvin) . Depending on such timing factors plus the relationship and 
rationale of fast. intermediate. or slow drug release as well as other release 
considerations. a particular design and tablet formulation strategy is 
often indicated. 
Many excellent review articles have been written on tablet technology, 
including various formulation aspects. Coop er [2] presented a review monograph 
on the contributions from 1964 to 1968 in the areas of tablet formulation, 
processing. quality standards, and biopharmaceutics. Later, Cooper 
and Rees [3] continued the review and included similar topics covering the 
period 1969 to 1971. Recent book chapters on tablets include those by 
Banker [4] and Sadik [5]. 
The present chapter will detail the general considerations of tablet 
product design; will describe a systematic approach to tablet design, ineluding 
the practical use of preformulation data; will describe the commonly 
used tablet excipients with particular emphasis on their advantages and limitations 
or disadvantages; and will present some general tablet formulation 
approaches. Extensive references to the literature should provide the 
reader with directed reading on topics where additional information may be 
obtained. While it is impossible to exhaustively cover as broad a topic as 
tablet formulation and design in one chapter of a book, it is the goal of 
this chapter to cover the major concepts and approaches, including the 
most recent thought bearing on validation. optimization, and programmatic 
methods related to the formulation, design, and processing of compressed 
The first step in any tablet design or formulation activity is careful consideration 
of the preformulation data. It is important that the formulator 
have a complete physicochemical profile of the active ingredients available, 
prior to initiating a for-mulation development activity. Compilation of this 
information is known as preformulation. It is USUally the responsibility of 
the pharmaceutical chemistry research area to provide the data shown below 
on the drug substances. 
1. Stability (solid state): light, temperature, humidity 
2. Stability (solution): excipient-drug stability (differential thermal 
analysis or other accelerated methods) 
3. Physicomechantcal properties: particle size, bulk and tap density, 
crystalline form, compressibility, photomicrographs, melting point, 
taste, color, appearance. odor 
4. Physicochemical properties: solability and pH profile of solution I 
dispersion (water, other solvents) 
5. In vitro dissolution: pure drug, pure drug pellet, dialysis of pure 
drug, absorbability. effect of excipients and surfactants

78 Peck, Baley, McCurdy. and Banker 
The basic purposes of the preformulation activity are to provide a 
rational basis for the formulation approaches, to maximize the chances of 
success in formulating an acceptable product, and to ultimately provide a 
basis for optimizing drug product quality and performance. From a tablet 
formulator's perspective. the most important preformulation information is 
the drug-excipient stability study. The question then, for a new drug, or 
a drug with which the formulator lacks experience. is to select excipient 
materials that will be both chemically and physically compatible with the 
The question is compounded by the fact that tablets are compacts; and 
while powder mixtures may be adequately stable, the closer physical contact 
of particles of potentially reactive materials may lead to instability. The 
typical preformulation profile of a new drug is usually of limited value to 
the formulator in assuring him or her that particular drug-excipient combinations 
will produce adequate stability in tablet form. An added problem 
is that the formulator would like to identify the most compatible excipient 
candidates within days of beginning work to develop a new drug into a tablet 
dosage form rather than to produce a series of compacts. place them on 
stability, and then wait weeks or months for this information. 
Simon [6], in reporting on the development of preformulation systems. 
suggested an accelerated approach, utilizing thermal analysis, to identify 
possibly compatible or incompatible drug-excipient combinations. In his 
procedure, mixtures are made of the drug and respective excipient materials 
in a 1: 1 ratio and subjected to differential thermal analysis. A 1: 1 ratio 
is used, even though this is not the ratio anticipated for the final dosage 
form, in order to maximize the probability of detecting a physical or chemical 
reaction, should one occur. The analyses are made in visual cells. and 
physical observations accompany the thermal analysis. The thermograms 
obtained with the drug-excipient mixtures are compared to thermograms for 
the drug alone and the excipient alone. Changes in the termograms of the 
mixture. such as unexpected shifts, depressions, and additions to or losses 
from peaks are considered to be significant. Simon [6] has given an example 
of the type of information which may be obtained from such a study 
by the data shown in Figure 1. The thermal peak due to the drug alone 
was lost when the thermal analysis was run on the drug in combination with 
the commonly used lubricant. magnesium stearate. This was strong evidence 
for an interaction between these materials. It was subsequently confirmed 
by other elevated-temperature studies that the drug did decompose rapidly 
in the presence of magnesium stearate and other basic compounds. Simon 
has concluded the differential thermal analysis can aid immensely in the 
evaluation of new compounds and in their screening for compatibility with 
various solid dosage form excipients. The combination of visual and physical 
data resulting from differential thermal analysis of drugs with exciplents is 
suggested as a programmatic approach to the very rapid screening of the 
drug-excipient combinations for compatibility. 
Following receipt of the preformulation information. the formulator may 
prepare a general summary statement concerning the drug and its properties 
relative to tablet formulation. This statement must often also take into account 
general or special needs or concerns of the medical and marketing 
groups for that drug. A typical statement might be as follows. 
Compound X is a white crystalline solid with a pyridine odor and bitter 
taste, which may require a protective coating (fllm or sugar). It displays 
excellent compressing properties and has not been observed to possess any

Tablet Formulation and Design 
In presence of 
1 I 
0 120 140 160 
180 120 140 160 
Figu re 1 Thermograms showing the melting endotherm of triampyzine sulfate 
and loss of the endotherm in the presence of magnesium stearate. 
polymorphs. It is nonhygroscopic, has low solubility in water, and in moderately 
volatile. It is an acidic moiety with a pK a of 3.1 and a projected 
dose of 50 to 100 mg. The compound is soluble in organic solvents and 
aqueous media at pH 7.5. Below pH 5 it is sparingly soluble. In the dry 
state it is physically and chemically stable. This product, while requiring 
coating protection, must be designed for rapid drug dissolution release 
(the drug is an acidic moiety, presumably best absorbed high in the gut). 
No severe chemical stability problems are foreseen. The volatility of the 
tableted form must be checked, and special packaging may be required. 
Tablet product design requires two major activities. First, formulation activities 
begin by identifying the excipients most suited for a prototype formulation 
of the drug. Second, the levels of those excipients in the prototype 
formula must be optimally selected to satisfy all process Iproduct quality constraints. 
A. Factors Affecting the Type of Excipient Used in a 
Tablet Formula 
The type of excipient used may vary depending on a number of preformulation 
, medical, marketing, economic, and process /product quality factors, 
as discussed in the following sections. 
Only those excipients found to be physically and chemically compatible with 
the drug should be incorporated into a tablet formula. Preformulation

80 Peck Baley, McCurdy, and Banker 
studies should also provide information on the flow and bonding properties 
of the bulk drug. Excipients that tend to improve on flow (glidants) and 
bond (binders) should be evaluated for use with poor-flowing and poorbonding 
compounds. respectively. 
At the conclusion of a preformulation study, it may be known which 
tableting process [direct-compression or granulation (wet/dry)] will be appropriate 
for the drug. If it is not known for certain which tableting 
process is most appropriate after preformulation, then initial formulation 
efforts should concentrate on a direct-compression method since it is most 
advantageous. Direct compression is the preferred method of tablet manufacture 
for the following four major reasons: (a) It is the cheapest approach 
since it is a basic two-step process (if components are of the proper particle 
size), involving only mixing and compressing, and it avoids the most 
costly process of unit operatin g, drying. (b) It is the fastest, most direct 
method of tablet production. (c) It has fewer steps in manufacture and 
fewer formulation variables (in simple formulations). (d) It has the potential 
to lead to the most bioavailable product (Which may be critical if bioavallability 
is a problem). 
The desired release profile for the tablet should be known early in tablet 
development. Immediate, controlled, and combinations of immediate and 
controlled release profiles require totally different approaches to formulation 
development. Immediate release tablets usually require high levels of disintegrants 
or the use of superdisintegrants. Controlled release are usually 
formulations of polymers or wax matrices. 
In many instances, the rate-limiting factor to absorption of a drug is 
dissolution. It may be necessary for the formulator to select excipients 
which may increase drug dissolution and enhance absorption. Solvang 
and Finholt [7] studied the effect of binder and the particle size of the 
drug on the dissolution rate of several drugs in human gastric juice. Surface 
active agents such as sodium lauryl sulfate may be needed to promote 
wetting of the drug. Alternatively, the use of disintegrants or superdisintegrants 
may improve dissolution. Hydrophobic lubricants may be used only 
at low levels or not at all. 
The targeting of drug delivery to various sites in the gastrointestinal 
tract is sometimes required to maximize drug stability, safety, or efficacy. 
This subject is discussed in detail in Section IV.A.2. Drugs that are acidlabile 
or cause stomach irritation should not be released in the stomach. 
The use of enteric coatings on tablets is the most common method of targeting 
the release of a drug in the small intestine. Tablets which are to be 
coated should be formulated to withstand the rigors of a coating process 
and to be compatible with the coating material. The use of alkaline excipients 
in the tablet may prove to weaken the integrity of the enteric coated 
The appearance of a tablet dosage form is usually not thought to have a 
large impact on the commercial success of a particular product. However. 
all tablets must meet a minimal elegance criteria. The appearance of a 
tablet can be evaluated by its color, texture, shape, size, and coating 
(when present), and any embossing information.

Tablet Formulation and Design 81 
Tablet appearance can be affected by the color and texture each excipient 
brings to a tablet formulation. Lactose, starch, and microcrystalline 
cellulose appear white to off-white when compressed. The inorganic diluents 
such as calcium sulfate, calcium phosphate, and talc produce more of a gray 
color in the tablet. Drugs will impact on the overall color and appearance 
of the tablet. Drug-excipient interactions may change the appearance of 
the tablet with time. The use of dyes may be required to improve the appearance 
of certain tablets. Relatively large amounts of stearates and high 
molecular weight polyethylene glycols produce glossy tablets. 
The tableting properties (flow and compressibility) of tablet formulations 
containing a low percentage of active  100 mg) are primarily dictated by 
the tableting properties of the excipients in the formulation. The formulator 
will frequently have numerous excipients to choose from because the 
drug does not dominate the behavior of the formulation during processing. 
However, if the tablet formula contains a large percentage of active, the 
formulator may be somewhat restricted in the choice of excipients. In order 
to be easily swallowed and remain elegant, tablet size and weight is 
limited in these formulations. Tablet formulas with a higher percentage of 
active can contain only minimal quantities of excipients. These excipients 
must therefore perform their functions at relatively low levels. The use of 
a more effective binder such as microcrystalline cellulose may be required 
to produce these tablets. Tablets with a high percentage of actives frequently 
require granulation methods of manufacture simply because excipients 
will not perform their desired function at low levels in a direct-compression 
Marketing may request a coated tablet product. The quality of a coating 
on a tablet can be greatly affected by the tablet formulation onto which 
it is applied. Tablets with low resistance to abrasion (high friability) will 
result in coatings that appear rough and irregular. Coating adhesion can 
be greatly affected by the tablet excipients. Hydrophilic excipients can 
promote greater contact with the coating and result in superior adhesion. 
Hygroscopic excipients or drugs will cause swelling of a coated tablet and 
result in rupture of the film with time. 
Embossing of compressed tablets is becoming increasingly popular. Embossing 
permits the tablet to have identifying information without requiring 
coating and printing operations. Embossing does exacerbate any picking 
or sticking problems usually observed during compression. This may necessitate 
higher levels of lubricants and glidant to alleviate these problems. 
Extreme care should be taken in designing tooling for embossed and scored 
tablets. It may take several design attempts to select a tooling design that 
will consistently produce acceptable embossed or scored tablets. Embossed 
tablets that are to be film-coated present additional coating problems such 
as bridging of the coat across a depression in a tablet. 
One factor often overlooked in the development of a tablet formula is the 
cost of the raw materials and the process of manufacture. Direct compression 
is usually the most economical method of tablet production as previously 
discussed. In spite of the more expensive excipients used in direct compression. 
the cost (labor, energy, and time) of granulating is usually greater. 
Franz et al , [8] showed that a thorough analysis of cost versus time relationships 
can be performed using simulations before selecting a tableting 
process. Some companies have preferred manufacturing processes and raw

82 Peck. Baley. McCurdy. and Banker 
materials. These general manufacturing processes and materials are considered 
the first choice when developing a new product. If it is demonstrated 
that the preferred manufacturing process or materials are not suitable 
for a new product. then alternative processes or materials are used. 
The use of preferred processes and materials helps keep the types of equipment 
needed to manufacture and materials in inventory at a minimum. thus 
reducing capital expenditures and material costs. Preferred manufacturing 
process and materials also makes it easier to automate a production facility 
for multiproduct use. 
Process /Product Quality 
Excipients should be selected that will enable the production of a tablet 
that will meet or exceed standard in-house quality tablet specifications. 
A formulator should be involved in the establishment of tablet specifications 
and be able to provide sound rationale for the critical specifications. Typical 
tests performed on tablets are as follows: 
Weight variation 
Disintegration time 
Water content 
Content uniformity 
Product quality is most often addressed at the tablet development stage. 
However. it is also important to monitor the processing quality of a formulation 
during development. Two reasons for monitoring processing quality 
during development are (a) to optimize the process as well as the product. 
and (b) to establish in-process quality control tests for routine production. 
It is more difficult to quantify the processing quality of a formulatlon than 
it is to meausre the product quality. Some measurements that could be 
performed on the process include 
Ejection force 
Take-off force 
Flow of lubricated mixture 
Press speed (maximum) 
Frequency of weight control adjustments 
Sensitivity of formula to different presses 
Tooling wear 
Effect of consolidation load (batch size) 
Hopper angle for acceptable flow 
Hopper orifice diameter for acceptable flow 
Compressional forces 
Environmental conditions (temperature. humidity. and dust) 
Each of the above processing parameters can become a source of trouble 
in scale-up or routine production. By monitoring these parameters in

Tablet Formulatton and Design 83 
development, it may be possible to adjust the formula or process early 
enough to alleviate the source of trouble. 
The expected production output (numbers of tablets) per unit time will 
determine what speed tablet press will be required for a particular tablet 
product. If the anticipated unit output for a tablet product is expected 
to be large. a high-speed press will be required. Attempts should be made 
in formulation development to design a tablet formula that will perform well 
on a high-speed press. A formula to run on a high-speed press should 
have excellent flow to maintain uniform die fill during compressing. It 
should have good bonding characteristics so that it can compress with a 
minimal dwell time. 
B. Experimental Approach to Developing a Prototype 
Tablet Formula 
Atter conducting an excipient compatibility study, a formulator may still 
have a wide choice of excipients available to use in the final tablet formula. 
The formulator must select a few exclpients from a list of chemically compatible 
exclpients. The formulator may later eliminate many drug-compatible 
exclpients by selecting only those excipients known to provide a much needed 
function in the tablet formula as dictated by medical, marketing, economic, 
or process/product qualtty concerns. The objective in screening excipients 
for a prototype tablet formula is to choose a combination of excipients that 
most completely achieves desirable tableting characteristics. Tablets made 
at this stage of experimentation can be made on a Carver, single-punch, 
or rotary press depending on the amount of drug available. Obviously, 
no evaluation of the flow properties of a mixture can be made on a Carver 
or single-punch press. The following Is a list of several experimental techniques 
that may be used to essist the formulator to develop a prototype 
Analysis of variance (ANOVA) 
Statistical screening designs (first-order designs) 
Plackett Burman 
Extreme vertices 
Analysis of Variance (ANOVA) 
The ANOVA approach involves making statistical comparisons of different 
tablet formulas. Each formula represents a different combination of excipients. 
The selection of a prototype formula is done by running an ANOVA 
on the results of all the tests performed.' The formula that is significantly 
better than the others tested becomes the prototype formula. 
Statistical Screening Designs (First-Order Designs) 
Plackett Burman Designs 
A statistical screening involves setting lower and upper limits on the 
levels of each excipient considered for use in a tablet formula. Usually 
no more than 10 excipients are being considered for use in the tablet at 
this point. An experimental design is ohosen that will enable a statistical 
test for the effect of each excipient on each process/product quality

84 Peck, Baley, McCurdy. and Banker 
Table 1 Twelve-Run Plackett-Burman Design 
x x x x x x x x x x x 
Trial 1 2 3 4 5 6 7 8 9 10 11 
1 + + + + + + 
2 + + + + + + 
3 + + + + + + 
4 + + + + + + 
5 + + + + + + 
6 + + + + + + 
7 + + + + + + 
8 + + + + + + 
9 + + + + + + 
10 + + + + + + 
11 + + + + + + 
characteristic (weight variation, hardness, friability, disintegration, dissolution, 
etcv) , 
This type of study requires at least n + 1 (n =number of excipients) 
trials to enable a statistical test. The type of statistical design employed 
in a screening study is referred to as a Plackett-Burman [9J design. Table 
1 shows a 12-run Plackett-Burman design. Using this design, as many as 
11 excipients could be screened for use in a tablet formula. Each column 
represents a different excipient. If seven excipients were to be screened, 
columns X1 to X7 would define the experimental design. Each row (tablet 
formula) in Table 1 represents combinations of high (+1) and low (-1) 
levels of each excipient (Xi) to be screened. The levels specified in each 
row are used to produce a tablet formula containing a fixed quantity of 
drug. Tablet formulas containing these mixtures of exctpients are compressed 
and evaluated. A first-order regression model is fit to data collected 
during the tablet evaluation. Statistical tests can be performed to 
determine whether each excipient affected the tablet quality in a significantly 
positive or negative manner. Excipients that did not provide a significant 
"positive" effect on tablet quality may be either retested in a second 
screening study at different levels of eliminated from durtehr consideration 
for inclusion into the tablet formula. A second statistical screen may 
be performed on excipients to refine excipient ranges to more appropriate 
levels. Once acceptable excipients and excipient ranges have been established, 
formulation optimization can proceed. 
Extreme Vertices 
The extreme vertices design [10J is usually used as an optimization 
technique. However, it can be used as a screening study if at least n + 1 
trials are run. The extreme vertices design is recommended when the

Tablet Formulation and Design 85 
number of components (excipients) is six or mor-e, A first-order model is 
fit to the data to test for significant eXcipients as was done in the PlackettBurman 
design. The disadvantage of using the extreme vertices design in 
tablet development is that tablet weight must be kept constant throughout 
the screening process. 
c. Experimental Approach to Optimizing a Prototype 
Tablet Formula 
A tablet formulation optimization study should be performed using an appropriately 
statistically designed experiment. Numerous experimental design 
texts [11,12] are available that can assist a formulator in selecting the appropriate 
experimental design. The extreme vertices design is not recommended 
in most tablet optimization studies unless tablet weight is to be 
held constant. It can be beneficial to have a statistician experienced in 
experimental designs select an appropriate design based on the established 
excipients and excipient ranges. All excipients should be varied in the 
optimization study to truly optimize the formulation. Excipients levels are 
USUally the only factors or variables in a formulation optimization study. 
To reduce the number of factors in a study, a ratio of two excipients can 
be used. However, the total quantity of those two excipients must be fixed 
in the formula. When using excipient ratios as factors, include a factor for 
tablet weight. Tablet Weight can then be varied as a factor. If a formulator 
suspects an interaction between an excipient and a particular process 
variable. the process variable should be considered for inclusion in the 
formulation optimization study. For example. in a sustained release directcompression 
tablet, compression force may impact on the release rate of the 
drug from the tablet. In this example. compression force should be included 
as a factor in a formulation optimization study. Usually, all other 
process variables are maintained constant. Process variables that cannot 
be held constant but are not expected to impact on the tablet characteristics 
should be "blocked" appropriately in the design. For instance, different 
lots of raw materials or bulk drug may be used in an optimization 
study. The different lots should be treated as blocks in the experimental 
design. This will allow for a statistical test for block (lot) effect at the 
data analysis stage of the experiment. In this example, blocks serve as 
a flag to signal the formulator that the quality of the raw materials is not 
well controlled. Since the use of blocks do not "cost" the formulator any 
additional trials. blocks should be used wherever possible. Statisticians 
experienced in experimental design frequently state that you cannot lose 
by blocking! 
It is important that all trails are performed in a randomized manner. 
After all the tablets have been manufactured, data analysis begins. A 
standard quadratic model is most often used to fit second-order experimental 
design data. Commercially available software (XSTAT. STATGRAPHICS, 
PCSAS. ECHIP) may be used to generate the coefficients and statistical 
tests on the raw data collected. This software will also provide a statistical 
analysis of the regression models produced. The analysis of the regression 
model provides the scientist information on how well the model explainer 
the data variation. If a particular regression model does not satisfactorily 
explain the data variation, transformations of the raw data can be tried to 
improve the fit of the model. For a regression model to be acceptable. the 
R 2 > O.75 the lack of fit should not be significant. and the residuals should

86 Peck, Baley, McCurdy, and Banker 
have no more than a few outliers. Once acceptable models have been established 
for each tablet characteristic, the scientist should examine the models 
to determine which main effects, interactions, or quadratic terms are significant. 
The formulator should then generate response surface plots of 
significant interactions as a function of the tablet characteristic. Response 
surface plots of significant main effects and quadratic terms will also help 
the formulator to understand the critical relationships between tablet characteristics 
and the formulation factors. 
Optimization of the final formulation can be performed using commercially 
available software (XSTAT, ECHIP, and PCSAS). Optimization invariably 
requires that constraints be placed on some or all of the critical response 
parameters. Constraints may also be placed on some or all of the 
factors as well. One critical tablet characteristic must be selected to optimize 
(minimize or maximize) while the other tablet characteristics and 
formulation factors are left constrained or unconstrained. For example. 
tablet friability could be constrained below 0.3% while dissolution rate is 
maximized. The mathematical algorithm used in specific optimization routines 
(software) varies. Optimization algorithms used in software routines 
are usually based on a simple method or a grid search method. The final 
formula determined to be optimal should be experimentally verified by manufacture 
and testing. Model predicted values for tablet characteristics 
should "agree" with actual experimental data collected on the optimal formula. 
D. Establishment of Excipient and Preliminary Process Ranges 
In light of the present interest in validating the product as well as the 
process of manufacture, it is to the formulator's advantage to establish 
excipient and process variable ranges. Having excipient and process ranges 
also allows production to make appropriate excipient or process changes without 
prior notification of the regulatory agencies. 
If it can be demonstrated that the excipient ranges used for conducting 
the optimization study produced acceptable tablets (Le , , all tablets produced 
were acceptable, then the excipient ranges used in the study should be 
used as final product ranges. However, if the excipient ranges used in 
the optimization study were not always acceptable, the ranges should be 
narrowed to acceptable limits. This can be done by performing constrained 
optimization of the critical response variables using registration specifications 
on the response variables as the constraining limits. 
E. Bioavailability Studies 
In vivo test procedures appropriate for tablets and other solid dosage 
forms are also the SUbject of Chapter 6 in Volume II of this series. In 
some cases in vivo testing of tablet formulations involves studies in animals 
prior to studies in humans: in other cases the tablet formulations are 
studied directly in humans. 
When in vivo studies in humans are undertaken, it may be desirable 
or even essential to conduct such studies with more than one formulation. 
This is particularly true if a goal of product design is product optimization. 
and a primary objective is to maximize bioavailability or response versus 
time profile. A bloavailability study should eventually be run comparing

Tablet Formulation and Design 87 
the optimized formulation, a formulation (within the excipient ranges) predicted 
to have the slowest dissolution, and a formulation (within the excipient 
ranges) predicted to have the fastest dissolution. The bioavailablhtv 
study results can be used to establish a correlation between the in vitro 
dissolution test and the in vivo bioavailability parameters. If the three 
formulations (optimal. slow, and fast-dissolving) turn out to be bioequivalent, 
the excipient ranges are valid from the in vivo performance viewpoint. 
If the three formulations are not bioequivalent, then the excipient ranges 
should be tightened using the in vitro/in vivo correlation. The specifications 
for the dissolution of the tablet should be set based on this correlation. 
F. Development of Stability Data for Tablet Formulations 
Stability data should be collected on the bulk drug as well as the final 
product stability. Stability on the bulk drug should be available in the 
preformulation data. Based on the results of the bulk drug stability testing, 
recommendations should be made about the storage conditions and the 
shelf life of the bulk drug. 
The final tablet formulation should be placed up on stability as soon 
as possible after its invention. Also. formulas that "cover" the proposed 
excipient ranges may also be placed up on stability. Stability data should 
be generated with the tablet in all the expected packaging configurations 
(i. e., blisters, plastic and glass bottles, etcv) , Ideally several lots of tablets 
should be put on stability using different lots of bulk drug. Having different 
lots of tablets containing different lots of bulk drug will give an indication 
of the lot-to-lot variability in product stability. Accelerated stability 
testing (high temperature. humidity, or intense lighting) can be helpful in 
[udging the long-term stability of a tablet package system. 
In addition to the stability data generated on the final formula, stability 
data generated on similar formulations can sometimes be used as supportive 
stability data. Usually there will be more supportive stability data available 
because the similar formulations were developed prior to the optimal 
Based on the product stability data, a formulator must recommend 
proper storage conditions, special labeling regarding storage, and an expiration 
date for each tablet package system. 
G. Development of Validation Data for Tablet Formulations 
As required under an NDA, process validation is the final step undertaken 
after the process has been scaled up to full production batch size. Under 
the concept of validation, an immense work load is placed on the pharmacy 
development group, the pilot plant group, and possibly the production department 
to achieve the goals of validation as previously defined. Because 
of the immense work load, some companies have created a group dedicated 
to assist formulators in Validating their manufactur-ing processes. As a 
rule of thumb, the less complex the manufacturing process, the better defined 
the drug and excipient specifications, the easier the validation process. 
The need to validate tablet products provides a great impetus to the use of 
optimization techniques in tablet product design. The data base required 
for product validation will often be adequate when development has

88 Peck, Baley, McCurdy, and Banker 
proceeded using optimization techniques. The validation of a new or reformulated 
tablet product requires two phases. In phase I the development 
team formulates the product and general process of manufacture. In phase 
II, emphasis is placed on the process validation of production scale batches. 
Phase II is usually accomplished at the production startup of a new or reformulated 
The objectives in phase I include: 
1. Producing an optimal formula and process. 
2. Identifying the most critical tablet characteristics and establishing 
specifications for the tablet. 
3. Quantifying relationships between the critical tablet characteristics 
and process I formulation variables. 
4. Establishing specifications for process/formulation variables to ensure 
that tablet specifications will be met. 
5. Proposing in-process tests for critical process variables and raw 
materials specifications for critical formulation variables when appropriate. 
6. Documenting above information. 
The objectives in phase II include 
1. Demonstrating that all manufacturing equipment and related systems 
(SOPs, equipment calibration, cleaning procedures, assays, 
packaging, and personnel training) have been qualified for use in 
the manufacture and testing of this product. 
2. Drafting a process validation protocol before manufacture of first 
production lots that specifies the procedures to be validated. This 
protocol should be written to challenge the proposed limits on the 
critical process/formulation variables. 
3. Running production/validation lots; collecting and analyzing data. 
4. Demonstrating that all product specifications have been met in spite 
of the challenges presented to the process. 
5. Documenting above information. 
Usually several production lots are required to complete phase II validation. 
The more process/formulation variables, the more production lots 
will have to used. If the production scale validation lots pass all the required 
specifications for that tablet product, the lots may be used for commercial 
A. Active Ingredients 
General Considerations 
Broadly speaking, two classes of drugs are administered orally in tablet 
dosage form. These are (1) insoluble drugs intended to exert a local effect 
in the gastrointestinal tract (such as antacids and absorbents) and (2) 
soluble drugs intended to exert a systemic drug effect following their dissolution 
in the gut and subsequent absorption. With each class of drugs 
very careful attention must be given to product formulation and design as 
well as to manufacturing methods in order to produce an efficacious and

Tablet Formulation and Design 89 
reliable product. The goal in designing tablet dosage forms for these two 
classes of drugs is different. When working with insoluble drugs whose 
action is usually strongly affected by surface phenomena (such as antacids 
and absorbents) it is critical that a product be designed that will readily 
redisperse the produce a fine particle size and large surface area. Accordingly 
the effect of formulation. granulation, and tableting on the surface 
properties of the material and the ability to regenerate a material in the 
gut with optimum surface properties are critical. 
In the case of drug products intended to exert a systemic effect, the 
design of a dosage form which rapidly disintegrates and dissolves mayor 
may not be critical, depending on whether the drug is absorbed in the upper 
gastrointestinal tract or more generally throughout the intestinal tract. 
and also based on the solUbility properties of the drug at or above its absorption 
site. Dosage forms must. however, be designed which do disintegrate 
or dissolve to release the drug in an available form at or above the 
region of absorption in the gut. 
The developmental pharmacist usually does not have a great deal of input 
into selecting the chemical form of an active ingredient. Drug-screening 
programs may not offer several salt or ester forms of the drug as candidates 
for a particular therapeutic claim. Instead the formulator. provided with 
small quantities of an active ingredient in a particular form to evaluate in 
the preformulation studies, is faced with the task of developing a tabletwhich 
may be capable of handling only drugs of the same physical and 
chemical properties as the small sample. When large batches become available, 
often months later, they frequently differ in physical properties, 
making formulation and processing modification necessary. Given the opportunity, 
the preformulation scientist may suggest a particular salt or 
crystal form of the drug that is more stable, more suitable for tableting, 
or more bioavailable. As an example, ethanol-recrystallized (ethyl) ibuprofen 
is the form of drug initially developed to produce ibuprofen tablets. 
The ethyl drug is poorly compressible and USUally must be tableted using 
wet g ranulution processing. Tablets made with the ethyl drug have a tendency 
to pick, stick. and laminate during compression. Methanol-recrystallized 
"methyl' ibuprofen [10] was subsequently developed. Methyl drug 
was capable of being tableted in a direct-compression formulation with no 
picking, sticking. or lamination problems. The difference between the 
crystal habits of the two drugs resulted in dramatically different tableting 
It is imperative that the physical properties of the active ingredient be 
thoroughly understood prior to the time of finalizing the formula. Indeed, 
these properties may provide a rational basis for a particular tablet design, 
such as rapid dissolution for a drug likely to be absorbed high in the upper 
gut, or the need for enteric or other forms of gastric protection for an 
acid-labile drug. 
Although almost all tablets will require the addition of nonactive components 
or excipients-to produce satisfactory drug release, to achieve 
acceptable physical and mechanical properties. and to facilitate their manufacture-
the formulator should not be anxious to begin adding excipients 
until the properties of the drug are thoroughly understood. If a substance 
possesses the proper crystalline structure, it can be compressed directly 
into a tablet without further treatment. Relatively few such materials 
(active or excipient) exist. and their number diminishes further if one 
considers only materials with therapeutic activity. Jaffe and Foss [13]

90 Peck, Baley, McCurdy, and Banker 
confirmed that generally drugs of cubic crystalline structure are compressible 
directly, since upon compression the crystals are fractured, and the 
fragments form a close-packed arrangement which readily consolidates on 
compression. In a CUbe, the structure is the same along each axis; thus, 
no alignment is necessary in order for ionic or van der Waals bonding to 
occur between the individual particles. Sodium choride has a cubic structure 
and is an example of a directly compressible material. 
In crystals which are not cubic, some realignment is necessary, which 
results in a reduced probability of bonding. Employing potassium chloride 
as a model, Lazarus and Lachman [14] found that the compaction of these 
crystals depended on many factors, such as particle size distribution, crystal 
shape, bulk density, and moisture content. If the drug to be formulated 
happens to possess a crystalline structure allowing for direct compaction, 
the formulator's task will be lessened. Rankell and Higuchi [15] have presented 
a theoretical discussion on the physical process which may be responsible 
for interparticulate bonding during compression. While the 
tableting aspects will be straightforward, the other requirements, such 
as acceptable friability, hardness, appearance, disintegration, and dissolution, 
must be met. 
It is extremely rare to find a drug system which does not involve the 
use of exeipients , The contribution of excipients will be discussed in 
Section IV.B. The treatment of processing which the active ingredient receives 
(alone or in combination with the excipients) will depend upon the 
dosage level, the physical and chemical properties of the active drug substance 
and the excipients used, the nature of the drug. its use, any absorption 
or bioavailability problems. and the granulation and tableting method 
employed. When potent drugs of limited solubility are involved. their particle 
size and uniform distribution throughout the tablet can dramatically affect 
the rapidity of their dissolution and absorption as well as content uniformity. 
However, if large dosage regimens of a soluble drug are considered, 
the effect of particle size is important-more from a processing standpoint 
than because of dissolution or absorption considerations. The relationship 
of various particle size factors to therapeutic effectiveness of drugs 
was discussed by Rieckmann [16]. He pointed out that one must be cautious 
in equating micronization, dissolution, and adsorption. especially with drugs 
such as nitrofurantoin, chloramphenicol. and spironolactone. 
The role of the active ingredient can then be considered in two broad 
systems: first, when the drug-excipient interactions are considered primarily 
from a pharmacological (dissolution and absorption) viewpoint; and 
second, where in addition to the concerns in the first area, significant 
processing questions must be answered. 
Bioavailability Considerations 
Before drugs can effectively pass through the gastrointestinal wall they 
must be in solution. Drugs which are only sparingly soluble in the gastrointestinal 
contents at or above the absorption site can have, as the controlling 
process affecting their absorption, the rate of drug solution in 
these fluids. In this type of system, the drug goes into solution at a slow 
rate ; absorption occurs almost immediately and is not, therefore, the ratelimiting 
step. In one study, Nelson [17] correlated the blood level coneentration 
of various theophylline salts with their dissolution rates. 
As noted earlier in this chapter and throughout this volume, drugs 
which exert a systemic effect must dissolve as a prerequisite to effective

Tablet Formulation and Design 91 
drug absorption. The various processes of tablet making, including the 
aggregation of drug into granular particles, the use of binders. and the 
compaction of the system into a dense compact, are all factors which mitigate 
against a rapid drug dissolution and absorption in the gastrointestinal 
tract. In considering in a general manner the availability of drugs from 
various classes of dosage forms, drugs administered in solution will usually 
produce the most available drug product-assuming the drug does not precipitate 
in the stomach or is not deactivated there. 
The second most available form of a therapeutic agent would be drug 
dispersed in a fine suspension, followed by micronized drug in capsule 
form, followed by uncoated tablets, with coated tablets being the least bioavailable 
drug product in general. In formulating and designing drug 
products as well as in considering methods of manufacture, the fact that 
the tablet dosage form is one of the least bioavailable forms (all other factors 
being equal) should be kept in mind. 
Many factors can affect drug dissolution rates from tablets. hence 
possibly drug bioavailability-incIuding the crystal size of the drug; tablet 
disintegration mechanisms and rates; the method of granulation; type and 
amount of granUlating agent employed; type, amount, and method of incorporation 
of disintegrants and lubr-icanta ; and other formulation and processing 
Levy et al . [18] showed the effects of granule size on the dissolution 
rate of salicylic acid. Salicylic acid of two mesh ranges, containing 300 
mg of aspirin and 60 mg of starch, were compressed at 715 kg cm- 2. The 
data are shown in Figure 2. 
Lachman et al . [19] studied the effect of crystal size and granule size 
on a delayed action matrix using tripelennamine hydrochloride. He noted 
that while granule and crystal size both affected release rate, in this instance 
the crystal size played a greater role than gr-anule size in dissolution 
Paul et al , [20] showed that with nitrofurantoin there was an optimal 
average crystal size of about 150 mesh, which resulted in adequate drug 
excretion (hence absorption and efficacy) but minimized emesis. This exemplifies 
a situation in which too rapid drug dissolution in the stomach 
may produce nausea and emesis; an intermediate release rate reduces this 
effect while achieving adequate bioavailability. 
Numerous accounts of the effect of particle size on the dissolution rate 
of steroids have been reported. In one study Campagna et al . [21] showed 
that, in spite of good disintegration, therapeutic inefficacy of prednisone 
tablets could occur. 
B. Nonactive I ngl"edients 
The selection and testing of nonactive ingredients or excipients in tablet 
formulas present to the formulator the challenge of predictive foresight. 
While the ability to solve problems when they occur is a valuable attribute, 
the ability to prevent the problem through adequate experimental design 
is a virtue. leads to more reliable and expeditious product development. 
and, when coupled with optimization methods. enables the formulator to 
tell how close a particular formula is to optimum conditions. 
It will become obvious to the forrnulator-, on reviewing the literature, 
that the total number of significant excipients currently in use is probably 
less than 25. These 25 materials fulfill the needs of the six major excipient

92 Peck. Baley, McCurdy. and Banker 
Ii g
0 120 w
100 til 
TIME (min) 
30 40 
Figure 2 Effect of granule size on the dissolution rate of salicylic acid 
contained in compressed tablets. Key: - 40- to 60-mesh granules;  60to 
SO-mesh granules. 
categories: diluents , binders. lubricants. disintegrants , colors, and 
sweeteners (flavors excluded). The United States Pharmacopeia (USP XIX) 
recognized the important role exeipients play in dosage form design by initiating 
a new section entitled "Pharmaceutic Ingredients. fI In time. official 
monographs may be developed for all the major or commonly used excipients. 
In 1974 the Swiss pharmaceutical companies, Ciba-Geigy. Hoffman-LaRoche. 
and Sandoz, joined together to publish in the German language an excipient 
catalog (Katalog Ptiarmazeutiscber Hilfsstoffe), covering almost 100 official 
and nonofficial excipients. The book contains general information, suppliers, 
tests. and specifications obtained from the literature or measured in the 
laboratories of the above companies. The development of an excipient codex 
was a major project of the Academy of Pharmaceutical Sciences of the American 
Pharmaceutical Association [22]. 
It will become apparent later in this section that many times the 25 or 
so excipients have been repeatedly evaluated over the past 50 years, and 
yet these same materials continue to stand the test of time. Rather than 
belabor the point we must simply be reminded that the tried and tested 
materials, by their longevity, deserve careful consideration. The formulator 
should not, however, be fearful of change or of evaluating new

Tablet Formulation and Design 93 
ingredients. Some formulators tend to "lock -inII on particular formulation 
types of approaches which have been successful in the past; the danger 
here is that one becomes dated. At the other extreme is the formulator 
who takes a quick look at a new disintegrant or binder, which then ends 
up in the formula months before sufficient data are available to make possible 
a sound judgment of total acceptability. Thus the best formulator 
is an individual who is constantly searching for new and better methods 
and systems, who avoids becoming sterotyped, and who is cautious and 
thoroughly analyzes new approaches without developing an undue proprietary 
or vested interest in them. 
Additives are usually classified according to Some primary function 
they perform in the tablet. Many additives will also often have secondary 
functions, which mayor may not be of a beneficial nature in good, solid 
design of oral dosage forms. Some fillers or diluents may facilitate tablet 
dissolution, which is beneficial, while others may impair dissolution. The 
most effective lubricants are water repellent by their nature, which may 
retard both disintegration and dissolution. 
Bavitz and Schwartz 123] concluded, in a paper evaluating common tablet 
diluents, that their proper choice becomes more critical when formulating 
water-insoluble drugs as opposed to water-soluble drugs. They showed 
that "inert ingredients" can profoundly affect the properties of the final 
dosage form. A knowledge of the properties of additives and how they 
affect the properties of the total formulation is necessary to provide guidelines 
in their selection. This is particularly true when the drug concentration 
is small. The drug plays a more significant role in determining 
the physical characteristics of the tablet as the drug concentration increases. 
Two major classifications of additives by function include those which 
affect the compressional characteristics of the tablet: 
Binders and adhesives 
Lubricants, anttadherents , and glidants 
and those which affect the biopharmaceutics, chemical and physical stability, 
and marketing considerations of the tablet: 
Flavors and sweeteners 
Miscellaneous components (e. g., buffers and adsorbents) 
Although diluents are normally thought of as inert ingredients, they can 
significantly affect the biopharmaceutic, chemical, and physical properties 
of the final tablet. The classic example of calcium salts interfering with 
the absorption of tetracycline from the gastrointestinal tract was presented 
by Bolger and Gavin [24]. The interaction of amine bases or salts with 
lactose in the presence of alkaline lubricants, and subsequent discoloration 
(as discussed by Costello and Mattocks [25] and Duvall et al , [26]), emphasized 
that excipient "inertness" may often not exist in the design of drug 
dosage form. 
Keller [27] reviewed the properties of various excipients while Kornblum 
[28,29] proposed preformulation methods of screening materials for use as

94 Peck, Baley, McCurdy, and Banker 
diluents. Simon [6] described rapid thermal analytical methods of screening 
for possible drug-excipient interactions. In another study Ehrhardt and 
Sucker {30] discussed rapid methods to identify a number of excipients 
used in tablet formulations. 
Usually tablets are designed so that the smallest tablet size which can 
be conveniently compressed is formed. Thus, where small dosage level 
drugs are involved I a high level of diluent or filler is necessary. If, however, 
the dosage level is large, little or no diluent will be required, and 
the addition of other excipients may need to be kept to a minimum to avoid 
producing a tablet that is larger than is acceptable. In such large drug 
dosage situations, nevertheless, excipient materials must often be added 
to produce a granulation or direct-compression mixture which may be compressed 
into acceptable tablets. 
Where moisture is a problem affecting- drug stability I the initial moisture 
level, as well as the tendency of the material to retain or pick up moisture, 
must be considered. The hygroscopic nature of excipients I as described 
by Daoust and Lynch [31], is an important consideration in formulation 
studies for the following reasons; 
1. Water sorption or desorption by drugs and excipients is not always 
reversible. Absorbed moisture may not be easily removed during 
2. Moisture can affect the way in which a system accepts aqueous 
granulating solutions. 
3. The moisture content and rate of moisture uptake are functions of 
temperature and humidity and should be considered. 
4. Moisture content in a granulation affects the tableting characteristics 
of the granulation. 
5. Hygroscopicity data can aid in the design of tablet-manufacturing 
6. Moisture-sensitive drugs should not be combined with hygroscopic 
7, Packaging materials should be chosen to suit the product. 
Sangekar et al , [32] reported on the percent moisture uptake of tablets 
prepared from various direct compression excipients. Figure 3 indicates 
that a range of 1. 7 to 5.6% uptake is possible, depending on the excipient 
used. Dicalcium phosphate, lactose anhydrous DTG, and lactose beadlets 
absorbed the minimum amount of moisture, while sorbitol and sucrose absorbed 
the maximum. Mannitol, dextrose, and monocalcium phosphate were 
shown to be intermediate. 
In selecting diluents, the materials will be found to contain two types 
of moisture, bound and unbound. The manner in which a diluent holds its 
moisture may be more important than the affinity of the material for moisture 
or the amount of moisture present. Calcium sulfate dihydrate, for 
example, contains 12% moisture on a mole-for-mole basis. The water is 
present, however, as bound moisture (as water of crystallization). Furthermore 
the tightly bound water is not liberated until a temperature of about 
SooC is reached (well above normal product exposure temperatures). Since 
calcium sulfate dihydrate is thermodynamically satisfied as to water content 
and moisture demand, it is not hygroscopic and absorbs little moisture. 
Since the bound water is generally unavailable for chemical reaction, 
CaS04' 2H20 has been widely used in vitamin tablets and other systems

Tablet Formulation and Design 95 
56 E6 
E8 w
48 IQ.. 
::> E2 Ii; 
a E5 
~ E7 w
C-' I- z
E3 a:: 
Q.. E, 
Z E4 e::t: 
4 8 12 16 20 24 28 32 36 40 44 48 
Figure 3 Direct-compression tablets with different exciptents (EI to ES)' 
common binder (microcrystalline cellulose). and common disintegrant 
(alginic acid). Mean percent moisture uptake across humidity levels of 
43, 65. 75, and 100% relative humidity at 2SoC. Key: EI dibasic calcium 
phosphate dihydrate (unmilled); E2, monobasic calcium phosphate monohydrate; 
E3' lactose anhydrous DTGj E4' lactose hydrous beadlets; ES' 
mannitol granular j E6, sorbitol crystalline, tablet type j E7' dextrose; E8' 

96 Peek, Baley. MeCurdy , and Banker 
which are moisture-sensitive. Such a system, containing tightly bound 
water but with a low remaining moisture demand, may be vastly superior to 
an anhydrous diluent (or other excipient) which has a high moisture demand. 
When using a hydrate or excipient containing water of crystallization or 
other bound water, careful attention must be paid to the conditions under 
which this water is released. 
The degree of cohesiveness which a diluent imparts to various drug 
substances when compacted into tablets becomes increasingly important when 
tablet size is a factor. Where size is not a factor, the ratio of the cohesiveness 
imparted by a diluent to its cost per kilogram should be considered. 
For example, if size is not a factor, a diluent that costs $3.08 per kilogram 
and is effective at a 10% concentration might be replaced by a diluent that 
must be present at a 25% concentration, that costs $0.66 per kilogram. 
Kanig [33] reviewed the ideal properties of a direct compaction diluent 
material, many of which hold true of any diluent. 
In special tablets, such as chewable tablets, taste and mouth-feel become 
paramount in diluent selection. In these specialized tablets a consideration 
of unique aging effects, such as increased hardness and reduced 
"chewability," must be carefully examined. 
The sensitivity of diluents to physicochemical changes caused by processing 
or manufacturing, both of wrdeh influence final tablet quality, should 
be considered in diluent selection. This is illustrated in Figures 4 and 5 
[34] by a comparative evaluation of excipients for direct-compression formulas. 
These figures indicate the variety of disintegration and hardness 
Lactose EFK 
Lac lose Anhydrous 
500 1000 1500 2000 2500 3000 
Figure 4 Disintegration time versus applied force for compacts of various 

Tablet Formulation and Design 
Avicel PH 102 
Lactose anhydrous
Lactose UK 
500 1000 1500 2000 
Figure 5 Crushing strength versus applied force for compacts of various 
profiles possible. Combinations of two or more exclpients generally provide 
a final disintegration-hardness spectrum which lies between the values for 
each material when used separately. The figures also show the sensitivity 
of various agents to alterations in properties (disintegration time and crushing 
strength) with changes in compressive load. Emcompress, for example, 
was very sensitive to a change in compressive load in this study whereas 
Celutab and dextrose monohydrate are almost totally insensitive. Ideally. 
the diluent selected will not be sensitive to processing variables. such that 
the quality of the final tablet features can degrade appreciably under the 
processing variables encountered in production. This is an important consideration 
in the validation of a product and its method of manufacturing: 
identifying the range of product quality features produced by the expected 
limits of the processing variables encountered in production, and designing 
product formulation and processes so as to minimize such variability. 
Lactose USP is the most widely used diluent in tablet formulation. It 
displays good stability in combination with most drugs whether used in the 
hydrous or the anhydrous form. Hydrous lactose contains approximately 
5% water of crystallization. The hydrous form is commonly used in systems 
that are granUlated and dried. Several suppliers offer various grades of 
hydrous and anhydrous lactose. The various grades have been produced 
by different crystallization and drying processes. It is most important not 
to assume that one form of lactose will perform in a similar manner as 
another form. Lactose is available in a wide range of particle size distributions. 
Nyqvist and Nicklasson [35] studied the flow properties of directly 
compressible lactose in the presence of drugs. While lactose is freely

98 Peck, Baley. McCurdy, and Banker 
(but slowly) soluble in water. the particle size of the lactose employed can 
affect the release rate of the medicinal. Recent studies indicate the T 50% 
(time required for 50% of the drug to dissolve) was decreased by a factor 
of 8 when micronized lactose (2 to 5 mg2 g-l) was used rather than unmicronized 
lactose (0.5 m2 g-l surface area). 
Lactose formulations usually show good drug release rates, are easy to 
dry (both in thrays and fluidized bed dryers). and are not sensitive to 
moderate variation in tablet hardness upon compression. They find exceptional 
application in tablets employing small levels of active ingredients 
(e. g.. steroids). The cost of lactose is low relative to many other diluents. 
As noted previously, lactose may discolor in the presence of amine drug 
bases or salts and alkaline lubricants. 
Lactose USP, anhydrous offers most of the advantages of lactose USP, 
hydrous. without the reactivity of the Maillard reaction. which leads to 
browning. Tablets generally show fast disintegration. good friability. and 
low weight variation. with an absence of sticking, binding. and capping. 
The applications of the anhydrous form have recently been evaluated by a 
number of investigators (36- 39]. Mendell [40] has reported on the relative 
sensitivity of lactose to moisture pickup at elevated humidities. Blister 
packages should be tested at elevated temperatures and humidity to establish 
their accept ability with lactose- based formulas. 
Lactose USP. spray-dried has improved flow and bond properties over 
the regular lactose due to the general spherical form of the aggiomerates , 
This shape can be affected by high - shear milling. The effect of particle 
diameter on particle and powder density, and angle of friction and repose I 
and the effect of orifice diameter on the flow rate have been studied by 
Alpar et al , [41] and Mendell [40]. Even when granulated. spray-dried 
lactose displays its flow and bond properties. It is commonly combined with 
microcrystalline cellulose and used as a direct -cornpaction vehicle. Alone. 
it usually must be used at a minimum concentration of 40 to 50% of the tablet 
weight for its direct-compaction properties to be of value. It has the 
capacity of holding 20 to 25% of active ingredients. Care must be exercised 
upon storage since loss of the usual 3% moisture content can adversely affect 
compressional properties. 
Brownley and Lachman [42] reported that. as with lactose USP, care 
must be taken in using spray-dried lactose since it tends to become brown 
due to the presence of 5-(hydroxymethyl)-2-furaldehyde, when combined 
with moisture. amines, phosphates, lactates. and acetates. Similar findings 
were reported by Duvall et al , [26J even in systems not containing amines. 
The employment of neutral or acid lubricants such as stearic acid appears 
to retard the discoloration. while alkaline lubricants (e. g., magnesium 
stearate) accelerate the darkening. Bases as well as drugs which release 
radicals (e. g . , amino salts) can bring about this browning, known as the 
Maillard reaction. Richman [43] reviewed the lubrication of spray-dried 
lactose in direct-compaction formulas and reported that this lactose form 
may affect the mechanism of action of lubricants. 
The cost of spray-dried lactose is moderate; however, the fact that it 
is not available from a large number of suppliers could limit its widespread 
USe. Tablets made with spray-dried lactose generally show better physical 
stability (hardness and friability) than regular lactose. but tend to darken 
more rapidly. 
It was reported by Henderson and Bruno [39J that the tableting characteristics 
of spray-dried lactose were inferior to those of lactose beadlets. 
However, the physical stability of the resulting products was similar.

Tablet Formulation and Design 99 
Starch USP may come from corn, wheat, or potatoes and finds application 
as a diluent. binder, and disintegrant. Tablets containing high concentrations 
of starch are often soft and may be difficult to dry, especially 
when a fluidized bed dryer is used. Commercially available starch USP 
may vary in moisture content between 11 and 14%. Certain specially dried 
types of starch are available at moisture levels of 2 to 4% at a premium 
price. Where the starch is used in a wet (aqueous) granulated system, the 
USe of specially dried starch is wasteful since normal drying techniques will 
result in a moisture level of 6 to 8%. Recent studies indicate that, in some 
drug systems, starch-initially at a moisture level of 10% or greater-may 
perform differently with respect to dissolution than starch at a 5 to 7% 
level, even though the final equilibrium moisture levels of the tablet are 
the same. 
There are also indications that I although starch reaches a moisture 
plateau of 11 to 14%. it often serves as a local desiccant to help stabilize 
moisture-sensitive drugs. This attribute can act in a negative fashion, 
however, as in steroid tablets. where the localization of moisture may result 
in reduced dissolution rates. 
In a study on the effect of granule size, compression force, and starch 
concentration on the dissolution rate of salicylic acid, Levy [44] showed an 
increase in dissolution rate with decreasing granule size, increasing precompression 
force, and increasing starch content. 
The effect of starch on the disintegration time of tolbutamide tablets 
was studied by Commons et al , [45]. They showed a critical starch concentration 
for different granule sizes of tolbutamide; however, disintegration 
times did not decrease with increasing starch levels. 
Schwartz et a1. [46] evaluated the incorporation of starch USP versus 
a modified cornstarch in various formulations. The modified starch generally 
exhibited improved processing characteristics and improved tablet properties. 
compared to starch USP. 
Directly compressible starch, marketed commercially as Starch 1500, is 
physically cornstarch. Chemically, compressible starch does not differ 
from starch USP. It is a free-flowing, directly compressible excipient, 
which may be used as a diluent, binder, and disintegrating agent. When 
compressed alone. it is self-lubricating and self-disintegrating, but when 
combined with as little as 5 to 10% of an ingredient that is not self-Iubrfeating, 
it requires additional lubricant and usually a gl1dant, such as colloidal 
silicone dioxide, at 0.25%. 
Starch 1500 contains about 10% moisture and is susceptible to softening 
when combined with excessive amounts (greater than 0.5%) of magnesium 
stearate. Direct compaction starches have been reported [47] to not affect 
the stability of aspirin where moisture may be a concern. Most of the formulas 
evaluated also contained microcrystalline cellulose. 
Underwood and Cadwallader [48] studied the effect of various starches 
on the dissolution rate of salicylic acid from tablets. They showed that 
the dissolution of the drug was most rapid from tablets containing a COmpressible 
starch (Fig. 6). 
Mannitol USP finds increasing application in the formulation of chewable 
tablets where mouth-feel and palatability are important considerations. Its 
mouth-feel is related to its negative heat of solution and its slow solubility, 
which is experienced by the user as a cool sensation during dissolution of 
the sugar. It has been reported to be about 72% as sweet as sucrose. One 
gram dissolves in 5.5 ml of water. Chewable vitamins and antacids are the 
primary application for this material, although certain regular chewable

100 Peck. Baley. McCurdy. and Banker 
0' 200 E
Cl 180 w
>..J 160 0
0 140 
..... z 120 ::> 
~ 100 et 
o 10 20 30 40 50 60 70 BO 90 100 110 
TIME (min) 
Figure 6 Dissolution rates of salicylic acid from tablets containing various 
starches, using the USP-NF method type 1 (basket, 100 rpm) at 37C. 
Key: - cornstarch; - potato starch;  rice starch;  arrowroot starch; 
D. compressible starch. 
tablets intended for swallowing do incorporate mannitol because of its nonhygroscopicity. 
Mannitol formulations, because of their poor flow properties, usually 
require higher lubricant levels (3 to 6 times as great) and higher glidant 
levels for satisfactory compression than other diluents. Kanig [49] has 
reported on studies to overcome these shortcomings by spray-congealing 
fused mannitol alone with sucrose or lactose. A wide range of tablet hardness 
can be obtained with mannitol-based tablets. Staniforth et a1. [50] 
crystallized mannitol to produce an excipient with an optimal particle size 
and surface coarseness for a direct -compresston excipient. Mannitol is a 
relatively expensive diluent, and attempts are usually made to reduce its 
quantity per tablet. A granular form of mannitol is now available as a 
direct-compression excipient. Mannitol has been shown to be chemically 
compatible with moisture-sensitive compounds. It picks up less than 1. 0% 
moisture at relative humidities as high as 90%.

Tablet Formulation and Design 101 
Sorbitol is an optical isomer of mannitol but differs dramatically from 
it in that sorbitol is hygroscopic at humidities above 65% and is more watersoluble 
than mannitol. It may be combined with an equal weight of dicalcium 
phosphate to form a direct compaction carrier. Mannitol and sorbitol 
are noncariogenic sugars and are of low nutritional and caloric content. 
Microcrystalline cellulose N. F., often referred to as Avicel, has found 
wide application in the formulation of direct-compaction products. Tablets 
prepared from the more widely used tablet grades PH 101 (powder) and 
102 (granular) show good hardness and friability. The flow properties 
of microcrystalline cellulose have been described by Mendell [40] as poor, 
by Fox et al , [51] as good. and by Livingstone [52] as very good, once 
again indicating that each additive must be evaluated in the formulator's 
own system. 
Numerous other investigators [53-58] have reviewed the applications 
of microcrystalline cellulose in tablet formulations. The capillarity of Avicel 
explains the penetration of water into a tablet, thereby destroying the cohesive 
bonds between particles. The hardness of the compressed tablet 
can significantly affect the disintegration time by breaking down the structure 
of the intermolecular spaces and destroying the capillary properties. 
Avicel is a relatively expensive diluent when compared with lactose 
USP or starch USP. Usually it is not used in tablets alone as the primary 
diluent unless the formulation has a specific need for the bonding properties 
of Avicel. It is capable of holding in excess of 50% active ingredients 
and has certain unique advantages in direct compression which may more 
than offset its higher cost. As a diluent, Avicel offers many interesting 
possibilities to control drug release rates when combined with lactose, 
starch, and dibasic calcium phosphate. Bavitz and Schwartz [23,37] have 
reported on various combinations for use with water-soluble and water-insoluble 
drugs. Avicel possesses the ability to function both as a binder 
and disintegrant in some tablet formulas, which may make it very useful 
in tablets which require improvement in cohesive strength, but which cannot 
tolerate lengthened disintegration times. 
Tablets containing high Avicel levels may be senstive to exposure to 
elevated humidities and may tend to soften when so exposed. 
Dibasic calcium phosphate dihydrate NF, unmilled, is commonly used 
as a tablet diluent. A commercially available free-flowing form is marketed 
as Emcompress and has been described for USe in tablet making by Mendell 
[40]. It is used primarily as a diluent and binder in direct-compaction 
formulas where the active ingredient occupies less than 40 to 50% of the 
final tablet weight. Emcompress is composed of 40- to 200-mesh material, is 
nonhygroscopic, and contains about 0.5% moisture. In direct-compaction 
formulas, 0.5 to 0.75% magnesium stearate is required as a lubricant. It 
shows no apparent hygroscopicity with increasing relative humidities (40 to 
Bavitz and Schwartz [23] showed the negative effect on dissolution of 
increasing the ratio of dibasic calcium phosphate to microcrystalline cellulose 
in a system containing an "insoluble" drug, indomethacin USP (Fig. 7). 
Formula IV (50:50) released 66% of the drug in 30 min. The amount released 
decreased to 18% and 10% in 30 min as the ratio of dibasic calcium 
phosphate to microcrystalline cellulose increased to 70: 30 (formula V) and 
84: 16 (formula VI), respectively. The study highlights the importance of 
carriers when insoluble drugs are employed.

102 Peck, Baley. McCurdy. and Banker 
0 60 - - 
:;) a:: 
0 40 
TIME (min) 
Figure 7 Drug release of an insoluble drug from direct-compression diluents 
diluents (see text). IV = microcrystalline cellulose N.F ./dibasic calcium 
phosphate N.F., 50; 50. V = microcrystalline cellulose N. F ./dibasic calcium 
phosphate N.F . 30: 70. VI = microcrystalline cellulose N.F ./dibasic calcium 
phosphate N.F . 16: 84.

Tablet Formulation and Design 103 
Khan and Rhodes [59] reviewed the disintegration properties of dibasic 
calcium phosphate dihydrate tablets employing insoluble and soluble disintegrating 
agents. The insoluble disintegrants showed a greater effect when 
compressional forces were varied than did the soluble disintegrants. 
The use of a medium coarse dicalcium phosphate dihydrate has been 
reported [60,61]. It has interesting applications in vitamin-mineral formulations 
as both a direct-compaction vehicle and as a source of calcium and 
Sucrose-based tablet diluent-binders are available under a number of 
trade names which include Sugartab (90 to 93% sucrose plus 7 to 10% invert 
sugar). Di-Pao (97% sucrose plus 3% modified dextrins), NuTab (95% sucrose, 
4% invert sugar, and 0.1 to 0.2% each of cornstarch and magnesium 
stearate) . 
All of the above sucrose-based diluent-binders find application in direct 
compaction tablet formulas for chewable as well as conventional tablets. All 
three demonstrate good palatability and mouth-feel when used in chewable 
tablets and can minimize or negate the need for artificial sweeteners. Due 
to their high sucrose level, they may exhibit a tendency to undergo moisture 
uptake. The initial moisture content is usually less than 1% on an 
"as-received" basis. 
NuTab is available to two grades, medium (40 to 60 mesh) and coarse 
(20 to 40 mesh), in white only. Mendes et al , [62] reported on the use of 
NuTab as a chewable direct compression carrier for a variety of products. 
The medium grade of Nu'I'ab , in moisture uptake studies, initially took on 
moisture more rapidly than the coarse; however, both reached the same 
equilibrium uptake of 3.3 to 3.5% after 2 weeks at 80% relative humidity. 
Di-Pae is available in one grade (40 to 100 mesh), the white and six 
colors, while Sugartab comes in one grade (20 to 80 mesh), the white only. 
Tablets made with these sucrose- based diluents at high levels do not 
disintegrate in the classical sense but rather dissolve. 
Confectioner's sugar N. F. may serve as a diluent in both chewable and 
nonchewable tablets. but does require granulation to impart bonding if 
present at significant levels. Powdered sugar is not pure sucrose; it contains 
Calcium sulfate dihydrate N. F. has been suggested as a diluent for 
granulated tablet systems where up to 20 to 30% of active ingredients are 
added to a stock calcium sulfate granulation. It is inexpensive, and has 
been reported to show good stability with many drugs. The recent lack of 
availability of an N.F. grade of material makes its choice as a diluent questionable. 
Two N. F. grades are marketed in the United States. 
Bavitz and Schwartz [23] showed the effect on dissolution rate of a 
calcium sulfate and microcrystalline cellulose based vehicle (product no. 
2834-125) when used with a water-soluble versus a water-insoluble drug. 
The water-soluble drugs showed a rapid release pattern while the waterinsoluble 
drug was released slowly (Fig. 8). 
Calcium lactate trihydrate granular N. F. has been used as diluent and 
binder in direct-compaction formulas with reasonable success. Its longterm 
availability should be reviewed before extensive studies are undertaken. 
Emdex and Celutab are hydrolyzed starches containing 90 to 92% dextrose. 
3 to 5% maltose, and the remainder higher glucose saccharides. 
They are free-flowing powders composed of spray-crystallized maltosedextrose 
spheres. Hydrolyzed starches are often used as mannitol substitutes 
in chewable tablets because of their sweet taste and smooth

fi} 60 
>..J o
::J a: o 
Peck, Baley, McCuY'dy, and Banker 
5 10 15 
TIME (min) 
20 25 30 
Figure 8 Release of a soluble (-. amitriptyline hydrochloride liSP) compared 
to an insoluble (- - -. hydrochlorothiazide liSP) drug from direct 
compression diluents (see text).

Tablet Formulation and Design 105 
mouth-feel. They show good stability with most drugs, but may react 
with drugs having active primary amino groups. when stored at high temperature 
and humidity. Tablets compressed using Emdex show an increase 
in hardness from 2 to 10 kg during the first few hours after compression. 
These materials contain 8.5 to 10.5% moisture, which must be considered 
when combining them with hydrolytically unstable drugs. 
Dextrose, commercially available as Cerelose, can be used as filler, 
carrier, and extender where a sweet material is desired, as in chewable 
tablets. It is available as a hydrate (Cerelose 2001) and in an anhydrous 
form (Cerelose 2401) where low moisture is needed. It can be used to 
partly replace spray-dried lactose in direct-compaction formulas. It requires 
higher lubricant levels than spray-dried lactose and has been shown 
to have a lesser tendency to turn brown than spray-dried lactose. A comparison 
of dextrose and spray-dried lactose has been presented by Duvall 
et al , {26J. 
Inositol has been used as a replacement diluent for chewable tablets 
employing mannitol, lactose, and a sucrose-lactose mixture. 
Hydrolyzed cereal solids such as the Maltrons and Mar-Rex have been 
suggested as lactose replacements. Except for economic considerations, 
their advantages are limited. 
Amylose, a derivative of glucose, possesses interesting direct-compaction 
properties and has been described for use in tablets {63]. Since 
amylose contains 10 to 12% water, its use with drugs subject to hydrolytic 
decomposition should be avoided. 
A list of miscellaneous tablet diluents would be extensive. Some additional 
materials used include Rexcel (food-grade natural source of 0.- and 
amorphous cellulose); Elcema (microfine cellulose, principally an a-cellulose) 
available in powder, fibrous. and granular forms; calcium carbonate; glycine; 
bentonite; and polyvinylpyrrolidone. 
Binders and Adhesives 
Binders or adhesives are added to tablet formulations to add cohesiveness 
to powders, thereby providing the necessary bonding to form granules, 
which under compaction form a cohesive mass or compact referred to as a 
tablet. The location of the binder within the granule can affect the quality 
of the granulation produced [64]. Granule strength is maximized when granulations 
are prepared by roller compaction followed by wet massing and 
spray drying [65]. The formation of granules aids in the conversion of 
powders of widely varying particle sizes to granules, which may more uniformly 
flow from the hopper to the feed system. and uniformly fill the die 
Granules also tend to entrap less air than powders used in a directcompression 
formulation. Table 2 summarizes some common granulating 
The primary criterion when choosing a binder is its compatibility with 
the other tablet components. Secondarily, it must impart sufficient cohesion 
to the powders to allow for normal processing (sizing, lubrication, compression, 
and packaging), yet allow the tablet to disintegrate and the drug to 
dissolve upon ingestion, releasing the active ingredients for absorption. 
Binder strength as a function of moisture has been reported by Healey et 
81. [66]. 
In a study [67] of a comparision of common tablet binder ingredients, 
the materials compressed were, in descending order of adhesive strength:

106 Peck. Baley. McCurdy, and Banker 
Table 2 Examples of Typical Granulating Systems 
System Concentration 
normally used used 
Material (% of granulating) (% of formula) 
Acacia 10-25 2-5 
Cellulose derivatives 5-10 1-5 
Gelatin 10-20 1-5 
Gelatin-acacia 10-20 2-5 
Glucose 25-50 2-25 
Polymethacrylates 5-15 5-20 
Polyvinylpyrrolidone 3-15 2-5 
Starch paste 5-10 1-5 
Sucrose 50-75 2-25 
Sorbitol 10-25 2-10 
Pregelatinized starch 2-5 1-10 
Tragacanth 3-10 1-4 
Sodium alginate 3-5 2-5 
glucose, acacia, gelatin, simple syrup, and starch. Although starch has 
the least adhesive strength of the materials in the list, it also has the 
least deleterious effect on general tablet disintegration rates of the materials 
listed. Different binders can significantly affect the drying rate and required 
drying time of a granulation mass, and the equilibrium moisture level 
of the granulation. 
Acacia, a natural gum, has been used for many years as a granulating 
solution for tablets. In solutions ranging from 10 to 25%, it forms tablets 
of moderate hardness. The availability of acacia has been uncertain over 
the past few years. and it should be avoided for that reason in new formulations. 
In addition to shortages. contamination by extraneous material 
and bacteria makes its use questionable. 
Tragacanth, like acacia, is a natural gum which presents similar problems 
to those of acacia. Mucilage is difficult to prepare and use. Thus, 
adding it dry and activating it through the addition of water works best. 
Such wet granulation masses should be quickly dried to reduce the opportunity 
for microbial proliferation. 
Sucrose, used as a syrup in concentration between 50 and 75%. demonstrates 
good bonding properties. Tablets prepared using syrup alone as 
a binder are moderately strong, but may be brittle and hard. The quantity 
of syrup used and its rate of addition must be carefully followed. especially 
in systems where overwetting occurs quickly. 
Gelatin is a good binder. It forms tablets as hard as acacia or tragacanth, 
but is easier to prepare and handle. Solutions of gelatin must be 
used warm to prevent gelling. Alcoholic solutions of gelatin have been 
used but without great success. Jacob and Plein [68] and Sakr et al , [69]

Tablet Formulation and Design 107 
have shown that increasing the gelatin content of tablets causes increases 
in their hardness, disintegration, and dissolution times. 
Glucose as a 50% solution can be used in many of the same applications 
as sucrose. 
Starch as a paste forms tablets which are generally soft and brittle. 
It requires heat to facilitate manufacture. Depending on the amount of 
heat employed, starch undergoes hydrolysis to dextrin and then to glucose. 
Thus, care in preparation of starch paste is necessary to produce a correct 
and consistent ratio of starch and its hydrolysis products , as well as 
to prevent charring. 
Cellulose materials such as methylcellulose and sodium carboxymethylcellulose 
(CMC) form tough tablets of moderate hardness. They may be used 
as viscous solutions or added dry and activated with water, which results 
in less effective granule formation. They are available in a wide variety 
of molecular weights which affect the viscosity of the solution as well as 
their swelling properties. 
Miscellaneous water-soluble or dispersible binders include alginic acid 
and salts of alginic acid, magnesium aluminum silicate, Tylose, polyethylene 
glycol, guar gum, polysaccharide acids, bentonites, and others. 
Combinations of the previously discussed binding agents often impart 
the desirable properties of each. Some typical combinations include: 
Gelatin + acacia 
Starch paste + sucrose (as a syrup) 
Starch + sucrose (as a syrup) + sorbitol 
Starch + sorbitol 
Some binders are soluble in nonaqueous systems, which may offer advantages 
with moisture-sensitive drugs. Most nonaqueous vehicle binders have as 
their main disadvantages the possible need for explosion-proof drying facilities 
and solvent recovery systems. A number of oven explosions have 
occurred in the pharmaceutical industry-related to the use of alcohol in 
wet granulation. Some manufacturers have used the approach of partially 
air-drying such granulations and then employing high air flow rates in 
their dryers to stay below the explosive limit of alcohol in air. While this 
approach may work for many years without incident, if a power failure 
occurs at the wrong time, alcohol vapor can build to the explosive limit, 
triggering an explosion when the power is turned back on. Great care 
should be taken in drying any granulation employing flammable solvents 
or in designing an oven system for such use. 
Polyvinylpyrrolidone (PVP) is an alcohol-soluble material which is used 
in concentrations between 3 and 15%. Granulations using a PVP-alcohol 
system process (granulate) well, dry rapidly, and compress extremely well. 
PVP finds particular application in multivitamin chewable formulations where 
moisture sensitivity can be a problem. 
Polymethacrylates (Eugragit NE30D, RS30D) can be used as binders in 
wet granulations. It is supplied as a 30% aqueous dispersion. Dilution 
with water prior to use is recommended. 
Hydroxypropylmethylcellulose (HPMC) and hydroxypropylcellulose 
(Klucel) are soluble in various organic solvents or cosolvent systems, as 
well as water. L-HPC is a low molecular weight, crosslinked form of KIucel. 
It functions not only as a tablet binder but also a tablet disintegrant. Unlike 
Klucel, it is not soluble in water. It has tremendous swelling capability

108 Peck, Baley. McCurdy, an d Banker 
which accounts for its disintegration property. L-HPC may be used in both 
direct-compression as well as wet granulation tablet formulas. Various 
grades of L-HPC may be used depending on whether the tablet is to be 
wet- granulated or directly compressed. 
Ethylcellulose (Ethocel) is used as alcohol solutions of 0.5 to 2.0% and 
affords moisture-sensitive components a protective coating. Vitamin A and 
D mixtures, which are usually sensitive to moisture, may be coated with 
ethylcellulose solution, dried, and granulated with conventional aqueous 
systems. Ethylcellulose may have a serious retardant effect on tablet disintegration 
and drug dissolution release. 
Pregelatinized starch (National 1551 and Starch 1500) can be blended 
dry with the various components of a tablet formula and activated with 
water at the desired time of granulation. In a direct-compression formulation, 
no more than 0.5% magnesium stearate should be used to prevent 
softening of the tablets. 
The purpose of a disintegrant is to facilitate the breakup of a tablet after 
admisinistration. Disintegrating agents may be added prior to granulation 
or dulling the lubrication step prior to compression or at both processing 
steps. The effectiveness of many disintegrants is affected by their position 
within the tablet. Six basic categories of disintegrants have been described: 
starches, clays, celluloses, algins, gums, and miscellaneous. It 
should be noted that many disintegrants have also been shown to possess 
binder or adhesive properties. Since disintegration is the opposite operation 
to granulation (agglomeration) and the SUbsequent formation of strong 
compacts, one must carefully weigh these two phenomena when designing 
a tablet. Khan and Rhodes [70] reviewed the water sorption properties of 
four tablet disintegrants: starch, sodium CMC, sodium starch glycolate, 
and a cation exchange resin. The different disintegration properties were 
related to the differing mechanisms by Which the disintegrants affect tablet 
rupture. Intergranular and extragranular disintegrating agents were reviewed 
by Shotton and Leonard [71]. The extragranular formulations disintegrated 
more rapidly than the intragranular ones, but the latter resulted 
in a much finer dispersion of particles. A combination of the two types of 
agents was suggested. Since the most effective lubricants are hydrophobic, 
water -repellent, and function by granule coating, it is not surprising that 
such materials may impede tablet wetting, disintegration, and dissolution. 
To overcome this problem, disintegrants such as starch are often combined 
with the lubricant to provide extragranular disintegration and to facilitate 
tablet wetting. Such combinations of lubricant and disintegrant which are 
added to tablet granulations prior to compression are termed running 
Starches are the most common disintegrating agents (Table 3) in use 
today. Ingram and Lowenthal [72] have attributed their activity as disintegrants 
to intermolecular hydrogen bonding which is formed during compression 
and is suddenly released in the presence of excess moisture. In 
a later study, Lowenthal [73] evaluated the effects of pressure on starch 
granules and showed that they do not regain their original shape when 
moistened with water. 
Lowenthal and Wood [74] showed that the rupture of the surface of a 
tablet employing starch as a disintegrant occurred where starch agglomerates 
were found. The conditions best suited for rapid tablet disintegration are

Tablet Formulation and Design 
Table 3 Starch Disintegrants 
Usual range 
Material ( %) 
Natural starch (corn, potato) 
Sodium starch glycollate (Primogel, Explotab) 
Pregelatinized starch (National 1551) 
Pregelatinized starch (Amijel) 
Modified cornstarch (Starch 1500) 
1-20 (4% 
a sufficient number of starch agglomerates. low compressive pressure. and 
the presence of water. 
Starches show a great affinity for water through capillary action, resulting 
in the expansion and subsequent disintegration of the compressed 
tablet. Formerly accepted swelling theories of the mechanism of action of 
starches as disintegrants have been generally discounted. In general, 
higher levels of starch result in more rapid disintegration times. However, 
high starch levels often result in a loss of bonding. cohesion. and hardness 
in tablets. It has been suggested [45] that an optimum starch level exists 
for many drugs such as tolutamlde , 
It is important to dry starch at 80 to 90C to remove absorbed water. 
Equally important is starch storage while awaiting use. since starches will 
quickly equilibrate to 11 to 13% moisture by plcking up atmospheric moisture. 
Sodium starch glycolate modified starches with dramatic disintegrating 
properties are available as Primogel and Explotab , which are low-substituted 
carboxymethyl starches. While natural predried starches swell in water to 
the extent of 10 to 25%. these modified starches increase in volume by 200 
to 300% in water. One benefit of using this modified starch is that disintegration 
time may be independent of compression force. However. hightemperature 
and humidity conditions can increase disintegration time, slowing 
dissolution of tablets containing this starch. 
Clays such as Veegum HV (magnesium aluminum silicate) have been 
used as disintegrants at levels ranging from 2 to 10%. The use of clays 
in white tablets is limited because of the tendency for the tablets to be 
slightly discolored. In general clays. like the gums. offer few advantages 
over the other more common, often more effective, and no more expensive 
disintegrants such as the starches (Including derivatives), celluloses  and 
Celluloses, such as purified cellulose, methyleellulose , sodium carboxymethylcellulose 
, and carboxymethylcellulose, have been evaluated as disintegrants 
but have not found widespread acceptance. A crosslinked form 
of sodium carboxymethylcellulose (Ac-Di-Sol) has been well accepted as a 
tablet disintegrant. Unlike sodium carboxymethyleellulose , Ac-Di-Sol is 
essentially water-insoluble. It has a high affinity for water which results 
in rapid tablet disintegration. Ac-Di-Sol has been classifed as a "superdisintegrant 
. 1I 
Microcrystalline cellulose (Avicel) exhibits very good disintegrant properties 
when present at a level as low as 10%. It functions by allowing water

110 Peck, Baley, McCurdy, and Banker 
water to enter the tablet matrix by means of capillary pores, which breaks 
the hydrogen bonding between adjacent bundles of cellulose microcrystals. 
Excessively high levels of microcrystalline cellulose can result in tablets 
which have a tendency to stick to the tongue, due to the rapid capillary 
absorption, dehydrating the moist surface and causing adhesion. 
Alginates are hydrophilic colloidal substances extracted from certain 
species of kelp. Chemically they are available as alginic acid or salts of 
alginic acid (with the sodium salt being the most common). They demonstrate 
a great affinity for water, which may even exceed that of cornstarch. 
Alginic acid is commonly used at levels of 1 to 5% while sodium alginate is 
used between 2.5 and 10%. Unlike starch, microcrystalline cellulose, and 
alginic acid, sodium alginates do not retard flow. * 
National 1551 and Starch 1500 are pregelatinized corn starches with 
cold water swelling properties. They swell rapidly in water and display 
good disintegrant properties when added dry at the lubrication step. When 
incorporated into the wet granulation process, pregelatinized starch loses 
some of its disintegrating power. 
Gums have been used as disintegrants because of their tendency to 
swell in water. Similar to the pregelatinized starches in function, they can 
display good binding characteristics (1 to 10% of tablet weight) when wet. 
This property can oppose the desired property of assisting disintegration, 
and the amount of gum must be carefully titrated to determine the optimum 
level for the tablet. Common gums used as disintegrants include agar, guar, 
locust bean, Karaya, pectin, and tragacanth. Available as natural and synthetic 
gums. this category has not found wide acceptance because of its inherent 
binding cap abilities. 
Miscellaneous aisin tegrants include surfactants, natural sponge, resins, 
effervescent mixtures, and hydrous aluminum silicate. Kornblum and 
Stoopak I 75] evaluated cross-linked PVP (Povidone-XL) as a tablet disintegrant 
in comparison with starch USP and alginic acid. The new material 
demonstrated superiority over the other two disintegrants tested in most of 
the experimental tablet formulations made as direct compaction or wet granulation 
systems. Povidone-XL also falls under the classification of superdisintegrant 
Lubricants, Antiadherents, and Glidants 
The primary function of tablet lubricants is to reduce the friction arising 
at the interface of tablet and die wall during compression and ejection. 
The lubricants may also possess antiadherent or glidant properties. 
Strickland [76] has described: 
Lubricants: Reduce friction between the granulation and die wall 
during compression and ejection 
Antiadherents: Prevent sticking to the punch and, to a lesser extent, 
the die wall 
Glidants: Improve flow characteristics of the granulation 
*A wide variety of grades are available from the Algin Corporation of 

Tablet Formulation and Design 111 
LUbrication is considered to occur by two mechanisms. The first is 
termed fluid (or hydrodynamic) lubrication because the two moving surfaces 
are viewed as being separated by a finite and continuous layer of 
fluid lubricant. A hydrocarbon such as mineral oil, although a poor lubricant, 
is an example of a fluid-type lubricant. Hydrocarbon oils do not 
readily lend themselves to application to tablet granulations and, unless 
atomized or applied as a fine dispersion, will produce tablets with oil spots. 
The second mechanism, that of boundary lubrication, results from the adherence 
of the polar portions of molecules with long carbon chains to the 
metal surfaces of the die wall. Magnesium stearate is an example of a 
boundary lubricant. Boundary-type lubricants are better than fluid-type 
lubricants since the adherence of a boundary lubricant to the die wall is 
greater than that of the fluid type. This is expected since the polar end 
of the boundary lubricant should adhere more tenaciously to the oxide metal 
surface than the nonpolar fluid type. 
The type and level of lubricant used in a tablet formulation is greatly 
affected by the tooling used to compress the tablets. Mohn [77] reviewed 
the design and manufacture of tablet tooling. Proper inspection of tablet 
tooling is critical to ensure that tooling continues to perform up to expectations. 
Capping of tablet is more often formulation -related; however, it 
can be caused by improper tooling. Compressing tablets at pressures 
greater than what the tooling was designed to handle can result in damage 
to punch heads. The use of cryogenic material treatments can increase 
tooling life. Vemuri [78] discussed the selection of the proper tooling for 
high speed tablet presses. 
Recommendations have been made to standardize tablet-tooling specifications 
by the IPT Section of the Academy of Pharmaceutical Sciences [79]. 
Mechtersheimer and Sucker [80] determined that die wall pressure is 
considerably greater when curve-faced punches are used to compress tablets 
instead of fat-faced punches. Additional lubricant is often needed in 
tablet formulations that are to be compressed with curved-face punches. 
Lubricants tend to equalize the pressure distribution in a compressed 
tablet and also increase the density of the particle bed prior to compression. 
When lubricants are added to a granulation, they form a coat around 
the individual particles (granules) which remains more or less intact during 
compression. This coating effect may also extend to the tablet surface. 
Since the best lubricants are hydrophobic, the presence of the lubricant 
coating may cause an increase in the disintegration time and a decrease in 
the drug dissolution rate. Since the strength of a tablet depends on the 
area of contact between the particles, the presence of a lubricant may also 
interfere with the particle-to-particle bond and result in a less cohesive 
and mechanically weaker tablet. Matsuda et al , [81] reviewed the effect 
on hardness and ejection force of two methods of applying the lubricant 
(stearic acid, magnesium stearate, calcium stearate. and talc) to statically 
compressed tablets prepared from a lactose granulation. In one method of 
addition the lubricant was incorporated into the granulation during preparation, 
while in the other it was added to (mixed with) the final granules. 
The mixing method gave better results for ease of ejection and tablet hardness 
than the incorporation method. 
As the particle size of the granulation decreases, formulas generally 
require a greater percent of lubricant. Danish and Parrott [82] examined

112 Peck, Baley, McCurdy, and Banker 
the effect of concentration and particle size of various lubricants on the 
flow rate of granules. For each lubricant there was an optimum concentration, 
not exceeding 1%, which produced a maximum flow rate. For a constant 
concentration of lubricant, the flow rate increased to a maximum rate 
as the size of the lubricant particles was decreased to 0.0213 em, A further 
reduction hindered the flow rate. Usually as the concentration of 
lubricant increases, the disintegration time increases and the dissolution 
rate decreases, as the ability of water to penetrate the tablet is reduced. 
The primary function of a lubricant is to reduce the friction between 
the die wall and the tablet edge as the tablet is being ejected. Lack of 
adequate lubrication produces binding, which results in tablet machine 
strain and can lead to damage of lower punch heads, the lower cam track, 
and even the die seats and the tooling itself. Such binding on ejection is 
usually due to a lack of lubrication. Such tablets will have vertically 
scratched edges, will lack smoothness or gloss, and are often fractured 
at the top edges. With excessive binding the tablets may be cracked and 
fragmented by ejection. Ejection force can be monitored as an indicator of 
adhesion problems during compressing studies [83]. A film forms on the 
die wall, and ejection of the tablet is difficult. 
Sticking is indicated by tablet faces which are dull. Earlier stages of 
sticking are often referred to as filming of the punch faces and may result 
when punches are improperly cleaned or polished or when tablets are 
compressed in a high humidity, as well as when lubrication is inadequate. 
Advanced states of sticking are called picking, which occurs when portions 
of the tablet faces are lifted or picked out and adhere to the punch face. 
Picking usually results from improperly dried granulations, from punches 
with incorrectly designed logos, and from inadequate glidant use, especially 
when oily or sticky ingredients are compressed. 
Capping and laminating, while normally associated with poor bonding, 
may also occur in systems which are overlubricated with a lubricant such 
as a stearate. Attempts have been made to measure the tendency of a 
powder to cap and stick when compressed based on theoretical calculations 
[83]. Rue et al , [84] correlated acoustic emissions during tableting of 
acetaminophen with lamination and capping events. Acoustic emission analysis 
demonstrated that capping occurs within the die wall during the decompression 
phase and not during ejection. Capping or lamination observed 
with curve-face punches can often be eliminated by switching to flat-faced 
Lubricants may be further classified according to their water solubility (as 
(as water-soluble or water-insoluble). The choice of a lubricant may depend 
in part on the mode of administration and the type of tablet being produced, 
the disintegration and dissolution properties desired, the lubrication and 
flow problems and requirements of the formulation, various physical properties 
of the granulation or powder system being compressed, drug compatibility 
considerations, and cost. 
Water-insoluble lubricants in general are more effective than watersoluble 
lubricants and are used at a lower concentration level. Table 4 
summarizes some typical insoluble lubricants and their usual use levels. 
In general lubricants, whether water-soluble or insoluble, should be 
200 mesh or finer and are passed (bolted) through a lOO-mesh screen 
(nylon cloth) before addition to the granulation. Since lubricants function 
by coating (as noted), their effectiveness is related to their surface area

Tablet Formulation and Design 
Table 4 Water-Insoluble Lubricants 
Usual range 
Material ( %) 
Stearates (magnesium. 
calcium. sodium) 
Stearic acid 
1/4- 2 
and the extent of particle size reduction. The specific lubricant, its 
surface area. the time (point) and procedure of addition. and the length 
of mixing can dramatically affect its effectiveness as a lubricant and the 
disintegration-dissolution characteristics of the final tablet. 
Glyceryl behapate (Comp ritol 888) is a new addition to the list of tablet 
lubricants. It has the unique classification of being both a lubricant and 
a binder. Therefore. it should alleviate both sticking and capping problems. 
When used with magnesium stearate in a tablet formula, its level 
should be reduced. The stability of aspirin has been extensively studied 
in conjunction with various lubricants. In combination with talc, the rate 
of decomposition has been related to the calcium content and loss on ignition 
of the talc source. Alkaline materials such as alkaline stearate lubricants 
may be expected to have a deleterious effect on the stability of 
aspirin-containing products. For those formulations that are not sufficiently 
lubricated with stearates, the addition of talc may be beneficial. Mechtersheimer 
and Sucker [80] also found that talc should be added prior to the 
lubrication step to optimize the tableting properties. When added together, 
talc and magnesium stearate provided acceptable lubrication. Magnesium 
lauryl sulfate has been compared to magnesium stearate as a tablet lubricant 
[83]. Higher levels of magnesium lauryl sulfate were required to 
provide an equivalent lubricantion as measured by tablet ejection force. 
However. harder and more compressible blends can be prepared with magnesium 
lauryl sulfate than with magnesium stearate at the same ejection 
Boron-coated tablet tooling has permitted the use of a lower lubricant 
level in some tablet formulations [85]. 
Water-soluble lubricants are in general used only when a tablet must 
be completely water-soluble (e. g . effervescent tablets) or when unique 
disintegration or, more commonly, dissolution characteristics are desired. 
Possible choices of water-soluble lubricants are shown in Table 5. Boric 
acid is a questionable member of the list due to the recognized toxicity of 
boron. A review of some newer water-soluble lubricants combined with talc 
and calcium stearate has been reported [86]. Polyethylene glycols and 20 
low melting point surfactants have been suggested as water-soluble lubricants 

114 Peck. Baley. McCurdy. and Banker 
Table 5 Water-Soluble Lubricants 
Usual range 
Material ( %) 
Boric acid 1 
Sodium benzoate + sodium acetate 
Sodium chloride 
Carbowax 4000 
Carbowax 6000 
Sodium oleate 
Sodium benzoate 
Sodium acetate 
Sodium lauryl sulfate 
Magnesium lauryl sulfate 
Methods of Addition. Lubricants are generally added dry at a point 
where the other components are in a homogeneous state. Thus, the lubricant 
is added and mixed for a period of only 2 to 5 minutes rather than the 
10 to 30 minutes necessary for thorough mixing of a granulation. Overmixing 
may lead to diminished disintegration-dissolution characteristics and loss of 
bonding in the tablet matrix. 
Lu bricants have also been added to granulations as alcoholic solutions 
(e.g., Carbowaxes) and as suspensions and emulsions of the lubricant 
material. In one study [88] various lubricants were added, without significant 
loss of lubricating properties, to the initial powder mixture prior to 
wet granulation. However, as a rule, powdered lubricants should not be 
added prior to wet granulation since they will then be distributed throughout 
the granulation particles rather than concentrated on the granule surface 
where they operate. In addition, powder lubricants added in this 
manner will reduce granulating agent and binder efficiency. 
Antiadherents are useful in formulas which have a tendency to pick 
easily. Multivitamin products containing high vitamin E levels often display 
extensive picking, which can be minimized through the use of a colloidal 
silica such as Syloid. Studies have indicated that Cab-O-Sil, although 
similar chemically, does not perform satisfactorily, probably because of its 
lesser surface area. 
Talc, magnesium stearate, and cornstarch display excellent punch-face 
or antiadherent properties. An extremely efficient yet water-soluble punchface 
lubricant is DL-Ieucine. The use of silicone oil as an antiadherent 
has been suggested [89]. Table 6 summarizes the more common antiadherents.

Tablet Formulation and Design 
Table 6 Antiadherents 
Usual range 
Material ( %) 
Talc 1-5 
Cornstarch 3-10 
Cab-O-Sil 0.1-0.5 
Syloid 0.1-0.5 
DL-Leucine 3-10 
Sodium lauryl sulfate <1 
Metallic stearatas <1 
In general materials that are good glidants are poor lubricants. Table 7 
lists a few of the common glidants. Glidants can improve the flow of granulations 
from hoppers into feed mechanisms and ultimately into the die cavity. 
Glidants can minimize the degree of surging and "starvation" often exhibited 
by direct-compaction formulas. They act to minimize the tendency of a granulation 
to separate or segregate due to excessive vibration. High-speed tablet 
presses require a smooth, even flow of material to the die cavities. When 
flow properties are extremely poor, and glidants are ineffective, consideration 
of forced-feed mechanisms may be necessary. The uniformity of tablet weights 
directly depends on how uniformly the die cavity is filled. 
Tablet Formulation and Design 
A review by Augsburger and Shangraw [90] of a series of silica-type 
glidants used decreased weight variation as a criterion of evaluation. The 
use of starch as a glidant has been widely practiced in tablet and capsule 
formulation. In general many materials commonly referred to as lubricants 
possess only a minimal Iubricattng activity, and are better glidants or 
Table 7 Glidants 
Usual range 
Material ( %) 
Talc 5 

116 Peck, Baley, McCurdy, and Banker 
antiadherents. Thus, a blend of two or more materials may be necessary to 
obtain the three properties. 
York [91] presented data indicating the relative efficiency of glidants for 
two powder systems and reported that the order of effectiveness was 
Fine silica> magnesium stearate> purified talc 
The mechanisms of action of glidants have been hypothesized by various investigators 
and include: 
1- Dispersion of electrostatic charges on the surface of granulations [92. 
2. Distribution of glidant in the granulation [94] 
3. Preferential adsorption of gases onto the glidant versus the granulation 
4. Minimization of van der Waals forces by separation of the granules [92] 
5. Reduction of the friction between particles and surface roughness by 
the glidant's adhering to the surface of the granulation [92.93] 
The most efficient means of measuring the effectiveness of a glidant in a 
powder blend is to compress the blend and determine weight variation. The 
use of shear cell and flowmeter data also gives some indication of the flow 
properties of a particular blend. A complete shear cell analysis of a powder 
blend can be performed to determine the appropriate hopper design (Le , 
angle from vertical, orifice diameter. hopper diameter, and material of con ~ 
struction). Shear-cell analysis also provides information on the tendency of 
a blend to consolidate with time and under a load. Excessive consolidation 
can result in a good-flowing formulation turning into a poor-flowing formulation. 
Nyqvist 195] correlated the frequency of tablet machine adjustments 
with shear cell and flowmeter data. The moisture content of dried granulations 
was found to impact on the flowability of the granules. 
The Running Powder. Since the best lubricants are not only water-dnsoluble 
but also water-repellent, and since lubricants function by coating 
the granulation to be compressed, it is not surprising that the lubricants 
used and the process of lubrication may have a deleterious effect on tablet 
disintegration and drug dissolution release. To overcome the tendency a 
second agent is often added to the lubricant powder to produce a less hydrophobic 
powder to be added as the lubricant system. The mixture of lubricant 
and a second. hydrophilic agent is called the running powder, since it is 
added to permit compression or running of the granulation on a tablet machine. 
The most common hydrophilic agent added to the lubricant is starch. The 
starch/lubricant ratio is typically in the range 1: 1 to 1: 4. 
Colors are incorporated into tablets generally for one or more of three purposes. 
First, colors may be used for identifying similar-looking products 
within a product line, or in cases where products of similar appearance exist 
in the lines of different manufacturers. This may be of particular importance 
when product identification (because of overdosing or poisoning and drug 
abuse) is a problem. Second, colors can help minimize the possiblity of mixups 
during manufacture. Third, and perhaps least important, is the addition 
of colorants to tablets for their aesthetic value or their marketing value.

Tablet Formulation and Design 117 
The difficulties associated with the banning of FD&C Red No. 2 (amaranth), 
FD&C Red No.4, and carbon black in 1976 should be a prime example of what 
may be the trend of the future. Other colors such as FD&C No. 40 and FD&C 
Yellow No. 5 have been questioned recently and will continue to be suspect 
for one reason or another. The pharmaceutical manufacturer can maximize 
the identification of his products through product shape and size, NDC number, 
and use of logos. One should not rely on color as a major means of eliminating 
in-house errors but should instead develop adequate general manufacturing 
practices to insure that mix-ups do not occur. 
Today the formulator may choose a colorant from a decreasing list of colors 
designated as D&C and FD&C dyes and lakes, and a small number of acceptable 
natural and derived materials approved for use by the U. S. Food and 
Drug Administration. Historically, drug manufacturers have, for the most 
part, restricted their choice of dyes to the FD&C list. Table 8 summarizes 
the colors available at this time. 
Dyes are water-soluble materials, whereas lakes are formed by the absorption 
of a water-soluble dye on a hydrous oxide (usually aluminum hydroxide) , 
which results in an insoluble form of the dye. 
The photosensitivity of lakes and dyes will be affected by the drug, exeipients, 
and methods of manufacture and storage of each product. Ultravioletabsorbing 
chemicals have been added to tablets to minimize their photosensitivity. 
Pastel shades generally show the least amount of mottling, especially in 
systems utilizing water-soluble dyes. Colors near the mid-range of the visible 
spectrum (yellow, green) will show less mottling than those at either extreme 
(blue, red). 
Methods of Incorporation 
Water-soluble dyes are usually dissolved in the granulating system for incorporation 
during the granulating process. This method assures uniform 
distribution through the granulation but can lead to mottling during the drying 
process. Colors may also be adsorbed onto carriers (starch, lactose, 
calcium sulfate, sugar) from aqueous or alcoholic solutions. The resultant 
color mixtures are dried and used as stock systems for many lots of a particular 
product. Water-soluble dyes may also be dry-blended with an excipient 
prior to the final mix. 
Lakes are almost always blended with other dry excipients because of 
their insoluble nature. In general, direct -compression tablets are colored 
with lakes because no granulation step is used. 
Flavors and Sweeteners 
Flavors and sweeteners are commonly used to improve the taste of chewable 
tablets. Cook [96] reviewed the area of natural and synthetic sweeteners. 
Flavor's are incorporated as solids in the form of spray-dried beadlets and 
oils, usually at the lubrication step, because of the sensitivity of these materials 
to moisture and their tendency to volatilize when heated (e. g. I during 
granulation drying). Aqueous (water-soluble) flavors have found little acceptance 
due to their lesser stability upon aging. 
Since oxidation destroys the quality of a flavor, oils are usually emulsified 
with acacia and spray-dried. Dry flavors are easier to handle and are generally 
more stable than oils. Oils are usually diluted in alcohol and sprayed onto 
the granulation as it tumbles in a lubrication tub . Use of a P-K V-blender 
with an intensifier bar has also been used. Oils may also be adsorbed onto 
an excipient and added during the lubrication process. Usually, the maximum

118 Peck, Baley, McCur'dy, and Banker 
Table 8 Status of Color Additives: Code of Federal Regulations (4-1-87) 
FD&C Blue No. 1 
FD&C Blue No.2 
D&C Blue No.4 
D&C Blue No.9 
FD &C Green No. 3 
D&C Green No.5 
D&C Green No.8 
D&C Orange No.4 
D&C Orange No.5 
D&C Orange No. 10 
D&C Orange No. 11 
D&C Orange No. 17 
FD&C Red No.3 
FD&C Red No. 4 
D&C Red No.6 
D&C Red No.7 
D&C Red No.8 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice. 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice. 
May be used in externally applied drugs in amounts 
consistent with current good manufacturing practice. 
May be used for coloring cotton and silk surgical 
sutures including sutures for ophthalmic use in 
amounts not to exceed 2.5% by weight of the suture. 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice. 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice. 
May be used in externally applied drugs in amounts 
not exceeding 0.01% by weight of the finished product. 
May be used for coloring externally applied drugs in 
amounts consistent with current good manufacturing 
May be used for coloring mouthwashes and dentifrices 
and for externally applied drugs in amounts not to 
exceed 5 mg per daily dose of the drug. 
May be used for coloring externally applied drugs 
in amounts consistent with current good manufacturing 
May be used for coloring externally applied drugs in 
amounts consistent with current good manufacturing 
May be used for coloring externally applied drugs 
in amounts consistent with current good manufacturing 
May be used for coloring ingested drugs in amounts 
consistent with current good manufacturing practice. 
May be used for externally applied drugs in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drugs such that the combined 
total of D&C Red No.6 and D&C Red No.7 
does not exceed 5 mg per daily dose of the drug. 
May be used for coloring drugs such that the combined 
total of D&C Red No.6 and D&C Red No.7 
does not exceed 5 mg per daily dose of the drug. 
May be used for coloring ingested drugs in amounts 
not exceeding 0.1% by weight of the finished product

Tablet Formulation and Design 
Table 8 (Continued) 
D&C Red No.9 
D&C Red No. 19 
D&C Red No. 21 
D&C Red No. 22 
D&C Red No. 27 
D&C Red No. 28 
D&C Red No. 30 
D&C Red No. 31 
D&C Red No. 34 
D&C Red No. 39 
FD&C Red No. 40 
D&C Violet No.2 
FD&C Yellow No.5 
FD&C Yellow No.6 
D&C Yellow No. 7 
D&C Yellow No. 10 
and for externally applied drugs in amounts consistent 
with current good manufacturing practice. 
May be used for externally applied drugs in amounts 
consistent with current good manufacturing practice. 
May be used for externally applied products in 
amounts consistent with current good manufacturing 
May be used for externally applied products in 
amounts consistent with current good manufacturing 
May be used for coloring drug product in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drug product in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drug product in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drug product in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drug product in amounts 
consistent with current good manufacturing practice. 
May be used for externally applied drugs in amounts 
consistent with current good manufacturing practice. 
May be used for coloring externally applied in 
amounts consistent with current good manufacturing 
May be used for external germicidal solutions not to 
exceed 0.1% by weight of the finished drug product. 
May be used in coloring drugs SUbject to restrictions 
and in amounts consistent with current good manufacturing 
May be used for coloring externally applied drugs in 
amounts consistent with current good manufacturing 
In general products containing FD&C Yellow No.5 
(tartrazine) must be so labeled. The Code of Federal 
Regulations should be consulted for use restrictions 
that may be added. 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice. 
May be used for externally applied drugs in amounts 
consistent with current good manufacturing practice. 
May be used for coloring drugs in amounts consistent 
with current good manufacturing practice.

120 Peck. Baley, McCurdy, and Banker 
Table 8 (Continued) 
D&C Yellow No. 11 
D&C Lakes, Ext. 
D&C Lakes, FD&C 
May be used for externally applied drugs in amounts 
consistent with current good manufacturing practice. 
Consult the current regulations for status. 
amount of oil that can be added to granulation without affecting the bond or 
flow properties is 0.75% (w Iw). 
Sweeteners are added primarily to chewable tablets when the commonly 
used carriers such as mannitol, lactose, sucrose, and dextrose do not sufficiently 
mask the taste of the components. 
Saccharin, which is FDA-approved , is about 400 times sweeter than 
sucrose. The major disadvantage of saccharin is its bitter aftertaste, which 
can sometimes be minimized by incorporating a small quantity (1%) of sodium 
chloride. The saccharin aftertaste is highly discernible to about 20% of the 
Aspartame, a nondrug approved artificial sweetener, is about 180 times 
sweeter than sucrose and is approved for use in beverages, desserts, and 
instant coffee and tea. It exhibits discoloration in the presence of ascorbic 
acid and tartaric acid. thus greatly limiting its use. Becuase of the possible 
carcinogenicity of the artificial sweeteners (cyclamates and saccharin), pharmaceutical 
formulators are increasingly attempting to design their tablet products 
without such agents. The following formulation represents such a system 
for a chewable antacid tablet. 
Example 1: Chewable Antacid Tablet, Aluminum Hydroxide, 
and Magnesium Carbonate Codried Gel (Direct Compression) 
Aluminum hydroxide and magnesium carbonate 
codrled gel (Reheis F-MA 11) 
Mannitol, USP (granular) 
Microcrystalline cellulose 
Calcium stearate 
Quanity per 
325.0 mg 
675.0 mg 
75.0 mg 
30.0 mg 
22.0 mg 
Blend all ingredients and compress using a 5/9-in. flat-faced 
level edge punch to a hardness of 8 to 11 kg (Strong-CobbArner 

Tablet Formulation and Design 121 
Adsorbents such as silicon dioxide (Syloid, Cab-O- Sil, Aerosil) are capable 
of retaining large quantities of liquids without becoming wet. This allows 
many oils, fluid extracts, and eutectic melts to be incorporated into tablets. 
Capable of holding up to 50% of its weight of water, silicon dioxide adsorbed 
systems often appear as free-flowing powders. This adsorbent characteristic 
explains why these materials function well in tablet formulations to alleviate 
picking, especially with high-level vitamin E tablets. Silicon dioxide also exhibits 
glidant properties and can play both a glidant and an adsorbent role in 
the formula. 
Other potential adsorbents include clays like bentonite and kaolin, magnesiurn 
silicate, tricalcium phosphate, magnesium carbonate, and magnesium oxide. 
Usually the liquid to be adsorbed is first mixed with the adsorbent prior to 
incorporation into the formula. Starch also displays adsorbent properties. 
In 1974 the U. S. Congress received a report on Drug Bioeq uivalence from 
the Office of Technology Assessment which noted as a major conclusion the 
potential influence of excipients on the bioavailability of many drug products. 
A further major comment made in the report, which has been largely overlooked 
as readers focused on the bioavailability issue, was a strong criticism 
regarding the current standards for excipients in the compendia. Obviously, 
if test methods for excipients are nonspecific and incomplete, especially as 
these properties may relate to bioavailability of drug products, compendial 
and other government standards cannot provide good assurance of the bioequivalence 
of marketed drug products. The report went on to note that many 
commonly used excipients (including those used in tablets and other solid dosage 
forms) were not even included in the compendia. 
The general notices of USP XX and NF XV contain broad, restrictive 
statements that require all excipients to be harmless in the amounts used, 
not to exceed the minimum amounts needed to produce the intended effect, 
not to impair the bioavailability or therapeutic effect of the drug(s) in the 
dosage form. and not to produce interference with any of the assays or tests 
required to determine adherence to compendial standards. Cooper [97] tabulated 
the various types of tests and standards applied to the 223 excipients 
listed in USP XIX and NF XIV. Each excipient has either a specific assay 
or an identity test. or both, together with various limit tests, which may include 
water content or loss on drying (for less than 80 exoipients) , tests for 
chloride. sulfate. arsenic, heavy metals, ash. residue on ignition. various 
specific or nonspecific impurities, tests for solubility or Insolubility (23 excipients), 
and tests for other specified physicochemical properties (24 excipients) 
A. Physicochemical Test Methods for Excipients 
While it has been known for some time that many (if not most) pharmaceutical 
excipients were lacking in characterizing physicochemical tests. the Swiss 
drug companies were the first to take corrective steps, when they specified 
certain standard physical tests for excipients in their Katalog Pharmazeutischer

122 Peck, Baley, McCurdy, and Banker 
Hilfsstoffe (Catalog of Pharmaceutical Excipients). John Rees of the Department 
of Pharmacy. University of Aston, Birmingham, England. has translated 
these tests for German to English, as they are given in the Swiss catalog. 
Five of the standard tests are given there, since they relate to excipients 
for tablets, and since detailed tests for these properties are not given in the 
current compendia. Other tests in the catalog will not be detailed (for vapor 
density, flash point, fire point. ignition temperature. explosive limits, or 
maximum working conditions concentration). 
The development of the Handbook of Pharmaceutical Bxcipients by the 
Academy of Pharmaceutical Sciences of the American Pharmaceutical Association 
in collaboration with the Pharmaceutical Society of Great Britain has produced 
a reference text with a comprehensive list of pharmaceutical excipients 
and suitable standards for each. This reference should prove to be invaluable 
to the formulator [22] . 
In selecting excipients for pharmaceutical dosage forms and drug products. 
the development pharmacist should be certain that standards exist and are 
available to assure the consistent quality and functioning of the excipient from 
lot to lot. 
A major task of the committee that worked on the Handbook of Pharmaceutical 
Excipients was the development of standard test methods for important 
excipient properties. Standard methods to evaluate over 30 physical properties 
were developed. 
The reader is urged to become familiar with the test methods, published 
in the Handbook, that allow comprehensive characterization of tablet excipient 
materials, especially the following: 
Flow rate 
Gel strength (binders) 
Lubricity (frictional) 
Microbiological status 
Moisture sorption 
Particle hardness 
Particle size distribution: 
(1) sieve analysis 
(2) air permeability 
Shear rate 
Tensile strength 
Volume, bulk 
Water absorption 
Water adsorption 
B. Tablet Formulation for I nternational Markets 
Many drug companies must consider regulatory requirements in many parts 
of the world when they undertake the formulation of new tablet products or 
reformulation of existing products. This is true not only for the largest drug 
companies with major international divisions, but is also the case for much 
smaller companies who market abroad through a separate foreign manufacturing 
or distributing company. or who hope (in the future) to license their product 
for foreign sale. Such formulations must take into account not only the acceptability 
of various excipients in the other countries and areas of the world 
of interest, but also the environmental restrictions of these countries which 
may impact on proposed manufacturing methods (e. g . the proposed solvents 
used, if any) and the worldwide availability of all excipient components in the 
required purity and specifications. While little information may be found in 
any literature compilation on this subject , Hess [98] presented a symposium 
paper in 1976 on the choice of excipients for international use; much of the 
following information has been drawn from this presentation. 
Excipients that are in use in the pharmaceutical industry for tablets or 
other oral dosage forms generally fall into one of the following categories: 
(1) excipients permitted in foodstuffs; (2) excipients described in

Tablet Formulation and Design 123 

pharmacopoeias; (3) newer excipients with no official status, but registered 
with health authorities in various countries of the world, and approved for 
use in some of these countries. 
Excipients permitted in foodstuffs are generally regarded as acceptable 
for like uses in drug products. Materials approved for excipient uses 
(e. g., fillers, surfactants, preservatives, binding agents) have usually 
been extensively tested in food and will be used in relatively low amounts 
as a tablet or pharmaceutical component compared to use as a food component. 
In general, an excipient listed in a major pharmacopoeia such as 
the United States, British. or European Pharmacopoeia can be used worldwide. 
An exception to this rule should be noted for Japan, where only exeipients 
named in one of the official Japanese compendia may be used. These 
compendia currently include: Japan Pharmacopoeia VIII, the Japanese 
Standards of Food Additives III. or the Special Koseisho Regulations. These 
compendia list some excipients not regularly used in the United States or 
Europe (e. g., calcium carboxymethylcellulose), while not listing such common 
ones as the free acid of saccharin (the sodium salt is listed) or diethyl 
phthalate (the dibutyl phthalate is listed). Polyvinylpyrrolidone. which 
was formerly acceptable, has now become restricted. Of the iron oxides 
only the red variety (Fe203) is permitted, while the use of the yellow 
(Fe203 monohydrate) and especially the black oxide (FeO'Fe203) seems 
doubtful. Koseisho, the Japanese health authority, also restricts the use 
of excipients with a pharmacological effect (e. g . citric and ascorbic acid) 
to one-fifth of the minimum daily dose. 
Pharmaceutical manufacturers must be careful to assure that excipients 
listed in pharmacopoeias, and made available by various suppliers around 
the world, do in fact comply with all the relevant pharmacopoeial specifications. 
In certain instances this may restrict the use of very similar, but 
not identical, compounds (e. g., cellulose ethers with different degrees of 
substitution) . 
The development of new materials for use as pharmaceutical excipients 
requires the demonstration of the absence of toxicity and freedom from adverse 
reactions. In OIOSt countries .today it is very difficult to obtain approval 
by regulatory agencies for the use of new excipient agents. Reportedly, 
the only clear recommendations for the type of toxicological data 
currently required on a new excipient are provided in the German regulations 
(1971) and the European Economic Community Directives (EEC 75/318, 
dated May 20, 1975). These regulations and directives call for acute toxicity 
studies in three animal species, observed over 14 days. If possible, 
the LD50 by the parenteral route should also be established in one species. 
The combined acute and long-term studies may be summarized as follows: 
Toxicological data on a new excipient: long-term oral administration 
Acute toxicity: to standard international protocols 
Repetitive administration: 6 months, 2 species (one nonrodent) 
Carcinogenicity: 1 species (18 months, mouse or 2 years, rat) 
Reproduction studies, segments 1, 11, and 111 (fertility, teratogenicity, 
effects on lactation): 1, 11, and 111 (rat); 11 (at least one other 
species nonrodent , e. g., rabbit) 
In the FDA-oriented countries (Australia and Canada in addition to the 
United States), 2-year repetitive-dose studies in rats and I-year studies in

124 Peck, Baley, McCurdy, and Banker 
dogs may be expected to be required rather than the 6-month studies described 
above. It may also be necessary to conduct mutagenicity studies. 
For excipients with any potential for complexation or drug binding. drug 
bioavailability studies will be required for products in which the excipient 
is incorporated. If the excipient is absorbed its ADME and pharmacokinetic 
profile may need to be established. In the event that the agent can be 
clearly demonstrated to not be absorbed from the gut, these later studies 
may be simplified I shortened, or omitted. This would assume the excipient 
is also a well-characterized high-purity agent. Excipients that are clearly 
known to be components of the normal human diet, such as I for example, 
a form of pure cellulose, are much easier to clear with regulatory agencies 
than a compound not normal to the diet, or for which no prior knowledge 
of human exposure or exposure effects exists. The very high cost of obtaining 
the necessary toxicological data for a unique new excipient agent 
makes it obvious that few totally new excipient agents will make their appearance 
in the future. 
Another consideration bearing on excipient use in international markets 
(that is expected to become increasingly important) is the subject of disclosure. 
Paragraph 10 of the 1976 Drug Law of the Federal Republic of 
Germany states that all active ingredients must be publicly declared. This 
requirement includes preservatives because of their antimicrobial activity. 
Whether dyestuffs with a weak allergenic potential should be included in 
this category is still debated. However, in countries such as Sweden, 
lists of drug preparations containing tartrazine and other azo dyestuffs 
have already been published. This obviously leads to a certain marketing 
disadvantage for these products. According to new regulations issued in 
November 1976. the azo dyestuffs tartrazine Sunset Yellow FCF, ponceau 
4R, and amaranth were not to be permitted in foodstuffs in Sweden after 
1979. Prohibitions or major restrictions against these, if not all, azo dyes 
may follow in the years ahead in other countries. Amaranth or FD&C Red 
No. 2 is currently prohibited in the United States, Taiwan, and Venezuela. 
The choice of the excipients to be used in any drug product is usually 
a compromise. This is even more the case in selecting excipients for international 
use, since technical performance must be balanced against local restrictions 
in some countries as well as cost and availability in all countries 
where the product is to be produced. 
Hess [98) has tabulated priorities of use for some common tablet and 
capsule excipients for international use. A number 1 indicates the highest 
priority for use based on all considerations (e. g., compatibility, availability, 
cost) . His tabulations of priority of use for ffller-s and disintegrants and 
for binders, gtidants , and lubricants are shown in Tables 9 and 10. 
In the last few years some powerful new disintegrants for tablets have 
appeared. They are of great assistance where long disintegration times or 
slow dissolution rates are a problem. The compounds have been grouped 
below according to their acceptability; it appears that sodium carboxymethyl 
starch creates the least problem worldwide, even though it is not listed yet 
in any pharmacopoeia. The new disintegrants are: 
Primogel, Scholten (NL): sodium carboxymethyl starch 
Nymcel, ZSB-10 mod, , Nyma (NL): sodium carboxymethylcellulose, low 
degree of substttution 
Plasdone XL, GAF (USA): cross-linked polyvinylpyrrolidone 
LH PC, Shinetsu (J): hydroxypropyl cellulose, low substitution

Tablet Formulation and Design 
Table 9 Priority for Use: Fillers, Disintegrants 
Avicel } Primogel 
Tricalcium phosphate 
OK (formaldehyde) 
OK (except primary amines) 
OK (technical problems) 
OK (hygroscopic point at 77.4% 
relative humidity) 
Somewhat less satisfactory than 
May lose water 
May accelerate hydrolytic 
Source: Adapted from Hess [981. 
Ac-Di-Sol, FMC Corp: internally crosslinked form of sodium carboxymethylcellulose 
of USP purity 
Starch is ranked as the most inert filler and disintegrant. It is also 
generally available worldwide in satisfactory quality at relatively low cost. 
Lactose, though not completely inert, is given a priority of I, based on its 
Table 10 Priority for Use: Binders, Glidants, Lubricants 
Starch paste 
Colloidal silica 
Magnesium stearate 
Calcium stearate 
Stearic acid 
Neutral fats 
Frequently accelerates degradation 
Better than PVP 
Rather worse than HPMC or starch 
Quite reactive 
Mostly OK 
Individual incompatibilities, no 
general rules 
Usually nonreactive 
Source: Adapted from Hess [98].

126 Peck, Baley, McCurdy, and Banker 
Table 11 Legal Status of Carotenoid Food Colors (April 1987) 
Country I3-Carotene 13 -Apocarotenal Canthaxanthin 
European Economic X 
Community Countries 
South American Countries X 
Switzerland X 
United States X 
New Zealand 
South Korea 
USSR and Eastern 
European Countries 
Source: Adapted from Hess [98J. 
worldwide availability and good technical properties. Mannitol, though inert. 
is ranked as second choice because of its less satisfactory technical 
properties. Sucorse is also quite inert and has comp ression properties simi1ar 
to those of lactose, but has a relatively low hygroscopicity point. is 
cariogenic, and is not a desired intake material in some patients. 
The preferred binder, for reasons cited previously, is starch paste. 
Hydroxypropylmethylcellulose (HPMC) and gelatin are less inert; gelatin 
promotes microbial growth, and polyvinylpyrrolidone is not acceptable worldwide. 
Colloidal silica, while being potentially reactive, has unique technical 
properties of combined binding. disintegrating, and lubricant action. Talc, 
though not reactive, is difficult to obtain in good and constant quality. 
Magnesium stearate is rated priority 1, based on availability. while it is 
recognized that different lubricants must be evaluated individually for compatibility 
in any particular application. See Table 10. 
The use of coloring agents to increase the elegance of coated and uncoated 
tablets, or for purposes of product identification, has changed rapid1y 
since 1975. The trend in international product development appears to 
be to use iron oxides and titanium dioxide as tablet colorants and carotenoid 
food colors in tablet coatings in place of FD&C dyes. The legal status of 
the carotenoid food colors is expected to expand in worldwide markets in 
the future. The status of these colors given in Table 11. 
Defined chemical composition and physical properties and defined chemical 
and microbiological properties are essential prerequisites for excipients 
in general, and for excipients for international use in particular. EXcipients 
should conform to the same stringent requirements in all these properties 
as must active ingredients. The most common problems with excipients used 
in international pharmaceutical manufacture are the presence of undesired 
impurities and unacceptable variations in technological performance. The

Tablet Formulation and Design 
Table 12 Microcrystalline Cellulose: Differences in Commercial Grades 
Native cellulose (cotton) 
Microcrystalline cellulose 
Elcema (Rehocel) 
30,000- 50,000 
Degree of 
Source: Huttenrauch and Keiner, Pharmazie, 31: 183 (1976). 
careful choice and continual monitoring of suppliers of excipients in international 
markets is essential. Suppliers who concentrate on the pharmaceutical 
and food industries are usually more reliable and better qualified to 
provide the high-quality products required by the drug industry. 
Drug companies engaged in international manufacture must be assured 
of reliable availability of the excipients they use. The quality and performance 
of excipients used at every manufacturing site must be consistent 
and reliable. Some of the most commonly employed newer classes of tablet 
excipients used internationally include microcrystalline cellulose, most of 
the new dtsintegr-ants , directly compressible excipients composed of lactose, 
various sugars, dicalcium phosphate, and special types of starches. In 
most cases when working with these specialized but very useful materials, 
one product cannot easily be replaced by another. For example, there are 
several brands of so-called microcrystalline cellulose available internationally. 
One type, known by the trade name of Avicel, is obtained by mechanical as 
well as acid treatment; another type (Elcema) is produced by mechanical 
treatment only. This leads to different degrees of crystallinity, which may 
be expected to have an influence on the effectiveness of each agent and on 
the properties of the dosage forms in which they are contained. The much 
higher level in the crystallinity of the Avicel product (Table 12) compared 
to the other microcrystalline forms accounts for its being a superior product 
as a disintegrant and directly compressible material. 
According to Hess [98] companies operating in international markets 
will usually employ brand name or specialty excipients only if they lead to 
a better product, usually one with better controlled bioavailability or one 
with superior mechanical or analytical properties. This will justify their use, 
their possibly higher price, and problems which may be encountered in importing 
these substances (including high import duties). In some countries, 
such as Mexico and India, such imports may not be possible at all or may be 
possible only with great difficulty. There are many difficult decisions, potential 
problems, and pitfalls in choosing excipients in a company which 
operates worldwide. Additional research and development and closer cooperation 
among the industries, the universities, and the regulatory agenciesto 
define the properties, the scope, and the use of pharmaceutical excipientswill 
be needed during the immediate future. In addition. the development 
of a catalog with standards for all the major excipients used in tablet 
making-which are accepted by regulatory agencies around the world-will 
provide a giant step forward for the quality assurance and standardization 
of products made in international markets.

Peck, Baley, McCurdy. and Banker 
1. M. C. Meyer, G. W. A. Slywka, R. E. Dann, and P. L. Whyatt. 
J. Pharm. Sci  63: 1693 (1974). 
2. J. Cooper, Advances in Pharmaceutical Sciences, Vol. 3, Academic 
Press, London, 1971. 
3. J. Cooper and J. Rees, J. Pharm. ScL, 61: 1511 (1972). 
4. G. S. Banker, Tablets and tablet product design, in American Pharmacy, 
7th ed., Lippincott, Philadelphia, 1974, Chap. 11. 
5. F. Sadik, Tablets, in Dispensing of Medication, 8th ed , (J. E. Hoover, 
ed.), Mack PUbl., Easton, PA., 1976, Chap. 5. 
6. T. H. Simon, Preformulation studies, physical and chemical properties 
evaluation-Paper 1 of a symposium entitled, A New Drug Entity A 
Systematic Approach to Drug Development, Am. Pharm , Assoc. Acad. 
Pharrn , Sci., Washington, D.C., 1967. 
7. S. Solvang and P. Finholt, J. Pharm. Sci., 59: 49 (1970). 
8. R. M. Franz, G. E. Peck, G. S. Banker, and J. R. Buck, J. Pharm. 
Sci. 69:621 (1980). 
9. R. L. Plackett and J. P. Burman, Biometrika, 33:305 (1946). 
10. U.S. Patent 4,476,248, October 9, 1984. 
11. V. L. Anderson and R. A. McLean, Design of Experiments: A Realistic 
Approach, Marcel Dekker, New York, 1974. 
12. R. H. Myers, Response Surface Methodology, 1976. 
13. J. Jaffe and N. E. Foss, J. Am. Pharm. Assoc., Sci. sa., 48: 26 
(1959) . 
14. J. Lazarus and L. Lachman, J. Pharm. ScL. 55:1121 (1966). 
15. A. S. Rankell and T. Higuchi, J. Pharm, sa., 57:574 (1968). 
16. P. Rieckmann, Pharm. Z., 34:1207 (1971). 
17. E. Nelson, J. Pharm. sa., 46: 607 (1967) 
18. G. Levy, J. M. Antkowiak, J. A. Procknal, and D. D. White, J. Pharm. 
Sci., 52: 1047 (1963). 
19. L. Lachman, J. Pharm. sa., 54: 1519 (1965). 
20. H. E. Paul, K. J. Hayer, M. F. Paul, and A. R. Borgmann, J. Pharm. 
Sci., 56:882 (1967). 
21. F. A. Campagna, G. Cureton, R. A. Mirigian, and E. Nelson, J. Pharm. 
Sci., 52:605 (1963). 
22. American Pharmaceutical Association, Handbook of Pharmaceutical Excipients, 
Washington, D. C., and the Pharmaceutical Society of Great 
Britain, London, 1986. 
23. J. F. Bavitz and J. B. Schwartz, Drug Cosmo lrui, , 60 (April 1976). 
24. W. P. Bolger and J. J. Gavin, N. Engl. J. Med., 261: 827 (1959). 
25. R. Costello and A. Mattocks, J. Pharm. ScL, 51:106 (1962). 
26. R. N. Duvall, K. T. Koshy, and R. E. Dashiell, J. Pharm. sa., 54: 
1196 (1965). 
27. O. Keller, Informationsaienst APV, 16: 75 (1970). 
28. S. S. Kornblum, J. Pharm. Sci. I 58: 125 (1969). 
29. S. S. Kornblum, Drug Cosmo Ind., 32:92 (1970). 
30. L. Ehrhardt and H. Sucker, Pharm. Ind., 32:92 (1970). 
31. R. G. Daoust and M. J. Lynch, Drug Cosmo Irui, , 93: 26 (1963). 
32. S. A. Sangekar, M. Sarli, and P. R. Sheth, J. Pharm. sa., 61:939 
(1972) . 
33. J. L. Kanig, Paper presented at the Emcompress Symposium, London, 

Tablet Formulation and Design 129 
34. C. F. Lerk, G. K. Bolhuis, and A. H. DeBoer, Pharm. Weekblad, 
109:945 (1974). 
35. H. Nyqvist and M. Nicklasson, Drug Dev , Ind. Pharm., 11: 745 (1985). 
36. N. H. Butuyios, J. Pharm. sa., 55:727 (1966). 
37. J. F. Bavitz and J. B. Schwartz, Drug Cosmo Ind., 114:44 (1974). 
38. C. F. Lerk and G. K. Bolhuis , Pharm. Weekblad, 108:469 (1973). 
39. N. L. Henderson and A. J. Bruno, J. Pharm. Sci., 59:1336 (1970). 
40. E. J. Mendell, Mfg. Chemist Aerosol News, 43:43(1972). 
41. O. Alpar, J. A. Hersey, and E. Shotton, J. Pharm. Pharmacal., 22: 
18-7S (1970). 
42. C. Brownley and L. Lachman, J. Pharm. Sci., 53: 452 (1964). 
43. M. D. Richman, M. S. thesis, University of Maryland (1963). 
44. G. Levy, J. Pharm. Sci., 52: 1039 (1963). 
45. K. C. Commons, A. Bergen, and G. C. Walker, J. Pharm. Sci., 57: 
1253 (1968). 
46. J. B. Schwartz, E. T. Martin, and E. J. Dehner, J. Pharm. Sci., 
64: 328 (1975). 
47. K. S. Manudhane, A. M. Contractor, H. Y. Kim, and R. F. Shangraw, 
J. Pharm. Sci., 58: 616 (1969). 
48. T. W. Underwood and D. E. Cadwallader, J. Pharm. SeL, 61:239 
(1972) . 
49. J. L. Kanig, J. Pharm. Sci., 53:186 (1964). 
50. J. N. Stainiford, et a1., Drug Dev. Ind. Pharm., 7:179-190 (1981). 
51. C. D. Fox, M. D. Richman, G. E. Reier, and R. R. Shangraw, Drug 
Cosmo Ind., 92: 161 (1963). 
52. J. L. Livingstone, Mfg. Chemist Aerosol News, 42: 23 (1970). 
53. O. A. Battista and P. A. Smith, Ind. Eng. Chem., 54:20 (1962). 
54. F. Jaminet and H. Hess, Pharm. Aeta Helv., 41: 39 (1966). 
55. G. M. Enezian, Prod. Prob. Pharm., 23:185 (1968). 
56. M. A. Shah and R. G. Wilson, J. Pharm. SeL, 57:181 (1968). 
57. M. D. Richman, C. D. Fox and R. F. Shangraw, J. Pharm. Sci., 54: 
447 (1965) 
58. G. E. Reier and R. F. Shangraw, J. Pharm. ScL, 55:510 (1966). 
59. K. A. Khan and C. T. Rhodes, J. Pharm. Sci., 64:166 (1975). 
60. J. Kalish, Drug Cosmo Jn d,.; 102: 140 (1968). 
61. F. Jaminet, Pharm. Acta Helv., 43:129 (1968). 
62. R. W. Mendes, M. R. Gupta, I. A. Katz, and J. A. O'Neil, Drug 
Cosmo Ind., 42 (1974). 
63. K. C. Kwan and A. Milosovich, J. Pharm. Sci., 55:340 (1966). 
64. H. Seager, I. Burt, J. Ryder, P. Rue, S. Murray, et al., Int. J. 
Pharm. Technol. Prod. Manuf., 1: 36 (1979). 
65. P. J. Rue, H. Seager, J. Ryder, and I. Burt, Int. J. Pharm. Teehnol. 
Prod. Manuf., 1: 2 (1980). 
66. J. N. C. Healey, M. H. Rubinstein, and V. Walters, J. Pharm. 
Pharmacal. 26: 41P (1974). 
67. G. I. Rubio, An. Fae. Quim. Farm. Chile, 9: 249 (1957). 
68. J. T. Jacob and E. M. Plein, J. Pharm. su., 52:802 (1968). 
69. A. M. Sakr , A. A. Kassem , S. A. A. Aziz , A. H. Shalaby, Mfg. Chem. 
Aerosol News, 43: 38 (1972). 
70. K. A. Khan and C. T. Rhodes, J. Pharm. sa., 64:447 (1975). 
71. E. Shotton and G. S. Leonard, J. Pharm. Sci., 65: 1170 (1976). 
72. J. T. Ingram and W. Lowenthal, J. Pharm. sa., 55:614 (1966). 
73. W. Lowenthal, J. Pharm. Sci., 61: 455 (1972).

130 Peck. Baley, McCurdy. and Banker 
74. W. Lowenthal and J. H. Wood, J. Pharm. ScL, 62:287 (1973). 
75. S. S. Kornblum and S. B. Stoopak, J. Pharm. ScL, 62:43 (1973). 
76. W. A. Strickland, Jr., Drug Cosmo Ind., 85: 318 (1959). 
77. B. Mohn , Pharm. Manuf., 3, 3D, 1986. 
78. S. Vemuri, Pharm. Tech., 6:98 (1982). 
79. Tableting Specification Manual, Academy of Pharmaceutical Sciences, 
American Pharmaceutical Association, Revised, 1981. 
80. B. Mechtersheimer and H. Sucker, Pharm. Tech., 10: 38 (l986). 
81. Y. Matsuda, Y. Minamida, and S. Hayashi, J. Pharm. sa., 65: 1155 
(1976) . 
82. F. Q. Danish and E. L. Parrott, J. Pharm. sa., 60:752 (1971), 
83. A. W. Holzer and J. Sjogren, Drug Dev , Ind. Pharm., 3: 23 (1977). 
84. P. J. Rue, P. M. R. Varkworth, W. P. Ridgway, P. Rough, D. C. 
Sharland, et al., Int. J. Pharm. Techno!. Prod. Manuf., 1: 2 (1979). 
85. T. B. Tsiftsoglou and R. W. Mendes, Pharm. Tech., 6: 30 (1982). 
86. J. MalyandA. Maras, Pharm. Ind., 29:494(1967). 
87. P. Fuchs, E. Schottky, and G. Schenck, Pharm. Ind., 32: 390 (1970). 
88. J. Cooper and D. Pasquale, Pharm. J., 181:397 (1958). 
89. N. T. Vegan, Medd, Norsk Farm. Selskap, 23:169 (1961). 
90. L. L. Augsburger and R. F. Shangraw, J. Pharm. sa., 55:418 (1966). 
91. P. York, J. Pharm. ScL, 64:1217 (1975). 
92. T. M. Jones, Symposium on Powders, Society of Cosmetic Chemists of 
Great Britain, Dublin, April 1969. 
93. M. Paleg and C. H. Mannheim, Powder Techno!.. 7:45 (1972). 
94. B. S. Neumann, Advances in Pharmaceutical Sciences, Vol. 2. Academic 
Press, London, 1967, pp. 194, 207. 
95. H. Nyqvist , Int , J. Pharm. Techno!. Prod. Manuf., 5:21 (1984). 
96. M. K. Cook, Drug Cosmo Ind., Sept. (1975). 
97. J. Cooper, Aust. J. Pharm. Sci., 7:9 (1978). 
98. H. K. Hess, Choice of excipients for international use. Paper presented 
to the Industrial Pharmaceutical Technology Section in a symposium 
entitled tl The Development of Drugs for International Markets, tl 
Am. Pharm , Assoc. Acad. Pharm , Sci , , 21st National Meeting, Orlando, 
FL, November 1976.

Compressed Tablets by Wet Granulation 
Fred J. Sandelln 
Schering-Plough Corporation and University of Tennessee. Memphis J 
Compressed tablets are the most widely used of all pharmaceutical dosage 
forms for a number of reasons. They are convenient, easy to use, portable, 
and less expensive than other oral dosage forms. They deliver a precise 
dose with a high degree of accuracy. Tablets can be made in a variety of 
shapes and sizes limited only by the ingenuity of the tool and die maker 
(i ;e , round, oval, capsule-shaped, square, triangular, etc.). 
Compressed tablets are defined as solid-unit dosage forms made by compaction 
of a formulation containing the drug and certain fillers or excipients 
selected to aid in the processing and properties of the drug product. 
There are various types of tablets designed for specific uses or functions. 
These include tablets to be swallowed per se j chewable tablets formulated 
to be chewed rather than swallowed, such as some antacid and vitamin 
tablets; buccal tablets designed to dissolve slowly in the buccal pouch; 
and sublingual tablets for rapid dissolution under the tongue. Effervescent 
tablets are formulated to dissolve in water with effervescence caused by the 
reaction of citric acid with sodium bicarbonate or some other effervescent 
combination that produces effervescence in water. Suppositories can be 
made by compression of formulations using a specially designed die to produce 
the proper shape. 
The function of tablets is determined by their design. Multilayer tablets 
are made by multiple compression. These are called layer tablets and 
usually consist of two and sometimes three layers. They serve several 
purposes: to separate incompatible ingredients by formulating them in 
separate layers, to make sustained or dual-release products, or merely for 
appearance where the layers are colored differently. Compression-coated 
tablets are made by compressing a tablet within a tablet so that the outer 
coat becomes the coating. As many as two coats can be compressed around 
a core tablet. As with layer tablets, this technique can also be used to 
separate incompatible ingredients and to make sustained or prolonged 

132 Bandelin 
release tablets. Sugar-coated tablets are compressed tablets with a sugar 
coating. The coating may vary in thickness and color by the addition of 
dyes to the sugar coating. Film-coated tablets are compressed tablets with 
a thin film of an inert polymer applied in a suitable solvent and dried. 
Film coating is today the preferred method of making coated tablets. It is 
the most economical and involves minimum time, labor, expense, and exposure 
of the tablet to heat and solvent. Enteric-coated tablets are compressed 
tablets coated with an inert substance which resists solution in gastric fluid, 
but disintegrates and releases the medication in the intestines. Sustained 
or prolonged release tablets are compressed tablets especially designed to 
release the drug over a period of time. 
Most drugs cannot be compressed directly into tablets because they 
lack the bonding properties necessary to form a tablet. The powdered 
drugs, therefore, require additives and treatment to confer bonding and 
free-flowing properties on them to facilitate compression by a tablet press. 
This chapter describes and illustrates how this is accomplished by the 
versatile wet granulation method. 
Whatever method of manufacture is used, the resulting tablets must meet a 
number of physical and biological standards. The attributes of an acceptable 
tablet are as follows: 
1. The tablet must be sufficiently strong and resistant to shock and 
abrasion to withstand handling during manufacture, packaging, 
shipping, and use. This property is measured by two tests, the 
hardness and friability tests. 
2. Tablets must be uniform in weight and in drug content of the individual 
tablet. This is measured by the weight variation test and 
the content uniformity test. 
3. The drug content of the tablet must be bioavailable , This property 
is also measured by two tests, the disintegration test and the dissolution 
test. However, bioavailability of a drug from a tablet, or 
other dosage form, is a very complex problem and the results of 
these two tests do not of themselves provide an index of bioavailability. 
This must be done by blood levels of the drug. 
4. Tablets must be elegant in appearance and must have the characteristic 
shape, color, and other markings necessary to identify the 
product. Markings are usually the monogram or logo of the manufacturer. 
Tablets often have the National Drug Code number printed 
or embossed On the face of the tablet corresponding to the official 
listing of the product in the National Drug Code Compendium of the 
Food and Drug Administration. Another marking that may appear 
on the tablet is a score or crease aeross the face, which is intended 
to permit breaking the tablet into equal parts for the administration 
of half a tablet. However, it has been shown that substantial variation 
in drug dose can occur in the manually broken tablets. 
5. Tablets must retain all of their functional attributes, which include 
drug stability and efficacy.

Compressed Tablets by Wet Granulation 
The size and, to some extent, the shape of the tablet are determined by 
the active ingredient( s) . Drugs having very small doses in the microgram 
range (e. g., folic acid, digitoxin, reserpine, dexamethasone, etc.) require 
the addition of fillers also called excipients to be added to produce a mass or 
or volume of material that can be made into tablets of a size that is convenient 
for patients. A common and convenient size for such low-dosage 
drugs is a 1/4-in. round tablet or equivalent in some other shape. It is 
difficult for some patients to count and handle tablets smaller than this. 
Tablets of this size ordinarily weigh 150 mg or more depending on the density 
of the excipients used to make up the tablet mass. 
As the dose increases, so does the size of the tablet. Drugs with a 
dose of 100 to 200 mg may require tablet Weights of 150 to 300 mg and 
round die diameters of 1/4 to 7/16 in. in diameter depending on the density 
and compressibility of the powders used. As the dose of the active ingrodient(
s) increases, the amount of the excipients and the size of the tablet 
may vary considerably depending on requirements of each to produce an 
acceptable tablet. While the diameter of the tablet may in some cases be 
fixed, the thickness is variable thus allowing the formulator considerable 
latitude and flexibility in adjusting formulations. 
As the dose, and therefore the size, of the tablet increases, the formulator 
uses his expertise and knowledge of excipients to keep the size of the 
tablet as small as possible without sacrificing its necessary attributes. Formulation 
of a tablet, then, requires the following considerations: 
1. Size of dose or quantity of active ingredients 
2. Stability of active ingredient(s) 
3. Solubility of active ingredient(s) 
4. Density of active ingredient(s) 
5. Compressibility of active ingredient(s) 
6. Selection of excipients 
7. Method of granulation (preparation for compression) 
8. Character of granulation 
9. Tablet press, type, size, capacity 
10. Environmental conditions (ambient or humidity control) 
11. Stability of the final product 
12. Bioavailability of the active drug content of the tablet 
The selection of excipients is critical in the formulation of tablets. Once 
the formulator has become familiar with the physical and chemical properties 
of the drug, the process of selecting excipients is begun. The stability of 
the drug should be determined with each proposed excipient. This can be 
accomplished as follows: In the laboratory, prepare an intimate mixture of 
the drug with an excess of each individual excipient and hold at 60DC for 
72 hr in a glass container. At the end of this period, analyze for the 
drug using a stability-indicating assay. The methods of accelerated testing 
of pharmaceutical products have been extensively reviewed by Lachman et 
al in The Theory and Practice of Industrial Pharmacy, 3rd Ed , , Lea and 
Febiger (1986).

134 Bandelin 
Table Suggested Excipient {Drug Ratio in Compatibility Studies 
Weight excipient per unit weight drug 
(anticipated drug dose, mg) 
Excipient 1 5-10 25-50 75-150 150 
Alginic acid 24 24 9 9 9 
Avicel 24 9 9 9 4 
Cornstarch 24 9 4 2 2 
Dicalcium phosphate 34 34 9 9 9 
Lactose 34 9 4 2 1 
Magnesium carbonate 24 24 9 9 4 
Magnesium stearate 1 1 1 1 1 
Mannitol 24 9 4 2 1 
Methocel 2 2 2 2 1 
PEG 4000 9 9 4 4 2 
PVP 4 4 2 1 1 
Sta-Rxa 1 1 1 1 1 
Stearic acid 1 1 1 1 1 
Talc 1 1 1 1 1 
aNow called starch 1500. 
Source: Modified from Akers, M. J., Can. J. Pharm. Sci., 11: 1 (1976). 
Reproduced with permission of the Canadian Pharmaceutical Association. 
The suggested ratio of excipient to drug is given in Table 1. 
are specified according to the function they perform in the tablet. 
are classified as follows: 
Fillers (diluents) 
These additives are discussed in detail later in this chapter. 

Compressed Tablets by Wet Gr-anulation 
A. Tablet Presses 
The basic unit of any tablet press is a set of tooling consisting of two 
punches and a die (Fig. 1) which is called a station. The die determines 
the diameter or shape of the tablet; the punches. upper and lower. come 
together in the die that contains the tablet formulation to form a tablet. 
There are two types of presses: single-punch and rotary punch. The 
single-punch press has a single station of one die and two punches. and 
is capable of producing from 40 to 120 tablets per minute depending on 
the size of the tablet. It is largely used in the early stages of tablet formulation 
development. The rotary press has a multiplicity of stations arranged 
on a rotating table (Fig. 2) in which the dies are fed the formulation producing 
tablets at production rates of' from a few to many thousands per 
minute. There are numerous models of presses. manufactured by anumber 
of companies, ranging in size, speed. and capacity. 
Figure 1 Two punches and die, comprises one station. (Courtesy of 
Pennsalt Chemical Corporation, Warminster. Pennsylvania.)

136 Bandelin 
Tablet presses consist of: 
1. Hoppers. usually one or two, for storing and feeding the formulation 
to be pressed 
2. Feed frame(s) for distributing the formulation to the dies 
3. Dies for controlling the size and shape of the tablet 
4. Punches for compacting the formulation into tablets 
5. Cams (on rotary presses) that act as tracks to guide the moving 
All other parts of the press are designed to control the operation of the 
above parts. 
B. Unit Operations 
There are three methods of preparing tablet granulations. These are (a) 
wet granulation, (b) dry granulation (also called "slugging"), and direct 
compression (Table 2). Each of these methods has its advantages and disadvantages. 
The first two steps of milling and mixing of the ingredients of the formulation 
are identical, but thereafter the processes differ , Each individual 
operation of the process is known as a unit operation. The progress or 
flow of materials through the process is shown in the schematic drawing 
(Fig. 3). 
Figure 2 Punches and dies on rotary tablet press. (Courtesy of Pennwalt 
Chemical Corporation, Warminister, Pennsyovania.)

Wet granulation Dry granulation 
Table 2 Steps in Different Methods of Tablet Manufacture (Unit Operations) 
1- Milling of drugs and 
2. Mixing of milled powders 
3. Preparation of binder 
4. Mixing binder solution 
with powder mixture to 
form wet mass 
5. Coarse screening of wet 
mass using 6- to 12- mesh 
6. Drying moist granules 
7. Screening dry granules 
with lubricant and 
8. Mixing screened granules 
with lubricant and 
9. Tablet compression 
1. Milling of drugs and 
2. Mixing of milled powders 
3. Compression into large, hard 
tablets called slugs 
4. Screening of slugs 
5. Mixing with lubricant and 
disintegrating agent 
6. Tablet compression 
Direct compression 
1. Milling of drugs and 
2. Mixing of ingredients 
3. Tablet compression 
s:: (\) ..... 
(\) ..... 
S..... ..... c

( b) 
Figure 3 Unit operations in three methods of tablet manufacture: (a) wet 
granulation, (b) dry granulation, and (c) direct compression.

Compressed Tablets by Wet Granulation 
( c) 
Figure 3 (Continued) 
This chapter is devoted to the first of these processes-the wet granulation 
The preliminary step of particle size reduction can be accomplished by 
a variety of mills or grinders such as shown in Figure 4. The next step 
is powder blending with a planetary mixer (Fig. 5) or a twin-shell blender 
(Fig. 6). The addition of the liquid binder to the powders to produce the 
wet mass requires equipment with a strong kneading action such as a sigma 
blade mixer (Fig. 7) or a planetary mixer mentioned above. The wet mass 
is formed into granules by forcing through a screen in an oscillating granulator 
(Fig. 8) or through a perforated steel plate in a Fitzmill (Fig. 9). 
The granules are then dried in an oven or a fluid bed dryer after which 
they are reduced in size for compressing by again screening in an oscillator 
or Fitzmill with a smaller orifice. The granulation is then transferred 
to a twin shell or other suitable mixer where the lubricant, disintegrant, 
and glidant are added and blended. The completed granulation is then 
ready for compression into tablets. 
Fluid bed dryers have been adapted to function as wet granulators as 
depicted by the schematic drawings Figs. 10 and 11. In the latter, powders 
are agglomerated in the drying chamber by spraying the liquid binder 
onto the fluidized powder causing the formation of agglomerates while the 
hot-air flow simultaneously dries the agglomerates by vaporizing the liquid 
phase. This manner of wet granulation has the advantage of reducing 
handling and contamination by dust and offers savings in both process 
time and space [1-3]. It also lends itself to automation; however, by its 
nature it has the disadvantage of being limited to a batch -type operation. 
Unlike the wet-massing method, fluidized granulation is quite sensitive to 
small variations in binder and processing. Conversion of granule preparation 
from the wet massing to the fluid bed method is not feasible without 
extensive and time-consuming reformulation [4- 8] . 
In one study it was noted that fluidized bed tablets were more friable 
than wet-massed tablets of the same tensile strength and attributes this to 
uneven distribution of the binder in the fluidized bed powders leading to 
drug-rich, friable areas on the surface and edges of the tablets causing 
breaking and chipping [91.

Figure II Tornado mill. (Courtesy of Pennwalt Chemical Corporation, 
Warminister, Pennsylvania.) 

compressed Tablets by We' Gronutalion 141 
Figure 5 Ross HDM 40 sanitary double planetary mixer. (Courtesy of 
Charles Ross 8. Son Co.. Happauge, New York.)

142 Bandelin 
Figure 6 Twin-shell blender. (Courtesy of Patterson-Kelley Company, 
East Strousberg, Pennsylvania.) 
In the past few years considerable improvements have been made in 
equipment available for fluidized bed drying. These have reduced the 
risk of channeling by better design of the fluid bed, improved design 
from a Good Manufacturing Practices viewpoint, and by means of in -place 
washing together with automatic controls. 
Several other methods of granulating not extensively used in the pharmaceutical 
industry but worthy of investigation are the following. 
Pan granulating is achieved by spraying a liquid binder onto powders 
in a rotating pan such as that used in tablet coating. The tumbling action 
of the powders in the pan produces a fluidizing effect as the binder is 
impinged on the powder particles. The liquid (water or solvent) is evaporated 
in the heated pan by a current of hot air and the vapors are carried 
off by a vacuum hood over the upper edge of the pan opening. 
Although pan granulation has found extensive application in other industries 
(e.g. agricultural chemicals). it has not found favor in the pharmaceutical 
industry. One reason may be the lack of acceptable design. 
Spray drying can serve as a granulating process. The drying process 
changes the size, shape, and bulk density of the dried product and lends 
itself to large-scale production [10]. The spherical particles produced 
usually flow better than the same product dried by other means because 
the particles are more uniform in size and shape. Spray drying can also 
be used to dry materials sensitive to heat or oxidation without degrading 
them. The liquid feed is dispersed into droplets, which are dried in seconds, 
and the product is kept cool by the vaporization of the liquid. 
Seager and others describe a process for producing a variety of drug formulations 
by spray drying [11-13]. 
Extrusion, in which the wet mass is forced through holes in a steel 
plate by a spiral screw (similar to a meat grinder), is an excellent method 
of granulating and densifying powders. It lends itself to efficient.

S .c. o
. F-I 
.~ s
a1 .... 
boO .... r.n

re 8 Oacillatin....nulator. (Courte.Y of Penns
Chemical Corpor.tion. 
Warministel', Ohio.)

Compressed Tablets by Wet Granulation 145 
Figure 9 Fitzmill. (Courtesy of The Fitzpatrick Company, Elmhurst, 
Illinois. )

146 Baruielin 
Figure 10 Fluid bed dryer. (Courtesy of Aeromatic, Inc , , South Somerville, 
New Jersey.) 
large-scale production as part of an enclosed continuous wet-granulating 
system protected from airborne contamination. 
The extruder can also act as a wet-massing mixer by providing a continuous 
flow of the binder into the screw cha.mber, allowing the spiral screw 
to act as the massing instrument as it moves the powder, infusing it with 
the liquid to form a wet mass that is then extruded to form granules. The 
extruder has the added advantage of being a small unit as compared with 
other mixers. and has a high production capacity for its size. It is easily 
cleaned and is versatile in its ability to produce granules of various size 
depending on the size of the plate openings used. 
Pellets can be prepared by spheroidization of the wet mass after extrusion 
The transfer of wet granulation technology from lab batches to production 
equipment, generally known as "scale-up," is a critical step because 
PATTERN _ _..... ~ 
Figure 11 Spray granulator. (Courtesy of Glatt Air Techniques, Inc., 
Ramsey, New Jersey.)

Compressed Tablets by Wet Granulation 147 
of the increased mass of the larger batches and different conditions in 
larger equipment. To attempt to anticipate granulation variation due to 
scale-up, intermediate pilot equipment facilitates the step-up to production 
quantities. This permits the use of various types of equipment or unit 
operations to determine which produces the best end result of the granulation 
process. Often, however, scale-up is limited to the available equipment, 
which limits. or locks in, the process. In this situation, it is incumbent 
on the formulator to utilize his or her expertise and experience 
in selecting excipients and binder which yield the best granulation and 
tablets with the equipment available [17-19]. 
Attempts to apply experimental design to scaling up the wet granulation 
process has not been rewarding so that, in practice, trial and error remains 
the most widely used procedure. 
Wet granulation research has greatly increased and expanded in the 
last decade because of the advent of new types of granulating equipment. 
Notable among these are the Lodige, Dioana , Fielder, and Baker-Perkins 
mixers. These are equipped with high-speed impellers or blades that rotate 
at speeds of 100 to 500 rpm. In addition to merely mixing the powders, 
they produce rapid and efficient wetting and densification of the powders. 
Most of these mixers are also equipped with a rotating chopper that operates 
at speeds of 1000 to 3000 rpm. This facilitates uniform wetting of the 
powders in a matter of minutes. Granule formation can be achieved by the 
controlled spraying or atomization of the binder solution onto the powders 
while mixing [20]. While these highly efficient mixers serve to optimize the 
wet granulation process, they also demand greater understanding of their 
effects on the individual fillers and binders as processed by the mixers 
Another mixer, blender, and granulator that has found application in 
the pharmaceutical industry is the Patterson-Kelley twin-shell liquid-solids 
Blender (Fig. 12). These twin-shell units are equipped with a jacket for 
Canted discs produce 
wide spray band 
---~----- Suspended 
Dispersion blades 
Size of aperture 
from a mist to droplets 
Figure 12 Twin-shell liquid-solid blender. (Courtesy of PattersonKelley 
Company, East Stroudsburg, Pennsylvania.)

148 Baruielin 
heating and cooling, a vacuum take-off, and a liquid dispersion bar 
through which a liquid binder can be added. As the blender rotates, 
liquid is sprayed into the powder charge through the rotating liquid dispersion 
bar, located concentric to the trunnion axis. The bar's dog-eared 
blades, rotating at 3300 rpm, aerates the powder to increase the speed and 
thoroughness of the blend. Granulation can be controlled by the rate of 
binder addition through the dispersion bar. After heating, the liquid of 
the binder is removed under reduced pressure. Mixing, granulating, 
heating, cooling, and removal of excess liquid are carried out in a continuous 
operation in an enclosed system, thereby protecting the contents from 
contamination and the adjacent area from contamination by the contents. 
Once the granulation process is completed, the remaining excipients can 
be added and blended by the simple rotating action of the blender. This 
unit is also known as a liquid-solids processor. 
Most powders cannot be compressed directly into tablets because (a) they 
lack the proper characteristics of binding or bonding together into a compact 
entity and (b) they do not ordinarily possess the lubricating and 
disintegrating properties required for tableting. For these reasons, drugs 
must first be pretreated, either alone or in combination with a filler, to 
form granules that lend themselves to tableting. This process is known as 
Granulation is any process of size enlargement whereby small particles 
are gathered together into larger, permanent aggregates [22] to render 
them into a free-flowing state similar to that of dry sand. 
Size enlargement, also called agglomeration, is accomplished by some 
method of agitation in mixing equipment or by compaction, extrusions or 
globulation as described in the previous section on unit operations [4,23, 
24] . 
The reasons for granulation as listed by Record [23] are to: 
1. Render the material free flowing 
2. Densify materials 
3. Prepare uniform mixtures that do not separate 
4. Improve the compression characteristics of the drug 
5. Control the rate of drug release 
6. Facilitate metering or volume dispensing 
7. Reduce dust 
8. Improve the appearance of the tablet 
Because of the many possible approaches to granulation, selection of 
a method is of prime importance to the formulator. 
A. Wet Granulation 
Wet granulation is the process in which a liquid is added to a powder in a 
vessel equipped with any type of agitation that will produce agglomeration 
or granules. This process has been extensively reviewed by Record [23], 
Kristensen and Schaefer [26], and Capes [27].

Compressed Tablets by Wet Granulation 149 
It is the oldest and most conventional method of making tablets. Although 
it is the most labor-intensive and most expensive of the available 
methods, it persists because of its versatility. The possibility of moistening 
powders with a variety of liquids, which can also act as carriers for 
certain ingredients, thereby enhancing the granulation characteristics, 
has many advantages. Granulation by dry compaction has many limitations. 
It does not lend itself to all tablet formulations because it depends on the 
bonding properties of dry powders added as a carrier to the drug thereby 
increasing the size of the tablet. In wet granulation, the bonding properties 
of the liquid binders available is usually sufficient to produce bonding 
with a minimum of additives. 
The phenomena of adhesion and cohesion may be defined as follows: 
adhesion is the bonding of unlike materials, while cohesion is that of like 
materials. Rumpf [28] identified mechanisms by which mechanical links are 
formed between particles. The following are involved in the bonding 
1. Formation of crystalline bridges by binders during drying 
2. Structures formed by the hardening of binders in drying 
3. Crushing and bonding of particles during compaction 
Wet granulation is a versatile process and its application in tablet formulation 
is unlimited. 
B. Advantages of Wet Granulation 
1. The cohesiveness and compressibility of powders is improved due 
to the added binder that coats the individual powder particles, 
causing them to adhere to each other so they can be formed into 
agglomerates called granules. By this method, properties of the 
formulation components are modified to overcome their tableting 
deficiencies. During the compaction process, granules are fractured 
exposing fresh powder surfaces. which also improves their compressibility. 
Lower pressures are therefore needed to compress tablets 
resulting in improvements in tooling life and decreased machine 
2. Drugs having a high dosage and poor flow and/or compressibility 
must be granulated by the wet method to obtain suitable flow and 
cohesion for compression. In this case, the proportion of the 
binder required to impart adequate compressibility and flow is 
much less than that of the dry binder needed to produce a tabletby-
direct compression. 
3. Good distribution and uniform content for soluble, low-dosage 
drugs and color additives are obtained if these are dissolved in 
the binder solution. This represents a distinct advantage over 
direct compression where the content uniformity of drugs and uniform 
color dispersion can be a problem. 
4. A wide variety of powders can be processed together in a single 
batch and in so doing, their individual physical characteristics are 
altered to facilitate tableting. 
5. Bulky and dusty powders can be handled without producing a 
great deal of dust and airborne contamination.

150 Bandelin 
6. Wet granulation prevents segregation of components of a homogeneous 
powder mixture during processing, transfering, and handling. 
In effect, the composition of each granule becomes fixed and 
remains the same as that of the powder mixture at the time of the 
7. The dissolution rate of an insoluble drug may be improved by wet 
granulation with the proper choice of solvent and binder. 
8. Controlled release dosage forms can be accomplished by the selection 
of a suitable binder and solvent. 
C. Limitations of Wet Granulation 
The greatest disadvantage of wet granulation is its cost because of the 
space. time, and equipment involved. The process is labor-intensive as 
indicated by the following. 
1. Because of the large number of processing steps, it requires a 
large area with temperature and humidity control. 
2. It requires a number of pieces of expensive equipment. 
3. It is time consuming, especially the wetting and drying steps. 
4. There is a possibility of material loss during processing due to 
the transfer of material from one unit operation to another. 
5. There is a greater possibility of cross-contamination than with the 
direct-compression method. 
6. It presents material transfer problems involving the processing 
of sticky masses. 
7. It can slow the dissolution of drugs from inside granules after tablet 
disintegration if not properly formulated and processed. 
A recent innovation in wet granulating, which reduces the time and 
energy requirements by eliminating the drying step, is the melt process. 
This method relies on the use of solids having a low softening or melting 
point which, when mixed with a powder formulation and heated, liquefy to 
act as binders [29,30]. Upon cooling. the mixture forms a solid mass in 
which the powders are bound together by the binder returning to the solid 
state. The mass is then broken and reduced to granules and compressed 
into tablets. Materials used as binders are polyethylene glycol 4000 and 
polyethylene glycol 6000 [31- 33], stearic acid [30], and various waxes 
[34,35] . 
The amount of binder required is greater than for conventional liquid 
binders (i .e . 20 to 30% of the starting material). 
Another advantage of the method is that the waxy materials also act as 
lubricants, although in some cases additional lubricant is required. 
A new variation of the granulating process known as "motature-acttvated 
dry granulation" [36] combines the efficiency of dry blending with the advantages 
of wet granulation. As little as 3% water produces agglomeration. 
The process requires no drying step because any free water is absorbed 
by the excipients used. After granulation, disintegrant and lubricant are 
added and the granulation is ready for compression. 
The complex nature of wet granulation is still not well understood, 
which accounts for the continuing interest in research on the process. 
One significant problem is the degree of wetting or massing of the powders. 
Wetting plays an exceedingly important roll in the compression characteristics

Compressed Tablets by Wet Granulation 151 
of the granules, and also in the rate of drug release from the final tablet. 
Some attempts at standardizing the wetting process have been made, particularly 
in the matter of overwetting [37 - 39]. Factors that affect wetting 
1. Solubility of the powders 
2. Relative size and shape of the powder particles 
3. Degree of fineness 
4. Viscosity of the liquid binder 
5. Type of agitation 
Although the wet granulation method is labor-intensive and time consuming, 
requiring a number of steps, it continues to find extensive application 
for a number of reasons. One reason is because of its universal 
use in the past, the method persists with established products where, 
for one reason or another, it cannot be replaced by direct compression. 
Although a number of these products might lend themselves to the directcompression 
method, to do so would require a change in ingredients to 
other excipients. A change of this nature would be considered a major 
modification requiring a careful review to evaluate the need to carry out 
additional studies or product stability, safety, efficacy, and bioavailability 
as well as the impact of pertinent practical and regulatory considerations. 
Since extensive data are likely to have been accumulated on the existing 
product(s), there is understandable reluctance on the part of the drug industry 
to undertake such changes unless dictated by compelling reasons. 
Another reason is that formulators prefer to use the wet granulation method 
to assure content uniformity of tablets where small doses of drug(s) andl 
or color additives are being dispersed by dissolving in the liquid binder. 
This procedure affords better and more uniform distribution of the dissolved 
material. The method is also singular for use in the granulation of 
drugs having a high dosage where direct compression, because of the necessity 
to add a considerable amount of filler to facilitate compaction, becomes 
unfeasible because of the resulting increase in tablet size. 
Excipients are inert substances used as diluents or vehicles for a drug. In 
the pharmaceutical industry it is a catch-all term which includes various subgroups 
comprising diluents or fillers, binders or adhesives, disintegrants, 
lubricants, glidants or flow promoters, colors, flavors, fragrances, and 
sweeteners. All of these must meet certain criteria as follows: 
1. They must be physiologically inert. 
2. They must be acceptable to regulatory agencies. 
3. They must be physically and chemically stable. 
4. They must be free of any bacteria considered to be pathogenic or 
otherwise objectionable. 
5. They must not interfere with the bioavailability of the drug. 
6. They must be commercially available in form and purity commensurate 
to pharmaceutical standards. 
7. For drug products that are classified as food. such as vitamins, 
other dietary aids, and so on, the excipients must be approved 
as food additives.

152 Bandelin 
8. Cost must be relatively inexpensive. 
9. They must conform to all current regulatory requirements. 
Certain chemical incompatibilities have been reported in which the filler 
interfered with the bioavailability of the drug as in the case of calcium 
phosphate and tetracycline [40] and the reaction of certain amine bases 
with lactose in the presence of magnesium stearate [41,42]. 
To assure that no excipient interferes with the utilization of the drug, 
the formulator must carefully and critically evaluate combinations of the 
drug with each of the contemplated excipients and must ascertain compliance 
of each ingredient with existing standards and regulations. 
Two comprehensive publications cataloging the various excipients used 
in the pharmaceutical industry are available. The first of these, published 
in German in 1974 by the combined Swiss Pharmaceutical firms of Ciba 
Geigy, Hoffman LaRoche, and Sandoz, and entitled Katalog Pharmaceutischer 
Hillstoff contains specifications, tests, and a listing of suppliers. More 
recently, the listing by the Academy of Pharmaceutical Science of the 
American Pharmaceutical Association entitled Handbook of Pharmaceutical 
Bxcipierits was published. 
The screening of drug-excipient and excipient-excipient interactions 
should be carried out routinely in preformulation studies. Determination 
of the optimum drug-excipient compatibility has been adequately presented 
in the literature [43- 45] . 
A. Fillers (Diluents) 
Tablet fillers or diluents comprise a heterogeneous group of substances 
that are listed in Table 3. Since they often comprise the bulk of the tablet, 
selection of a candidate from this group as a carrier for a drug is of 
prime importance. Since combinations are also a possibility, consideration 
should be given to possible mixtures. 
Calcium sulfate, dihydrate, also known as terra alba or as snow -white 
filler, is an insoluble, nonhygroscopic, mildly abrasive powder. Better 
grades are white, others may be greyish white or yellowish white. It is 
the least expensive tablet filler and can be used for a wide variety of 
Table 3 Tablet Fillers 
Insoluble Soluble 
Calcium sulfate, dihydrate 
Calcium phosphate, dibasic 
Calcium phosphate, tribasic 
Calcium carbonate 
Modified starches 
(carboxymethyl starch, etc.) 
Microcrystalline cellulose 

Compressed Tablets by Wet Granulation 153 
acidic, neutral, and basic drugs. It has a high degree of absorptive 
capacity for oils and has few incompatibilities. Suggested binders are 
polymers such as PVP and methylcellulose, and also starch paste. See 
Example 1 for a typical formulation. 
Determination of final tablet weight: Since the amount of starch added 
as starch paste in the massing procedure was not known, it is necessary 
to determine the amount added to find the tablet weight for pressing. One 
method of doing this is to weigh the completed granulation before pressing 
and determine the tablet weight as follows: 
Weight of completed granulation == tablet weight 
Theoretical number of tablets 
Calcium phosphate, dibasic is insoluble in water, slightly soluble in 
dilute acids, and is a nonhygroscopic, neutral, mildly abrasive. fine white 
powder. It produces a hard tablet requiring a good disintegrant and an 
effective lubricant. Its properties are similar to those of calcium sulfate. 
but it is more expensive than calcium sulfate and is used to a limited extent 
in wet granulation. If inorganic acetate salts are present in the formulation, 
the tablets are likely to develop an acetic odor on aging. It can 
be used with salts of most organic bases, such as antihistamines. and with 
both water- and oil-soluble vitamines. Best binders are starch paste, PVP, 
methylcellulose, or microcrystalline cellulose. See Example 2. 
Tricolcium phosphate Is an insoluble. slightly alkaline, nonhygroscopic, 
abrasive, fine white powder. It is used to a limited extent in wet granulation 
, It should not be used with strong acidic salts of weak organic bases 
or in the presence of acetate salts. It should not be used with the watersoluble 
B vitamines or with certain esters such as vitamin E or vitamin A 
acetate or palmitate. 
Calcium carbonate is a dense, fine, white, insoluble powder. It is 
available in degrees of fineness. Precipitated calcium carbonate of a very 
fine particle size is used as a tablet filler. It is inexpensive, very white, 
nonhygroscopic. and inert. It cannot be used with acid salts or with 
acidic compounds. Its main drawback, when used as a filler, is that when 
granulated with aqueous solutions. care must be taken not to overwet by 
adding too much granulating liquid or overmixing because this produces a 
sticky, adhesive mass that is difficult to granulate, and tends to form hard 

granules that do not disintegrate readily. For this reason, it is best used 
in combination with another diluent such as starch or microcrystalline cellulose. 
Calcium carbonate. in common with calcium phosphates, can serve as a 
dietary source of calcium. It also serves as an antacid in many products. 
A tablet with unique mouth-feel and a sweet, cooling sensation. See Example 
Microcrystalline cellulose (Avicel) is a white, insoluble , nonreactive, 
free-flowing, versatile filler. It produces hard tablets with low -pressure 
compression on the tablet press. It produces rapid, even wetting by its 
wicking action. thereby distributing the granulating fluid throughout the 
powder bed. It acts as an auxiliary wet binder promoting hard granules 
with less fines. It lessens screen blocking and promotes rapid, uniform 
drying. It promotes dye and drug distribution thus promoting uniform 
color dispersion without mottling. Microcrystalline cellulose also serves as a 
disintegrant, Iubrlcant , and glidant. It has an extremely low coefficient

154 Baruieliri 
Example 1: Phenylpropanolamine Hydrochloride Tablets 
Phenylpropanolamine hydrochloride 
Calcium sulfate, dihyd rate 
10% Starch paste* 
Starch 1500 (StaRx) (disintegrant) 
Magnesium stearate (lubricant) 
Quantity Quantity per 
per tablet 10,000 tablets 
(mg) (g) 
60 600 
180 1800 
q.s. q .s . 
12 120 
6 60 
*Starch paste is made by mixing 10% starch with cold water and heating 
to boiling with constant stirring and until a thick, translucent white 
paste is formed. 
Mix the phenylpropanolamine hydrochloride with the calcium sulfate in 
a sigma blade mixer for 15 min, then add sufficient starch paste to form 
a wet mass of suitable consistency. Allow to mix for 30 min. Pass the 
wet mass through a no. 14 screen and distribute on drying trays. Dry 
in a forced-air oven at 120 to 130F or in a fluid bed dryer. When dry, 
screen through a no. 18 mesh screen, place in a twin-shell blender, add 
the starch 1500 starch and the magnesium stearate, blend for 6 to 8 min, 
and compress the completed granulation on a tablet press using 3/8-in. 
standard cup punches. 
Example 2: Diphenylhydramine (Benadryl) Tablets 
Quantity Quantity per 
per tablet 10,000 tablets 
Ingredients (mg) (g) 
Diphenhydramine hydrochloride 25 250 
Calcium phosphate, dibasic 150 1500 
Starch 1500 (StaRx) 20 200 
10% PVP in 50% alcohol q.s. q.s. 
Stearic acid, fine powder 75 75 
Microcrystalline cellulose 25 250 
Mix the diphenylhydramine hydrochloride, calcium phosphate, dibasic, 
and the starch in a planetary mixer. Moisten the mixture with the polyvinylpyrrolidone 
solution and granulate by passing through a 14-mesh 
screen. Dry the resulting granules in an oven or fluid bed dryer at 
120 to 130F. Reduce the size of the granules by passing through a

155 Compressed Tablets by Wet Granulation 
no. 20 mesh screen and dry. Add the stearic acid after passing 
through a 30-mesh screen and the microcrystalline cellulose in a twinshell 
blender for 5 to 7 min. Compress to weight using 5/16-in. standard 
concave punches. 
Important: In all formulations where an indeterminate amount of granulating 
agent is added, weigh the dried granulation after all other ingredients 
(e. g., lubricant, disintegrant, etc.}, which were not part of 
the wet granulation, and calculate the weight for compression of the 
tablet as illustrated in Example 1. 
Example 3: Calcium Carbonate-Glycine Tablets 
Quantity Quanity per 
per tablet 10,000 tablets 
Ingredients (mg) (g) 
Calcium carbonate, precipitated 400 4000 
Glycine (aminoacetic acid) 200 2000 
10% starch paste q.s. q .s . 
Light mineral oil (50 to 60 SUS) 6.5 65 
Mix the calcium carbonate and the glycine in a sigma blade or planetary 
mixer for 10 min. Add the starch paste with constant mixing until sufficiently 
moistened to granulate. 
Important: Powders are considered to be sufficiently moistened to granulate 
when a handful of the wet mass can be squeezed into a solid, 
hand-formed mass that can be broken in half with a clean fracture 
while the two halves retain their shape. (This method of determining 
when powders are adequately moistened to granulate holds true for most 
wet granulations.) Then force the wet mass through a no. 12 screen 
and dry the resulting granulation in a forced-air oven at 130 to 140F 
or in a fluid bed dryer. Size the granules by passing thorugh a no. 
12 mesh screen. Reduce the particle size by forcing through a no. 18 
mesh screen. Using a 30-mesh screen, separate out all particles passing 
through the screen. Finally, add the light mineral oil in a tumble 
mixer. Mix for 8 min and compress to weight with 7!16-in. punches 
and dies. 
of friction, both static and dynamic, so that it has little lubricant requirement 
itself. However, when more than 20% of drug or other excipient is 
added, lubrication is necessary. It can be advantageously combined with 
other fillers such as lactose, mannitol, starch, or calcium sulfate. In 
granulating, it makes the consistency of the wet mass less sensitive to 
variations in water content and overworking. This is particularly useful 
with materials which, when overwet or overmixed, become claylike, forming 
a mass that clogs the screens during the granulating process. When dried, 
these granules become hard and resistent to disintegration. Materials that

156 Bandelin 
Example 4: Calcium Carbonate and Water Only 
I ngredient Quantity 
Calcium carbonate 
Example 5; Calcium Carbonate Plus 
Microcrystalline Cellulose and Water 
Calcium carbonate 
Avicel PH-l0l 
1000 9 
300 ml 
1000 9 
100 9 
300 ml 
cause this problem are clays such as kaolin and certain other materials 
such as calcium carbonate. This is illustrated by Examples 4 and 5. 
The material of Example 4 produces a sticky mass, which is difficult to 
granUlate, whereas that of Example 5 produces a nonsticky mass, which 
can be granulated through a no. 12 screen. 
Microcrystalline cellulose added to a wet granulation improves bonding 
on compression and reduces capping and friability of the tablet. 
For drugs having a relatively small dose, microcrystalline cellulose used 
as a filler acts also as an auxiliary binder, controls water-soluble drug 
content uniformly. prevents migration of water-soluble dyes, and promotes 
rapid and uniform evaporation of liquid from the wet granulation. 
Although the usual method of making wet granulations is a two-step 
procedure, Avicel granulations can be prepared by a one-step procedure. 
In the two-step procedure, the drug and fillers are formed into granules 
by wetting in the presence of a binder, drying the resulting moist mass, 
and passing through a screen or mill to produce the desired granule size. 
These granUles are then blended with a disintegrant and lubricant, and, 
if necessary, a glidant as in the following formulation (Example 6). 
In the one-step method, the lubricant is included in the wet granulation 
contrary to what is USUally taught concerning the necessity for small 
particle size of these substances in order to coat the granules to obtain easy 
die release. Apparently, in the comminution of the granulation, sufficient lubricant 
becomes exposed to perform its intended function (Example 7). 
The quantities used in the one-step formulations are the same as those 
used in the two-step formulations. This method eliminates the usual mixing 
step for incorporating lubricants. It is also a good idea to incorporate a 
disintegrant in the wet granulation so that the granules will also disintegrate 
readily when the tablet breaks up. The practice is valid and can be widely 
used with modifications in one-step formulations. The materials and

Compressed Tablets by Wet Granulation 
Example 6: Two-Step Avicel Granulation 
I ngred ients Percent 
Drug q.s. 
Avicel PH-l01 q ,s , 
Confectioners sugar 2.5 
Starch 1500 5.0 
Starch paste, 10% q ,s , 
Talc 3.0 
Magnesium stearate 0.5 
Sodium lauryl sulfate 1.0 
Note: The amount of Avicel is replaced 
by the amount of the drug. 
Blend the first four ingredients and pass 
through a no. 1 perforated plate (round 
hole) in a Fitzrnlll , hammers foreward. 
Add the starch paste to the powder to 
form a uniform wet mass. Dry at 140F. 
Reduce the granule size by passing 
through a 20-mesh wire screen in a 
Fitzmill with knives foreward , medium 
speed. Transfer the dry granules to a 
twin-shell blender, add the last three 
ingredients, blend, and compress into 
tablets at the predetermined weight. 
quantities used in the one-step method are essentially the same as those 
in the two-step method. 
Example 7 illustrates the one-step method. 
In the above formulation, if the amount of the drug is less than 10% of 
the total tablet weight, up to 30% of the Avicel may be replaced with calcium 
sulfate dihydrate. 
Avicel PH-lOl mixed with starch and cooked until the starch forms a 
thick paste makes an excellent wet granulating mixture. Using 60% Avicel 
and 40% starch as a 10% paste makes the wet mass easier to push through 
a screen, forms finer granulations and harder granules on drying with 
fewer fines than with starch paste alone. 
Lactose, also known as milk sugar, is the oldest and traditionally the 
most widely used filler in the history of tablet making. In recent years, 
however, with new technology and new candidates, other materials have 
largely replaced it. Its solubility and sweetening power is somewhat less 
than that obtained with other sugars. It is obtained by crystallization 
from whey, a milk byproduct of cheese manufacture. Chemically, lactose 
exists in two isomeric forms, ex and 13. In solution, it tends to exist in 
equilibrium between the two forms. If it is crystallized at a temperature

158 Baruielin. 
Example 7: One-Step Avicel Granulation 
Ingredients Percent 
Drug q.s 
Avice! PH-101 
Confectioners sugar 
Starch 1500 
Polyethylene glycol 6000 
Magnesium lauryl sulfate 
50% alcohol 
q .s , 
In a planetary mixer, blend all of the 
ingredients except the polyethylene glycol 
6000 and the hydroalcoholic solution. Dissolve 
1 part of polyethylene glycol 6000 
in 1 part (w Iv) of the 50% alcohol by 
heating to SOoC. Add this solution to the 
blended powders with constant mixing in 
a sigma blade mixer until uniformly moist. 
Spread the wet mass On trays and dry 
in an oven at SOC. Pass the dry mass 
through a no. 2 perforated plate in a 
Fitzmill, knives foreward. Compress 
to predetermined size and weight. The 
use of alcohol is not essential, but it 
gives better control of wetting the powders 
and promotes more rapid drying. 
over 93C., S-lactose is produced that contains no water of crystallization 
(it is anhydrous). At lower temperatures, a-lactose monohydrate (hydrous) 
is obtained. 
a-Lactose monohydrate is commercially available in a range of particle 
sizes from 200- to 450-mesh impalpable powder. The spray-dried form is 
used for the direct-compression method of producing tablets. Lactose is a 
reducing sugar and will react with amines to produce the typical Maillard 
browning reaction. It will also turn brown in the presence of highly alkaline 
compounds. Lactose is also incompatible with ascorbic acid, salicylamide, 
pyrilamine maleate, and phenylephrine hydrochloride [46}. Nevertheless, 
it has a place in tableting by the wet granulation method in the 
sense that on wetting some goes into solution thereby coating the drug and 
offering an amount of protection and slow release where rapid dissolution 
is not req uired . 
Sucrose can be used as both a filler and as a binder in solution. It 
is commercially available in several forms: granular (table sugar), fine

Compressed Tablets by Wet Granulation 
Example 8: Vitamin B12 Tablets 
(1) Vitamin B12 (cyanocobalamin, USP) 
(2) Lactose, anhydrous, fine powder 
(3) 10% Gelatin solution 
( 4) Hyd rogenated vegeta bl e 0 iI (S terotex) 
per tablet 
55 ].1g* 
150 mg 
q .s , 
5 mg 
10,000 per 
0.55 g 
1500 g 
q .s . 
50 g 
*lncludes 10% manufacturing overage. 
Dissolve the vitamin 8 12 in a portion of the gelatin solution. Slowly add 
this to the lactose in a sIgma blade mixer with constant mixing. Add sufficient 
additional gelatin solution to form a wet mass suitable to granulate. 
Pass through a no. 14 mesh screen and dry in a suitable dryer. Reduce 
the granule size by passing through a no. 20 mesh screen. Add the 
Sterotex to the granules in a twin-shell blender and blend for 5 min. Compress 
using 1/2-in. punches and dies. This procedure forms hard tablets 
that do not disintegrate readily but dissolve rather slowly. 
granular. fine. superfine, and confectioners sugar. T he latter is the most 
commonly used in wet granulation formulations and contains 3% cornstarch 
to prevent caking. It is very fine, 80% passing through a 325-mesh screen. 
When used alone as a filler, sucrose forms hard granulations and tablets 
tend to dissolve rather than disintegrate. For this reason, it is often used 
in combination with various other insoluble fillers. It is used in chewable 
tablet s to impart sweetness and as a binder to impart hardness. In this 
role it may be used dry or in solution. When used as a dry filler, it is 
usually granulated with water only or with a hydroalcoholic binder. Various 
tablet hardnesses can be obtained depending on the amount of binder used 
to granulate. The more binder, the harder the granulation and the tablet. 
If a mixture of water and alcohol is used, softer granules are produced. 
Sucrose has several disadvantages as a filler. Tablets made with a 
major portion of it in the formulation tend to harden with time. It is not 
a reducing sugar but with alkaline materials, it turns brown with time. 
It is somewhat hygroscopic and tends to cake on standing. 
Dextrose has found some limited use in wet granulation as a filler and 
binder. It can be used essentially in the same way as sucrose. Like sucrose, 
it tends to form hard tablets, especially if anhydrous dextrose is 
used. It has the same disadvantages of both lactose and sucrose in that 
it turns brown with alkaline materials and reacts with amines to discolor. 
Mannitol is a desirable filler in tablets when taste is a factor as in 
chewable tablets. It is a white, odorless, pleasant-tasting crystalline powder 
that is essentially inert and nonhygroscopic. It is preferred as a 
diluent in chewable tablets because of its pleasant, slightly sweet taste and 
its smooth, cool, melt-down mouth-feel. Its negative heat of solution is

160 Bandelin 
responsible for its cool taste sensation. Mannitol may be granulated with a 
variety of granulating agents but requires more of the solution than either 
sucrose or lactose and approximately the same as dextrose. The moisture 
content of these granulations after overnight drying at 140 to 150F for 
sucrose, dextrose, and mannitol was less than 0.2%, except for dextrose 
granulations made with 10% gelation and 50% glucose, in which case the 
moisture content was 1.15 and 0.2%, respectively. In all lactose granulations, 
the moisture content was between 4 and 5%. Mannitol and sucrose were the 
lowest, having about the same moisture content. It was found. however, 
that mannitol, although requiring more granulating solution, generally gave 
a softer granulation than either sucrose or dextrose. 
B. Binders 
Binders are the "glue" that holds powders together to form granules. They 
are the adhesives that are added to tablet formulations to provide the cohesiveness 
required for the bonding together of the granules under compaction 
to form a tablet. The quantity used and the method of application 
must be carefully regulated, since the tablet must remain intact until swallowed 
and must then release its medicament. 
The appearance, elegance, and ease of compression of tablets are dir-
eotly related to the granulation from which the tablets are compressed. 
Granulations, in turn, are dependent on the materials used, processing 
techniques, and equipment for the quality of the gr-anulation produced. Of 
these variables, none is more critical than the binder used to form the granulation, 
for it is largely the binder that is fundamental to the granulation 
particle size uniformity, adequate hardness, ease of compression, and general 
quality of the tablet [47- 50] . 
Binders are either sugars or polymeric materials. The latter fall into 
two classes: (a) natural polymers such as starches or gums including 
acacia, tragacanth, and gelatin. and (b) synthetic polymers such as polyvinylpyrrolidone, 
methyl- and ethylcellulose and hydroxypropylcellulose. 
Binders of both types may be added to the powder mix and the mixture 
wetted with water, alcohol-water mixtures, or a solvent, or the binder may 
be put into solution in the water or solvent and added to the powder. The 
latter method, using a solution of the binder, requires much less binding 
material to achieve the same hardness than if added dry. In come cases, 
it is not possible to get granules of sufficient hardness using the dry method. 
In practice, solutions of binders are usually used in tablet production. 
Reviews of binders and their effects are available [23,26,51,52]. A guide 
to the amount of binder solution required by 3000 g of filler is presented 
in Table 4. 
A study on the addition of a plasticizer to the binder solution on the 
tableting properties of dicalcium phosphate, lactose, and paracetamol 
(acetaminophen) indicated that it improved the wet-massing properties of 
the granulation. Including a placticizer in the binder increased the tensile 
strength, raised the capping pressure. and reduced the friability of all 
the tablets. The plasticizers used in this study were propylene glycol, 
polyethylene glycol 400, glycerine, and hexylene glycol [531. 
A list of commonly used binders is given in Table 5. These are treated 
in detail as discussed in the following paragraphs.

Compressed Tablets by Wet Granulation 
Table 4 Granulating Solution Required by 3000 g of Filler 
Volume of Filler 
granulating solution 
required (ml) Sucrose Lactose Dextrose 
10% Gelatin 200 290 500 
50% Glucose 300 325 500 
2% Methylcellulose 290 400 835 
(400 cps) 
Water 300 400 660 
10% Acacia 220 400 685 
10% Starch paste 285 460 660 
50% Alcohol 460 700 1000 
10% PVpa in water 260b 340b 470b 
10% PVpa in alcohol 780b 650b 825b 
10% Sorbitol in water 280b 440b 750b 
bDerived by the author, not from source noted below. 
Source: Taken in part from the Technical Bulletin, Atlas Mannitol, leI 
Americas, Wilmington, Delaware, 1969. 
Starch in the form of starch paste has historically been, and remains, one 
of the most used binders. Aqueous pastes usually employed range from 5 
to 10% in concentration. Starch paste is made by suspending starch in 1 
to 1- 1/2 parts cold water, then adding 2 to 4 times as much boiling water 
with constant stirring. The starch swells to make a translucent paste that 
can then be diluted with cold water to the desired concentration. Starch 
paste may also be prepared by suspending the starch in cold water and 
heating to boiling in a steam -jacketed kettle with constant stirring. Starch 
paste is a versatile binder yielding tablets that disintegrate rapidly (see 
Example 9) and in which the granulation is made using starch as an internal 
binder and granulated with water only. 
An example of granulation made by massing with starch paste as an internal 
binder rather than an external binder when wetted with water only 
as in Example 9 is given in Example 10. 
Pregelatinized starch is starch that has been cooked and dried. It can 
be used in place of starch paste and offers the advantage of being soluble 
in warm water without boiling. It can also be used as a binder by adding 
it dry to the powder mix and wetting with water to granulate as indicated 
in Example 9. 
Starch 1500 is a versatile, multipurpose starch that is used as a dry 
binder, a wet binder, and a disintegrant. It contains a 20% maximum cold 
water-soluble fraction which makes it useful for wet granulation. It can be

162 Bandelin 
Table 5 Binders Commonly Used in Wet Granulation 
Binder Usual concentration 
Cornstarch. USP 
Pregelatinized cornstarch 
Starch 1500 
Gelatin (various types) 
Methylcellulose (various 
viscosity grades) 
Sodium carboxymethylcellulose 
Ethylcellulose (various 
viscosity grades) 
Polyvinyl alcohol (various 
viscosity grades) 
Polyethylenene glycol 6000 
5-10% Aqueous paste 
5-10% Aqueous solution 
5-10% Aqueous paste 
2-10% Aqueous solution 
10-85% Aqueous solution 
5-20% Aqueous solution 
5-20% Aq ueou s , alcoholic. or hydroalcoholic 
2-10% Aqueous solution 
2-10% Aqueous solution 
2-15% Alcoholic solution 
2-10% Aqueous or hydroalcoholic solution 
10-30% Aqueous, alcoholic, or hydroalcoholic 
Example 9: Aminophyll ine Tablets 
Quantity Quantity per 
per tablet 10,000 tablets 
Ingredients (mg) ~} 
Aminophylline 100 1.0 
Tricalcium phosphate 50 0.5 
Pregelatinized starch 15 0.15 
Water q.s. q .s , 
Talc 30 0.3 
Mineral oil, light 2 0.02 
Mix the aminophylline, tricalcium phosphate, and starch 
and moisten with water with constant mixing. Pass through 
a 12-mesh screen and dry at 110F. Size the dry granulation 
through a 20-mesh screen; add the talc and mix in 
a suitable mixer for 8 min. Add the mineral oil, mix for 5 
min, and compress with 5/16-in. standard concave punches.

Compressed Tablets by Wet Granulation 
Example 10: Pseudoephedrine Tablets 
Pseudoephed ri ne hyd roch lorid e 
Calcium sulfate, dihydrate 
Citric acid, fine powder 
Starch (as starch paste) 
Sterotex (hydrogenated vegetable oil) 
Alginic acid (disintegrant) 
FD&C Yellow No. 6 
per tablet 
Quantity per 
10,000 tablets 
(5 mg) 
Mix the pseudoephedrine hydrochloride, citric acid, and calcium sulfate in 
an appropriate mixer for 15 min. Dissolve the FD&C Yellow No. 6 in the 
water used to make the starch paste, or dissolve the dye in a small quantity 
of water and add to the prepared paste. Add the starch paste sufficient 
to form a suitable wet mass and granulate through a 14-mesh screen. Dry 
at 120 to 130F. Reduce the granules by passing through an 1a-mesh 
screen, add the alginic acid, mix, and compress with 5/16-in. standard 
cup punches. 
dry-blended with powder ingredients and granulated with ambient temperature 
water. The water-soluble fraction acts as an efficient binder, while 
the remaining fraction aids in the disintegration of the tablet. It also will 
not present overwetting problems as commonly experienced with pre gelatinized 
Approximately 3 to 4 times as much starch is req uired to achieve the 
same tablet hardness as with starch paste. 
Gelatin. If a still stronger binder is needed, a 2 to 10% gelatin solution 
may be used. Gelatin solutions should be made by first allowing the gelatin 
to hydrate in cold water for several hous or overnight, then heating the 
mixture to boiling. Gelatin solutions must be kept hot until they are used 
for they will gel on cooling. Although gelatin solutions have been extensively 
used in the past as a binder, they have been replaced to a large 
extent by various synthetic polymers. such as polyvinylpyrrolidone, methylcellulose 
, et c . 
Gelatin solutions tend to produce hard tablets that req uire active disintegrants. 
The solutions are generally used for compounds that are difficult 
to bind. These solutions have another disadvantage in that they 
serve as culture media for bacteria and molds and, unless a preservative is 
added, they are quickly unfit to use. 
Sucrose solutions are capable of forming hard granules. Some gradation 
of tablet hardness can be achieved by varying the concentration of sucrose 
from 20 to 85% depending on the strength of binding required. 
In ferrous sulfate tablets, sucrose acts both as a binder and to protect 
the ferrous sulfate from oxidizing.

164 Bandelin 
Example 11: Ferrous Sulfate Tablets 
per tablet 
Ingredients fmg) 
Ferrous sulfate, dried 
Corn starch 
Sucrose as a 70% wIw syrup 
Explotab (sodium carboxymethyl starch) 
Magnesium stearate 
q .s , 
Mix the ferrous sulfate and the starch; moisten with 
the sugar solution to granulate through a 14~mesh 
screen. Dry in a tray oven overnight at 130 to 140F. 
Size through an 18-mesh screen, add the Explotab, talc, 
and magnesium stearate, and compress to weight using 
3/8-in. deep-cup punches. The reason for the deep-cup 
punches is that ferrous sulfate tablets need to be coated 
and tablets prepared with deep-cup punches lend themselves 
better to the coating process in that the edges 
at the perimeter are less obtuse than the standard 
punch tablets. 
Sugar solutions are good carriers for soluble dyes, producing granulations 
and tablets of uniform color. Sugar syrups are used to granulate 
tribasic phosphate excipient, which USUally requires a binder with greater 
cohesive properties than starch paste. Some other compounds for which 
sugar is indicated include aminophylline, acetophenetidin, acetaminophen, 
and meprobamate. 
Acacia solutions have long been used in wet granulation, but now they 
have been largely replaced by more recently developed polymers such as 
polyvinylpyrrolidone and certain cellulose derivates. However, for drugs 
with a high dose and difficult to granulate , such as mephenesin, acacia is 
a suitable binder. It produces hard granules without an increase in hardness 
with time as is the case with gelatin. One disadvantage of acacia is 
that it is a natural product and is often highly contaminated with bacteria, 
making it objectionable for use in tablets. Tragacanth is another natural 
gum which, like acacia, has been used in 5 to 10% solutions as a binder. 
It does not produce granulations as hard as acacia solutions. Like acacia. 
it often has a high bacterial count. In the following formula, a soluble 
lubricant, polyethylene glycol 6000, is added to the acacia solution to assist 
both in tableting and in disintegration of the tablet (Example 12). 
Polyvinylpyrrolidone has become a versatile polymeric binder. This compound, 
first developed as a plasma substitute in World War II, is inert and 
has the advantage of being soluble both in water and in alcohol. Although 
it is slightly hygroscopic, tablets prepared with it do not. as a rule, harden 
with age, which makes it a valuable binder for chewable tablets (Example

Compressed Tablets by Wet Granulation 
Example 12: Mephenesin Tablets 
per tablet 
Ingredients (mg) 
Mephenesin 400 
Acacia, 10% aqueous solution with 
1%polyethylene glycol 6000 
Add sufficient acacia-polyethylene glycol 6000 solution 
to the mephenesin in a planetary or other suitable 
mixer to granulate; pass the wet mass through 
a 12-mesh screen and dry in an oven or other suitable 
dryer at 130 to 140F. Force the dry granules 
through a 16-mesh screen, add the talc and the 
starch in a tumble mixer. mix for 10 to 15 min. 
and compress using 1/2-in flat-face, bevel edge 
13). Generally, it is better to granulate insoluble powders with aqueous or 
hydroalcoholic solutions of PVP and to granulate soluble powders with PVP 
in alcoholic solution. Effervescent tablets comprising a mixture of sodium 
bicarbonate and citric acid can be made by wet granulation using solutions 
of PVP in anhydrous ethanol since no acid-base reaction occurs in this 
anhydrous medium. Anhydrous ethanol should always be used in this granulation 
and not anhyrous isopropanol, since the latter leaves a trace of its 
odor in the tablets no matter how, or how long, the granulation has been 
dried. A concentration of 5% PVP in anhydrous ethanol produces a granulation 
of good compressibility of fine powders of sodium bicarbonate and 
citric acid. and makes the vigorous effervescence and rapid dissolution of 
the resulting tablets. Polyvinylpyrrolidone is also an excellent binder for 
chewable tablets, especially of the aluminum hydroxide- magnesium hydroxide 
type (Example 12). The inclusion of 2 to 3% of glycerine (based on the 
final weight of the tablet) tends to reduce hardening of these tablets with 
age. It is a versatile and excellent all-purpose binder used in approximately 
the same concentration as starch, but considerably more expensive. 
Methylcellulose in aqueous solutions of 1 to 5%, depending on the viscosity 
grade, may be used to granulate both soluble and insoluble powders. 
A 5% solution produces granulations similar in hardness to 10% starch paste. 
It has the advantage of producing granulations that compress readily, producing 
tablets that generally do not harden with age. Methylcellulose is' 
a better binder for soluble excipients such as lactose, mannitol, and other 
sugars. It offers considerable latitude in binding strength because of the 
range of viscosity grades available. Low-viscosity grades, 10 to 50 cps, 
allow for higher working concentrations of granulating agent than higher 
viscosity grades, such as the 1000 to 10,000 cps grades.

166 Btuuieiiri 
Example 13: Chewable Antacid Tablets 
per tablet 
Ingredients (mg) 
Aluminum hydroxide, dried gel 
Magnesium hydroxide, fine powder 
Sugar, confectioners lOX 
Mannitol, fine powder 
Polyvinylpyrrolidone, 10% solution in 50% 
alcohol solution 
Magnesium stearate 
Cab-O-Sil M-5 
Oil of peppermint 
Mix the first four ingredients in a suitable mixer. Add 
the glycerine to the PVP solution and use to moisten 
the powder mix. Granulate by passing through a 14-mesh 
screen and dry at 140 to 150F. Mix the oil of peppermint 
with the Cab-O-Sil and the magnesium stearate, mix, and 
size through a 20-mesh screen. Mix well and compress 
using 1/2-in. flat-face, bevel edge punches. 
*10 milligrams of dry PVP may be added to the powder mix 
and granulated with 50% hydroalcoholic solution instead of 
the PVP solution. This, however, is about 3 times as 
much as is required when used in solution. 
Sodium carboxymethylcellulose (sodium CMC) in concentrations of 5 to 
15% may be used to granulate both soluble and insoluble powders. It produces 
softer granulations than PVP, and tablets have a greater tendency 
to harden. It is incompatible with magnesium, calcium, and aluminum salts, 
and this tends to limit its utility to some extent. Although producing 
softer gr-anulatlons , these generally compress well. However, tablets have 
a relatively long disintegration time. 
Ethylcellulose is insoluble in water and is used in alcohol solutions. Like 
methylcellulose, it is available in a range of viscosities, depending on the 
degree of substitution of the polymer. Low-viscosity grades are usually 
used in concentrations of 2 to 10% in ethanol. It may be used to granulate 
powders which do not readily form compressible granules, such as aceteminophen, 
caffeine, meprobamate, and ferrous fumarate (Example 14), and 
it offers a nonaqueous binder for medicaments that do not tolerate water 
(Example 15). 
Polyvinyl alcohols are water-soluble polymers available in a range of viscosities. 
As granulating agents they resemble acacia but have the advantage 
of not being heavily laden with bacteria. They are film-formers and their 
granulations are softer than those made with acacia, yielding tablets that

Compressed Tablets by Wet Granulation 
Example 14: Ferrous Fumarate Tablets 
per tablet 
Ingredients (mg) 
Ferrous fumarate, fine powder 
Ethylcellulose 50 cps, 5% in ethanol 
q.s. (approx. 10 mg) 
Slowly add the ethylcellulose solution to the ferrous fumarate in 
a double-S arm mixer with constant mixing until sufficiently 
moist to granulate. Force through a Hi-mesh screen and dry 
in a suitable dryer. Transfer the dry granulation to a tumble 
mixer, add the Stearowet and the Cab-O-Sil, mix, and compress 
using 3/8-in. standard cup punches. 
*Stearowet is a mixture of calcium stearate and sodium lauryl 
sulfate. This combination of hydrophobic and a hydrophilic 
lubricant tends to decrease the disintegration time of the tablets. 
disintegrate more readily and generally do not harden with age. Viscosities 
lending themselves to tablet granulation range from 10 to 100 cps. 
Polyethylene glycol 6000 may serve as an anhdrous granulating agent 
where water or alcohol cannot be used. Polyethylene glycol 6000 is a white 
to light yellow unctuous solid melting at 70 to 75C and solidifying at 56 to 
Example 15: Ascorbic Acid Tablets 
per tablet 
Ingredient (mg) 
Ascorbic acid, 20-mesh granules 
Ethylcellulose 50 cps, 10% in ethanol 
Explotab (sodium carboxymethyl starch) 
Calcium silicate 
q .s. (approx. 4 mg) 
In a rotating drum or coating pan add the ethylcellulose solution 
slowly to the ascorbic acid with rapid rotation of the drum. 
Dry with warm air directed into the rotating drum or pan 
equipped with an exhaust system to remove alcohol vapor. When 
dry. transfer to a tumble mixer, add the Explotab and the calcium 
silicate, mix, and compress with 13/32-in. punches.

168 Bandelin 
Example 16: Polyethylene 6000 granulation 
Ingredients per tablet 
Drug q.s. 
Filler, calcium sulfate dihydrate, or 
dlcalciurn phosphate, or lactose, or 
any other suitable filler 
Polyethylene glycol 6000 up to 30% of the 
above mixture* 
Magnesium stearate 
Aerosil 200 
q i s . 
q .s . 
Uniformly mix the drug with the filler and the polyethylene 
glycol 6000 and pass through a pulverizer 
using a no. 20 screen. Spread on trays and place in an 
OVen at 75 to 80C for 3 hr. Cool the heated mass to 
room temperature and screen through an l8-mesh screen, 
blend with the balance of the ingredients, and compress 
into tablets of proper weight. 
*Because of variation of drug and filler, the amount of 
polyethylene glycol 6000 needs to be determined on an 
experimental basis for each formula. 
A procedure for making tablets by this method has been given by Shah 
et al , [29] in which polyethylene glycol 6000 acts as the binding agent 
(Example 16). 
Another method described by Rubenstein {32] carries out the granulation 
in a coating pan modified so that the pan contents can be heated to 60C. 
The disintegrant is charged into the pan followed by 4% of polyethylene glycol 
6000 in powder form. The heated pan is then rotated to melt the polyethylene 
glycol. The drug is then added and the whole mass is tumbled and 
heated for 5 min. The molten PEG 6000 acts as a binder covering the surface 
of the powders. After thoroughly mixing, the heat is discontinued 
and the mass allowed to cool to room temperature. During the cooling period, 
the PEG 6000 solidifies coating the powders to produce granules. The resulting 
granules are free flowing but require the addition of a glidant (0.2% 
Aerosil 200) for tableting. The granules are not self-lubricating and require 
the addition of a lubricant to permit tableting. 
Sustained Release Applications 
Binders as waterproofing agents having been used to obtain sustained or 
prolonged release dosage forms. By granulating or coating powders with 
relatively insoluble or slowly soluble binders (i.e. shellac. waxes, fatty acids 
and alcohols, esters and various synthetic polymers), tablets having delayed 
or prolonged release properties have been formulated. This application is 
discussed later in this chapter.

Compressed Tablets by Wet Granulation 
C. Lubricants 
Lubricants are used in tablet formulations to ease the ejection of the tablet 
from the die, to prevent sticking of tablets to the punches, and to prevent 
excessive wear on punches and dies. They function by interposing a film 
of low shear strength at the interface between the tablet and the die wall 
and the punch face. Lubricants should be carefully selected for efficiency 
and for the properties of the tablet formulation. 
Metal stearates because of their unctuouse nature and available small 
particle size, are probably the most efficient and commonly used lubricants. 
They are generally unreactive but are slightly alkaline (except zinc), and 
have the disadvantage of retarding tablet disintegration and dissolution because 
of their hydrophobic nature [59,63,64]. Of the metal stearates, magnesium 
is the most widely used. It also serves as a glidant and antiadherent. 
Butcher and Jones [59] showed that particle size, packing density, and 
frictional shear tests are necessary to evaluate the quality and suitability 
of commercially available stearates as lubricants. 
Stearowet C, because of its surfactant component, is less likely to interfere 
with disintegration and dissolution. Sodium lauryl sulfate is an auxiliary 
lubricant as well as a surfactant. 
In instances where lubrication is a problem, an internal and an external 
lubricant can be used in conjunction with each other as given in Example 17. 
Allow the gelatin to soak in 70% of the water for several hous or overnight. 
Heat to 80F, add the polyethylene glycol 6000, stir until dissolved, 
and cool slowly to 110 to 120F. Add water, maintaining the temperature in 
the above range. The solution must be used at this temperature because it 
will gel on cooling. 
Stearic acid is a less efficient lubricant than the metal stearates. It 
melts at 69 to 70C. so that it does not melt under usual conditions of 
storage. It should not be used with alkaline salts of organic compounds 
such as sodium barbiturates, sodium saccharin, or sodium bicarbonate. With 
these compounds it has a tendency to form a gummy, sticky mass that causes 
sticking to the punches. 
Numerous studies of lubricants indicate that there is no universal lubricant 
and that the formula, method of manufacture, and the formulators knowledge 
and experience determine the choice and amount used [56- 60] . 
In selecting a lubricant, the following should be considered: 
1. Lubricants markedly reduce the bonding properties of many excipients. 
2. Overblending is one of the main causes of lubrication problems. 
Lubricants should be added last to the granulation and tumbleblended 
for not more than 10 min. 
3. The optimum amount of lubricant must be determined for each formulation. 
Excess lubricant is no more effective but rather interferes 
with both disintegration and bioavailabiltty by waterproofing the 
granules and tablet. 
4. Lubricant efficiency is a function of particle size; therefore, the 
finest grade available should be used and screened through a 100 
to 300-mesh screen before use. 
Ragnarssen et al. [61] found that a short mixing time for magnesium 
stearate in excipient blends resulted in poor distribution of the lubricant 
but did not impair the lubrtcating efficiency in tablet compression.

170 Bandelin 
Example 17: Analgesic-Decongestant Tablets 
mg per 
I ngredients tablet 
Aceta mi nophen 
Pseudoephedrine hydrochloride 
Chlorpheniramine maleate 
10% gelatin-5% polyethylene glycol 6000 
aqueous solution* 
Microcrystalline cellulose 
Starch 1500 
Stearowet C 
Cab-O-Sil (silica aerogel) 
Mix the acetaminophen, pseudoephedrine hydrochloride, 
chlorpheniramine maleate, and sucrose, and granulate 
with the gelatin -polyethylene 6000 solution, passing 
the wet mass through a 12~mesh screen. Dry at 130 
to 140F and size through an l8-mesh screen. Add 
the Cab-O-Sil, Starch 1500, and microcrystalline 
cellulose in order and blend for 15 min. Finally, add 
the Sterowet C and blend for 3 min. Compress using 
7/l6-in. standard cup punches. 
*Preparation of gelatin-polyethylene glycol 6000 solution. 
Another study [62] found that prolonged rmxing time tends to limit or 
reduce lubricant effectiveness and that glidants should be added first and 
intimately blended after which the luhricant is added and blended for a 
relatively short time. 
Insufficient lubrication causes straining of the tablet press as it labors 
to eject the tablet from the die. This may cause a characteristic screeching 
sound and straining of the press parts involved. Another indication of insufficient 
lubrication is the presence of striations or scratch marks on the 
edges of the tablets. 
Lubricants fall into two classes: water-insoluble and water-soluble. A 
listing of the hydrophobic and the soluble lubricants is given in Table 6. 
Hydrogenated vegetable oils, commercially available as Sterotex and 
Duratex, are bleached, refined. and deodorized hydrogenated vegatable oils 
of food grade. They are usually available in spray-congealed form. While 
the particle size is not as small as may be desirable, the establishment of 
appropriate blending times with specific granulations can aid in the distribution 
on the granules through attrition of the lubricant powder. These have 
special application where alkaline metal stearates cannot be used, or where 
their metallic taste may be objectional as in tablets or lozenges to be dissolved 
in the mouth. Example 18 illustrates this use.

Compressed Tablets by Wet Granulation 
Table 6 Lubricants: Typical Amounts Used 
Metal stearates, calcium, magnesium, zinc 
Stearowet C: a water-wettable mixture of calcium stearate 
and sodium lauryl sulfate 
Stearic acid, fine powder 
Hydrogenated vegetable oils (Sterotex, Duratex) 
Light mineral oil 
Sodium benzoate 
Sodium chloride 
Sodium and magnesium lauryl sulfate 
Polyethylene glycol 4000 and 6000 (Carbowax 4000 
and 6000), fine powder 
Amount used 
in granulations 
(% w/w) 
High-Melting Waxes. Numerous food grade waxes are available, and 
while these are not generally used as lubricants, they offer possibilities for 
investigation. Waxes of both mineral sources (ceresin) and vegetable sources 
(carauba) offer possibilities as Iubricants , 
Talc acts as both lubricant and glidant. It is less efficient than the previously 
mentioned products and larger quantities are required for adequate 
lubrication. It has the disadvantage of retarding disintegration. Smaller 
quantities can be used in conjunction with other lubricants. It is essential 
that talc used in tableting be asbestos-free and, to this end, each lot should 
be accompanied by a certification from the supplier to this effect. 
Starch is derived from a number of sources: corn, potatoe, rice, and 
tapioca. It may exist as dry granules, powder, swollen granules, in solution, 
and may be used as a filler, binder, disintegrant and film-former. It 
is available both as hydrophilic and hydrophobic corn starch. 
Pharmaceutically cornstarch is the item of commerce most commonly used. 
Although there are much more efficient lubricants, starch because of its 
multiple properties is often included in formulations as an auxiliary lubricant 
because of its many applications in tablet making by the wet granulation 
Mineral oil. Light mineral oil having a Saybolt viscosity of 50 to 60 SUS 
(approximately 8 centistokes) is a liquid lubricant with universal application 
because it is unreactive, odorless, tasteless, and can be easily sprayed onto

1 72 Bandelin 
Example 18: Medicated Throat Lozenges 
Ingredients per tablet 
Sucrose, fine powder (1 OX confectioners sugar 
Acacia, fine powder 
Citric acid, fine powder 
10% Gelatin solution 
Hexyl resorcinol 
Hydrogenated vegetable oil (Sterotex, Duratex) 
Ethanol 95% 
15 mg 
q .s , 
12 mg 
10 mg 
2.4 mg 
160 mg 
0.04 ml 
Mix the sucrose, acacia, and citric acid and mass with the gelation 
solution. Granulate through an 8-mesh screen and dry 
at 130 to 140F. Dissolve the met hol , benzocaine, and the 
hexylresorcinol in the ethanol and distribute on the granulation 
in a twin-shell blender. Spread on trays in an oven and remove 
alcohol with forced air at ambient temperature. Transfer the 
granulation to a tumbel blender, add the hydrogenated vegetable 
oil, blend for 5 min, and compress with 3/4-in. flat-face, bevel 
edge pu nches . 
granulations. It should be sprayed onto the formulation in a closed container. 
preferably in a twin-shell or double-cone blender equipped with a spray head 
or an intensifier bar. On compression. tablets lubricated with mineral oil 
often show mottling with oil spots on the surface of the tablet. This is more 
noticeable with colored tablets, especially dark colors. This mottling disappears 
after a day or two as the oil disperses in the tablet. One disadvantage 
of mineral oil as a lubricant is that the granulation. after the addition 
of the oil. must be compressed within 24 to 48 hr because the oil has a 
tendency to penetrate into the granules and thereby lose its effectiveness 
as a lubricant. Mineral oil is a largely neglected but excellent lubricant 
that greately reduces die wall friction and sticking to punches. 
Sodium benzoate and sodium chloride have limited application in pharmaceuticals 
but find some use in household products. Sodium benzoate is essentially 
tasteless and can be used in tablets intended to be chewed or 
allowed to dissolve in the mouth. 
Sodium and magnesium lauryl sulfate are water-soluble surfactants that 
can be used instead of the metal stearates to counteract their waterproofing 
properties as tablet lubrtcants . Studies indicate that granulations run on a 
rotary tablet press using both magnesium stearate and magnesium lauryl sulfate 
as lubr-icants , produced tablets having less variation in physical properties 
with the latter than with the former. It appears that magnesium lauryl 
sulfate is at least as efficient as magnesium stearate and has the advantage 
of reduced interference with dissolution [65.66].

Compressed Tablets by Wet Granulation 173 
Magnesium lauryl sulfate also has less taste than the sodium salt and is 
therefore better for chewable tablets. 
Polyethylene glycol 4000 and 6000, also known as Carbowax 4000 and 
5000, are water-soluble lubricants that find considerable use in tablet manufacture 
and in the formulation of chewable tablets. They are generally unreactive 
and can be used with sensitive ingredients such as aspirin, ascorbic 
acid, and other vitamins. 
As with other lubricants, the smaller the particle size, the greater 
the distribution on granules, which makes for more efficient lubrication. 
Solid polyethylene glycols in very fine powder are not commercially available; 
however, they may be micronized if cooled to -10 to - 20C. 
Polyethylene glycol 6000 can be used in aqueous, alcoholic, or hydroalcoholic 
solution with various binders thereby obtaining a binder-lubricant 
combination that can be used in wet massing. Solutions may also be spr-ayed 
or atomized onto powders in a fluidized bed granulator or in a twin-shell or 
double-cone blender equipped with a vacuum takeoff to remove solvent thus 
applying both binder and lubricant. 
Recently I two new additions to the field of lubricants have been proposed. 
These are sodium stearyl fumarate and glyceryl behenate [67]. Using magnesium 
stearate for comparison, these were added to granulations of lactose 
and salicylic acid and compressed with equivalent force on an instrumented 
tablet press. The new lubricants showed less effect on tablet strength and 
had a lesser effect on dissolution rate of the active ingredients than did 
magnesium stearate. Magnesium stearate and sodium stearyl fumarate were 
effective at 1 to 3% levels whereas glyceryl behenate required 3% for effective 
In tablet formulation, a lubricant often permits the resolution of several 
production problems that are related to compression. LUbrication facilitates 
glidancy of granules during material flow, eliminates binding in the die, and 
minimizes picking and sticking to punch-face surfaces on compression. Mixing 
time in the scale-up of tablet production is greatly influenced by the 
type of mixing equipment and by the batch size. Vigorous mixing shortens 
the time required for the distribution of the disintegrant and the batch size I 
due to the shear weight on the charge I influences the mixing time because 
of the increased flow of particles in tumble- I twin-shell, and double-conetype 
mixers. The release characteristics and performance criteria of the 
final tablet (such as physical integrity and stability) depend on lubricantexcipient 
interaction and the manner in which these materials are affected 
by mixing. 
D. Disintegrants 
Disintegrant is the term applied to various agents added to tablet granulation 
for the purpose of causing the compressed tablet to break apart (disintegrate) 
when placed in an aqueous environment. Basically I the disintegrant's 
major function is to oppose the efficiency of the tablet binder and 
the physical forces that act under compression to form the tablet. The 
stronger the binder, the more effective must be the disintegrating agent in 
order for the tablet to release its medication. Ideally I it should cause the 
tablet to disrupt, not only into the granules from which it was compressed, 
but also into the powder particles from which the granulation was prepared 
[68-71] .

174 Bandelin 
There are two methods used for incorporating disintegrating agents into 
tablets. These methods are called external addition and internal addition. 
In this, the disintegrant is added to the sized gr-anulation with mixing just 
prior to compression. In the internal addition method, the disintegrant is 
mixed with other powders before wetting the powder mixture with the granulating 
solution. Thus, the disintegrant is incorporated within the granule. 
When this method is used, part of the disintegrant can be added internally 
and part externally. This provides immediate disruption of the tablet into 
the previously compressed granules while the disintegrating agent within 
the granules produces further erosion of the granules to the original powder 
particles. Although this latter is an attractive theory, it is only 
partially effective in practice because any disintegrating agent bound within 
the granules loses some of its disruptive force due to its encasement by the 
binder. Nevertheless, the two-step method usually produces better and 
more complete disintegration than the usual method of adding the disintegrant 
to the granulation surface only. 
Disintegrants constitute a group of materials that, on contact with water, 
swell, hydrate, change in volume or form, or react chemically to produce a 
disruptive change in the tablet. This group includes various forms of 
starch, cellulose, algins, vegetable gums, clays, ion exchange reins, and 
acid-base combinations. A list of commonly used tablet disintegrants and 
the amounts usually used are given in Table 7. 
Starch is the oldest and probably the most widely used disintegrant 
used by the pharmaceutical industry. Regular cornstarch USP, however, 
has certain limitations and has been replaced to some extent by modified 
starches with specialized characteristics to serve specific functions. Starch 
1500 is a physically modified cornstarch that meets all the specifications of 
pregelatinized starch NF. It is somewhat unique in that it lends itself well 
Table 7 Disintegrants: Typical Amounts Used 
Starch USP 
Starch 1500 
Avicel PH 101, PH 102 (microcrystalline cellulose 
Solka floc (purified wood cellulose) 
Alginic acid 
Explotab (sodium starch glycolate) 
Guar gum 
Polyclar AT (polyvinylpyrrolidone, crosslinked PVP) 
Amberllte IPR 88 (ion exchange resin) 
Methylcellulose, sodium carboxymethylcellulose, hydroxypropylmethylcellulose 
in granulation 
(% w Iw) 

Compressed Tablets by Wet Granulation 175 
to conventional manufacturing techniques I especially to wet granulation. 
There are many classical theories that attempt to explain the mode of action 
of disintegrants, especially starches. One theory is that the disintegrant 
forms pathways throughout the tablet matrix that enable water to 
be drawn into the structure by capillary action, thus leading to disruption 
of the tablet. An equally popular concept relates to the swelling of starch 
grains on exposure to water, a phenomenon that physically ruptures the 
particle-particle bonding in the tablet matrix. Neither of these mechanisms 
explains the dramatic explosion that often takes place when tablets containing 
starch are exposed to water. Unique work carried out by Hess [72J 
would seem to suggest that on compression there is a significant distortion 
of the starch grains. On exposure to water, these grains attempt to recover 
their original shape, and in so doing release a certain amount of 
stress which, in effect, is responsible for the destruction of interparticulate 
hydrogen bonds and causes the tablet to be literally blown apart. Starch 
thus functions as the classical disintegrant. Starch 1500, by virtue of its 
manufacturing process, retains the disintegrant qualrties of the parent cornstarch. 
These qualities make it a versatile disintegrating agent as both an 
internal and external disintegrant in tablet formulations (Example 19). 
Avicel (microcrystalline cellulose) is a highly effective disintegrant. 
It has a fast wicking rate for water, hence, it and starch make an excellent 
combination for effective and rapid disintegration in tablet formulations. 
One drawback to its use is its tendency to develop static charges with increased 
moisture content, sometimes causing striation or separation in the 
granulation. This can be partially overcome by drying the cellulose to 
remove the moisture. When wet-granulated, dried, and compressed, it 
does not disintegrate as readily as when unwetted. It can be used with 
almost all drugs except those that are moisture- sensitive (such as aspirin, 
penicillin, and vitamins) unless it is dried to a moisture content of less than 
1% and then handled in a dehumidified area. 
Solka floc (p urified wood cellulose) is a white, fibrous, inert, neutral 
material that can be used alone or in combination with starch as a disintegrating 
agent for aspirin, penicillin, and other drugs that are pH - and 
moisture- sensitive. Its fibrous nature endows it with good wicking properties 
and is more effective when used in combination with clays such as 
kaolin I bentonite, or Veegum. This combination is especially effective in 
tablet formulations possibly having a high moisture content (such as ammonium 
chloride, sodium salicylate, and vitamins). 
Alginic acid is a polymer derived from seaweed comprising D-mannuronic 
and L- glucuronic units. Its affinity for water and high sorption capacity 
make it an excellent disintegrant. It is insoluble in water, slightly acid 
in reaction, and should be used only in acidic or neutral granulations. It 
can be used with aspirin and other analgesic drugs. If used with alkaline 
salts or salts of organic acids, it tends to form soluble or insoluble alginates 
that have gelling properties and delay disintegration. It can be successfully 
used with ascorbic acid, multivitamin formulations, and acid salt s of 
organic bases. 
Explotab (sodium starch glycolate) is a partially substituted eat-boxymethyl 
starch consisting of granules that absorb water rapidly and swell. 
The machanism by which this action takes place involves accelerated absorption 
of water leading to an enormous increase in volume of granules. 
This results in rapid and uniform tablet disintegration. Explotab is official 
in the N.F. XVI.

1 76 Bandelin 
Example 19: Multivitamin Tablets 
Vitamin A (coated) 
Vitamin D (coated) 
Vitamin C (ascorbic acid, coated) 
Vitamin B1 (thiamine mononitrate) 
Vitamin B2 (riboflavin) 
Vitamin B6 (pyridoxine hydrochloride) 
Vitamin B12 (cyanocobalamin) 
Calcium pantothenate 
Sodium saccharin 
Mannitol NF (fine powder) 
Starch 1500 (internal disintegrant) 
Magnesium stearate 
Starch 1500 (external disintegrant) 
Per tablet 
5000 USP units 
400 USP units 
60 mg 
2 mg 
1.5 mg 
2 l1g 
3 mg 
10 mg 
0.3 mg 
350 mg 
65 mg 
10 mg 
12 mg 
40 mg 
q .s . 
Blend the mannitol, saccharin, and internal Starch 1500 
with 10% of the riboflavin and all the other vitamins 
except A, D, and C. Granulate this blend with water. 
Dry at 120F, pass through a 15-mesh screen, and add 
the flavor. Mix the ascorbic acid with the magnesium 
stearate; mix the vitamins A and D with the remainder of 
the ri boflavi n , Add these and the talc and the external 
Starch 1500 to the previous mixture and mix well. Compress 
using 7/16-in., flat-face, bevel edge punches. 
Guur gum is a naturally occurring gum that is marketed under the 
trade name Jaguar. It is a free-flowing, completely soluble, neutral polymer 
composed of sugar units and is approved for food use. It is available 
in various particle sizes and finds general use as a tablet disintegrant. It 
is not sensitive to pH, moisture content, Or solubility of the tablet matrix. 
Although an excellent disintegrant, it has several drawbacks. It is not 
always pure white, and it sometimes varies in color from off-white to tan. 
It also tends to discolor with time in alkaline tablet. 
Polyclar AT (Polyplasdone XL and Polyplasdone XL10) are crosslinked, 
insoluble homopolymers of vinylpyrrolidone. Polyplasdone XL ranges in 
particle size from 0 to 400 + um , and Polyplasdone XL10 has a narrower 
range and smaller particle size (0 to 74 um) , which makes for better distribution 
and reduced mottling in tablet formulations. Tablet hardness

Compressed Tablets by Wet Granulation 177 
and abrasion resistance are less affected by the addition of Polyplasdone 
XL as compared to starches, cellulose, and pectin compounds [73]. A 
tendency toward tablet capping is reduced [74]. Polyplasdone XL disintegrants 
do not reduce tablet hardness and provide rapid disintegration 
and improved dissolution [75-77]. Polyplasdone, due to its high capillary 
activity, rapidly draws water into the tablet causing swelling which exceeds 
the tablet strength, reu sIting in spontaneous tablet disintegration. 
Amberlite IPR 88 (ion exchange resin) has the ability to swell in the 
presence of water thereby acting as a disintegrant. Care must be taken 
in the selection of a resin as a disintegrant since many resins have the 
ability to adsorb drugs upon them. Anionic and cationic resins have been 
used to absorb substances and release them when the charge changes. 
Methyl cellulose, sodium carboxymethylcellulose, and hydroxypropylcellulose 
are disintegrants to some extent depending on their ability to 
swell on contact with water. Generally, these do not offer any advantage 
over more efficient products such as the starches and microcrystalline cellulose. 
However, in certain cases they may be of benefit when used in 
conjunction with the above. 
E. Glidants 
Glidants are materials that improve the flow characteristics of granulations 
by reducing interparticulate friction. They increase the flow of materials 
from larger to smaller apertures, from the hopper into the die cavities of 
the tablet press. 
The effects produced by different glidants depend on (a) their chemical 
nature in relation to that of the powder or granule (i .e . the presence 
of unsaturated valences, ionic or hydrogen bonds on the respective surfaces 
that could interact chemically) and (b) the physical factors including 
particle size, shape, and distribution of the glidant and various other formulation 
components, moisture content, and temperature. In general, hydrophilic 
glidants tend to be more effective on hydrophilic powders, and the 
opposite is true for hydrophobic glidants. For any particular system there 
is usually an optimum concentration above which the glidant may start to 
act as a antigIidant [78]. This optimum depends, among other factors, 
on the moisture level in the granUlation [79]. 
When fine particles of less than the optimum for flowability are added 
to a bulk powder of similar chemical constitution, there is often an improvement 
in the rate of flow through an orifice [80]. The improvement is dependent 
on the size and concentration of the fine particles; the smaller the 
particles, the lower the concentration required to produce an increased 
Some glidants commonly used and suggested concentrations for optimum 
glidant effect are shown in Table 8. 
The silica-type glidants are the most efficient probably because of their 
amall particle size. In one study [81], it was found that all silica-type 
glidants improved the flow properties of granulations as reflected in increased 
tablet weight and in decreased weight variation in the tablets. 
Chemically, the silica glidants are silicon dioxide. They are available as 
two types, both insoluble: (a) the pyrogenic silicas prepared by burning 
silicon tetrachloride in an atmosphere of oxygen and (b) the hydrogels, 
which are prepared by the precipitation of soluble silicates. The pyrogenic

Table 8 Commonly Used Glidants and 
Usual Concentration Range 
Silica aerogels 
Cab-O-Sil M- 5 
Aerosil 200 
QUSO F-22 
Calcium stearate 
Magnesium stearate 
Stearowet C* 
Zinc stearate 
Calcium silicate 
Starch, dry flow 
Starch 1500 
Magnesium lauryl sulfate 
Magnesium carbonate, heavy 
Magnesium oxide, heavy 
1. 0-10.0 
1. 0-5.0 
silicas are generally composed of smaller particles that tend to be more 
spherical in shape. Pyrogenic silicas are available in both hydrophilic and 
hydrophobic form [82]. The particle size of most commercially available 
silicas used as glidants range in size from 2 to 20 nm and have an enormous 
surface area averaging 200 to 300 m2 g-l. 
There are no specific rules dictating the amount of any glidant required 
for a particular granulation. Glidants differ not only in chemical properties 
but also in physical characteristics such as size, frictional properties. 
structure, and density. For these reasons the amount of glidant varies 
with the material to which it is added. Since it is the purpose of the glidant 
to confer fluidity on the granulation, this property may be measured 
by one of several methods [83]. One method is the determination of the 
angle of repose {84, 851. When powdered material is allowed to fall freely 
from an orifice onto a flat surface. the material deposited forms a COne. 
The base angle of the COne is referred to as the angle of repose. By 
this method it has been found. for example, that the repose angle of a 
sulfathiazole granulation increases with decreasing particle size. Talc added 
in small quantities reduces the angle of repose, indicating greater flow, but 
tends to increase the repose angle at higher concentrations, thus becoming 
an antiglidant. The addition of fines causes a marked increase in the repose 
Another method of determining the effect of glidants on the flow properties 
of a granulation is that of allowing a given amount of granulation. 
with and without glidant , to flow through an orifice ranging in size from

Compressed Tablets by Wet Granulation 179 
3/8 to 1 in. in diameter depending on the size of the granules, and observing 
the efflux time. The glidant efficiency factor may then be determined 
as follows: 
f :::: rate of flow in presence of glidant 
rate of flow in absence of glidant 
Since many materials used as glidants are also efficient lubricants, a 
reduction in Interparticulate friction may also be encountered. This reduction 
can occur in two ways: (a) The fine material may adhere to the 
surface rugosity, minimizing the mechanical interlocking of the particles. 
(Rugosity refers to surface roughness or deviation of shape from spherical. 
The coefficient of rugosity is defined as the ratio of actual surface area, 
as determined by a suitable method, to the geometric surface area found 
by microscopy.) (b) Certain glidants, such as talc and silica aerogels, roll 
under shear stresses to produce a "ball bearing" effect or type of action I 
causing the granules to roll OVer one another. 
Many powders acquire a static charge during handling, in mixing, or 
in an induced die feed. The addition of 1% or more of magnesium stearate 
or polyethylene glycol 4000 or 2% or more of talc effectively lowers the accumulated 
Magnesium oxide should be considered an auxiliary glidant to be used 
in combination with silica-type glidants, especially for granulations that 
tend to be hygroscopic or somewhat high in moisture content. Magnesium 
oxide binds water and keeps the granulation dry and free flowing. 
That anomalies exist in the action of glidants has been pointed out 
[86] in some cases of the physical and mechanical properties of mixtures of 
lactose, paracetamol , and oxytetracycline when small amounts of silica 
glidants are added to them. Owing to the differing propensities to coat 
the particles of the host powders, the silica aerogels act as a glidant for 
lactose and paracetamol but as an antiglidant for oxytetracycline. 
Selection of glidants must be determined by the formulator by trial and 
error since there is no way of predicting which will be effective in a specific 
Multilayer tablets are tablets made by compressing several different granulations 
fed into a die in succession, one on top of another, in layers. Each 
layer comes from a separate feed frame with individual weight control. Rotary 
tablet presses can be set up for two or three layers. More are possible 
but the design becomes very special. Ideally, a slight compression 
of each layer and individual layer ejection permits weight checking for 
control purposes. 
A. Advantages of Multilayer Tablets 
1. Incompatible substances can be separated by formulating them in 
separate layers as a two-layer tablet or separating the two layers 
by a third layer of an inert substance as a barrier between the 

180 Baruielin 
2. Two layer tablets may be designed for sustained release-one 
layer for immediate release of the drug and the second layer for 
extended release, thus maintaining a prolonged blood level. 
3. Layers may be colored differently to identify the product. 
B. Layer Thickness 
Layer thickness can be varied within reasonable proportions within the 
limitations of the tablet press. Thinness is dependent on the fineness of 
the granulation. 
C. Sizes and Shapes 
Size is limited by the capacity of the machine with the total thickness 
being the same as for a single-layer tablet. Many shapes other than 
round are possible and are limited only by the ingenuity of the die maker. 
However, deep concavities can cause distorition of the layers. Therefore, 
standard concave and flat-face beveled edge tooling make for the best appearance, 
especially when layers are of different colors. 
D. Granulations 
For good-quality tablets with sharp definition between the layers, special 
care must be taken as follows: 
1. Dusty fines must be limited. Fines smaller than 100 mesh should 
be kept at a minimum. 
2. Maximum granule size should be less than 16 mesh for a smooth, 
uniform scrape-off at the die. 
3. Materials that smear, chalk, or coat on the die table must be 
avoided to obtain clean scrape-off and uncontaminated layers. 
4. Low moisture is essential if incompatibles are used. 
5. Weak granules that break down easily must be avoided. Excessive 
amounts of Iubzication , especially metallic stearates, should be 
avoided for better adhesion of the layers. 
6. Formulation of multilayer tablets is more demanding than that of singlelayer 
tablets. For this reason, selection of additives is critical. 
E. Tablet Layer Press 
A tablet multilayer press is simply a tablet press that has been modified 
so that it has two die-filling and compression cycles for each revolution of 
the press. In short, each punch compresses twice, once for the first 
layer of a two-layer tablet and a second time for the second layer. Threelayer 
presses are equipped with three such compression cycles. 
There are two types of layer presses presently in use-one in which 
each layer can be ejected from the press separately for the purpose of 
weight checking, and the second in which the first layer is compressed so 
hard that the second layer will not bond to it, or will bond so poorly 
that upon ejection the layers are easily separated for weighing. Once the

Compressed Tablets by Wet Granulation 181 
proper weight adjustments have been made by adjusting the die fill , the 
pressure is adjusted to the proper tablet hardness and bonding of the 
One hazard of layer tablet production is the lack of proper bonding 
of the layers. This can result in a lot of 100,000 tablets ending up as 
200,000 layers after several days if the layers are not sufficiently bonded. 
In a two-layer tablet press, two hoppers above the rotary die table 
feed granulated material to two separate feed frames without intermixing. 
Continuous, gentle circulation of the materials through the hoppers and 
feed frames assures uniform filling without segregation of particle sizes 
that would otherwise carryover to the second layer and affect layer weight, 
tablet hardness, and, in the case of differently colored granulations, the 
appearance of the tablet. The same procedure is followed in the three-layer 
press with three hoppers for the three granulations instead of two. 
Certain single-layer or unit tablet presses are equipped with two precompression 
stations prior to the final compaction. This provides highspeed 
production by increasing dwell time of the material under pressure 
making for harder, denser tablets. 
Prolonged or sustained release tablets can be made by the wet granulation 
method using slightly soluble or insoluble substances in solution as binding 
agents or low-melting solids in molten form in which the drug may be incorporated. 
These include certain natural and synthetic polymers, wax 
matrices, hydrogenated oils, fatty acids and alcohols, esters of fatty acids. 
metallic soaps, and other acceptable materials that can be used to granulate, 
coat, entrap, or otherwise limit the solubility of a drug to achieve a 
prolonged or sustained release product. 
Freely soluble drugs are more difficult to sustain than slightly soluble 
drugs because the sustaining principle is largely a waterproofing effect. 
Ideally, the ultimate criterion for a sustained release tablet is to achieve 
a blood level of the drug comparable to that of a liquid product administered 
every 4 hr. fO this end, prolonged release dosage forms are designed to 
release the drug so as to provide a drug level within the therapeutic range 
for 8 to 12 hr with a single dose rather than a dose every 4 hr (Fig. 13). 
They are intended as a convenience so that the patient needs to take only 
one dose morning and evening and need not get up in the night. 
Prolonged drug forms are not without disadvantages. Since gastrointestinal 
tracts are not all uniform, certain individuals may release too much 
drug too soon and experience toxic or exaggerated response to the drug, 
whereas others may liberate the drug more slowly and not receive the 
proper benefit or response anticipated. This is especially true of older 
people whose gastrointestinal tract is less active than that of the younger. 
Also. where liberation is slow, there is danger of accumulation of the drug 
after several days resulting in high blood levels and a delayed exaggerated 
Prolonged release products may be divided into two classes: 
1. Prolonged release 
2. Repeat action

Drug concentration in blood 
1st dose 2nd dose 3rd dose 
Drug concentration in blood 
1st dose 
figure 13 Conventional versus prolonged release dosage forms. 
(Left) repeated doses of conventional drug, and (right) single dose 
of ideal controlled release drug. 
A prolonged (or sustained) release product is one in which the drug is 
initially made available to the body in an amount sufficient to produce the 
desired pharmacological response as rapidly as is consistent with the properties 
of the drug and which provides for the maintenance of activity at 
the initial level for a desired number of hours. 
A repeat action preparation provides for a single usual dose of the drug 
and is so formulated to provide another single dose at some later time after 
administration. Repeat action, as defined here, is difficult to achieve and 
most products on the market today are of the sustained release type. 
Many varied materials have been used in practice to achieve prolonged 
release dosage forms. The following example illustrates the Ubiquitous 
nature shown in Example 20. 
Prolonged release tablets must be tested for the rate of drug release 
by the prescribed in vitro laboratory method. Each product has an inherent 
release rate based on properly designed clinical trials of blood concentration 
and excretion in humans which is compared to the concentration and pharmacological 
activity resulting from the usual single-dose schedule of the drug 
administered in solution. 
Once established, the in vitro testing based on the above is valuable 
for manufacturing control purposes to assure batch-to-batch uniformity of 
drug release. 
Typical examples of release rates by laboratory tests are illustrated in 
Example 21. 
Different drugs require different time release patterns depending on 
the half-life of the drug in the blood. 
A prolonged release tablet containing two drugs in a single granulation 
has been patented in Example 22. 
Some formulations are so constructed as to separate the ingredients 
into two formulations, one for immediate release and one for prolonged

Compressed Tablets by Wet Granulation 
Example 20: Ferrous Sulfate Prolonged Release Tablets 
mg per 
Ingredients tablet 
Ferrous sulfate, anhydrous, fine powder 
Lactose, fi ne powder 
Methocel E 15LV 
Ethylcellulose, 50 cps, 15% in 95% ethanol 
Magnesium stearate, fine powder 
Mix the ferrous sulfate and the lactose and granulate 
with the ethylcellulose solution and dry at 120 to 130F. 
(It will be necessary to granulate several times to achieve 
25 mg per tablet of ethylcellulose. The batch must be 
weighed after each addition until the proper weight is 
attained. ) 
In a twin-shell blender, add the Cab-O-Sil and blend 
for 5 min, next add the magnesium stearate and blend 
for 2 min. Compress with 13/32-in.-deep cup punches. 
Coat the tablets with cellulose acetate phthalate solution 
in alcohol and ethyl acetate. 
release. The following formulation illustrates this by employing a two-layer 
tablet for the formulations. 
Still another type is a tablet containing the prolonged release drug(s) 
in the core tablet and the immediate release dose in the coating as is illustrated 
by Example 24. 
Prolonged release tablets have also been prepared by incorporating the 
drug in a granulation for immediate release and in another granulation for 
prolonged release. then mixing the two granUlations and compressing as 
given in Example 24. 
Example 21: Typical In Vitro Drug 
Release Rates 
Percent cumulative release 
Product A Product B 
28 36 
26 44 
54 58 
71 74 
82 86

Example 22: Prolonged Release Hydrochlorothiazide 
with Probenecid Tablets [87] 
Cellulose acetate phthalate (5% solution 
in acetone) 
Magnesium stearate 
mg per 
Mix the hydrochlorothiazide, probenecid with the 
lactose and 20 mg of starch, granulate with the 
cellulose acetate phthalate solution; pass the wet 
mass through a 10-mesh screen. Dry at 120 to 
130F. Screen through a 20-mesh screen, incorporate 
the magnesium stearate and the remaining 
starch, and compress into tablets. 
Example 23: Prolonged and Immediate Release 
Tablet Containing Pentaerythritol Tetranitrate 
Two-Layer Tablets 
I mmediate Release Layer 
Pentaeryth ritol tetran itrate 
Calcium sulfate, dihydrate 
Starch paste, 10% 
Magnesium stearate 
Prolonged Release Layer 
Penterythritol tetranitrate 
mg per 
q .s , 

Compressed Tablets by Wet Granulation 
Example 23 (Continued) 
Prolonged Release Layer 
Acacia, powdered 
cae-o-su M-5 
mg per 
Procedure for immediate release layer: Mix 
the first four ingredients and granulate with 
the starch paste through a 12-mesh screen. 
Dry at 130 to 140F and size the dry granulation 
through a 20-mesh screen, add the 
magnesium stearate, and blend for 3 min. 
Hold for compressing on the following layer. 
Procedure for prolonged release layer: Melt 
the beeswax and add all of the ingredients 
except the Cab-Oe-Sll with constant stirring 
and heating to maintain the molten state. 
Allow to cool and granulate by passing the 
mass through an 18-mesh screen; blend in 
the cse-o-su. 
Compression: On a two-layer tablet press, 
first compress the immediate release layer 
with 7/16-in. flat-face, bevel edge punches; 
then compress the prolonged release layer 
on top of it. Check the tablets for layer 
Example 24: Antihistamine Decongestant Prolonged Release Tablet 
Brompheniramine maleate 
Phenyl propanolami ne hyd rochloride 
Calcium sulfate, dihydrate 
Zein granulating solution* 
Zinc stearate 
in core 
mg in 
*Zein granulating solution is prepared as follows: 
Zein G-20oa 
Propylene glycol 
100 g 
10 g

186 Bandelin 
Example 24 (Continued) 
Stearic acid 
Ethyl alcohol, 90% 
10 g 
200 ml 
Dissolve the stearic acid in the alcohol at 35 to 40F, next 
add the propylene glycol and then the zein with constant agitation 
until all is in solution. 
aZe in G-200 is a protein derived from corn. It is resinlike and 
is acceptable for food use. Zein resists microbial decomposition. 
Granulating procedure for core tablet: Mix the three drugs with 
the calcium sulfate and the kaolin, and moisten with the zein 
granulating solution until evenly wetted. Granulate by passing 
through a 12-mesh screen and dry at 120 to 130F. Pass the 
dry granulation through an t s-mesh screen, add the zinc stearate, 
and compress with 5/16-in .-deep cup punches. 
Sugar coating: Dissolve the three drugs for immediate release in 
a solution of 810 g of sucrose, 80 g of acacia in 400 ml water, and 
apply as a sugar coating in a coating pan. 
Example 25: Chloroprophenpyridamine Tablets [88] 
Ingredients Pounds 
Prolonged release granulation-A 
Chloroprophenpyridamine maleate, 50 mesh 5.0 
Terra alba, 60 mesh 45.0 
Sucrose, 75% w/v aqueous solution 15.0 
Cetyl alcohol 10.0 
Stearic acid 5.0 
Glyceryl trilaurate 20.0 
The cetyl alcohol, stearic acid, and glyceryl trilaurate 
are melted together. The chloroprophenpyrldamine 
maleate and terra alba are added to the melted mixture 
with stirring. After mixing, the mixture is cooled until 
congealed to a hard mass. The mass is ground and 
sieved through a 30-mesh screen. The sucrose syrup 
is added to the powder obtained and thoroughly mixed 
to mass the powder. The resulting product is ground 
through a 14-mesh screen. The granules thus formed 
are dried at 37C and sieved through aa 18-mesh 

Compressed Tablets by Wet Granulation 
Example 25 (Continued) 
Ingredients Pounds 
Immedi ate rei ease g ranul ation- B 
Chloroprophenpyridamine maleate, 60 mesh 5.0 
Terra alba, 60 mesh 65.0 
Dextrose, 40 mesh 20.0 
Lactose, 60 mesh 4.0 
Starch, 80 mesh 5.0 
Gelatin, 13% aqueous solution 1.0 
Mix the chloroprophenpyridamine maleate, terra alba, 
lactose, dextrose, and starch and mass with the gelatin 
solution. Granulate through a 14-mesh screen and dry 
at 40C. Sieve the dried granules through an 18-mesh 
Mix equal quantities of the prolonged release granulation-
A and immediate release granulation-B and compress 
into 200-mg tablets. 
Example 26: Prednisolone Tablets [89]
mg per 
I ngredients tablet 
Prednisolone 5.0 
Dicalcium phosphate 
Aluminum hydroxide, dried gel 
Sugar, as syrup 
Magnesium stearate 
Blend the first three ingredients and wet with 
15 ml of syrup having a sugar concentration 
Of 850 gIL. Screen through a 20-mesh screen 
to form granules and dry at 60C for 12 hr. 
The dried material is then passed through 
a 20-mesh screen to form final granules. 
These granules are blended with the magnesium 
stearate and compressed into tablets. This 
formulation is claimed to have a disintegration 
time of 12 hr.

188 Bandelin 
Drugs may also be prepared in prolonged release form by adsorbing 
on acceptable materials such as ionic synthetic resins, aluminum hydroxide, 
and various clays. The following example presents the use of aluminum 
hydroxide and an aqueous granulating liquid (Example 26). 
Prolonged action drug tablets have also been prepared with drugs 
bound to ion exchange resins that permit slow displacement of the drug 
from the drug-resin complex when it comes into contact with the gastrointestinal 
fluids. The displacement reaction of drug-resin complex may be 
described by the following equation; 
(R-SO -H N-R') - (X-Y) 3 3 
where X is H or some other cation and Y is CiaI' some other anion. The 
opposite of this would occur if an acidic drug were bound to an anion exchange 
resin with CiaI' other anion causing drug displacement. 
Preparation of drug-ion exchange complexes are described in several 
patents [89-93]. Drug in solution in excess or less than the amount required 
by stoichiometric considerations is exposed to a suitable resin displacing 
the cation or anion, as the case may be, for the resin. After 
washing with water, the resin is dried and is then incorporated into a 
tablet granulation. 
Although tablet presses have become more complex over the years as a result 
of numerous modifications, the compaction of material in a die between 
upper and lower punches remains essentially the same. The main differences 
that have been made are increase in speed, mechanical feeding of the 
material from the hopper into the die, and electronic monitoring of the 
press. Precompression stations allow for the elimination of air from the 
gr-anulation by partially compressing the tablet material prior to final pressing 
of the tablet. This makes for harder. firmer tablets with less tendency 
toward capping and lower friability. The number of tablets a press can 
produce is determined by the number of tooling stations and the rotational 
speed of the press. Large presses can produce as many as 10,000 tablets 
per minute. All these advancements and innovations, however, have not 
decreased the problems often encountered in production, and in fact have 
increased the problems because of the complexities of the presses and the 
greater demands of quality. 
The production of faulty or imperfect tablets creates problems that 
range from annoying to serious. These are time consuming and costly. 
Imperfections may arise from causes inherent in the granulation to improper 
machine adjustment and lor tooling. 
A. Binding 
Binding in the die or difficult ejection is USUally due to insufficient lubrication. 
It is the resistance of the tablet to ejection from the die. This can 
cause the tablet press to labor and squeak producing tablets with rough 
edges and vertical score marks on the edges. This may be overcome by,

Compressed Tablets by Wet Granulation 189 
1. Increasing lubrication 
2. Using a more efficient lubricant 
3. Improving the distribution of the lubricant by screening through 
an 30-mesh screen and mixing with a portion of fines screened 
from the granulation 
4. Reducing the size of the granules 
5. Increasing the moisture content of the granulation 
6. Using tapered dies 
7. Compressing at a lower temperature and/or humidity. 
B. Sticking, Picking, and Filming 
Sticking is usually due to improperly dried or lubricated granulation causing 
the tablet surface to stick to the punch faces. Contributing to this are 
tablet faces that are dull, scratched, or pitted. This condition usually becomes 
progressively worse. 
Picking is a form of sticking in which a small portion of granulation 
sticks to the punch face and grows with each revolution of the press, 
picking out a cavity on the tablet face. 
Filming is a slow form of picking and is largely due to excess moisture 
in the granulation, high humidity, high temperature, or loss of highly 
polished punch faces due to wear. These may be overcome by 
1. Decreasing the moisture content of the granulation 
2. Changing or decreasing the lubricant 
3. Adding an adsorbent (Le . , silica aerogel, aluminum hydroxide, 
microcrystalline cellulose) 
4. Polishing the punch faces 
5. Cleaning and coating the punch faces with light mineral oil, lowviscosity 
C. Capping and Laminating 
Capping occurs when the upper segment of the tablet separates from the 
main portion of the tablet and comes off as a cap. It is usually due to air 
entrapped in the granulation that is compressed in the die during the compression 
stroke and then expands when the pressure is released. This 
may be due to a large amount of fines in the granulation and/or the lack 
of sufficient clearance between the punch and the die wall. It is often 
due to new punches and dies that are tight fitting. Other causes may be 
too much or too little lubricant or excessive moisture. 
Lamination is due to the same causes as capping except that the tablet 
splits and comes apart at the sides and is ejected in two parts. If tablets 
laminate only at certain stations, the tooling is usually the cause. The 
following should be tried to overcome capping and laminating: 
1. Changing the granulation procedure 
2. Increasing the binder 
3. Adding dry binder such as pregelatinized starch, gum acacia, 
powdered sorbitol, PVP, hydrophilic silica, or powdered sugar 
4. Increasing or changeing lubrication

190 Bandelin 
5. Decreasing or changing lubrication 
6. Using tapered dies 
7. Decreasing the upper punch diameter by 0.0005 in. to 0.002 in. 
depending on the size 
D. Chipping and Cracking 
Chipping refers to tablets having pieces broken out or chipped, usually 
around the edges. This may be due to damaged tooling or an improperly 
set takeoff station. These problems are similar to those of capping and 
laminating, and are annoying and time consuming. Cracked tablets are 
usually cracked in the center of the top due to expansion of the tablet, 
which is different from capping. It may occur along with chipping and 
laminating and lor it may be due to binding and sticking. It often occurs 
where deep Concave punches are used. These problems may be overcome 
by one or more of the following: 
1. Polishing punch faces 
2. Reducing fines 
3. Reducing granule size 
4. Replacing nicked or chipped punches 
5. Adding dry binder such as pregelatinized starch, gum acacia, PVP, 
spray-dried corn syrup, powder-ed sugar, or finely powdered gelatin 
Solving many of the manufacturing problems requires an intimate knowled 
ge of granulation processing and tablet presses, and is acquired only 
through long study and experience. 
The foregoing are just a few of the problems of tablet manufacture 
that are encountered in production by the pharmaceutical scientist, and as 
new technologies develop, new problems arise. 
For decades wet granulations have been processed on a purely empirical 
basis, often on a small scale. If tablet compression ran smoothly, reproducibility 
of the granulation was unimportant. Today, however, highspeed 
presses, demanding specifications, GMP regulations, and validation 
requirements have given rise to the need for more and greater effort to 
assure uniformity and reproducibility of the gr-anulation. Experience indicates 
that formulation and process variables greatly influence the performance 
characteristics of the final product. Recent developments in techniques 
utilizing various high-shear mixers. granulating by extrusion, spray drying, 
pan granulating, and fluid bed agglomeration have presented new areas of 
investigation. Fast -running, automated processes demand greater control 
through instrumental and computer monitoring for satisfactory scale-up from 
laboratory to production scale. It is to this end that more research needs 
to be directed. 
1. D. E. Fonner, G. S. Banker, and J. Swarbrick, J. Pharm. sa., 55: 
181-186 (1966). 
Z. P. J. Sherington and R. Oliver. Granulation, Heyden and Son Ltd., 
Philadelphia (1981).

Compressed Tablets by Wet Granulation 191 
3. T. Schaefer and O. Worts, Arch. Pharm. Chem. Sci. Ed., 6:69-72 
(1978) . 
4. T. Schaefer and O. Worts, Arch. Pharm. Chim. Sci. Ed., 6:14-25 
(1978) . 
5. M. E. Aulton and M. Banks, Int. J. Pharm. Techno!. Prod. Manufact., 
2:24-28 (1981). 
6. T. Schaefer and O. Worts, Arch. Pharm. Chem., Sci. Ed., 6:5:178193 
7. K. T. Jaiyeoba and M. S. Spring, J. Pharm. Pharmacol., 33:5-11 
8. K. V. Sastry and D. W. Fuerstenau, in K. V. Sastry (ed.), 
Agglomeration, AIME, New York, 1977, p , 381. 
9. M. J. Gamlen, H. Seager, and J. K. Warrach, Int. J. Pharm. Tech. 
Prod. Manufact., 3:(4)108-114 (1982). 
10. E. Nuernberg, Acta Pharm. Techno/., 26:39-42 (1980). 
11. H. Seager and C. R. Trask, Mfg. Chem. Aerosol News, Dec. 10, 1976. 
12. H. Seager, Mfg. Chem. Aerosol News, April 2, 1977. 
13. D. M. Jones, Pharm. Tech., 9:50 (1985). 
14. H. J. Malinowski and W. E. Smith, J. Pharm'. Sci., 63:285-288 (1974). 
15. H. J. Malinowski, J. Pharm. Sci., 64:1688-1692 (1975). 
16. R. C. Rowe, Pharm. Ind., 46:119-125 (1985). 
17. T. Schaefer et al , , Pharm, lrui., 48:1083-1095 (1986). 
18. R. Kinget and R. Kernel, Acta Pharm. Techno!., 31: 57-61 (1985). 
19. H. Lewenberger, Acta Pharm. Technol., 29:274-283 (1983). 
20. P. Holm et al., Pharm. Ind., 45:886-888 (1983). 
21. T. Schaefer et al ; , Arch. Pharm. Chem., Sci. Ed., 14:1-8 (1986). 
22. Chemical Engineer's Handbook, 5th ed., McGraw-Hill, New York, 1975, 
pp. 53-58. 
23. P. C. Record, Int. J. Pharm. Prod. Dev., 1:32-39 (1980). 
24. W. A. Knepper (ed.), Agglomeration, John Wiley and Sons, New York, 
1962, pp. 365-383. 
25. C. Orr, Particulate Technology, Macmillan, New York, 1966, pp. 400432. 
26. H. G. Kristensen and To Schaefer, Drug. Dev. Ind. Pruirm , , 13: (4 
and 5),803-872 (1987). 
27. C. E. Capes, Particle size enlargement, in Vol. I, Handbook of Powder 
Technology, J. C. Williams and T. Allen (eds.), Elsevier, Amsterdam, 
28. H. Rumpf, Chem. Ind. Techno!. 46: (1974). 
29. R. C. Shah, P. V. Raman, and P. L. Sheth, J. Pharm. Sci. 66:15541561 
30. M. H. Rubenstein and B. Musikabhumma, Drug Dev. Ind. Pharm., 
6:451-58 (1976). 
31. K. Pataki et al , , Proe, Con]; Appl. Chern. Unit Oper. Processes, 
3( 4) : 258- 270 (1983). 
32. M. H. Rubenstein, J. Pharm. Pharmacol., 28: Suppl , 67P (1976). 
33. J. L. Ford and M. H. Rubenstein, Pharm. Helv , Acta, 55: 1-14 
(1980) . 
34. P. Flanders et al , Proc. Pharm. Tech. 5th Conf., 2:40 (1986). 
35. J. C. McTaggert et al , , Int. J. Pharm., 19:139-145 (1984). 
36. 1. Ullah et aI., Pharm. Tech., 11: 48-54 (1987). 
37. J. T. Carstensen et aI., J. Pharm. Sci., 65:992-997 (1976). 
38. M. A. Zoglio et al., J. Pharm. Sci., 65:1205-1208 (1976).

192 Bandelin 
39. A. Rogerson et a1.. J Pharm. Pharmacol., 28: Suppl. 63D (1976). 
40. W. P. Bogar and J. J. Gavin, N. Engl. J. Med . 261:827 (1959). 
41. R. Costello and A. J. Mattocks. J. Pharm. Sci., 51:106-110 (1962). 
42. R. M. Duvall et al . J. Pharm. Sci., 54: 607- 611 (1965). 
43. A. L. Jacobs, Pharm. Mfg. June; pp. 43-45 (l985). 
44. M. J. Akers, Can. J. Pharm. su., 11:1-8 (1976). 
45. L. Lachman and P. DeLuca, Theory and Practice of Industrail Pharmacy, 
2nd Chap. 2, Lea and Febiger, Philadelphia, 1976. 
46. S. A. Botha. A. P. Lotter, and J. L. dePreez, Drug. Dev . Ind. 
Pharm., 13:1197-1215 (1987). 
47. M. T. Sclosserman and A. S. Feldman. J. Soc. Cos. Chem., 24:357- 
361 (1973). 
48. E. Doelker and E. J. Shotton, J. Pharm , Sci., 66:193-196 (1977). 
49. J. Carstensen et al . , J. Pharm. Sci., 65: 992- 998 (1976). 
50. A. A. Chalmers and P. H. Elsworthy, J. Ptuirm . Pharmaco1., 28:228230 
51. Z. T. Chowhan and Y. D. Chow, Int. J. Pharm. Prod. Mfr., 2:2934 
52. J. L. Fabricans and J. Cicala, J. Drug. Dev. Ind. Pharm . 13:12171227 
53. J. I. Wells, D. A. Bhatt, and K. A. Khan, J. Pharm. Pharmacal., 
32:55-59 (1980). 
54. H. N. Wolkoff and G. Pinchuk, U.S. Patent 3,511,914 (1970). 
55. Tablet Making. Tech. Bull. 14-100.500A, Pennwalt-Stokes, Warminster, 
56. Y. Matenda, V. Minamida, and Hayashi, J. Pharm. sa., 65: 1158- 
1162 (1976). 
57. R. Appino, G. Banker. and G. DeKay. Drug Stand., 27:193-195 (1959). 
58. A. C. Carman. Soc. Chem. Ind . London, 57:225-231 (1939). 
59. A. E. Butcher and T. M. Jones, J. Pnarm, Pharmacal., 24:163-166 
(1972) . 
60. E. Nelson et al ; , J. Am. Pharmaceut. Assn . Sci. Ed . 43:596-601 
(1954) . 
61. G. Ragnarssen, A. W. Holtzer, and S. Sjogren, Int. J. Pharm . 3: 
127-131 (1979). 
62. G. K. Bolhuis and C. F. Lerk, J. Pharm. Pharmacal., 33:790 (1981). 
63. P. J. Jarosz and E. L. Parrott, Drug Dev . Ind. Pharm., 10:259-273 
(1984) . 
64. M. E. Johnsson and M. Nicklasson, J. Pruirm, Pharmacal., 38:51-54 
(1986) . 
65. M. 1. Blake, J. Am. Pharmaceut. Assn., New Series, 11: 603 (1971). 
66. A. C. Caldwell and W. J. Westlake. J. Pharm. sa., 61:984-988 
(1972) . 
67. N. H. Shah et al., Drug. Dev , Ind. Pharm., 12:329-246 (1986). 
68. A. A. Khan and C. T. Rhodes, J. Pharm. su., 64:447-453 (1975). 
69. K. A. Khan and D. J. Rooke, J. Pharm. Pharmacal., 28:6336 (1976). 
70. L. A. Bergman and F. J. Bandelin, J. Pharm. sa., 54:445-448 
(1965) . 
71. E. Shotton and G. S. Leonard, J. Pharm. Sci., 65:1170-1176 (1976). 
72. H. Hess. Pharm. Tech . 2:36-57 (1978). 
73. R. Huttenrauch and 1. Kleiner. Pharmazie, 28: 40-48 (1973). 
74. A. H. Bronnsack , Pharm. Ind., 38:40-45 (1978). 
75. D. Gessinger and A. Stamm, Pharm. Ind., 42:189-195 (1980).

Compressed Tablets by Wet Granulation 193 
76. H. Flasch, G. Asmussen, and N. Heinz, Ar. Forsch., 28:1-12 (1978). 
77. K. A. Kahn and D. J. Rooke, J. Pharm. Pharmacol., 28: 633 (1976). 
78. H. G. Kristensen, Dan. Tidsskr. Farm., 45:114-120 (1971). 
79. N. Pilpel, in Advances in Pharmaceutical Sciences. Vol. 3, Chap. 3, 
H. S. Bean (ed.), Academic Press, London, 1971. 
80. G. Gold et aI., J. Pharm. Sci., 57:667-671 (1968). 
81. R. Tawashi, Pharm. Ind., 26:682 (1964). 
82. Tech. Bull. 49, Aerosil in Pharmaceuticals and Cosmetics, Degussa, 
New York. 
83. Silanox Bull., Cabot Corp.. Boston. 
84. E. J. Nelson, J. Am. Pharmaceut. Assn., Sci. Ed., 44:435-441 (1955). 
85. T. M. Jones, J. Soc. Cos. Chem., 21:483-489 (1970). 
86. S. Varthalis and N. Pilpel, J. Pnarm, Pharmacal., 29:37 (1977). 
87. J. E. Baer, Br. Patent 646,426 (1967). 
88. E. V. Svedres, U.S. Patent 2,793,999 (1957). 
89. E. Schuster, U.S. Patent 3,469,545 (1971). 
90. S. P. Retyet al , , Br. Patent 862,242 (1958). 
91. E. L. Gustus, U.S. Patent 4,101,390 (1977). 
92. C. R. Hamilton, U.S. Patent 4,473,957 (1982). 
93. F. J. Keller, U. S. Patent 3,687,465 (1975).

Compressed Tablets by Direct Compression 
Ralph F. Shangraw 
The University of Maryland School of Pharmacy, Baltimore, Maryland 
Until the late 1950s the vast majority of tablets produced in the world 
were manufactured by a process requiring granulation of the powdered 
constituents prior to tableting. The primary purpose of the granulation 
step is to produce a free-flowing and compressible mixture of active ingredients 
and excipients. The availability of new excipients or new forms of old 
excipients, particularly fillers and binders, and the invention of new (or the 
modification of old) tablet machinery have allowed the production of tablets 
by the much simpler procedure of direct compression. However, in spite 
of its many obvious advantages, tableting by direct compression has not 
been universally adopted even in those cases where it would seem to be 
technically feasible and advantageous. The reasons for this can be understood 
only by reviewing the development of direct-compression technology 
and the decision-making steps involved in selecting one manufacturing 
process over another. 
The term direct compression was long used to identify the compression 
of a single crystalline compound (usually inorganic salts with cubic crystal 
structures such as sodium chloride, sodium bromide, or potassium bromide) 
into a compact without the addition of other substances. Few chemicals 
possess the flow, cohesion, and lubricating properties under pressure to 
make such compacts possible. If and when compacts are formed. disintegration 
usually must take place by means of dissolution-which can take a considerable 
length of time, delaying drug release and possibly causing physiological 
problems such as have occurred in potassium chloride tablets. 
Note: A glossary of direct-compression excipients, trade names, and 
supplies can be found on page 243. 

196 Shangraw 
Furthermore. the effective dose of most drugs is so small that this type of 
direct compression is not practical for most drug substances. 
Pellets of potassium bromide are directly compressed for use in infrared 
spectrophotometry. and disks of pure drug have been directly compressed 
for the study of intrinsic dissolution rates of solids. However, there are few 
examples today of direct compression as classically defined in the literature. 
The term direct compression is now used to define the process by which 
tablets are compressed directly from powder blends of the active ingredient 
and suitable excipients (including fillers, disintegrants, and lubricants), 
which will flow uniformly into a die cavity and form into a firm compact. 
No pretreatment of the powder blends by wet or dry granulation procedures 
is necessary. Occasionally, potent drugs will be sprayed out of solution 
onto one of the excipients. However, if no granulation or agglomeration is 
involved, the final tableting process can still be correctly called direct compression. 
The first significant discussion of the concept of direct compression 
was presented by Milosovitch in 1962 [1]. 
Increasingly, there has been a trend toward integrating traditional wet 
granulation and direct-compression processes wherein triturations of potent 
drugs or preliminary minigranulations are added to direct-compression filler 
binders and then compressed. These techniques will be described later in 
the chapter. 
The advent of direct compression was made possible by the commercial 
availability of directly compressible tablet vehicles that possess both fluidity 
and compressibility. The first such vehicle was spray-dried lactose, which, 
although it was subsequently shown to have shortcomings in terms of compressibility 
and color stability, initiated the "direct-compression revolution" 
[2]. Other direct-compression fillers were introduced commercially in the 
1060s. including: Avice! (microcrystalline cellulose), the first effective dry 
binder/fl11er [2) j Starch 1500, a partially pregelatinized starch that possesses 
a higher degree of flow ability and compressibility than plain starch while 
maintaining its disintegrant properties; Emcompress, a free-flowing compressible 
dicalcium phosphate; a number of direct-compression sugars such as 
Nutab, Di- Pac , and Emdex; and a variety of sorbitol and mannitol products. 
The relatively minimal compression properties of spray-dried lactose were 
improved by enhanced agglomeration of smaller crystals and the problems 
of browning due to impurities in the mother liquid were corrected. At the 
same time major advances were made in tablet compression machinery, such 
as improved positive die feeding and precompression stages that facilitate 
direct-compression tableting. By the beginning of the 1980s, the excipients 
and machinery had become available to make possible the direct compression 
of the vast majority of tablets being manufactured. It is important to understand 
why this has not occurred. 
The simplicity of the direct-compression process is obvious. However, 
it is this apparent simplicity that has caused so many initial failures in 
changing formulations from wet granulation to direct compression. Direct 
compression should not be conceived as a simplified modification of the granulation 
process for making tablets. It requires a new and critical approach 
to the selection of raw materials, flow properties of powder blends. and 
effects of formulation variables on compressibility. During the wet granulation 
process the original properties of the raw materials are, to a great

Compressed Tablets by Direct Compression 197 
extent. completely modified. As a result, a new raw material, the granulation, 
is what is finally subjected to compression. Many inadequacies in the 
raw materials are covered up during the granulation step. This is not 
true in direct compression and therefore the properties of each and every 
raw material and the process by which these materials are blended become 
extremely critical to the compression stage of tableting. If direct compression 
is approached as a unique manufacturing process requiring new approaches 
to excipient selection, blending, and compressibility, then there 
are few drugs that cannot be directly compressed. If this is not done, 
failures are very likely to be encountered. 
The process of wet granulation is historically embedded in the pharmaceutical 
industry. It produces in a single process (although many steps may 
be involved) the two primary requisites for making a reproducible tablet 
compact (i. e , , fluidity and compressibility). The various methods of 
granulation as well as the steps involved in the process of granulation and 
the materials used are reviewed in an article by Record [4] and described 
extensively in Chapter 3 of this book. 
The advantages of the wet granulation process are well established and 
the advent of high-shear mixers and fluidized bed granulation and drying 
equipment has made wet granulation a more efficient process today than it 
was a quarter of a century ago. The advantages include the fact that it 
(a) permits mechanical handling of powders without loss of mix quality; 
(b) improves the flow of powders by increasing particle size and sphericity; 
(c) increases and improves the uniformity of powder density; (d) improves 
cohesion during and after compaction; (e) reduces air entrapment; (f) reduces 
the level of dust and cross-contamination; (g) allows for the addition 
of a liquid phase to powders (wet process only); and (h) makes hydrophobic 
surfaces hydrophilic. 
On the other hand, the granulation process is subject to a great many 
problems. Each unit process gives rise to its own specific complications. 
The more unit processes, the more chance for problems to occur. Granulation 
essentially involves the production of a new physical entity, the 
granule. It is therefore necessary to control and validate all the steps involved 
in making a new material (the granulation) and to assure that this 
final material is in fact reproducible. 
In addition to blending, problems include (a) type, concentration, rate 
of addition, distribution, and massing time of the binder solution; (b) 
effects of temperature, time, and rate of drying on drug stability and distribution 
during the drying process; and (c) granule size and segregation 
during the dry screening and subsequent final granulation blending. Each 
of these factors often involves a considerable effort in regard to both process 
and equipment validation. 
When taken as an aggregate, these problems can be imposing, and it is 
easy to see why direct compression has both a scientific and economic appeal. 
However, it certainly offers no panacea for the unwary or unthinking formulator.

198 Shangraw 
A. Advantages 
The direct-compression process assumes that all materials can be purchased 
or manufactured to specifications that allow for simple blending and tabletIng.
The most obvious advantage of direct compression is economy. It is 
safe to say that there would be a relatively minor interest in the process 
of direct-compression tableting if economic savings were not possible. 
Savings can occur in a number of areas, including reduced processing time 
and thus reduced labor costs I fewer manufacturing steps and pieces of 
equipment, less process validation I and a lower consumption of power. Two 
unit processes are common to both wet granulation and direct-compression 
tableting: blending and compression. Prior micronization of the drug may 
be necessary in either process. Although a number of pieces of equipment, 
such as granulators and dryers, are not needed in preparing tablets by 
direct compression, there may be a need for greater sophistication in the 
blending and compression equipment. However I this is not always the case. 
The most significant advantage in terms of tablet quality is that of 
processing without the need for moisture and heat which is inherent in most 
wet granulation procedures. and the avoidance of high compaction pressures 
involved in producing tablets by slugging or roll compaction. The unnecessary 
exposure of any drug to moisture and heat can never be justified j it 
cannot be beneficial and may certainly be detrimental. In addition to the 
primary problem of stability of the active ingredient I the variabilities encountered 
in the processing of a granulation can lead to innumerable tableting 
problems. The viscosity of the granulating solution-which is dependent 
on its temperature, and sometimes on how long it has been preparedcan 
affect the properties of the granules formed. as can the rate of addition. 
The granulating solution I the type and length of mixing. and the 
method and rate of wet and dry screening can change the density and particle 
size of the resulting granules, which can have a major effect on fill 
weight and compaction qualities. The drying cycles can lead not only to 
critical changes in equilibrium moisture content but also to unblending as 
soluble active ingredients migrate to the surfaces of the drying granules. 
There is no question that. when more unit processes are incorporated in 
production, the chances of batch-to-batch variation are compounded. 
Probably one of the least recognized advantages of direct compression 
is the optimization of tablet disintegration, in which each primary drug particle 
is liberated from the tablet mass and is available for dissolution. The 
granulation proeess , wherein small drug particles with a large surface area 
are PIgluedII into larger agglomerates, is in direct opposition to the principle 
of increased surface area for rapid drug dissolution. 
Disintegrating agents. such as starch. added prior to wet granulation 
are known to be less effective than those added just prior to compression. 
In direct compression all of the disintegrant is able to perform optimally, 
and when properly formulated, tablets made by direct compression should 
disintegrate rapidly to the primary particle state. However. it is important 
that sufficient disintegrant be used to separate each drug particle if ideal 
dissolution is to occur. One bioavailability advantage of making tablets by 
wet granulation has never been fully appreciated. The wetting of hydrophobic 
drug surfaces during the granulation step and the resulting film of

Compressed Tablets by Direct Compression 199 
hydrophilic colloid that surrounds each drug particle can certainly speed 
up the dissolution process providing that each one of the primary drug particles 
can be liberated from the granule. Although this is not as likely to 
occur in a tablet made by direct compression as in one made by granulation, 
it is possible to add a wetting agent in the dry blend of powders to 
enhance dissolution rates. Prime particle disintegration in direct-compression 
tablets depends on the presence of sufficient disintegrating agent and 
its uniform distribution throughout the tablet matrix. High drug concentrations 
can lead to cohesive particle bonding during compression with no 
interjecting layer of binder or disintegrating agent. 
Although it is not well documented in the literature, it would seem 
obvious that fewer chemical stability problems would be encountered in 
tablets prepared by direct compression as compared to those made by the 
wet granulation process. The primary cause of instability in tablets is 
moisture. Moisture plays a significant role not only in drug stability but 
in the compressibility characteristics of granulations. While some directcompression 
excipients do contain apparently high levels of moisture, this 
moisture in most cases is tightly bound either as water of hydration (e. g. , 
lactose monohydrate) or by hydrogen bonding (e.g., starch, microcrystalline 
cellulose) and is not available for chemical degradation. The role 
of moisture is discussed further under the description of individual excipients. 
One other aspect of stability that warrants increasing attention is the 
effect of tablet aging on dissolution rates. Changes in dissolution profiles 
are less likely to occur in tablets made by direct compression than in those 
made from granulations. This is extremely important as the official compendium 
now requires dissolution specifications in most solid dosage form 
B. Concerns 
On the basis of the distinct advantages listed above, it is difficult to understand 
why more tablets are not made by the direct-compression process. 
To understand this fully, one must have an appreciation of not only the 
technology, but the economics and regulation of the pharmaceutical industry. 
The technological limitations revolve mainly about the flow and bonding 
of particles to form a strong compact, and the speed at which this must be 
accomplished in an era of ever-increasing production rates. 
With an increased emphasis on dissolution and bioavailability, many drugs 
are commonly micronized. Micronization invariably leads to increased interp 
artteulate friction and decreased powder fluidity, and may also result in 
poor compressibility. Very often a decision has to be made as to whether 
to granulate a micronized powder-which may result in a longer dissolution 
time-or to directly compress a slightly larger particle size of the drug. 
In either case the decision should be based on in vivo blood studies as 
well as in vitro dissolution tests. 
The choice of excipients is extremely critical in formulating directcompression 
tablets. This is most true of the filler- binder, which often 
serves as the matrix around which revolves the success or failure of the 
formulation. Direct-compression filler-binders must possess both compressibility 
and fluidity. In most cases they are specialty items available 
from only one supplier and often cost more than comparable fillers used

200 Shangraw 
in granulations. In addition, there is a need to set functionality specifications 
on properties such as compressibility and fluidity, as well as on 
the more traditional physical and chemical properties. These specifications 
must be rigidly adhered to in order to avoid lot-to-Jet variations in raw 
materials, which can serioualy interfere with tableting qualities. This is 
as true of the drug substance as it is of the excipients. The costs of 
raw materials and raw material testing are thus higher in direct compression. 
However, this increased cost is often more than offset by the economies 
described earlier. 
Many active ingredients are not compressible in either their crystalline 
or their amorphous forms. Thus, in choosing a vehicle it is necessary to 
consider the dilution potential of the major filler-binder (I.e., the proportion 
of active ingredient that can be compressed into an acceptable compact 
utilizing that filler). Fillers-binders range from highly compressible materials 
such as microcrystalline cellulose to substances that have very low 
dilution capacity such as spray-dried lactose. It is not possible to give 
specific values for each filler because the dilution capacity depends on the 
properties of the drug itself. In some cases it is necessary to employ 
tablet presses with precompression capabilities in order to achieve an acceptable 
compact at a reasonable dilution ratio. 
Outside of compressibility failures, the area of concern most often mentioned 
by formulators of direct-compression tablets is content uniformity. 
The granulation process does lock active ingredients into place and, provided 
the powders are intimately dispersed before granulation and no dryinginitiated 
unblending occurs after wetting. this can be advantageous. Directcompression 
blends are subject to unblending in postblending handling steps. 
The lack of moisture in the blends may give rise to static charges that can 
lead to unblending. Differences in particle size or density between drug 
and excipient particles may also lead to unblending in the hopper or feed 
frame of the tablet press. 
The problems of unblending can be approached in either of two ways. 
The traditional approach involves trying to keep particle sizes or densities 
uniform. Ideally the vehicle itself (drug and/or filler binder) should incorporate 
a range of particle sizes corresponding as closely as possible to 
the particle size of the active ingredients. This range should be relatively 
narrow and should include a small percentage of both coarse and fine particles 
to ensure that voids between larger particles of drugs or filler excipients 
are filled by smaller sized particles. In SUch an approach, Avice! 
or Starch 1500 could be used to fill voids between larger excipient particles 
such as Emdex or Emcompress. The problem can also be solved by ordered 
blending which is discussed in detail later in the chapter. 
One other technical disadvantage of direct compression related to blending 
is the limitation in coloring tablets prepared in this manner. There is 
no satisfactory method for obtaining tablets of a uniformly deep color. 
However, it is possible through the use of highly micropulverized lakes 
preblended or milled with filler-s such as Starch 1500 or microcrystalline 
cellulose to obtain a wide variety of pastel shade tablets. 
Lubrication of direct-compression powder blends is, if anything, more 
complicated than that of classical granulations. In general the problems 
associated with lubricating direct-compression blends revolve around both 
the type and amount needed to produce adequate lubrication and the softening 
effects that result from Iu brication. It may be necessary to avoid

Compressed Tablets by Direct Compression 201 
the alkaline stearate lubricants completely in some direct-compression 
The most common approach to overcome the softening as well as hydrophobic 
effects of alkaline stearate lubricants is to substantially limit the 
length of time of lubricant blending often to as little as 2 to 5 min. In 
fact, it is probably advisable in all direct-compression blending not to include 
the lubricant during the majority of the blending period. Lubricants 
should never be added to direct-compression powder blends in a highshear 
mixer. In addition, the initial particle size of the lubricant should 
be carefully controlled. Another approach is to abandon the alkaline 
stearate lubricant and use hydrogenated vegetable oils such as Sterotex, 
Lubrrtab , and CompritoI. In such cases, higher concentrations are necessary 
than would be used to lubricate granulations of similar filler/drug 
mixtures with magnesium stearate. 
Outside of the limitations imposed by vehicle and formulation, there 
are economic and regulatory considerations necessary in making a decision 
to convert present products or to develop new products utilizing directcompression 
It is interesting to note that, except for spray- dried lactose, all directcompression 
exeipient.s were developed after the 1962 Kefauver-Harris 
amendmant to the Food, Drug and Cosmetic Act, which placed very strigent 
restrictions on dosage form as well as drug development. There is 
no question that this has led to a much more conservative approach to 
product development and formulation. Because of a 3- or 5-year or longer 
interval between formulation and marketing, many product development 
pharmacists hesitate to develop direct-compression formulations with unproven 
excipients. Of even greater uncertainty today is the physical 
specifications of the drug substance after its production has been scaled up 
to commercial proportions. In addition, there are increasing pressures to develop 
formulations that will be accepted internationally. In this respect, direct 
compression is much more widely used in the States than in Europe, although 
this situation is rapidly changing. Direct compression is more likely 
to be used by noninnovator companies because by the time patents have expired, 
the physical properties of the drug substance are more clearly defined. 
Complicating this picture in the past was the sampling of experimental 
direct-compression excipients that were never marketed commercially or 
were subsequently withdrawn, leading to instability in the specialty excipient 
marketplace. Lot -to-Iot variation in common direct-compression fillers 
commercially available today is rare. Of equal importance is the number 
of companies that have tried direct-compression formulations that failed 
when placed in full-scale production. In many cases this could be attributed 
to a failure to appreciate the complexities of the direct-compression 
technology, failure to set adequate specifications on raw materials, and 
failure of lot -to-Iot reproducibility in the drug substances, particularly 
high- dose active ingredients. 
In order to reduce the likelihood of raw material failure, it is advisable 
to set quality specifications on particle size, bulk density fluidity, and 
even compressibility. The latter can be easily done using a Carver press 
or single-punch machine under carefully prescribed conditions and determining 
the breaking strengths of resulting compacts. 
The major advantages and concerns for the wet granulation and directcompression 
processes are contrasted in Table 1.

202 Shangraw 
Table 1 Comparison of Direct-Compression and Wet Granulation Processes 
for Making Tablets 
Wet granulation 
Direct Compression 
Harder tablets for poorly compressible 
Excellent in most cases 
Potential problem for high-dose 
Many formulations may require a 
Cannot micronize high-dose drugs 
Larger with greater range 
Particle Size
Lower with narrower range 
Content Uniformity 
Massing and drying induced 
High Or low shear 
Less sensitive to lubricant softening 
and overblending 
Segregation may occur in mass 
transport, hopper, and feed 
Low shear with ordered blending 
Minimal blending with magnesium 
Often problems with granules 
Lower levels usually necessary 
1. Drug wetted during processing 
2. Drug dissolution from granules 
may be a problem 
3. Generally slower than direct 
Increase in equipment, labor, time, 
process validation, energy 
1. No wetting, may need surface 
active agent 
2. Dissolution may be slower if 
larger size drug crystals used 
3. Generally faster than wet 
Increase in raw materials and their 
quality control

Compressed Tablets by Direct Compression 203 
Table 1 (Continued) 
Wet granulation Direct compression 
Flexibility of Formulation 
Granulation covers raw material 
1. Problems with heat or moisture 
2. Dissolution rate may decrease 
with time 
Properties of raw materials must 
be carefully defined 
1. No heat or moisture added 
2. Dissolution rate rarely 
May be faster 
Less dusty 
Attitude of Equipment Suppliers 
Very negative 
T ableting Speed 
May require lower speed 
More dusty 
Deep or pastel (dyes or lakes) Pastel only (lakes only) 
A. General Considerations 
Direct-compression excipients, particularly flllez--bdnder-s , are specialty ex~ 
cipients. In most cases they are common materials that have been modified 
in the chemical manufacturing process to impart to them greater fluidity 
and compressibility. The physical and chemical properties of these specialty 
products are extremely important if they are to perform optimally. It is 
most important for the direct-compression formulator to understand that 
there is no chance to cover up flaws in raw materials in direct compression 
as there is in the wet granulation process. 
Many factors influence the choice of the optimum direct-compression 
filler to be used in a tablet formulation. These factors vary from primary 
properties of powders (particle size. shape. bulk density. solubility) to 
characteristics needed for making compacts (flowability and compressibility) 
to factors affecting stability (moisture). to cost. availability. and governmental 
acceptability. It is extremely important that raw material specifications 
be set up that reflect many of these properties if batch-to-batch

Table 2 Factors Influencing Choice of Direct-Compression Fillers 
1. Compreasibilttyf 
a. Alone 
b. Dilution factor or capacity 
c. Effect of lubricants, glidants, disintegrants 
d. Effect of reworking 
2. Flowabilitya 
a.' Alone 
b. In the finished formulation 
c. Need for glidant 
3. Particle Sizea and Distribution 
a. Effect on flowability 
b. Effect on compressibility 
c. Effect on blending 
d. Dust problems 
4. Moisture Content and Typea 
a. Water of hydration (lactose, dextrose, dicalphosphate) 
b. Bound and free moisture 
c. Availability for chemical degradation 
d. Effect on compressibility 
e. Hydroscopicity 
5. Bulk Densitya 
ti volume of tablet a. Compression ra 10 = --'--------bulk 
volume of powder 
b. Effect of handling and blending 
6. Compatibility with Active Ingredient 
a. Moisture 
b. pH 
c. Effect on assay 
7. Solubility (in GI Tract) 
a. Rate of dissolution 
b. Effect of pH 
8. Stability of Finished Tablets 
a. Color 
b. Volume 
c. Hardness 
9. Physiological Inertness 
a. Toxicity 
b. Reducing sugar 
c. Osmotic effect 
d. Taste and mouth-feel (if appropriate) 
10. Cost and Availability 
11. Governmental Acceptability 
a. United States and foreign countries 
b. Master File 
C. GRAS status 
d. Compendia! standards (N.F.) 
~eed to set purchase specifications for each lot of raw material. 

Compressed Tablets by Direct Compression 205 
manufacturing uniformity is to be assured. This is particularly true in 
the case of the filler-binders because they often make up the majority of 
the tablet weight and volume. However, this fact is still not fully appreciated 
by pharmaceutical formulators and production personnel. A list of 
factors involved in the choice of a filter-binder can be found in Table 2. 
Most all of the classic tablet fillers have been modified in one way or 
another to provide fluidity and compressibility. In viewing the scanning 
electron photomicrograp hs of the variou s direct-compression filler-binders, 
one is taken with the fact that none of the products consist of individual 
crystals. Instead, all of them are actually minigranulations or agglomerations 
that have been formed in the manufacturing process by means of cocrystallization, 
spray drying, etc. The resulting material thus is able to 
deform plastically in much the same manner as the larger particle size granules 
formed during the traditional wet granulation process. The key to 
making any excipient or drug directly compressible thus becomes obvious 
and the possibility of making all tablets by direct compression appears to 
be within the scope of present technology. 
B. Soluble Filler-Binders 
Spray-dried lactose is the earliest and still one of the most widely used 
direct-compression fillers. It is one of the few such excipients available 
from more than a single supplier. In spite of many early problems, this 
material revolutionized tableting technology. 
Coarse and regular grade sieved crystalline fractions of a.-lactose monohydrate 
have very good flow properties but lack compressibility. However 
spray drying produces an agglomerated product that is more fluid and compressible 
than regular lactose [1]. 
In the production of spray-dried lactose, lactose is first placed in an 
aqueous solution which is treated to remove impurities. Partial crystallization 
is then allowed to occur before spray-drying the slurry. As a result 
the final product contains a mixture of large a.-monohydrate crystals and 
spherical aggregates of smaller crystals held together by glass or amorphous 
material. The fluidity of spray-dried lactose results from the large 
particle size and intermixing of spherical aggregates. The compressibility 
is due to the nature of the aggregates and the percentage of amorphous 
material present and the resulting plastic flow, which occurs under compaction 
The problem of compressibility of spray-dried lactose is still real and 
troublesome. The compressibility of spray-dried lactose is borderline, and 
furthermore, it has relatively poor dilution potential. Spray-dried lactose 
is an effective direct-compression filler when it makes up the major portion 
of the tablet (more than 80%), but it is not effective in diluting high-dose 
drugs whose crystalline nature is, in and of itself, not compressible. 
Furthermore, spray-dried lactose does not lend itself to reworking because 
it loses compressibility upon initial compaction. 
Spray-dried lactose has excellent fluidity, among the best for all directcompression 
fillers. It contains approximately 5% moisture, but most of 
this consists of water of hydration. The free surface moisture is less than

206 Shangraw 
0.5% and does not cause significant formulation problems. It is relatively 
nonhygroscopic . 
Spray-dried lactose is available from a number of commercial sources 
in a number of forms [5]. Because the processing conditions used by 
different manufacturers may vary. all spray-dried lactoses do not necessarily 
have the same properties particularly in terms of degree of agglomeration, 
which influences both fluidity and compressibility. Alternative sources 
of supply should be validated. as is true of all direct-compression fillers. 
When spray-dried lactose was first introduced. two major problems existed. 
The one that received the most attention was that of browning [2}. 
This browning was due to contaminants in the mother liquid, mainly 5-hydroxyfurfural, 
Which was not removed before spraying. This browning 
reaction was accelerated in the presence of basic amine drugs and catalyzed 
by tartrate. citrate. and acetate ions [6]. Although the contaminants are 
now removed during the manufacturing process in most commercial products. 
the specter of browning still remains. However, at the present time. there 
appears to be no more danger of browning in spray-dried lactose than in 
any other form of lactose. 
After many abortive attempts to improve on spray-dried lactose, a much 
more highly compressible product was introduced in the early 1970s [7]. 
This product, called Fast- Flo lactose, consists mainly of spherical aggregates 
of microcrystals. These microcrystals are lactose monohydrate, and 
they are held together by a higher concentration of glass than is present 
in regular spray-dried lactose. During the manufacturing process the 
microcrystals are never allowed to grow but are agglomerated into spheres 
by spray drying. Because it is much more compressible, it has replaced 
regular spray-direct lactose in many new direct-compression formulations. 
Because of the spherical nature of the spray-dried aggregates, Fast- 
Flo lactose is highly fluid. It is nonhygroscopic and, as is the case with 
most spray-dried lactose. contaminants that could lead to browning are removed 
in the manufacturing process. Tablets made from Fast-Flo lactose 
are three to four times harder than those made from regular spray-dried 
lactose when compressed at the same compression force. An agglomerated 
form of lactose that is more compressible than spray-dried but less compressible 
than Fast-Flo lactose is marketed under the name Tabletose. 
Anhydrous lactose is a free-flowing crystalline lactose with no water 
of hydration, first described in the literature in 1966 [8]. The most common 
form of anhydrous lactose is produced by crystallization above 93C 
which produces the 13 form. This is carried out on steam-heated rollers, 
the resultant cake being dried, ground, and sieved to produce the desired 
size. It is available in a white crystalline form that has good flow properties 
and is directly compressible. Its compressibility profile (compression 
force versus hardness) is similar to that of Fast-Flo lactose. Anhydrous 
lactose can be reworked or milled with less loss of compactability than occurs 
with other forms of lactose. However. anhydrous lactose contains a relatively 
high amount of fines (15 to 50% passes through a 20D-mesh screen), so that 
its fluidity is less than optimal. The use of a glidant SUch as Cab-O-Sil 
or Syloid is recommended if high concentrations are included in a formulation. 
At high relative humidities anhydrous lactose will pick up moisture, forming 
the hydrated compound. This is often accompanied by an increase in 
the size of the tablets if the excipient makes up a large portion of the total 
tablet weight. At a temperature of 45C and a relative humidity of 70%. 
plain anhydrous lactose tablets will increase in size by as much as 15% of

Compressed Tablets by Direct Compression 207 
their original volume. Much has been made of the fact that anhydrous 
lactose contains less moisture than regular lactose and thus is a better 
filler' for moisture- sensitive drugs. In fact, the surface moisture of the 
anhydrous and hydrous forms is about the same (0.5%) and the water of 
hydration does not play a significant role in the decomposition of active 
ingredients. Anhydrous lactose possesses excellent dissolution properties, 
certainly as good as, if not better than, a-lactose monohydrate. 
Anhydrous lactose possesses excellent dissolution properties which is 
due in part to the fact that it is predominantly S-lactose. The intrinsic 
dissolution rate is considerably faster than a-lactose monohydrate. Lactose 
N.F., anhydrous, direct tableting, is available in the United States from 
Sheffield products while both high- S- and high-a-content anhydrous lactose 
are produced by DMV in Europe. Dehydration of the hydrous form must 
occur above 130C in order to obtain stable anhydrous crystals needed for 
pharmaceutical use. A number of excellent articles on the various types 
of lactose and their tableting properties have been published by Lerk, 
Bolhuis, and coworkers [9-14]. 
Sucrose has been extensively used in tablets both as a filler, usually in 
the form of confectioners sugar, and in the form of a solution (syrup), 
as a binder in wet granulations. Attempts to directly compress sucrose 
crystals have never been successful. but various modified sucroses have 
been introduced into the direct-compression marketplace. One of the first 
such products was Di-Pac , which is a cocrystallization of 97% sucrose and 
3% highly modified dextrins [15]. Each Di-Pac gnanule consists of hundreds 
of small sucrose crystals "glued" together by the dextrin. Di-Pae has 
good flow properties and needs a glidant only when atmospheric moisture 
levels are high (greater than 50% relative humidity). It has excellent color 
stability on aging, probably the best of all the sugars. 
Di-Pac is a product that points out the need for setting meaningful 
specifications in purchasing raw materials for direct compression. The concentr-
ation of moisture is extremely critical in terms of product compressibility. 
Compressibility increases rapidly in a moisture range of 0.3 to 
0.4%, plateaus at a level of 0.4 to 0.5%. and rises again rapidly up to 0.8% 
when the product begins to cake and lose fluidity [16]. The moisturecompressibility 
profile of Di-Pac is closely related to the development of 
monomolecular and multimolecular layers of moisture on both the internal 
and external surfaces of the sucrose granules-a process that increases 
hydrogen bonding on compression. The dilution potential of Di-Pae and 
most other sucroses is only average. ranging from 20 to 35% active ingredients. 
While a moisture concentration of 0.4% is probably optimal for most 
pharmaceuticals. material of high moisture content is extremely advantageous 
when making troches or candy tablets. Interestingly, as moisture levels 
increase, lubricant requirements decrease. Tablets containing high concentrations 
of Di-Pac tend to harden slightly (1- to 2-kg units) during the 
first hours after compression, or when aged at high humidities and then 
dried. This is typical of most direct-compression sucroses or dextroses. 
Like all direct-compression sucroses , the primary target products are 
chewable tablets, particularly where artificial sweeteners are to be avoided. 
Both the process for making cocrystallized sucrose products and their properties 
are described in an article by Rizzuto et al. [17].

208 Shangraw 
Nutab is a directly compressible sugar consisting of processed sucrose, 
4% invert sugar (equimolecular mixture of levulose and dextrose), and 0.1 
to 0.2% each of cornstarch and magnesium stearate [181. The latter ingredients 
are production adjuncts in the granulation process by which the product 
is made and are not intended to interject any disintegrant or lubricant 
activity in a final tablet formulation. NuTab has a relatively large particle 
size distribution which makes for good fluidity but could cause blending 
problems if cofillers and drugs are not carefully controlled relative to particle 
size and amounts. In formulations NuTab has poor color stability 
relative to other direct-compression sucroses and Iactoses . 
One of the most dramatic modifications of natural raw materials for improving 
tableting characteristics is directly compressible dextrose marketed under 
the name Emdex [19J. This product is spray-crystallized and consists of 
90 to 92% dextrose, 3 to 5% maltose, and the remainder higher glucose 
polysaccharides. It is available as both an anhydrous and a hydrous product 
(9% moisture). Reports indicate that the anhydrous form is slightly 
more compressible than the monohydrate; but the compressibility of both 
is excellent, being second only to microcrystalline cellulose when not diluted 
with drugs or other excipients. The most widely used product is the monhydrate 
and the water of hydration does not appear to affect drug stability. 
At approximately 75% relative humidity both forms of Emdex become quite 
hygroscopic, particularly if they have been milled or sheared on the surface 
of a die table. Above 80% relative humidity both products liquefy. 
Tablets produced from Emdex show an increase in hardness of approximately 
2 kg at all levels of initial hardness up to 10 kg. The increase occurs in 
the first few hours after compression with no further significant hardening 
on long-term storage under ambient conditions. However, hardness increases 
do not result in significant changes in rates of dissolution. 
Emdex possesses the largest particle size of all the common direct-compression 
excipients. Blending problems can occur if blends of other smaller 
particle size excipients are not used to fill in voids. This filler lends itself 
to ordered blending, where the micronized drug is first blended with 
the large particle size Emdex, before other excipients are added to the 
blender. The micronized drug becomes lodged in the pores on the surfaces 
of the large spheres and are apparently held in place with sufficient attractive 
force to prevent dislodging during subsequent blending operations. 
Sorbitol is one of the most complex of all direct-compression fillers. It is 
available from a number of suppliers in various direct-compression forms. 
However, sorbitol exists in a number of polymorphic crystalline forms as 
well as an amorphous form. Failure of many suppliers to fully appreciate 
the ramifications of these crystalline forms on both compressibility and 
stability has caused major problems among users. The less stable (). and 
S) polymorphic forms of sorbitol will convert to the more stable form (y), 
which often results in dendritic growth (small, hairlike crystals). This 
causes a caking of particles and is accentuated by the presence of moisture. 
More stable products such as Sorbitol 834 and NeoSorb 60, consisting almost 
solely of the y form, are now available and overcome most of the stability

Compressed Tablets by Direct Compression 209 
problems. However, all y-sorbitols are not crystallized in the same way 
and thus still have different compressibilities and lubricant requirements. 
At the present time interchange of one directly compressible form for 
another is not recommended without some validation of processing characteristics. 
The complexities of sorbitol and the modification of its crystalline 
structure to influence tableting properties are described by DuRoss 
[20], while an evaluation of ascorbic acid and gamma sorbitol tablets is 
presented by Guyot-Hermann and Leblanc [21]. 
Sorbitol is widely used as the sole ingredient in "sugar-free" mints 
and as a vehicle in chewable tablets. It forms a relatively hard compact, 
has a cool taste and good mouth-feel , However, it is hygroscopic and will 
clump in the feed frame and stick to the surfaces of the die table when tableted 
at humidities greater than 50%. 
Lubricant requirements increase when the moisture content of the sorbitol 
drops below 0.5% or exceeds 2%. 
Recently, there has been an increased interest in direct-compression mannitol. 
Mannitol does not make as hard a tablet as sorbitol but is less sensitive 
to humidity. Mannitol is widely used in the direct compression of 
reagent tablets in clinical test kits where rapid and complete solubility 
is required and can be lubricated sufficiently for this purpose using 
micronized polyethylene glycol 6000. One company has developed a highly 
specialized technique to produce beads of sensitive biological materials and 
mannitol or sorbitol for direct compression [22,23]. Its use as a filler in 
chewable tablets is limited by its cost, although its cool mouth feel is 
highly attractive. Mannitol also exists in a number of polymorphic forms 
and this phenomenon should be explored if a lot of mannitol behaves in a 
peculiar fashion. Debord et al , [24] tested four polymorphic forms of 
mannitol, two of which they obtained in pure state. Different forms were 
shown to have different compression characteristics. 
A free-flowing agglomerated maltodextrin is available for direct-compression 
tableting under the name Maltrin. The product is highly compressible. 
completely soluble. and has very low hygroscopic characteristics. 
C. Insoluble Filler-Binders 
One of the most widely used tablet excipients starch. does not in its natural 
state possess the two properties necessary for making good compacts: 
compressibility and fluidity. There have been many attempts to modify 
starch to improve its binding and flow properties. The only modification 
of starch that has received widespread acceptance in direct compression 
is Starch 1500. Starch 1500 is more fluid than regular starch and meets 
the specifications for pregelatinized starch, N.F. Starch 1500 consists of 
intact starch grains and ruptured starch grains that have been partially 
hydrolyzed and subsequently agglomerated [25]. It has an extremely high 
moisture content (12 to 13%), but there is little indication that this moisture 
is readily available to accelerate the decomposition of moisture-sensitive 
drugs [26].

210 Shangraw 
Although Starch 1500 will readily compress by itself. it does not form 
hard compacts. Its dilution potential is minimal. and it is not generally 
used as the filler-binder in direct compression, but as a direct-compression 
filler disintegrant. The major advantage of Starch 1500 is that it retains 
the disintegrant properties of starch without increasing the fluidity and 
compressibility of the total formulation, which is not the case with plain 
starch. Because Starch 1500, like all starches, deforms elastically when a 
compression force is applied, it imparts little strength to compacts. As few 
clean surfaces are formed during compaction, lubricants, particularly the alkaline 
stearate lubricants, tend to dramatically soften tablets containing high 
concentrations of Starch 1500, Lubricants such as stearic acid or hydrogenated 
vegetable oils are preferred in such formulations. 
The first widespread use of cellulose in tableting occurred in the 1950s 
when a floc cellulose product, Solka- Floc I was introduced as a filler disintegrant. 
Solka-Floc consists of cellulose that has been separated from 
wood by digestion and formed into sheets that are mechanically processed 
to separate and break up individual fibers into small pieces. This converts 
the cellulose into a free- flowing powder. However, this material has poor 
fluidity and compressibility. and is not used as a direct-compression excipient. 
The most important modification of cellulose for tableting was the isolation 
of the crystalline portions of the cellulose fiber chain. This product, 
microcrystalline cellulose (Avicel), was introduced as a direct-compression 
t ableting agent in the early 1960s and stands today as the single most important 
tablet excipient developed in modern times [3]. Although it was 
developed with no though of tableting in mind, its properties are close to 
optimal. Microcrystalline cellulose is derived from a special grade of purified 
alpha wood cellulose by severe acid hydrolysis to remove the amorphous 
cellulose portions, yielding particles consisting of bundles of needlelike 
microcrystals. Microcrystalline cellulose for direct-compression tableting 
comes in a number of grades, the most widely used of which is PH 101, 
which was the original product, and PH 102, which is more agglomerated 
and possesses a larger particle size, resulting in slightly better fluidity 
but with no significant decrease in compressibility. 
Microcrystalline cellulose is the most compressible of all the direct-compression 
fillers and has the highest dilution potential. This can be explained 
by the nature of the microcrystalline particles themselves, which 
are held together by hydrogen bonds in the same way that a paper sheet 
or an ice cube is bonded [27]. Hydrogen bonds between hydrogen groups 
on adjacent cellulose molecules account almost exclusively for the strength 
and cohesiveness of compacts. When compressed, the microcrystalline cellulose 
particles are deformed plastically due to the presence of slip planes 
and dislocations on a microscale, and the deformation of the spray-dried 
agglomerates on a macroscale. A strong compact is formed due to the extremely 
large number of clean surfaces brought in contact during the plastic 
deformation and the strength of the hydrogen bonds formed. 
Other factors are important in the ability of a comparatively small 
amount of microcrystalline cellulose to bind other materials during compaction, 
the low bulk density of the microcrystalline cellulose, and the broad 
range of particle sizes. An excipient with a low bulk density will exhibit 
a high dilution potential on a weight basis I and the broad particle size

Compressed Tablets by Direct Compression 211 
range provides optimum packing density and coverage of other excipient 
Microcrystalline cellulose has an extremely low coefficient of friction 
(both static and dynamic) and therefore has no lubricant requirements itself. 
However, when more than 20% of drugs or other excipients are added, 
lubrication is necessary. Because it is so compressible, microcrystalline 
cellulose generally withstands lubricant addition without significant softening 
effects. However, when ,.high concentrations (greater than 0.75%) of 
the alkaline stearate lubricants are used, and blending time is long, the 
hardness of tablets compressed at equivalent compression forces is lower. 
Because of cost and density considerations, microcrystalline cellulose 
is generally not used as the only filler in a direct-compression tablet but 
is more often found in concentrations of 10 to 25% as a filler-binder-disintegrant. 
Although it is not as effective a disintegrant as starch in equivalent 
concentrations, it can be used as the only disintegrant at levels of 20% 
or higher and has an additive effect with starch at lower levels. Hard 
compacts of microcrystalline cellulose disintegrate rapidly due to the rapid 
passage of water into the compact and the instantaneous rupture of hydrogen 
bonds. The fluidity of microcrystalline cellulose is poor compared to 
that of most other direct-compression fillers because of its relatively small 
particle size. However, comparisons with other direct-compression fillers 
based on a weight per unit time flow through an orifice are misleading due 
to its inherently low-bulk density [28]. A comparison of the relative volumetric 
and gravimetric flow rates of typical direct-compression fillers can be 
seen in Table 3. Small amounts of glidant are recommended in many formulations 
containing high concentrations of microcrystalline cellulose. 
Tablets made from higher concentrations of microcrystalline cellulose 
soften on exposure to high humidities due to moisture pickup and loosening 
of interparticulate hydrogen bonds. This softening is often reversible when 
tablets are removed from the humid environment. Cycling of temperature 
and moisture over a period of time can cause both increases or decreases 
of equilibrium hardness, depending on the total formulation. 
Because microcrystalline cellulose is highly oompressible, self-lubricating, 
and a disintegrant, attempts have been made to use it as the only fillerbinder 
in tablets containing drugs with low doses. It has been found that 
formulations containing more than 80% microcrystalline cellulose may slow 
the dissolution rates of active ingredients having low water solubility. Apparently, 
the small particles get physically trapped between the deformed 
microcrystalline cellulose particles, which delays wetting and dissolution. 
This phenomenon can be easily overcome by adding portions of water-soluble 
direct-compression excipients such as Fast-Flo lactose. 
During the middle 1980s, a number of cellulose products were introduced 
into the marketplace to compete with Avicel. These products represent a 
continuum from floc to crystalline celluloses , some of which meet N. F. specifications 
for microcrystalline cellulose (Le., Emcocel). Personen and Paronen 
[29] compared the crystallinity, particle size, densities, flow, and binding 
properties of Emcocel and Avicel PH 101. 
However, the most complete comparative evaluation of microcrystalline 
cellulose products was conducted by Doelker et al . [30]. They studied 
the tableting characteristics of N.F. grade microcrystalline celluloses produced 
by seven manufacturers. The powders were examined for moisture 
content, particle size, densities, flow, and tableting properties (on an instrumented 
press) by measuring diametral crushing force of the compacts.

212 Shangraw 
Table 3 Volumetric and Gravimetric Comparative Flow Rates of Selected 
Direct- Compression Fillers 
Volumetric flow 
rate based on 
Poured Gravimetric poured bulk 
bulk density flow rate density 
Filler-binder (g cm-3) (kg min-I) (L in. -I) 
Microcryst alline 0.314 1. 300 4.140 
Powdered 0.531 1. 499 2.823 
Pregelatinized 0.589 1. 200 2.037 
Hydrous lactosed 0.650 2.200 3.385 
Compressible 0.694 3.747 5.399 
Dibasic calcium 0.933 4.300 4.609 
aAvicel PH-102, FMC Corp. Philadelphia, Pennsylvania. 
b Elcema G-250, Degussa Corp., Teterboro, New Jersey. 
CStarch 1500, Colorcon, Inc., West Point, Pennsylvania. 
dFast-Flo, Foremost Whey Products, Barzboo , Wisconsin. 
eDi-Pac, Amstar Corp., New York, New York. 
fDi-Tab, Stauffer Chemical Co , , Westport, Connecticut. 
Source: From Pharm. Tech., 7(9}, 94 (l983). 
Great differences in packing and tableting properties and in sensitivity to 
the addition of a lubricant were generally observed between products from 
various manufacturers. In contrast, lot-to-lot variability was quite acceptable. 
Using an empirical scale, the authors rated the various products 
and found Avicel and Emcocel to overall outperform other products. However, 
the functionality of microcrystalline cellulose depends as much on 
physical form as it does on crystalline content. Equivalence of microcrystalline 
products varies with desired functionality and substitutions of one 
product for another must be validated. Often less compressible microcrystalline 
cellulose can be substituted for Avieel with acceptable results because 
products may have been overly formulated with microcrystalline cellulose to 
begin with. 
It should be remembered that the effectiveness of microcrystalline cellulose 
as a binder decreases as moisture is added to it in processing. Thus 
microcrystalline cellulose is effective as a binder in direct compression, 
slugging, roller compaction, or when added to a granulation in the freeflowing 
mix directly before compression. Its binding advantages in granulation 
decrease with an increase in water addition.

Compressed Tablets by Direct Compression 213 
Another form of cellulose advocated for direct compression is microfine 
cellulose, (Elcema). This material is a mechanically produced cellulose 
powder which also comes in a granular grade (G-250), which is the only 
form that possesses sufficient fluidlty to be used in direct compression. 
Microfine cellulose is a compressible, self-disintegrating, antiadherent form 
of cellulose that can be made into hard compacts. However, unlike microcrystalline 
cellulose, it possesses poor dilution potential, losing its compressibility 
rapidly in the presence of noncompressible drugs. It is not a 
particularly effective dry binder due to the large particle size of the G-250 
granules and the resistance to fracture under compression. Microfine cellulose 
forms few fresh or clean surfaces during compression because of 
the lack of slip planes and dislocations in the cellulose granules. Thus 
little interparticulate binding occurs, and sufaces "contaminated" by lubricant 
during mixing show little inclination to form firm compacts. 
Inorganic Calcium Salts 
The most widely used inorganic direct-compression filler is unmilled diealcium 
phosphate, which consists of free-flowing aggregates of small microcrystals 
that shatter upon compaction. This material is available in a 
tableting grade under the names Emcompress or DiTab. Dicalcium phosphate 
is relatively inexpensive and possesses a high degree of physical and 
chemical stability. It is nonhygroscopic at a relative humidity of up to 
80%. Dioaleium phosphate in its directly compressible form exists as a dihydrate. 
Although this hydrate is stable at room and body temperature, 
it will begin to lose small amounts of moisture when exposed to temperatures 
of 40 to 60C [31]. This loss is more likely to occur in a humid environment 
than a dry environment. This anomaly is theorized to occur because 
at low humidities and high temperatures, the outer surfaces of the particles 
lose water of hydration and become case-hardened, preventing further loss. 
In a humid environment the loss continues to occur. When combined with a 
highly hygroscopic filler like microcrystalline cellulose, the loss of moisture 
may be sufficient to cause a softening of the tablet matrix due to weakening 
of the Interpartleulate bonds and to accelerate decomposition of moisturesensitive 
drugs like vitamin A. 
The fluidity of dicalcium phosphate is good, and glidants are generally 
not necessary. While it is not as compressible as microcrystalline cellulose 
and some sugars (Fast-Flo lactose, Emdex) , it is more compressible than 
spray-dried lactose and compressible starch. It apparently deforms by 
brittle fracture when compressed, forming clean bonding surfaces. Lubricants 
exert little softening effect on compacts. 
Because it is relatively water-insoluble, tablets containing 50% or more 
of dicalcium phosphate disintegrate rapidly. Diealcium phosphate does dissolve 
in an acidic medium, but it is practically insoluble in a neutral or 
alkaline medium. Therefore, it is not recommended for use in high concentrations 
in combination with drugs of low water solubility. This is of 
particular concern in formulating tablets that may be used in geriatric 
patients where the incidence of achlorhydria is significant. 
Dicalcium phosphate dihydrate is slightly alkaline with a pH of 7.0 to 
7.3, which precludes its use with active ingredients that are sensitive to 
even minimal amounts of alkalinity. Tricalcium phosphate (TriTab) is less 
compressible and less soluble than dicaIcium phosphate but contains a 
higher ratio of calcium ions [32]. Calcium sulfate, dihydrate N.F., is 
also available in direct-compression forms [Delaflo, Compactrol}.

214 Shangraw 
Cel-O-Cal is the first significant direct-compression tablet filler specifically 
designed to combine the advantages of dissimilar materials by the 
method of coprocessing. It consists of 30 parts of microcrystalline cellulose 
and 70 parts of anhydrous calcium sulfate coprocessed in a spray dryer. 
It combines the compressibility and dislntegrant advantages of microcrystalline 
cellulose with the cost advantages of calcium sulfate. The product is 
significantly more compressible than a physical mixture of its component 
parts and produces tablets of much lower friability. It is also less subject 
to lubricant softening effects due to its larger particle size. Because 
Cel-O-Cal is composed of two substances that are not water-soluble, care 
should be taken in using it in formulation of drugs with low water solubility 
particularly if the product is to be wet -granulated. 
Calcium Carbonate 
Calcium carbonate has been used in the past as a tablet filler even though 
it does have a significant pharmacological effect (antacid). It is available 
from a number of suppliers in directly compressible forms. There has been 
a renewed interest in calcium carbonate in the United States because of 
its use as a nutritional supplement in the prophylaxis of osteoporosis. 
Although its effectiveness for this condition has been questioned, numerous 
calcium supplements, ineluding combinations with vitamin D and multivitamins 
are being marketed. Calcium carbonate is available in a number of 
forms including precipitated, ground oyster shells and mined limestone. 
There is no evidence that anyone of these sources provides a nutritionally 
superior product and all have similar dissolution profiles. They do differ 
in terms of degree of whiteness, particle size, and impurities. Calcium 
carbonate has been coprocessed with various binders to make it directly 
compressible. The solUbility of calcium carbonate does depend on pH. 
The effectiveness of calcium carbonate as a source of calcium in achlorhydric 
patients has been questioned. 
On the other hand. calcium carbonate is much more soluble than either 
dicalcium phosphate, tricalcium phosphate, or calcium sulfate. The use of 
these other substances even in normal patients would appear to be even 
less justified. 
A glossary of direct compression excipients, trade names, and suppliers 
can be found at the end of the chapter. 
More than in any other type of tablets, successful formulations of directcompression 
tablets depend on careful consideration of excipient properties 
and optimization of the compressibility, fluidity, and lubricability of powder 
blends. The importance of standardizing the functional properties of the 
component raw materials and the blending parameters cannot be overstressed. 
Preformulation studies are essential in direct-compression tableting 
even for what would appear to be a simple formulation. 
A. Compressibility 
Formulation should be directed at optimizing tablet hardness without applying 
excessive compression force while at the same time assuring rapid tablet

Compressed Tablets by Direct Compression 215 
disintegration and drug dissolution. In those cases where the drug makes 
up a relatively minor proportion of the tablet, this is usually no problem, 
and concern revolves around homogeneous drug distribution and content 
uniformity. Often much simpler excipient systems can be utilized, and 
factors such as relative excipient costs become more important. In those 
cases where the drug makes up the greater part of the final tablet weight, 
the functional properties of the active ingredient and the type and concentration 
of the excipient dominate the problem. Often the decision resolves 
about the question of what is the least amount of excipient necessary to 
form an acceptable and physically stable compact. In regard to the active 
ingredient it is important to determine the effect of particle size on compressibility 
as well as the effect of crystalline form (crystalline or amorphous) 
on compressibility. It may be necessary to granulate the active 
ingredient by slugging to improve compressibility and increase density. 
The most effective dry binder is microcrystalline cellulose. It can add 
significant hardness to compacts at levels as low as 3 to 5%. It should 
always be considered first if the major problem in the formulation is tablet 
hardness or friability. It has been used at levels as high as 65% to bind 
active ingredients with extremely poor compressibility characteristics. No 
other direct-compression excipient acts as well as a dry binder in low concentrations. 
The compressibilities of varying fillers have been discussed 
as they relate to individual substances. Most disintegrating agents (such 
as starch) or glidants have negative effects on compressibility, although 
compressible starch is better than plain cornstarch. 
A comparison of the relative compressibilities of various direct-compression 
fillers using magnesium stearate and stearic acid as lubricants is presented 
in Figures 1 and 2. As can be seen, microcrystalline cellulose is 
by far the most compressible of the substances tested. Magnesium stearate 
causes a softening of compacts to the point that Starch 1500 cannot be tableted. 
However, the relative compressibility of the fillers remains constant. 
'" B 
'" !~
6 ~ 
0 400 800 1200 
Compression Force (kg) 
.... AvicelpH 101 
... Nu-Tab 
.... Di-Pac 
... Anhyd. Lac. 
_ Fast Flo Lac. 
-0- Emcompress 
.... Elcema G250 
- Starch 1500 
Figure 1 Excipient compressibility with 2% stearic acid as lubricant.

l 8 
=II 6 c: 
III :::t: 
0 400 800 1200 1600 
Shang raw 
... Avicel pH 101 
... Fast Flo Lac. 
...... Anhyd Lac 
.... Di-Pac 
... Nu-Tab 
-0- Emcompress 
... Elcema G250 
Compression Force (kg) 
Figure 2 Excipient compressibility with 0.75% magnesium stearate as 
It is possible to compare the relative compressibility of a variety of 
direct-compression lactoses in a similar manner (Fig. 3). As can be seen, 
there can be as great as a twofold difference in the compressibility of two 
different forms at equivalent compression forces. 
It might be expected that compressibility properties would be additive 
(Le . , that a mixture of microcrystalline cellulose and spray-dried lactose 
would have a compressibility profile of some proportionate value between 
those of the individual ingredients). For instance, Lerk et al , [33] showed 
an additive effect between most lactose fillers when they were combined 
with other lactoses or microcrystalline cellulose. However, an antagonistic 
behavior was demonstrated by blends. of fast-dissolving vehicles such as 
dextrose or sucrose with cellulose or starch products. For instance, almost 
all combinations of microcrystalline cellulose and compressible dextrose 
gave poorer compressibility profiles and longer disintegration times than 
either ingredient alone. Bavitz and Schwartz [34] showed essentially additive 
effects in hardness when blending fillers, but their work did not 
include either sucrose or dextrose. 
Almost all disintegrating agents retard compressibility as well as fluidity 
due to particle size. In order to have optimal disintegration into primary 
particles, it is desirable to have the particle size of the disintegrating agent 
as small as possible, preferably smaller than that of the active ingredient. 
This is not always possible. 
One of the major advances in the development of direct-compression 
technology and its adoption by industry has been the introduction of the 
"euper-dlaintegrurrts , II These agents, which include Croscarmellose N.F. 
(AcDiSol), Crospovidone N.F. (Polyplasdone XL), and sodium starch 
glycolate N. F. (Explotab and Primoge}) I allow for faster disintegration of 
tablets, and lower use levels, therefore minimizing the softening effect

Compressed Tablets by Direct Compression 217 
and fluidity problems encountered when high levels of starch are used. 
Fortunately. direct-compression formulations generally do not require as 
high a disintegrant concentration as wet granulation because the problem 
of intragranular disintegration does not exist. 
As direct-compression blends may not possess ideal compressibility. 
operational problems may be reduced by the use of one or two precompression 
stages or use of large compression rolls. 
It is generally concluded that direct-compression formulations are less 
compressible than wet granulation formulations. Obviously. this depends 
to a great extent on the materials used. However, when direct -compression 
and wet-granulated formulations of norfloxacin were compared in a recent 
publication, it was found that the direct-compression formulation was superior 
not only in terms of disintegration and dissolution. but was also more 
compressible [35]. 
B. Fluidity 
The fluidity of tablet blends is important not only from the direct effect 
on uniformity of die fill and thus uniformity of tablet weight, but also from 
the role it plays in blending and powder homogeneity. Because of the 
overall smaller particle size encountered in direct-compression blends, 
fluidity is a much more serious problem than in the case of gr-anulations. 
A comparison of the bulk densities and particle size of SOme of the most 
common direct-compression fillers can be found in Table 4. 
It is important that fludity specifications be placed on all active ingredients 
and fillers that make up more than 5% of a final tablet formulation. 
Fluidity of active ingredients becomes a factor when the drug has been 
micronized to improve dissolution rate or provide more key particles of 
fIl 6 :l c .... Fast Flo "2
'" .- Anhyd. (Shaf.) 
:I: 4 .. DCl21, Hi Beta ..... DCl30. Hi Alpha ... Sp. Dried(Fore) 
2 -o- OCL11,Sp.Dried ... Zeparox 
0 400 BOO 1200 1600 2000 
Compression Force (kg) 
Figure 3 Compressibility profiles of different directly compressible 
l actoses ,

218 Shangraw 
Table If Physical Specifications of Direct-Compression Fillers 
Spray-dried lactose 
Fast~Flo lactose 
Anhydrous lactose 
Avicel pH 101 
Avicel pH 102 
Starch 1500 
( %) 
0.25- 0.5 
Bulk density 
(g mlr I) 
Particle sizeb 
100% through 30 
30- 60% on 140 
15- 50% through 200 
O.5-1. 5% on 60 
25-65% on 140 
15- 45% through 200 
16% on 60 
65% between 60- 200 
20% through 200 
1% on 20 
20% max. through 100 
3% max. on 40 
75% min. on 100 
5% max. through 100 
50% min. on 60 
10% max. through 120 
1% max. on 60 
7% through 200 
8% max. on 60 
45% on 200 
0% on 8 
0.5% max. on 40 
90% through 100 
5% max on 40 
15 max through 200 
aContains 4.5% water of hydration. 
bMesh size of screen.

Compressed Tablets by Direct Compression 219 
drug per tablet. If the amount of drug is small, this problem can be overcome 
by a proper choice of excipient fillers. However, when the drug 
makes up higher proportions of the tablet weight, the use of glidants in 
addition to careful selection of tablet fillers is necessary. The most effective 
glidants are the micronized silicas such as Cab-O-Sil and Syloid. 
They are generally used in concentrations of 0.1 to 0.25%. At higher levels 
the weight variation of tablets will often increase, and tablet hardness per 
specific die volume fill becomes less [36]. However, higher concentrations 
may be helpful as antiadherents, and may reduce filming and picking problems 
on punch faces. 
Most direct-compression fillers are purposely designed to give good 
flow properties. In most cases, fluidity in terms of volume (not weight) 
flow per unit time is directly related to particle size (Table 3). The two 
fillers with poorest flow appear to be microcrystalline cellulose and 
compressible starch. However, flow of these materials is not as poor as 
is often recorded when gravimetric flow and not volumetric flow data are 
presented [28]. 
The trend toward higher tablet machine output has necessitated the 
development of more sophisticated feeders because in older designs the dwell 
time of the die cavity in contact with the feeder was not adequate to allow 
uniform filling. This problem can become even more critical in direct compression 
because of the smaller mean particle size of direct-compression 
powder. There are two basic approaches to increasing die-feeding efficiency: 
(a) to force material into the die cavity; (b) to imp rove flow properties 
of material directly above the die cavity so that the material will 
naturally flow downward. The latter approach appears to be the more 
realistic and serves as the basis for most tablet machine modifications for 
improvement of die fill. One such system, designed by the Manesty Corporation, 
employs a rotary feeder with two horizontal paddles, which rotate 
in opposite directions. The paddle speeds can be synchronized with the 
main drive. It is possible that the use of such positive die-feeding equipment 
may be necessary if optimum fluidity cannot be obtained through 
careful selection of ingredients and choice of their concentrations. 
C. Content Uniformity 
Highly fluid powder blends facilitate unblending. The narrower the particle 
size range of all components and the more alike the particle densities, 
the less chance for unblending or segregation. It is important to note 
that it is the particle density and not the bulk density that is important 
in segregation. Cellulose and starch products tend to have lower true 
densities than sugars or inorganic chemicals. However, the small and 
angular particle shape of microcrystalline cellulose makes it difficult for 
higher density particles to sift down through the spaces between the blend 
of materials. Major problems with segregation can occur in spherically 
shaped fillers, particularly if the particle is large and spherical, such as 
is the case with compressible dextrose (Emdex). In such cases it is necessary 
to select other excipients to fill the empty spaces or to purposely preblend 
a micronized active ingredient with the large-particle filler. This 
approach is recommended by Ho and Crooks [37], who blended sulfaphenazole 
(mean particle diameter of 2 um) with coarse direct-compression tablet 
fillers, and then studied the blends, using a sampling method and electron 
microscopy. After mixing with a 180- to 250 urn fraction of direct-compression

220 Shangraw 
sucrose (DiPac) for 100 min, the standard deviation of 200-mg samples 
containing 4 mg of sulfaphenazole was equivalent to that predicted for a 
random mix. The mix did not appear to segregate during mixing or vibration. 
It is theorized that blending of the filler particles first (with lubricant, 
etc ,') or simply blending all materials at once would have interfered 
with the surface attraction of drug particles to filler and resulted in decreased 
homogeneity. There are a number of other excellent articles on 
ordered blending that point out its importance to direct compression [3840] 
D. Lubrication 
Lubrication has always been one of the most complicated and frustrating 
aspects of tablet formulation. The lubrication of direct-compression powder 
blends is, if anything, more complicated than that of classical granulations. 
In general, the problems associated with Iubricating direct-compression 
blends can be divided into two categories: (a) type and amount needed 
to produce adequate lubrication; (b) the softening effects of lubrication. 
Because the overall mean particle size of direct -compression blends is 
less than that for gr-anulations , higher concentrations of lubricants are 
often needed. The recognized need for small particle size of lubricants 
in granulations is of even greater importance in direct compression. 
Because there are already many more surfaces covered with lubricant 
in direct-compression blends, the softening effect upon compression is 
magnified. This is particularly true in direct-compression fillers that exhibit 
almost no fracture or plastic flow on compression. Even when all 
surfaces of a gr-anulation are covered by a layer of lubricant, significant 
clean surfaces are formed during compression. In most instances standard 
blending times will result in complete coverage of these surfaces. The same 
blending times in direct-compression blends mayor may not cover all primary 
surfaces. Thus length of blending becomes much more critical in 
direct compression than in lubrication of tablet granulations. If blended 
long enough. alkaline stearate lubricants will shear off and completely 
cover all exposed particle surfaces. It may be necessary to avoid the alkaline 
stearate lubricants completely in some direct-compression formulations. 
The influence of the duration of lubricant and excipient mixing on the 
processing characteristics of powders and on the properties of compacts 
prepared by direct compression was studied by Shah and Mlodozeniec [41]. 
T hey found that ejection force, hardness, disintegration. and dissolution 
of directly compressed tablets of lactose and microcrystalline cellulose were 
all significantly affected by blending times. The properties of directly 
compressed tablets can also be dramatically affected by the type of blender. 
which can be a major problem when scaling up from laboratory to production 
equipment [42J. When operated at the same rotation speed, the decrease 
in crushing strength of tablets was much faster for the large industrial 
mixers than for the laboratory blenders. Lubrication of direct-compression 
formulations is one of the more complex and difficult problems faced by a 
pharmaceutical formulator. 
The compressibility of direct-compression filler-binders can be more easily 
understood by viewing the morphology of individual particles. As was

Compressed Tablets by Direct Compression 221 
mentioned previously, most direct-compression fillers are mtntgranutattons 
in which the raw material itself has in some way been agglomerated or 
granulated after being chemically or physically modified. 
The scanning electron microscope has provided a unique tool to visualize 
such modifications while at the same time allowing for a q ualttative assessment 
of product quality. The scanning electron microscope was dramatically 
used by Hess to depict the nature of pharmaceutical compacts 
and the effects of compression force and disintegrating agents on tablet 
morphology [43J. The use of scanning electron photomicrographs for the 
characterization of direct -compreaston excipients was first reported by 
Shangraw et al , [44,45) and updated in a later article that further reviewed 
the usefulness of scanning electron microscopy in studying excipient 
properties [46). 
As can be seen in Figures 4 and 5, the spnay drying of lactose can 
result in agglomerates consisting of small a-monohydrate crystals held together 
by amorphous glass. These agglomerates now have the p rerequisite 
flow and deformation properties to make them compressible. The cocrystallization 
of sucrose with modified dextrins changes the poorly compressible 
sucrose crystals into a highly deformable dense aggregate of crystallites 
(Figs. 6 and 7). 
It was not possible to utilize fibrous cellulose as a tableting agent until 
it was mechanically formed into a large-particle floc that improved flow 
characteristics but with little improvement in compressibility (Fig. 8). 
However, it was the acid hydrolysis of cellulose and the subsequent spray 
drying of the more crystalline portions of the fibers into a free-flowing 
powder that revolutionized direct-compression tablet ing . This product, 
microcrystalline cellulose (Fig. 9), not only forms extremely hard compacts, 
but has the ability to improve the compressibility of other substances when 
it is added in concentrations of 10 to 30%. 
A scanning electron photomicrograph of unmilled dicalcium phosphate 
provides evidence of the aggregates of crystallites that shatter upon compaction 
to give tablet strength (Fig. 10). The agglomeration of starch 
Figure 4 Crystalline lactose. N.F. (non-spray- dried).

222 Shangraw 
Figure 5 Lactose, N.F. Spray-dried. (Fast-Flo). 
Figure 6 Sucrose, N.F. (crystalline).

Compressed Tablets by Direct Compression 223 
Figure 7 Compressible sugar, N.F. (Dipac) . 
Figure 8 Powdered cellulose, N.F. (Elcema 250).

Figure 9 Microcystalline cellulose. N. F. (A viceI pH 102) 
with partially hydrolyzed starch to form a free-flowing compressible granulation 
can be seen in Figure 11. 
One of the most significant contributions to the literature of pharmaceutical 
eXcipients is The Handbook of Pharmaceutical Excipients [47]. Of 
particular interest to those concerned with morphology and functionality 
are the book's scanning electron photomicrographs of almost all tablet fillers 
and disintegrating agents. A wide range of data is also presented 
for products that have the same chemical composition yet different morphologies. 
Such data include information about particle size. compressibility. 
and moisture sorption. 
Figure 10 Dibasic calcium phosphate. USP unmilled (Di-Tab, Emcompress).

Compressed Tablets by Direct Compression 225 
Figure 11 Pregelatinized starch N.F. compressible (Starch 1500). 
As it has become more and more apparent what makes chemical substances 
compressible and also what enhances their dissolution rates, it has become 
increasingly obvious that emphasis in tablet formulation has been misplaced. 
There is nothing less compressible or less rapidly soluble than a perfectly 
pure crystalline material. Yet for a century there has been an emphasis 
on producing the purest possible drug crystals. It is then up to the pharmaceutical 
formulator to take those crystals and mask the inadequacies 
of compressibility and dissolution inherent in them by means of external 
excipients. A more logical approach would be to supply the drug in an 
impure form (with known quantities of known impurities) so that the crystals 
are actually flawed or in fact do not exist as large crystals but as 
aggregates of microfine crystals. Although this has not yet been done for 
drug substances, pregranulations of some common drugs are available commercially. 
Ascorbic acid has long been available in a number of powder or granular 
forms. Ascorbic acid is commonly crystallized in monoclinic, platelike 
crystals. The term g ranular simply means large crystals (similar to granular 
sugar), not a granulation in terms of aggregated powders. 
In the mid 1970s Roche marketed ascorbic acid C-90 in which micronized 
ascorbic acid particles are granulated with starch paste. The product appears 
to be extruded through! a compactor and then ground. Each large 
particle is actually a granule of ascorbic acid and pasted starch, and is 
much more compressible than the pure crystalline material. However, the 
product does have an extremely wide variation in particle size, and addition 
of some filler-binder, such as microcrystalline cellulose, is recommended to 
optimize compressibility. More recently. Roch marketed a C- 95 ascorbic acid 
that contains only 5% excipients and utilizes methylcellulose rather than 
starch as the binder. Takeda Chemical Industries markets both a C-97 
direct-compression ascorbic acid and SA-99, a direct-compression sodium 

226 Shangraw 
Because of the increasing popularity of acetaminophen as an analgesic, 
it was only natural that a modification of this substance to improve compressibility 
would be attempted. Acetaminophen generally occurs as large monoclinic 
crystals. a crystal form which is not easily deformed and resists compaction. 
A direct-compression form of acetaminophen is available commercially 
from Mallinckrodt containing 90% acetaminophen and 10% of partially 
pregelatinized starch under the name COMPAP [48]. The spherical nature 
of the particles indicates that the material is prepared by spray drying; 
each particle is almost a perfect minigranule. Deformation can occur along 
any plane and multiple clean surfaces are formed during the compaction 
process. Moreover, each granule consists of hundreds of small crystals 
with wetted surfaces which optimize dissolution. Tablets with rapid dissolution 
can be easily formed by the addition of small concentrations of 
AcDiSol (2%) and lubricant (0.5% magnesium stearate). A self-lubricating 
version of this material is also available (COMPAP-L) as well as a combination 
of acetaminophen and codeine (Codacet-Bb) , 
Another direct-compression acetaminophen product is marketed by 
Monsanto under the name DC- 90 [49]. This product is prepared by fluidized 
bed granulation instead of spray drying. It has a compressibility 
profile similar to that of COMPAP but is only available in the self-lubricating 
form. Both products exhibit rapid dissolution profiles when formulated 
with effective disintegrant systems. The compressibility of both materials 
can be enhanced by the addition of 10 to 20% microcrystalline cellulose. 
The different morphologies or these products is debicted in Figure 12a 
and b. 
Figure 12 Direct-compression acetaminophen: (a) Compap (Mallinckrodt); 
(b) DC 90 (Monsanto).

Compressed Tablets by Direct Compression 227 
In 1982, Mallinckrodt introduced a directly compressible ibuprofen 
product under the name DCI. However. this product contains only 63% 
active ingredient and appears to be a classic granulation with little innovation. 
In some respects the term direct-compression is a misnomer when applied 
to any of these products. However, it is apparent that these products 
will continue to multiply and provide convenient intermediate materials 
for manufacturing companies with limited processing equipment. In many 
ways, they resemble the slugged aspirin/starch (90/10) granulations that 
became popular in the post-World War II period and are still commercially 
There is no reason to believe that it would not be possible to convert 
any active ingredient into a compressible form by crystal modification. 
The question remains as to whether or not this technique will be applied 
to drug substances or if pharmaceutical formulators will be forced to continue 
working with noncompressible, poorly soluble pure crystals. 
It is in the area of dry granulation and mixed processing systems where 
the most recent impact of direct -compression technology has taken place. 
When initially developed, direct compression was thought of as an 
all-or-nothing system. Gradually the integration of direct compression 
with various granulation processes has occurred. These include: 
1. Use of direct-compression excipients in postgranulation running 
2. Optimization of granulations prepared by roll compaction and 
3. Semi- or pseudogranulations , mini- or microg'ranulations , preblending 
of triturations 
4. Matrix for controlled relase granules or beads 
The use of microcrystalline cellulose, which was originally thought of 
as a direct-compression binder-filler. in the postgranulation running powder 
for increasing tablet hardness has been a common practice almost since 
its introduction. Subsequently. microcrystalline cellulose has gained acceptability 
in mini-or microgranulations in which small quantities of wet binders 
are used but are more thoroughly distributed in loosely agglomerated powders 
[50]. This allows for the maximization of the effect of both the wet 
binder and the dry binder. However, care in the granulation step has to 
be taken because the overwetting of the granules tends to reduce the 
binding effectiveness of the microcrystalline cellulose. 
A unique modification of this process was proposed by Ullah using a 
process called "moisture-activated dry granulation" (MADG) [51]. In this 
procedure, the binder (polyvinylpyrrolidone) is blended with the drug 
plus filler, a small amount of water is added, and the combination is then 
mixed thoroughly. Microcrystalline cellulose is subsequently added to 
sorb the small amount of moisture present. No traditional drying step is 
involved. The granulation tends to be nondense, with a relatively small 
particle size.

228 Shangraw 
Direct compression has had a significant impact on the particle size 
originally thought necessary for tablet manufacture. Formulators have 
come to realize that with the use of glidants, much smaller mesh materials 
can be used as granulations and the particle size of granules can in fact 
approach the particle size of direct-compression fillers. In fact, as was 
stated earlier, most direct-compression fillers are nothing more than microor 
The innovative use of compressible excipierrts for increasing the compressibility 
of a difficult material to tablets is illustrated by one approach 
to manufacturing BOO-mg ibuprofen tablets [52J. Ibuprofen has a very 
low bulk density, low melting point, POOl' compaction properties, and tablets 
produced by wet gr-anulations may age due to scintering. The patent 
for a stable high-dose high-bulk-density ibuprofen granulation describes 
the preparation of a dry granulation of croscarmellose and ibuprofen by 
roll compaction or chilsonation , and the subsequent blending of the granulation 
with additional croscarmellose and microcrystalline cellulose to produce 
a tablet. One might argue that this process is not direct compression, 
but the fact of the matter is that without the unique sorbent and disintegrating 
properties of croscarmellose and the unusual dry-binding properties 
of microcrystalline cellulose in the post blend powder, this product 
would not be possible. 
A further modification of the direct-compression process is the use of 
premixed triturations of potent drug substances with one or more fillers 
and the subseq uent addition of other fillers and binders before the final 
blend is directly compressed. This process is now being used successfully 
for making tablets of such potent drugs as clonidine with tablet strengths 
of 0.1, 0.2, and 0.3 mg. Preparation of tablets of this strength by direct 
compression would have been thought impossible 10 years ago. 
More recently, two potassium supplements have been introduced into 
the marketplace that involve the compression of coated potassium chloride 
crystals into directly compressed tablet matrices. One product is made by 
coating KCI crystals with a solution/suspension of paraffin, acetyl tributyl 
citrate. ethylcellulose. and silicon dioxide in isopropanol. The coated crystals 
are then blended with microcrystalline cellulose. rice starch, magnesium 
stearate, and talc, an d then compressed. The tablets are easily crushed 
and can be administered as a powder without changing the release characteristics 
of the KCI. 
A similar potassium chloride tablet with a strength of 20 meq has also 
been marketed. The tablet is extremely hard but disintegrates into the 
primary coated Kel crystals very rapidly. Microcrystalline cellulose and 
crospovidone act both as compressible cushioning agents during compaction 
and disintegrating agents during the very rapid breakup that occurs on 
exposure to fluids. which allows the tablet contents to be administered as 
a suspension if so desired. 
In spite of the slow adoption of direct-compression tableting by the pharmaceutical 
industry, there is every indication that its acceptance will continue 
to grow. Its use in the manufacture of generic drug and

Compressed Tablets by Direct Compression 229 
nonprescription drug products, where innovation is easier to apply and justify 
economically, is now widespread. As was mentioned in the last section, 
there is an increasing inclination to integrate aspects of direct compression, 
dry granulation, and wet granulation in product manufacture. Coprocessing 
of excipients and active ingredients to provide drum-to-hopper 
t ableting of raw materials will no doubt also increase in volume. It is 
difficult to envision significant new filler-binders because the basic building 
materials that are both chemically and physiologically acceptable have 
already been modified. However, there will be a continuing search for dry 
binders that can mimic or exceed the properties of microcrystalline cellulose 
and to discover a lubricant with the functionality of magnesium stearate 
but without its hydrophobic properties. 
As indicated above, the development of formulations for direct compression 
is both an art and a science. All formulations are highly dependent on the 
properties of the raw materials including the drug substance. It is not desirable 
to change sources of supply or grades of raw materials without validating 
effects on fluidity, compressibility, and solubility. This applies to 
the active ingredient also, par-ticularly in a high-dose drug. Following is a 
collection of formulations taken from the literature (Examples 1 to 25) illustrating 
many of the points discussed in the chapter. These are guide formulations 
only and results may vary depending on the properties of the drug 
substance and the type of blender or tablet press used. A number of them 
have been taken or adapted from formularies available from FMC, Food and 
Pharmaceutical Products Division and Edward Mendell Co . , Inc. 
Example 1; Aspirin Tablets USP (325 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Aspirin. USP 80.0 325.0 
( 4G--mesh) 
2. Avicel PH 12.0 48.0 
3. Cornstarch, N.F. 8.0 32.0 
100.0 405.0 
Note: Hardness of finished tablets can be improved by 
replacing corn starch with Starch 1500 with no resultant 
decrease in disintegration. Use of stearic acid is optional 
depending on aspirin type and concentration of Avicel. 
Blend all the ingredients for 20 min. Compress into tablets 
using 7/16-in. standard concave tooling.

230 Shangraw 
Example 2: Aspirin-Caffeine Tablets 
Composition per tablet 
Ingredient ( %J (mg) 
1. Aspirin. USP 80.0 384.00 
(40-mesh crystal) 
2. Caffeine, USP 3.30 15.84 
3. Avicel PH 10.00 48.00 
4. Cornstarch, N.F. 5.95 28.56 
5. Stearic acid, N.F. 0.75 3.60 
100.0 480.00 
Blend all ingredients in a P-K blender or equivalent for 
20 min. Compress into tablets using 7/16-in. standard 
concave tooling. 
Example 3: Acetaminophen Tablets USP (325 mg)
Composition per tablet 
Ingredient ( %) (mg) 
1. Acetaminophen, 56.5 325.0 
USP. granular 
2. Solka Floc-BW 100 20.9 120.0 
3. Emcocel 18.8 108.3 
4. Cab-O-Sil M-5 0.5 3.0 
5. Explotab 2.5 14.40 
6. Magnesium stearate, 0.7 4.30 
100.0 575.0 
Mix 1, 2, and 3 together for 10 min. Add 4 and 5 and 
blend for 10 min. Add 6 and blend for 5 min. and 
compress at maximum compression force. 
Note: Harder tablets can be made by replacing additional 
portions of Solka Floc with Emcocel.

Compressed Tablets by Direct Compression 
Example 4: Acetaminophen Tablets USP (325 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Acetamino phen 70.00 325.00 
2. Avicel PH 101 29.65 138.35 
3. Stearic acid, N.F. 0.35 1. 65 
(fine powder) 
100.00 465.00 
Note: If smaller crystalline size acetaminophen is desired 
to improve dissolution, it would be necessary to use a 
higher proportion of Avicel and to use PH 102 in place of 
PH 101, and to use a glidant. All lubricants should be 
screened before adding to blender. 
Blend 1 and 2 for 20 min. Screen in 3 and blend for 
an additional 5 min. Compress tablets using 7/16-in. 
standard concave or flat bevel tooling. 
Example 5: Analgesic Tablets 
1. Asprin, USP 
2. Salicylamide, USP 
3. Acetaminophen, USP 
(large crystals 
or granular) 
4. Caffeine, US? 
5. Avicel PH 101 
6. Stearic acid 
(powder), N. F. 
7. Cab-O-Sit 
( %) 
per tablet 
Blend all the ingredients except 5 for 20 min. Screen 
in 5 and blend for an additional 5 min. Compress into 
tablets using 7/16-in. standard concave tooling.

Example 6: Propoxyphene Napsylate-Acetaminophen 
(APAP) Tablets (100/650 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. 90% Pregranulated 93.01 722.19 
2. Propyoxyphene 11.49 100.00 
napsylate, USP 
3. Avicel PH 102 4.00 34. 77 
4. Ac-Di-Sol 1. 00 8.70 
5. Cab-O-Sir 0.15 1. 30 
6. Magnesium stearate, 0.35 3.04 
100.00 870.00 
Note: Pregranulated APAP is available from both 
Mallinckrodt and Monsanto in directly compressible forms 
containing 90% active ingredient. 
Screen 2 and 6 through a 40-mesh sieve. Screen 5 
through a 20~mesh sieve. Blend 1, 2, 3, 4, and 5 in 
a twin-shell blender for 15 min. Add 6 and blend for 5 
min. Compress using precompression force equal to 
one-third the final compression force. 
Example 7: Chewable Ascorbic Acid Tablets (100 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Ascorbic acid, 12.26 27.60 
USP (fine crystal) 
2. Sodium ascorbate, 36.26 81.60 
3. Avicel PH 101 17. 12 38.50 
4. Sodium saccharin 0.56 1. 25 
(powder) , N.F. 
5. DiPac 29.30 66.00 
6. Stearic acid, N.F. 2.50 5.60 
7. Imitation orange 1. 00 2.25 
juice flavor 

Compressed Tablets by Direct Compression 
Example 7: (Continued) 
Composition per tablet 
Ingredient ( %) (mg) 
8. FD&C Yellow 0.50 1. 10 
No. 6 dye 
9. cse-o-sn 0.50 1. 10 
100.00 225.00 
Note: It is not possible to make chewable ascorbic acid 
tablets with over 50% active ingredient. ather directcompression 
sugars such as Emdex could be used to replace 
DiPac. Magnesium stearate should be avoided in 
ascorbic acid formulations. Addition of a higher concentration 
of Avice! will not usually increase tablet hardness. 
Blend all ingredients, except 6, for 20 min. Screen in 
the stearic acid and blend for an additional 5 min. Compress 
into tablets usinq 7/16-in. standard concabe tooling. 
Example 8: Ascorbic Acid Tablets, US? (250 mg) 
1. Ascorbic acid, 
USP (fine crystal 
or granular) 
2. Avicel PH 101 
3. Starch 1500 
4. Stearic acid, N.F. 
(powder) or 
5. Cab-a-Sil 
( %) 
per tablet 
Note: It is important to use free-flowing types of ascorbic 
acid due to the high concentration in the formulation. 
Ascorbic acid concentration could be increased slightly by 
using more Avicel and less Starch 1500. 
Stearic acid, Sterotex, Compritol 888, and Lubritab are 
interchangeable in most formulations. 
Blend all the ingredients, except 4, for 25 min. Screen 
in 4 and blend for an additional 5 min. Compress into 
tablets using 7/16-in. standard concave tooling.

Example 9: Thiamine Hydrochloride Tablets, USP (100 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Thiamine hydro- 30.0 100.00 
chloride, USP 
2. Avicel PH 102 25.0 83.35 
3. Lactose, N.F. 42.5 141. 65 
4. Magnesium stearate, 2.0 6.65 
5. Cab-O-Sil 0.5 1. 65 
100.0 333.30 
Note: Anhydrous lactose could be replaced with Fast-Flo 
lactose with no loss in tablet quality. This would reduce 
(the need for a glidant (which is probably present in too 
high a concentration in many formulations). (Usually 
only 0.25% is necessary to optimize fluidity.) Blend all 
ingredients, except 4, for 25 min. Screen in 4 and blend 
for an additional 5 min. Compress using 13/32-in. standard 
concave tooling. 
Example 10: "Maintenance" Multivitamin Tablets
Composition per tablet 
Ingredient ( %) (mg) 
1. Vitamin A acetate 5.5 11. 0 
(dry form 500 IU and 
500 D2 per mg) 
2. Thiamine monoitrate, 0.8 1. 65 
3. Riboflavin, USP 1.1 2.20 
4. Pyridoxine HCI, USP 1.0 2.10 
5. 1%Cyanocobalamin 0.1 0.22 
(in gelatin) 
6. D-Cal ci urn pantothenate, 3.75 7.50 
1. Ascorbic acid, USP 33.25 66.50 
(fine crystals) 
8. Niacinamide 11.0 22.00 

Compressed Tablets by Direct Compression 
Example 10: (Continued) 
9. Emcompress 01" DiTab 
10. Microcrystalline 
cellulose, N. F. 
11. Talc USP 
12. Stearic acid, N. F. 
13. Magnesium stearate, 
N. F. (powder) 
( %) 
per tablet 
Note: This formulation could be converted into a 
chewable tablet by adding 40 to 50% sugar filler (Le., 
OJ-Pac and a small quantity of saccharine or aspartame). 
Blend all ingredients in a suitable blender. Compress 
at a tablet weight of 200 mg using 3/8-in. standard 
concave tooling. 
Example 11: Geriatric Formula Vitamin Tablets 
Composition per tablet 
Ingredient ( %) (mg) 
1. Ferrous sui fate, USP 30.00 156.00 
95% Ethecal granulation 
2. Thiamine mononitrate, 1. 09 6.00 
3. Riboflavin, USP 1. 00 5.50 
4. Niacinamide, USP 6.00 33.00 
5. Ascorbic acid, C-90 17.45 96.00 
6. Calcium pantothenate, 0.73 4.00 
7. Pyridoxine HCI, USP 0.14 0.75 
B. Cyanocobalamin, 0.82 4.50 
0.1% spray-dried 
9. AcDisol 2.00 11.00 
10. Stearic acid N.F. 2.00 11.00 

236 Shangraw 
Example 11: (Continued) 
11. Magnesium stearate 
12. CeloCal 
{ %} 
per tablet 
1. 38 
Prepare a premix of items 2. 3, 6. 7. Mix in other ingred 
ients except 10 and 11 and blend for 15 min. Add 
10 and mix for 5 min. Add 11 and blend for an additional 
5 min. Compress using oval punches (1 = 0.480in 
. w ::: 0.220 x cup = 0.040-in.). Sugar or film coat. 
Example 12: Pyridoxine HCI Tablets (10 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Pyridoxine HCI, 5.0 10.00 
2. Emcornpress 92.5 185.00 
3. Emcosoy 2.0 4.00 
4. Magnesium stearate. 0.5 1. 00 
100.0 200.00 
Blend 1 and 2 together for 10 min in a twin-shell 
blender. Add 3 and blend for an additional 10 min. 
Add 4 and blend for 5 more min and compress.

Compressed Tablets by Direct Compression 
Example 13: Sodium Fluoride Chewable Tablets (2.2 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Sodium fluoride 2.0 2.200 
2. Emdex 96. 75 106.425 
3. Artificial grape flavor 0.25 0.275 
5.5. (Crompton and 
4. Color. grape 53186 0.25 0.275 
(Crompton and Knowles) 
5. Magnesium stearate. 0.75 0.825 
100.00 110.000 
Mix ingredient 1 and one-third of 2 for 10 min. Add remaining 
amount of 2 and 4 and mix thoroughly for 20 min. 
Add 3 and blend for 10 min. Add 5 and blend 5 additional 
min and compress. 
Example 14: Chewable Antacid Tablets 
Composition per tablet 
Ingredient ( %) (mg) 
1. FMA-11* (Reheis 25.2 400.00 
Chemical Co. ) 
2. Syloid 244 3.2 50.00 
3. Emdex 69.3 1100.00 
4. Pharmasweet powder 1.3 20.00 
(Crompton and Knowles) 
5. Magnesium stearate, 1.0 16.00 
100.0 1586.00 
Note: An appropriate flavor may be added. 
*Aluminum hydroxide/magnesium carbonate co-dried gel. 
Mix 1 and 2 together for 5 min. Screen through 30- 
mesh screen (if ingredients no already prescreened) and 
mix for 10 to 15 min. Add 3 and 4 and blend thou roughly 
for 10 to 15 min. Add 5. blend 5 min. and compress. 

238 Shangraw 
Example 15 : Calcium Lactate Tablets (10 gr) 
Composition per tablet 
Ingredient ( %) (mg) 
1- Calcium lactate, * US? 71.25 470 
2. AcDiSol 1. 25 10 
3. Avicel PH 101 10.00 80 
4. Stearic acid, N.F. 2.50 20 
5. Magnesium stearate, 0.50 4 
6. CeioCaJ 14.50 116 
100.00 800 
*Equivalent to calcium lactate pentahydrate 650 mg. 
Mix ingredients 1, 2, 3, and 6 for 10 min. Add 5 and 
blend for an additional 5 min. Compress on Stokes 551 
using l/2-in. standard concave upper bisect punches. 
Example 16: Pyrilamine Meleate Tablets, USP (25 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1- Pyrilamine maleate, 12.50 25.00 
2. Avicel PH 101 17.00 34.00 
3. Lactose, N.F. 68.40 136.80 
4. AcDiSol 1. 00 2.00 
5. Cab-O-Sit 0.35 0.70 
6. Stearic acid, N.F. 0.25 0.50 
7. Magnesium stearate, 0.50 1. 00 
100.00 200.00 
Screen 1, 6, and 7 through 40-mesh sieve. Belnd 1 and 
3 for 3 min in V blender. Add 2, 4, and 5 to step-2 and 
blend for 17 min. Add 6 to step 3 and blend for 3 min. 
Add 7 to step 4 and blend for 5 min. Tablet using 
5/16-ln standard concave punches to a hardness of 
5.5 kg.

Compressed Tablets by Direct Compression 239 
Example 17: Doxylamine Succinate Tablets USP 
Composition per tablet 
Ingredient ( %) (mg) 
1. Doxylamine succinate, 6.4 25.13 
2. Syloid 244 0.85 3.35 
3. Solka Floc, BW100 4.05 16.70 
4. Emcompr-ess 83.95 331.82 
5. Explotab 5.0 20.0 
6. Magnesium stearate, N.F. O. 75 3.0 
100.00 400.0 
Screen 6 through 30 mesh screen and blend with 2 for 10 
to 15 min. Add 3 and one-third of 4 and mix for 10 min. 
Add remaining 4 and blend for 10 min. Add 5 and blend 
for 5 to 7 min. Add 6 and blend for 3 to 5 min. 
Example 18: Amitriptyline HCI Tablets US? (25 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Amitri pty line HCI, U.S.P. 22.73 25.0 
2. Fast- Flo lactose 59.52 05.47 
3. Avlcel PH 102 15.00 16.50 
4. Ac-Di-Sol 2.00 2.20 
5. Cab-O-SO 0.25 0.28 
6. Magnesium stearate, 0.50 0.55 
100.00 110.0 
Screen 1, 2, and 6 through a 40-mesh screen. Blend 1, 
2, 3, 4, and 5 in a suitable twin-shell blender for 5 min 
using intensifier bar. Blend above mixture for an additional 
5 min without the intensifier bar. Add 6 and blend 
for another 5 min. Compress.

240 Shangraw 
Example 19: Furosemide Tablets USP (40 mg) 
1. Furosemide, USP 
2. Avicel, PH-l02 
3. AcDiSol 
4. Fast-Flo lactose 
5. Cab-O-Sil 
6. Stearic acid, N.F. 
7. Magnesium stearate, N.F. 
( %) 
1. 50 
1. 00 
per tablet 
1. 60 
Screen 5 through a lO-mesh sieve. Screen 6 and 7 through 
a 40-mesh sieve. Blend 1t 2. and 4 in twin-shell blender 
without intensifier bar for 1 min and then blend with aid of 
intensifier bar for 0.5 min and without intensifier bar for 
1.5 min. Add 3 and 5 and blend for 3 min. Add 6 and 
blend for 3 min. Add 7 and blend for 5 min. Discharge 
blender and pass blend through 40-mesh sieve using oscillating 
granulator. Charge blender with sieved blend and 
blend for 5 min. Compress using 6/16-in. flat-faced, 
beveled edge punches. Compression force as needed 
to give a tablet of 6-kg hardness. 
Example 20: Allopurinol Tablets (300 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Allopurinol, USP 55.74 300.00 
2. Emcompress 37.2 200.00 
3. Explotab 3.8 20.50 
4. Talc 1.8 10.00 
5. Cab-O-Sil 0.5 2.50 
6. Magnesium stearate, N.F. 1.0 5.00 
100.0 538.00 
Blend 1 and 2 for 10 min. Add 3 and blend for 10 more 
min. Add 4 and 5 and blend 3 to 5 min. Add 6 and 
blend 5 more min.

Compressed Tablets by Direct Compression 
Example 21: Chlorpheniramine Maleate and 
Pseudoephedrine HCI Tablets (4/60 mg) 
1'. Chlorpheniramine 
maleate. USP 
2. Pseudoephedrine HCI. 
3. Avicel PH-101 
4. Fast-Flo lactose 
5. AcDiSol 
6. Cab-O-Sil 
7. Stearic acid, N.F. 
8. Magnesium stearate. 
( %) 
51. 36 
1. 00 
per tablet 
1. 1 
Screen 2, 7, and 8 through 40-mesh sieve. Blend 1, 2, 
and 3 in V blender for 3 min. Add 4, 5, and 6 to step 
2 and blend for 17 min. Add 7 to step 3 and blend for 
3 min. Add 8 to step 4 and blend for 5 min. Tablet 
to a hardness of 5.3 kg using 5/16-in standard concave 
Example 22: Penicillin V Potassium Tablets USP 
(250 mg; 400 IU) 
Composition per tablet 
( %) (mg) 
50.00 250.00 
24.25 121. 25 
22.00 110.00 
1. Penicillin V potassium, 
2. Avicel PH 102 
3. Ditab or Emcompress 
(unmilled dicalcium 
4. Magnesium stearate, 
Blend 1, 2. and 3 for 25 min. Screen in 4 and blend 
for an additional 5 min. Compress using 7/16-in. 
standard concave tooling.

242 Shangraw 
Example 23: Quinidine Sulfate Tablets USP (200 mg) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Quinidine sulfate, USP 55.85 200.0 
2. Avicel PH 102 40.25 144.0 
3. Cab-O-Sil 0.50 1.8 
4. Stearic acid, N.F. 2.50 9.0 
5. Magnesium stearate, 0.90 3.2 
100.10 358.0 
Blend 1, 2, and 3 for 25 min. Screen in 4 and 5 and 
blend for 5 min more. Compress using 3/8-in. standard 
concave tooling. 
Example 24: Chlorpromazine Tablets USP (100 mg) 
( %) 
per tablet 
1. Chorpromazine hydrochloride, 
2. Avice! PH 102 
3. Ditab or Emcompress 
4. Cab-O-Sil 
5. Magnesium stearate, 
28.0 100.00 
35.0 125. 00 
35.0 125.00 
0.5 1.74 
1.5 5.25 
100.0 357.00 
Blend all the ingredients, except 5, for 25 min. 
Screen in 5 and blend for an additional 5 min. C 
Compress into tablets using 11/32-in. tooling.

Compressed Tablets by Direct Compression 243 
Example 25: lsosorbide Dinitrate Tablets (10 mg, oral) 
Composition per tablet 
Ingredient ( %) (mg) 
1. Isosorbide dinitrate 20.00 40.00 
(25% in lactose) 
2. Avicel PH 102 19.80 39.60 
3. Fast-Flo lactose 59.45 118.90 
4. Magnesium stearate, 0.75 1. 50 
100.00 200.00 
Blend 1, 2, and 3 in a P-K blender for 25 min. Blend 
in 4 for 5 min. Compress into tablets using 5!16-in. 
standard concave tool ing. 
Glossary of Trade Names and Manufacturers 
Trade name 
Avicel 101, 
Compritol 88 
Elcema G-250 
Chemical! description 
Croscarmellose, N.F. 
Lactose N. F. (anhydrous 
direct tableting) 
Microcrystalline cellulose, N.F 
Glyceryl behenate , N.F. 
Lactose, N. F. (various types) 
Direct-compression calcium 
Compressible sugar, N.F. 
Compressible sugar, N. F. 
Dibasic calcium phosphate, 
USP (unmilled) 
Powdered cellulose, N.F. 
FMC Corporation, 
Philadelphia, PA 19103 
Sheffield Chemical, 
Union, NJ 07083 
DMV Corp., 
Veghel, The Netherlands 
FMC Corp., 
Philadelphia, PA 19103 
Gattefose Cor-p , , 
Elansford, NY 10523 
DMV Corp , , 
Veghel, Holland 
J. W.S. Delavau co., 
Philadelphia, PA 19122 
Desmo ChemicaI Corp., 
S1. Louis, MO 63144 
American Sugar Co., 
New York, NY 10020 
Stauffer Chemical Co . , 
Westport, CT 06880 
D-6000 Frankfurt (Main) 

Glossary of Trade Names and Manufacturers (Continued) 
Trade name 
Neosorb 60 
Solka Floc 
Sorbitol 834 
Sta-Rx 1500 
(Starch 1500) 
Microcrystalline cellulose, 
Dibasic calcium phosphate, 
USP special size fraction 
Dextrates, N. F . 
Sodium starch glycolate, N.F. 
Lactose, N.F. (spray 
Hydrogenated vegetable oil, 
Agglomerated maltrodextrin 
Sorbitol, N.F. 
(direct - compression) 
Compressible sugar, N.F. 
Crospovidone, N.F. (crosslinked 
Sodium starch glycolate, N.F. 
(carboxymethyl starch) 
Cellulose floc 
Sorbitol, N.F. (crystalline 
for direct compression) 
Lactose N.F. 
Pregelatinized starch, N.F. 
Hydrogenated Vegetable oil, 
Edward Mendell Co., 
Carmel, NY 10512 
Edward Mendell Co., 
Carmel, NY 10512 
Edward Mendell cc., 
Carmel, NY 10512 
Edward Mendell Co., 
Carmel, NY 10512 
Foremost Whey Products 
Banaboo, Wi. 
Edward Mendell Co., 
Carmel, NY 10512 
Grain Processing Corp., 
Muscatine, IA 52761 
Roquette Corp , , 
645 5th Avenue 
New York. NY 10022 
Ingredient Technology, 
Inc. , 
Pennsauken, NJ 08110 
GAF Corp . , 
New York, NY 10020 
Generichem Corp., 
Little Falls, NJ 07424 
Edward Mendell Co., 
Carmel, NY 10512 
ICI United States, 
Wilmington, DE 19897 
Foremost Whey Products, 
Baraboo, Wi. 53913 
DMV Corp., 
Vehgel, Holland 
Colorcon, Inc., 
West Point, PA 19486 
Capital City Products Co., 
Columbus, OH 43216

Compressed Tablets by Direct Compression 
Glossary of Trade Names and Manufacturers (Continued) 
Trade name 
Chemical / description 
Trade name identifying a number 
of direct-compression 
sugars including sucrose, 
fructose, dextrose 
Lactose, N. F. hydrous 
(for direct compression) 
Tricalcium phosphate anhydrous 
direct compression 
Coprocessed product containing 
30% calcium carbonate and 
70% microcryst alline cellulose 
Edward Mendell Co., 
Carmel, NY 10512 
Fallek Chemical co., 
New York, NY 10022 
(Product of Meggle 
Stauffer Chemical Co., 
Westport, CT 06881 
FMC Corp., 
Philadelphia, PA 19103 
1. G. Milosovitch, Drug Cosmet. Ind., 92, 557 (1963). 
2. W. C. Gunsel and L. Lachman, J. Pharm. ScL, 52, 178 (1963). 
3. C. D. Fox et al . Drug Cosmet. Ind. 92. 161 (1963). 
4. P. C. Record, Int. J. Pharm. Tech. and Prod. Mfr., 1)2),32 (1980). 
5. S. Pearce. Mfr. Chemist, 57(6), 77 (1986). 
6. R. A. Castello and A. M. Mattocks, J. Pharm. ScL, 51, 106 (1962). 
7. J. T. Hutton and G. Palmer, U.S. Patent 3,639,170 (1972). 
8. N. A. Butuyios, J. Pharm. ScL. 55, 727 (1966). 
9. G. K. Bolhuis et al , Drug Dev. Ind. Pharm., 11(8), 1657 (1985). 
10. H. Vromans et al., Acta Pharm. Suec., 22, 163 (1985). 
11. H. Vromans et al , , Pharm. Weekblad, Sci. Ed., 7, 186 (1985). 
12. DeBoer et al , , Sci. Ed, 8,145 (1986). 
13. H. V. VanKamp et al., Int. J. Pharm., 28, 229 (1986). 
14. H. V. VanKamp et aI., Acta Pharm. Suec., 23, 217 (1986). 
15. C. P. Graham et al., U.S. Patent 3,642,535 (1972). 
16. S. E. Tabibi and G. Hollenbeck, Int. J. Pharm., 18, 169 (1984). 
17. A. B. Rizzuto et a1.. Pharm. Tech., 8(9), 132 (1984). 
18. C. B. Froeg et al., U.S. Patent 3,639,169 (1972). 
19. H. D. Bergman et al., Drug Cosmet. Ind., 109, 55 (1971). 
20. J. DuRoss, Pharm. Tech., 8(9), 32 (1984). 
21. A. M. Guyot-Hermann and D. Leblanc, Drug Dev , Ind. Pharm., 11, 
551 (1985).

246 Shangraw 
22. A. Briggs, Develop. BioI. Standards, 36, 251 (1977). 
23. A. Briggs and T. Maxwell, U.S. Patent 3,932,943 (1976). 
24. B. Debord at 81., Drug Dev. Ind. Pharm, , 13, 1533 (1987). 
25. R. Short and F. Verbanac, U.S. Patent 3,622,677 (1971). 
26. K. S. Manudhane et aI., J. Pharm. sci., 58, 616 (1969). 
27. G. E. Reier and R. F. Shangraw, J. Pharm. sa., 55, 510 (1966). 
28. J. W. Wallace at al , , Pharm. Tech. 7(9), 94 (1983). 
29. T. Personen and P. Paronen, Drug Dev . Ind. Pharm., 12, 2091 (1986). 
30. E. Doelker et al., Drug oe, Ind. Pharm., 13,1847 (1987). 
31. A. D. F. Toy, Phosphorous Chemistry in Everyday Living, Am. Chern. 
Soc. Press, Washington, D. C., 1976, p , 57. 
32. X. Hou and J. T. Carstensen, Int. J. Pharm., 25,207 (1985). 
33. C. F. Lerk et 81 Pharm. Weekblad, 109, 945 (1974). 
34. J. Bavitz and J. B. Schwartz, Drug Cosmet. Ind., 114,44 (1974). 
35. A. V. Katdare and J. F. Bavitz, Drug Dev , Ind. Pharm., 13,1047 
(1987) . 
36. L. L. Augsburger and R. F. Shangraw, J. Pharm. Sci.. 55, 418 
(1966) . 
37. R. Ho et al,.; Drug Dev; Ind. Pharm., 3, 475 (1977). 
38. J. N. Staniforth, Int. J. Pharm. Tech. Prod. Manuf., 3(Suppl) 1, 
(1982) . 
39. J. Verraes and R. Ktnget , Int. J. Pharm. Tech. Prod. Manuf., 1(3), 
38 (1980). 
40. J. Staniforth and J. Rees, J. Pharm. Pharmacol., 35, 549 (1983). 
41. A. C. Shah and A. R. Mlodozeniec, J. Pharm. Sci., 66, 1377 (1977). 
42. G. K. Bolhuis et aI., Drug Dev , Ind. Pharm., 13, 1547 (1987). 
43. H. Hess, Pharm. Tech., 2(9), 36, (1978). 
44. R. Shangraw et aI., Pharm. Tech., 5(9), 68 (1981). 
45. R. Shangraw et aI., Pharm. Tech., 5(10),44 (1981). 
46. R. Shangraw, Pharm. Tech., 11(6), 144 (1987). 
47. American PharmaceuticaI Association and the Pharmaceutical Society of 
Great Britian, Handbook of Pharmaceutical Excipients, American Pharmaceutical 
Association, Washington, D.C. (1986). 
48. Ani! Salpekar , U.S. Patent 4,600,579 (1986). 
49. Steve Vogel, U.S. Patent 4,439.453, (1984). 
50. E. J. deJong, Ptutrtn, Weekblad, 104, 469, (1969). 
51. 1. UlIah et aI., Pharm. Teciu , 11(9), 48, (1987). 
52. R. Franz, U.S. Patent 4,609,675, (1986).

Compression-Coated and Layer Tablets 
William C. Gunsel* 
Ciba-Geigy Corporation 
Summit. New Jersey 
Robert G. Dusel 
Lachman Consultant Services. Inc. 
Westbury. New York 
In the early 1950s, two major developments in tableting presses occurred. 
Machines for compressing a coating around a tablet core and machines for 
making layer tablets appeared on the market. They were accepted enthusiatically 
through the 1960s, but the compression-coating technique is rarely 
employed today in the manufacture of new products because of the advent 
of film coating with its relative simplicity and its cost advantages. 
The chief advantage was the elimination of water or other solvent in 
the coating procedure. Thus there is no need for a barrier coating to prevent 
water from penetrating the cores-possibly softening them or initiating 
an undesired reaction. Such barriers, if efficient. slow down disintegration 
and dissolution. The dry coating is applied in a single step (in contrast to 
the repeated applications of different syrups), reducing the time required 
to evaporate the water and eliminating the necessity of cleaning the coating 
pan each time it becomes heavily encrusted with dried syrup. With dry 
coating. incompatible substances can be separated by placing one of them 
in the core and the other in the coating. There may be some reactivity 
at the interface but this should be negligible in the dry state. In addition, 
if a drug tends to discolor readily or develop a mottled appearance because 
of oxidation or sunlight, these problems can be minimized by incorporating 
the drug in the core tablet. 
Compression-coated tablets function like sugar-coated or film-coated tablets 
in that the coating may cover a bitter substance. conceal an unpleasant 
or mottled appearance, or provide a barrier for a substance irritating to 
the stomach or one inactivated by gastric juice. The advent of film coating 
"'Currently retired. 

248 Gunsel and Dusel 
dissipated much of the advantage of dry coating since larger quantities of 
tablets can be coated in a short time with film-formers dissolved in organic 
or aqueous solvents. These films dry so rapidly that there is scarcely sufficient 
time for a reaction to occur. Most recently, the deposition of films 
out of aqueous solution and suspension has become feasible. Recent advances 
in coating equipment, such as the side vented pans, have increased the efficiency 
of the aqueous coating operation to a point where even asprin tablets 
may be aqueous coated without significant hydrolysis. This has greatly 
increased the popularity of film coating over compression coating. Films 
produce a minimal increase in the size and weight of the core tablets; monograms 
and other devices on the core remain legible. 
While sugar coating a tablet may increase its weight by 50 to 100% of 
the core weight. the compression-coated tablet requires a coating that is 
about twice the weight of the core. If the cores are composed mainly of 
materials of low bulk density, such as fats and waxes. the amount of coating 
(by weight) must be even greater to assure a uniform volume of material 
surrounding the core. 
Another application of the compression-coated dosage form is in sustained-
release preparations. A coating containing the immediate-release 
portion is compressed around a slowly releasing core. This gives a far 
more accurate dose than is the case with sugar coating. In the latter, the 
immediate-release portion must be applied in increments; the cores do not 
pick up weight equally. As the process continues, those with increased 
surface area gain at the expense of those with less. Thus at the end of 
a coating run. tablet weights and drug content may vary as much as 20% 
for individual tablets, depending on the number of coats of active ingredient 
required. With compression coating, monograms and other markings 
may be impressed in the coating-in contradistinction to the printing of 
sugar-coated tablets in a separate step. The latter also requires complete 
inspection to sort out imperfect printing. 
A. History of Compression Coating 
The availability of compression-coating machines in the 1950s generated 
great interest; nevertheless, the idea was not new. As early as 1896, P. J. 
Noyes of New Hampshire acquired a British patent for such a device [11. 
The machine was a rotary press with two hoppers which supplied the granulation 
of the bottom and top coating. Between them was a third hopper from 
which the previously compressed core tablets passed through a tube with a 
reciprocating finger into the die. As each tablet was deposited On the bottom 
layer of coating, the die table paused in its rotation to allow good centering 
of the core. Then the process continued with deposition of the top 
layer, compression, and ejection. 
The next advance occurred in 1917 when F. J. Stokes [21 patented a 
machine which fed the cores onto a toothed disk. The cores passed from 
the disk into the dies. The timing of the disk was controlled by a starwheel, 
which was actuated in turn by projections on the turret. Another 
innovation was the embedding of the core into the bottom fill by the fall of 
the upper punch. The patent indicates that this was a layer press, but 
that a coated tablet was feasible if the cores had smaller diameters than the 
dies into which they were deposited.

Compression-Coated and Layer Tablets 249 
In 1935, the DeLong Gum Company of Massachusetts obtained a British 
patent [3] for a machine to compress a sugar composition onto chewing gum. 
The purpose of the invention was to protect the gum from the atmosphere. 
Biconvex cores were punched out of sheets of gum and deposited by the 
machine between two layers of coating. The concave faces of the punches, 
the convexity of the cores, and the lubricity of the coating contributed to 
automatic centration. A device with fingers might also aid in the placement 
of the cores, according to the claim made; the mechanics of this unit was 
not described. The inventor also mentioned that the product could be distinctively 
In 1937, Kilian, a German inventor, received a British patent [4] for a 
unit which compressed tablets on one machine and held them in the upper 
punches. These punches had rods passing lenthwise through them. The 
compression wheel was recessed so that it could compress the cores without 
activating the core rod. The cores were carried around the turret to the 
transfer mechanism. At this point the upper punches passed under a roller 
which pressed down the core rods, ejecting the cores onto the transfer 
plate. The plate carried the cores to the coating machine. It is evident 
that the Manesty DryCota adopted the idea of two machines running synchronously 
from this patent. Kilian, however, in cooperation with Evans 
Medical Supplies, Ltd., developed for sale the Prescoter which is a single 
rotary press. In the operation of this machine, the cores are fed from a 
vibrating hopper onto a feed plate which carries them to the dies, the 
process then resembling that of the Stokes machine. A reject device operating 
by the difference in hardness eliminates any coreless tablet. 
8. Available Equipment 
There are three principal designs in compression-coating machines. TwO of 
them provide for putting the coating on cores that were compressed on 
another machine; one provides for the compression of the core on one side 
of the machine with almost instantaneous transfer to the other side of the 
machine for the application of the coating. An example of the first type 
is the Colton Model 232 (Fig. 1). Previously compressed cores are fed by 
a vibrating feeder unit (A) onto a circular feeding disk (B) which is rotated 
clockwise or counterclockwise, as desired, by a variable- speed motor. 
The disk is tapered slightly downward from its center to its edge. A vibrator 
(C) gently agitates the disk so that the core tablets separate into a 
single layer. Around the periphery of the disk is a plastic ring (D) which 
prevents the tablets from piling up or escaping from the core selector ring 
immediately below (not visible in the photograph). The selector ring has 
33 V-shaped slots around its inner edge, which engage the cores. The 
cores are picked out of the slots by transfer cups (E) connected to a vacuum 
system through flexible tubing (F). The cups, which are spring loaded, 
are guided into contact with the cores by means of the cam (G) and the 
pins (H). The core centering ring and the transfer cups are synchronized 
with the speed of the die table. 
In practice, a bottom layer of coating enters the die from the hopper 
(I) and feed frame (J). At the same time (see Fig. 2), a core is picked 
up by a transfer cup which is guided by another cam (A) into the die (not 
visible) . The vacuum is interrupted, and the core rests on the bed of

250 Guneel and DU8el 
Figure 1 Colton Model 232. Refer to text 
coating. A metering feed plate (B) passes under a hopper (C) and feed 
frame (D)-and over the die into which it deposits the top layer of granulation. 
The whole is then compressed in the usual manner by passing the 
punches between the compression rolls. If a transfer cup does not contain 
a core, the vaeuum will suck the bottom layer of coating from the die. The 
metered amount of the top layer is then insufficient to form a tablet and 
will be expelled. If a core is not deposited, the pin in the transfer cup 
activates a microswitch which shuts off the press. Figure 3 is a schematic 
providing another view of the machine. 
The machine has 33 compression stations. It can produce a maximum 
of 900 tablets per minute. the largest tablet being 5/8 in. in diameter. It 
can handle cores p reviou sly made with flat - faced. sh allow concave. standard 
concave, cap sule- shaped, or oval punch tips. 
There are a number of problems in the operation of this machine. When 
one transfer cup fails to pick up a core, vacuum is lost-to the extent that 
cores picked up by the other nozzles are held insecurely and fallout before 
they can be deposited in the dies. Some cores are picked up inaccurately 
because, being constantly in motion. they do not slide all the way 
into the slots of the COre centering ring. They are then deposited offcenter 
or at a tilt and sometimes become visible in the surface of the coated 
tablet. The pins in the transfer cups, which are supposed to ensure deposition 
of the cores in the dies. become bent or jam; then cores are 
crushed, or the machine is frequently stopped by the tripping of the microswitch 
when the cores are retained. Parts of crushed cores can be carried 
beyond the point of deposition. fallon the die table, and be swept into the 
feed frame. Blockage may occur. Cores prepared with standard concave 
or deep concave punches tend to shingle or overlap on the core centering 
ring and cannot be picked up properly. Capped tablets will also disrupt 
the feeder which inserts the core tablets into the transfer device, producing

Compression-Coated and Layer Tablets 
Figure 2 Colton Model 232. Refer to text. 
a high level of rejected "ooreless" tablets. Tablets with flat faces or shallow 
convexities behave much better. 
The Stokes Model 538 is a modified 27-station BB2 double rotary machine 
with one set of compression rolls removed. As can be seen in Figure 4, 
the previously manufactured cores are loaded into a vibrating hopper (A) 
which moves them into a flexible feeder tube (B). The cores pass down 
the tube against a wheel mounted vertically, behind the housing (C). The 
top surface of the wheel is level with the die table. It contains 9 holes 
bored through the center; these are connected to a vacuum system by means 
of which the cores are carried to a transfer mechanism (D) mounted horizontally. 
This device contains 14 V-shaped slots in a link-chain system. 
A star-wheel, which is synchronized with the die table by means of bushings

Toblet Duster 
Metering Feed Tobie 
Unreleased Core Detector 
Plastic Retaininq Ring 
Direction of Fe('d Table 
Colton Model 232, schematic (Vector Corp.). Figure 3

Compression-Coated and Layer Tablets 253 
B:.: gl 
.......... F 
Figure 4 Stokes Model 538. Refer to text (Stokes Division, Pennwalt 
Corporation) . 
on the turret, guides the V-slots over the vacuum wheel and the dies. As 
the core tablet enters the V-slot, a spring-loaded pin rests with a slight 
pressure upon it. As the core passes over the die, which now contains 
the lower layer of coating from the hopper (E) and feed frame (F), the pin 
presses the core into the coating. At this moment the lower punch drops, 
leaving room for the deposition of the top layer of coating provided by a 
hopper and feed frame at the back of the machine. If a core is missing, an 
electrical sensing device on the feed mechanism detects the fact and activates 
a time-delay solenoid-which, in turn. releases a brief blast of compressed 
air, which blows the defective tablet into a reject chute (G) installed 
just before the normal tablet take-off. This machine can produce 700 tablets 
per minute, with diameters up to 5/8 in.; it can handle special shaped such

254 Gunsel and Dusel 
as ovals and capsules; and it is much simpler to set up and operate than 
the Colton Model 232. 
Nevertheless, difficulties occur with the Stokes machine also. Cores 
may clog the feeding tube: the vacuum wheel occasionally fails to hold a 
tablet: and the cores do not always slide accurately into the V-stots or fall 
into the center of the dies. In the last instance, some cores will be partially 
visible in the surface of the completed tablet. The reject mechanism catches 
more than one tablet in its jet of air and blows good as well as bad tablets 
into the reject chute. The good, however, can be salvaged by inspection. 
The Manesty DryCota is illustrated in Figure 5. It is essentially two 
heavy-duty D3 presses with a transfer device between, the three parts of 
the machine joined and kept in synchronization by a common drive shaft. 
The core tablets are compressed in the normal manner on the left-hand 
press (A). Upon ejection, they are brought up flush with the surface of 
the die adjacent to the right-hand side of the feed frame. They rise up 
into cups (B) on transfer arms (e) and are carried across the bridge (D) 
to the coating side of the machine (E). The transfer arms are positioned 
precisely by means of rings projecting below the upper punch guides. The 
rings engage a semicircular recess in each arm. The arms are spring loaded 
for positive fit. The bridge is perforated and connected to a vacuum pump, 
which removes loose dust and small particles from the cores and prevents 
transfer of core granulation into the coating granulation. 
The feed frame (F) on the coating turret is narrowed in its central portion 
to allow the transfer arms to pass. Granulation flows from the hopper 
(G) into the front of the feed frame and fills the bottom layer of coating 
into the die. A transfer arm is guided over this die: the core falls out of 
the cup as the lower punch is pulled down to make room. Simultaneously 
the upper punch (H) drops down on its cam track and taps the top of the 
transfer cup to assure positive release of the core. The die then passes 
beneath the back portion of the feed frame where the top layer of coating 
is applied. Then the whole is compressed together at (I). While the other 
two machines require a hopper, a feed frame. and fill adjustment for each 
of the bottom and top coatings, the DryCota requires only one hopper and 
feed frame. It does, however, have two weight adjustments. One is for 
the total amount of coating: the second, at (J). adjusts the bottom fill so 
that the top and bottom layers are of equal thickness. 
In the operation of the machine, the weight and hardness of the cores 
are adjusted first. Once these parameters are satisfactory, the transfer 
arms and cups are installed. Now, as the ceres are being transferred to 
the coating turret. the weight of the coating and the hardness of the tablets 
are established. Tablets are cut or broken in half to determine if the cores 
are centered. If not, the bottom fill is adjusted until centration is satisfactory. 
The weight and hardness of the cores can now be routinely 
checked while the machine is running. A lever behind the control box is 
depressed, causing a portion of the lower cam track to be raised and to 
eject the core and coating just before the compression wheels. A fixed 
blade, mounted across and close to the die table, diverts the ejected materials 
around the compression wheels to the discharge chute. The cores can 
now be separated from the coating granulation and tested for compliance 
with specifications, and any needed corrections can be made. 
There is a positive arrangement for detecting coreless tablets. The 
transfer cup is actually composed of two parts; a die with a cylindrical 
vertical bore through it, in which the core is tapped, and a pin with a

Compression-Coated and Layer Tablets 255 

Figure 6 Manesty DryCota: front microswitch. 
Gunsel and Dusel 
wide flange on top, which rests on the tablet. As Figure 6 demonstrates, 
when a core is in the cup. the flange (A) is raised and passes the knife 
blade (B) of a microswitch (C) without disturbing it. When the flange is 
resting on the die. signifying that a core has not been picked up. the 
switch is tripped and stops the machine. This switch is mounted at the 
front of the machine. As shown in Figure 7. at the back is another microswitch 
(A), which is actuated when the pin is up. indicating that the core 
has not been deposited. Again the machine is stopped. It is not necessary 
Figure 7 Manesty DryCota: rear rnicroswitch.

Compression-Coated and Layer Tablets 
Figure 8 Manesty DryCota: CenterCota unit. Refer to text. 
that the machine be wired to stop it in case of a reject. Alternatively, a 
gate in the discharge chute can be activated to divert the reject (along with 
several other tablets) into a separate receptacle while the machine continues 
to run. When the core is not deposited, it is forced out of the cup when the 
pin passes under, and is depressed by, an inclined ramp (B). The core 
falls into a small depression in the bridge of the transfer unit and thus cannot 
return to the core turret. 
The largest DryCota is a 23-station machine capable of producing 900 
tablets per minute with a maximum diameter of 5/8 in. The machine can also 
. be fitted with a unit called the CenterCota (Fig. 8), which enables previously 
compressed cores to be dry coated. It consists of a vibratory hopper (A) 
which guides the cores to a flexible tube (B). The cores pass down the 
tube to be engaged by U-slots mounted on transfer arms like those described 
above. The slots guide the cores to the dies of what is normally the coreforming 
turret. Thus the DryCota could be used for compressing a coating 
on cores that had been specially treated with a barrier coating, for example, 
to obviate a reaction between the two parts of the tablet or to provide a 
delayed release.

258 Gunsel and Dusel 
Apart from its low output, the DryCota has several drawbacks: the 
core tablets cannot be analyzed before they are coated, and they cannot be 
pretreated unless the CenterCota device is added to the machine. The first 
problem can be compensated for in large measure by analyzing the granulation 
beforehand. Although the cores are dedusted as they cross the bridge 
to the coating side, particles from the core granulation may be carried over 
to mingle with the coating and show up in the surface of the finished tablet. 
This event is most apt to occur if the upper core punches are worn and form 
a small ring (flash) around the top of the tablet. Flakes from this ring then 
falloff. Precompressed cores, having been vigorously vibrated beforehand 
on a tablet deduster, tend to be free of flash. Of course, good manufacturing 
practice would require the replacement of worn punches. 
Another machine available is the Kilian Preseoter , which operates like 
the Stokes machine except that the vacuum wheel is absent. The modern 
Prescoter is a single rotary machine and does not resemble the machine described 
in the 1937 British patent mentioned above. 
The newer model dry-coating machines have a number of distinctive 
features. Perfect feed-in of tablet cores is achieved by setting the circumferential 
speed of the core centering table equal to that of the turntable 
and employing an involute curve. A photoelectric tube is used to detect 
whether a tablet has its eore or not; if not, the tablet is rejected. As a 
doublecheck , coreless tablets are detected by measuring, with a load cell, 
the pressure difference between tablets with cores and those without. 
C. Comparison of Compression-Coating Machines 
The advantage of the Colton 232 and Stokes 538 is that the cores can be 
compressed on machines of much greater output-as many as 10,000 tablets 
per minute. The cores may be assayed before coating. When the core tablet 
is prepared on a separate machine, the hardness must be sufficient to retain 
its integrity during the bulk transfer and feeding into the die. This increased 
hardness often requires additional lubricant which will reduce the 
powder coating bond strength and therefore increase the level of rejects. 
Such cores will be firm enough to be handled in packaging machinery without 
incurring damage and therefore should be able to withstand transfer 
on the coating machines. It is almost futile to assign a numerical standard 
to the hardness requirement; hardness varies with the composition, thickness, 
shape, and diameter of tablets. A core 3/B in. in diameter with a hardness 
of 5 se units may be very satisfactory in one instance and oompletely inadequate 
in another. Although hardness testers measure resistance to crushing, 
which is important in dry coating, the resistance of a tablet to transverse 
fracture is more important. Unfortunately, there is no satisfactory way to 
make this measurement; there is only the subjective test of breaking the tablet 
with the fingers and listening for a distinct report of breakage-the 
The coating granulation tends to bond poorly to hard cores because of 
the latter's surface density. Then the strength of the coating depends mainlyon 
its own cohesiveness. The core may be likened to a peanut in a shell. 
The principal area of weakness is over the edge of the core. Increasing 
the eoattng thickness may compensate for the weakness. The advantage of 
the Manesty DryCota is that the core need only be firm enough to hold together 
while being carried a short distance across the bridge of the press 
to the coating turret. Thus the surface of the core is rather porous,

Manufacturer and model designations 
Table 1 Condensed Specifications for Compression-Coating Machines 
Compression coaters 
Manesty 900 
Colton Stokes Hata 
Specification 232 538 Core Coating HT-AP44-C 
Maximum tablet diameter (in.) 5/8 5/8 9/16 5/8 7/16 
Maximum depth of fill (in.) 1/2 11/16 7/16 5/16 
Number of compression stations 33 27 23 23 44 
Maximum output (tablets per min) 900 500 950 900 1,540 
Pressure (tons. in. - 2) 3 4 6 6 5.5 
aSeveral standard presses for comparison. 
Standard Machinesa 
Stokes Colton Beta 
585-1 247-41 Press 
7/16 7/16 5/8 
11/16 3/4 11/16 
65 41 16 
10,000 4,300 1,500 
10 4.5 6.5 
ci l:Q 
l:Q o' ;::s
c.... (ll 

t'" c -e 
a;' ..... 
t..:l en 

260 Gunsel and Dusei 
permitting penetration by the granules of the coating. On final compression. 
core and coating are densified simultaneously and bound firmly together. 
Since each of the machines described is a modification of equipment 
used in normal tablet operations, the manufacturers stress the latter use 
also. However, when such machines are purchased, they are usually devoted 
exclusively to dry coating. Only a research laboratory or a small 
business would employ them for multiple purposes. Their low output is 
extremely disadvantageous. The Manesty, being composed of two presses, 
is more productive than the others because it can turn out twice as many 
plain tablets as coated ones. 
Table 1 details the manufacturers' specifications for each machine. For 
comparison purposes. several high- speed machines are listed. 
Information about formulations for compression coating is Characterized by 
its paucity. A few workers have published some of their experiences; 
a few have obtained patents on compositions. Several authors have prepared 
review articles in which they have set down general rules for successful 
use of the dry-coating technique but there are few specifics. 
It is no easy task to obtain optimum quality in a tablet-a task to be 
attacked anew for each active ingredient and sometimes for each strength 
of the same medicinal chemical. For compression coating. where two formulas 
are involved for each product, the task can be even more difficult. Also 
of course, no one knows what the optimum formula is; the formulator usually 
settles for that composition which satisfies certain standards of hardness, 
friability, disintegration time. dissolution time. and stability. as well as the 
clinical requirement of effectiveness. Almost every new therapeutic agent 
presents problems of formulation which cannot be solved by some pet formulation. 
Nevertheless. there are compositions available which can be tried 
and, with some changes, found satisfactory. 
A. Core Tablets 
Almost any formula which will produce a firm tablet is satisfactory for all 
the machines described. There are a number of compressible fillers and 
compositions on the market which may be combined with the medicament, 
disintegrants , glidants , dry binders, and each other in an infinite number 
of proportions. They are economical to use because they eliminate the need 
for wetting to form granules and subsequent drying. There is no need to 
mill them, although screening may be required to break up agglomerates. 
On the other hand, the presence of the drug may interfere with the cohesion 
of the filler. Seldom does one find a substance like sodium chloride 
or potassium chloride which is inherently directly oompressible. When the 
amount of drug is small, the content uniformity may be poor; the drug 
may not distribute well because static charge develops during blending with 
the vehicle. The addition of starch. with its high moisture content. is useful 
for dissipating the charge. The fluidity of the vehicle may lead to 
segregation of the active ingredient on the tablet press. Here. the presence

Compression-Coated and Layer Tablets 261 
of microcrystalline cellulose in the formula can reduce the tendency to demix. 
Often also. large quantities of the compressible excipient may be needed 
for good cohesion. Since core tablets should be kept small, it is better to 
change to a wet granulation formula. 
Some materials currently available are spray-dried lactose, anhydrous 
lactose, microcrystalline cellulose, dicalcium phosphate, granular mannitol, 
sucrose, hydrolyzed starch derivatives (Emdex, Starch 1500 NF), and compositions 
of sucrose, invert sugar, starch, and magnesium stearate (NuTab). 
Some typical formulas using these materials are shown. 
Example 1: Typical Core Granulation 
I ngredient Quantity 
Active ingredient 
Starch NF 
Magnesi urn stearate NF 
Lactose NF anhydrous 
Example 2: Typical Core Granulation 
I ngredient Quantity 
Active ingredient 
Microcrystalline cellulose NF 
Magnesi um stearate NF 
Lactose NF (s pray-d ried) 
Example 3: Typical Core Granulation 
q .s . 
I ngredient Quantity 
Active ingredient 
Sodium starch glycolate NF 
Magnesium stearate NF 
Diabasic calcium phosphate USP 
1. 0% 

262 Gunsel and Dusel 
Example 4: Typical Core Granulation 
I ngredient Quantity 
Active ingredient 
Sodium starch glycolate NF 
Stearic acid NF 
1. 0% 
q , s , 
I n these four examples the drug is comminuted 
to a fine particle size, the other ingredients 
are passed through a 20 mesh screen if they 
are agglomerated, and the materials are blended 
for 15 to 20 min in a planetary, ribbon, or 
double-arm blender. The starches are included 
to promote disintegration. In the second 
example, the microcrystalline cellulose 
improves cohesion, disintegration, and compressibility. 
With dicalcium phosphate, additional 
magnesium stearate is needed for 
die release. 
An example of an active ingredient which could be formulated for direct 
compression is chlorisondamine chloride, a quaternary ammonium ganglionic 
blocker, used for the treatment of hypertension. It had a extremely unpleasant, 
bitter taste which had to be masked to make it acceptable. It 
had one physical attribute that was useful for tableting; namely, it was 
readily compressible. 
Example 5: Chlorisondamine Chloride Tablets 
Quantity per 
I ngredient tablet 
Chlorisondamine chloridea 
Lactose, NF spray-dried 
Magnesium stearate NF 
55.55 mg 
43.70 mg 
0.75 mg 
100.00 mg 
Break up any aggregates by passing all materials 
through a 20 mesh screen. Blend for 20 
min in a double-arm mixer. 
aContains 10% alcohol of crystallization.

Compression-Coated and Layer Tablets 263 
This formulation was suitable also for 25-mg and 100-mg cores, which 
were other desired strengths of the drug. The 25-mg core was compressed 
with 3/16-in. diameter standard concave punches; the 50-mg core with 1/4in. 
punches; and the 100-mg core with 5/16-in. punches. The level of 
magnesium stearate was established at 0.75% to overcome resistance of the 
tablets to extrusion. When these cores were covered with an inert composition, 
the 50-mg strength had a hardness value of 12 SC units and disintegrated 
in 10 min. The coating formula used was the same as in Example 
A second type of formula is a two-phased one in which the drug and 
the fillers are formed into granules by wetting them in the presence of an 
adhesive, drying the resultant moist mass, and passing it through a mill 
to obtain a convenient particle size. These granules are then blended with 
a disintegrant, if necessary, and a lubricant. It is a good idea to incorporate 
a disintegrant in the wet phase so that the granules will also readily 
disintegrate after the core tablet breaks up. Some thought must be given 
to the milling step since. in general, the granules should be relatively 
coarse so that the surfaces of the cores will be somewhat porous and permit 
penetration by the coating material for good bonding. Wire mesh or 
perforated plates for milling with openings of 10 to 16 mesh should be selected, 
the smaller openings for the smaller cores. 
Examples of core granulations prepared by the two-phase system are 
Example 6: Core Granulation (Two-Phase) 
I ngredient Quantity 
Active ingredient 
Dibasic calcium phosphate NF 
Starch NF 
lactose, NF, impalpable 
Povidone USP 
Purified water USP 
Magnesium stearate NF 
Note: The amount of lactose NF is reduced 
by the amounts of the drug. 
Blend the first four ingredients and pass 
them through a #1 perforated plate (roundhole 
screen) on a Fitzmill operating at medium 
speed with hammers forward. Prepare a 
solution by suspending the povidone in 
water. Add this povidone solution to the 
blended powders and mix until the mass is

264 Gunsel and Dusel 
Example 6: (Continued) 
uniformly moist. Spread the mass on trays 
and dry at 50 cC to a moisture content of 
2 to 3%. Pass the dried material through a 
20 (wire) mesh screen on a Fitzmil! running 
at medium speed with knives forward. 
Return the granules to a mixer and add 
the magnesium stearate. Mix for 10 min. In 
this formula, the lactose and calcium phosphate 
are the fillers; the starch is an internal 
disintegrant; the povidone is the 
binder. The magnesium stearate is the dierelease 
Example 7: Core Granulation (Two-Phase) 
Ingredient Quantity 
Active ingredient q .s , 
Mannitol USP q . s , 
Hydroxypropylmethylcellulose, NF 2.0% 
Purified water USP q.s. 
Sodium starch glycolate NF 4.0% 
Magnesium stearate NF 1. 0% 
Blend the drug and the mannitol. Dissolve 
the hydroxypropylmethylcellulose in water. 
Add to the powders and mix until the batch 
is uniformly moist and granular in appearance. 
Dry on trays at SOcC. Pass the 
dried materials through a Tornado mill 
equipped with a 16 (wire) mesh screen and 
running at medium speed with knives forward. 
Return the granules to the blender, add the 
sodium starch glycolate and magnesium 
stearate. Mix until uniformly dispersed (5 
to 10 min). Compress into tablets. 
Example 8: Core Granulation (Two-Phase) 
I ngredient Quantity 
Active ingredient 
Lactose NF implapable 
Starch 1500 NF 

Compression-Coated and Layer Tablets 
Example 8: (Continued) 
I ngredient Quantity 
Purified water USP 
Sodium lauryl sulfate NF 
Magnesi urn stearate NF 
1. 0% 
Blend the first four ingredients in a doublearm 
or planetary mixer. Moisten the powders 
with sufficient water to form a uniformly 
moist, granular mass. Pass the wet 
mass through a #4A screen on a Fitzmill operating 
at low speed with hammers forward. 
Spread the batch on trays and dry at 45C, 
until the moisture content is 2 to 3%. Pass 
the dried material through a #6 perforated 
plate on a Tornado mill running at medium 
speed with knives forward. 
Return the resultant granules to a twinshell 
blender, add the sodium lauryl sulfate, 
and magnesium stearate. Mix for 10 min. 
Compress at the predetermined weight and 
tablet dimensions. The sodium lauryl sulfate 
is the disintegrant in this formula. 
An unusual type of formula is the single-step granulation patented by 
Cooper et al , [5]. Also referred to as self-lubricating, the method calls 
for the blending of glidant and lubricant in the wet stage. 
It may seem unusual to include the lubricants in the wet granulation 
step, a procedure contrary to what is usually taught about the necessity 
for fine particle size of these substances in order to obtain easy dierelease. 
Nevertheless, the idea is valid and is presently used in a majority 
of one company's solid dosage forms. Apparently, in the comminution step, 
enough of the lubricant becomes exposed to perform its intended function. 
The quantities used in these one-phase formulas are the same as those in 
two-phase formulas. This procedure eliminates the usual mixing step to 
incorporate the lubricants. Any losses of materials in processing are in 
proportion to their presence in the formula. 
Example 9: Core Granulation (One-Phase) 
I ngredient Quantity 
Active ingredient 
Lactose NF impalpable 
Sucrose NF 

266 Gunsel and Dusel 
Example 9: (Continued) 
Ingredient Quantity 
polyethylene glycol 6000 NF 
Starch NF 
Talc USP 
Magnesium stearate NF 
Purified water USP 
Anhydrous alcohol 
After suitable screening to break up any 
aggregates, blend drug, lactose, sucrose, 
starch, talc, and magnesium stearate in a 
planetary or ribbon blender. Dissolve the 
polyethylene glycol in a mixture of purified 
water and alcohol at 50C. (The volume 
of the mixture is 20% larger than the weight 
of the polyethylene glycol.) Add this solution 
to the blended powders, mixing u ntil 
granules form, using additional 50% alcohol 
if necessary. Dry the moist mass at 45 
to 50C until moisture content is 1.0 to 
2.5%. Pass the dried material through a 
#5 perforated plate on a Tornado mill 
running at medium speed with knives forward. 
The batch is now ready for compression 
at the desired shape and weight. 
The use of the alcohol is not essential, 
but gives a better control of the wetting 
of the blended powders and promotes 
more rapid drying of the granulation. 
B. Coating Granulations 
Coating granulations also have some special requirements so that they will 
make a physically stable tablet. They require excellent cohesiveness as 
well as the ability to adhere tightly to the COre. They should be plastic 
enough to expand slightly with the slight swelling of the core after the extrusion 
of the completed tablet from the die. The maximum size of the 
granules must be less than the space between the deposited core and the 
walls of the die so that the granules will readily fill the space. Preferably 
the granules should be about one- fourth the width of this space. Good 
centratlon of the core is necessary to obtain a coating of equal strength 
all around. Although it is possible to apply a coating of only 1/32 in. on 
the edges of the core I 3/64 in. is better because the granulation can more

Compression-Coated and Layer Tablets 267 
easily fill the space, and there is leeway for slight off-centering. Centration 
can be critical if an enteric coating is being applied. Uniformity of 
coverage eliminates thin areas which may break down and release the contents 
of the core too early. Centration is also critical if the tablet is bisected 
with the intent of providing a divided dose. Misalignment of cores 
will make for unequal doses when the tablet is halved. 
Centration is affected by the mechanics of the machine, its rotational 
speed, and the quality of the coating granulation. The adjustment of the 
press must be made according to the manufacturer's specifications and will 
not be discussed here. The speed of the machine tends to centrifuge the 
core tablet toward the periphery of the table and opposite to the direction 
of rotation. Reducing the speed of the press will overcome this tendency. 
However, this is not an economical solution to the problem. The answer 
lies in the formulation of the coating. The granules should be relatively 
soft, somewhat like lactose, rather than hard like sucrose. Such granules 
prevent the core from sliding on the bottom layer a f coating. The fall of 
the upper punch on top of the core while the latter is being deposited is 
also helpful. To provide safteness in the granulation, plastic materials 
such as gelatin and polyethylene glycol should be included in the formula. 
The amount of granulating liquid should be kept to a minimum, and granulating 
time should be restricted, to prevent excessive activation of the 
Because the edges of compression-coated tablets are thicker than those 
of ordinary tablets, a somewhat larger amount of lubricant is needed to 
facilitate extrusion from the die. If 0.5% of magnesium stearate is sufficient 
for a plain tablet, about 50% more is necessary for dry-coated products. 
The amount of stearic acid or of hydrogenated vegetable oils, which are 
much less efficient, should be about double that of magnesium stearate. 
However, the amounts of these and similar lubricants may be reduced if 
polyethylene glycol 4000, 6000, or 20,000 is part of the formula since they 
also have lubricity. 
Any excipient that is suitable for a standard tablet or core tablet is 
suitable for the dry-coating formulation. It is customery, however, to 
use the same materials in the coating as in the core, a practice based on 
the theory that like substances w1ll bond better to like than to different 
ones. Nevertheless I a better criterion is the cohesiveness and plasticity of 
the formula: cohesiveness because the continuity of the coating depends 
on its strength around the edge of the core, and plasticity so that it can 
absorb the expansion of the core after the completed tablet is released from 
the die. This is especially important with the DryCota because there is 
only the briefest time for lateral expansion of the core, while the other 
machines the cores are prepared well ahead of time and can thus be seasoned. 
Wolff [6J has recommended that 2% of acacia be included in the formula to 
achieve bonding and has said that 1.75% of gelatin imparted satisfactory 
plasticity. His examples also reveal an extensive use of sugar in his 
coating formulas, a substance which is very cohesive. Cooper et al . [5] 
have relied mainly on tragacanth and sucrose for bonding and polyethylene 
glycol 6000 for plasticity and lubrication. Examples of typical formulas 
for coatings are shown in Examples 10 and 11. 
A formula which is resistant to moisture penetration is Example 12. 
A formula for an enteric coating given by Blubaugh et al. [7] is shown 
in Example 13.

268 Gunsel and Dusel 
Example 10: Typical Coating Granulation 
Ingredient Quantity 
Lactose NF impalpable 
Confectioners sugar NF 
Acacia NF spray dried 
Starch NF 
Gelatin NF 
Magnesium stearate NF 
Soluble dye 
Purified water USP 
q .5. 
Blend the first four materials until homogeneously 
mixed. Dissolve the dye in 
sufficient water and the gelatin in 5 
times its weight of water, using heat. 
Combine the dye and gelatin solutions 
and add to the mixed powders. Continue 
mixing until a moist, uniformly 
colored mass is formed. Pass the mass 
through a #4 perforated plate on a 
Fitzmill running at low speed with hammers 
forward. Spread on trays and 
dry at 45C to a moisture content of 
2 to 3%. Pass the dried granules 
through a #27 perforated plate on a 
Tornado mill operating at medium 
speed, knives forward. Return the 
granules to the mixer and add the 
magnesium stearate. Blend for 5 min. 
The granulation is ready for compression. 
Example 11: Typical Coating Granulation 
Ingredient Quantity 
Lactose NF (spray-d ri ed) q . s . 
Confectioners sugar NF 2.0% 
Acacia NF (spray-dried) 2.0% 
Polyethylene glycol 6000 NF 4.0% 
Talc USP 3.0% 
Magnesium stearate NF 0.5%

Compression-Coated and Layer Tablets 
Example 11: (Continued) 
Ingredient Quantity 
Soluble dye q. s 
Purified water USP q.s. 
Blend all ingredients except the dye and 
polyethylene glycol in a double-arm mixer. 
Dissolve the dye in a minimum 
amount of water and the polyethylene 
glycol in 1.2 times its weight of water 
at a temperature of 50C. Combine 
the two solutions and add slowly to the 
mixed powders. Mix for about 30 min 
or until a uniformly colored and moist 
mass is formed. Spread on trays and 
dry at 45C until the moisture content 
is 1 to 3%. Alternatively, the batch 
may be dried in a vacuum tumbler 
dryer with a jacket temperature ranging 
from 35 to 60C. Pass the dried granules 
through a #5 perforated plate on 
a Tornado Mill operating at medium 
speed with knives forward. The granulation 
is ready for compression. (If 
a drug is incorporated into the coating, 
the amount of lactose is reduced to 
compensate. ) 
Example 12: Typical Coating Granulation 
(Moisture Resistant) 
Calcium sulfate dihydrate 
Mannitol NF 
Tragacanth NF 
Acacia NF 
Talc USP 
Magnesi um stearate NF 
Purified water USP 
q .s , 

270 Gunsel and Dusel 
Example 12: (Continued) 
Blend the calcium sulfate, mannitol, 
tragacanth, talc, magnesium stearate, 
and colorant. Make a mucilage of the 
acacia with the water and add to the 
mixed powders. Pass the moist mass 
through a #4A perforated plate on a 
Fitzmill operating with knives forward. 
Spread on trays and dry in an oven at 
45C. Pass the dried material through 
a 20 mesh screen on the same mill. The 
granulation is ready for compression. 
Example 13: Enteric Coating 
Ingredient Quantity 
Triethanolamine cellulose 
acetate phthalate 
Lactose NF 
Magnesium stearate NF 
Purified water 
1. 0% 
Mix the triethanolamine cellulose acetate 
phthalate and lactose in a mixer with a 
Z-type agitator. Dissolve the colorant 
in the water and add to the mixed powders. 
Use sufficient water to make a 
tacky mass. Dry to the mass at 26C 
and a relative humidity of 30%. Pass 
the dried batch through a #2 perforated 
plate on a Fitzmill. Blend the granules 
with the magnesium stearate. Compress 
around core tablets using punches and 
dies 3/32 in. larger in diameter. The 
coating is to be a minimum of 1/32-in. 
thick. This coating withstands disintegration 
for 2 hr at pH 5.5. But 
of pH 5.6, disintegration occurs in 100 
to 110 min. At pH 7.5, it is 10 to 12 
In Examples 10, 11, and 12 one may substitute an active ingredient for 
part of the major excipient. Typical mesh patterns for the formulations 
in these examples would be:

Compression-Coated and Layer Tablets 271 
Caught on 
screen 10 11 12 13 
20 mesh 0% 2.5-7.5% 0% 0% 
40 mesh 15-25% 25-35% 42% 42% 
60 mesh 25-35% 10-20% 20% 20% 
80 mesh 10~20% 2.5-7.5% 12% 12% 
100 mesh 5-15% 2.5-7.5% 3% 3% 
Pan 2.5-7.5% 35- 45% 23% 23% 
C. Problem Solving 
Since core formulations can be, and are developed on standard machines, 
problems relating to hardness, friability, capping, extrusion from the dies, 
and disintegration can be solved before resorting to compression-coating 
machines. But it is otherwise with the coating formulations. They must 
be evaluated on the specific equipment available. The coating may cap off 
the cores because there is an excess of fine powder in the granulation: 
the amount of glidants, disintegrants, and lubricants should be no more 
than 10% of the batch, since these are powders with little cohesiveness. 
An excess of fines may be due to powdering in the mill because the granulation 
is weak and needs more binder, or because it is too hard and brittle. 
When a drug is present, it may affect the adhesive quality of the 
binder selected and require the choice of a different one. Fines may also 
be caused by the selected and require the choice of a different one. Fines 
may also be caused by the selection of the screen for milling. It is advisable 
to prepare a granulation and divide it into several parts. Each part 
should then be passed through a different screen-and at two different 
speeds. Then, each part should be used to coat the same batch of cores, 
and the physical parameters of the tablets should be evaluated. 
The granulation may be too dry; since water improves bonding, an increase 
would be needed. It can be obtained by adding starches or materials 
which tend to hold on to water, like sucrose or povidone. Perhaps the 
drying conditions need to be altered, with a reduction in temperature or 
drying time. 
The cores may have been compressed too hard and their surfaces densified 
so that the coating cannot bond. Hard cores tend to be elastic rather 
than plastic. Upon release of pressure when the tablet is ejected from the 
die, the rebound of the core pops the top off the tablet. 
Improper centration of the core either vertically or horizontally produces 
weak edges, and the coating will not hold together. Figure 9 illustrates 
faulty placement of cores within the envelope of coating. Windheuser and 
Cooper [8] ascribed poor centration to the poor flow characteristics of a 
granulation. They also believed that hard granules allowed the centrifugal 
force applied by the rotating turret to move the core off-center. Along the 
same lines, Lachman et al , [9] compared the bed of coating to a liquid.

272 Gunsel and Dusel 
Figure 9 Examples of off-centering. Faults in compression coating: (a) 
unequal coating; (b) cocking; (c) and (d) off-center. 
The rotation of the press would cause the liquid to move from the level. 
They found that fine granules caused the least movement of the cores. 
This failure of the cores to orient themselves on the bottom bed of 
coating is frequently found with those composed mainly of waxy substances. 
The core falloff-center or they land in a cocked position. The fault lies 
in the relative humidity in the compression booth. For good core deposition. 
the relative humidity should be at least 35%. and preferably between 40 and 
50%. Temperatures above 75F can soften wax cores and cause sticking in 
the transfer cups or V-slots. Either the temperature in the booth should 
be kept under control or the core granulation should be refrigerated for 
24 to 48 hr before being compressed. One drum of granulation should be 
used up before a second drum is removed from the refrigerator and brought 
into the compression room. 
If there is an incompatibility between the drugs in a combination tablet, 
or if the core is sensitive to moisture, the moisture content of the granulations 
should be kept to a minimum. Excipients such as mannitol and anhydrous 
lactose are preferable to sucrose in such cases. Longer drying times 
or more severe drying conditions are useful. The relative humidity in the 
compression area may need to be lowered also. 
When steel dies have been used for some time, they develop compression 
rings. The diameter of the ring is larger than that of the rest of the die. 
The tablet is slightly imprisoned by the ring, and a greater force than normal 
is applied to the circumference of the tablet during ejection. The bond 
at the top of the tablet weakens, and capping results. One must be sure 
that the dies are in good condition or invest in carbide insert dies which 
have an extremely long life without showing wear. The use of deep cup 
punches may also result in capping because the compression force is so unequal 
between the edge and the center of the tablet that, again, the adhesion 
of coating to core is weakened. 
Capping is not always obvious at the time that the tablets are being 
compressed. It may occur at some time later. To determine quickly if this 
event may occur. the formulator has several means. One common way is to 
attempt to force the coating off by pressing the thumbnail at the point 
where the top (or bottom) of the tablet meets the side. The tablet may be 
cut in half with a sharp knife or razor blade, and then an attempt may be 
made to pull off the coating. Another common test is to shake 20 or 30 
tablets vigorously in the cupped hands. More scientifically. a friability test 
may be performed. This test can be continued beyond the normal 4 min 
until the tablets do break up. This longer trial can be used to compare 
formulations. An even more pertinent test is to run the tablets two or more 
times through an automatic tablet counter and bottling machine and examine 
the tablets for damage.

Compression-Coated and Layer Tablets 273 
As a rule of thumb, the weight of the coating is about twice the weight 
of the core, provided that the granulations of both have similar bulk densities 
and the coating diameter is 3/32 in. larger than that of the core. This 
ratio provides enough material for a covering about 3/64 in. thick around 
the core. This margin can be made greater if the amount of drug in the 
coating is large. To make it less means risking cores that show through 
the coating or weakness at the tablet edge. If the core contains materials 
of low bulk density, then the amount of coating must be increased to give 
adequate coverage. 
A variation of the compression -coated tablet is the inlay, dot. or bull's-eye 
tablet. Instead of the core tablet being completely surrounded by the 
coating, it s top surface is completely exposed. With a yellow core and a 
white coating, the tablet resembles a fried egg. The preparation of such 
tablets requires that the top layer of coating be eliminated. Only the bottom 
layer of coating is deposited in the die and the core is placed upon it. 
The compression wheels then embed the core in the granulation, displacing 
some of the latter to form the sides, and finally press the whole into a 
tablet. Figure 10 shows two views of inlay tablets. With the Stokes, 
Colton, or Kilian machines, no alterations in equipment are needed. The 
feed frame and hopper which normally provide the top coating are not installed. 
With the Manesty DryCota, which utilizes a two-compartment feed 
frame for coating, it is necessary to block off the second part so that the 
granulation is diverted away from the dies and around the turret. 
This dosage form has a number of advantages over the compressioncoated 
tablet. It requires less coating material, only about 25 to 50% more 
than the weight of the core. The core is visible, so coreless tablets are 
readily detected. The reduction in the amount of coating makes for a 
thinner tablet. There is (of course) no concern with the capping of the 
top coating. 
This form can be useful in sustained-release preparations to reduce the 
size and weight of the tablet. The slow-release portion, Which contains 2 
to 3 times the amount of active ingredient. becomes the coating, and the 
immediate-release portion becomes the core. A specific example is the 100mg 
PBZ Lontab. As marketed, it was composed of a slow-releasing core 
weighing 200 mg and containing 67 mg of tripelennamine hydrochloride USP. 
The coating weighed 350 mg and contained 33 mg of the drug. The complete 
tablet weighed 550 mg and was 7/16 in. in diameter and 5.7 mm thick. When 
Figure 10 Inlay tablet: (a) cross-section; (b) view from above.

274 Gunsel and Dueel 
the core became the outer shell, the immediate-release portion could be reduced 
to 130 mg at 9/32-in. diameter, and the complete tablet made 3/8 in. 
in diameter with a thickness of only 4.2 mrn, a tablet more easily swallowed. 
The 100 mg PBZ Lontab is no longer marketed. It has been replaced by a 
conventional wax matrix release tablet. Another example is a European preparation 
containing 25 mg of hydrochlorothiazide in the bulls-eye and 600 
mg of potassium chloride in the outside portion. The latter contains a 
waxy substance to retard release and obviate gastrointestinal irritation. 
Thus the inlay is available immediately for its diuretic activity. To surround 
the potassium chloride with a granulation containing the hydrochlorothiazide 
would result in a tablet at least 1/2 in. in diameter-and in a 
great waste of materials. Only a layer tablet would be a reasonable alternative.
Since the inlay tablet requires the use of the same equipment as the 
dry-coated tablet, the problems encountered are similar. Poor centration 
is a much more obvious defect in the inlay tablet, however. Also, any 
color reactions due to incompatibility between the core and coating are 
The types of formulations previously cited are suitable for this kind of 
product. * There will be no difference in the output of the machines. 
Layer tablets are composed of two or three layers of granulation compressed 
together. They have the appearance of a sandwich because the edges of 
each layer are exposed. This dosage form has the advantage of separating 
two incompatible substances with an inert barrier between them. It makes 
possible sustained-release preparations with the immediate-release quantity 
in one layer and the slow -release portion in the second. A third layer 
with an intermediate release might be added. The weight of each layer can 
be accurately controlled, in contrast to putting one drug of a combination 
product in a sugar coating. Two-layer tablets require fewer materials than 
compression-coated tablets, weigh less, and may be thinner. Monogram s 
and other distinctive markings may be impressed in the surfaces of the 
multilayer tablets. Coloring the separate layers provides many possibilities 
for unique tablet identity. Analytical work may be simplified by a separation 
of the layers prior to assay. Since there is no transfer to a second 
set of punches and dies, as with the dry-coating machine, odd shapes 
(such as triangles, squares, and ovals) present no operating problems except 
for those common to keyed tooling. 
A. History of Layer-Tablet Presses 
F. J. Stokes, in his 1917 patent [2], indicated that his machine was a layer 
press, the first layer or tablet being compressed on another machine. The 
idea was apparently not pursued by the pharmaceutical industry at that time, 
*Dorsey Laboratories of Lincoln, Nebraska hold a patent [10] on the inlay 
concept and marketed several products in this dosage form.

Compression-Coated and Layer Tablets 275 
but the electrical industry developed the idea for the production of bimetallic 
contacts, which are actually two layers of metal bonded together. 
The earliest machines fed controlled volumes of each separate granulation 
on top of each other and compressed them together at one pressing 
station. The later machines were engineered to compress each layer separately 
before the deposition of the next granulation, with a final compression 
for the complete tablet. Since, in these machines, the excess g ranulation 
from each feed frame could not be permitted to circulate around the 
turret and commingle, wipe-off blades covering the entire face of the die 
table had to be installed. The excess was thus directed into pots at the 
side of the press and manually returned to the appropriate hopper. Suction 
tubes were needed to remove any fine dust that escaped under the 
scraper blades. The latest refinement has been the force feeders which 
retain the individual granulations. But some powder escapes from these 
also. and the same arrangement as described above is installed on the 
presses to prevent one granulation from contaminating the other. 
In the operation of the older type of machine, the granulation for the 
first layer is placed in the hopper, and the machine is adjusted until the 
desired weight is achieved with consistency; then the second hopper is 
filled with its granulation, and the same procedure is followed until the 
correct total tablet weight is obtained. In this. the single-compression 
method, the delineation between layers tends to be a little uneven. It is 
also difficult to make weight adjustments during a run. 
B. Layer-Tablet Presses 
Of the modern machines, there are two types which differ mainly in the 
way the layers are removed for weight and hardness checking. In one, 
the first layer of the first two layers are diverted from the machine; in 
the other, the first layer is made so hard that the second layer will not 
bond to it or will bond only weakly; upon ejection of the completed tablet, 
the layers may be easily separated and tested individually. 
Figure 11 illustrates the operation of a three-layer press with force 
feeders. The line (A) represents the die table. A granulation is placed 
in the first hopper and flows into the feed frame (B). The machine is 
started. and the volume of granulation in the die is adjusted by the weightadjustment 
cam (e). The upper and lower punches are brought together 
by the precompression rolls (D) and (E) to form a weak compact. Part of 
the lower cam track (F) is then raised hydraulically to eject the first layer, 
which is swept off the die table (A) by a wipe-off blade (G) affixed to the 
back edge of the second feeder (H). Samples are weighed, and hardness 
is determined. The operator makes any necessary corrections. When conditions 
are satisfactory, the ejection cam is lowered, and the entire procedure 
is repeated for the second layer, using feed frame (H), weightadjusting 
cam (I). tamping rolls (J) and (K), ejection cam (L), and wipeoff 
blade (M). The weight of the second layer is determined by the difference 
between the two weighings. The sequence is again repeated for 
the third layer by means of feed frame (N), weight adjustment (0) and 
final compression rolls (P) and (Q), with the completed tablet being removed 
from the machine by the Wipe-off blade (R) [to the right of the 
first feed frame (B)]. 
When a layer is ejected l the upper tamping roll is lowered slightly to 
exert more pressure upon the layer. This action will prevent damage to

2nc1 LAYER 
WT Checl( 
,. I 
: .'F 
~.I : .. 
Figure 11 Schematic of a layer press. Refer to text (Thomas Engineering). 
l: en 

Table 3 Specifications for Some Layer Presse~ ~
Manufacturer and model designation I 
Manesty Stokes ttl c. 
Manesty Rotapress Killian Versapress Fette Hata Vector Q 
Specification Layerpress mk I1a RU-3S 560-1 P- 3002 HT AP55L-DU Magna ;::s 
Number of dies 47 61 20 45 55 55 90 Q -e 
Maximum pressure (tons) 6.5 6.5 8.5 4 20(kN) 9 10 ...., 
Maximum tablet diameter 7/16 7/16 3/4 7/16 1/2 1/2 7/16 roo .... 
(in. ) ~ 
Maximum depth of fill (in.) 11/16 11/16 9.16 11/16 5/16 5/16 
Maximum layer thickness 
(prior to pressing) (In , ) 
First layer 1/4 7/16 1/4 7/16 11/16 - 3/4 
Second layer 1/4 1/4 1/4 1/4 11/16 - 3/4 
Third layer 1/4 - 1/4 - - - 3/4 
Maximum output (TPM) 1,500 5,550 417 2,100 4,125 3,850 5,000

278 Gunsel and Dusel 
the layer as strikes the take-off blade and is directed into the collection 
box. Once the lower punches have cleared the next filling station. they 
are quickly pulled down by a lowering cam so that they are not struck by 
the upper punches. The latter are already descending into the dies to 
make the next tamping or compression stroke. 
The leading and trailing edges of each feed frame are equipped with 
wipe-off blades which divert any powders that escape from the feeders into 
collection boxes. The blade on the trailing edge of the first feed frame 
guides the completed tablets down the chute (G) to the collection bin. 
Vacuum tubes at each filling unit suck away any powder- or granulation 
that remains in the lower punch faces during weight checks. Although the 
punches are raised flush with the die table at this time and do not drop as 
they pass under the feed frames. they do trap a small amount of material 
in the depressions in their tips. 
If an adjustment in the weight or thickness of the first or second layer 
is necessary, then the weight of each succeeding layer will probably need 
correction, since weight is related to the fill volume. 
The second type of machine is similar to the one described above, except 
for the manner in which weight checking is handled. Instead of a cam 
arrangement for ejecting the layers, the pressure on the first layer is increased, 
and the layer is made so hard that the next layer will not bond to 
it. Thus both layers are easily separated for weighing. This effect is 
achieved by activating a pneumatic cylinder which raises the lower tamping 
roll. There is an adjustment to control the distance that the compression 
roll may rise. Embossed or engraved upper punches provide a key between 
layers and tend to hold them together. Gentle shaking may be required to 
separate the layers in this case. Table 3 provides specifications for several 
typical layer presses currently available. 
As with compression-coated tablets, the granulation for layer tablets should 
be readily compressible for good bonding between layers. Dustlike fines 
should be kept to a minimum; the less dust, the cleaner the scrape-off at 
each feed frame. It may be necessary to separate out that fraction of a 
granulation which is finer than 70 or 80 mesh. Such material is not discarded 
but added to the next lot and regranulated. Lubricants, however, 
must be finely divided. their efficiency depending on the degree of fineness. 
Since these lubricant fines cannot be avoided, the quantities used should be 
kept minimal. The metallic stearates present an additional difficulty in that 
they interfere with the bonding of the -layers . Stearic acid and the hydrogenated 
fats are better lubricants from this point of view. Nevertheless, 
granules should be small, less than half the thickness of the layers; otherwise, 
the lines of demarcation between layers will be uneven. 
Equal weights of granulation will not necessarily lead to equal thickness 
of the layers. That will depend on the compression ratios of the formulations. 
It may be compensated for by adjusting the weights required for 
each layer. (It is not necessary. however, that each layer have the same 
thickness.) The shape of the punches also plays a role: punches with

Compression-Coated and Layer Tablets 279 
Figure 12 Cross sections of layer tablets. 
beveled edges or concave faces will make the top and bottom layers of a 
three-layer tablet appear thinner than the middle one. Flat-faced tooling 
will produce equal thickness of the layers, but unfortunately the edges of 
the tablets will tend to chip readily. Figure 12 shows cross sections of 
layer tablets and illustrates how the shape of the upper punch determines 
the shape of the layers. If the upper punch faces have monograms or 
other markings, the bonding between layers will be strengthened because 
the devices will act as keys between the layers. Additionally, precompression 
lengthens dwell time and aids in bonding. The formulas previously 
given for compression-coated tablets will serve as a guide for the development 
of formulations for layer tablets, with the exception of two of those 
for direct compression (Examples 1 and 2), Which are composed entirely of 
fine substances. 
An illustrative formula is one for an analgesic-antipyretic decongestant 
containing aspirin and phenylpropanolamine. A thin layer of placebo is 
placed between them to negate the chemical incompatibility of the active 
Example 14: First Layer of 
Analgesic-Antipyretic Decongestant 
Quantity per 
Ingredient tablet 
Phenylpropanolamine Hel USP 12.50 mg 
Lactose NF (spray-dried) 55.00 mg 
Microcrystalline cellulose NF 28.00 mg 
Colloidal silicon dioxide NF 1.25 mg 
Stearic acid NF 1.25 mg 
Screen where necessary to break down agglomerates 
or lumps (30 mesh screen is satisfactory) 
and blend the phenylpropanolamine, 
lactose, colloidal silicon dioxide and 
stearic acid.

Example 15: Second Layer of 
Analgesic-Antipyretic Decongestant 
Gunsel and Dusel 
Lactose NF (spray-dried) 
Microcrystalline Cellulose NF 
Colloidal silicon dioxide NF 
Stearic acid NF 
Quantity per 
26.00 mg 
54.00 mg 
1.00 mg 
1. 00 mg 
Pass the lactose and microcrystalline cellulose 
through a 30 mesh screen and blend them in 
a suitable mixer. Add the stearic acid and 
colloidal silicon dioxide. Mix for 10 min. 
Example 16: Third Layer of 
Analgesic-Antipyretic Decongestant 
Aspirin 40 mesh crystals 
Starch 1500 NF 
Colloidal silicon dioxide NF 
Stearic acid NF 
Quantity per 
81. 0 mg 
19.0 mg 
0.5 mg 
1.0 mg 
Blend in a suitable mixer until homogeneous 
(10 to 15 min). Compress the three layers 
together using 3/8-in. diameter, flat-faced, 
beveled-edge punches. The weight of each 
layer is: 
First layer, 98 mg 
Second layer, 82 mg 
Thl rd layer, 101. 5 mg 
The top layer is the last layer to be pressed. 
Since it is the aspirin portion, it will be most 
resistant to extrusion from the dies. 
Layer presses find employment in the manufacture of chewable antacid 
tablets. A possible formula for such a product follows. The mannitol provides 
pleasant mouth -feel and sweetness, and the saccharin enhances the 
latter. Peppermint flavoring has a long and honorable association with antacid 
preparations. The sucrose acts as the binder, although. of course. it 
also contributes to the taste of the tablet.

Compression-Coated and Layer Tablets 
Example 17: Fi rst Layer of Chewable 
Antacid Tablet 
Magnesium oxide heavy USP 
Mannitol USP 
Sucrose NF 
Saccharin sodium USP 
Purified water USP 
Magnesium stearate NF 
Peppermint oil NF 
200.0 mg 
400.0 mg 
60.0 mg 
1.0 mg 
7.0 mg 
4.0 mg 
Blend the magnesium oxide, mannitol, and 
saccharin in a double-arm mixer. Dissolve 
the sucrose in double its weight of water 
and add to the blended powders. Continue 
mixing until a moist, granular mass 
is formed, using additional purified water 
if necessary. Pass the batch through a 
#5 perforated plate on a Fitzmill operating 
at low speed with hammers forward. 
Spread the material on trays and dry at 
SOoC. Pass the dried granules through 
a 12-mesh screen on a Fitzmill running at 
medium speed with knives forward. Return 
the granules to the mixing machine 
and add the peppermint oil. When the 
oil has been thoroughly dispersed, add 
the magnesium stearate. (If the oil is 
not added before the lubricant. the tablet 
will have oil spots on its surface.) 
Compress the layer at 672 mg using 
S18-in. diameter punches with flat faces 
and beveled edges. 
Example 18: Second Layer of Chewable 
Antacid Tablet 
Aluminum hydroxide 
(dried gel) USP 
Mannitol USP 
Saccharin sodium USP 
Starch NF 
Purified water 
200.0 mg 
400.0 mg 
0.6 mg 
32.0 mg 
160.0 mg

282 Gunsel and Dusel 
Example 1B: (Continued) 
Ingredient Quantity 
Peppermint oil NF 
Magnesium stearate NF 
3.4 mg 
7.0 mg 
q . s. 
Blend the aluminum hydroxide, mannitol, 
and saccharin. Dissolve the color in the 
water and add the starch. Heat the mixture 
on a waterbath until the starch jells 
and forms a paste. Use the paste to 
granulate the blended powders. Add 
more water, if necessary, to form a lumpy 
mass. Pass this mass through a #5 perforated 
plate on a Fitzmill running at low 
speed with hammers forward. Spread the 
material on trays and dry at 50C. Pass 
the dried granules through a 12 mesh 
screen on a Fitzmill running at medium 
speed with knives forward. Return the 
granules to the mixer. Add the flavor 
first and then the magnesium stearate. 
Compress at 643 mg onto the first layer. 
A recent search of the literature has shown that no significant advances 
have been reported in the field of three-layer tablets. 
From the older literature [11] there is this example of a three-layer 
tablet. Today. this formulation would be unacceptable from a safety standpoint 
because of the FD&C Yellow #5. chloroform. and phenacetin. However, 
it is a good example of a typical three-layer tablet. 
Example 19: Bottom Layer of Three-Layer 
Ingredient Quantity 
Acetylsalicylic acid 
FD&C Yellow No. 5 
210.0 9 
4.0 9 
30.0 g 
10.0 9 
Mix thoroughly and pass the mixture 
through a hammer mill. Add sufficient

Compression-Coated and Layer Tablets 
Example 19: (Continued) 
chloroform to obtain a wet granulation. Reduce 
the granules to a range of 20 to 40 
mesh and dry overnight at a temperatu re 
of 120 to 140F. 
Example 20: Middle Layer of Three-Layer 
Ingredient Quantity 
Phenyltoloxamine dihydrogen 
Powdered sugar 
Distilled water 
Magnesium stearate 
150.0 g 
15.0 g 
25.0 g 
4.0 9 
0.4 g 
3.3 g 
3.0 9 
Blend the phenacetin, caffeine, and phenyltoloxamine 
dihydrogen citrate. Prepare a 
paste by heating the starch and sugar in 
the water. Add the paste to the powders 
and form granules. Dry the moist mass 
overnight at 120 to 140F. Reduce the 
mass to granules of about 20 mesh. Blend 
the granules with the magnesium stearate. 
Example 21: Top Layer of Three-Layer 
Ing redient Quantity 
Potassium phenethicillin 
FD&C Red No. 3 
Magnesium stearate 
125.00 g 
0.03 g 
q i s , 
3.00 g 
Blend the first two ingredients and pass 
them through a hammer mill. Add sufficient 
chloroform to make a hard rubber-like 
mass. Break up the mass and dry overnight 
at 120 to 140F. Reduce the dried

284 Gunsel and Dusei 
Example 21: (Continued) 
material to about 20 mesh granules. Blend 
the granules with the magnesium stearate. 
Using a three-layer press, compress 
the bottom layer at 254 mgt the middle 
layer at 197.4 mg t and the top layer at 
128.03 mg. 
Today FD&C Yellow No.5 would not 
be used with acetylsalicylic acid because 
of the possi bility of allergic reactions. 
Although compression-coated and layer tablets are a modest fraction of 
solid oral dosage forms, they provide two additional alternatives in solving 
formulation problems. They tend to be more expensive to manufacture than 
other tablets (except tablet triturates) because of the multiple granulations 
needed and the slowness of the special presses used. 
1. Noyes, P. J., British Patent 859996 (1896). 
2. Stokes, F. J., U.S. Patent 1,248,571 (1917). 
3. DeLong Gum Company, British Patent 439,534 (1935). 
4. Kilian, F., British Patent 464,903 (1937). 
5. Cooper, J., Pasquale, D., and Windheuser, J., U.S. Patent 2,857,313 
(1958) . 
6. Wolff, J., U.S. Patent 2,757,124 (1956). 
7. Blubaugh, F., Zapapas, J., and Sparks, M., J. Amer. Phram. Assoc. 
[Sci. Ed.], 47,12:857-870 (1958). 
8. Windheuser, J. and Cooper, J., J. Amer. Pharm. Assoc. [Sci. Ed.], 
45,8:543 (1956). 
9. Lachman, L., Speiser, P., and Sylwestrowicz, H., J. Pharm. Sci . 
52,4: 379- 390 (1963). 
10. Boswell, C., U.S. Patent 3,048,526 (1962). 
11. Buehwalter , F., Granatek, A., and DeMurio, M., U.S. Patent 
3,121,044 (1964). 
Remington's Practice of Pharmacy, 17th ed . , Mack PUb., Easton, Pa , , 1985. 
Ritschel, W. A., Die Tablette, Editio Cantor KG, Aulendorf i , Wuertt. I 
Germany, 1966.

Effervescent Tablets 
Raymond Mohrle 
Warner-Lambert Company, Morris Plains, New Jersey 
Effervescence is defined as the evolution of bubbles of gas from a liquid 
as the result of a chemical reaction. Effervescent mixtures have been 
known and used medicinally for many years. Effervescent powders used 
as saline cathartics were available in the eighteenth century and were subsequently 
listed in the official compendia as compound effervescent powders. 
These were more commonly known commercially as "Seidlitz Powders." Effervescent 
mixtures have been moderately popular over the years since 
along with the medicinal value of the particular preparation, they offered 
the public a unique dosage form that was interesting to prepare. In addition, 
they provided a pleasant taste due to carbonation which helped to 
mask the taste of objectionable medicaments. When tableting equipment was 
developed, these granular materials began to be compressed into tablets 
that offer some advantages over the powdered dosage forms. Effervescent 
tablets are convenient, easy-to-use, premeasured dosage forms. They cannot 
spill as can the powdered preparations. They can be individually packaged 
to exclude moisture, thereby avoiding the problem of product instability 
of the unused contents during storage. 
Only two effervescent tablets (both potassium supplements) are listed 
in the current USP [11. However, a wide range of effervescent tablets 
have been formulated over the years. These include dental compositions 
containing enzymes [2], contact lens cleaners [31, washing powder compositions 
[41, beverage sweetening tablets [5], chewable dentifrices (6], denture 
cleansers [71, surgical instrument sterilizers [8], analgesics [91, effervescent 
candy [10], as well as many preparations of prescription pharmaceuticals 
such as antibiotics [11,12], ergotamines [13J, digoxin [141, 
methadone [15] and L-dopa [16]. Preparations for veterinary use have 
also been developed [17]. 

286 Mohrle 
Some types of effervescent tablets are illustrated by the formulations 
in Section VIII of this chapter; however, all of them can generally be categorized 
into two distinct classes depending on the intended use of their 
solutions (Le , , is the resultant solution ingestible or not suitable for ingestion?). 
Effervescent tablets are not meant to be ingested or used without 
prior dissolution, usually in water. The ultimate use of the tablet solution 
plays a major role in the formulation of the product, specifically in 
the choice of raw materials to be used. Many substances have useful properties 
in the formulation of tablets whose solutions are not ingestible while 
possessing at the same time additional properties that render them useless 
if the solution is to be ingested (Le . , boric acid as a tablet lubricant, 
sodium bisulfite as an acid source, or sodium bicarbonate as a source of 
carbon dioxide in a sodium-free potassium supplement). 
Several investigators have studied changes in the bioavailability of a 
drug when delivered from an effervescent tablet. Many studies have been 
done with aspirin. Some indicated that significant differences in the absorption 
kinetics of aspirin were observed between effervescent and conventional 
or enteric coated tablets [18-20] and that the differences could be 
attributed to gastric emptying rate and rapid tablet dissolution [21]. No 
significant differences were observed in other studies with aspirin [22,23] 
or acetaminophen [24J effervescent tablets. Another investigator reported 
increased bioavailability of phenylbutazone from an effervescent dosage 
form [25]. 
The use of enteric coated effervescent tablets to improve the absorption 
of sodium aminosalicylate or L-dopa from the intestine has also been studied 
[26,27J. As the cellulose acetate phthalate- or hydroxypropyl methylcellulose 
phthalate-coated tablets reached the upper part of the intestine, rapid 
disintegration ensued causing an increased rate of absorption and a more 
prolonged blood level of the drugs as compared to conventional compressed 
tablets. The clinical effectiveness of a foaming antacid tablet was reported 
to be significantly higher than that of placebo in a study using a chewable 
tablet containing an effervescent matrix of alginic acid and sodium bicarbonate 
A. General Characteristics 
In many respects, the principles that apply to the production of conventional, 
noneffervescent tablets apply to the production of effervescent tablets, and 
are covered in greater detail in other sections of this volume. Much of the 
processing and process equipment are the same, as are the general properties 
of tablet granulations needed to produce a satisfactory tablet, such 
as particle shape, particle size, and uniformity of distribution to produce 
a free-flowing granulation suitable for use with high-speed rotary tablet 
presses. In addition, the granulations must be compressible either through 
the inherent properties of the raw materials or through the use of additives 
or specialized processing to impart the desired compressive properties. 
One property of the raw materials chosen for use in effervescent tablets, 
perhaps somewhat more important than for conventional tablets, is moisture 
content. The reaction most often employed for tablet disintegration in an 
effervescent tablet formulation is that between a soluble acid source and

Effervescent Tablets 287 
an alkali metal carbonate to produce carbon dioxide gas, the latter serving 
as the tablet disintegrant. This reaction proceeds spontaneously when the 
acid and carbonate components are mixed in water. The reaction can also 
occur-to a lesser degree-in the presence of small amounts of water 
bound to or adsorbed on the raw materials used in the formulation. If 
this reaction does occur after the tablet is prepared and packaged, it will 
cause the product to become physically unstable and decompose. Once 
initiated, the reaction will proceed even more rapidly since a byproduct of 
the reaction is additional water. For these reasons, raw materials either 
in the anhydrous state, with little or no adsorbed moisture, or with water 
molecules bound in a stable hydrate are preferred. Some water is needed, 
however, for binding purposes since a completely anhydrous granulation 
usually will not be compressible. Raw materials can be carefully chosen 
to provide the water needed for binding purposes as explained further in 
Section III of this chapter. 
Solubility is another raw material property especially important in the 
formulation of effervescent tablets. If the tablet components are not soluble, 
the effervescent reaction will not occur and the tablet will not disintegrate 
quickly. The rate of solubility is perhaps even more important than solubility 
per se since a slowly dissolving soluble substance can hinder tablet 
disintegration and provide a slowly soluble, often objectionable residue after 
the tablet disintegrates. Ideally, all the tablet components should have 
similar rates of solubflity . 
B. Acid Sources 
The acidity needed for the effervescent reaction can be derived from three 
main sources: food acids, acid anhydrides, and acid salts. The food 
acids are the most commonly used. They occur in nature and are used as 
food additives; they are all ingestible. 
Food Acids 
Citric Acid 
Citric acid is the most commonly used food acid, being readily abundant 
and relatively inexpensive. It is highly soluble, of high acid strength, and 
available in fine granular, free- flowing, anhydrous, and monohydrate food 
grade forms. Powdered forms are also commercially available. It is very 
hygroscopic, and care must be taken to prevent exposure to and storage 
in high-humidity areas if it is removed from its original container and not 
suitably repackaged. 
Tartaric Acid 
Tartaric acid is also used in many effervescent preparations, being 
readily available commercially. It is more soluble than citric acid and is 
also more hygroscopic. It is as strong an acid as citric acid, but more 
must be used to achieve equivalent acid concentration since it is diprotic, 
whereas citric acid is triprotic. 
Malic Acid 
Malic acid is also available in sufficient quantity for possible use in effervescent 
preparations. It is also hygroscopic and readily soluble. Its acid

288 Mohr-Ie 
strength is less than citric or tartaric acids but high enough to provide sufficient 
effervescence when combined with a carbonate source. It has a smooth, 
tart taste that does not "burst" in flavor as does the tart taste of citric acid. 
Fum aric Acid 
Fumaric acid, although as strong an acid as citric acid, is not generally 
useful in effervescent tablets due to its extremely low solubility in water. 
It is virtually nonhygroscopic in nature and is the most economical of the 
food acids. A cold water soluble form of furmaric acid is available (Monsanto 
Co , , St. Louis, Missouri). The increase in solubility is due to the addition 
of 0.3% dioctyl sodium sulfosuccinate; however, even this additive has 
not made fumaric acid adaptable for effervescent products. 
Adipic and Succinic Acids 
Neither of these food acids has been used extensively in effervescent 
products since they are far less soluble than citric acid in the temperature 
range in which most effervescent products are used. They are also less 
available and less economical. Both have the advantage, however, of being 
nonhygroscopic. Both have been reported to be useful as tablet lubricants 
as discussed in Section II. G. 
Acid Anhydrides 
Anhydrides of food acids are of possible value in effervescent products. 
When mixed with water, they are hydrolyzed to the corresponding acid, 
which can react with the carbonate source present to produce effervescence. 
If the hydrolytic rate is controlled, acid will be continuously produced throughout 
the solution, resulting in a sustained, high-volume, effervescent effect. 
Water cannot be used in the manufacture of products containing anhydrides 
since they would be converted to the acid prior to product use. Succinic 
anhydride is commercially available and has been used in a denture soak composition. 
It is reported that the acid anhydride reduces caking tendencies 
by acting as an internal desiccant in addition to increasing carbon dioxide 
evolution [29]. Citric anhydride has been reported in the literature [30]. 
Acid Salts 
Certain acid salts are useful in the formulation of effervescent products. 
Sodium Dihydrogen Phosphate 
This compound, also known as monosodium phosphate, is available commercially 
in granular and powdered anhydrous forms. It is readily soluble 
in water, producing an acid solution of about pH 4.5. It readily reacts 
with carbonate or bicarbonate to produce effervescence when dissolved. 
Disodium Dihydrogen Pyrophosphate 
This compound, also called sodium acid pyrophosphate, is another acid 
salt that has been used in effervescent tablets. It is also readily available 
and is soluble in water, producing an acid solution. 
Acid Citrate Salts 
Use of both sodium dihydrogen citrate and disodium hydrogen citrate 
has been reported in an effervescent composition [31]. Both are readily 
soluble and produce acid solutions suitable for ingestion.

Effervescent Tablets 
Sodium Acid Sulfite 
This raw material, also known as sodium bisulfite, produces an acid solution 
capable of releasing carbon dioxide gas from a carbonate source. 
Sodium bisulfite is not suitable for ingestion but may have application in 
the formulation of effervescent tablets for uses such as toilet bowl cleaners. 
It is a strong reducing agent and is not compatible with oxidizing agents. 
C. Carbonate Sources 
Dry. solid carbonate salts provide the effervescent in most effervescent 
tablets-carbon dioxide gas. Both the bicarbonate and carbonate forms 
are useful with the former being more reactive and used most often. 
Sodium Bicarbonate 
Sodium bicarbonate is the major source of carbon dioxide in effervescent 
systems. It is completely soluble in water, nonhygroscopic, inexpensive. 
abundant, and available commercially in five particle size grades ranging 
from a fine powder to a free -flowing uniform granule. It is ingestible and 
is. in fact. widely used as an antacid either alone or as part of antacid 
products. It is used extensively in food products as baking soda, and as 
a component of dry chemical and soda/acid fire extinguishers. It is the 
mildest of the sodium alkalies, having a pH of 8.3 in an aqueous solution 
of 0.85% concentration. It yields approximately 52% carbon dioxide. 
Sodium Carbonate 
Sodium carbonate, also known as soda ash, can be a useful raw material 
to the formulator of effervescent tablets. In addition to its effect as a 
source of carbon dioxide, it is useful as an alkalizing agent due to its high 
pH of 11. 5 in an aq ueous solution of 1% concentration. Sodium carbonate 
also exhibits a stabilizing effect when compounded into effervescent tablets 
due to its ability to absorb moisture preferentially. preventing the initiation 
of the effervescent reaction. (This phenomenon is discussed in more detail 
in Section VI.) For this reason, the anhydrous form is preferred over the 
hydrated forms that are also available. 
Potassium Bicarbonate and Potassium Carbonate 
Both of these salts can be used in effervescent tablets. especially when 
the sodium ion is undesirable or needs to be limited, as in the case of antacid 
products in which dosage is dependent on the amount of sodium recommended 
for ingestion. They are more soluble than their sodium counterparts 
and are significantly more expensive. The range of commercially available 
forms may be less satisfactory to the formulator than the wide range available 
for the sodium salts. 
Sodium Sesq uicarbonate 
This material, used primarily in the laundry industry, is a compound 
consisting of equal molar amounts of sodium carbonate and sodium bicarbonate 
and twice the molar amount of water. It is soluble in water, with a 
pH of 10.1 at a 2% concentration. It may be useful in effervescent tablets; 
however. mixtures of sodium bicarbonate and sodium carbonate will usually 
suffice in this application. The dihydrate form may also present a stability 
problem in some applications.

290 Mohrle 
Sodium Glycine Carbonate 
This material is a complex of aminoacetic acid and sodium carbonate. 
It is reported [32] to have the following advantages over other carbon 
dioxide sources: directly compressible granules; greater water solubility; 
less alkalinity; more heat stability; does not yield free water or reaction, 
and therefore provides the tablet with greater stability in the presence of 
trace amounts of water. The economics of the product may be a disadvantage 
in some formulations. 
L- Lysine Carbonate 
The preparation of this material is described in the literature [33]. It 
can be used in effervescent mixtures for preparing sparkling drinks and 
pharmaceutical compositions, especially when alkali metal ions are not desired. 
The material is a white crystalline powder that is very soluble in water. 
However, commercial availability is not apparent. 
Arginine Carbonate 
Use of this material has been reported [34] in an effervescent product 
free from alkaline earth metals. Tablets incorporating citric acid and 
arginine carbonate provided a source of the amino acid for various medicinal 
Amorphous Calcium Carbonate 
Preparation and use of this material has been described in the literature 
[35]; however, it is not yet commercially available. This material, which 
does not show a crystalline state upon X-ray analysis, remains stable without 
reverting to a crystalline form for a significant period of time. The 
preparation was reported for effervescent compositions that are sodium-free 
and highly palatable with excellent carbonation. 
D. Other Effervescent Sources 
The gas produced during effervescence need not always be carbon dioxide, 
although this is the one most frequently used. The evolution of oxygen gas 
can be used as a source of effervescence in certain products, particularly 
denture cleansers. Tablets have been compounded [36] in which a raw 
material known as anhydrous sodium perborate or effervescent perborate 
has been used. This raw material is prepared by heating either sodium 
perborate monohydrate or tetrahydrate under controlled conditions to drive 
off the hydrated water molecules. When it is mixed with water, copious 
volumes of oxygen gas are liberated, producing effervescence. 
Another method of generating oxygen gas to serve as an effervescent 
tablet disintegrant is the reaction between a peroxygen compound that 
yields active oxygen on mixture with water, e.g., sodium perborate monohydrate 
or sodium percarbonate, and a chlorine compound that liberates hypochlorite 
on contact with water, e. g., sodium dichloroisocyanurate or calcium 
hypochlorite [37]. The evolution of oxygen gas, which occurs best 
in alkaline media, proceeds as the peroxygen compound is decomposed by 
the chlorine compound. 
A recent U. S. patent [38] describes the preparation and use of an effervescent 
material prepared by the absorption of a gas, such as carbon 
dioxide, into an anhydrous base medium composed of an inorganic oxide

Effervescent Tablets 291 
material, such as zeolite aluminostlicate . Upon contact with water, the gas 
is desorbed from the inorganic matrix producing effervescence. This 
process is most useful for semisolid applications such as toothpastes and 
hand cleaners. 
E. Binders and Granulating Agents 
Binders are materials that help to hold other materials together. Most 
materials require some binder to assist in the formulation of a granulation 
suitable for tablet compression. Compared to conventional tablets, the use 
of binders in effervescent tablet formulation is limited, not because binders 
are unnecessary but because of the two-way action of the binders themselves. 
The use of any binder, even one that is water-soluble, will retard 
the disintegration of an effervescent tablet. In granulations that require 
a binder for tableting, a proper balance must be chosen between granule 
cohesiveness and desired tablet disintegration. Binders such as the natural 
and cellulose gums, gelatin, and starch paste are generally not useful due 
to their slow solubilfty or high residual water content. Dry binders such 
as lactose, dextrose, and mannitol can be used but are often not effective 
in the low concentrations normally permissible in effervescent tablets due to 
their disintegration-hindering properties as well as weight/volume restraints. 
Most effervescent tablets are composed primarily of ingredients needed 
to produce effervescence or to carry out the function of the tablet. 
Usually there is little room for excipients, which are needed in large 
concentrations to be effective. Polyvinylpyrrolidone (PVP) is an effective 
effervescent tablet binder. This material is usually added to the powders 
to be granulated either dry, and subsequently wetted with the granulating 
fluid, or in a solution with aqueous, alcoholic, or hydroalcoholic granulating 
fluids. Isopropanol and ethanol exert no binding effects themselves but are 
used in granulating fluids as solvents for the dry binders such as PVP. 
Water is useful both as a solvent for dry binders and as a binder itself. 
A small amount of water carefully added, and controlled to prevent initiation 
of the effervescent reaction, is very effective as a binder because of a 
partial dissolution of the raw materials followed by subsequent crystallization 
on drying. Procedures for manufacturing effervescent tablets using this 
technique are discussed in Section III. The hazards and solvent recovery 
problems associated with the organic solvents are common to the manufacture 
of both effervescent and conventional tablets. 
F. Diluents 
Due to the nature of the ingredients in an effervescent tablet, there is 
normally little need for added dUuents. The effervescent materials themselves 
are usually present in large enough quantity to preclude the use of 
diluents to achieve the desired tablet bulk. Sodium bicarbonate is as useful 
and inexpensive a filler as any, provided the extra effervescence and 
solution pH effects do not pose a problem. Other materials that are considered 
should be readily soluble, available in a particle size similar to that 
of the other ingredients in the product, and crystalline in nature to provide 
adequate compressibility. Examples are sodium chloride and sodium 
sulfate. Both of these substances are relatively dense and may be useful 
in producing a more dense tablet compaction if desired.

292 Mohrle 
G. Lub ricants 
Of all the ingredients compounded into effervescent tablets, the lubricant 
is one of the most important because without this material production of 
effervescent tablets on high -speed equipment would not be possible. Effervescent 
granulations are inherently difficult to lubricate, partly due to 
the nature of the raw materials used and partly due to the rapid tablet 
disintegration usually required. Many substances are effective lubricants 
in certain concentrations but inhibit tablet disintegration at these same 
concentrations. When the concentration is lowered to permit the tablet to 
properly disintegrate, the lUbricating efficiency of the material is lost or 
so greatly diminished that it is no longer useful. If a clear solution is 
desired when the tablet disintegrates, the problem is even greater since 
the most efficient Iubrteants are water-insoluble and will leave a cloudy 
solution once dispersed. 
Excellent articles pertaining to the fundamental aspects of tablet lubrication 
and the mechanism of action and evaluation of tablet lubricants have 
been published [39,40]. In the latter article, 70 materials were evaluated 
as tablet lubricants, some of them water-soluble and therefore of particular 
interest to the effervescent tablet formulator. 
Intrinsic lubrication is provided by those materials that are compounded 
directly into the tablet as the granulation is being prepared. This is the 
most efficient and most used method. The magnesium, calcium, and zinc 
salts of stearic acid are the most efficient substances commonly used. Concentrations 
of 1% or less are usually effective; however, they are not watersoluble, 
can hinder tablet disintegration, and produce cloudy solutions. 
Talc and powdered polytetrafIuoroethylene are also insoluble in water but 
generally permit more rapid tablet disintegration. The water-soluble or 
dispersible materials discussed in the remaining paragraphs of this section 
can be used. All are less efficient than the stearates but may provide the 
needed properties if the concentration is high enough. All solid materials 
must be finely divided, and in some cases micronized, to act efficiently. 
Liquids are more easily handled if they are dispersed on a granulation component 
prior to addition. 
Powdered sodium benzoate and micronized polyethylene glycol 8000 are 
effective water-soluble lubricants. It has been found in one case that the 
addition of sodium benzoate promotes tablet disintegration rather than prod 
ucing an inhibiting effect [41]. An improvement in the efficiency of 
sodium benzoate was seen by the incorporation of paraffins I dimethicone I 
or polyoxyethylene glycols [42]. 
Sodium stearate and sodium oleate are soluble in low concentrations; 
therefore, a combination of small amounts of both may be effective. The 
taste of these materials may be objectionable for an ingested product. Cottonseed, 
corn, and mineral oils all have Iubricating properties and will disperse 
in water. Polyvinylpyrrolidone [43], powdered sodium acetate, and 
impalpable boric acid have also been used as soluble Iubrioants as well as 
powdered adipic acid [44], powdered succinic acid [45], and powdered 
fumaric acid [46]. An interesting soluble Iubr-ieant , although rarely used 
in effervescent tablets due to its extremely high cost, is L-Iycine. This 
amino acid is highly efficient, having a stereochemical structure similar to 
that of graphite. It is most often used to lubricate noneffervescent hypodermic 
tablets that must completely dissolve prior to injection.

Effervescent Tablets 293 
The surfactants that are contained in some formulations to provide cleaning 
or detergent solutions also act as lubricants. Sodium lauryl sulfate is an 
effective lubricant but can hinder tablet disintegration if present in too 
high a concentration. Magnesium lauryl sulfate will also provide lUbricating 
properties with a minimal disintegration -hindering effect. A mixture of 
spray-dried magnesium lauryl sulfate powder and micronized polyethylene 
glycol polymers has been found to be an excellent water-soluble lubricant 
for effervescent denture cleanser tablets [47]. 
Acetylsalicylic acid crystals provide adequate Iubrtoating properties so 
that effervescent analgesic formulations containing this substance at effective 
dose levels usually do not require additional lubricants. 
Extrinsic lubrication is provided by a mechanism that applies a lubricating 
substance to the tableting tool surface during processing. In one 
method, a film of melted wax is sprayed onto the tool surfaces after one 
tablet is ejected and before the granulation for the next tablet enters the 
die cavity. Accurate spray synchronization with minimal volume delivery 
and precise spray placement were troublesome when this experimental system 
was adopted for high-speed tablet production. Another method makes 
use of an oiled felt washer attached to the lower punch below the tip, 
which wipes the die cavity with each tablet ejection. Neither of these 
methods is as good as adding lubricating substances directly to the granulation, 
as directed above. 
Another Iubr-icating procedure can be used with tablet presses having 
two compression cycles used for the production of multilayer tablets [48,49]. 
A tablet containing a high concentration of lubricant is compressed at the 
first compression station. As this tablet is ejected, a film of lubricant is 
deposited over the die wall and punch surfaces. 
The effervescent tablet of interest is compressed at the second compression 
station, lubricated by the thin film previously deposited. Elegant effervescent 
tablets can be produced in this manner; however, the output of 
the double-rotary press is cut in half. The Iubrioattng tablets can be milled 
and reused but this further adds to the cost of the effervescent tablets. 
In addition, it is necessary to use care to prevent the lubricant and effervescent 
granulation from becoming mixed. 
The role of the lubricating substances can be eased somewhat if the 
following mechanical means are employed. All tablets expand slightly after 
compression due to the elasticity of their ingredients. The use of outwardtapered 
dies can promote an easier escape for an expanding compaction as 
it leaves the die cavity. It is also beneficial if the tableting tools are 
coated with materials having a low frictional resistance. Many materials, 
such as polytetrafluoroethylene, have been applied to tableting tools but 
h ave rapidly worn off during processing. Electroplating all compression 
surfaces with chromium, which resists wear, is helpful. 
H. Other Ingredients 
Effervescent tablets may contain ingredients other than those previously 
mentioned. All are related to some function of the tablet other than its 
effervescent system, and in some cases may consist of a large portion of the 
tablet. These ingredients include drugs such as analgesics, decongestants, 
antihistamines, potassium supplements, and antacids; oxidizing agents such 
as sodium per-borate or potassium monopersulfate are commonly found in

294 Mohrle 
denture cleaning compositions; flavoring, coloring, or sweetening agents are 
usually contained in tablets whose solutions are ingested. Often these materials 
can influence the perceived attractiveness of the effervescent solution. 
As with any formulation, all ingredients of an effervescent tablet 
must be carefully balanced to achieve the desired properties. 
A. Special Conditions 
The processing of effervescent tablets, although similar in many ways to 
the processing of conventional tablets, presents certain problems and employs 
methods that are not often found with the latter. Special environmental 
conditions are required. Low relative humidity and moderate-tocool 
temperatures in the processing areas are essential to prevent the granulations 
or tablets from sticking to machinery and from picking up moisture 
from the air. which can lead to tablet instability. 
The storage of unopened containers of raw materials need not be restricted 
to a low relative humidity area. Normal warehouse conditions are 
usually sufficient since the containers of most hygroscopic raw materials 
contain moisture barriers of some type to protect their contents. Once 
the container is opened, however, the unused portion should be protected 
from moisture by transfer to suitable containers or by storage in a lowhumidity 
area. Once effervescent reactants are mixed. storage in a lowhumidity 
area is essential, since adsorbed moisture can initiate the effervescent 
A maximum of 25% relative humidity at a controlled room temperature 
of 25C (72F) or less is usually satisfactory to avoid problems due to atmospheric 
moisture. Relative humidity is more correctly expressed as grains 
of moisture per pound of air at a specified temperature. If the amount of 
moisture remains constant while the temperature increases. the volume a 
pound of air occupies will increase and the relative humidity will fall. As 
such, relative humidity expressed in terms of percent or grains of moisture 
per pound of air must be accompanied by a value for temperature; otherwise 
the term lacks definition. A atudy of the geographic and chronological 
distribution of relative humidity in an effervescent manufacturing area illustrates 
the need to pay particular attention to environmental moisture 
control [50J. 
B. Equipment 
The processing equipment used to produce conventional tablets is adaptable 
to the production of effervescent tablets provided the operations conducted 
with the various mixers. blenders, mills, granulators, tablet presses, and 
ovens are done in a low-moisture atmosphere. Specialized equipment known 
as a Topo gr-anulator- has al so been used [51J. In order to produce effervescent 
granules, this self-contained device controls the effervescent 
reaction which occurs during processing with the addition of a solvent to 
the dry ingredients following by quick vacuum drying. This process is 
repeated until a surface passivation is reached that increases product stability. 
This phenomenon is discussed further in Section VI of this chapter.

Effervescent Tablets 
C. Wet Granulation 
The principle in preparing a granulation for effervescent tablets is basically 
the same as for conventional tablets. Wet-granulating techniques involve 
mixing the dry ingredients with a granulating fluid to produce a workable 
mass. The mass, which may be plastic and cohesive in nature, is reduced 
to an optimum particle size distribution and dried to produce a compressible 
granulation. Alternate procedures in which the formed mass is dried before 
particle size reduction are also possible. 
A more unconventional granulation for effervescent tablets is simply a 
mixture of loosely adhering particles to which a very small amount of granulating 
fluid (0.1 to 0.5%) has been added. The mixture I which appears 
dry, is tableted, directly followed by drying. A discussion of this process 
appears in Section IILC.3. Wet granulations can be prepared in three different 
manners: with the use of heat, with nonreactive liquids, and with 
reactive liquids. 
With Heat 
This classical method of preparing effervescent granulations involves the 
release of water from hydrated formulation ingredients at a low temperature 
to form the workable mass. The ingredient most often used for this purpose 
is hydrous citric acid which, when fully hydrated, contains about 
8.5% water. This process is very sporatic and difficult to control to achieve 
reproducible results. Often done in a static bed, the reaction is not uniform 
throughout the bed because the release of water, being temperaturedependent, 
is not uniform throughout the depth of the bed. A different 
approach to the preparation of effervescent granules with heat, not intended 
for, but adaptable to, granulations for tableting, has been reported in the 
literature [52]. The use of a special mixer, which generates the heat required 
to start the effervescent reaction solely by the frictional resistance 
of the mixer contents to turbulent, high - speed mixing, is described. 
With Nonreactive Liquids 
This method is more commonly employed and is similar to that used to prepare 
granulations for conventional tablets. Granulating fluids such as ethanol 
or isopropanol, in which the effervescent ingredients and most of the 
remaining ingredients are not soluble, are most often used. The granulating 
fluid is slowly added to the premixed formulation components in a 
suitable mixer until the fluid is uniformly distributed. Binders, which are 
required in many formulations. can be added to the dry ingredients and 
activated as the mass is wetted. Alcohol-soluble binders, such as PVP. 
can be dissolved in the granulating fluid prior to addition to the bulk. 
Binders added in this manner are usually more effective and can be used in 
lower concentrations with fewer negative effects on tablet disintegration. 
Once the mass is uniformly wetted, it is manually transferred to trays and 
dried in an oven. Automated systems have been designed to remove the 
granulation from the mixer and pass it through an oven on a continuous 
basis. (The latter method is more suitable for loosely bound particulate 
granulations. ) After the granulations are dried, they are reduced to the 
desired particle size by using appropriate mills Or granulators, and are collected 
in containers for future use or transferred directly to other mixers

296 Mohrle 
for further processing prior to tableting. The denture cleanser tablet 
formulation (Example 7) is an example of this process. 
The characteristics (e.g., uniformity, compressibility, and flowability) 
of a granulation to be tableted are the same for effervescent tablets as for 
conventional tablets, and are discussed elsewhere in this volume. 
An advantage of granulating with nonreactive liquids is that not all the 
ingredients of a formulation need to be subjected to contact with the granu1ating 
fluids or to the heat of the drying process. In some formulations, 
it may be desirable to granulate the acidic and basic effervescent components 
separately to eliminate any reaction. Heat-labile compounds can be 
added subsequently to the granulation phase, and bulk raw material-handling 
requirements can be reduced if some of the formulation components have inherently 
suitable tableting characteristics and need not be granulated. 
One disadvantage is that some processing is still required after the 
granulation has been dried and ground. Most often. this entails additional 
mixing to blend more of the separate granulations or add heat-labile compounds 
or tableting lubricants. Additional grinding of the granulations 
can occur due to the attrition in the mixers. which may be detrimental to 
the granulatton particles. Another disadvantage is that the vapors from 
the granulating fluids are often hazardous and must be exhausted or condensed 
and collected. In any case, suitable ventilation must be provided 
to prevent dangerous levels of these solvents from accumulating. 
With Reactive Fluids 
One of the most effective gr-anulating agents for effervescent mixtures is 
water. Due to the fact that the effervescent reaction is initiated with 
water, obvious care must be taken to adequately control such a process if 
the effervescent character of the finished tablet is to be maintained. Often 
this process is difficult to control since the granulated mass must be quickly 
dried to stop the effervescent reaction. A process using water as a granulating 
agent for a mass-produced effervescent tablet has been developed 
and used for many years to produce tablets with final uses ranging from 
antacids to reconstitutable mouthwashes to denture cleanser tablets. 
This granulation process is based on the addition of small amounts of 
water (0.1 to 0.5%) to a blend of raw materials that possesses the uniformity, 
compressibility, and flowability needed to produce good-quality tablets, but 
which lacks the needed binding properties. The added free water acts as 
a binder. In practice, the water is usually added in the form of a flne 
spray to selected formulation components while mixing in a ribbon blender. 
When uniform distribution of the water has been achieved, the remaining 
constituents are consecutively added with adequate mixing to distribute the 
water throughout the mass. The fonnulation and process described for bath 
salt tablets (Example 8) illustrates this method. 
The ingredients selected to receive the water spray should readily release 
the adsorbed water to the rest of the formulation components rather 
than adsorb and bind it internally. After the formulation is complete, the 
free-flowing granulation is transferred directly to the tablet-compressing 
machines and tableted while moist. The compressed tablets are then passed 
through an oven, which causes the water to be removed or bound internally 
as water of crystallization and thus stabilized. Substantial increases in 
tablet hardness are usually experienced during the heating process. By 
using more than one blender to feed a common granulation transfer system, 
a continuous flow of granulation can be directed to the tablet presses from

Effervescent Tablets 297 
granulation prepared on a batch-to-batch basis. One distinct disadvantage 
of this process is that formulations which contain ingredients susceptible 
to attack from moisture and lor heat cannot be prepared without some degradation 
A wet granulation method using the simultaneous addition of water and 
application of heat in a vacuum oven has been described [53]. It is claimed 
that the tablets produced from the granulation had better carbon dioxide 
release properties with more rapid effervescence than from tablets produced 
by conventional means. 
O. Dry Granulation 
Dry granulation can be accomplished with the use of special processing 
equipment known as a roller compactor or chilsonator. These machines 
compress premixed powders between two counterrotating rollers under extreme 
pressure. The resultant material is in the form of a brittle ribbon, 
sheet, or piece-depending on the configuration 0 f the roller. The compressed 
material is reduced to the proper size for tablet granulation purposes. 
The toilet bowl cleanser tablets described in Example 10 are prepared 
by this process. 
Another dry granulation procedure is slugging, in which slugs or 
large tablets are compressed using heavy-duty tablet-compacting equipment 
and are subsequently ground to the desired granulation characteristics. 
Both of these processes are used for materials that ordinarily will not compress 
using the more conventional wet granulation techniques and require 
precompression to increase density or exclude entrapped air due to porosity. 
A simple blending of raw materials, which after mixing are suitable for 
direct compression into tablets, can also be considered a form of dry granulation. 
Measurements have been made of the mechanlcal properties of effervescent 
raw materials and mixtures to predict compressibility when directly 
compressed [54]. Fumaric acid had the best compression properties 
among the acids tested, while sodium bicarbonate was the best among the 
E. Fluidized Bed Granulation 
The production of effervescent granules that can be used to prepare effervescent 
tablets has been accomplished using fluidized bed granulation. 
A dry mixture of the powdered form of an acid and carbonate source 
is suspended in a stream of hot air, forming a constantly agitated, fluidized 
bed. An amount of granulating fluid, usually water, is introduced in a 
finely dispersed form causing momentary reaction before it is vaporized. 
This causes the ingredients to react to a limited extent forming single granules 
of the two reactive components. The granules are larger than the 
powder particles of the starting materials and suitable for compression into 
tablets after drying has been completed in the fluidlzea bed apparatus. 
This procedure has the advantage of ingredient mixing, granulattng, and 
drying all in one piece of equipment with minimal loss of carbon dioxide. 
Preparations containing aspirin and acetaminophen have been made using a 
fluidized bed temperature of 60 to 64C [55,56] with the effervescent granulation 
dried to a water content of 0.25%. The effect of the ratio of citric 
acid, sodium bicarbonate, and the PVP content of the granulating fluid as

298 Mohrle 
well as the temperature and rate of the input air was studied in a factorial 
design using a fluidized bed apparatus [57J. Granule size, fine-powder 
content, and the dissolution rate of aspirin tablets made using the resultant 
granules were measured. All the parameters affected the value of the 
fine-powder content; however, only the ratio of the reactants and the PVP 
concentration in the granulation fluid affected the dissolution rate of the 
tablets. It was concluded that a maximum content of 20% fine powder was 
optimum to achieve the desired tablet dissolution time of 120 sec. 
F. Pretableting Operations 
In many cases. after the effervescent portion of the granulation is prepared. 
certain materials are added that were purposely withheld during the granulation 
process. These materials are most often those that would be degraded 
by the heat or moisture present in granulation preparation (e.g 
acetylsalicylic acid, enzymes. and gragrance oils) or are those added in 
the final stages before tableting. such as lubricants. LUbricating powders 
are added as near to the end of the granulation process as possible in order 
to coat the rest of the granulation and provide their maximum effect. 
Ingredients present in small amounts should be added using geometric dilution 
techniques to ensure even distribution throughout the granulation. 
If liquid ingredients such as fragrances or oil lubricants are to be incorporated 
into the formulation, it is desirable to separately mix them thoroughly 
with a small portion of the total granulation or one of the formulation ingredients. 
and then add this wetted mixture to the remainder of the granulation. 
If the oils are added directly to the granulation, an even distributian 
is difficult to obtain. and small lumps are likely to occur throughout 
the granulation. Oils are effectively distributed when premixed with granuIar 
sodium bicarbonate. 
The ideal granulating and tableting operation from a cost and efficiency 
point of view is one of direct compression without prior granulation. This 
process, which may be feasible for some effervescent tablets, is difficult 
to carry out in general. In order to be directly compressible, the particle 
size distribution of the raw materials used in the formulation should be 
roughly the same and have inherent compressible properties. Many of the 
raw materials used in effervescent tablets are available in a fine granular 
form. Others, which are available in larger particle sizes, can be ground 
to the desired size. The problem occurs when a large portion of the formulation 
is composed of particles that are smaller than average. In this case 
granulation is required. The addition of small amounts of finely powdered 
substances can usually be accommodated if the bulk of the formulation is 
granular and free flowing , in which case the fine particles fill in the voids 
among the larger particles and become thoroughly mixed throughout the 
granulation. If the granulations are properly prepared, tableting operations 
will run smoothly. 
G. Tableting 
Effervescent granulations are tableted in the same manner as conventional 
tablet granulations (discussed in detail in other sections of this volume). 
Common process controls are tablet weight, thickness, and hardness.

Effervescent Tablets 299 
Once the tablet presses are operating and have been properly adjusted. 
these parameters will be relatively constant if the granulation is of good 
quality. Significant variations indicate the development of problems. and 
the tablets leaving the press should be examined closely tor signs of difficulty. 
If problems occur, they will most often be caused by insufficient 
binding (evidenced by laminating or capped tablets) and inadequate lubrication 
(evidenced by tablet surface picking and die wall sticking). Since 
many effervescent tablets are large in diameter. laminations or capping can 
be detected easily by snapping the tablet between the thumb, forefinger. 
and middle finger across the diameter of the circular surface of the tablet. 
Examination of the broken interface will reveal the presence of a lamination 
if definite layers or striations Can be seen within the tablet. As the severity 
of lamination increases, capping becomes evident (I ,e . the top surface 
of the tablet splits from the body of the tablet). A downward adjustment 
in tablet hardness may eliminate this problem if tablets of adequate quality 
can be produced at hardness levels below that at which capping occurs. 
A sudden drop in tablet hardness with a concomitant increase in the pressure 
adjustment (which normally raises the hardness) is indicative of the 
failure of the binding system at that pressure. A reduction in pressure 
should result in a return to tablets of expected quality. 
Evidence of lubrication difficulties can be observed by a loss in the 
gloss or shine on the surface of the tablet when held so that light is seen 
reflecting from it. Granulation sticking to the tablet tools will produce. in 
the tablet surface I small indentations called picking. Careful observation 
of the tablet edge can detect early stages of die wall sticking. seen as lines 
or scratches perpendicular to the tablet face. These are caused by small 
amounts of granulation adhering to the die wall. If not remedied. this situation 
will increase in severity and the tablets will not eject freely from the 
die cavity. 
Modifications in the binder and lubrication systems contained in the 
formulation can solve these problems; but as previously mentioned. the effects 
of both binders and lubricants are detrimental to tablet disintegration 
and, in the case of lubricants. the hardness of tablets. Formulations must 
be individually tailored to achieve adequate binding and lubrication with 
minimal negative effects to the finished tablets. A complete factorial design 
experiment has been carried out to study the influence of compression 
force, drug content. and particle size of ingredients on the hardness of 
effervescent aspirin tablets [58]. The interactions between compression 
force and drug content as well as compression force and ingredient particle 
size were found to be significant. The interactions between drug content 
and ingredient particle size as well as the interaction among all three parameters 
had no marked effect on tablet hardness. Information gathered 
from studies of this type can be useful in preparing high-quality effervescent 
The preparation of two-layer effervescent tablets is possible but requires 
special tablettng equipment. It is more difficult since adequate 
binding and lubrication are needed for both layers. which usually differ 
from each other in composition. This technique is used to separate active 
ingredients for stability purposes and to create a visual difference between 
layers with the use of colors [59]. Compositional differences will allow 
each layer to effervesce at a different rate. This is useful for a color 
effect in solution when different dyes are used or for functional reasons

300 Mohrle 
when release of one ingredient into solution prior to a second is desired. 
The pH of the solution can be controlled in this manner if, for instance. 
the more rapidly soluble layer is acidic, which is subsequently neutralized 
ane even alkalinized as the basic layer dissolves. In conventional tablets, 
the separation of drugs for stability reasons is usually accomplished by 
encapsulation. Bncepsulatad materials frequently are not acceptable in effervescent 
preparations. if clear solutions are desired. due to their slow 
rates of solubility in water. 
Molded rather than compressed effervescent tablets have been prepared 
[60]. These tablets, which contain about 30% void space, are rapidly soluble 
in iced liquids. They are formed by triturating acid and carbonate 
powders with a limited amount of water containing up to 10% of a volatile 
organic solvent such as ethanol. The wet mass is molded into a tablet form 
and dried at 50C. Evaporation of the volatile solvent causes the void 
space. which permits very rapid solution in cold liquids. Tablets containing 
sweetening agents, analgesics. and disinfectants have been produced 
using this procedure. 
The large-scale manufacture of effervescent tablets is best done using a 
batch-continuous type of procedure. As with most tablet-making processes 
that require a granulation step. a continuous feed-in and feed-out system 
is not suitable. An exception would be an extrusion process that allows 
for a continuous flow of material during granulation. 
Two different processes are illustrated in Figures 1 and 2. The process 
in Figure 1 requires more equipment and space than that depicted in 
I I 
I  I IL___.., 
I  I 
I L______ ---- ..J 
Figure 1 Manufacturing flow chart.

Effervescent Tablets 30. 
I I .2 
I EOUIPMENT I L ________
----- --- -- - - 
Figure 2 Manufacturing flow chart. 
Figure 2 if a continuous flow of material is to be obtained. In this process 
the raw materials are brought from a storage or staging area to the manufacturing 
area and weighed into proper batch quantities. Any p reproceaaing 
that may be required, such as grinding, is done before weighing takes place 
since the yield from exact batch quantities subjected to preprocessing will 
be less than 100% due to factors such as loss in the equipment, spillage, 
and dust generation. 
Some manufacturers prefer to weigh the raw materials in an area other 
than the manufacturing area to reduce the chances of a compounding error 
in the quantity or in the specific raw material weighed. Once weighed, the 
raw materials are transferred to the appropriate mixers and the mass to be 
granulated is prepared. Smaller mixers are used for blending raw materials 
such as liquids or coloring substances prior to transfer to the larger blenders. 
Two blenders are used to prepare the granulation to provide a continuous 
flow of material. As one batch, just prepared. is being transferred 
to the drying ovens, a second batch is in the process of preparation. The 
time needed to prepare one batch should not be longer than the time needed 
to empty the other blender if a continuous flow of granulation is to be maintained. 
After passing through the drying oven, the mass is passed through 
appropriate equipment to produce the granules of desired size satisfactory 
for tableting. At this point. the granulation must be collected if heatlabile 
ingredients have been withheld for addition after the heating process 
has been completed. Appropriate quantities of the granulation are then reweighed 
and mixed with the withheld ingredients in the final blenders. 
Two blenders are also used at this point to provide a continuous flow of 
material as described above. Once mixed, the just completed granuletton 
is transferred to the tablet press; the compressed tablets to the stripwrapping 
equipment (see Section VILe for details of this operation); and 
the wrapped tablets to automatic cartoning and finally to storage as

302 Mohrle 
finished goods ready for shipment, after the needed quality assurance 
If no additional materials are to be added after granulating, the granu1ation 
can be transferred directly to the tablet press for compaction. as 
indicated by the dotted line in Figure 1, bypassing the second weighing 
and mixing operations. All process areas contained in the dashed line 
area should be maintained at the proper temperature and low humidity as 
described in Section III.A. The strip~wrapping equipment should be separated 
from the granulating and tableting equipment to minimize the possiblity 
of poor sealing characteristics due to airborne dust. 
The process illustrated in Figure 2 is that used when the granulation 
is prepared with a reactive fluid and no additional components are added 
after the granulatton is completed as described in Section III.C. 3. This 
process is similar to that described above I up to the point when the granulation 
leaves the blenders in which it is mixed. From the blender. the 
loosely bound granulation is transferred directly to the tablet press and 
the compacted tablets are passed through the stabilizing oven, the last 
portion of which is actually a cooling area to reduce temperature of the 
tablets before packaging. Note that in both processes. only the oven inlet 
and outlet are in the dehumidified area. Due to the high cost of temperature 
and humidity control equipment and the energy required for its 
operation, economic advantages can be realized if the environmental controlled 
area is kept to a minimum. The strip-wrapping and cartoning procedures 
are as previously described, with a separate area provided for the 
strip-wrapping equipment. 
The important parameters in the evaluation of effervescent tablets can be 
divided into physical and chemical properties. The evaluation of their effectiveness 
in their intended use (e.g., whether a denture cleanser tablet 
solution actually cleans dentures) is not discussed. 
A. Physical Parameters 
Tablet disintegration time is One of the most important characteristics since 
the visual effect of the dissolving tablet and its subsequent carbonation are 
the main reasons for the use of effervescent systems, other than providing 
a mechanism for tablet disintegration. Obviously, there is little advantage 
over compressed, noneffervescent tablets if rapid disintegration is not obtained. 
Previously discussed factors that can hinder tablet disintegration 
are excessive concentrations of water-insoluble materials 01' too efficient 
binder systems. Excessive tablet hardness can also reduce the expected 
rapidity of tablet disintegration. As with conventional tablets, disintegration 
is distinct from dissolution, since an effervescent tablet can disintegrate 
into slowly soluble fragments or particles. Usually this is a distinct negative 
since a slowly soluble residue is unsightly and the full effect of the 
functional ingredients is not Obtained unless they are in solution. A properly 
formulated effervescent tablet will disintegrate and dissolve quickly, 
usually in 1 or 2 min.

Effervescent Tablets 
Table 1 Volume and Temperature of Water Used in Effervescent Tablet 
Disintegration Testing 
Water volume Water temperature 
Tablet (ml) (OC) 
Antacidlanalgesic 120-180 15-20 
Denture cleanser 120-150 40-45 
Flavored beverage 180-240 10-15 
Mouthwash 20-30 25-25 
Toilet bowl cleaner 4000-6000 20-25 
Disintegration time tests involve placing the tablet in a standard volume 
of water at a specified initial water temperature and recording the time in 
which the tablet disintegrates. The volume and temperature of the water 
depend on the type of product being tested. It is most realistic if both 
are those to be used by the consumer. Examples are given in Table 1. 
often, effervescent tablets will float to the top of the solution prior to 
complete disintegration, making accurate disintegration time determination 
difficult. This occurs when the density of the tablet mass and bubbles adhering 
to it become less than the density of the solution in which it is disintegrating. 
Careful observation of the floating tablet as it crumbles is 
needed at this point to determine the actual disintegration time. 
Two other important effervescent tablet physical parameters are hardness 
and friability. As with conventional tablets, these criteria are interrelated, 
depending on the formulation components. Generally with effervescent 
tablets, the harder the tablet, t he lower the friability. Both of 
these parameters are indicative of how well the tablet wUl withstand the 
rigors of handling after compression. Automated packaging systems provide 
the most abuse to the tablet surfaces after compression. Many marketed 
effervescent tablets are large in diameter and chip easily at the edges 
during handling. 
The choice of proper tableting tools, especially beveled edge configurations, 
can minimize edge chipping. The relative hardness of effervescent 
tablets can be modified by adjusting the ratio of tablet thickness to tablet 
diameter. The closer this ratio is to I, the harder the tablet wUl be. 
Often this approach is not useful, however, because thick tablets are difficult 
to properly package in individual, hermetically sealed pouches. As 
the tablet becomes thicker, a greater strain is placed on the pouch seal 
area, increasing the probability of leaking pouches. This will be discussed 
in greater detail in the Section VII.C.5 of this chapter. Tablet hardness 
is measured using standard hardness testers available to the pharmaceutical 
industry. A Roche friabilator is useful in measuring tablet friability. 
Another important physical parameter is tablet weight. This is a function 
of the formulation and compressing equipment adjustments, as is the 
case with conventional tablets. Good manufacturing practice will result in 
tablets that conform to compendial weight variation tests discussed elsewhere 
in this volume.

304 MOhY'le 
B. Chemical Parameters 
An interesting chemical property. perhaps unique to effervescent tablets. 
is the solution pH generated when the tablet dissolves. Due to the nature 
of the effervescent system components. buffer systems are furmed. and 
thus discrete pH readings can be obtained. The consistent measurement 
of solution pH is a sign of good distribution of raw materials within the 
tablet. A wide variation in solution pH from tablet to tablet is indicative 
of a nonhomogeneous granulation directly prior to tableting , A consistent 
pH difference from that normally observed for a product in a particular batch 
of tablets is indicative of a compounding or raw material weighing error. The 
pH of the solution is important for taste reasons in a product meant for ingestion. 
Often antacid products are formulated to YIeld a slightly acidic pH to 
augment the taste of the solution. particularly if citrus or berry flavors are 
used. Products that are mint flavored are best formulated so that the solution 
is neutral or slightly alkaline. Solution pH can be functionally important 
for some effervescent products. A toilet bowl cleaner should be acidic rather 
than alkaline to dissolve calcium and iron deposits from the porcelain fixture. 
A denture cleanser can be acidic for maximum calculus solubility. or 
neutral to slightly alkaline for potentiation of the typical oxidizing agents 
used in these formulations. 
Solution pH is measured with suitable instrumentation in standardized 
water volumes and temperatures. It is conveniently done following disintegration 
time measurements. The pH should be measured at a specific 
time after the tablet has been placed in the water since it is not unusual 
for effervescent solutions to change in pH on standing. This is due to 
the constant breakdown of carbonic acid to carbon dioxide gas and water 
within the solution. If slowly soluble materials are present. adequate time 
should be given for the ingredients to dissolve (after tablet disintegration 
occurs) before pH measurements are made. 
Another important chemical parameter for tablets containing assayable 
active ingredients is content uniformity assay. This is the same for any 
tableted dosage form and is discussed elsewhere in this volume. A 10% 
variation from the theoretical amount of the active ingredient compounded 
into the product is usually acceptable. 
The stability of effervescent tablets can be discussed in two distinct parts. 
One deals with the degradation of drugs or other functionally active ingredients. 
the other with the stability of the effervescent system itself. 
They are not mutually exclusive. however. since if the portion of the tablet 
is unstable and has decomposed. the stability of the active assayable components 
is of little concern to the formulator. This section deals with the 
stability of the effervescent system common to all effervescent tablets and 
not the stability of particular components compounded into effervescent 
systems that are peculiar to each formulation. 
Effervescent systems are not stable in the presence of moisture. 
Trace amounts of moisture can activate the effervescent system during prolonged 
storage and decompose the tablet prior to use. To make matters 
worse. effervescent tablets are hygroscopic and will absorb enough water 
to initiate degradation if not properly packaged.

Effervescent Tablets 
A. Methods of Achieving Stability 
The elimination or inactivation of free water within the effervescent system 
is the key to stability, aside from manufacturing effervescent tablets in 
controlled low-humidity atmospheres. A choice of the proper types of raw 
materials is essential. Unless a hydrated form of a raw material is chosen 
purposely. all raw materials used should be in their anhydrous form. 
Many anhydrous raw materials that are prepared from hydrated forms become 
hygroscopic and readily adsorb water vapor from the air. In such 
cases, drying of these raw materials prior to use can be critical. The 
knowledge of possible hydrates formed during processing (which are not 
present at the outset) is useful, especially if free water is used as a granulating 
Materials such as citric acid and sodium carbonate will form hydrates 
readily. It is possible that the tablets containing hydrates formed during 
processing will appear to be stable when first examined but will slowly decompose 
as these hydrates are released with time. Some materials. such 
as anhydrous citrate salts. will form stable hydrates and act as effective 
internal desiccants to actually increase the stability of the tablet with time 
under certain conditions. Finely divided silica gel has also been used as 
an internal desiccant in effervescent tablets [61]. 
Finely divided anhydrous sodium carbonate has been found to be an 
effective stabilizing agent for effervescent tablets when incorporated into 
formulations at about 10% wIw of the sodium bicarbonate concentration. It 
is theorized that the anhydrous salt preferentially absorbs any minute trace 
of free water present, producing stable hydrated forms. Another method 
of sodium carbonate stabilization is found in the patent literature [62]. In 
this method. the sodium bicarbonate used in the formulation is heated so 
that 2 to 10% wIw is converted to sodium carbonate. Stabilization of the 
effervescent system results from the chemical change that occurs on the 
outer particle surface forming a barrier to hinder reaction with the acid 
source contained in the formulation. 
Surface passivation of a solid acid has been found to improve effervescent 
product stability [63,64]. The acid and a carbonate source are heated 
from 40 to 80C in a closed vessel. A polar solvent such as water is introduced 
and rapidly vacuum-dried. The degree of passivation is measured by 
monitoring the pressure increase in the vessel during processing. The 
procedure is repeated until no further pressure increase is observed, indicating 
that surface passivation has been achieved. It is claimed that the 
dry mixtures are highly stable even on storage under tropical conditions. 
In another case [65]. crystals of solid acid are coated with calcium carbonate 
which adheres to the acid crystal surface by a bonding layer formed 
during processing. 
The addition of substances that decrease the hygroscopicity of the effervescent 
mixtures can also provide a stabilizing effect. Encap~ulation of 
the acid and lor carbonate phases has been accomplished using PVP and 
hydroxypropylcellulose [66J, methacrylic acid polymers [67]. and maltodextrin 
[68]. In general. however. tablet solubility is reduced due to the 
slowly soluble nature of the encapsulating materials. Stabilization of the 
effervescent system is also possible if the sodium bicarbonate is mixed with 
a dilute solution of gum followed by drying and particle size reduction [69]. 
or if as little as 0.5% of lactic albumin. casein. or soya bean albumin is 
mixed with the total effervescent preparation [70].

306 Mohrl.e 
Another method of achieving effervescent stability was accomplished by 
producing a cored tablet in which the inner core of effervescent materials 
was protected from moisutre absorption by coating with a sugar alcohol 
such as sorbitol [71l. This tablet was meant for oral use rather than for 
dissolution in water. 
B. Stability Testing and Shelf-Life 
The stability testing and shelf-life prediction of' effervescent tablets are not 
complicated, and the usual Arrhenius equation kinetic principles can be applied 
to the data obtained from the following tests. Each tablet is hermetically 
sealed in a standard size aluminum foil laminate pouch. The pouches are 
placed at 25. 37, 45, and GOoC after the thickness of the tablet and the 
foil pouch is measured and recorded. If decomposition occurs. small amounts 
of carbon dioxide gas will be released into the pouch. causing it to swell. 
The degree of swelling. as measured by increase in pouch thickness, is related 
to the amount of gas evolved. An apparatus can be constructed so 
that the initial thickness of the packaged tablet is assigned a zero reading 
on an adjustable measurement scale While a constant weight is applied. An 
increase less than 1/16 in. is considered negligible. Even though the decomposition 
of the product may be small, pouch swelling is considered an 
important criterion of stability. Most effervescent products are sold packaged 
in this manner, and swollen packages are not readily accepted by the 
Another method of measuring the stability of the effervescent system 
with time is to assay the tablet themselves for total carbon dioxide content. 
This is easily done following established procedures for baking powder assays 
as developed by the Association of Agricultural Chemists [72]. This 
method uses liquid volume displacement equipment known as the Chittick 
apparatus. A tablet sample is crushed, and a portion of powder is accurately 
weighed and placed in a flask, into which is introduced an acid-water 
solution. The amount of carbon dioxide liberated from the sample is measured 
volumetrically by the displacement of a non-carbon-dioxide-adsorbing 
solution contained in a graduated cylinder. It is essential that the solution 
be nonabsorbent of carbon dioxide. The weight percent of carbon dioxide 
is then calculated using temperature and pressure corrections. 
Tablet disintegration time is another measure of effervescent stability. 
If carbon dioxide is lost due to chemical decomposition within the dosage 
form, the tablet will not disintegrate as rapidly as when it was initially 
prepared. Using the techniques to measure disintegration time previously 
described in this chapter, a record is kept of the tablets' disintegration 
characteristics when stored at elevated temperatures for varying lengths 
of time. If the disintegration time exceeds the previously established acceptable 
limit when stored for less than 3 months at 45C. 6 months at 37C, 
or 24 months at 25C, evidence of decomposition exists and should be investigated. 
Quantification of an effervescent reaction to monitor the stability of 
selected effervescent tablet systems has been studied [73]. Two devices 
were developed to monitor the reactivity of pharmaceutical effervescent systems. 
The first device monitored carbon dioxide pressure generation during 
the effervescent reaction in a specially constructed cylindrical plastic pressure 
vessel that allowed mixing of the sample tablet and water after the

Effervescent Tablets 307 
unit had been sealed. At standard time intervals the pressure was read 
from the pressure gage fitted to the vessel and recorded. The dissolution 
time of the tablet was observed through a transparent position of the pressure 
vessel and was also recorded. The second device utilized a doublecantilever 
beam and an electomagnetic proximity transducer to measure the 
weight loss attributed to carbon dioxide loss to the atmosphere. Tablet 
dissolution time was also observed and recorded. A correlation coefficient 
of 0.937 was calculated from a plot of the relationship of pressure generated 
versus weight loss for a series of experimental effervescent tablets. Using 
these data an index of reactivity was calculated that can be used to quantitate 
the effervescent activity from a particular system. Loss of reactivity 
with time as a quantitative measure of stability of the system can therefore 
be monitored using this technique. 
Further work was done combining these techniques with mercury intrusion 
porosity measurements to determine the effects of compression pressure, 
water vapor, and high temperature on effervescent tablet stability [74]. It 
was found that compression pressure was not a factor in tablet stability. 
The stability was dependent, however, on the tablet formulation, storage 
conditions, and the length of time the tablet was stored. 
C. In-Process Stabi Iity Measurements 
Obviously, it is not an acceptable technique to place samples of each batch 
of tablets at elevated temperatures and wait for swelling to occur to determine 
if the tablets are unstable. Quick, accurate, in-process quality assurance 
methods are needed to determine if each batch of tablets will be 
stable for the expected shelf life of the product. Since any decomposition 
is triggered by trace amounts of water, several methods have been devised 
to measure the residual water content either directly or indirectly. 
Conventional loss-an-drying methods are not useful for effervescent 
systems containing carbonates since the heat generated in the test apparatus 
will drive off carbon dioxide gas, producing false weight-loss readings. 
Water assay using the Karl Fischer titration procedure usually is not useful 
since the water content being measured is too low to be determined accurately. 
A better method, but still not ideal, is vacuum drying to a constant 
tablet weight over concentrated sulfuric acid. This procedure, aside from 
being time consuming and potentially hazardous due to the acid used, lacks 
the accuracy needed. It also probably would not detect initially stable 
hydrates furmed during processing, which could subsequently decompose 
and release free water, initiating the effervescent reaction. 
An acceptable technique using a modified Parr calorimeter (illustrated 
in Fig. 3) has been used for many years. The tablets are sealed inside a 
closed chamber fitted with a pressure gage with a scale ranging from 0 to 
60 lb in. -2. Enough tablets should be placed in the chamber so as to leave 
a minimum air space in order to avoid erroneous readings due to air expansion. 
As heat is applied externally from a constant temperature source. 
any trace amount of water will be liberated, causing the effervescent reaction 
to begin and release carbon dioxide. The pressure from the gas 
evolved is measured on the gage, being directly related to the amount of 
potentially troublesome water contained in the tablet. Through experimentation, 
it is possible to produce a stable effervescent tablet that, When tested 
using this procedure, gives a mid-range reading on the pressure gage. To

308 Mohrle 
Figure 3 Modified calorimeter used for stability measurement. 
do this, a correlation among the moisture content of the tablets, elevatedtemperature 
stability testing, the test bath temperat ure , and the time of 
exposure in the bath must be made. 
In practice, several batches of the product are prepared with moisture 
content varying from batch to batch. These are packaged and placed in 
environments with a range of controlled temperatures such as 25, 37, 45, 
and 60oC. With time. differences in the stability of the test products will 
become evident. and a dividing line between a stable and an unstable product 
can be determined. Concurrently, calorimeter tests are conducted at 
varying temperatures and lengths of exposure to the temperatures until a 
reasonable range of values relating to the moisture content of each batch 
of product is determined. These data are then correlated with those of 
the elevated temperature tests. resulting in a specification for a stable 
product as measured by the calorimeter test. The established specification 
and test method can easily be incorporated into quality assurance procedures 
since the calorimeter method is rapid and reliable. In a comparison of two 
products, data accumulated for one product-even though both are similar 
in effervescent composition-should not be related to the other, especially 
if addrtives in the first differ from those in the second. Each product 
should be thoroughly tested according to the above procedure and assigned 
its own stability specification. An example of determining a specification 
for an effervescent tablet with a diameter of 0.75 in. and thickness of 0.20 
in. follows. 
Enough tablets are placed in a stainless tube with an internal diameter 
of 0.90 in. so that the top tablet is 0.25 in. below the top of the tube, in

Effervescent Tablets 309 
this case 10 tablets. The pressure gage is attached, the device is sealed 
and placed in a constant temperature bath so that the liquid in the bath 
covers the stainless steel tube. Trials are conducted at 75, 80, and 85C 
and pressure readings are recorded at 30, 45, 60, 90, and 120 min. Similar 
data are obtained for two additional batches of the same formulation with 
different moisture contents resulting from varying oven-drying procedures. 
The data obtained are shown in Table 2. Additional stability testing with 
these same products using the packet puffing and carbon dioxide measurement 
testing described above indicate that only the low moisture level tablets 
were acceptably stable after 3 months storage at 45C. Therefore the 
data in Table 2 for the low moisture level tablets can be used to determine 
the specification. A good choice for this example would be to accept any 
batch whose readings are not greater than 15 psig when tested for 60 min. 
Any of the values could be used. However, it is best to avoid the 
low or very high pressure readings On the pressure gage scale for accuracy 
and to allow enough time in the bath to adequately heat the tube contents 
causing decomposition, if it is to occur. 
Table 2 Pressure Readings (psig) Obtained During Stability Specification 
Determination Testing for 8 Particular Effervescent Tablet Formulation 
Bath Time (min) 
(OC) 30 45 60 90 120 
HIgh Moisture Level Tablets 
75 8 14 18 27 37 
80 12 20 27 40 54 
85 15 26 38 51 60+ 
Medium Moisture Level Tablets 
75 6 12 16 22 28 
80 9 17 23 31 41 
85 13 24 33 46 52 
Low Moisture Level Tablets 
75 4 6 10 13 16 
80 5 8 15 18 23 
85 7 9 20 26 34

310 Mohrle 
A. Moisture Control 
Since effervescent tablets are hygroscopic, they must be protected from 
atmospheric moisture if a reasonable shelf life is to be expected. Any absorption 
of moisture will initiate the effervescent reaction; therefore packages 
for effervescent tablets must have hermetic seals regardless of the 
type of container. Multiuse containers, such as tubes or bottles, must 
have closures that can be resealed after each tablet is removed. Packaging 
operations must be conducted in Iow-humidtty environments (maximum 25% 
relative humidity at 25C) similar to those required for granulating and 
tableting if the long-term stability of the tablets is to be maintained. 
B. Packaging Configurations and Materials 
Effervescent tablets are usually packaged in glass, plastic. or metal tubes 
or individual foil pouches joined to form a conveniently sized strip of tablets. 
Glass offers the highest degree of moisture protection of the nonflexible 
packaging materials; however, inherent limitations exist, such as breakage 
and cost of shipping a heavy package. Since individual packaging in glass 
is economically infeasible, moisture-proof closures for these multiple-use 
containers must be used. Metal caps with a waxed, aluminum foil, pulpbacked 
cap liner usually prove satisfactory when repeatedly opened and 
closed. If properly closed after each use, moisture is excluded from the 
interior of the package. Many effervescent tablets are rather large, approaching 
1 in. in diameter, and do not lend themselves to random filling 
in a glass bottle as would smaller tablets. These tablets are packaged by 
stacking them one on another in a glass or plastic tube slightly larger in 
diameter than the tablets and about 5 in. high. 
In this manner, a minimum of air space surrounds the tablet prior to 
use. Since moisture can enter a glass container only through the closure, 
the top tablet serves as a desiccant and protects the rest of the tablets in 
the package. Once opened, however, protection from moisture is diminished 
because the air space becomes greater and greater as the tablets are used. 
This can be especially troublesome if humid air is permitted to enter the 
tube. In any event, the tablets should be used promptly or the last few 
will be nonreactive when placed in water, due to complete reaction which 
has slowly occurred in the container prior to use. Plastic tubes are not 
as protective as glass due to the moisture vapor permeability of plastic packaging 
materials. Tablets with a low order of hygroscopicity can be satisfactorily 
packaged in plastic tubes with moisture-proof closures. Special 
caps can be constructed with a chamber containing silica gel or some other 
desiccant that will preferentially absorb moisture vapor entering through the 
closure. Extruded, seamless metal tubes, often made from aluminum, have 
been used commonly in Europe to package effervescent tablets. These are 
impervious to moisture as are glass tubes. Tightly fitting plastic snap caps 
that may contain a desiccant chamber are used as closures. 
Effervescent tablets are most frequently strip-wrapped in individual 
pouches arranged in conveniently sized strips and stacked in a paperboard 
box. Each tablet is hermetically sealed in its own container and is not exposed 
to the atmosphere until the time of use. Many different flexible 
packaging materials are available for packaging, but few are suitable for

Effervescent Tablets 311 
protecting effervescent tablets from moisture vapor or physical damage. 
Some effervescent tablets produced in Europe are available packaged in 
thermoformed plastic blisters with foil backing. This type of packaging requires 
that the tablet s be pushed through the foil backing by pressing on 
the blister. The tablets packaged in this manner must be hard enough so 
as not to break when they are removed from their package. Most largediameter, 
relatively thin effervescent tablets cannot be made hard enough 
to withstand the force required to remove them from this type of packaging. 
A transparent film known as Aclar has a very low moisture-vapor transmission 
rate and will suitably protect effervescent tablets, but it is too 
axpensive to be competitive with the standard material used industrywide 
(Le , . heat-sealable, aluminum foil laminates). 
Aluminum foil is a flexible, absolute barrier to gases, water vapor, and 
light. It is nontoxic and immune to microbiological attack. It has excellent 
heat conductivity, thereby making it an excellent choice for heat-sealing 
strip -p ackaging op erations . 
Aluminum foil laminates are composed of several layers of different materials 
bonded together. A primer or wash is applied to the surface of 
one layer to promote bonding of the adjacent layer. Shellac or ethyl acrylic 
acid copolymers are commonly used as primers. The outside layer of the 
laminate is typically some form of paper, perhaps glassine, bond, or calendered 
(compressed) pouch paper. This layer provides a surface for printing, 
protects the foil against abrasion, and provides mechanical support for 
the entire laminate. A printed laminate allows identification of each tablet 
unit after removal from the strip. The next layer is polyethylene-about 
0.005 in. thick - which bonds the paper to the aluminum foil layer. The 
aluminum foil can range in thickness from 0.00035 in. to 0.002 in. Foil of 
0.001 in. thickness will impart the needed barrier properties to the packaging 
laminate. Thinner materials can be used, but a loss in moisture protection 
can occur due to the possibility of pinholes in the foil through 
which moisture vapor will pass. The thinner the foil is rolled, the more 
pinholes will be present. The inside layer consists of a heat-sealable material 
such as polyethylene, also about 0.001 in. thick. Laminates are also 
available with heat seals consisting of acrylic copolymers such as Surlyn. 
A typical laminate structure is illustrated in Figure 4. 
New laminates containing a stretchable aluminum foil alloy sandwiched 
between two plastic stretchable films have been developed in Europe 
(Alusuisse Metals Company, Singen, West Germany) for use on equipment 
which produces a foil blister by mechanically drawing the laminate into a 
machined cavity without the use of heat. The outer film, which is usually 
biaxially oriented nylon or polypropylene, provides strength to the laminate 
to prevent foil rupture during the cold forming process. The inside layer, 
which comes in contact with the product, is usually made from polyvinylchloride 
or polyethylene depending on the compatability requirements of 
the product. 
c. Strip Wrapping 
In a typical packaging operation, a diagram of which is shown in Figure 5, 
two sheets (or webs) of the laminate converge and pass between a psir of 
matching heated cylinders, each containing exactly corresponding cavities 
appropriate in depth and dimension to the tablet to be packaged. The

Effervescent Tablets 
Figure 4 Typical packaging laminate structure. 
tablets are fed between the converging sheets synchronously with the cylinder 
cavities so that they are not crushed. The two sheets of foil laminate 
around each cavity are heated by contact with the cylinder surface and 
subjected to pressure between the cylinders, forming the heat seal. The 
cylinders are engraved with a knurled or cross- hatched pattern to ensure 
an effective seal. As the formed pouch leaves the heated cylinders, the 
temperature of the laminate falls, causing the two heat seal layers to bond. 
The sheet is then automatically cut into the proper configuration and perforated 
to allow the removal of one pouch without disturbing the sealed area 
Figure 5 Strip wrap packaging.

Effervescent Tablets 
Figure 6 Poor foil laminate seals: (a) foil fracture, (b) wrinkling in the 
laminate. (c) foreign matter in seal. and (d) stress on the foil. 
of the adjacent pouch. The seal integrity of the completed pouches is of 
prime importance because without a good seal, moisture will enter the pouch 
and decompose the tablet prior to use by the customer. 
Poor seals in high-speed strip-wrapping operations can result from a 
number of sources, illustrated in Figure 6 and discussed in the following 
Temperature of Sealing Roller Too Low 
High-speed equipment is capable of wrapping in excess of 800 tablets per 
minute. At these speeds. the contact time between the foil laminate and 
the heat sealing roller is short. Even though aluminum is a good conductor 
of heat. it may not transfer the heat from the sealing roller to the thermoplastic 
heat-seal material fast enough to effect a good seal. If, at a maximum 
sealing roller temperature, adequate sealing does not take place and 
production speeds cannot be decreased to extend the contact time between 
the laminate and the roller, preheating of the laminate is advisable. This can 
be accomplished by the use of preheat rollers over which the laminate passes 
immediately prior to contact with the sealing roller. The preheat rollers will 
heat the laminate to a point just below the melting point of the thermoplastic 
heat seal and facilitate complete sealing in a relatively short period of time.

314 Mohrle 
Foreign Matter in Seal Area 
A common problem leading to poor seals is the presence of dust or tablet 
chips or pieces in the seal area. This is especially true if the tablets are 
not hard enough to possess a low order of friability and easily chip or 
break when subjected to the rigors of the packaging equipment. If tablets 
are vertically fed and dropped between the sealing rollers, it is possible 
for troublesome quantities of dust to fall onto the heat seal surface of the 
laminate prior to sealing. Adequate vacuum systems along the tablet feed 
track will minimize, if not eliminate, this problem. 
Wrinkling in the Laminate 
Uneven tension for the foil rolls or misregistration of the two laminates as 
they feed between the sealing rollers can cause a puckering, folding, or 
wrinkling of the foil laminate. Leaks are possible in this area due to the 
formation of a channel from the atmosphere to the interior of the pouch 
through which moisture can pass. A defect-free laminate and careful packaging 
equipment adjustment can remedy this problem. 
Foil Fracture 
Often foil fractures are found parallel to the inside seals, but not parallel 
to the cross-seals, as the packaged tablets leave the sealing rollers. This 
is caused by too much sealing pressure between the heat seal rollers. The 
pressure between the rollers is constant; therefore, much greater pressure 
is applied to the laminate at the side seals when the rollers are sealing an 
area across the nonsealed centers of the pouches. While the cross-seals 
are being formed, the rollers are completely touching and the pressure is 
less and evenly distributed across the roller. This phenomenon can be 
eliminated by careful packaging equipment adjustment. 
S tress on the Foil 
Pouch size in relation to the tablet diameter and thickness is an important 
factor in producing adequately packaged tablets. The tablet thickness 
must not be so great as to put an undue stress on the foil laminate during 
the sealing operation and immediately afterward. At this point, the thermoplastic 
heat-seal materials are still hot and in the process of binding. A 
thick tablet can physically pull them apart and seriously lessen the seal 
integrity. Coordination of tablet size and pouch configuration will obviate 
this problem. Satisfactory relationships between tablet and pouch size are 
shown in Table 3. 
A change in tablet shape from flat-faced to one with a deeply beveled 
edge may help also. Design patents [75-77] have been issued for tablets 
of this shape. Due to an extreme bevel on the tablet, the angle at the 
point of laminate contact is much less than that present with a flat-faced 
tablet. Less stress is transferred to the side seal area, thereby reducing 
the possibility of laminate separation before heat-seal binding occurs. 
A recent innovation from European packaging machine manufacturers 
has been the modification of thermoform plastic, blister pack equipment to 
produce a packet containing a formed aluminum blister (Uhlmann Packaging 
Systems, Fairfield, NJ). An example of this equipment is shown in Figure 
7. The vacuum-draw, heated dies used to form plastic blisters have been 
replaced with a unit that mechanically draws the laminate into a machined

Effervescent Tablets 
Table 3 Satisfactory Dimensional Relationships Between Tablets 
and Foil Laminate Pouch to Avoid Excessive Laminate Stress 
Pouch size Tablet diameter Tablet thickness 
(in. x in.) (in. ) (in. ) 
2.25 x 2.25 1.00 0.22 
2.00 x 2.00 1.00 0.16 
2.00 x 2.00 0.75 0.19 
1.50 x 1. 50 0.63 0.16 
cavity to form the aluminum laminate packets. Tablets are fed into the 
packets while the laminate containing the formed packets moves through the 
machtre in a horizontal plane rather than vertically as with the stripwrapping 
equipment described above. After the tablets have been placed 
in the packets, a printed, thin, lidding foil laminate is positioned on top of 
the tablet packets, heat is applied, and a sealed packet is produced. Tablets 
can be removed by tearing the laminate from the edge or pushing the 
tablet through the lidding laminate. 
In addition to its modern appearance, this package is more economical 
to produce since less laminate is required to package a given number of 
tablets due to the orientation of the tablets in the package. It is also 
possible to package tablets that are thicker than vertical strip-wrapping 
equipment can accommodate. The cavity depth is limited only by the degree 
to which the foil laminate can be stretched during packet formation. Examples 
of formed aluminum packages for tablets are shown in Figure 8. 
D. Package Integrity Testing 
To be certain an effervescent tablet reaches the ultimate user with the 
same quality as originally produced and packaged, tests are performed on 
the seal integrity of various packaging configurations. Clearly. the integrity 
of any package is only as good as its closure. For effervescent 
tablets. an impervious package with a loose-fitting cap or imperfect heat 
seal is as good as if the cap were left off or the heat seal area left unbounded. 
Hermetic packaging is required if effervescent tablets are to 
attain a reasonable shelf life of 2 to 3 years. The ultimate testing procedure 
is to store packages for their expected shelf life under the most 
severe humidity and temperature conditions that they will encounter, once 
sold. Since this is not practical, accelerated testing procedures have been 
developed that simulate long-term storage in adverse environments. Packages 
containing effervescent tablets are stored in chambers regulated at 
constant high humidity and temperatures, such as 80% relative humidity at 
37C and 80% relative humidity at 25C. If the relative moisture content of 
the product is determined before the study is started, changes in moisture 
content with time can be monitored. These changes may be due to moisture 
seeping into the product through the closure or through the package itself

Figure 7 Aluminum blister-forming machine. (Courtesy Uhlmann Packaging Systems t Inc.. Fairfield t NJ.) 
Co.l .... 

Effervescent Tablets 317 
Figure 8 Formed aluminum packages for tablets.

318 Mohrle 
if it is made of a material not completely impervious to moisture-vapor transmission, 
such as polyethylene bottles or thin aluminum foil with pinholes. 
The point at which the package is no longer protective will be determined 
by the rise in moisture content of each particular tablet formulation 
and is governed by the relative hygroscopicity of the tablet. Since this 
is so. tablets that have low hygroscopicity may be suitably packaged in 
less expensive. less protective containers. If a product shows little or no 
moisture pickup after being stored in a chamber at 80% relative humidity 
at 37C for 3 months, the package is considered satisfactory. Test conditions 
in the high-humidity and high-temperature chambers should be dynamic 
and not static. Air should be freely circulating about the packages 
to maximize the similarities between the test conditions and those that 
would actually occur in the field. Products and packaging that can pass 
the most severe laboratory testing are sure to be stable in the field. 
An extrapolation of accelerated test data to actual field conditions can be 
made after some testing under field conditions to ensure the predictive accuracy 
of laboratory testing. It is important that packages prepared on 
production equipment be used for any testing that will be the basis of projections 
to field conditions. Data gathered for packages wrapped on laboratory 
or experimental equipment should only be used as a guide, due to differences 
among machinery and the speeds at which they operate. 
Obviously, one cannot afford the expense or the time to wait 3 months 
to test representative samples of the packages produced on a day-to-day 
basis; therefore, several methods to test seal integrity rapidly (especially 
seals of aluminum foil laminates) have been devised. 
Vacuum Underwater Method 
The most commonly used method involves the application of vacuum to the 
pouches while they are SUbmerged in water. A representative sample of 
pouches' is placed in a water-filled chamber under a weighted plate to keep 
the pouches from floating during the test. The chamber is sealed and a 
500- to 635-mmHg vacuum drawn and maintained for 3 min. The vacuum is 
then slowly released over an additional 2- to 3-min period. 
Seal and foil defects can be located by a small stream of bubbles rising 
from a particular point on the pouch. After testing, the pouches should 
be removed from the water, allowed to dry, and carefully opened for examination. 
Water that has been drawn into the pouch during the decreasing 
vacuum phase will initiate the effervescent reaction. Tablets enclosed in 
leaking pouches can be identified easily in this manner. This method, although 
indicative of truly leaking pouches, has a distinct disadvantage: 
the difficult balance that must be made between (a) the vacuum needed to 
put enough stress on the seal to promote failure of poorly sealed pouches 
and (b) the maximum vacuum allowable without creating additional leaking 
pouches due to distortion of the foil laminate. Whether this balance is 
possible to achieve is open to question. This problem does not exist with 
the alternative methods that follow. 
Detection of Tracer Material Sealed Within the Pouch 
In this method a tracer material such as dry carbon dioxide or helium gas 
is sealed into the pouch with the tablets. The pouches to be tested are 
placed in a small, sealed chamber to which a vacuum is applied. The effluent 
from the chamber is passed through an infrared spectrophotometer

Effervescent Tablets 319 
sensing device calibrated for the specific tracer being used (Modern Controls 
, Minneapolis, MN). If the pouches are adeq uately sealed, none of 
the tracer escapes from the pouches and no response is given by the instrument. 
If a leak exists. the tracer is detected and an alarm is sounded. Systems 
can be devised for various tracer substances. some of which. perhaps. 
may be part of the formulation. thereby obviating the need for extraneous 
addition of the tracer. This method will not detect grossly unsealed areas 
from which the tracer has escaped prior to testing. However. customary 
visual examination will detect these gross defects. 
Purging with Detectable Gas 
This method is similar to the one just described. except that the pouches 
are placed in a vessel that is subsequently pressurized with the tracer 
gas as noted above. If the pouches have seal or foil defects. the gas will 
enter the pouch and mix with the contents. The pressure is released. and 
the pouches are tested as described above. The sensitivity of the instrument 
must be such that the concentration of the tracer gas, now diluted 
with the gaseous contents of the pouch. can still be detected. 
Infrared Seal Inspection 
A nondestructive infrared test method has been developed to detect sealing 
flaws (Barnes Engineering Co . Stamford. CT). A transport mechanism 
holds the sealed package and passes the seal across a focused radiation 
heat source that produces a thermal gradient in the seal. An infrared 
microscope located opposite the heat source can directly measure the temperature 
differences along the heated strip of the flexible seal. When the 
seal is uniform. heat dissipates at a uniform rate and the infrared microscope 
output is uniform. If there are voids. occluded matter. or wrinkles 
in the seal. the heat transfer rate is reduced and a sharp negative change 
is recorded by the microscope. Each unit produced can be screened in 
this manner with an automatic system designed to reject only those strips 
that contain defects. This system is not designed to detect defects other 
than those in the seal area. Tests with knurled or cross-hatch seals have 
presented problems in the past due to uneven heat distribution caused by 
the seal configuration. This method is most applicable to flat-seal areas 
without distortion-seldom used to package effervescent tablets. 
Electronic Airtightness Tester 
This relatively new, patented test method [78] was developed at the WarnerLambert 
Company to quickly and nondestructively test the hermetic seal integrity 
of packages and containers. Using vacuum and position sensors and 
analog and digital processing techniques. the unit will quickly and accurately 
determine the seal integrity of a wide variety of packages including those 
used for effervescent products. The instrument features include nondestructive 
testing with no package preconditioning necessary. It is simple 
to operate and produces objective results. This instrument utilizes a microprocessor 
controller and associated software to determine the degree of 
package airtightness based on the package's response to an external vacuum. 
The unit includes a vacuum chamber, pump. microprocessor, and both vacuum 
and displacement transducers. The package that has at least one flexible

320 Mohrle 
surface is positioned inside the vacuum chamber and the door is latched. 
A linear displacement transducer is lowered into contact with the expandable 
surface of the package. The unit is activated and a dedicated microprocessor 
begins the test sequence. The chamber air is gradually evacuated and 
the package thickness is monitored in response to the changing vacuum. 
Over the course of 5 to 30 sec. the microprocessor analyzes the data and. 
based on the expansion and contraction of the package in response to the 
vacuum. determines whether or not the package is airtight. As the test 
proceeds, the data points collected are graphically displayed on a monitor 
screen and a go Ino-go determination appears. The microprocessor also 
computes a linear regression of the package expansion as a function of 
vacuum. The principle of the test is that if the package is airtight, the 
expansion of the package will track the applied vacuum, expanding and 
contracting as the vacuum increases or decreases. Consequently. the linear 
regression will have a high degree of correlation. If the package is not 
airtight. it either will not expand or will expand initially and begin to collapse 
as the head space is vented under the external vacuum. In either 
case, the expansion of the package will not behave as a linear function of 
vacuum and the correlation will be low. The decision regarding the acceptability 
of the package seal integrity is made based on the variables measured 
during the test. 
This test instrumentation is being used in a manufacturing environment 
to provide a quick determination of the suitability of the package coming 
off the packaging line. It can provide the line operator important information 
regarding the performance of the packaging machine so that corrective 
actions can be accomplished before a large quantity of defective goods are 
This instrument has been tested against the gas detection methods described 
above and was found to be as accurate without the need to purge 
or f111 the packages with a tracer or detection gas. A limited number of 
units are under fabrication and are available to the pharmaceutical, food. 
and confectionery industries from the Consumer Products Package Development 
Department, Warner-Lambert Company, Morris Plains. NJ 07950. 
The following formulations and suggested manufacturing procedures illustrate 
the principles discussed in the text of this chapter. 
Example 1: Antacid Effervescent Tablets 
Ingredient Quantity 
1- Citric acid, anhydrous (granular) 
2. Sodium bicarbonate (granular) 
3. Sodium bicarbonate (powder) 
4. Citrus flavor (spray-dried) 
5. Water 
1180 g 
1700 g 
50 9 
30 9

Effervescent Tablets 
Example 1: (Continued) 
Thoroughly blend 1, 2, and 4 in a planetary mixer. 
Quickly add all of 5 and mix until a workable mass 
is formed. Granulate through a 10-mesh screen using 
an oscillating granulator. Spread evenly on a paperlined 
drying tray and dry in a forced-draft oven at 
70C for 2 hr. Remove from oven, cool, and regranulate 
through a 16-mesh screen. Place granulation 
in a tumble blender and add 3. Mix well. 
Compress 1-in. flat-faced, beveled edge tablets 
each weighing 3.10 g. Package in glass tubes or 
aluminum foil. 
Example 2: Antacid-Analgesic Effervescent Tablets 
Ingredient Quantity 
1. Acetylsalicylic acid (SO-mesh crystals) 
2. Monobasic calcium phosphate (powder) 
3. Sodium bicarbonate (granular) 
4. Citric acid, anhydrous (granular) 
325 9 
165 9 
1700 g 
1060 9 
Convert 3 to 7-9% sodium carbonate by placing in a 
forced-draft oven set at 100DC for 45 min, with two 
mixings at 15-min intervals. Cool the converted bicarbonate 
and mix with 2 and 4 in a tumble blender. Add 
1 and mix for 10 min. Compress 1-in.-diameter flatfaced, 
beveled edge tablets each weighing 3.25 g. 
Stabilize tablets in a forced-draft oven at 60DC for 1 hr. 
Cool and package in glass tubes or aluminum foil. 
Example 3: Potassium Chloride Effervescent Tablets [79] 
Ingredient Quantity 
1. Glycine hydrochloride 1338 g 
2. Potassium chloride 597 g 
3. Potassium bicarbonate 1001 g 
4. Potaasl um citrate 216 g 
5. Polyvinylpyrrolidone 77g 
6. Polyethylene glycol 8000 (powder) 115 g 
7. Saccharin 20 9 
8. SiIlea gel (fumed) 5 g

322 Mohrle 
Example 3: (Continued) 
I ngredient Quantity 
9. L-Leucine (pulveri zed) 
10. Citrus color 
11. Citrus flavor (spray-dried) 
12. Isopropyl alcohol 
34 9 
3 9 
5 9 
6 9 
Grind together 1, 2, 3, and 4. Mix the ground materials 
in a tumble blender for 15 to 20 min. Granulate the 
mixed powders with a solution of 5, 6, and 7 dissolved in 
12. Spread the granulation on trays and dry in a forcedai 
r oven at 50 to 55C until the alcohol odor Is gone. 
Pass through a 12-mesh screen. Place granulation in a 
tumble blender and blend in 8, 11, and 9. Compress 
l-in.-diameter, flat-faced, beveled edge tablets each 
weighing 3.41 g. Package in aluminum foil. 
Example 4: Flavored Beverage Effervescent Tablets 
I ngredient Quantity 
1. Sodium bicarbonate (granular) 
2. Sodium carbonate, anhydrous 
3. Citric acid, anhydrous (granular) 
4. Aminoacetic acid 
5. Flavor (spray-dried) 
6. Color 
7. Light mineral oil 
8. Water 
735 g 
80 9 
1300 9 
50 9 
50 9 
5 g 
15 9 
4 9 
Premix 7 with 200 9 of 1. Disperse 6 on 35 9 of 1. 
Place 3 in the bowl of a planetary mixer. Start mixer 
and slowly add 8; mix thoroughly. Add to mixer in sequence, 
while mixing, the remainder of 1, 2, 4, 5, the 
color dispersion, and the mineral oil dispersion; mix until 
uniform. Compress 3!4-in., flat-faced, beveled edge tablets 
weighing 2.23 g each. Pass through curing oven; 
cool; and package in aluminum foil.

Effervescent Tablets 
Example 5: Stannous Fluoride Mouthwash Effervescent 
Tablets [801 
Ingredient Quantity 
1. Malic acid 
2. Sodium bicarbonate 
3. Sodium carbonate 
4. Stannous fluoride 
5. Color 
6. Flavor 
7. Sweetener 
8. Sorbitol 
9. Polyethylene glycol 8000 (powder) 
10. Sodium benzoate (fine powder) 
11. Simethicone 
420 9 
290 9 
70 g 
21 g 
3 g 
20 g 
4 9 
110 g 
30 9 
30 9 
2 g 
Coat 10 with 11 using a twin-shell blender with intensifier 
bar activated. Blend 1, 2, 3, 4, 9, the blend 
of 10 and 11, 8, 5, 7, and 6 in a ribbon blender. 
Compress 11. 1-mm-diameter, shallow concave tablets 
each weighing 480 to 500 mg. Package in aluminum 
Example 6: Children's Decongestant Effervescent 
Cold Tablets 
Ingredient Quantity 
1. Acetylsalicylic acid, USP (crystals) 81 9 
2. Pseudoephedrine hydrochloride 30 9 
3. Fruit flavor (spray-d ri ed) 20 g 
4. Fruit color 2 9 
5. Sod i urn bica rbonate (granular) 550 9 
6. Citric acid, anhydrous (granular) 325 g 
7. Citric acid, anhydrous (powder) 325 g 
8. Water 
Convert 5 to 7-9% sodium carbonate by placing in a 
forced-draft oven at 100C for 45 min, with two mixings 
at 15-min intervals. Cool the converted bicarbonate 
and mix with 6 and 7 in a planetary mixer for 10 
min. Quickly add B and mix until the water is evenly

324 Mohrle 
Example 6: (Continued) 
distributed and a mild reaction occurs. Immediately 
transfer to paper-lined drying trays and spread evenly. 
Place trays in a forced-draft oven at 700 e for 2 hr. 
Remove from oven, cool, and granulate through a 
12-mesh screen. Mix together 2, 3, and 4. Mix the 
dried granulation, the 2-3-4 premix and 1 in a tumble 
blender until uniform. Compress 5/S-in.-diameter, 
flat-faced, beveled edge tablets weighing 1.33 g each. 
Stabilize the tablets by heating in a forced-draft oven 
at 60C for 1 hr. Cool and package in aluminum foil. 
Example 7: Denture Cleanser Effervescent Tablets 
Ingredient Quantity 
1. Potassium monopersulfate 
2. Citric acid, anhydrous (granular) 
3. Sodium bicarbonate (granular) 
4. Sodium chloride 
5. Sodium perborate monohydrate 
6. Sodium sulfate 
7. Polyvinylpyrrolidone 
S. Isopropyl alcohol 
9. Sodium lauryl sulfate 
10. Color 
11. Oil of peppermint 
12. Magnesium stearate 
800 g 
575 g 
SOD g 
320 g 
320 g 
225 g 
100 g 
170 g 
10 g 
2 g 
16 g 
20 g 
Blend 3, 4, 5, 6, and 7 in a planetary mixer. Add 
S and mix until the mass is uniformly wet. Spread 
wetted mixture on trays about 1-in. deep. Dry in 
forced-d raft oven at 70C for 16 hr. Pass d ri ed 
granulation through an Hi-mesh screen using an 
oscillating granulator. Mix 1 and 2 in a tumble blender. 
Add 1500 g of the dried, screened granulation 
and tumble until well mixed. Distribute 9, 10, and 
11 in 265 g of the dried, screened granulation and 
add to the tumble blender. Mix thoroughly. Add 
12 to the tumble blender and mix well. Compress 
l-in. -diameter flat-faced, beveled edge tablets 
weighing 3.19 g each. Package in aluminum foil.

Effervescent Tablets 
Example 8: Bath Salt Effervescent Tablets 
Ingredients Quantity 
1. Monosodium phosphate anhydrous 
2. Citric acid, anhydrous 
3. Sodium bicarbonate (fine granular) 
4. Surfactant 
5. Blue color 
6. Simethicone 
7. Encapsulated fragrance 
8. Water 
3200 g 
630 g 
2500 g 
1 g
50 9 
16 9 
Thoroughly mix 6 with 100 g of 3 on which 5 has been 
previously distributed. Add 7 and mix thoroughly; 
set aside. Place 1 in a ribbon blender. Slowly add 8 
while mixing and mix thoroughly. While mixing, slowly 
add 2, 2400 9 of 3, the 3-5-6-7 premix, and 5. Mix 
well. Compress 1-in. -diameter flat-faced, beveled 
edge tablets each weighing 6.4 g. Pass through a 
forced-draft oven to stabilize, cool, and package six 
tablets to a container. (Six tablets are dissolved in 
a 25-gallon tub to yield a water softening, lightly colored, 
and lightly fragranced bath.) 
Example 9: Feminine Hygiene Solution Effervescent 
Ingredient Quantity 
1. Sodium lauryl sulfate 70 g 
2. Simethicone 15 g 
3. Sodium bicarbonate 345 g 
4. Monosodium phosphate, anhydrous 440 g 
5. Citric acid, anhydrous (granular) 655 9 
6. Sodium chloride 865 g 
7. Water 2 g 
Thoroughly blend 2 with 145 g of 3 in a planetary 
mixer. Place 5 and 6 in a pony mixer; energize the 
mixer and blend for 1 min. Continue mixing and 
slowly add 7. Mix for 1 min or until uniform. Continue 
mixing and add consecutlve'y 4, simethicone 
premix, 300 g of 3 and 1. Mix for 3 min until

326 Mohrle 
Example 9: (Continued) 
thoroughly blended. Compress 3/4-in.-diameter flatfaced, 
beveled edge tablets each weighing 2.39 g. 
Place tablets on a paper- lined drying tray and stabilize 
in a forced-draft oven at 90C for 30 min. Remove 
from oven, cool, and package in aluminum foil. (Each 
tablet is dissolved in 1000 ml of 40C water prior to 
use. ) 
Exarr pie 10: Toilet Bowl Cleaner Effervescent Tablets 
Ingredient Quantity 
1. Sodium bisulfate 1200 g 
2. Sodium bicarbonate 250 g 
3. Detergent 30 g 
4. Color 2 g 
5. Frag ranee oi I 10 g 
Disperse 4 and 5 on 3, using geometric dilution techniques. 
Place 600 g of 1 in a tumble blender. Add 
the color Ifragrance premix and blend for 1 min. Add 
20 g of 3 and blend for 1 min. Add 600 g of 1 and 
blend for 2 min. Roller-compact or slug the granulation 
to densify. Granulate the compacted sheets or 
slugs by passing through a 12-mesh screen. Place 
granulation in the tumble blender and add 109 of 3. 
Blend thoroughly. Compress on heavy-duty tablet 
equipment or form compacts using briquetting equipment. 
each compact weighing 149.2 g. A suitable tablet size 
would be 2-3/4 in. in diameter and about 7/8 in. thick. 
I ndividually wrap each tablet in an aluminum foi I pouch. 
1. United States Pharmacopeia XXI, 854, United States Pharmacopeia! Convention, 
Inc, , Rockville, MD (1985). 
2. U.S. Patent 3,590,121 (1971). 
3. Eur. Patent 93784A (1983). 
4. Fr. Patent 2,477,174 (1980). 
5. U.S. Patent 4,552,771 (1985). 
6. U.S. Patent 3,962,417 (1976). 
7. U.S. Patent 4,417,993 (1983). 
8. Br. Patent application 2,095.556 (1982). 
9. Br. Patent 1,328,591 (1970). 
10. U.S. Patent 4,127,645 (1976). 
11. R. Gaedeke, C. Sander, P. Schmidt, and H. Steinitz, Med. RUn. 
(Munich), 67:1173-1175 (1972).

Effervescent Tablets 327 
12. U.S. Patent 4,289,751 (1981). 
13. A. J. Blowers, E. G. Cameron, and E. R. Lawrence, Br. J. Clin. 
Pract., 35:188-190 (1981). 
14. U.S. Patent 4,147,768 (1976). 
15. U.S. Patent 3,980,766 (1976). 
16. U.S. Patent 3,961,041 (1976). 
17. U.S. Patent 4,153,678 (1979). 
18. W. D. Mason and N. Winer, J. Pharm. ScL, 72:819-821 (1983). 
19. G. Ekenved, R. Elofsson, and L. Solvell, Acta Pharm. Suec., 12:323332 
20. L. M. Ross-Lee, M. J. Elms, B. E. Cham, F. Bochner, 1. H. Bunce, 
and M. J. Eadie, Eur. J. Clin. Pharmacol., 23:545-551 (1982). 
21. W. D. Mason and N. Winer, J. Pharm. ScL, 70:262-265 (1981). 
22. D. Orton, R. Treharne Jones, T. Kaspi, and R. Richardson, Br: J. 
en. Pharmacol., 7:410-412 (1979). 
23. M. Zdrakovic, G. T. Pedersen, N. A. Klitgaard, and E. Rasmussen, 
Arch. Pharm, Chemt , Sci. Ed., 9:25-33 (1981). 
24. V. Sanno, J. Tuomisto, and M. Airaksinen, Acta. Pharm. Fenn., 92: 
77-83 (1983). 
25. Z. El-Din, A. T. Nouth, A. H. EI-Gawad, and M. EI-Shaboury, 
Biopharm. Pharmacokinet. Eur. Congr. 2nd, 1: 395-399 (1984). 
26. U.S. Patent 3,131,123 (1964). 
27. K. Nishimura, K. Sasahara, M. Arm, T. Nitanai, 1. Yoshihiko, T. 
Morioka, and E. Nakajima, J. Pharm. ScL, 73:942-946 (1984). 
28. F. L. Lanza, V. Smith, J. A. Page-Castell, and D. O. Castell, So. 
Med. J., 79: 327-330 (1986). 
29. U.S. Patent 4,180,467 (1979). 
30. A. J. Repta and T. Higuchi, J. Pharm. Sci" 58:1110-1113 (1969). 
31. Eur. Patent 11,489 (1978). 
32. Technical Bulletin, Edward Mendell Co., Carmel, N.Y. 
33. Belg. Patent 880,749 (1978). 
34. Fr. Patent 2,547,501 (1983). 
35. U.S. Patent 4,237,147 (1980). 
36. U.S. Patent 2,931,776 (1960). 
37. U. S. Patent 3,936,385 (1976). 
38. U.S. Patent 4,592,855 (1986). 
39. W. A. Strickland, Jr., E. Nelson, L. Busse, and T. Hlguchi , J. Am. 
Pharm. Assoc. Sci. ea., 45: 51-55 (1956). 
40. W. A. Strickland Jr., T. Higuchi, and L. Busse, J. Am. Pharm. 
Assoc. Sci. Ed., 49: 35- 40 (1960). 
41. U.S. Patent 3,821,117 (1974). 
42. Br. Patent 1,255,437 (1969). 
43. U.S. Patent 3,136,692 (1961). 
44. Br. Patent 1,221,038 (1971). 
45. U.S. Patent 4,536,389 (1985). 
46. Br. Patent 1,178,294 (1967). 
47. U.S. Patent 3,976,601 (1976). 
48. F. E. J. Sendall, J. N. Staniforth, J. E. Rees, and M. J. Leatham, 
Pharm. J., 230:289-294 (1983). 
49. U.S. Patent 4,007,052 (1976). 
50. J. P. Armandou and A. G. Mattha, Pharm. Acta. Helv., 57: 287-289 
51. Braz. Patent 8,150 (1982).

328 Mohrle 
52. R. W. Murray, J. Pharm. ser.. 57:1776-1779 (1968). 
53. J. Joachim, J. Corboda, G. Joachim, and H. Delonea, J. Pharm. Belg., 
38: 251-257 (1983). 
54. S. I. Saleh, A. Aboutaleb, C. Boymond , and A. Stamm, Expo.-Congr. 
Int. Technol. Pharm., 3rd, 2:38-48 (1983). 
55. U.S. Patent 4,614,648 (1986). 
56. Belg. Patent 781,358 (1971). 
57. A. Devay, J. Uderazky, and I. Racz, Acta. Pharm. Techno!.. 30: 
239- 242 (1981). 
58. H. M. EI-Banna and S. N. Savtin, Sci. Pharm., 48:369-377 (1980). 
59. U.S. Patent 4,256,599 (1981). 
60. U.S. Patent 4,004,036 (1977). 
61. Br. Patent, 1,374,105 (1970). 
62. U.S. Patent 3,105,792 (1963). 
63. Eur. Patent 76,340A (1983). 
64. Eur. Patent 76,340B (1984). 
65. Br , Patent, 2,148,117 (1985). 
66. Br. Patent, 1,270,781 (1969). 
67. U. 8. Patent 4,579,742 (1986). 
68. Technical Bulletin, Durkee Industrial Foods, Cleveland, OH. 
69. U.S. Patent 2,984,543 (1961). 
70. U.S. Patent 3,875,073 (1975). 
71. U.8. Patent 4,127,645 (1978). 
72. W. Horowitz, (ed.), Official Methods of Analysis of the Association of 
Official Agricultural Chemists, 9th ed., 1960, pp. 97-98, Association 
of Official Agricultural Chemists, Washington, D. C. 
73. N. R. Anderson, G. S. Banker, and G. E. Peck, J. Pharm. ScL, 
71: 3- 6 (1982). 
74. N. R. Anderson, G. S. Banker, and G. E. Peck, J. Pharm. ScL, 
71: 7-13 (1982). 
75. U.S. Design Patent 274,846 (1984). 
76. U.S. Design Patent 275,614 (1984). 
77. U.8. Design Patent 275,615 (1984). 
78. U. S. Patent 4,663,964 (1987). 
79. U.S. Patent 3,903,255 (1975). 
80. U.S. Patent 4,267,164 (1981).

Special Tablets 
James W. Conine* and Michael J. Pikal 
Eli Lilly and Company, Indianapolis. Indiana 
Most tablets are intended to be swallowed, the active ingredients being absorbed 
from the gastrointestinal tract. There are some special types of tablets. 
however, which are intended for administration in other ways. Most of 
the tablets discussed in this chapter are intended for adsorption through the 
mucosal lining of the mouth, either sublingually (Le ,.; from the area beneath 
the tongue) or buccally (i. e , , from the area between the cheek and gum) [1]. 
In addition, molded tablets for other applications and other modes of administration 
will be briefly discussed. 
A. Effect of the Site on Absorption 
Drugs can be absorbed into the bloodstream from many of the surfaces of 
the body (e.g., gastrointestinal, nasal, rectal, dermal) to which the drug 
can be applied and held in position for a sufficient time for absorption to 
take place. A compound should be formulated so that it can be properly 
administered for the particular surface through which it will be absorbed. 
The use of swallowed medication is by far the most common means of introducing 
drugs into the general circulatory system. When absorbed from 
the stomach or intestinal tract, the drug passes through the membrane 
lining into the capillaries to the superior mesenteric vein, then through 
the portal vein and liver into the inferior vena cava, before reaching the 
heart and arterial circulation which distributes the drug throughout the 
body. This route selectively channels compounds through the liver, which 
is the body's major organ of detoxication. Metabolism by the liver can 
greatly reduce the amount of active compound ultimately reaching the target 
*Currently retired. 

330 Conine and Pikal 
Absorption of drugs through the highly vascular mucosal lining of the 
mouth moves the drug through the sublingual or buccal capillaries and 
veins to the jugular vein and superior vena cava-directly into the heart 
and arteria circulation without first passing through the liver. This 
route can be effective when drugs absorbed through the gastrointestinal 
tract are destroyed by extensive hep atic detoxication. For example, :n 
rats naltrexone and naloxone were found to have less than 1% bioavailability 
from oral dosage as a result of extensive first -p ass metabolism, 
while buccal availability was 63 and 71%, respectively {2]. The sublingual 
and buccal areas offer convenient sites to deposit and hold a tablet on an 
absorbing surface over a time sufficient for absorption to take place. 
Some recent work has been directed toward the determination of the 
degree of enzymatic hydrolysis of peptides that occurs at different mucosal 
sites. In a study in rabbits using enkephalins as models, peptide hydrolysis 
was found to be twice as great in nasal compared to buccal mucosa, 
and in both of these areas it was much less than that found in ileal mucosa 
[3] . Inhibition of aminopeptidase activity has been shown to occur in the 
presence of the penetration enhancers sodium desoxyeholate , sodium glycolate, 
and polyoxyethylene- 9-lauryl ether [4]. 
B. Effect of the 0 rug on Absorption 
The practice of chewing leaves or other parts of plants so that alkaloids 
or other compounds are absorbed through the lining of the mouth to prod 
uce central or systemic effects is common in several cultures. In Malaysia 
and the South Pacific the areca or betel nut is chewed in combination 
with shell lime and the leaves of Piper betel. The natives of Peru have 
a history of chewing coca leaves with or without lime, which predates the 
Spanish conquest. In our own society there is an appreciable market for 
smokeless tobacco products. Also the use of nicotine chewing gum and 
other buccally absorbed nicotine products have been used to help break 
the tobacco-smoking habit. 
Absorption of drugs through the mucous membrane lining of the mouth 
has been described as the passive diffusion of the un-ionized form of the 
drug from the aqueous phase (in the saliva) to the lipid phase (in the 
membrane) {5]. The work of Walton and Lacy [6] and of Walton [7,9] 
established that there is a direct relationship between the oil/water partition 
coefficient and drug absorption. Absorption of the drug is more 
or less independent of the absolute solubility of the drug in either the 
aq ueous or lipid phase. 
Table 1 shows the inverse relationship between the oil/water partition 
coefficient and the ratio of sublingual to subcutaneous dose for some of 
the drugs studied by Walton. A comparison of the sublingual and subcutaneous 
dose is used since this is a measure of the ability of the drug 
to penetrate the membrane lining of the mouth. Satisfactory absorption 
of compounds over a wide oil/water partition coefficient range of 40 to 
2000 has been observed. Compounds with coefficients in the 20 to 30 
range are borderline for effective administration by the sublingual route. 
For compounds with oil/water partition coefficients of less than 20, the 
effective sublingual doses are several times the subcutaneous doses. 
Buccal administration of morphine sulfate has been reported to provide 
a similar degree of postoperative analgesia to an equal dose administered 
intramuscularly [10]. Peak plasma levels were somewhat lower following

Special Tablets 
Table 1 Comparison of Oil/Water Partition Coefficient 
Compared to Sublingual/Subcutaneous Dosage Ratio [9] 
partition SUblingual/subcutaneous 
Drug coefficient ratio 
Cocaine 28 2 
Apomorphine 20 2 
Heroin 17 3 
Strychnine 21 4 
Thebaine 12 >4 
Emetine 9 >6 
Atropine 7 8 
Morphine 0.15 10 
Hydromorphine 0.2 15 
Codeine 2.0 15 
buccal dosage, but total bioavallability was 40 to 50% greater. Nitroglycerin 
has a very high partition coefficient of 1820 [9] and is extremely effective 
when administered sublingually. However, as the oil/water partition coefficient 
increases beyond 2000, the solubility in the saliva is usually not 
enough to supply an adequate concentration for transfer through the mucous 
membrane. Since nitroglycerin is a liquid, absorption of the undissolved 
compound directly into the membrane possibly explains its very 
rapid absorption and pharmacological response. 
A number of studies by Beckett and coworkers [5,11,12] demonstrated 
that the relationship of pKa to absorption from the lining of the mouth is 
similar to the results observed in the gastrointestinal tract [13]. It has 
been found that. by buffering a solution of the drug which is held in the 
mouth, absorption depends on partitioning the un-ionized form into the 
lipid phase. Basic drugs which are administered as salts become better 
absorbed as the pH is raised. thereby converting more of the salt into 
the base. For example. buccal absorption of amphetamine does not occur 
below pH 6.6, but over 60% absorption occurs at pH 9.0 [11]. The saliva 
ordinarily maintains the pH of the mouth between 5.6 and 7.6. The use 
of buffered solutions or tablets makes it possible to control the pH somewhat 
outside this range in order to enhance the absorption of some drugs. 
When two compounds have the same pKa, the compound with the greater 
oil/water solubility ratio will be better absorbed (Fig. 1). In this series 
of n-alkanoic acids (from 4 to 12 carbons), all with pKa from 4.82 to 4.85 
at 25C. the absorption increases as the chain length and oil/water solubility 
ratio increase. Compounds which contain no ionizable groups are less 
affected by pH changes, although buccal absorption of nitroglycerin is 
greater below pH 5.0 [14].

0:: o
Conine and Pikal 
Figure 1 Buccal absorption of n-alkancic acid in humans. Key: 4 butyric; 
v Valeric j  hexanoic; 0 heptanoic; x octanoic; ... nonanoic; 0 decanoic; 
 undaeanoic ; o: dodecanoic. [From Ho, N. F. H., and Higuchi, W. I., 
J. Pharm, Sci  60: 537 (1971). Reproduced with permission of the copyright 
There is good evidence that peptides are absorbed buccally. The thyrotropin-
releasing hormone protirelin when administed through buccal absorption 
from a paper disk produced increases in the thyrotropin and prolactin 
levels of human subjects (15]. However, buccal doses were 100 
times the intravenous doses used in the study [15]. 
Theoretical physical models have been proposed to accurately describe 
the mechanism of absorption from the lining of the mouth [16,17]. The 
model for the n-alkanoic acids whose absorption is described in Figure 1 
consists of a three-compartment system where the first and third are aq ueous 
compartments separated by the second, which is a lipid layer. The 
first compartment or mucosal side is the bulk aqueous drug solution, and 
the third or sclerosal side is an aqueous layer at pH 7.4, which is the pH 
of the blood. There is assumed to be a perfect sink after the third compartment. 
The pH of the first compartment is either the natural pH or 
one adjusted by buffers. 
C. Currently Marketed Buccal and SUblingual Drugs 
In addition to good absorption, the ideal drug for sublingual or buccal use 
should be small in dose, usually not more than 10 to 15 mg. The drug

Special Tablets 
Table 2 Drugs Marketed as SUblingual or Buccal Tablets 
Ergoloid mesylates 
Ergotamine tartrate 
Erythrityl tetranitrate 
Isoproterenol hydrochloride 
Isosorbide dinitrate 
0.5-1 mg 
2 mg 
5-10 mg 
10-15 mg 
2.5-5 mg 
0.15-0.6 mg 
5-20 mg 
1-3 mg 
oral dose 
0.6-1 mg 
30 mg 
10-20 mg 
2.5-6 mg 
10-40 mg 
2.5-6 mg 
should not be highly ionic or at least should be capable of being buffered 
in tablet form if it is to result in satisfactory absorption. The ideal compound 
should not have an undesirable taste. since bitter or bad -tasting 
compounds will stimulate saliva flow. The major drugs which are currently 
marketed as sublingual or buccal tablets are listed in Table 2. These consist 
of nitrate esters. isoproterenol hydrochloride. and hormones. They 
represent a select group of compounds for which this is currently the most 
effective means of administration. Nitroglycerin, which is the most widely 
used sublingual drug, has placed in the top 100 of most prescribed drugs 
for the past several years [18]. The sublingual response to nitroglycerine 
is more rapid than that from the gastrointestinal tract and more effective, 
since it avoids the destructive first passage through the liver [19]. 
A number of other products besides those listed in Table 2 have at one 
time or another been commercially available either as sublingual or buccal 
tablets. Estradiol and progesterone, which were once administered buccally 
have been replaced by orally active agents having the same activity. Because 
there is some inconvenience in the administration of sublingual and 
buccal tablets, particularly in the latter, products designed for absorption 
through the mucosal lining of the mouth are USUally those for which this is 
the only satisfactory nonparenteral method of administration. After the 
sublingual or buccal tablet has been placed in position, the patient should 
avoid eating, drinking. chewing. smoking. and possibly talking. in order 
to keep the tablet in place. SWallowing of saliva should also be avoided, 
since the saliva may contain dissolved drug, and ingestion through the 
gastrointestinal tract is usually much less efficient than absorption through 
the oral mucosa.

334 Conine and Pikal 
The molded tablet was originally introduced by Fuller in 1878 [20]. Only 
a year earlier Brunton [21] described the first use of sublingual drug 
therapy when he utilized nitroglycerin in the treatment of angina pectoris. 
Sublingual tablets are intended to be placed beneath the tongue and held 
there until absorption has taken place. They must dissolve or disintegrate 
quickly, allowing the medicament to be rapidly absorbed. Therefore, sublingual 
tablets are frequently formulated as molded tablets. 
Molded tablets may also be used for buccal absorption, may be swallowed. 
may be used to prepare solutions for topical application, or (as in the 
past) may be used for injection. Molded tablets are also referred to as 
tablet triturates; the designation comes from the early practice of preparing 
tablets from triturations. Official triturations were 10% dilutions of finely 
divided potent drugs in lactose. A dilution of this type made it easier to 
handle the drug and divide it more accurately into single doses. The trituration 
could be further diluted with lactose to make the correct tablet 
Molded tablets designed to be dissolved in a small amount of water to 
make an aqueous solution which can be administered parenterally are known 
as hypodermic tablets. Current standards of sterility cannot be met by 
by the usual method of handling hypodermic tablets in multiple-dose containers. 
The removal of one tablet would-under most conditions-expose 
the remaining tablets to possible contamination. Technical advances which 
have increased the availability of sterile parenteral products have eliminated 
the need which once existed for the hypodermic tablet [22]. The formulations 
for hypodermic tablets are similar to those which will be described 
for tablets triturates. 
A. Formulations for Molded Tablets 
Molded tablets are usually prepared from soluble ingredients so that the 
tablets are completely and rapidly soluble. They contain, in addition to 
the drug, an excipient or base of lactose. dextrose, sucrose, mannitol, or 
other rapidly soluble materials or mixtures of these ingredients. Commercial 
lactose is the monohydrate or Cl form and is the most common excipient. 
I3-Lactose, which is an anhydrous form produced by crystallization above 
93. 5C, has been also used as an excipient and is reported to be more 
readily soluble than a-lactose. Tablets containing insoluble excipients may 
be prepared from finely divided kaolin. calcium carbonate. calcium phosphate, 
or other insoluble powders; but such tablets are not often eneountered 
today. To insure rapid solubility of the soluble tablets, the excipients 
are usually put through a fine screen or 120-mesh bolting cloth. 
After the excipient is blended with the drug, the powder mix is moistened 
with the solvent. which is most commonly aqueous alcohol. Other 
volatile solvents such as acetone or hydrocarbons might also be used. 
Antioxidants, such as sodium bisulfite, and buffers or other ingredients 
may be added to improve the physical and chemical stability of the product. 
A variety of materials have been tested in nitroglycerine tablets to 
stabilize them against decreases in the content uniformity of the tablets 
which occur during aging. Problems unique to nitroglycerin tablets will 
be discussed in Section III of this chapter. To increase the hardness and 
reduce the erosion on the edges of the tablets during handling, agents

Special Tablets 335 
such as glucose, sucrose, acacia, or povidone have been added to the solvent 
mixture. This should be done with care, since. if used in excessive 
amounts I such agents can decrease the rate of solubility of the tablets. 
Formulations for molded tablets are usually very simple and contain no 
insoluble ingredients. Placebo tablets can be prepared which contain only 
lactose. Typical formulas for several molded tablets are listed here. 
Example 1: Codeine Phosphate Tablets (30 mg) 
Quantity per 
Ingredient tablet 
Codeine phosphate powder 
Lactose (bolted) 
Sucrose (powder) 
Alcohol-water (60:40) 
30.0 mg 
17.5 mg 
1.5 mg 
q .s , 
Screen and blend the powders; add alcoholwater 
(60: 40) to moisten and mold tablets. 
Example 2: Scopolamine Hydrobromide Tablets 
(0.4 mg) 
Quantity per 
Ingredient tablet 
Scopolamine hydrobromide 
Lactose (bol ted) 
Sucrose (as 85% syrup) 
Alcohol-water (60: 40) 
0.4 mg 
35.0 mg 
0.3 mg 
Screen and blend the powders: moisten the blend 
with alcohol-water (60:40) to which the surcrose 
syrup has been added, and mold the tablets. 
Example 3: Nitroglycerine Tablets (0.4 mg) 
Quantity per 
Ingredient tablet 
Trituration of nitroglycerin 
( 10% on lactose) 
Lacotose (bolted) 
Polyethylene glycol 4000 
Alcohol-water (60: 40) 
4.4 mg 
32.25 mg 
0.35 mg 
qvs , 
Screen and blend the powders; moisten the blend 
with alcohol-water (60: 40) to which the polyethylene 
glycol 4000 has been added, and mold the tablets.

336 Conine and PikaI 
B. Hand Molding of Tablets 
The method and equipment used for hand molding tablets have changed 
little since they were originally described by Fuller [20]. The powder 
mixture must be blended carefully to insure that a homogeneous mixture is 
obtained. On a very small scale, this is usually done in a mortar. The 
solvent mixture is added to make a workable mass without overwetting the 
powder. The mold plate is placed on a smooth tile or glass plate, and the 
mass is forced into the tablet mold with sufficient pressure, uniformly applied. 
to insure that all tablets have the same weight (Fig. 2). This can 
be done with either an ordinary spatula or a special spatula resembling a 
short-bladed putty knife. The mold plates contain anywhere from 50 to 
several hundred die holes and are made of metal, hard rubber, or plastic. 
To remove the tablets for drying, the mold plate is placed on top of a 
plate which has projecting pegs that coincide with the die holes (Fig. 3). 
By pressing the mold plate down onto the pegs, the tablets are forced out 
of the dies onto the tops of the pegs. The tablets are then removed from 
the pegs to drying. There are usually two longer guide pins (one at each 
Figure 2 Hand molding of dispensing tablets.

Special Tablets 337 
Figure 3 Molded dispensing tablets ready for removal from mold plate. 
end of the peg plate) which coincide with the holes in the mold plate so 
that the pegs can be precisely guided, and no damage to the soft tablets 
results. The ends of the plates differ in shape so that they can be put 
together correctly in only one way. This feature gives the process better 
reproducibility and the tablets greater uniformity. Since the weight uniforIDity 
normally increases with tablet density. the molds should be packed 
fairly tightly to minimize the weight variation. However. the uniformity of 
weight normally attained with compressed tablets cannot be achieved with 
molded tablets. 
Some molding problems can be related directly to the solvent. Application 
of too little solvent may result in a soft tablet. On the other hand, too 
much solvent will result in tablet shrinkage upon drying. In addition to 
the irregular shape due to shrinkage, the tablets may become case-hardened 
and less readily soluble. Similiar problems result if aqueous alcohol of incorrect 
solvent proportions is used. The most satisfactory range for lactosebased 
tablets is 50 to 60% alcohol. When the water content is low. the

338 Conine and Pikal 
resulting tablets are poorly bonded and will tend to powder and wear on 
the edges. With high water content, the tablets will become harder and 
less readily soluble. 
The tablets are removed from the pegs and allowed to dry in ambient 
air currents, or the drying may be accelerated by placing the tablets in 
a forced-air oven. As the tablets dry, the solvent migrates to the surface 
and may carry the active ingredient or other soluble components to the 
tablet surface [23,24]. This can produce a nonhomogeneous distribution 
of drug throughout the tablet. Solvent-mediated migration of the drug may 
affect the stability, particularly if the active component is photosensitive 
or is subject to oxidation [23]. Although drug migration has been reported 
in studies of granulation drying [25,26] and can be readily demonstrated 
in the migration of soluble dyes during drying, drug migration in molded 
tablets has received little attention. A change to a different solvent or 
mixture can minimize migration and thereby result in an improved tablet. 
Also, a change to an excipient which has greater attraction for the drug 
in the solvent system will also reduce the amount of migration which occurs 
during tablet drying. Care should be exercised to avoid choosing an excipient 
which will bind the drug so tightly that it is not easily removed 
from the excipient in vivo. 
When formulating tablets, a placebo can be made in order to determine 
the expected tablet weight. If the dose is quite small (for example, less 
than 1 mg) a direct substitution of drug for excipient can be made. If a 
larger portion of the tablet consists of the drug itself, the density of the 
drug as well as that of the excipient needs to be considered in determining 
the finished tablet weight. 
C. Machine Molding of Tablets 
Equipment is available for the large-scale production of molded tablets. 
The blending of the dry mix may be carried out in any of the pharmaceutical 
mixers capable of producing a homogeneous mixture of dry powders. 
Depending on the lot size, the entire lot or only a portion of the dry mix 
may be moistened for molding at one time. A Colton production-size molding 
machine is shown in Figure 4. The dampened mass is placed in a hopper 
(A) which is equipped with a revolving blade, and the mass is allowed to 
drop into one of four circular sections in the rotating circular feed plate 
(B). The feed plate is set above the mold or die plate (C), but they are 
on different centers so that only about 30% of the mold plate is covered by 
the feed plate. The mold plate contains four sets of die holes. In the 
first step of the molding operation, the mass which was dropped into the 
feed plate is moved over one set of dies into which the foot of the packing 
spinner (D) uniformly forces the tablet mass. The packing spinner has a 
spring which can be adjusted to regulate the force (and correspondingly 
the amount of tablet mass filled into the die) and thus to control the tablet 
weight. The mold plate moves to the second position, in which the top 
surfaces of the tablets are smoothed off by the foot of the smoothing spinner 
(E). Any excess powder is removed from the die plate by a rake-off 
(F) in the third position. At the fourth and final position. the tablets 
are ejected onto a conveyer belt (G) by a nest of carefully fitted punches 
(H) which match the dies. The tablets are air dried at room temperature 
as they move along the belt to drop onto a drying tray. Depending on

Special Tablets 
(D) Packing Spinner Unit IV ...,..., t 
(8) Feed Plate 
/ (Not Visible) 
Figure 4 Colton machine for preparing molded tablets. 
the tablet size and the number of dies in a set, the production rate varies 
from 100,000 to 150,000 tablets per hour. The belt drying can be accelerated 
by electrical heating units. warm air currents, or infrared heat lamps 
which are directed onto the conveyer belt. 
At the end of the conveyer belt I the tablets are dropped onto a drying 
tray where they will undergo completion of the drying process. They are 
sampled at this time to check the tablet weight. Weighing of the damp tablets 
at this point gives an estimate of what the dry weight will be and can 
be used to determine what packing spinner adjustments need to be made 
to achieve the correct tablet weight. 
The remaining solvent in the tablets can be removed by air drying on 
trays in a rack or in a circulating-air oven at 100 to 120F for up to 1 hr. 
Microwave drying of 1 to 3 min may be used to reduce the exposure time 
during the drying process. The tablets should be dedusted on a vibrating 
screen or by passing the screen holding the tablets over an exhaust unit 
prior to final evaluation and packaging.

340 Conine and Pilal 
D. Evaluation of Molded Tablets 
The USP now recognizes separate uniformity of dosage unit specifications 
for molded and compressed tablets. Content uniformity standards for molded 
tablets are met if not less than 9 out of 10 tablets taken from a sample 
of 30 as determined by the content uniformity method lies within the range 
of 85.0 to 115.0% of label claim. no unit is outside the range of 75.0 to 
125.0% of label claim, and the relative standard deviation of 10 tablets is 
less than or equal to 6.0%. 
If two or three dosage units are outside the range of 85.0 to 115.0% 
but not outside the 75.0 to 125. 0% range, or if the relative standard deviation 
is greater than 6. 0%. or if both conditions prevail, an additional 20 
units are tested. The uniformity requirement is met if not more than three 
tablets of the 30 are outside the range of 85.0 to 115.0% of label claim. and 
none lies outside 75.0 to 125.0%. and the relative standard deviation of the 
30 tablets does not exceed 7.8%. 
The disintegration test for sublingual tablets is run in the USP disintegration 
apparatus without disks, using water at 37  2C. All six tablets 
should disintegrate completely within the time limit specified in the monograph 
(2 min for nitroglycerin tablets). If one or two of the tablets fail 
to disintegrate completely, a repeat test is made on 12 more, and not less 
than 16 of the total 18 tablets should disintegrate in the specified time [27]. 
If the molded tablets are intended to be completely soluble, a solUbility 
test should be required which includes both rate and completeness of solution 
in a specified amount of water. Dissolution tests have been established 
for many tablets, but they usually are done in large volumes of water. 
For sublingual nitroglycerin tablets. where only small volumes of saliva 
would ordinarily be encountered in actual use. methods have been established 
using very small amounts of media [28,29]. 
One method places the individual tablet on a Millipore illter (0.45 mm) 
in the upper chamber of a plastic Millipore Swinnex 25 filter holder. One 
ml of water is flushed through the chamber at 30-sec intervals up to 2 min, 
and samples at each time interval are collected and assayed [28]. 
In a second method designed specifically for nitroglycerin. a tablet is 
dropped into 5 ml of water purged with nitrogen to remove any oxygen, in 
a cell containing a rotating platinum electrode. The system is operated until 
no further increase in reduction potential is observed. From the data, 
the amount of nitroglycerin in solution at any time interval is obtained [29]. 
Stability studies on each formulation are needed to establish the shelf 
life of the product for both physical and chemical evaluation. Specific procedures 
and methods are in the literature for many drugs. Potency changes 
on aging should be monitored, and special attention should be psid to physical 
changes such as color development, decreased solubility of the tablet, 
and changes in disintegration time and dissolution rate. Special tests developed 
for the evaluation of sublingual nitroglycerin tablets will be discussed 
in the next section. 
A. Mechanisms of Potency Loss 
Since nitroglycerin is a liquid with a significant vapor pressure at ambient 
temperatures. and since each tablet contains only a small amount of

Special Tablets 341 
10- 3 
> 10- 4 
20 25 30 35 40 
Temperature (oc) 
45 50 
Figure 5 Vapor pressure of pure nitroglycerin as a function of temperature. 
nitroglycerin (0.15 to 0.6 mg), the formulation, manufacture. and packaging 
of nitroglycerin tablets present some special problems. Nitroglycerin 
tablets potentially can lose potency in four ways: loss to the atmosphere 
by evaporation. intertab1et migration, sorption by packaging materials, and 
chemical decomposition. The first three mechanisms of potency loss, although 
perhaps not unique to nitroglycerin. are certainly not common modes 
of potency loss in pharmaceuticals. 
The vapor pressure of pure nitroglycerin (Fig. 5), although it increases 
sharply with an increase in temperature, is equal to only about 10- 4 x the 
vapor pressure of water [30]. Due to the minute levels of nitroglycerin in 
tablets, even this slight volatility is sufficient to result in significant losses 
in potency when nitroglycerin tablets are exposed to ambient air currents 
for a few days. Loss of nitroglycerin from conventional tablets spread in 
a monolayer and exposed to ambient (rv25C) air currents is illustrated in 
Figure 6. The term conventional tablets refers to molded tablets formulated 
only with nitroglycerin and lactose, and perhaps a small amount of sucrose 
to serve as a binder. 
The "drafty" environment (Fig. 6) is a location near an air vent while 
the "draft-free" location represents more normal room air circulation. The 
vertical lines represent the 90% confidence error limits for the mean value 
of 30 single-tablet assays. The increases in error limits as the tablets age 
reflect the decrease in content uniformity observed as the tablets lose potency. 
The data shown in Figure 6 are qualitatively aimilar to corresponding 
data reported by other workers [29,31]. although exact agreement

342 Conine and Pikal 
12 10 6 8 
4 2 o 
'iii 300 
:::t -:.. i
 200 ... 
Figure 6 Potency loss of nitroglycerin from conventional tablets exposed 
to ambient air currents (n.,25C). 
between different laboratories cannot be expected due to variations in air 
currents. Clearly, the unnecessary exposure of tablets to air currents 
during manufacture or storage should be avoided. However, with reasonable 
care in the manufacturing process. the drying step is the only phase 
of manufacturing where potency losses via evaporation could be significant. 
During drying, the storage air currents and elevated temperatures 
needed to remove water and alcohol from the freshly molded tablets will 
also remove a measurable amount of nitroglycerin-the amount volatilized 
depending on the drying methodology and the tablet formulation. Data 
typical of potency loss in a forced-air drying oven operated at 40C are 
shown in Figure 7 [32]. The tablets are O.4-mg stabilized tablets which 
contain povidone at a level of 1% of the tablet weight. The povidone is included 
to stabilize the content uniformity. The tablets not only show a 
significant loss in potency beyond about 1 hr but, as might be expected. 
the potency loss depends on the tablet location within the dryer. Since 
essentially all the alcohol and excess water is removed after about 1 hr of 
drying. drying in excess of 1 hr serves only to decrease the mean potency 
and to magnify the effect of tablet location on tablet potency. 
Since the rate of nitroglycerin loss for a given tablet will depend on 
the temperature, air velocity. and partial pressure of nitroglycerin in the 
immediate vicinity of that tablet. these variables should be uniform throughout 
the drying oven. The difference between the two curves in Figure 7 
is probably due to a lower temperature and a higher pressure of nitroglycerin 
for the air near the air exhaust port. Prolonged drying and lack 
of uniform drying will result in tablets suffering variable potency loss. resulting 
in poor content uniformity.

Special Tablets 343 
Although, in principle, nitroglycerin will leak from loosely sealed containers, 
the leak rate would be negligible for any closure likely to be used. 
For example, Fusari [31] found that 100 tablets stored in a glass bottle 
without a closure lost only about 2% in potency during 1 month of storage 
at ambient conditions. Thus, heroic efforts to seal the containers are unnecessary. 
(Nitroglycerin sorption by packaging components is a more 
serious problem and will be discussed Iater , ) 
Intertablet Migration 
On aging for several months, conventional nitroglycerin tablets normally develop 
very poor content uniformity with only minor losses in potency [30, 
33]  This phenomenon is illustrated in Figure 8 for a lot consisting of 
0.3-mg conventional tablets. For fresh tablets (8 days old), the assays 
(wt% nitroglycerin in each of 30 tablets) are clustered tightly around the 
mean value and the content uniformity parameter e, defined as the relative 
standard deviation for assay (wt% nitroglycerin) of 30 tablets, in only 3.9%. 
As the tablets age (at 25C in closed glass containers), a greater range of 
assay values is observed until, at 50 days, a significant number of both 
subpotent and superpotent tablets are found. The content uniformity parameter 
0 is 13.3%, significantly higher than found for the fresh tablets. 
Most of the loss in content uniformity occurs during the first 2 months 
after manufacture (Fig. 9). The data shown represent mean values for 2 
lots (153 days), 3 lots (88 days), and between 7 and 11 lots for all other 
points. Although all single lots show qualitatively the same behavior as 
1.0 --;::------------------, 
/'Tablets Near Air Exhaust 
o 1 2 3 
4 5 
Figure 7 Potency loss of tablets (O.4-mg nitroglycerin) in a forced-air 
drying oven at 40C. Tablets contain 0.36 mg povidone added to stabilize 
content uniformity.

0::: 6.9% 
0.6 0.7 
0= 13.3% 
0::: 3.9% 
0.8 0.9 1.0 1.1 
(wt %)Nitroglycerin in Tablet 
Conine and Pikal 
50 Days 
21 Days 
8 Days 
Figure 8 Loss of content uniformity on aging: O.3-mg conventional tablets. 
shown in Figure 9, significant quantitative differences do exist (Le., some 
lots develop poorer content uniformity than others). While the data shown 
in Figure 9 refer only to conventional tablets manufactured by Eli Lilly 
(prior to December 1972), conventional tablets manufactured by Parke-Davis 
exhibit qualitatively the same behavior [33]. 
The observation that some tablets increase in potency while others decrease 
is a most unusual observation that is attributed to the phenomenon 
of capill8l'Y condensation [30]. Any liquid that is condensed in a capillary 
tube will have a lower vapor pressure and, therefore, lower free energy G 
than the same liquid in the bulk state. This reduction in vapor pressure 
becomes more pronounced the smaller the diameter of the capillary, and it 
is significant only for very small capillaries. Nitroglycerin tablets contain 
a significant number of cracks and pores which behave as small capillary 
tubes; due to nonuniformity in the molding process, the volume of such 
small pores exhibits significant variation within a group of nominally equivalent 
tablets. Thus, freshly prepared tablets exhibit significant and variable 
deviations from equilibrium due to a number of empty or partially filled 
small pores. As the tablet system (e.g., 100 tablets in a bottle) ages and 
approaches equilibrium, nitroglycerin is transferred from regions of high 
free energy (i.e., nitroglycerin coated on the lactose surface) to the empty 
or partially filled small pores, which are states of lower free energy. This

Special Tablets 
12 - 
b 8
2 -f---.,.---r---,----r---r-----,------,--,..,Ir~.-----' 
o 20 40 60 80 100 120 
Tablet Age (days) 
140 160 1095 
Figure 9 The content uniformity parameter " as a function of tablet age: 
Conventional tablets (0. 3-mg nitroglycerin). 
transfer is shown schematically in Figure 10. Here, the relative vapor 
pressure PIP'. where P is the vapor pressure of nitroglycerin in a given 
state and P' is the vapor pressure of bulk nitroglycerin (where surface 
effects are negligible), is lowered from about 1. 0 to 0.9 by the transfer 
process. Thus fl G, the free energy change for this process is negative 
and the change is spontaneous in the thermodynamic sense. 
Conventional Tablet 
"Typical" Pore 
Surface Phase 
PIP "'1 
Stabilized Tablet 
AG >0 _ NoMigration r---------, "1tff \ Nitroglycerin U "Typical" Pore 
PIP- NO.9 
Surface Phase 
PIP- <0.8 
Figure 10 Mechanism for the migration effect (illustrated for transfer to 
empty pores). Top, conventional tablet. Bottom. stabilized tablet.

346 Conine and Pikal 
Table 3 Nitroglycerin Absorption by Polymer Films 
Absorption of 
nitroglycerin (wt%) 
Cryst allinity 
Polymer type (X-ray) 25C 37C 
Vinyl (I) Amorphous 28.9 24.8 
Vinyl (II) Amorphous 25.6 20.8 
Vinyl (III) Amorphous 25.6 20.8 
High-density Highly 0.030 0.028 
polyethylene (IV) crystalline 
Low-density Very weak 3.0 2.3 
polyethylene (V) cryst alline 
Ionomer (I X) Essentially 0.81 0.89 
aThe film numbers in parentheses correspond to those in reference [37]. 
The ionomer is Surlyn 1604 (DuPont). 
Since a given tablet is not an isolated system, intertablet as well as 
Intr-at ablet transfer takes place, resulting in intertablet potency variations 
of the same order of magnitude as the intertablet variations in the volume 
of small pores. In summary, the migration effect is a direct result of the 
volatility of nitroglycerin, the presence of small pores, and an intertablet 
variation in the volume of small pores. The mechanism of stabilization 
shown in Figure 10 will be discussed in Section III. B . 
Sorption by Packaging 
Because nitroglycerin is volatile and has a great affinity for many common 
packaging materials, nitroglycerin tablets may suffer significant potency 
losses via sorption by the packaging [31,33, 34-37]. For example, conventiona! 
tablets strip packaged in an aluminum foil and low-density polyethylene 
laminate lost about 90% of their nitroglycerin to the package [35]. 
As the data in Table 3 illustrate, plastics vary greatly in their affinity 
for nitroglycerin. These data were generated [37] by allowing the polymer 
films (or plastics) to absorb nitroglycerin from a 10% trituration of nitroglycerin 
and lactose until equilibrium was attained, and thus indicate the 
solubility of nitroglycerin in the plastic. Vinyls absorb the most nitroglycerin, 
and high -density polyethylene, due to its high crystallinity. absorbs 
the least. The ionomer (IX), although less crystalline than the lowdensity 
polyethylene film (V), absorbs significantly less nitroglycerin. 
This effect is believed due to the chemical composition of the ionomer. An 
ionomer has a chemical composition similar to that of polyethylene except 
that the ionomer contains structurally bound anions (Le . carboxyl ions) 
and their corresponding counterions (Le,; Na+ ions). One might speculate 
[37] that the electrostatic field of the ions is sufficient to "salt out" nitroglycerin 
in much the same way that electrolytes decrease the aqueous solubility 
of many nonpolar solutes.

Special Tablets 347 
While stabilized tablets show less nitroglycerin loss to packaging [29, 
37J even stabilized molded tablets show excessive loss of potency in most 
types of strip packaging [29,37]. An aluminum foil and thermoplastic 
polymer laminate appears to be necessary for achievement of stability in a 
unit-dose strip package comparable to the stability in conventional packaging 
(100 tablets in a screw-capped glass bottle) [37]. The aluminum 
foil is necessary to eliminate potency loss by diffusion through the package 
and evaporation to the atmosphere. The thermoplastic polymer is needed 
to allow the package to be sealed by a heat-sealing process. Obviously, 
the thermoplastic polymer must not absorb excessive amounts of nitroglycerin. 
Stabilized nitroglycerin tablets do maintain acceptable potency and 
content uniformity when strip packaged in an aluminum foil and Surlyn 
1604 laminate [37]. 
Not even the standard commercial package (100 tablets in an amber glass 
bottle with a screw cap) is free of package absorption problems. The 
stuffing used to retard tablet breakage absorbs nitroglycerin, and the cap 
liners cause some loss of nitroglycerin by absorption-and perhaps by diffusion 
through the liner facing into the bulk of the liner. Cotton stuffing 
appears to absorb about 5 times as much nitroglycerin as rayon stuffing 
[33], at least with 0.4-mg conventional tablets. Rayon stuffing absorbs 
about the equivalent of two O.4~mg tablets when packaged with 0.4-mg 
conventional tablets [33]. Tablets packaged with vinyl cap liners offer 
the least protection against potency loss while Excelloseal is only slightly 
better. Tin foil, Mylar (polyethylene terephthalate), and Aclar (a fluorohalocarbon) 
offer the best protection against potency loss [33]. 
Chemical Decomposition 
Although chemical stability is normally not a problem with conventional nitroglycerin 
tablets, both polyethylene glycol 400 and povidone (molecular 
weight '" 35,000), which are used to stabilize content uniformity, may accelerate 
the hydrolysis of nitroglycerin. 
Chemical decomposition via hydrolysis is illustrated by the data in 
Table 4 [37] for nitroglycerin-providone-lactose systems. Both 1, 2~dinitroglycerin 
and 1,3-dinitroglycerin were present in the aged samples in 
roughly eq ual amounts. Dinitroglycerin content is expressed as weight 
percent of the total nitroglycerin compounds. Within the uncertainty of 
the data, both the nitroglycerin loss and the dinitroglycerin content were 
independent of the povidone concentration above a weight ratio of 0.6. 
Although the thin-layer chromatographic assay [37] is only semiquantitative, 
the data demonstrate that a significant fraction of the nitroglycerin loss 
was due to hydrolysis of the trinitroester to dinitroglycerin species. 
The high -temperature stability of tablets containing povidone is compared 
with that of other formulations in Table 5 [37J. Potency loss at 
high temperature was significantly greater with the povidone-containing 
formulation. Analysis by thin-layer chromatography showed significant 
amounts of the 1,2- and 1,3-dinitroglycerin species in the aged povidone 
formulation but only trace amounts in the other formulations. Although 
povidone-containing tablets show poor high-temperature stability. the stability 
at 25C is satisfactory (approximately 3 to 4% potency loss per year) 
[37, 38J . 
Polyethylene glycol 400 has also been shown to accelerate potency loss 
of nitroglycerin from tablets [39] and from solution [40J. However. tablets 
with satisfactory stability have been formulated with PEG 400 at a

348 Conine and Pikal 
Table Jj Hydrolysis of Nitroglycerin in Nitroglycerin-Polyvinylpyrrolidone 
Nitro loss (%)a Dinitroglycerin content (%)b 
PVP /Nitro 
weight 1. 5 yr/25C + 1. 5 yr /25C + 
ratio 1.5 yr/25C 1 mof50oC 1.5 yrf25C 1 mo/50oC 
0.22 2 11 1 2 
0.65 7 22 4 7 
1.04 12 22 3 9 
1. 56 4 8 
2.13 5 8 
Note; Samples were prepared by dry-blending polyvinylpyrrolidone (PVP) 
and 10% nitroglycerin trituration on 8-lactose. 
aDetermined from nitroglycerin assay on initial and aged samples. 
bExpressed as weight percent of total nitroglycerin compounds (Le., dinitroglycerin 
and trinitroglycerin), determined by semiquantitative thinlayer 
Source: From Pikal, M. J. Bibler, D. A., and Rutherford, B., J. Pharm. 
Sci.. 66; 1293 (1977). Reproduced with permission of the copyright owner. 
Table 5 Potency Loss of 0.3-mg Tablets at High 
Temperature: A Comparison of Formulations 
Potency loss 
Conventional tablet 
(no stabilizer) 
Stabilized tablet 
(1% polyvinylpyrrolidone) 
Stabilized tablet 
(polyethylene glycol)a 
5 mo/45C 
Note: Tablets stored in screw-cap glass bottles with 
rayon stuffing, 100 tablets per bottle. 
~itrostat (Parke Davis). 
Source: From Pikal, M. J., Bibler, D. A.. and 
Rutherford, B., J. Pharm. sa., 66: 1293 (1977). Reproduced 
with permission of the copyright owner.

Special Tablets 349 
o 0.2 0.6 1.0 1.4 1.8 2.2 2.6 3.0 
Weight Ratio, AdditivelNitro - 
Figure 11 Nitroglycerin vapor pressure reduction at 25C by selected 
additives (A, PEG 1000 powder; 0, PEG 400 tablet; ., PEG 400 powder). 
[From Pikal, M. J., Lukes, A. L., and Ellis, L. F., J. Pharm. SeL, 65: 
1278 (1976).] 
weight ratio of glycol to nitroglycerin of 0.85 [31]. This apparent anomaly 
is resolved when stability is examined as a function of the weight ratio of 
PEG 400 to nitroglycerin. It appears that below a weight ratio of approximately 
1, PEG does not significantly affect the st ability of nitroglycerin, 
but at high weight ratios ('V 2), PEG 400 causes extensive hydrolysis of 
nitroglycerin even at 25C [41]. 
B. Stabilization of Content Uniformity 
Empirical observations have indicated that the addition of PEG 400 or 4000 
at a weight ratio of glycol to nitroglycerin of 0.85 would stabilize the content 
uniformity. Similar observations have been made for the addition of 
povidone [38]. These additives are soluble in nitroglycerin [30] at the 
levels used and decrease the vapor pressure of nitroglycerin (Fig. 11). 
Polyethylene glycol 4000 data, not shown in Figure 11, are nearly identical 
to the data shown for the other glycols up to the maximum solubility of 
PEG 4000 in nitroglycerin (weight ratio of 0.9) [41]. The data for di - (2ethylhexyl) 
phthalate are included only for comparison. This material is 
not used as a tablet additive. 
The vapor pressures of nitroglycerin in aged tablets and the corresponding 
content uniformity parameters are summarized in Table 6 [30]. 
The first three rows refer to conventional tablets. and the last four rows 
refer to commercial stabilized tablets (povidone or PEG additive). The relative 
vapor pressure PIP' is the vapor pressure of nitroglycerin in the tablet 
P divided by P', the vapor pressure of pure bulk liquid nitroglycerin.

350 Conine and Pikal 
Table 6 Vapor Pressure and Content Uniformity of Aged Nitroglycerin 
Tablet Formulations 
Relative Content 
vapor uniformity 
Weight pressure 
Potency ratio PIP' a Number 
Additive (mg) (additive ING) at 25C ( %) of lots 
Nonea 0.6 0 0.97 12 (2) 
Nonea 0.4 0 1. 01 12 (3) 
Nonea 0.3 0 0.90 13 ( 6) 
Povidones 0.6 0.59 0.76 5.4 (8) 
Povidonea 0.4 0.89 0.52 5.7 (13) 
Povldones 0.3 1.19 0.31 5.8 (5) 
Polyethyleneb 0.6 0.85 0.64 4.3
c (12) 
glycol (400 
or 4000) 
Note: Tablet age 6 months to 5 years. 
~li Lilly. 
bp arke Davis. 
CEstimated from the published [31] relative standard deviation data calcu1ated 
from nitroglycerin content per tablet and the weight variation given 
for one lot. 
The number in parentheses after the content uniformity parameter (J is the 
number of lots used to generate the average content uniformity listed. 
The value of (J is approximately 3% at the date of manufacture for both 
conventional tablets (Fig. 9) and povidone-stabilized tablets [38]. Thus, 
while conventional tablets show an increase in the content uniformity parameter 
of about 11% on aging, povidone-stabilized tablets (Table 6) and 
PEG-stabilized tablets [31J show an increase of only 2 to 3%. Note that, 
although the stabilized formulations yield reduced nitroglycerin vapor pressures 
from 24 to 69%, all stabilized formulations are equally effective in 
preventing the large increase in content uniformity parameter characteristic 
of conventional tablets. 
The role of the additive in content uniformity stabilization is believed 
to be a reduction of vapor pressure sufficient to make it thermodynamically 
impossible for a significant quantity of nitroglycerin to be transferred from 
the lactose surface to a small pore. This mechanism is illustrated (bottom 
of Fig. 10) for transfer to an empty pore. Here the nitroglycerin on the 
lactose surface is in solution with the additive, giving a relative vapor 
pressure less than 0.76 (Table 6). Most of the small pores in a tablet are 
only small enough to lower the vapor pressure of nitroglycerin by about 
15% [30]. For purposes of illustration. a typical pore in Figure 10 is assumed 
to be small enough to lower the vapor pressure by 10% (Le.,

Special Tablets 351 
PIP' = 0.9). Thus the transfer of nitroglycerin from a system of lower 
vapor pressure (PIP' ~ 0.76) to a region of higher vapor pressure (PIP' = 
0.9) would result in a positive free-energy change (flO > 0), and the 
process is therefore thermodynamically impossible. Note that the role of 
the stabilizing additive is not to minimize the migration rate by slowing 
the rate of volatilization. Reduction of the rate of volatilization is not particularly 
important within the context of the migration effect. 
Although absorption by the packaging materials is not normally the 
major cause of poor content uniformity, it should be noted that reduction 
of the vapor pressure of nitroglycerin in the tablet will reduce the extent 
of package absorption and, therefore, will also reduce content uniformity 
problems arising from package absorption. 
C. Testing Procedures 
Vapor Pressure 
Since all mechanisms of potency loss except chemical decomposition depend 
directly on the vapor pressure of nitroglycerin in the tablet. any evaluation 
of a proposed formulation should include vapor pressure measurement or 
determination of some property strongly correlated with vapor pressure. 
The vapor pressure of nitroglycerin in molded tablets may be measured 
directly by a modification of the gravimetric Knudsen effusion technique 
[30,42]. Here the sample is placed in a chamber having a small orifice in 
the top, and the chamber is suspended from one arm of a high-vacuum 
microbalance. The rate of mass loss through the orifice is determined in 
a high vacuum (10- 6 torr). For pure materials the vapor pressure is calculated 
directly from the proportionality between the rate of mass loss and 
the vapor pressure. However, for nitroglycerin tablets. vaporization of 
water present as an impurity may result in an appreciable "background" 
mass loss. and nonequilibrium effects may be present (Le . the nitroglycerin 
vapor may be unable to escape from the sample rapidly enough to 
maintain the equilibrium vapor pressure in the Knudsen cell). Thus. the 
rate of nitroglycerin loss is not directly proportional to the vapor pressure. 
Special procedures and data analysis are needed to extract vapor pressure 
data from the rates of mass loss [42]. 
In view of the special equipment and complex data analysis needed for 
the direct measurement of vapor pressure, convenience may dictate that an 
alternate property be measured that is strongly correlated with vapor pressure. 
The open dish evaporation test used for this purpose has been described 
in promotional literature as well as scientific literature [29,31]. 
Here tablets are placed in a single layer in an open glass dish and are exposed 
to normal laboratory air currents. Circulating air evaporates SOme 
of the nitroglycerin, causing loss of potency, which is monitored by assay 
for nitroglycerin as a function of time. Results of such a test may be 
seen in Figure 6. Although the test is simple and. if carefully done, capable 
of providing evaporation rates which are. as a first approximation, 
proportional to the initial vapor pressure of nitroglycerin in the tablets 
[41], great care must be exercised to insure that the air currents are uniform 
and reproducible or the data obtained are too imprecise to be useful. 
For example, the data in Figure 6 illustrate qualitatively the difference observed 
when the air currents differ. 
A modified open dish evaporation test [43] is illustrated by the schematic 
shown in Figure 12. The flow of air over a set of tablets is measured

Valve --fWI 
Flexible Tubing --11 
Rubber Stopper 
GasDispersion Tube 
1 liter Inverted Beaker 
Single layer of Tablets 
Inverted go x 500mm 
Crystallization Dish 
Compressed Air 
Conine and Pikal 
Figure 12 Controlled flow rate evaporation test: schematic diagram. 
and controlled by the flow meter valve. Moreover, placing the tablets inside 
the inverted beaker ensures that only air initially devoid of nitroglycerin 
is being passed over the tablets. Thus, the modified evaporation 
test standardizes the evaporation conditions and allows more reproducible 
data to be obtained. An example of data obtained [43] with this procedure 
is shown in Figure 13. The tablets were 0.4-mg, stabilized with 0.36 mg 
of povidone. An increase in flow rate from 2 to 4 .13 hr- 1 clearly increases 
the evaporation rate. Data obtained at 6 ft 3 hr- 1 (not shown) 
were essentially the same as the data for 4 ft 3 hr- 1 Evidently, at flow 
rates greater than approximately 4 ft 3 hr- 1, gas phase diffusion of nitroglycerin 
through the tablet matrix is rate controlling for loss of nitroglycerin. 
Isothermal thermogravimetric analysis has also been used as a measure 
of nitroglycerin volatility [44]. The weight loss of two tablets is followed 
for 1.5 to 4 hr at so-c with a nitrogen flow rate of 20 ml min-I. To avoid 
loss of water of hydration, anhydrous lactose should be used to formulate 
the tablets. If loss of nitroglycerin via decomposition is ignored, the 
thermogravimetric experiment at BOOC is probably equivalent to a controlled 
open dish evaporation test where the rate is accelerated by increased temperature. 
ThUs, it is reasonable to assume that the rate is proportional 
to the vapor pressure of nitroglycerin (at 80C), with the proportionality 
constant being some unknown function of the nitroglycerin diffusion coefficient 
and the tablet porosity. To the extent that the rate is sensitive 
to porosity, intertablet variation in porosity could result in variable results 
since only two tablets are used in a given experiment. 
The authors of the foregoing study [44] did not address either decomposition 
or variations of rate with tablet porosity. If one assumes that 
these potential problems are minor, thermogravimetric analysis offers a 
rapid method for a relative measurement of nitroglycerin vapor pressure 
at elevated temperatures.

Special Tablets 353 
>. 0.5 
0 0.4 ci
ii 0.2 :e 
. 0.1 -0.09 
0 0.08 
III 0.07 .. IL 0.06 
0 4 8 12 16 20 24 28 
Figure 13 Potency loss of 0.4-mg tablets at selected air flow rates at 
25C. Nitroglycerin tablets containing 0.36 mg povidone added to stabilize 
content uniformity. 
Package Adsorption 
So-called package adsorption may be detected by a solvent extraction of 
all packaging material which is in vapor phase contact with the tablets, 
followed by an assay for nitroglycerin. Ethanol was found to be a suitable 
solvent for most types of strip packaging [37]. Simple rinsing of the packaging 
is normally not sufficient to remove absorbed nitroglycerin. Extraction 
times of 1 to 2 days may be necessary (37]. 
Content Uniformity Stability 
The content uniformity should be determined shortly after manufacture by 
single-tablet assay (30,33] of large tablet samples (about 30 or more). 
The content uniformity parameter 0" should be about 5% or less for freshly 
manufactured tablets. After the tablets are packaged in the containers 
of interest, (J should be determined at monthly intervals for several months. 
Normally, if poor content uniformity is going to develop, a significant increase 
is e will be obvious after 2 to 3 months of storage of 25C (Figs. 8 
and 9). 
Chemical Stability 
The chemical stability is best studied by storage of a large number of tablets 
(more than 100). in glass bottles with no stuffing and with foil cap 
liners, so that package absorption is negligible. Thus, any potency loss 
can be attributed to chemical decomposition. Thin-layer chromatography 
is also useful in that trace levels (about 2%) of dinitroglycerin and

Conine and Pikal 
mononitroglycern may be detected-to confirm decomposition via hydrolysis 
[36,41). Since the decomposition rate increases sharply with increasing 
temperature (Table 4) [37,41). accelerated stability studies may be used 
for a preliminary evaluation of any proposed formulations. Storage at 
50C for 1 to 2 months normally results in decomposition at least as extensive 
as that of storage at 25C for 2 years. 
The effect of humidity (moisture content) on stability may be studied 
by first placing bottles of tablets without caps in a closed chamber of 
fixed relative humidity to equilibrate for about 24 hr. Constant humidity 
is conveniently maintained by a mixture of a salt and its saturated aqueous 
solution , The bottles are then closed, and the stability test is started. 
Simulated Patient Use Tests 
Tests designed to simulate the conditions generated when a patient repeatedly 
opens the bottle and removes a tablet have also been used. For 
example, a bottle is opened, the rayon stuffing is descarded, and an initial 
assay is obtained for a small tablet sample (perhaps 3 tablets). The 
bottle is then opened daily for a fixed time to simulate a patient's removal 
of a tablet. Each week a small tablet sample is taken for assay until approximately 
15 tablets remain, at which time the remaining tablets are 
assayed to obtain a measure of average potency and content uniformity. 
However, since no measurable nitroglycerin will evaporate during this procedure 
(the entire bottle volume contains less than 0.2 ug nitroglycerin 
in the vapor state), this type of test offers no real scientific advantage 
over the testing procedures described previously. 
The requirements for SUblingual tablets are speed of absorption and a correspondingly 
rapid physiological response, which are normally best achieved 
with a rapidly soluble molded tablet. However, compressed subungual tablets 
have also been prepared which disintegrate quickly and allow the active 
ingredient to dissolve rapidly in the saliva-and to be available for 
absorption without requiring the complete solution of all the ingredients of 
the formulation. Erythrityl tetranitrate , isosorbide dinitr-ate , and isop 1'0terenol 
hydrochloride are marketed as compressed tablets for sublingual 
use. Compressed nitroglycerin tablets have been described in the literature 
{22, 24]; formulations for these tablets contain large amounts of cellulosic 
material and may also contain lubricants, gfidants , flavors, coloring 
agents, and stabilizers. 
Compared to molded tablets, compressed tablets of this type normally 
have less weight variation and better content uniformity. The USP [27] 
requirement for the uniformity of dosage units is met if each of the 10 
tablets tested lie s within the range of 85.0 to 115.0%of label claim and the 
relative standard deviation is less than or equal to 6.0%. If one unit is 
outside the 85.0 to 115.0% range and no unit is outside 75.0 to 125.0% of 
label claim, an additional 20 tablets are tested, and the requirements are 
met if not more than one out of 30 tablets is outside the 85.0 to 115.0% 
range but none lies outside of 75.0 to 125.0% of label claim and the relative 
standard deviation of the 30 dosage units does not exceed 7.8%. The 
tablets are also harder and less fragile, thereby avoiding weight and potency 
loss that occur by the erosion of the molded target edges.

Special Tablets 
Example 4: Nitroglycerin Tablets (0.3 mg, 
Di rect-Comp ression ) 
Nitroglycerin (10% of 
microcrystalline cellulose) 
Microcrystalline cellulose 
Coloring agent 
Quantity per 
3.0 mg 
2.0 mg 
29.0 mg 
q i s , 
Screen and blend the powders and compress 
into tablets. 
Compressed nitroglycerin tablets were reported to have a rapid disintegration 
time of from 3 to 7 sec by the USP method for sublingual tablets 
[27], as well as rapid response time as measured by an increase in pulse 
rate of 10 to 13 beats/min within 3 min in human volunteers [24]. However, 
in some clinical patients, these compressed tablets did not appear 
to disintegrate or release the medication for absorption. In these subjects, 
the compressed tablets either gave no response or gave a delayed response 
when compared with molded tablets [45,46]. 
Example 5: Nitroglycerin Tablets (0.3 mg, 
Microcrystalline cellulose 
Anhydrous lactose 
Starch, USP 
Coloring agent 
Nitroglycerin (as the spirit) 
Calcium stearate 
Quantity per 
21. 00 mg 
5.25 mg 
3.00 mg 
0.30 mg 
0.30 mg 
0.15 mg 
Blend the excipients and the coloring agent, 
and granulate with an ethanol solution of 
povidone and nitroglycerin. After the 
granulation is dried and milled, it is 
blended with the calcium stearate and 

356 Conine and Pikul 
Perhaps insufficient saliva is present to allow complete removal of nitroglycerin 
from the absorbent cellulose. A strong negative psychological 
effect resulting from the presence of undissolved cellulose in the patient's 
mouth has also been suggested as the reason for product failure [29,47]. 
The methods for evaluation of compressed sublingual tablets are the same 
as those given for molded sublingual tablets. 
The purpose of buccal tablets is the same as that of sublingual tablets 
(Le . , absorption of the drug through the lining of the mouth). While the 
advantage of sublingual medication is rapid response, buccal tablets are 
most often used when replacement hormonal therapy is the goal. Although 
completeness of absorption is desired, a high rate of absorption is not desirable. 
Flat, ellipitcal or capsule-shaped tablets are usually selected for 
buccal tablets, since they can be most easily held between the gum and 
cheek. The parotid duct empties into the mouth at a point opposite the 
crown of the second upper molar, near the spot where buccal tablets are 
usually placed. This location provides the medium to dissolve the tablet 
and to provide for release of the medication. 
Methyl testosterone and testosterone propionate are the most commonly 
used buccal tablets. The following formulation is an example of a typical 
buccal tablet. 
Example 6: Methyltestosterone Buccal 
Tablets (10 mg) 
Lactose, USP 
Sucrose, USP 
Acacia, USP 
Talc, USP 
Magnesium stearate, USP 
Quantity per 
10 mg 
86 mg 
87 mg 
10 mg 
6 mg 
1 mg 
q .s , 
Put the drug and excipients through a 
GO-mesh screen and blend. Moisten with 
water to make a stiff mass; pass through 
an a-mesh screen and dry at 40C. Reduce 
the particle size by passing the dried 
granulation throguh a 10-mesh screen; 
blend in lubricants and compress.

Special Tablets 357 
Compressed buccal tablets are prepared either by the procedures 
used for granulation (as described) or by direct compression. In Example 
6 the formulation contains no disintegrants, so the tablet will dissolve 
slowly. Flavoring agents and sweeteners are sometimes added to make 
the tablets more palatable, but this practice has been criticized since increased 
flow of saliva may result. It is important to minimize the swallowing 
of saliva during the time that the buccal tablet is held in place, 
since compounds administered by the buccal route are either not absorbed 
from the gastrointestinal tract or are rapidly metabolized on the first pass 
through the liver. Since buccal tablets are to be held in the mouth for 
relatively long periods of time (from 30 to 60 min), particular care should 
be taken to see that all the ingredients are finely divided so that the tablets 
are not gritty or irritating. 
Water-soluble cyclodextrans have been used as adjuvants to enhance 
the absorption of steroidal hormones from the lining of the mouth. To 
prepare these materials a 40% aqueous solution of 2-hydroxypropyl or 
poly-l3-cyclodextran was saturated with the steroid, freeze-dried, and compressed 
into tablets. Testosterone derivatives administered sublingually 
as either tablets or solutions produced serum levels two to three times 
greater than that of the drug alone or when it was solubilized with polyethylene 
glycol 20 sorbitan monooleate [48]. This elevated serum level 
was seen only when the adsorption took place from the oral cavety but 
not from the GI tract where the absorbed steroid would be removed on 
first pass through the liver and by direct metabolism in the intestinal tissues. 
A number of formulations designed as long-acting buccal tablets have 
been published in the patent literature. The basis for these formulations 
is the use of viscous natural or synthetic gums or mixtures of gums which 
when present in the formulations. can be compressed to form tablets which 
absorb moisture slowly to form a hydrated surface layer from which the 
medicament slowly diffuses and is available for absorption through the buccal 
mucosa. If the tablet can be maintained in place, absorption can take 
place for periods up to 8 hr. 
Several patents cover the use of hydroxypropyhnethylcellulose (HPMC) 
alone or blended with hydroxypropylcellulose (HPC) , ethylcellulose (EC), 
or sodium carboxymethylcellulose (SCMC) as Synchron carriers [49- 52] . 
Some restrictions are made on the USP type, viscosity, or moisture level 
of the HPMC. The HPMC may be treated with oxygen or moisture to oxidize 
or hydrolyze it prior to incorporating it into the formulation or it 
may be used in the untreated form. Release profiles of the drug from 
tablets of this type follow a zero-order rate [53]. 
Tablets have also been prepared using polyacrylic copolymer (Carbapol 
934, B. F. Goodrich Chemical Co.) blended with HPC [54] or sodium 
caseinate [55] for long-acting buccal absorption. Other tablet bases include 
sodium polyacrylate (PANA) combined with carriers such as lactose, 
microcrystalline cellulose, and mannitol [56]. Natural gums such as locust 
bean gum, x anthan , and guar gum have also been utilized [57,58]. 
Some polymers have mucosal adhesive properties that aid in holding 
the tablet in position at the adsorption site between the gum and the 
cheek or lip. PANA and Carbapol 934 have been reported to possess 
these properties [54,56]. Two-layer tablets have been prepared with an 
adhesive and a nonadhesive layer [54]. An in vitro method to measure 
the adhesiveness of various materials to mucus has been developed based

358 Conine and Pikal 
on the force required to detach a glass plate coated with the test substance 
from isolated mucous gel [59]. Time must be allowed to hydrate 
the materials in order to obtain a satisfactory evaluation. Carbapol 934, 
SCMC. tragacanth and sodium alginate had good mucosal adhesive properties, 
whereas povidone and acacia were poor when measured by this method. 
Following are examples of long-acting buccal tablets: 
Example 7: Nitroglycerin Buccal Tablets 
(2 mg) [50] 
Quantity per 
Ingredient tablets 
Nitroglycerin on lactose (1: 9) 
Stearic acid 
Lactose anhydrous 
20 mg 
16 mg 
10 mg 
2 mg 
0.4 mg 
q .s , 70 mg 
The cellulose ethers are blended with the lactose 
and then the nitroglycerin dilution and 
lubricant are added and mixed. The tablets 
are then compressed from the powder blend. 
Example 8: Prochlorperazine Maleate Buccal 
Tablets (5 mg) [57] 
Prochlorperazine maleate 
Locust bean gum 
Xanthan gum 
Sucrose powder 
Magnesium stearate 
Quantity per 
1.5 mg 
1.5 mg 
3 mg 
47.5 mg 
0.5 mg 
1.0 mg 
A blend of prochloperazine maleate, the gums, 
and sucrose is granulated with a solution of 
povidone in aqueous alcohol. After the granulation 
is sized and blended with the lubricants , 
it is compressed into tablets.

Special Tablets 359 
Weight variation. content uniformity. hardness. and friability are determined 
by the same procedures used for compressed tablets. The disintegration 
evaluation differs in that the test for buccal tablets is run in 
water at 37C, according to the USP emthod [27] for uncoated tablets 
using disks. The requirement is that 16 out of 18 tablets should disintegrate 
within 4 hr. A long dissolution time is allowed since buccal tablets 
are normally designed to release the medication slowly. The usual 
disintegration time for a compressed tablet would be between 30 and 60 
Tablets have been designed for vaginal administration in the treatment of 
local infections as well as for systemic absorption and absorption into the 
vaginal tissue. The vaginal wall consists of highly vascular tissue providing 
the potential for- excellent absorption across the membrane lining. 
The venous circulation from this area drains through the hypogastic vein 
directly into the 'inferior vena cava thus bypassing .the portal vein and 
avoiding the rapid destruction of those drugs which are susceptible to 
first-pass metabolism in the liver. 
Only those compounds that have specific use in treating the femal reproductive 
system are usually administered by the vaginal route, although 
many drugs are well absorbed this way and would give effective blood 
levels. The absorption of compounds used to treat local vaginal infections 
is not necessarily desirable, since this could lower the effective concentration 
of the drug directly in contact with the infecting organism. Many 
types of products have been designed for vaginal administration including 
creams, gels, suppositories, powders, solutions, suspensions, and sponges 
as well as tablets, which are the subject of this discussion. Estrogens 
have been administered to increase the level in the vaginal tissue in the 
treatment of atropic vaginitis and further absorption into the system is 
not seen as beneficial. Progesterones such as flugestone acetate have 
been administered intravaginally on sponges to syncronize estrus in sheep 
and other domestic animals [60]. Tablets could also be used but are not 
considered the dosage form of choice for this purpose. 
Vaginal absorption follows first-order kinetics and has been described 
as two parallel pathways, a lipoidal and an aqueous pore pathway. This 
is based on a study of absorption of aliphatic acids and alcohols in the 
rabbit [61]. The plasma level of propranolol in women after vaginal administration 
has been shown to be much higher than when the product is 
given orally [62]. Cyclodextran formulations of hydrophilic drugs such as 
amino glycosides , B-Iactam antibiotics, and peptides are reported to be 
more readily absorbed from the nasal cavity, vagina, and rectum than 
when the drug is administered alone [63]. 
Despite the demonstrated effectiveness of systemic absorption through 
the vaginal wall, the most frequent use of vaginally administered medication 
and especially tablets is in the treatment of localized vaginal infections 
such as Candida albicans, yeast. and Haemophilus vaginalis. The most 
commonly used drugs in the treatment of these infections are nystatin. 
clotrimazole, and sulfonamides. The formula and design of vaginal tablets 
should aim for the slow dissolution or erosion of the tablet in the vaginal

360 Conine and Pikal 
Example 9: Triple-Sulfa Vaginal Tablets 
Quantity per 
I ngredient tablet 
Guar gum 
Magnesium stearate 
166.7 mg 
166.7 mg 
166.7 mg 
400 mg 
400 mg 
60 mg 
30 mg 
10 mg 
The sulfonarnides , urea, lactose, and guar gum 
are blended together and granulated with water. 
After drying, the granulation is sized and 
blended with the starch and lubricant and then 
secretions, as a rate sufficient to provide an effective level of medication 
for as long a time as possible. The same approach is also used in the formulation 
of buccal tablets and troches. The tablet should remain in one 
piece during dissolution and not break into fragments. Vaginal tablets 
weigh from 1 to 1-1/2 g, are flat with an oval-. pear-, or bullet-shaped 
silhouette, and are usually not coated. They can be inserted with the aid 
of an applicator that is provided by the manufacturer. The treatment for 
these infections usually is one to two tablets once or twice daily for 2 
Since these tablets are not subject to peristaltic action. a method of 
testing tablet disintegration under static conditions has been devised that 
should give a more realistic measure of what could be expected in actual 
use and also serve as a quality control test [64]. The apparatus is simple 
and consists of two pieces of no. 9 mesh screen wire. The tablet is 
placed on a screen and covered with the second screen and placed in distilled 
water at 37C in the horizontal position. The disintegration time is 
defined as the time when the two screens touch. This method has been 
used for both effervescent and noneffervescent tablets. 
Sustained release-type formulations that were previously described 
under buccal tablets also include references to and examples of vaginal 
tablets using HPMC or vegetable gums [49,65]. 
Rectal administration of drugs is an old and accepted means of treatment 
for both conditions requiring systemic absorption and the alleviation of 
local symptoms. The small veins of the lower colon drain through the

Special Tablets 361 
inferior mesenteric vein into the portal vein thus exposing any absorbed 
compound to potential first-pass metabolism. Other veins from this part 
of the colon flow into the vena cava so that first-pass metabolism is avoided. 
The proportion of absorbed compound channeled through each of these two 
pathways has not been determined [66]. The availability of the medication 
for absorption depends on the release of the drug from the dosage form 
and its dissolution. The volume and nature of the rectal fluid, its buffer 
capacity. p'H, and surface tension playa large part in this but are subject 
to wide variation. even within single subjects, resulting in variability 
of absorption by this route of administration [66]. There is also the possibility 
of premature expulsion of the dosage form before sufficient absorption 
takes place. 
The suppository has been the customary means of rectal administration 
of medication. Suppository vehicles are most often cocoa butter or some 
other fatty material with similar properties that depend on melting at body 
temperature to release the dru g . Water- soluble solid PEG vehicles have 
been used, but the rate of solution controls the release. PEG bases have 
been criticized as being irritating [67]. The limitations of theobroma oil 
and the PEG vehicles. which include the promotion of decomposition of 
some drugs, and the handling requirements, such as the necessity of refrigeration, 
have restricted the usage of the suppository dosage form. 
The physician, when treating illnesses in which the patient may be 
temporarily unable to swallow tablets or capsules due to nausea, asthmatic 
attacks. or other conditions that make swallowing difficult, may instruct 
the patient as an alternative to administer the tablet or capsule rectally. 
Although this is usually an emergency measure. some consideration should 
be given to the design of tablets for rectal administration. In the review 
article of de Bleay and Polderman [66]. the rational for the design of rectal 
delivery forms including capsules is discussed but fails to consider 
Tablets, which disintegrate rapidly in very small volumes of water to 
form pastes. can be formulated. Such tablets, which disintegrate under 
static conditions in an amount of water equal to only a few times the tablet 
weight, present the drug in a form for absorption equivalent to that 
from the suppository, unless the presence of a lipid base promotes absorption. 
Unfortunately, little is known about rectal delivery from tablets 
and additional studies would be required to demonstrate the extent of bioavailability. 
Tablets offer some distinct advantages over suppositories 
[66] in not requiring refrigeration as well as demonstrating better product 
stability. even at room temperature. Suppositories containing such compounds 
as aspirin and penicillin G sodium have limited product stability, 
even under refrigeration. Tablets of these products are quite stable and 
can be readily formulated. 
The oil or PEG suppository bases act as their own lubricants for insertion. 
but the HPMC film coat on the tablet could also provide some lu bricant 
action. especially in the presence of water, even in small amounts. 
If this should prove to be inadequate, further lubrication could be supplied 
by a jelly. The ultimate usefulness and acceptability of rectal tablets 
awaits further studies. 
The following formulation for tablets will disintegrate to form a paste 
within a few minutes in the presence of four to five times its weight in 
water under static conditions in a water bath. The addition of highefficiency 
disintegrants , such as eroscarmellose sodium and crosslinked 
povidone, will also produce a rapid tablet disintegration.

Example 10: Rectal tablet Prochlorperazine 
(25 mg) 
Conine and Pikal 
Core Tablet 
Magnesium stearate 
Coating (aqueous solution) 
PEG 6000 
Propylene glycol 
Quantity per 
25 mg 
600 mg 
210 mg 
30 mg 
52 mg 
14 mg 
8 mg 
The prochlorperazine is blended with the lactose 
and starch and the mixture granulated is with 
an aqueous solution of povidone. After drying 
and sizing, the granulation is blended with the 
lubricants and additional starch and then compressed. 
The tablets can then be film-coated 
with the HPMC solution. 
Tablets which are to be added to water or other solvents to make a solution 
containing a fixed concentration of the active ingredient are known 
as dispensing tablets. Most commonly they are used to prepare antiseptic 
solutions such as mercuric chloride or cyanide at dilutions of 1/1000. The 
tablets are usually large and contain no insoluble materials since they will 
be made into a clear solution. Because of their toxic nature, they are 
made in disinctive, unusual shapes such as diamond, triangle, or coffinshaped. 
In order to call further attention to their toxicity, they are 
marked either with the word poison Or with skull and crossbones. They 
are also packaged in bottles of distinctive shape with knurled or rough 
edges, so that anyone picking up the container would be aware that it is 
a toxic item. 
The following is a formula for a dispensing tablet suitable for preparing 
1 pint of 1/1000 mercuric chloride solution.

Special Tablets 
Example 11: Mercuric Chloride Dispensi ng 
Quantity per 
Ingredient tablet 
Mercury bichloride 
Potassium alum (powdered) 
Tartaric acid 
Soluble dyes 
Ethanol-water (75: 25) 
475 mg 
510 mg 
65 mg 
q .s , 
The preparation of dispensing tablets is similar 
to that described for small hand-molded tablets. 
The powders are screened, blended, 
then moistened with ethanol-water (75:25), 
and molded-as described earlier and shown in 
Figures 2 and 3. 
A tablet can also be made for the preparation of eye drops. 
Example 12: Tablet for Ophthalmic Drops of 
Neomycin Sulfate [68] 
Boric acid 
Neomycin sulfate 
Sodium sulfate 
Phenyl mercuric nitrate 
Quantity per 
100 rng 
125 mg 
275 mg 
2.5 mg 
The powders are finely milled and blended, and 
the blend is then compressed into tablets. The 
tablet is dissolved into sufficient sterile water to 
make 50 ml of solution. 
Other types of dispensing tablets which have also been used include 
topical local anesthetics, such as cocaine, and antibiotics. such as bacitracin, 
which are used for topical application or irrigation. 
The technology of tablet production offers an economical and efficient 
means of manufacturing solid units of accurate weight and composition. 
The use of the tablet form has spread far beyond pharmaceutical usage 
into almost every aspect of daily life. These tablets cover a wide range

364 Conine and Pikol 
of shapes and sizes from molded tablets through a variety of compressed 
tablets designed to fill specific needs. 
Reagent tablets have been prepared that are relatively stable and provide 
the ingredients as a single unit to conveniently perform qualitative 
and quantitative tests away from the laboratory. Tablets have been formulated 
to be used to enable diabetics to estimate urinary sugar levels and 
the presence of acetone and other aldehydes or ketones in the urine. 
The addition of the tablet to a premeasured amount of water yields a standard 
reagent solution for a single test. Other tablets are available to determine 
the presence of albumin in the urine and for the detection of occult 
blood. The tablets contain all the ingredients req uired for the test 
and, if necessary, any Lubr-icants or binders that will not interfere with 
the sensitivity of the test. Although tests utilizing reagent tablets have 
provided useful information over the years, their use is now being challenged 
by the convenient and sophisticated paper strip tests and rapid 
tests utilizing biotechnology. 
Tablets fulfill countless other needs (e. g ., Halizone for water purlfication, 
artificial sweeteners, nutritional ingredients for diet control, cleaners 
for dentures, general cleaners and disinfectants, fertilizers for house 
plants, and even Easter egg colors. These represent but a few examples 
of the extension of tablet utilization into nonpharmaceuticaI areas. 
1. M. Gibaldi and J. L. Kanig, J. Oral Ther. Pharmacol.. 1: 440 (1965). 
2. M. A. Hussain, B. J. Aungst, A. Kearney, and E. Shefter, Pharm. 
Res., 3(Suppl), 97S (1986). 
3. H. Choi and V. H. L. Lee, Pharm. Res., 3(Suppl), 70S (1986). 
4. D. Gallardo and V. H. L. Lee, Pharm. Res., 3(Suppl), 73S (1986). 
5. A. H. Beckett, R. N. Boyes, and E. J. Triggs, J. Pharm. Pharm acol. , 
20; 92 (1968). 
6. R. P. Walton and C. F. Lacey, J. Pharmacol. Exp. Ther., 54:61 
7. R. P. Walton, Proc. Soc. Exp. BioI. Med., 32:1486 (1935). 
8. R. P. Walton, Proc. Soc. Exp. Biol. Med., 32:1488 (1935). 
9. R. P. Walton, J. Am. Med. Assn . 124:138 (1944). 
10. M. D. D. Bell et al., Lancet, Vol. I, 71 (1985). 
11. A. H. Beckett and E. J. Triggs, J. Pharm. Pharmacol., 19(5uppl), 
31S (1967). 
12. A. H. Beckett and A. C. Moffat, J. Ptuirm, Pharmacol., 20(Suppl), 
239S (1968). 
13. L. S. Shanker, J. Med. Pharm. Chern., 2;343 (1960). 
14. W. A. Ritschel, J. Pharm. Sci.. 60: 1683 (1971). 
15. R. Anders, H. P. Merkle, W. Schurr, and R. Ziegler, J. Pharm. 
Sci., 72:1481 (1983). 
16. N. F. H. Ho and W. 1. Higuchi, J. Pharm. Sci  60:537 (1971). 
17. K. R. M. Vora, W. 1. Higuchi, and N. F. H. Ho, J. Pharrn. sct., 
61: 1785 (1972). 
18. Am. Druggist, Feb. 1987, p. 19. 
19. M. H. Litchfield, J. Pharm. sa., 60:1599 (1971). 
20. R. M. Fuller, New Remedies, 7: 69 (1878). Republished from Medical 
Record, 13: 185 (1878).

Special Tablets 365 
21. R. L. Brunton, The Gaulsonian Lectures Delivered before the Royal 
College of Surgeons, Macmillan, London, 1877. 
22. M. D. Richmond, C. D. Fox, and R. F. Shangraw, J. Ptuirm, sci., 
54:447 (1965). 
23. I. A. Chaudry and R. E. King, J. Pharm. ScL, 61:1121 (1972). 
24. F. W. Goodhart, H. Gucluyildiz, R. E. Daly, L. Chefetz, and F. C. 
Ninger, J. Pharm. Sci., 65:1466 (1976). 
25. J. W. Warren and J. C. Price, J. Pharm. Sci., 66:1406 (1977). 
26. J. W. Warren and J. C. Price, J. Pharm. ScL, 66:1409 (1977). 
27. United States Pharmacopeia XXI, Mack PUblishing, Easton, PA 1984. 
28. C. A. Gaglia, Jr., J. J. Lomner, B. L. Leonard, and L. Chefetz, 
J. Pharm. Sci., 65:1691 (1976). 
29. B. Dorsch and R. Shangraw, Am. J. Hosp. Pharm., 32:795 (1975). 
30. M. J. Pikal, A. L. Lukes, and L. F. Ellis, J. Pharm. sa., 65: 1278 
(1976) . 
31. S. A. Fusarf , J. Pharm. sci., 62:2012 (1973). 
32. M. L. Broderick, Eli Lilly 81 Co., unpublished data. 
33. S. A. Fusari, J. Pharm, Sci . 62:122 (1973). 
34. D. Banes, J. Pharm. sa., 57: 893 (1968). 
35. B. A. Edelman, A. M. Contractor, and R. F. Shangraw, J. Am. 
Pharm. Assn., NS11, 30 (1971). 
36. D. P. Page, N. A. Carson, C. A. Buhr, P. E. Flinn, C. E. Wells, 
and M. T. Randall, J. Pharm. Sci., 64: 140 (1975). 
37. M. J. Pikal, D. A. Bibler, and B. Rutherford, J. Ptuirm, Sci., 66: 
1293 (1977). 
38. M. J. Pikal, Eli Lilly 81 Co., unpublished data. 
39. D. Stephenson and J. F. Humphreys-Jones, J. Pharm. Pharmacol., 
3:767 (1951). 
40. P. Suphajettra, J. Strohl, and J. Lim, J. Pharm. ScL, 67;1394 
(1978) . 
41. M. J. Pikal, A. L. Lukes, and J. W. Conine, J. Pharm. Sci. 73: 
1608 (1984). 
42. M. J. Pika! and A. L. Lukes, J. Pharm. sct., 65: 1269 (1976). 
43. S. Shah, Eli Lilly 81 Co. I unpublished results. 
44. H. Gucluyildiz, F. W. Goodhart, and F. C. Ninger, J. Ptiarm, Sci., 
66:265 (1977). 
45. B. Greer, Am. J. Hosp. Pharm., 32; 979 (1975). 
46. J. Zeitz and N. Schwartz, Am. J. Hosp. Pharm., 33; 209 (1976). 
47. R. M. Gabrielson, Am. J. Hosp. Pharm., 33:209 (1976). 
48. J. Pitha, S. M. Harmon, and M. E. Michel, J. Pharm. ScL, 75:165 
(1986) . 
49. H. Lowery, U.S. Patent 3,870,790 (1975). 
50. H. Lowery, U.S. Patent 4,259,314 (1981). 
51. J. M. Schor, U.S. Patent 4,357,469 (1982). 
52. J. M. Schor, A. Nigalaye, and N. G. Gaylord, U.S. Patent 
4,389,393 (1983). 
53. S. S. Davis, J. W. Kennerley, M. J. Taylor, J. G. Hardy, and 
C. G. Wilson, Int. Congr. Symp. Ser.-R. Soc. Med. 1983, 54(Mod. 
Concepts Nitrate Delivery Syst.) 29- 37. Through C. A. 99: 110574s 
54. Y. Suzuki, H. Ikura, and G. Yamashita, U.S. Patent 4,292,299 
55. G. L. Christenson and H. E. HUber, U.S. Patent 3,594,467 (1971).

366 Conine and Pikal 
56. W. Tanaka, A. K. Yoshida, T. Terada, and H. Ninomiya , U.S. 
Patent 4,059,686 (1977). Through Transdermal and Related Delivery 
Systems, (D. A. Jones, edv) , Noyes Data Corp., Park Ridge, NJ, 
57. K. Sugden, Br. Patent Appl. GB 2,165,451. Through C.A. 105: 
30103q (1986). 
58. G. L. Christenson and H. E. Huber, U.S. Patent 3,590,117 (1971). 
59. J. D. Smart, J. W. Kellaway, and H. E. C. Worthington, J. Pharm. 
Pharmacal., 36:295 (1984). 
60. M. B. Kabadi and Y. W. Chien, J. Pharm. ScL, 73:1464 (1984). 
61. S. Hwang, E. Owada , L. Suhardja, N. F. H. Ho, and G. L. Flynn, 
J. Pharm. Sci., 66:781 (1977). 
62. L. G. Patel, S. J. Warrington, and R. N. Pearson, Br. Med. J., 
287: 1247 (1983). 
63. Y. Uda, S. Hirai, and T. Yashiki , U.S. Patent 4,670,419 (1987). 
64. M. Yamaguchi and K. Tanno, Yakauzaigaku, 45: 41 (1985). Through 
C.A., 103:59192n (1985). 
65. J. M. Schor, U.S. Patent 4,226,849 (1980). 
66. C. J. deBlaey and J. Polderman, in Drug Design, Vol. 9 (E. J. 
Ariens, ed.), Academic Press, New York, 1980. 
67. J. G. Wagner, Btopharmaceuttcs and Relevant Pharmacokinetics, 
Drug Intelligence Publications, Hamilton, ILL, 1971- 
68. I. Smazynski and L. Krowizynski, A cta Pol. Pharm.. 25: 291 (1968). 
Through C.A., 70: 31654r (1969).

Chewable Tablets 
Robert W. Mendes Massachusetts College of Pharmacy and Allied Health 
Sciences. Boston, Massachusetts 
Aloysius O. Anaebonam Fisons Corporation. Rochester. New York 
Jahan B. Daruwala E. 1. du Pont de Nemours & Company, Inc . 
Wilmington, Delaware 
Chewable dosage forms, such as soft pills, tablets, gums, and, most recently, 
"chewy squares," have long been part of the pharmacist's armamentarium. 
Their possible advantages, compared to solid dosage forms intended 
to be SWallowed, include better bioavailability through bypassing 
disintegration (and perhaps enhancing dissolution), patient convenience 
through the elimination of the need for water for SWallowing, possible use 
as a substitute for liquid dosage forms where rapid onset of action is 
needed, improved patient acceptance (especially in pediatrics) through 
pleasant taste, and product distinctiveness from a marketing perspective. 
There are, of course, limitations to the use of chewable tablets. Badtasting 
drugs and those having extremely high dosage levels present the 
formulator with significant obstacles to be overcome. These will be discussed 
in considerable detail throughout this chapter. 
Chewable tablets represent the largest market segment of the chewable 
dosage forms, with chewing gums and the new chewy squares accounting 
for a much smaller percentage. Of the chewable market. from a therapeutic 
perspective, antacids account for the largest segment, with pediatric 
vitamins next. For both physiological and psychological reasons, children 
up to the young teens usually have trouble swallowing tablets and capsules: 
often. this problem continues into adulthood. As a result, most products 
for children are formulated as liquids. A notable exception to this is represented 
by the over-the-counter (OTC) and prescription (Rx) vitamin 
products. For these, chewable tablets are preferable because of their 
patient acceptability and better stability. Additionally. several OTC cough/ 
cold products and some analgesics are alternatively available as chewables. 
Formulation considerations of importance primarily revolve around taste; 
children tend to be particularly sensitive in their preferences for various 
flavors. sweetness levels, etc. 

368 Mendes et al. 
The swallowing problems associated with the very young may also be 
assumed to exist among the elderly. Despite this, the authors have been 
unable to identify a single drug product formulated as a chewable and 
specifically targeted to this population. 
Formulation considerations here would be similar. except that the preference 
would be different: less sweet, less flavor emphasis, etc. 
Various factors involved in the formulation of a chewable tablet can be 
schematically represented as shown in Figure 1. The first four formulation 
factors shown in the schematic diagram are common to regular (swallowed) 
and chewable tablets; however, the organoleptic properties of the 
active drug substance (or substances) are of primary concern here. A 
formulator may use one or more approaches to arrive at a combination of 
formula and process that results in a product with good organoleptic properties. 
Such a product must have acceptable flow, compressibility, and 
stability characteristics. Generally as the required amount of active substance 
per tablet gets smaller and less bad tasting, the task of arriving at 
an acceptable formulation becomes easier due to the fact that a greater 
number of formulation options are available. Conversely, extremely badtasting 
and lor high-dose drugs are difficult to formulate into chewable 
The factors of flow, lubrication, disintegration, compressibility, and 
compatibility-stability have been described in depth in Chapters 1 through 
4. The organoleptic considerations will be elaborated here. 
A. Taste and Flavor 
Physiologically, taste is a sensory response resulting from a chemical stimulation 
of the taste buds on the tongue. There are four basic types of 
tastes; salty, sour, sweet, and bitter. Salty or sour tastes are derived 
from substances capable of ionizing in solution [1]. Many organic medicinal 
compounds stimulate a bitter response even though they may not be 
capable of ionizing in an aqueous medium. Most saccharides, disaccharides, 
some aldehydes, and a few alcohols give a sweet taste. Substances incapable 
of prod ucing a sensory stimulation of the buds are referred to as 
bland or tasteless. 
The term [usvor generally refers to a specific combined sensation of 
taste and smell (olfaction). For example, sugar has a sweet taste but no 
flavor whereas honey has a sweet taste and a characteristic smell-the 
combination of the two being known as honey flavor. 
B. Aroma 
Pleasant smells are generally referred to as aromas. For example, a wellform 
ulated , or ange- flavored, chewable tablet should have a characteristic 
sweet and sour taste and an aroma of fresh orange.

Amount of active substances as a 
percent of total tablet weight 
Compatability  Stability 
Organoleptic Considerations 
'" 1 
Cold remedies 
Good taste and mouthfeel 
Acceptable bioavailability and " bioactivity , 
Acceptable stability and quality 
Economical formula and process 
Taste panels 
Blood levels (for adsorbed drugs) 
In vitro vs. in vivo correlation 
for antacids 
Stability (chemical, physival, 
Quality control and assurance 
""'3 g. 
Molecular complexes 
Formation of salts or derivatives 
Artificial sweeteners 
Solid dispersions 
Ion exchange 
Spray congealing and coating 
Granulation and Coating 
Use of amino acid and protein hydrolysates 
Inclusion complexes 
Figure 1 Flow chart of various aspects to be considered in connection with chewable tablets. 

370 Mendes et al. 
C. Mouth-feel 
The term mouth-feel is related to the type of sensation or touch that a 
tablet produces in the mouth upon chewing. As such, it has nothing to 
do with chemical stimulation of olfactory nerves or taste buds. However, 
for a formulation to be successful, the overall effect in the mouth is important. 
In general, gritty (e. g., calcium carbonate) or gummy textures are 
undesirable, whereas a soothing and cooling sensation (e. g. , mannitol) 
with smooth texture is preferred. 
D. After Effects 
The most common after effect of many compounds is aftertaste. For 
example, some iron salts leave a "rusty" aftertaste; saccharin in high 
amounts tends to leave a bitter aftertaste. 
Another common after effect is a numbing sensation of a portion of the 
whole surface of the tongue and mouth. Bitter antihistamines such as pyribenzamine 
hydrochloride and promethazine hydrochloride are typical of this 
class of drugs. 
E. Assessment of the Formulation Problems 
Wherever feasible and practical, the first step in the formulation of a 
chewable tablet is to obtain a complete profile of the active drug. This 
usually leads to the most efficient formulation of a stable and quality product 
as the drug usually dictates the choice of fillers. carriers, sweeteners. 
flavor compounds, and other product modifiers. 
The drug profile ideally should contain information on the following: 
Physical properties 
Taste, aftertaste, and mouth- feel 
Physical form: crystal, powder, amorphous solid, oily liquid, etc. 
Melting or congealing temperature 
Existence of polymer-phs 
Moisture content 
Aqueous solubility 
Active drug's stability on its own 
Compressibility if applicable 
Chemical properties 
Chemical structure and chemical class 
Major reactions of this chemical class 
Major incompatible compounds or class of compounds 
Drug dose and any limit on final dosage size 
Any other relevant information 
This active drug profile Would eliminate potentially incompatible excipients , 
flavors. and the like at the outset. leading to the use of excipients that 
would best compliment the drug chemically. physically. and organoleptically. 
The choice of excipients and other product modifiers would involve judgment. 
balancing their cost with their functionality. The use of bw- calorie

Chewable Tablets 371 
and non-sugar-based excipients may represent a marketing advantage, 
especially with consumers concerned about calorie intake and dental caries. 
Almost invariably, the formulation problem involves at least one of the 
following: undesirable taste, bad mouth- feel, or aftertaste. The desired 
product should prevent or minimize stimulation of the taste buds, contain 
a suitable flavor and sweetener, and achieve good mouth- feel and compressibility. 
The following techniques are used to solve one or more of the 
A. Coating by Wet Granulation 
Wet granulation , which is discussed in detail in Chapter 3, historically has 
been the method of choice for preparing drugs for compression. 
This process may be described as one which agglomerates drug particles 
through a combination of adhesion and cohesion using a wetting agent 
and binder. Generally, binders are classified as hydrophilic gums (e. g. , 
acacia), sugars (e. g., sucrose), starches (e, g., natural or modified corn), 
and polymers (e. g., povidone, cellulose derivatives, gelatin), which have 
the property of becoming sticky when wetted with water or another suitable 
solvent. This method consists of the mixing of the ingredients in a solidsliquids 
processor to form a dampened, agglomerated mass that may then be 
subdivided, dried, and sized to form a suitable tree- flowing and compressible 
Although this process is primarily intended to impart flowability and 
compressibility to impalpable substances, under certain conditions it may 
be useful in the application of coatings to drug particles in order to mask 
or reduce their taste. Example 10 illustrates the use of ethylcellulose (a 
water-insoluble polymer) to coat ascorbic acid through wet granulation to 
improve its stability and assist in taste masking. In this case, the drug 
is wet-granulated with an anhydrous solution of polymer in a planetary 
processor, dried, sized, and blended with a directly compressible sweetener 
and other ingredients to produce a material suitable for compressing. 
In general, this is the simplest approach to taste masking. Wet granulation 
may be accomplished as described above with or without the inclusion 
of additional excipients such as lactose, sucrose, mannitol, sorbitol, other 
sugars, or starches. Although this approach is similar to that for the wet 
granulation of nonchewable tablets, some fundamental concepts should be 
kept in mind. Whenever possible, the granulating/coating agent should 
form a flexible rather than brittle film, have no unpleasant taste or odor 
of its own, be insoluble in saliva but not interfere with drug dissolution 
after swallowing-, Ideally, sweet fillers, such as sugars, should be included 
in the granulation. Disintegrant should preferably be included in the wet 
granulation to ensure proper dissolution of the granules after chewing. 
While the procedure described above involves the use of classical wet 
granulation processing, the desirability of utilizing more modern techniques 
should not be overlooked. The use of flUidized bed or air suspension 
coating may represent a more efficient approach. 
In this technique, drug particles to be coated are fluidized by means 
of suspension in a controlled, high-velocity. warm air stream directed

372 Mendes et al, 
through a perforated plate into a coating chamber. The drug particles 
undergo cyclic flow past an atomizing nozzle delivering coating agent in 
solution or suspension. The sprayer may be mounted either to spray upward 
from the bottom (Wurster style) or downward from the top, as depicted 
in Figure 2. As the particles become coated. they are removed 
from the spray field, dried by the warm air stream, and returned for recoating. 
This cycling continues until the desired coating thickness has 
been achieved. Fluidization of the drug particles provides increased Surface 
exposure for more efficient and uniform coating and drying. Since 
evaporation occurs over the entire surface in a very short time, particle 
temperature does not increase. This permits the coating of heat-labile 
drugs without concern for degradation. Figure 3 is a comparative electron 
micrograph showing acetaminophen particles before and after air suspension 
Although it is not within the scope of this discussion to detail the optimization 
of the process, the formulator should recognize the importance of 
factors such as par-ticulate properties of the drug. viscosity of the coating 
liquid. design and placement of the spray nozzle. and velocity and temperature 
of the fluidizing air. 
Although taste improvement by coating is attractive in its simplicity, it 
should be understood that this method may only suffice for mildly to 
moderately unpleasant tasting drugs. For those that are extremely bitter, 
sour, or otherwise difficult, more heroic methods will most certainly be 
B. Microencapsulation 
Microencapsulation is a method a f coating drug particles or liquid droplets 
with edible polymeric materials. thereby masking the taste and forming 
relatively free- flowing microcapsules of 5- to 5000-11m size [2- 4] . A number 
of methods have been described in the literature [5,6]. but phase 
separation or coacervation technique appears to be more relevant and suitable 
for taste-masking applications. The process essentially consists of 
three steps [5]; 
1. Formation of three immiscible phases: a liquid-manufacturing 
vehicle phase, a core material drug phase. and a coating material 
2. Depositing the liquid polymer coating by sorption around the core 
material under controlled physical mixing of the three phases 
3. Rigidizing the coating. usually by thermal crosslinking or desolvation 
techniques, to form a rigid microcapsule 
The resultant coated granules not only mask the taste of a drug but also 
minimize any physical and chemical incompatibility between ingredients [2]. 
The size distribution of the drug can be narrow or relatively broad, and 
the process is applicable to a variety of compounds regardless of pharmacological 
classification. Typical coating materials include carboxymethylcellulose, 
cellulose acetate phthalate, ethylcellulose, gelatin, polyvinyl 
alcohol. gelatin-acacia, shellac, and some waxes-with the choice depending 
on the specific application. The encapsulated drug is isolated from the 
liquid-manufacturing vehicle as a free-flowing powder. In general. the 
encapsulated drug is then blended with direct-compression vehicles

Chewable Tablets 
Product Container 
Air Diffuser Plate 
t Air Inlet Plenum 
Coatin Partition 
S ra Nozzle 
Air Distribution Plate. 
Air Inlet Plenum 
Figure 2 (a) Top spray fluidized bed system. (b) Bottom spray 
fluidized system. (Courtesy Nortec Development Associates, Inc.)

374 Mendes et aZ. 
Figure 3 (a) Uncoated acetsminophan, (b) Coated (taste-masked) 
acetaminophen. (Courtesy Nortec Development Associates, Inc.)

Chewable Tablets 375 
(described later in this chapter), other diIuents, artificial sweeteners, 
flavors, and lubricants for tableting. 
A typical taste-masking application of microencapsulation has been 
described by Bakan and Sloan [2] for acetaminophen tablets (Example 1). 
It must be borne in mind that, upon compression, the structure of the 
coating is disrupted and should be expected to lose some of its protective 
barrier. Furthermore, the extent of mastication and the length of time 
that a drug remains in the mouth also play an important role in determining 
the extent of taste masking, especially for very bitter drugs. Appropriate 
compensatory measures, such as choosing the right coating material 
and the extent of the applied coat, along with the solubility and particle 
size considerations, must be taken into account before arriving at an acceptable 
microencapsulated form suitable for further blending with excipients 
and subsequent tableting. As discussed earlier, the mouth- feel characteristics 
of a chewable tablet formulation are important; it is noteworthy 
that microcapsules larger than 60 mesh are unsatisfactory [2], and that 
smaller particle sizes (about 100 to 120 mesh) should give a good mouthfeel, 
and at the same time have adequate flow for uniform blending and 
compression. Eliminating or minimizing a potential incompatibility is also 
an inherent advantage with this technique. 
Farhadieh [4] has used coagulable water- soluble egg albumin as the 
coating medium for masking the taste of erythromycin derivatives. The 
process involves suspending the drug particles in an aqueous solution of 
egg albumin at pH 7 to 10, and stirring the suspension with a liquid alkane 
containing a surfactant to form an emulsion. The emulsion is then heated 
to 50 to 80C with stirring, so as to coagulate the albumin in the form of 
Example 1: Acetaminophen Tablets (Microencapsulated, 
Quantity per 
I ngredient tablet 
Microcapsules ('iJ100 mesh) 
Acetami nop hen 
Coating (cellulose-wax) 
Mannitol (major diluent) 
Microcrystalline cellulose (Avicel) 
Guar gum 
Mint, spice, and peppermint flavors 
Magnesium stearate 
327 mg 
35 mg 
393 mg 
755 mg

376 Mendes et a1. 
microcapsules around the drug. The solid microcapsules are separated 
from the suspension and dried. This can be made into chewable tablets 
by dilution with direct-compression grade mannitol, flavor, artificial 
sweeteners, and lubricant and subsequent compression. Heat stability 
(physical and chemical) of the tablets made with denatured protein should 
be carefully evaluated along with flavor retention characteristics of the 
formulation over a prolonged period of time before considering the formulation 
as acceptable. 
The above discussion points out certain obvious advantages of the 
process, such as considerable flexibility in the choice of coating materials, 
particle size, and minimization of incompatibilities'. However, it should be 
pointed out that compared to the conventional wet granulation process, 
microencapsulation is more expensive and does require specialized equipment 
and knowledgeable personnel. The economics of the process and its 
practicality for a given drug should be expected to improve as the required 
dose per tablet decreases. Limitations imposed due to extensive patent 
coverage should also be taken into account. 
C. Solid Dispersions 
Bad-tasting drugs can be prevented from stimulating the taste buds by 
adsorption onto substrates capable of keeping the drugs adsorbed while in 
the mouth but releasing them eventually in the stomach or gastrointestinal 
tract. A good example of such an application is the adsorption of dextromethorphan 
hydrobromide onto magnesium trisilicate substrate (Example 2) 
[7]. The adsorbate is commercially available in the form of micronized 
powder with a drug content of 10% wIw. It must be noted that besides 
the dextromethorphan portion of the compound, the bromide ion contributes 
significantly to the undesirable taste of the drug. This is indeed true of 
many other bromide salts of medicinal compounds. A formulator attempting 
to formulate an adsorbate may consider many substrates such as bentonite, 
Veegum, and silica gel. 
D. Adsorbate Formation Techniques 
Solvent Method 
Generally the formation of an adsorbate involves dissolving the drug in a 
solvent I mixing the solution with the substrate I and evaporating the solvent-
leaving the drug molecules adsorbed upon the substrate. The variables 
of the process, such as choice of solvent, substrate, proportions, 
mixing conditions, rate of evaporation, and temperature, must be optimized 
to give the desired product. 
Melting Method 
Here the drug or drugs and a carrier are melted together by heating. 
The melted mixture is then cooled and rapidly solidified in an ice bath 
with vigorous stirring. The product is then pulverized and sized. Heatlabile 
drugs, volatile drugs, and drugs that decompose on melting are 
obviously unsuitable for this method. The method is simple with low cost 
and no problem of residual solvents as are encountered in the solvent 
evaporation method.

Chewable Tablets 
Example 2; Cough Preparation (Chewable) 
Quantity per 
Ingredient tablet 
10% Dextromethorphan HBr adsorbate (micronized) 
Mouthwash flavor (Givaudan FS098) 
Magnesium stearate 
Sorbitol (crystalline) 
76. S mg* 
2. S mg 
10.0 mg 
10.0 mg 
1301. 0 mg 
1400.0 mg 
*Contains 7. S mg active drug plus 2% excess. 
Pass the sorbitol through a 10-mesh screen to deagglomerate the particles. 
Premix the adsorbate, benzocaine, and flavor with one-fourth of 
the required amount of sorbitol for 10 min. Add the rest of the sorbitol 
and mix for another 10 min. Finally, add the magnesium stearate and 
mix for a further 3 min. Compress to a hardness of about 6 kp using 
S/8-in. diameter tooling. 
Comments: A noteworthy feature of the above example is that the total 
drug content (dextromethorphan HBr and benzocaine) of the tablet is 
about 10 mg, which requires a tablet of 1,400 mg weight to give satisfactory 
taste-masking and mouth-feel characteristics. It must also be 
pointed out that the compression hardness of 6 kp gives the tablet 
chewable characteristics. Furthermore, the required S/8-in. tooling 
would put some restriction on the tablet press that could be used for 
the product. 
Chiou and Riegelman [8] reported a third method of preparing solid 
dispersions with limited application. This method is a combination of certain 
aspects of the solvent and melting methods. 
E. Ion Exchange 
Ion exchange has been defined by Wheaton and Seamster [9] as the reversible 
interchange of ions between a solid and a liquid phase in which there 
is no permanent change in the structure of the solid. The solid is the 
ion exchange material while the ion could be a drug. When used as a drug 
carrier. ion exchange materials provide a means for binding drugs onto an 
insoluble polymeric matrix and can effectively mask the problems of taste 
and odor. in drugs to be formulated into chewable tablets. 
Ion exchange resins can be classified in four major groups: strong 
acid cation. weak acid cation, strong base anion. and weak base anion 
exchange resins. 
(1) Strong acid cation exchange resins are best exemplified by the 
principal sulfonated styrene-divinylbenzene copolymer products such as 
Amberlite IRP- 69 (Rhorn and Haas) and DOWEX MSC-l (Dow Chemical).

378 Mendes et al. 
These resins can be used for masking the taste and odor of cationic 
(amine-containing) drugs prior to their formulation into chewable tablets. 
These resins are spherical products prepared by the sulfonation of styrenedivinylbenzene 
copolymer beads with the SUlfonating agent of choice: sulfuric 
acid. chlorosulfonic acid I or sulfur trioxide. The use of a nonreactive 
swelling agent is generally required for rapid and uniform swelling 
with minimum breakage. 
CH=C~ 6
sulfonating acid 
swelling agent 
Strong-acid resin 
Strong acid cation exchange resins function throughout the entire pH 
range. A schematic of a strong acid cation exchange resin in use follows: 
- + + - + + 
Resin (SO 3) A + D ------ resin (SO 3) D + A 
Cation exchange resin + drug - resin-drug complex + displaced ion 
(2) The most common weak acid cation exchange resins are those prepared 
by crosslinking an unsaturated carboxylic acid such as methacrylic 
acid with a crosslinking agent such as divinylbenzene. 
-C-O./ -CH-CH - I'~ 2 
-CH-CH2 - 
CH =CH2 0---"" CH =CH2 
C: CH2 + 
methacrylic acid 
Weak acid resin 
Examples include DOWEX CCR- 2 (Dow Chemical) and Amberlite IRP- 65 (Rhom 
and Haas). Weak acid cation exchange resins function at pH values above 6.

Chewable Tablets 379 
(3) Strong base anion exchange resins are quaternized amine resins 
resulting from the reaction of triethylamine with chloromethylated copolymer 
of styrene and divinylbenzene. Examples include Amberlite IRP- 276 (Rhom 
and Haas) and DOWEX MSA-A (Dow Chemical). These strong base anion 
exchange resins function throughout the entire pH range. 
(4) Weak base anion exchange resins are formed by reacting primary 
and secondary amines or ammonia with chloromethylated copolymer of styrene 
and divinylbenzene. Dimethylamine is usually used. These weak 
base anion exchange resins function well below pH 7. 
Important properties to be considered when using an ion exchange 
resin include particle size, shape, density, porosity, chemical and physical 
stability, and ionic capacity. 
The rate and extent of drug desorption from these resins in vivo will 
be controlled by the diffusion rate of the drug through the polymer phase 
of the resin, as well as the selectivity coefficient between the drug and 
the resin. 
Ion exchange resins, especially weakly acidic cation exchange resins, 
have certain adsorptive mechanisms that have been utilized in the stabilization 
of the nonionic vitamin B12 (cyanocobalamin) for many years [10]. 
A formulator must thoroughly investigate the various available types of 
pharmaceutical grade resins available for specific applications and check 
their approval status for oral use in the amounts anticipated. The quantity 
of resin required per unit quantity of drug to achieve effective taste 
masking and lor stability improvement is a limiting factor as the dose of 
drug per tablet increases. 
F. Spray Congealing and Spray Coating 
In a broad sense, the process of spray congealing involves cooling (or congealing) 
of melted substances in the form of fine particles during their 
travel from a spray nozzle to the distant vicinity of a spraying chamber 
held at a temperature below their melting point. If a slurry of drug 
material insoluble in a melted mass is spray-congealed, one obtains discrete 
particles of the insoluble material coated with the congealed substance. 
The application is best exemplified by the taste masking of thiamine 
mononitrate 1 riboflavin. pyridoxine hydrochloride. and niacinamide by 
fatty acids or monoglycerides and diglycerides of edible fatty acids. These 
are commonly available (as Rocoat vitamins) in the form of relatively freeflowing 
powders having the composition shown in Table l. 
It must be noted that the weight of the active substance is approximately 
one-third that of the spray-congealed preparation. For small-dose 
entities, such as the vitamins, spray congealing is ideally suited. The influence 
of the coating on the bioavailability of the drug must be considered 
before considering this method as the means of improving the taste of the 
drug. Polyethylene glycols (Carbowaxes) of molecular weights between 
4000 and 20,000 are suitable for spray congealing especially where their 
solubility would represent an added advantage. 
As opposed to spray congealing. the spray-coating process involves 
the spraying of a suspension of the drug particles in a solution of the 
coating material through an atomizer into a high-velocity stream of warm 
air. The coarse droplets delivered by the atomizer consist of drug

Examples of Coated Vitamins
Coating agent 
Mendes et aI. 
Vitamin content 
(% wIw) 
Thiamine mononitrate (B 
Riboflavin (B 2) 
Pyridoxine HCI (B 
Mono- and diglycerides 
of edible fatty acids 
Mono- and diglycerides 
of edible fatty acids 
Mono- and diglycerides 
of edible fatty acids 
Stearic acid 
33. 3 
particles enveloped by coating solution. As the solvent evaporates, the 
coating material encapsulates the drug particle. 
The atomizers typically used in such a system may be of the pneumatic 
type in which atomization is accomplished through pressurization through 
an orifice) or the rotating disk which functions through centrifugal force. 
Since drying is nearly instantaneous, the drug particle is subjected to 
little temperature increase, making this process suitable for heat-labile 
drugs. As with all spray-drying techniques, concentration, viscosity, 
spray rate, temperature, and velocity are factors that require optimization. 
Examples of applications of the method include the spray coating of flavor 
oils and the coating of sodium dicloxacillin [11] and vitamins A and D. 
The coating of the antibiotic sodium dicloxacillin, or some other tetracyclines, 
involves a mixture of ethylcellulose and spermaceti wax (as coating 
materials) dissolved in methylene chloride. A suspension of micronized 
antibiotic in this solution, upon spray drying, results in a free- flowing 
product suitable for further compounding into a chewable formulation, [It 
should be noted that in July 1987 the United States Consumer Product 
Safety Commission (CPSC) rejected a proposed rule that would have declared 
methylene chloride a hazardous substance in consumer products. 
Instead, the commission voted to require chronic hazard warning labels on 
consumer products containing more than 1% methylene chloride. This 
points up the significant interest among various regulatory agencies with 
regard to the use of organic solvents. It is encumbant on the formulator 
to ascertain the regulatory status of solvents with respect to their use in 
pharmaceutical products and processes, as well as possible restrictions due 
to environmental concerns.] 
Vitamins A and D are fat-soluble and, as such, are unsuitable (too 
oily) for incorporation into chewable formulations, primarily because of 
their physical form and poor stability due to oxidation. Two commercially 
available, protectively coated forms of vitamin A or vitamins A and D 
together exemplify the application of the technology under discussion. 
Crystalets: Contain vitamin A acetate) available with or without 
vitamin D2 as fine, free- flowing particles in a matrix of gelatin, 
sugar, and cottonseed oil, stabilized with BHT, BHA, and sodium 

Chewable Tablets 381 
Beadle ts: Contain vitamin A acetate and vitamin D2 beadlets as fine, 
free-flowing particles in a gelatin matrix with sugar and modified 
food starch, stabilized with BHT, BHA, methylparaben, propylparaben. 
and potassium sorbate . 
It should be noted that the coating method for antibiotics described above 
was primarily for masking the taste, whereas vitamins A and D are essentially 
tasteless and are spray-dried for reasons of stability and ease of 
G. Formation of Different Salts or Derivatives 
This approach differs from the others previously discussed in that an 
attempt is made to modify the chemical composition of the drug substance 
itself, so as to render it less soluble in saliva and thereby less atirnulating 
for the taste buds, or to obtain a tasteless or less bitter form. Even if 
one is successful in preparing a new salt or a derivative of a bitter drug, 
the legalities of its new drug status from a regulatory point of view must 
be considered. Moreover. the solubility. stability, compatibility. and bioavailability 
aspects of the "new" compound must also be kept in mind. If 
a less bitter tasting salt form or a tasteless derivative can be obtained. 
this would represent the best approach to taste masking. Since there is 
no coating that can be broken during chewing, no problem will be encountered 
with respect to unpleasant aftertaste. 
H. Use of Amino Acids and Protein Hydrolysates 
By combining amino acids, their salts, or a mixture of the two [12], it is 
possible to substantially reduce the bitter taste of penicillin. Some of the 
preferred amino acids are sarcosine, alanine. taurine. glutamic acid. and 
especially glycine. The taste of ampicillin is markedly improved by granulating 
with glycine in the usual manner and subsequently blending this 
mixture with additional glycine, starch. lubricants, glidants, sweeteners, 
and flavors before compression. 
I. Inclusion Complexes 
In inclusion complex formation, the drug molecule (guest molecule) fits into 
the cavity of a complexing agent (host molecule) forming a stable complex. 
The complex is capable of masking the bitter taste of the drug by both 
decreasing the amount of drug particles exposed to the taste buds and lor 
by decreasing the drug solubility on ingestion, both activities leading to a 
decrease in the obtained bitterness associated with the drug. The forces 
involved in inclusion complexes are usually of the Vander Waals type. and 
one of the most widely used complexing agents in inclusion type complexes 
is e-cyclodextrin, a sweet, nontoxic, cyclic oligosaccharide obtained from 
Three primary methods have been reported for the preparation of 
cyclodextrin inclusion compounds [13]. Two of these are laboratory scale, 
while the other is industrial scale.

382 Mendes et al. 
Laboratory Methods 
(1) Equimolar quantities, or a lO-fold excess of water-soluble substances, 
are dissolved directly in concentrated hot or cold aqueous solutions 
of the cyclodextrins. The inclusion compounds crystallize out immediately 
or upon slow cooling and evaporation. 
(2) Water-insoluble drugs are dissolved in a non-water-miscible organic 
solvent and shaken with a concentrated aqueous solution of cyclodextrins. 
The inclusion compounds crystallize at the interface between the layers, or 
as a precipitate. The crystals must then be washed with solvent to remove 
uncomplexed drug and dried under appropriate conditions to remove residual 
Industrial Method 
The drug substance is added to the cyclodextrin and water to form a 
slurry which undergoes an increase in viscosity with continued mixing. 
This may concentrate to a paste that can be dried, powdered, and washed. 
If the inclusion compounds are readily soluble in water, or decompose on 
drying, it may be advisable to use lyophilization to accomplish drying. 
This may provide the additional advantages of easy redispersibility and 
improved dissolution rate. 
Inclusion-type complexes can also increase the stability of the guest 
molecule by shielding it from moisture, oxygen, and light, which can degrade 
the drug molecule via hydrolysis, oxidation, and photodegradation, 
J. Molecular Complexes 
Molecular complex formation involves a drug and a complexing organic molecule 
and, like inclusion complexes, can be used in the masking of the 
bitter taste or odor of drugs by forming complexes that would lower the 
aqueous solubility of the drug and thus the amount of drug in contact 
with the taste buds. 
Higuchi and Pitman [14J reported the formation of a molecular complex 
between caffeine and gentisic acid leading to a decrease in caffeine solubility. 
One would consequently expect a decrease in the bitter taste of 
caffeine if the above complex were used in a chewable caffeine tablet 
form ulation. 
The SUbject of tablet excipients in general has been extensively covered in 
Chapters 3 and 4. Special consideration, however, needs to be given to 
those materials that form the basis for chewable tablet formulation. The 
acceptability in the marketplace of chewable tablets will be primarily determined 
by taste and, to a lesser degree, appearance. Therefore, appropriate 
selection and use of components that impact on these properties are 
of extreme importance. Of course, the formulator must not become so concerned 
with these properties as to lose sight of other pharmaceutical and 
biomedical considerations; the resultant product must be as pure, safe, 
efficacious, and stable as any other.

Chewable Tablets 383 
The processes described in Chapters 3 and 4 (wet granulation, dry 
granulation, direct compaction) are as applicable to chewables as to any 
other type of tablet. The concerns such as moisture content and uptake, 
particle size distribution, blending and loading potentials, flow and compressibility 
are no less important, and must be addressed by the formulation/
process development pharmacist as for any product. However, in the 
case of chew ables , the new concerns of sweetness, chewability , mouthfeel, 
and taste must also be considered. Major excipients, such as fillers 
or direct-compaction vehicles. have the major role in the outcome of these 
concerns; process. a lesser (but certainly not minor) role. 
Many of the excipients commonly used in tablet formulation are especially 
applicable for use in chewable tablets due to their ability to provide 
the necessary properties of sweetness and chew ability. In general. 
these fall into the sugar category, although a combination of bland excipient 
with artificial sweeteners may provide a satisfactory alternative. 
The following descriptions of chewable excipients have been compiled 
from general references [15-19} and from other specific references as 
Uko-Nne and Mendes [20} reported on the development of dried honey 
and molasses products marketed for use in chewable tablets. Hony-Tab 
[21] and Mala-Tab [22} are marketed by Ingredient Technology Corporation. 
and consist of 60 to 70% honey or molasses solids codried with wheat 
flour and wheat bran. Both are fr-ee-flowing compressible materials with 
characteristic colors. odors, and tastes that limit their primary applicability 
to the vitamin/food supplement field. These were evaluated with 
vitamin C, vitamin E, and vitamin B complex, as well as with wheat germ 
and bran. 
Two other molasses derivatives are also available that may have applicability 
to specialized chewable tablets. Granular molasses is a cocrystallized 
aggregate of molasses, syrup, and caramel marketed as CrystaFlo by 
Amstar Corporation. It contains up to 94% sucrose and 2% invert, with no 
carriers or flow agents [23}. It is a coarse, free- flowing, granular 
material with the color and taste of molasses. Moisture content is not 
more than 1%. Brownulated, also marketed by Amstar Corporation, has 
similar properties [24]. Neither has been reported on. relative to evaluation 
as tableting agents. 
Compressible sugar, which mayor may not be designated uN.F. ." consists 
chiefly of sucrose that has been processed and combined with other 
constituents in such a way as to render the product directly compressible. 
The N. F. permits the addition of starch, dextrin, invert sugar. and lubricants. 
Compressible sugar is white. odorless, has a sweet taste (equal to 
that of sucrose) and acceptable mouth-feel, a high degree of water solubility, 
and demonstrates good compressibility under normal conditions. It 
has a moisture content of less than 1% and is nonhygroscopic , thus contributing 
to good overall product compatibility and chemical stability, 
despite a tendency toward discoloration when stored at high temperature. 
Its compressibility is markedly effected by moisture content and lubricant 
Generally. equilibrium moisture content is approximately 0.4%; higher 
levels usually produce harder tablets with reduced chewability  while 
lower levels may produce unacceptably soft tablets. Tablet strength also 
may be adversely affected by increased lubricant levels; normally, a magnesium 
stearate concentration between 0.5 and 0.75% is adequate.

Table 2 Common Chewable Tablet Excipients 
Common name Trade name Source Particle size LOD Comments 
Brown sugar Brownulated Amstar 92% on 50 mesh 0.7% Dark brown 
92% sucrose 
bulk density 0.67 g Iml 
Molasses granules CrystaFlo Amstar 100% on 12 mesh 1% Dark brown 
92% sucrose 
bulk density O. 67 glml 
Cornp ressible molasses Mola-Tab Ingredient Technology 50% on 60 mesh 4% Dark brown 
10% thru 120 mesh 70% solids 
Compressible honey Hony-Tab Ingredient Technology 50% on 60 mesh 4% Golden yellow 
10% thru 120 mesh 60% solids 
Compressible sugar Di-Pac Amstar 75% on 100 mesh 0.5% Bulk density 0.64 g/ml 
NuTab Ingredient Technology 50% on 60 mesh 0.5% Bulk density 0.72 g/ml 
10% thru 120 mesh

e,., co 
Dextrose Ifructose 1m altose Sweetrex Mendell 3% on 20 mesh 7% Bulk density 0.6-0.9 
25% thru 100 mesh g/ml 
heat of solution -18 
Dextrates Emdex Mendell 3% on 20 mesh 9% Bulk density 0.68 g Iml 
25% thru 100 mesh 
Lactose DT Sheffield 20% on 60 mesh 1% 
50% thru 100 mesh 
Fast-Flo Foremost 25- 65% on 140 mesh 5.5% 
Mannitol - I CI - Americas 75% on 80 mesh 0.3% BUlk density 0.6 g Iml 
heat of solution - 28. 9 
Sorbitol Sorb-Tab I CI- Americas 
Tablet type Pfizer 33% on 60 mesh 1% Heat of solution - 26.5 
22% thru 120 mesh cal/g 
bulk density 0.7 g/ml

386 Mendes et al. 
Each of the major suppliers produces a different product based on 
composition and process. Di-Pac (Amstar Corporation) consists of a cOcrystallized 
3% highly modified dextrins and 97% sucrose [25]. The former 
acts to interrupt the crystal structure of the latter, thereby improving 
its compressibility. NuTab (Ingredient Technology Corporation) is a chilsonated 
mixture of 4% invert sugar (dextrose and levulose) and 96% sucrose, 
with approximately 0.1% each cornstarch and magnesium stearate as 
processing aids [26]. The agglomerates thus formed are very dense and 
compressible. Several other products are also available from other suppliers; 
they tend to be very similar in their properties. 
Dextrose is the sugar obtained through the complete hydrolysis of 
starch. Its sweetness level is approximately 70% that of sucrose, and it 
is available in both anhydrous (but more hygroscopic) and a monohydrated 
form. Equilibrium moisture content of the former is approximately 1% at 
up to 75% relative humidity; the latter, approximately 10% at up to 80% RH. 
It occurs as a colorless to white crystal or as a white granular powder. 
Dextrose is suitable for use in wet granulation with added binder; its compressibility 
is not sufficient for direct compression. For the latter use, 
Dextrates, a spray-crystallized combination of 95% dextrose with various 
maltoses and higher glucose saccharides, is marketed as Emdex (Edward 
Mendell Co.) [27]. It is free- flowing, compressible, moderately hygroscopic 
(except at high relative humidity where liquification may occur), 
and stable. Deformation during compression occurs over many planes, 
resulting in extremely hard tablets at relatively low compressional force 
levels. Tablets harden markedly during the first few hours after compression. 
Because of the high equilibrium moisture content of dextrose and its 
potential for reaction with amines, dextrose and Dextrates may present 
problems in some applications. A related commercially available product is 
Sweetrex (Edward Mendell Co.), a combination of approximately 70% dextrose 
and 30% fructose with minor amounts of related saccharides [28]. It 
is slightly sweeter than sucrose, and its other properties are similar to 
other dextrose- related excipients. 
Lactose is a monosaccharide produced from whey, a byproduct of the 
processing of cheese. Although generally acknowledged as the most 
widely used pharmaceutical excipient in the world. its applicability to 
chewable tablets is minor at best, due to its extremely low sweetness 
level (15% of sucrose). This deficiency requires the addition of an artificial 
sweetener of sufficient potency to overcome lactose's blandness. 
Assuming that such an addition is acceptable, lactose may be considered a 
very useful filler. For wet granulation applications, regular pharmaceutical 
grades (hydrous fine powders) are available. 
For direct compression, an anhydrous grade having good flow and 
compressional characteristics is available as Lactose DT (Sheffield Products 
Co. ). This product has the appearance of granulated fine crystals that 
easily deform under pressure, providing excellent compressibility [29]. 
Another directly compressible form is Fast-Flo Lactose (Foremost Foods 
Co. ), an aggre gated microcrystalline a-lactose [30 I, Although more 
flow able , it is less compressible than anhydrous lactose. It is also more 
prone to discoloration upon exposure to high temperature and humidity. 
Both often require higher than normal lubricant levels in their formulations. 
Mannitol is a white, crystalline polyol approximately 50% as sweet as 
sucrose. It is freely soluble in water and, when chewed or dissolved in

Chewable Tablets 387 
the mouth, imparts a mild cooling sensation due to its negative heat of 
solution. This combined with an exceptionally smooth consistency has 
made mannitol the excipient of choice for chewable tablet formulations. 
In powder form, it is suitable only for wet granulation in combination 
with an auxiliary binder. For direct- compression applications, a granular 
form ("tablet grade") is available (ICI Americas). Mannitol has a low 
moisture content, is nonhygroscopic, and the equilibrium moisture content 
remains at approximately 0.5% up to a relative humidity of approximately 
85% [17]; these properties, combined with those related to sweetness and 
mouth- feel, represent significant advantages for the formulation of chewable 
Sorbitol is a slightly sweeter and considerably more hygroscopic isomer 
of mannitol. For direct compression, it is available commercially as SorbTab 
(ICI Americas) and Crystalline Tablet Type (Pfizer Chemical). Although 
similar in that both are aggre gated microcrystals covered with dendrites, 
the structures of these materials are sufficiently different to provide 
somewhat different compressional characteristics. Relative humidities 
greater than 50% at 25C should be avoided. Equilibrium moisture content 
rises lO-fold (from 2.7 to 28.4%) between 64 and 75% RH [17]. 
Artificial sweeteners are a class of excipients that are of significant 
importance to chewable tablet formulations. As noted above, none of the 
other sugars are as sweet as sucrose, which itself is often not sweet 
enough to mask the bitterness or sourness of many drugs. Presently, 
there is considerable regulatory disagreement worldwide concerning the 
use of these materials; some are approved for use in some countries but 
not others. Although this form ulation problem 1s not related solely to 
artificial sweeteners, but in fact to virtually all classes of excipients (and 
drug's), it is probably a greater problem only with colorants. Three 
materials appear to be usable from other than a regulatory perspective: 
aspartame, cyclarn ate. and saccharin. All have potency (sweetness) levels 
many times that of sucrose. permitting the use of very low concentrations 
(less than 1%) to cover most bitter drugs. Other semisynthetic sweeteners, 
derived from glycyrrhiza, have enjoyed some degree of popularity over the 
years. These are much sweeter than sucrose but less sweet than saccharin. 
It is recommended that the formulator validate the current regulatory acceptance 
of the intended sweetener prior to its use for a particular product 
and market country. 
From the perspective of consumer acceptance. taste is almost certainly the 
most important parameter of the evaluation of chewable tablets. Taste is a 
combination of the perceptions of mouth- feel, sweetness. and flavor. 
Mouth-feel is affected by the heat of solution of the soluble components 
(negative being preferable), smoothness of the combination during chewing. 
and hardness of the tablet. These factors are directly and almost entirely 
related to the active ingredient and major excipients. 
Sweetness. at an appropriate level. is a necessary background to any 
flavor. The primary contributors to sweetness in a chewable tablet are 
the drug, natural sweeteners, and artificial sweetness enhancers that may 
be incorporated in the form ulation ,

388 Mendes et cl, 
A. Sweeteners 
Most of the excipients described in the previous section as appropriate 
bases for chewable tablets have. as their major property, a level of sweetness 
that contributes positively to the overall taste of the product. Often, 
the sweetness imparted by these excipients is insufficient to overcome the 
bad taste of the drug. In these cases, the formulator must often use 
artificial enhancers to increase the overall sweetness impact. 
Table 3 presents a compilation of the most common artificial and synthetic 
sweeteners used in pharmaceutical products, their relative sweetness 
levels, and pertinent comments. It is important to note that, with all excipient 
materials, it is the responsibility of the formulator to ascertain the current 
regulatory status of the material in the country for which the product 
is intended. At the present time, the acceptability of saccharins and cyclamates 
varies from country to country, while aspartame and glycyrrhizin 
are generally (though perhaps not universally) recognized as safe. In 
addition to use restrictions, label requirements may apply. 
The obvious major advantage to the use of artificial sweeteners is their 
relative potencies> which may range from 50 to 700 (compared to sucrose) 
depending on choice and conditions of use. For example. the relative 
sweetness of saccharin decreases as the sweetener level is increased. Furthermore, 
as the saccharin concentration is increased. the level of unpleasant 
aftertaste increases. 
Glycyrrhizin (Magnasweet) is a licorice derivative with an intense, late, 
long-lasting sweetness [31]. These properties permit its use as an auxiliary 
sweetener to boost sweetness level while overcoming aftertaste. Typical use 
levels are 0.005 to 0.1%, with higher concentrations tending to lend a 
slight licorice flavor. 
Aspartame (NutraSweet) is the most recently introduced artificial 
sweetener, having been approved for use in the United States in 1981. 
Its relative sweetness level is approximately 200, and its duration is 
greater than that of natural sweeteners [32]. It enhances and extends 
citrus flavors. Aspartame's dry stability is said to be excellent at room 
temperature and a relative humidity of 50%, while in solution it is most 
stable at pH 4. Its typical usage level in chewable tablets is 3 to 8 mg 
per tablet. 
B. Flavors 
Flavoring agents, both natural and artificial, are available in a variety of 
physical forms from a large number of suppliers specializing in these 
materials. Virtually all offer technical support services, which will be 
addressed in the section on flavor formulation. Forms available include 
water-miscible solutions. oil bases, emulsions, dry powders, spray-dried 
beadlets, and dry adsorbates. A typical flavor house might catalog 50 or 
more basic flavors, while having the capability of producing several 
hundred combinations for a given application. 
C. Flavor Selection and Formulation 
Initially, the inherent taste of the active drug must be evaluated to determine 
its probable contribution to the formulation, Next, a decision must

Chewable Tablets 
Table 3 Approximate Relative Sweetness of 
Different VehiCles and AUxiliary Sweeteners 
Relative sweetnessb 
a Cyclamates 
Saccharins a 
Dextrose (glucose) 
Fructose (levulose) 
ucrose IS 
status must be checked before 
taken as a standard of 1 for 
be made relative to formulation components that would impact on both the 
pharmaceutical properties and organoleptic characteristics of the tablet. 
Throughout the steps in formulation development, these considerations 
must be maintained and eventually optimized. The goal must be a baseline 
formulation having acceptable properties such as hardness, friability, and 
dissolution, while providing a suitable mouth- feel and sweetness background 
for flavoring. Appropriate selection of processes and excipients discussed 
earlier in this chapter, and others, will lead to the development of such a 
Having succeeded in the preparation of one or more unflavored bases, 
the development pharmacist should next prepare several basic flavored 
preference samples. These should be designed to narrow the flavor focus 
to one or more groups or categories of flavor preferred by decision makers 
within the company. Tables 4 and 5 provide general guidelines for such 
preliminary choices based on baseline taste and drug product type. 
In creating the preliminary flavor samples. the pharmacist should 
recognize age-dependent preferences. Children have a high tolerance for 
sweet and low tolerance for bitter; as age progresses, tolerance for bitter 
taste increases as the taste buds and olfactory centers lose sensitivity. 
Generally. mild tastes are less fatiguing and therefore better choices. 
Menthol, spices. and mint flavors tend to anesthetize the taste buds and 
reduce flavor reaction. Vanilla, on the other hand. tends to enhance 
other flavors [331. 
As stated previously, flavor houses generally provide flavor development 
services to their customers. Once preliminary samples have been

390 Mendes et al. 
Table II Suggested Flavor Groups for General Baseline Taste Types 
Sweet Vanilla, stone fruits, grape, berries, maple, honey 
Sour (acidic) 
Citrus, cherry, rasp berry, straw berry, root beer, 
anise, licorice 
Nutty, buttery, butterscotch, spice, maple, melon, 
raspberry, mixed citrus, mixed fruit 
Licorice, anise I coffee, chocolate, wine, mint, grapefruit, 
cherry, peach, raspberry, nut, fennel, spice 
Mint, chocolate, cream, v anilla 
Grape, burgundy, lemon-lime 
Source: Adapted from Refs. 34 and 35. 
used to determine flavor category, the pharmacist should turn the remaining 
flavor development activities over to the flavor chemists. This will 
require that the company provide samples of the baseline granulation (and 
tablets) for the chemist to use. The more information that can be provided 
(under confidentiality agreements or other arrangements), the greater 
the probability that the supplier can produce the desired result. 
Usually a broad range of samples will be created for evaluation by 
small, informal groups within the company. As development progresses, 
tablet samples should be evaluated by formal taste panels with appropriate 
statistical planning to assure final selection of a product with a high probability 
of success in the marketplace. 
D. Flavor Quality Assurance 
The qualification of multipte vendors for a specific flavor is a virtual impossibility. 
Each supplier will produce and provide a proprietary combination 
Table 5 Several Commonly Recommended Flavor Applications 
Mint (peppermint, spearmint) 
Mint anise 
Bavarian cream 
Cherry cream punch 
Anise birch 
Black currant 
Rum peach 
Spice vanilla 
Wild cherry 
Menthol- eucalyptus 
Fresh pineapple 
Passion fruit 
Toasted nut 
Source: Adapted from Refs. 34 and 35.

Chewable Tablets 391 
with similar, but not absolutely identical, flavors. While the application of 
these flavors should be interchangeable, they should be processed through 
the raw materials quality control system as unique components. Purchasing 
contracts and acceptance specifications should be tightly drawn to 
assure that batch-to- batch variation is minimized, since these complex 
materials are critical to the market acceptance of the product. 
This complexity also leads to a potential for instability. Flavors are 
highly susceptible to decomposition and/or loss of potency through exposure 
to elevated temperature and humidity. The supplier should provide 
storage and shelf life information, and such instructions should be followed. 
The stability of a flavor compound. in its raw form or in a finished 
product , is difficult to follow. Although gas chrom atography may be 
capable of determining the myriad of components in the flavor, very minor 
chemical changes which are analytically undetectable could alter the taste. 
In reality, unlike most pharmaceutical ingredients for which the user 
shares responsibility with the supplier, the flavor user must rely almost 
entirely on the reputation and integrity of the supplier. 
E. Flavor IColor Integration 
The final aspect of taste psychology requires that the flavor and color 
match or correspond. A mismatch may detract from consumer acceptance. 
Table 6 provides a general guideline for such matching. 
Table 6 Flavors and Corresponding Color Guidelines 
Flavor Color 
Cherry, wild cherry, tutti- frutti , raspberry, 
strawberry, apple 
Chocolate, maple, honey, molasses, butterscotch, 
walnut, burgundy, nut. caramel 
Lemon, lime, orange, mixed citrus, custard, banana, 
cherry, butterscotch 
Lime, mint, menthol, peppermint, spearmint, pistachio 
VaniIla I custard, mint I spearmint, peppermint, 
nut, banana. caramel 
Grape I plum, licorice 
Mint, blueberry, plum, licorice, mixed fruit 
Pink to red 
Yellow to orange 
Off white to white 
Violet to purple 
For speckled tablets, color of speckling or background should correspond to 
flavors chosen.

392 Mendes et al. 
Colorants are used in the manufacture of chewable tablets for the following 
1. To increase aesthetic appeal to the consumer 
2. To aid in product identification and differentiation 
3. To mask unappealing or nonuniform color of raw materials 
4. To complement and match the flavor used in the formulation 
Colorants are available either as natural pigments or synthetic dyes. 
However, due to their complexity and variability, the natural pigments 
cannot be certified by the Food and Drug Administration (FDA) as are the 
synthetic dyes. Certification is the process by which the FDA analyzes 
batches of certifiable dyes to ascertain their purity levels and compliance 
with specifications and issues a certified lot number. 
The Food Drug and Cosmetic Act of 1938 created three categories of 
coal tar dyes, of which only the first two are applicable to the manufacture 
of chewable tablets. 
1. FD&C colors: These are colorants that are certifiable for use in 
foods, drugs, and cosm etics. 
2. D&C colors: These are dyes and pigments considered safe for use 
in drugs and cosmetics when in contact with mucous membranes or 
when ingested. 
3. External D&C: These colorants, due to their oral toxicity, are not 
certifiable for use in products intended for ingestion but are considered 
safe for use in products applied externally. 
Two main forms of colorants are used in the manufacture of chewable 
tablets depending on whether the process of manufacture is by wet granulation 
or direct compression. 
A. Dyes 
Dyes are chemical compounds that exhibit their coloring power or tinctorial 
strength when dissolved in a solvent [36]. They are usually 80 to 93% 
(rarely 94 to 99%) pure colorant material. Dyes are also soluble in propylene 
glycol and glycerine. 
Certifiable colorants, both "primary" and "blends" of two or more primary 
colorants. are available for use in a number of forms including 
powder, liquid, granules, plating blends, nonflashing blends, pastes, and 
dispersions [37]. For the formulation of chewable tablets the powders, 
liquids or dispersions are used in the wet gr-anulation stage of tablet manufacture. 
The powders are first dissolved in water or appropriate solvent 
and used in the granulation process. 
Dyes are synthetic, usually cheaper, and are available in a wider range 
of shades or hues with higher coloring power than the natural pigments. 
The physical properties of dyes (particle size, variation in the grinding 
and drying process, different suppliers) are usually not critical in 
terms of their ability to produce identically colored systems. The tinctorial 
strength of a dye is directly proportional to its pure dye content. This 
means that 1 unit of a 92% pure dye is equivalent to 2 units of a 46% pure

Chewable Tablets 393 
dye. Dyes are generally used in the range of 0.01 to 0.03% in chewable 
tablet formulations. and the particle size range of dyes is USUally between 
12 and 200 mesh. Dyes used in the wet granulation step are usually dissolved 
in the gr-anulation fluid; the granulation and drying operations must 
be optimized to prevent or minimize dye migration. This problem is exceptionally 
important when dye blends are used. since a "chromatogr-aphic" effect 
may be obtained as the different primary dyes migrate through the granules 
at different rates leading to nonuniform colored granules. 
Solutions of dyes should be made in stainless steel or glass-lined tanks 
(for minimization of dye-container incompatibility) with moderate mixing 
and should routinely be filtered to remove any undissolved dye particles. 
Dye solutions in water that are intended to be stored for 24 hr or more 
should be adequately preserved to prevent microbial contamination. Suitable 
preservatives include propylene glycol, sodium benzoate with phosphoric 
acid or with citric acid. 
During storage, use, and processing, dyes should be protected against 
1. Oxidizing agents. especially chlorine and hypochlorites. 
2. Reducing agents, especially invert sugars, some flavors, metallic 
ions (especially aluminum, zinc, tin, and iron), and ascorbic acid. 
3. Microorganisms, especially mold and reducing bacteria. 
4. Extreme pH levels; especially FD&C Red #3 which is insoluble in 
acid media and Should not be used below pH 5. O. Also, effects of 
fading agents such as metals are greatly enhanced by either very 
high or low pH values. 
5. Prolonged high heat-s-only FD&C Red #3 is stable On exposure to 
prolonged high heat. Thus I dyes should be processed at low to 
moderate temperatures and should have only very brief exposure 
to moderate or high heat levels. The negative activity of reducing 
and oxidizing agents is greatly enhanced by elevated temperatures. 
6. Exposure to direct sunlight-FD&C Red #40 and FD&C Yellow #5 
have moderate stability to light, while FD&C Blue #2 and FD&C Red 
#3 have poor light stability. It is important to minimize the exposure 
of products to direct sunlight, especially products containing 
dye blends. 
To compensate for losses due to fading and other dye loss during 
processing and storage, some formulators add a slight excess of dye at the 
beginning. This approach should be cautiously employed since One can 
obtain unattractive shades when too much color is added at the beginning 
in an attempt to provide for time-dependent or processing color loss. 
Regulations covering all aspects of colorants. including their procedures 
for use, provisionally and permanently certified and uncertified color additives, 
and use levels and restrictions for each coloring additive. are covered 
in the Code of Federal Regulations 21 CFR parts 70 through 82. 
Regulatory updates on color additives should be monitored by formulators 
using colorants. These updates and revisions are published in the 
Federal Register. 
Concerns still persist about the safety of absorbable dyes despite completed 
studies done so far. This has led dye manufacturers and suppliers 
to develop and test nonabsorbable dyes, which are considered safer by 
virtue of their nonabsorption from the gastrointestinal tract. Long-term

394 Mendes et al. 
feeding studies using animals are continuing since 1977 and. if the profiles 
are good, approvals may be expected in the not too distant future. 
B. Lakes 
Lakes have been defined by the FDA as the "aluminum salts of FD&C watersoluble 
dyes extended on a substratum of alumina." Lakes prepared by 
extending the calcium salts of the FD&C dyes are also permitted but to 
date none has been made. Lakes also must be certified by the FDA. 
Lakes, unlike dyes, are insoluble and color by dispersion. Consequently, 
the particle size of lakes is very critical to their coloring capacity 
or tinctorial strength. Generally, the smaller the particle size, the higher 
the tinctorial strength of lakes due to increased surface area for reflected 
Lakes are formed by the precipitation and absorption of a dye on an 
insoluble base or substrate. The base for the FD&C lakes is alumina hydrate. 
The method of preparation of the alumina hydrate and the conditions 
under which the dye is added or absorbed determines the shade, 
particle size, dispersability, as well as tinctorial strength. Other important 
variables are the temperature, concentration of reactants, final pH, and the 
speed and type of agitation [36]. 
Lakes contain 1 to 45% pure dye but, unlike dyes, the tinctorial strength 
is not proportional to the pure dye content. Also, the shade or hue of a 
lake varies with the pure dye content. 
Particle size of lakes is in the range of 0.5 to 5 um , but the micrometer 
and submicrometer size leads to significant electrostatic cohesive 
forces causing particles to agglomerate to 40 to 100 urn , For effective use 
in direct-compression formulation of chewable tablets, lakes should preferably 
be deagglomerated to their original particle size ranges by premixing 
them with some of the inert ingredients in a formulation using highshear 
mixers and finally incorporating the rest of the ingredients. 
FD&C lakes are available in six basic colors: one yellow, one orange, 
two reds (a pink-red and an orange-red), and two blues (a green-blue 
and a royal blue). Blends are available to provide more lake colors as 
needed including brown, green, orange, red. yellow. and purple. 
Lakes are used in chewable tablets made by direct compression in a 
concentration range of 0.1 to 0.3%. They possess a higher light and heat 
stability than dyes, are quite inert, and are compatible with most ingredients 
used in chewable tablets. 
Lakes are usually used in the direct-compression method of chewable 
tablet manuracturing. However, some unique cases of wet granulation may 
also call for the use of lakes. These unique cases involve "chromatographic 
effects" previously described when blends of soluble dyes are used 
in the gr-anulation step. This problem can be resolved by using lake 
blends instead of dye blends since lakes, being insoluble. do not migrate 
during massing and drying of granules. 
Table 7 gives some physical and chemical properties of certified 

181.)/.... l'l!.l:lh..tll l:l.uJ vlH;:IlJJ ....a1 .t~_r\;)~l1t;.b o l L.t;.~l1jjl;.u I..-_LJ..:> 
(g/100 ml) 
Stability to at 25C 
FD&C Name Chemical Tinctorial 
(common name) class Light Oxidation pH Change strength Hue Water 25% EtOH 
Red No. 3 Xanthine Poor Fair Poor V. good Bluish pink 9 8 
Red No. 40 Monoazo V. good Fair Good V. good Yellowish red 22 9.5 
Yellow No. 6 Monoazo Moderate Fair Good Good Reddish 19 10 
(Sunset Yellow FCF) 
Yellow No. 5 Pyrazolone Good Fair Good Good Lemon yellow 20 12 
Green No. 3 TPM
Fair Poor Good Excellent Bluish green 20 20 
(Fast Green FCF) 
Blue No. 1 TPMa 
Fair Poor Good Excellent Greenish blue 20 20 
(Brilliant Blue FCF) 
Blue No. 2 Indigoid V. Poor Poor Poor Poor Deep blue 1.6 0.5 
~riphenyl methane. 
<:'nl ('p' 'r'l Rf ~"rl fr -rn Ppf ~.

396 Mendes et at 
A. General Considerations 
Four important aspects of chewable tablet manufacture are the proper incorporation 
of the coloring agent, assurance of necessary particle size distribution) 
maintenance of correct moisture content I and achievement of proper 
tablet hardness. All of these are the routine responsibility of the manufacturing 
department once the parameters have been established during 
development. It is therefore critical that process development and scale-up 
considerations be thoroughly explored in order to ensure the establishment 
of proper specifications. 
As with all types of tablets, if the granulating process involves wet 
granulation, the extent of wetting and the rate and extent of drying must 
be defined. Overwetting can be expected to produce harder granules that 
may have poor compressional characteristics, resulting in softer and more 
friable tablets. Due to the lesser degree of particle deformation I these 
tablets often have a gritty mouth- feel when chewed. Overwetting during 
granulation also leads to longer drying times in order to achieve the 
desired moisture level or. worse. a higher moisture level due to failure to 
compensate through adjustment of the drying cycle. Improper wetting 
and drying may also adversely affect the particle size distribution, leading 
to ineffective postgranulation blending. poorer flow, and increased weight 
Also, the method and appropriate order for the addition of the flavor 
and color must be determined if wet granulation is being used. Since most 
flavor substances are volatile, they cannot be subjected to elevated temperature. 
For this reason, they cannot be incorporated prior to granulation; 
rather, flavors are added (often as premixes) in the final blending operation 
of the process. The color, if in the form of a lake, would be incorporated 
in the same step. The concentrations of these ingredients normally 
do not exceed 0.1% and generally would be even lower. An important consideration 
is the assurance of uniform blending; rarely would analytical 
methods be used to establish flavor or color uniformity. Since the final 
blending step may require the combining of a 99; 1 materials ratio, the 
establishment and validation of this operation is extremely important. 
The uniformity of color incorporation needs to be viewed from the perspective 
of performance. If color is used in the form of a dye added to a 
wet granulation, the final blending operation usually consists of the addition 
of uncolored (white?) powders (lubricants, etc.) to colored granules. 
It is assumed that the white powder will uniformly coat the colored granules, 
thus resulting in an even distribution of color. However. When these 
granules fracture during compression, the uniformity will often be disturbed, 
resulting in tablets having lighter or deeper color on the opposite background; 
this is referred to as "mottling. n 
On the other hand, if color is added as a lake following granulation 
(or to a direct-compression blend), then the blending operation consists 
of the addition of colored powder to uncolored (white?) granules. Again, 
it is assumed that the colored powder will uniformly coat the white 
granules , However, during compression, the granules fr acture and release 
fresh white material to the surface, resulting in white spots on a colored 
background, or vice versa ("speckling"). In either case, the result is a 
less than elegant finished product. Since the visible problem in both cases

Chewable Tablets 397 
may be reduced through lessening the color contrast between the materials, 
two common approaches are the use of low concentrations of light colors 
and the use of high-intensity mixing of reduced particle size materials in 
order to assure thorough blending. 
B. Antacids 
Antacid products compose a rather large percentage of the over-the-counter 
(OTC) drug market. Efficacy studies [38,39] have questioned the comparative 
efficacy of chewable antacid tablets to their suspension antacid counterparts 
due to the state of hydration of the latter. However, the inconvenience 
of carrying a bottle of liquid and a measuring device is obvious. 
Consequently, the user is faced with a choice between convenience and 
possibly greater efficacy. Most choose the convenience of the solid form 
at least when away from home, and probably all of the time. 
Few antacid tablets specifically formulated for swallowing are presently 
marketed. All other solid antacid products currently available are in the 
form of chewable tablets, chewing gum, or chewy squares. 
From a formulation perspective, antacids present extreme difficulty due 
to the nature and quantity of the active ingredients. They are generally 
metallic, astringent, chalky, and lor gritty I thus providing a combination 
of bad taste and bad mouth-feel to be overcome. In addition, the usually 
high dosage levels required result in very large tablets (typically 5/8-in. 
diameter, 700 to 1000 mg weight), with two tablets the normal dose. This 
quantity of material, coupled with the frequency of dosing, may lead to 
taste fatigue even with a good-tasting product; one that is poor or mediocre 
will quickly lose acceptability in the marketplace. 
Table 8 provides a list of the commonly used antacids; generally, these 
are used in combinations of two or more to provide better therapeutic action. 
Table 8 Common Antacid Drugs and Some TypiCal 
Dose Ranges 
Aluminum hydroxide 
Calcium carbonate 
Magnesium hydroxide/oxide 
Magnesium trisilicate 
Aluminum carbonate 
Dihydroxyaluminum aminoacetate 
Dihydroxyaluminum sodium carbonate 
Magnesium carbonate 
Magnesium gluconate 
Potassium bicarbonate 
Sodium bicarbonate 
Source: Compiled from Ref. 40. 
80- 600 mg 
194- 850 mg 
65- 400 mg 
20-500 mg

398 Mendes et al. 
In addition to the antacid components I other ingredients are often 
found in these products as adjunct actives. These include simethicone 
(dimethicone, dimethyl polysiloxane) at a level of 20 to 40 mg per tablet 
as an antiflatulent. Peppermint oil, approximately 3 mg per tablet, is sometimes 
used as a carminative. Alginic acid, 200 to 400 mg, is also used by 
at least one company. 
Following are two examples of antacid tablet formulations. Example 3 
is a dextrose (Dextrates)- based direct-compaction system. while Example 4 
is a sucrose (compressible sugar)- based product. 
Example 3: Chewable Antacid Tablets 
Ingredient mg /tablet 
FMA-11 * (Reheis Chemical) 
Syloid 244 
Pharmasweet Powder (Crompton and Knowles) 
Magnesium stearate 
Total Weight 
1. Mix FMA-11 and Syloid together for 5 min. Screen 
through 30-mesh screen (if ingredients not already prescreened) 
and mix for 10 to 15 min. 
2. Add Emdex and Pharmasweet to step 1 and blend 
thoroughly for 10 to 15 min. 
3. Add magnesium stearate to step 2, blend 5 min, and 
"'Aluminum hydroxide/magnesium carbonate codried gel. 
Note: An appropriate flavor may be added in step 2. 
Source: Ref. 27. 
Comments: FMA-ll is a very fine powder that tends to 
"sllp'' rather than flow, thereby leading to blending problems 
and overfilling of the tablet machine feed frame. Syloid, a 
synthetic silica. acts as a glidant to improve flow characteristics. 
Pharmasweet, in conjunction with Emdex, provides sufficient 
sweetness to compensate for the bland chalkiness of the 

Chewable Tablets 
Example 4: Antacid Tablet Direct Compression 
Ingredient mg Itablet 
Aluminum hydroxide and magnesium carbonate codried gel 
(Reheis FMA-1l) 
Di-Pac DTE 
Microcrystall ine cell ulose (Avicel) 
Calcium stearate 
Mix all ingredients and compress on standard S/8-in. flat-face bevel edge 
punch to a hardness of 8-11 SCA Units. 
Source: Ref. 25. 
Comments: This formulation is similar to Example 3. Two differences are 
the slightly higher drug load (29% versus 25%) and the replacement of 
Dextrates with Compressible Sugar. If desired, the formula could alternatively 
be wet-granulated using water to wet the sugar and cellulose. 
C. Cough ICoid Analgesics 
The primary appeal of products in this category is the pediatric market 
into the teens. Generally, levels of the drugs are one-quarter or less of 
the adult dose, which would require the use of a large number of tablets 
by an adult. Despite the extremely large market, the industry has failed 
to exploit the potential of adult strength chewable tablets and the patient 
convenience such products might provide. 
Drugs commonly encountered include aspirin, acetaminophen, chlorpheniramine, 
phenylpropanolamine, pseudoephedrine, and dextrcmethcrphan, 
These may be used alone or in various combinations with appropriate attention 
to possible incompatibilities. The one common property all of these 
drugs share is unpleasant taste. Aspirin is acidic and astringent; the 
others are all very bitter. 
None, except acetaminophen, present compressibility problems; aspirin 
is relatively compressible, and the others are used in low doses and therefore 
low percentage compositions. Acetaminophen has inherently poor compressibility, 
although newer, directly compressible forms are marketed by 
some suppliers. These have previously been gr-anulated by the producer 
to prepare them for tableting; the alternative for chewable tablets is a wet 
granulation process, as illustrated in Example 5. 
The other drug products mentioned can be produced by direct compression. 
as shown in Examples 6 and 7. 
Incompatible drugs that may be desirable in combination, such as 
aspirin and phenylpropanolamine, require special treatment as they would 
in a nonchewable tablet. The drugs must be kept separated; this can be 
accomplished through the multilayer technology or through coating one or 
both drugs prior to blending.

400 Mendes et al. 
Example 5: Chewable Acetaminophen Tablet (Wet Granulation) 
Quantity per 
Ingredient tablet 
Mannitol, USP 
Sodium saccharin 
Acetaminophen, N.F. (5. B. Penick, coarse granular) 
Binder solution 
Peppermint oil 
Syloid 244 
Banana, Permaseal F-4932 
Anise, Perrnasaal F-2837 
Sodium chloride (powdered) 
Magnesium stearate 
*lncludes 5.4 mg acacia and 16.2 mg gelatin. 
First, prepare a binder solution consisting of: 
720.0 mg 
6.0 mg 
120.0 mg 
21. 6 mg* 
0.5 mg 
0.5 mg 
2.0 mg 
2.0 mg 
6.0 mg 
27.5 mg 
906.0 mg 
Acacia (powdered) 
Gelatin (granular) 
15 9 
45 9 
q, s , ad 400 rnl 
Screen the mannitol and sodium saccharin through a 4o-mesh screen. 
Blend thoroughly with the acetaminophen. Using 180 ml of binder solution 
per' 000 tablets, granulate and dry overnight at 140 to 150F. Screen 
through a 12-mesh screen. Adsorb the peppermint oil onto the Syloid 244 
and mix with the flavors and sodi urn chloride. Blend this flavor mixture, 
the dried granulation, and the magnesium stearate. Compress on 1/2-in. 
flat-face bevel edge punches to a hardness of 12 to 15 kg. 
Comments: Acetaminophen is generally regarded as very difficult to compress, 
and usually is processed by wet granulation in order to permit 
higher drug loading. The acacia-gelatin binder provides high tablet 
strength; the solution should be freshly prepared to prevent microbial 

Chewable Tablets 
Example 6: Chewable Children's Antihistamine Tablets 
Ingredients mg /tablet 
Phenylpropanolamine HCI 
Chlorpheniramine maleate 
Magna Sweet 165 (MacAndrews & Forbes) 
Flavor, Artificial Red Punch. S.D. (Crompton & Knowles) 
Color, cherry (Crompton & Knowles) 
Magnesium stearate 
Total weight 
1. Mix phenylpropanolamine and Emdex together for 10 min. 
2. To a small portion of 1 add chlorpheniramine and Magna Sweet and 
mix for 15 to 20 min. 
3. Mix 1 and 2 together. Add flavor and blend for 10 to 15 min. 
4. To 3 add color and mix for 20 to 25 min. 
S. Add magnesium stearate, blend 5 min and compress. 
Source: Ref. 27. 
Comments: The combination antihistami ne-decongestant is commonly used 
for both allergy and upper respiratory tract infections. The apparent 
potential incompatibility between the amine drugs and dextrate excipient 
does not cause discoloration except in the presence of moisture and heat. 
Often, it is desirable to add aspirin or another analgesic for pain and 
fever; the combination should be expected to exhibit poor stability unless 
separated by techniques such as the Use of multilayer tablets. 
D. Vitamins/Minerals/Food Supplements 
It has long been common medical practice to supplement the diet with vitamin-
mineral products from infancy into elderliness. Although such products 
may be presumed unnecessary for those who consume an appropriate diet. 
it is generally recognized that the likelihood of widespread proper dietary 
habit is low. 
In infancy. vitamin supplements are provided as liquids for "dropper" 
dosage. Usually. at around the age of 2 or 3, children are switched to a 
chewable multivitamin. with or without fluoride depending on local water 
supplies. Because of the age group for which these products are intended, 
the combinations tend to be limited and of relatively low dosage. At least 
one manufacturer. however. produces a higher strength product for older 
children. There appears to be no product in the marketplace intended 
for adults. 
Children's vitamins in recent years have become extremely complex 
from a manufacturing and tooling perspective. Marketing pressures have 
dictated the adoption of extraordinarily detailed shapes such as cartoon 
characters. animals. etc., designed to stimulate sales through appeal to 
children. Such tablets require punches and dies with numerous compound

Example 7: Children's Buffered Aspirin 
Chewable Tablet 
Mendes et al, 
Aluminum hydroxide dried gel 
Aspi rln , 4Q-mesh crystals 
Mafco Magnasweet-150 
Orange flavor (F&F no. 11598) 
mg {tablet 
13 mg 
81 mg 
2 mg 
8 mg 
93.4 mg 
0.6 mg 
2 mg 
1. Blend the NuTab and aluminum hydroxide 
dried gel for 10 min. 
2. Add the aspirin and blend for an additional 
5 min. 
3. Premix the Primogel, talc, flavor. and 
Magnasweet and pass through a so-mesh 
4. Add the premix and blend for an additional 
5 min. 
Comments: The combination of NuTab and 
Magnasweet adds sufficient sweetness to offset 
the tartness of the aspi rin and the orange 
flavor. In the dry state, there is no incompatible 
reactivity between the acidic aspirin 
and alkaline aluminum hydroxide. 
curves and punch faces with many detail markings. These require optimized 
formulations and manufacturing processes in order to ensure acceptable 
appearance quality levels. 
The total active ingredient content is high and consists of a combination 
of difficult tastes. Barry and Weiss [41] described the basic taste 
characteristics of various vitamins as follows: 
Vitamin A acetate. vitamin D2 (ergocalciferol). and vitamin E (DLtocopheryl 
acetate): "substantially tasteless" 
Vitamin Bl (thiamine hydrochloride or nitrate): "yeasty," bitter 
Vitamin B6 (pyridoxine hydrochloride); slightly bitter. slightly salty 
Vitamin B12 (cyanocobalamin): tasteless 
Niacinamide: very bitter 
Vitamin C (ascorbic acid): Sour 
Vitamin C (sodium ascorbate): less sour, salty. somewhat "soapy" 
Calcium pantothenate: bitter 
Biotin: tasteless 
Folic acid: nearly tasteless 
Minerals (e. g. iron salts): metallic

Chewable Tablets 403 
A multivitamin and mineral mixture will have a combination of bitter 
plus sour plus salty plus metallic tastes. The sourness can be depressed 
by adding sweetness via the vehicle (e. g., mannitol) and additional 
sweetener (e. g., saccharin sodium). The sourness is further depressed 
by a careful choice of the ratio of ascorbic acid to sodium ascorbate so as 
to retard the acidity, and by adding a citrus flavor that corresponds to 
the degree of tartness chosen [41]. Ferrous fumarate and ferric pyrophosphate 
are relatively tasteless compared to other iron salts. A further 
reduction in the metallic taste of ferrous fumarate has been accomplished 
by a patented coating process in which the iron salt is coated with at least 
one of the following: a monoglyceride or a diglyceride of a saturated fatty 
acid, using spray-congealing technique [42]. Other mineral salts that are 
"practically nonmetallic" in taste include manganese glycerophosphate, zinc 
oxide, magnesium oxide, and dibasic calcium phosphate, all of which can 
be used to provide the corresponding trace metals desired in the vitaminmineral 
combination formula. Finally, the bitterness must be masked. For 
best results, the Br-complex group of vitamins are chosen in individually 
coated forms known as Rocoat vitamins, which are prepared by spray congealing 
of the vitamins with monoglycerides and diglycerides of edible 
fatty acids. The end product has a vitamin/fat ratio of 1: 3. Niacinamide 
is also available in this form. Vitamin B12 is available in gelatin (0.1%) or 
as Stablets (1%). Vitamins A and D are also available as free- flowing 
powders protected in a matrix 0 f gelatin, sugars or starches, and preservatives-
and are known as Crystalets or Beadlets. Vitamin E is available as 
an adsorbed dry powder or as microbeadlets [41]. After a choice has 
been made of the minerals, the physical form of the B- complex vitamins, 
ni acinamide, vit amins A, E, and D, and the ratio of ascorbic acid to sodium 
ascorbate, the final flavoring must be chosen. The approach [41,43] is to 
blend out the overall taste of the formula so that the vitamin taste becomes 
part of the flavor impression by adding appropriate flavors and flavor enhancers 
that complement the tartness, saltiness, sourness, and sweetness 
already present in the baseline formulation. The complementary flavors 
recommended [41,43] are citrus, mint, apricot, cherry, orange, peach, 
strawberry, rasp berry, wintergreen, pineapple, and cherry (among many 
other potential candidates). 
A typical directly compressible multivitamin with iron is illustrated in 
Example 8. 
The most common single-vitamin product is vitamin C, which often is 
desirable in chewable tablet form. Since ascorbic acid is extremely sour 
tasting, additional steps are usually taken to improve the flavor. A combination 
of ascorbic acid and sodium ascorbate, both of which are available 
in direct-compression form, is less sour and therefore easier to flavor (see 
Example 9). Another approach (Example 10) requires coating the ascorbic 
acid with ethylcellulose to reduce its solubility and therefore its sourness. 
Generally, citrus flavors are preferred to compliment the taste. 
An excellent account of the stability and incompatibility of various 
vitamins has been given by Macek [45]. The following is a brief summary 
of the most pertinent aspects as they relate to chewable tablets. 
General: Minimum exposure to heat and moisture during processing 
and in final product (around 1%) is highly desirable. Vitamins A, 
B10 B2, B12, C, and pantothenic acid are relatively more unstable.

404 Mendes et a1. 
Example 8: Chewable Multivitamin Tablets 
Vitamin A acetate (Roche) 
Vitamin 0 
Vitamin O
Vitamin E, 50% SO (Roche) 
Ascorbic acid 90% (Roche) 
Folic acid 
Vitamin 8 
(Rocoat 33-1/3%) 
Vitamin 8 
(Rocoat 33-1/3%) 
Vitamin 8 
(0.1% SO-Roche) 
Niacinamide (Rocoat 33-1/3%) 
Ferrous fumarate, coated 
Pharmasweet Powder (Crompton & Knowles) 
Natural orange flavor 5.0. (Crompton & 
Color Orange No. 53182 (Crompton & 
Magnesium stearate 
Total weight 
mg Itablet Equivalent to 
12.50 5000 IU 
0.58 400 IU 
33.00 15 IU 
67.00 60 mg vito C 
0.40 0.4 mg 
5.20 1. 7 mg 
6.00 2.0 mg 
6.00 6.0 l.lg 
60.00 20.0 mg 
q..s , 
1. Mix Vitamins 01, O2, folic acid, and 8 12 with niacinamide for 15 min. 
2. To 1 add vitamins A, E, ascorbic acid, 82. B6' ferrous fumarate, 
small portion of Emdex, and mix thoroughly for 15 min. 
3. To 2 add remaining Emdex, flavor, and Pharmasweet and mix for 
10 to 1S min. 
4. Add color to 3 and blend thoroughly until it is evenly distributed. 
5. Add magnesium stearate to 3, blend 5 min, and compress. 
Source: Ref. 27. 
Comments: Common practice for multivitamin preparation is to use wet 
granulation in order to process the large amount and number of multiple 
active ingredients. This example demonstrates the feasibi lity of using 
direct compaction-in this case based on Emdex-despite the presence of 
11 actives.

Example 9: Vitamin C Chewable Tablets (250 mg) 
Ingredients A B C 
Sodi um ascorbate (SA- 99) 1 170.5 170.5 170.5 
Ascorbic acid (C-97) 1 103.5 103.5 103.5 
336.0 sucrose 
Compressible natural 
389.8 sugar 
Crystalline sorbitol 335.3 
Sodium saccharin 0.7 
FD&C Yellow #6 Lake (iet-rn ilied) 2.0 2.2 2.0 
Flavoring 5.0 5.5 5.0 
Magnesium stearate 3.0 4.0 3.0 
Total 620.0 675,5 620.0 
1Takeda Chemical Industries. 
2Di-Pac, Amstar Corp. 
3Sweetrex, Edward Mendell Co. 
Actives, lake, flavor, and sweeteners are mixed for 25 min 
in a P-K blender. Magnesium stearate is screened, added, 
and blended for an additional 10 min. 
Source; Ref. 44. 
Example 10: Ascorbic Acid Chewable Tablets 
(250 mg) 
Ingredients mg Itablet 
Ascorbic acid (10% excess) 
Ethocel 7 cps, 10% in isopropanol 
Sta-Rx 1500 
Sodi um sacchari n 
FD&C lake 
Magnesium stearate 
1. Granulate the ascorbic acid with the ethylcellulose 
in isopropanol in a planetary mixer. 
2. Dry overnight at 50oC; screen through a 
3. Add the NuTab and Sta-Rx 1500 and mix 
for 15 min in a P-K blender without 
4. Add the sodium saccharin, lake, flavor, and 
magnesium stearate, previously premixed 
and screened. 
5. Blend for 5 additional min.

406 Mendes et at 
Coating vitamins individually or together in compatible groups is 
desirable to minimize incompatibilities. 
Vitamin A: Sensitive to oxidation. Palmitate and, more commonly, 
acetate esters coated with gelatin or gelatin-starch are used 
(Crystalets and Beadlets). The all-trans isomer is most active 
biologically but chemical assay will not differentiate it from the 
other isomers. 
Vitamin B1: Sensitive to oxidation and reduction. A pH environment 
above 3 to 4 is undesirable. Thiamine mononitrate is preferred 
since incompatibility with pantothenic acid is diminished by the 
nitrate salt. Pentothenyl alcohol or calcium pantothenate and 
thiamine mononitr-ate are preferred. Coated vitamin is preferable. 
Vitamin B2: Sensitive to light. The riboflavin base of 51-phosphate 
sodium salt is used. Coated vitamin is preferable. A pH in the 
alkaline range is undesirable. Protection from red ucing agents is 
Vitamin C: Very sensitive to oxidation, and is a strong reducing 
agent. Presence of copper or iron increases oxidation rate. 
Sodium ascorbate or a mixture of the salt and free acid may be 
Vitwnin B12: Susceptible to loss of activity by reducing agent (e. g. , 
ascorbic acid). A pH environment of 4 to 7 is optimal. Other 
detrimental factors could be ferrous salts, decomposition products 
of thiamine, and some flavors. Vitamin B12 resin complex, 1% 
(Stablets) or 0.1% in gelatin concentrate is preferred. 
Pantothenic acid: The acid itself is not used. Calcium salt is preferred 
since it is a soluble crystalline powder as opposed to the 
acid or the alcohol, both of which are viscous oils. The salt and 
the acid are sensitive to acid, base, and heat. The optimal pH is 
6. The incompatibility with thiamine is discussed above. 
Davis [46] prepared vitamin C and multivitamin formulations to evaluate 
coarse powder sorbitol as a chewable tablet base. Of concern was the 
potential of hardness gain and attempts to ameliorate the problem through 
the incorporation of various starch excipients. He found that the inclusion 
of 5 to 8% pregelatinized modified cornstarch (Dura Gel DGD) helped maintain 
proper hardness during storage without affecting hygroscopicity Iweight 
A. In-Process Organoleptic Evaluation 
Organoleptic evaluation takes place at various stages in the development of 
a chewable tablet. These follow in sequence at various stages as shown 
in Table 9. 
The evaluation of the drug substance itself (stage 1) has already been 
briefly discussed in this chapter. Stages 1. 2, and 3 are generally carried 
out by the formulating pharmacist either alone or in collaboration with a 
small taste panel within a development laboratory. Since organoleptic 
evaluation is subjective in nature, it is necessary to have the terminology. 
comparative standards, and test conditions well defined and controlled for

Chewable Tablets 
Table 9 Various Stages of Organoleptic Evaluation 
Stage of organoleptic 
Evaluation of the drug 
substance itself 
Evaluation of coated (e. g. , 
granulated) or treated 
(e. g., adsorbed) drug 
Evaluation of unflavored 
baseline formulation 
Evaluation of flavored 
baseline formulation 
Final selection and product 
accept ance test 
It involves: 
Characterization and comparison 
of the drug substance in 
an absolute sense or against a 
known reference standard 
Comparison against the pure 
drug as well as different 
coatings or treatment 
Comparisons among different 
vehicles, proportions of 
vehicles, or other formulation 
variables (except flavors) in 
presence of coated or treated 
Comparison among different 
flavored formulations 
Comparison between two "topcandidate" 
form ulations and lor 
a competitive product 
meaningful results. For example, Borodkin and Sundberg [47] evaluated 
methapyrilene, dextromethorp han. ep hedrine, and pseudoep hedrine for 
their basic bitterness (stage 1), followed by the taste comparison of these 
drugs after adsorption onto a polycarboxylic acid ion exchange resin 
(stage 2). The resin adsorbates were further coated with a 4: 1 ethyl 
cellulose-hydroxypropylmethylcellulose polymer at various concentrations 
of coatings. Comparisons were made against pure drug, adsorbed drug, 
and adsorbed drug with variable coating percentages (stage 2). The 
coated adsorbates were blended with other tableting excipients, such as a 
sweetener, magnesium stearate, and mannitol vehicle in standard proportions 
(I, e., baseline unflavored formulations). These formulations. in 
tableted forms. were compared with the respective coated and uncoated 
adsorbates and the pure drugs (stage 3). For comparative quantitation, 
caffeine solutions were chosen as the standards for bitterness intensity on 
a scale of 0 to 3, 3 being strong bitterness (0.2% caffeine), 2 being 
moderate (0.1% caffeine), 1 being slight (0.05% caffeine), x being threshold 
(0.001% caffeine), and 0 being no taste (water) [47]. 
The panel in the caffeine-dextromethorphan study consisted of at least 
seven members of each sample, using a so-called time intensity method in 
which the sample equivalent to one dose is held in the mouth (or chewed, 
in the case of tablets) for 10 sec. The readings are taken immediately 
and at several intervals over a period of 15 min. The type of quantitative 
information generated in the study is shown, using the bitterness scale of 
o to 3, in Table 10.

408 Mendes et al. 
Table 10 Bitterness Evaluation of Dextromethorphan HBr at Various 
Stages of Product Development 
Form of Degree of bitterness after time 
dextromethorp han 
hydrobromide 10 Sec 1 min 2 min 5 min 10 min 15 min 
Uncoated drug >3 >3 2.5 1.5 1 0.5 
Uncoated adsorbate 2 2 1.5 0.5 x 0 
Uncoated adsorbate 1.5 1.5 1. 5 to 2 1.5 1 x 
in tablet 
Powder adsorbate 0.5 x to 0.5 x x 0 0 
coated with 25. 4% 
Adsorbate powder x to 0.5 0.5 x x x 0 
coated with 25. 4% 
polymer in tablet 
Source: From Ref. 47. 
Table 10 serves to illustrate that, although adsorption does reduce 
bitterness, it is necessary to reduce it further by polymeric coating. The 
uncoated adsorbate in the mannitol- based tablet formulation is much more 
bitter than the tablet made with the coated adsorbate. Another interesting 
observation is that there is apparently little or no difference in the bitterness 
of the coated adsorbate in powder form or when compounded with the 
mannitol- based tablet formulation under the conditions of this evaluation. 
A possible explanation is the fact that the coating is disrupted to some 
extent during compression as well as during mastication of the tablet, thus 
exposing the bitter substance. At this point in development the formulator 
should overcome the residual bitterness by a proper choice of flavors. 
Stages I, 2, and 3 (Table 9) involve the evaluation of the extent of taste 
masking in unflavored preparations while stages 4 and 5 involve the evaluation 
of the preferences among flavored baseline formulations. Although the 
extent of testing is a matter of relative quantitation, and hence relatively 
easy to control, the preference testing is a matter of determining individual 
choices of the panelists. which are subject to significant variability. Thus. 
the selection of the panelists and their number are important factors in 
establishing a preference-testing panel. For example, during the final 
selection and product acceptance test (stage 5, Table 9) of a chewable 
multivitamin product for children, as many as 100 or more panelists representing 
the Ultimate consumer age group are necessary [41]. Certain additional 
guidelines are noteworthy insofar as taste panels are concerned. 
1. Conditions of testing must be optimized. These conditions include 
the temperature of the sample (e. g., a freshly removed sample 
from a refrigerator or a 40C stability oven win taste significantly

Chewable Tablets 409 
different than the same sample at 25C). a well-ventilated room. 
the absence of distracting noise. and the presence of muted light. 
2. Rapid succession (frequency) of samples quickly fatigues the 
tongue, thus leading to erroneous conclusions. Sampling time 
should be standardized. 
3. A certain "washout" treatment between samples is often recommended 
(e. g. fixed quantity of water with a saltine cracker). 
4. Any unsolicited comments (e. g . chalky. cooling, delicious. 
tangy. gritty) must be recorded. 
5. The panelists should not be grouped but rather should be individualized 
(e. g. in a compartmented room), and the panel should 
reflect the age and sex of the eventual consumer [41]. 
6. Whenever possible, the dosage should be close to the actual intended 
7. It is imperative that a statistician be involved in the design of 
the test protocol. 
8. For chewable vitamin formulations, no more than two samples 
should be given for comparison, and the order of submission 
should be randomized. 
9. It is best to have panelists who have had no prior experience 
with the test products. 
10. Questions concerning acceptance, rejection, preference, and 
similar items should be reserved until after the panelist has 
made the selection. 
B. Chemical Evaluation 
This aspect involves total drug assay and content uniformity testing if 
Assay for Drug Content 
A suitable analytical method (chromatographic, titrimetric, spectrophotometric, 
etc.) is used to determine the active drug content on a representative 
sample (usually an aliquot of 20 randomly selected tablets after pulverization). 
The recovered amount of active drug is then expressed as percent 
of labeled drug content. The obtained value of drug content should be 
within established limits. 
Dosage Uniformity 
This test is done to ensure that the batch of tablets is uniform as to the 
content of active ingredient per dosage unit within specified limits. If the 
drug level in the dosage unit is high. then a weight variation test is sufficient 
to indicate uniformity of drug content in the dosage units. However. 
if the drug dose is low compared to the weight of the dosage unit. as is 
usually the case with chewable tablets where provision is made for a large 
use of sweet excipients, coating agents, and/or for taste masking and 
mouth-feel. then individual assay of the given number of randomly selected 
dosage units is done to obtain drug content in the various samples tested. 
The coefficient of variation gives an indication of the uniformity or nonuniformity 
of the tested units in the batch. The U.S. Pharmacopeia (USP) 
gives in detail the protocol and acceptance criteria for the determination of

410 Mendes et al, 
dosage form uniformity for conventional tablets, which is also applicable to 
chewable tablets. 
In Vitro and In Vivo Evaluation (Antacid Tablets) 
Since antacids represent a sizable proportion of chewable tablets, a description 
of their in vitro and in vivo evaluation is considered important. 
Antacid tablets are meant to exert their effect in the stom ach , and 
hence the gastric bioactivity is of prime concern. Analogous to the rate 
and extent of bioavailability, an antacid preparation should be evaluated 
for its rate and extent of action and total acid-consuming capacity during 
in vitro and in vivo testing. 
In the United States a product may be labeled as an antacid only if it 
meets a prescribed test [48], which in principle is as follows: the tablets 
are comminuted to a particle size between 20 and 100 mesh (U.S. standard 
sieve), and an accurately weighed amount equivalent to the minimum labeled 
dosage is mixed with 40 ml of water under standard conditions for 1 min. 
Then 10 ml of 0.5 N HCI is added to the slurry and stirred under fixed 
conditions for 10 min. The pH of the mixture is then read. If the pH is 
below 3.5, the product is not permitted to be labeled as an antacid. 
The U. S. Food and Drug Administration (FDA) has also defined the 
minimum requirement for an antacid product in terms of its acid-neutralizing 
cap acity [48]. 
The sample is prepared in essentially the same manner as described 
above-up to the addition of water. A standard volume of 1. 0 N HCI is 
then added and mixed for a fixed period of time, immediately followed by 
the back-titration of the excess acid with 0.5 N NaOH to a stable pH of 3.5. 
The total number of milliequivalents (mEq) of the acid neutralized by the 
product under test are then calculated. The requirement is that no product 
shall be marketed with an acid-neutralizing capacity below 5 mEq. The 
capacity is expressed in terms of the dosage recommended per minimum 
time interval or, if the labeling recommends more than One dosage, in terms 
of the minimum dosage recommended per minimum time interval. For compliance 
purposes, the value determined by this test at any time (during 
the expiration period of the product) must be at least 90% of the labeled 
While the determination of the acid-neutralizing capacity is an important 
in vitro parameter, the onset (rate) and duration of the neutralizing 
action are equally important. Smyth et al , [49J, for example, studied 
these aspects and their correlation with in vivo results in human subjects 
using two in vitro methods known as Bachrach titration [50] and modified 
Beekman procedure [51]. In principle, both methods are based on the 
neutralization of an acid by the antacid preparation under study. 
Using Bachrach titration [50]. Smyth et al. [49] showed that a fixed 
quantity of a tablet powder in a fixed volume of water had an initial pH of 
8.66, which was initially lowered to 3.5 (onset) within 60 sec by the addition 
of 1.1 ml of 0.8 N HCl. To maintain the pH at 3.5 (for at least 30 
sec), 4.9 ml of the acid (capacity) was required, and the endpoint was 
reached at 13.8 min (duration). Further addition of the acid resulted in 
the lowering of the pH below 3.5 (i. e., capacity exhausted). These data, 
when compared to those for suspension containing an equivalent stoichiometric 
quantity of the active ingredient, indicated that the suspension 
required a larger volume of the acid (1. 8 ml) for initial onset, and that it 
took a longer time (75 sec) for the acid to initially bring the pH to 3.5.

Chewable Tablets 411 
Further, more acid (5.4 ml) was needed to reach the endpoint. Thus all 
observations seemed to lead to the conclusion that apparently the suspension 
was somewhat superior to the chewable tablets containing the same active 
ingredient. A similar statement is found in the literature elsewhere [52] 
(i. e., that antacids in tablet form are less effective than liquid or powdered 
preparations). In another in vitro experiment. however, Smyth et al, [49] 
found the two dosage forms to be comparable. 
In the same study [49], the second in vitro method, known as the 
modified Beekman procedure [51], consisted of adding the antacid preparation 
(tablet. powder, or suspension) to a 50-ml volume of 0.1 N HCI at 
37C. With continuous agitation, more acid was added (by a pump) continuously 
at a fixed rate, and the ant acid- acid mixture was continuously 
removed at an equal rate to keep the overall volume constant. The pH 
was continuously monitored with an electrode dipped in the antacid-acid 
mixture. The results are shown in Table 11. 
The Bachrach method estimated the suspension to be slightly better in 
onset and capacity whereas the Beekman procedure indicated the two to 
be equal. Smyth et al. [49] showed by in vivo tests in human subjects 
that no significant difference existed between the two. The study in 
human subjects involved a controlled set of conditions of fasting, standard 
meals at specific times, a fixed volume of water I and time dosing. The 
criterion used for evaluation was the monitoring of the actual intragastric 
pH by means of a device known as a Heidelberg capsule [53,54]. The precalibrated 
capsule is capable of sending a radiotelemetric pH recording 
signal from within the stomach. The capsule is attached to a nonwettable 
surgical string and swallowed. The position of the capsule is controlled 
by the length of the string. The details of the test conditions are not 
within the scope of this chapter but it is sufficient to say that a fairly 
accurate means of intragastric pH monitoring has been clearly demonstrated 
and correlated with the in vitro data. The study discussed above describes 
the methods of comparison between antacids in two different dosage forms; 
however, the principles are equally applicable to the comparison of two or 
more chewable antacid tablet formulations. 
C. Physical Evaluation 
The physical evaluation involves the following: 
(1) Tablet physical appearance. As one of the quality control procedures, 
tablets should be inspected for smoothness, absence of cracks, 
Table 11 In Vitro Evaluation of Antacids by Modified Beekman Method 
Speed Capaci ty for 
Onset (time in min Duration buffering 
(time to reach to reach (time in min (maximum pH 
Preparation pH 3 initially) maximum pH) above pH 3) reached) 
Suspension Immediate 5.0-7.5 26.5- 27. 5 5.62-5.7 
Tablet Immediate 2.5-5.0 25.5- 28.0 5.72-6.1 
Source: Ref. 49.

412 Mendes et al. 
chips, and other undesirable characteristics. If the tablets are colored I 
this would include examination for mottling and other evidence of nonuniform 
color distribution except where they are used intentionally. A suitable 
magnifying glass may be used to appropriately view the samples. 
(2) Hardness. The hardness test is performed to provide a measure of 
tablet strength. Tablets should be hard enough to withstand packaging 
and shipping but not so hard as to create undue difficulty upon chewing. 
Tablet hardness is determined using equipment from various suppliers 
that measure the force needed to break up the tablets. Usually a random 
sample of tablets (10 to 20) that have been allowed to age for at least 24 
hr after production (to ensure equilibration of stresses and forces within 
the tablet) are individually tested and the mean hardness value determined. 
The coefficient of variation is also determined. High variations in tablet 
hardness values are not unusual although abnormally high coefficients of 
variation may indicate excessive weight variation, blend nonuniformity, 
poor tooling control, etc. 
(3) Friability. The friability test gives an indication of the tablets' 
ability to resist chipping and abrasion on handling during packaging and 
shipping. Usually for conventional tablets a friability value of 1% or less 
is desirable, while for chewable tablets (due to the lower hardness of the 
tablets) friability values of up to 4% are acceptable. The friability test is 
done using a Roche friabilator or its modification. 
In this test at least 20 tablets weighing at least 6 g are accurately preweighed. 
The tablets are rotated in a Roche friabilator through 100 revolutions, 
dedusted, and reweighed. The percent friability is determined from 
the weight loss. 
(4) Disintegration. This test initially may not appear appropriate for 
chewable tablets as these tablets are to be chewed before being swallowed. 
However, patients, especially pediatric and geriatric, have been known 
to swallow these chewable dosage forms. This test would thus indicate 
the ability of the tablet to disintegrate and still provide the benefit of the 
drug if it is accidentally swallowed. Tablets should preferably pass the 
USP disintegration test for uncoated tablets. 
(5) Dissolution. The dissolution test measures the rate of dissolution 
of the drug from the dosage form in vitro. It is usually expressed as the 
extent of dissolution (percent of drug content) of the drug occurring after 
a given time under specified conditions. This test is necessary to help in 
the prediction of the behavior of the drug in the dosage form after ingestion 
and as a quality control tool for checking batch-to-batch uniformity. 
Chewable tablets should preferably be tested in two forms: intact (in case 
the dosage form is accidentally SWallowed) and partially crushed (to simulate 
chewing). The USP describes the procedures for routine dissolution 
testing. Apparatus 1 (rotating basket) of the USP protocol may be appropriate 
for testing of partially crushed dosage forms while apparatus 2 
(rotating paddle) may be suitable for testing whole tablets. This dissolution 
test on the two forms mentioned above is particularly necessary in 
chewable tablets formulated from active drug present in a matrix or where 
the active drug has been coated using different methods and means for 
bitter taste masking. Such treatment may alter the dissolution rate of the 
untreated drug. 
As discussed earlier, many taste-masking applications involve some 
sort of coating, barrier, or adsorption to mask the taste of the drug.

Chewable Tablets 413 
As compared to regular (swallowed) tablets, such applications in chewable 
tablets may result in a somewhat delayed release of the drug in the 
stomach. The formulator should be careful to ensure that a proper 
balance is achieved between the desired levels of taste masking and the 
rate and extent of release in the stomach. Proper judgment is necessary 
to determine how much and what sort of testing is necessary to ensur-e 
that this balance is achieved. 
D. Stabi Iity Testi n9 
Stability testing of dosage forms or drug products is carried out to evaluate 
time-dependent changes, if any, occurring with the dosage form. Stability 
testing may be either accelerated or real time under ambient conditions. 
Accelerated stability testing is used to predict quickly potential 
changes that may occur in a product. However I it must be pointed out 
that results obtained under stress conditions may not be obtained under 
ambient conditions. Accelerated storage conditions include high temperatures, 
high relative humidities, and high light intensities. There are 
three areas of major concern in the stability testing of chewable tablets; 
organoleptic, chemical, and physical stability. 
Data obtained from chemical evaluation of the tablets at elevated temperature 
and humidity stress conditions are considered most useful. The 
Arrhenius equation relates kinetic rates at different temperatures and, 
when appropriate, can be used to extrapolate and thus obtain the expected 
reaction rate at room temperature I which would be used to tentatively 
determine the stability of the product under test or for overage determination 
as with vitamin tablets. The FDA's Stability Guidelines [54} describe 
in considerable detail the appropriate conditions for conducting stability 
studies in order to achieve the necessary goals and objectives. The stability 
testing of chewable tablets would include all tests for conventional 
tablets plus tests unique to chewable tablets. 
Fundamental to all types of stability evaluations in flavored chewable 
tablets is the fact that the flavor is a complex mixture, often eonsiating 
of as many as 50 or more ingredients. The picture is further complicated 
by the fact that the flavors are then incorporated in tablets where they 
come in contact with the active and inert ingredients. A flavored chewable 
tablet is therefore prone to many problems-with greater potential for stability 
problems than its nonflavored , regular (swallowed) counterpart. 
Some generalization of the flavor composition is necessary to an understanding 
of its implication on the stability of a flavored tablet. A flavor 
may consist of, for example, a combination of the following; 
Alcohols (e.g., ethyl alcohol, butyl alcohol, glycerol) 
Aldehydes (e.g., benzaldehyde, butyraldehyde, citral, and vanillin) 
Ketones (e. g., methyl amyl ketone) 
Esters (e. g., ethyl acetate. butyl butyrate, methyl salicylate) 
Essential oils (e. g., anise oil, lemon oil, orange oil) 
Plant extractives (e. g., lovage, fenugreek) 
Acids (e.g., citric acid, tartaric acid) 
Carbohydrates (e. g., sugar, dextrose, molasses-used mainly as 
Others (e. g., silica gel-used as a carrier for adsorbed flavors)

414 Mendes et al. 
These flavor compounds are either very reactive (e. g. aldehydes). 
volatile (e. g . the essential oils and alcohols). or prone to hydrolysis 
(e. g. I the esters), and. as such, formulations containing them must be 
carefully evaluated during the stability study. 
The formulator is generally responsible for the organoleptic evaluation 
during a given stability study. Logically and most desirably. a preliminary 
stability study should have been conducted after the evaluation of a flavored 
baseline formulation. and before the final selection, acceptance. and 
final stability testing. This ensures that the candidate products are marketable. 
pending the final selection. Conducting the final stability evaluation 
after the final selection and acceptance testing may be risky since a 
well-designed selection and testing trial is generally expensive and time 
consuming, requiring considerable paper work to organize. 
Any stability study involves the comparison of an initial value (or an 
initial observation) with subsequent readings (or observations) taken at 
various time intervals under various conditions of storage. This definition 
poses a problem unique to the stability study of chewable tablets especially 
with regard to the organoleptic properties, since it requires the formulator 
to have an accurate initial reading of the organoleptic characteristics so 
that the future stability samples can be compared against it. This is best 
accomplished by comparing the stability samples with a freshly prepared 
lot of the same formula. The judgment of the person making such a comparison 
is of paramount importance in the stability study. since it is impractical 
to have a taste panel evaluate the flavor at each stability checkpoint. 
The organoleptic evaluation should be an integral part of a stability 
protocol, and should be conducted and recorded at reasonable intervals in 
the same manner as other physical and chemical results are recorded. 
Other tests in the stability program would Include 
1. Active drug content determination using a validated stability indicating 
assay method. 
2. Change, if any. in physical characteristics of the tablets-mottling 
of colored tablets. color migration, appearance of spots on tablet 
surfaces, crystallization of active drug on tablet surfaces, odor 
development. etc. 
3. Changes in tablet hardness. friability. dissolution rate and/or 
extent of dissolution, increase in disintegration time. 
4. Moisture content of tablets-moisture pickup by tablets may lead 
to soft tablets that crumble and are gummy upon chewing. If 
tablets lose moisture, they may become brittle. leading to increase 
in their friability. Also, the hardness of the tablet may 
5. Stability of the coating systems-the polymers used in taste-m asking 
processes should not degrade, leading to exposure of the active 
drug particles. The coat and matrix should also be stable, thus 
ensuring taste protection. 
6. Stability of the colorants-the color of colored tablets should not 
fade or shift with time. Color stability testing would include 
methods such as tristimulus matching with standards and with 
initial values.

Chewable Tablets 
Chewable tablets represent an example of a specialized tablet type specifically 
designed to be chewed prior to swallowing. They are primarily used 
for children's vitamin supplements and cough/cold/analgesic products, and 
for adult antacids. Despite their potential appeal to the adult population 
for general medicinal use, the chewable tablet form has not been exploited 
in the marketplace for such applications. 
Chewable products must be formulated in such a way as to provide acceptable 
taste and mouth- feel despite the usually bad taste of most drugs. 
Consequently, an even greater than usual challenge is presented to the 
formulating pharmacist. This challenge should be looked on as an opportunity 
to demonstrate a fUll range of knowledge and skills necessary to 
bring a less-than-commonplace product to market acceptance. 
1. C. H. Best and N. B. Taylor, Physiological Basis of Medical Practice 
(J. R. Brobeck, ed . ) , 9th ed , , Williams and Wilkins, Baltimore, 
1973, Chap. 5, Sec. 8. 
2. J. A. Bakan and F. D. Sloan, Drug Cosmo Ind., 110(3): 34 (March 
3. L. J. Luzzi, J. Pharm. Sci., 59:1367 (1970). 
4. B. Farhadieh, U.S. Patent 3,922,379 (1975). 
5. J. A. Bakan, in Theory and Practice of Industrial Pharmacy (L. 
Lachman, H. Lieberman, and J. Kanig, eds.), 3rd ed . , Lea and 
Febiger, Philadelphia, 1986. 
6. N. N. Salib, Pharm. Ind., 34:671 (1972). 
7. Roche Chemical Div, , Trade Literature no. STP641, HoffmannLaRoche, 
Nutley, NJ. 
8. W. L. Chiou and S. Riegelman, J. Pharm. Sci., 60:1281 (1971). 
9. R. M. Wheaton and A. H. Seamster, A basic reference on ion exchange, 
form no. 177-194- 86, Dow Chemical Company, Midland, MI. 
10. Amberlite IRP-64 Technical Bulletin no. 5086J /232, Rhom and Haas 
Company, Philadelphia, PA, 1983. 
11. A. P. Granatek and M. P. DeMurio, U.S. Patent 3,459,858 (1969). 
12. D. Hoff and K. Bauer, U.S. Patent 3,872,227 (1975). 
13. W. Saenger, Angew. Chem. Int. Ed. Eng!.. 19:344 (1980). 
14. T. Higuchi and I. H. Pitman, J. Pharm. su., 62: 55 (1973). 
15. United States Pharmacopeia XXI/National FormUlary XVI, United 
States Pharmacopeial Convention, Rockville, MD, 1985. 
16. The Merck Index X, Merck and Co., Rahway, NJ, 1983. 
17. Handbook of Pharmaceutical Excipients, American Pharmaceutical Association, 
Washington, DC, 1986. 
18. R. W. Mendes and S. B. Roy, Pharm. Tech., 2(9):61 (1978). 
19. R. F. Shangraw, J. W. Wallace, and F. M. Bowers, Pharm. Tech 
5(9):69 (1981). 
20. S. D. UkoNne and R. W. Mendes, Pharm. Tech., 6(11):104 (1982).

416 Mendes et aZ. 
21. Hony- Tab Technical Literature, Ingredient Technology Corporation, 
Pennsauken. NJ. 1982. 
22. Mola- Tab Technical Literature. I ngredient Technology Corporation, 
Pennsauken. NJ. 1982. 
23. CrystaFlo Technical Literature. Amstar Corporation. New York, 1985. 
24. Brownulated Technical Literature, Amstar Corporation, New York. 
25. Di-Pac Technical Literature, Amstar Corporation, New York, 1985. 
26. NuTab Technical Literature. Ingredient Technology Corporation, 
Pennsauken. NJ. 1982. 
27. Emdex Technical Literature, Edward Mendell Co . Carmel. NY. 1986. 
28. Sweetrex Technical Literature, Edward Mendell Co . Carmel, NY, 
29. Lactose Technical Literature, Sheffield Products, Kraft, Jnc , , Norwich, 
NY, 1985. 
30. Lactose Technical Literature, Foremost Foods Co., San Francisco, CA, 
31. Magnasweet Technical Information, MacAndrews and Forbes Co., 
Camden. NJ. 1987. 
32. Nutrasweet Technical Overview, NutraSweet Co., Skokie, IL, 1986. 
33. The PFC Index, Pharmaceutical Flavor Clinic, Division of Foote and 
Jenks, Camden, NJ, 1986. 
34. Flavor Guidelines for the Pharmaceutical Industry, Food Materials 
Corp., Chicago, IL. 1979. 
35. F. Wesley, Pharmaceutical Flavor Guide, Fritzsche Brothers. Inc; , 
New York, NY, 1957. 
36. All about lake pigments. Technical Bulletin, Warner-Jenkinson 
Company. St. Louis, MO, 1986. 
37. Certified food colors. Technical Bulletin, Warner-Jenkinson Company, 
St. Louis, MO, 1982. 
38. J. R. B. J. Brouwers and G. N. J. Tytgat, J. Piuirm, Sci., 30:148 
(1978) . 
39. C. K. Svensson and T. H. Wiser, Drug Intell. Clin, Pharm., 15: 120 
40. Handbook of Non-Prescription Drugs, 8th ed, , American Pharmaceutical 
Association, Washington, DC, 1986. 
41. R. H. Barry and M. S. Weiss, J. Am. Pharm. Assn., 10:601 (1970). 
42. J. Raymond, U.S. Patent 3,458,623 (1969). 
43. T. L. Fisher and J. F. Cassens, Presentation on Pharmaceutical 
Flavors, Philadelphia Discussion Group, Academy of Pharmaceutical 
Sciences, 1976. 
44. N. Kitamori , K. Hemmi , M. Maeno, and H. Mirna, Pharm. Tech., 6(10): 
56 (1982). 
45. T. J. Macek, Am. J. Pharm., 132: 433 (1960). 
46. J. D. Davis, Drug Cosmo Ind., 128(1):38 (1981). 
47. S. Borodkin and D. P. Sundberg, J. Pharm. ScL, 60:1523 (1971). 
48. Code of Federal Regulations, Title 21. Food and Drugs, Sec. 331.1, 
pp. 132-136. 
49. R. D. Smyth, T. Herczeg, T. A. Whatley, W. Hause. and N. H. 
Reavy-Cantwell, J. Pharm. s, , 65:1045 (1976). 
50. W. H. Steinbert, H. H. Hutchins, P. G. Pick, and J. S. Lazar, J. 
Pharm. ser.. 54: 625 (1965). 
51. S. M. Beekman, J. Am. Pharm. Assn. 49: 191 (1960).

Chewable Tablets 417 
52. R. A. Locock, Can. Pharm. J., 104: 86 (1971). 
53. E. Johannesson, P.-D. Magnusson, N.-D. Sjoberg, and A. SkovJensen, 
Scand. J. Gastroenterol., B:65 (1973). 
54. J. C. McAlhany, Jr., D. R. Yarbrough III, M. G. Weidner, Jr., and 
R. Ravenel. Am. Surg. 35: 836 (1969). 
55. Draft guideline for stability studies for human drugs and biologics, 
Food and Drug Administration, Rockville, MD, 1985.

Medicated Lozenges 
David Peters* 
Warner-Lambert Company, Morris Plains J New Jersey 
Lozenges are flavored medicated dosage forms intended to be sucked and 
held in the mouth or pharynx [1,85]. They may contain vitamins, antibiotics, 
antiseptics, local anesthetics, antihistamines, decongestants, corticosteroids, 
astringents, analgesics, aromatics, demulcents, or combinations 
of these ingredients [2]. The oropharyngeal symptoms which lozenges are 
intended to relieve are commonly caused by local infections and occasionally 
by allergy or drying of the mucosa from mouth breathing. 
Lozenges may take various shapes, the most common being the flat, 
circular, octagonal, and biconvex forms. Another type, called bacilli, are 
in the form of short rods or cylinders. A soft variety of lozenge, called 
a pastille, consists of medicament in a gelatin or glycerogelatin base or in 
a base of acacia, sucrose, and water. Confections (now obsolete) are 
heavily sugared soft masses containing medicinal agents [3]. 
Two types of lozenge bases have gained wide usage because of their 
ready adaptation to modern high-speed methods of product manufacture. 
These two lozenge forms, which will be discussed in detail, include hard 
(or boiled) candy lozenges and compressed tablet lozenges. 
Hard candy is a mixture of sugar and other carbohydrates that are kept in 
an amorphous or glassy condition [4]. This form can be considered a solid 
syrup of sugars generally having from 0.5 to 1. 5% moisture content. 
Essentially, the preparation of hard candy lozenges can be considered 
an art. Many of the formulations used in confectionary manufacturing, and 
the rationale used for solving problem areas, are based on experience and 
intuition rather than scientific deduction. The confectionary equipment 
utilized by the manufacturer of lozenges is suitable for the preparation of 
*Current affiliation: Treworgy Pharmacy, Calais, Maine 

Figure 1 Mixing of flavors and medicinals by hand. Preparation of 1- or 
2-kg laboratory batches enables the formulator to evaluate potential problem 
areas that may develop when flavor or medicament is incorporated into 
hard candy base. (From Ref. 24.) 
candies but is not designed to produce a controlled and reproducible medicated 
candy with close tolerances as to size. weight, and quantity of drug 
concentration per unit dose. The formulator must gain a comprehensive 
knowledge of the physical and chemical qualities of raw materials in the 
product and become familiar with all aspects of candy base production in 
order to prepare a medicated product that conforms to the specifications 
for good manufacturing procedures (Figures 1 and 2). A review of possible 
shelf life problems must be determined through stability testing after 
the product is manufactured. The formulator. in essence, is required to 
bring a scientific approach to an empirical art. 
A. Raw Materials 
Sugar (Sucrose) 
Various grades and types of sugars are av8iJ.able in commerce that may be 
suitable for incorporation into hard candy, but the two with the greatest 
utility are cane and beet sugars [4.80]. 
Sucrose is prepared commercially from sugar cane. beet root, or sorghum. 
The sugar cane is crushed and the juice (amounting to about 80%)

Medicated Lozenges 421 
is expressed with roller mills. treated with lime to clear the syrup and 
then with carbonic acid gas to remove excess lime. The juice is then concentrated 
in vacuum pans until crystallization of sucrose is complete. The 
crystals and the syrup are separated by centrifugation-with the resulting 
syrup (a byproduct) known as molasses. Beet sugar is made by a similar 
process but is more difficult to purify. 
Refined sugar from either raw cane or beet sugars is prepared by dissolving 
the sugar in water. clarifying. filtering, and finally decolorizing 
the solution by treatment with charcoal. The water-clear solution is evaporated 
under reduced pressure to the crystallizing point [5]. 
Cane and beet sugars are now chemically and physically identical and 
therefore cannot be distinguished from each other in the 'refined state. 
At one time. though. there were significant differences in the purity and 
shelf life among products prepared with each type of sugar. Beet sugar 
contained many impurities. producing a final product containing batch-tobatch 
differences in color. The candies had a tendency to grain (exhibit 
sugar crystallization) and pick up excessive moisture. Advances in sugar 
refining have led most manufacturers to indicate that these differences no 
longer exist, with only geographic considerations and availability determining 
which is used. 
Today liquid sugar with a solids content of 67% w/w (Table 1) is used 
almost exclusively in the manufacture of confections, as all continuous candy 
base manufacturing equipment requires a constant supply of sugar syrup 
and corn syrup during cooking. Manufacturers can prepare the syrup as 
Figure 2 Motorized drop-former. Lozenges manufactured in the laboratory 
are suitable for stability evaluation of medicament, flavor, and color 
prior to manufacture of production batches. (From Ref. 24.)

422 Peters 
Table Physical Constants of Sucrose Solutions 
Degrees Degrees Index of Specific Weight (lb) 
Brix (% of Baume refraction gravity of 1 US gal. 
sugar) (modulus 145) at 68F at 68F at 68F 
67.0 36.05 1. 4579 1. 3309 11. 08 
68.0 36.55 1. 4603 1.3371 11.13 
69.0 37.06 1. 4627 1. 3433 11.18 
70.0 37.56 1. 4651 1. 3496 11.23 
71. 0 38.06 1. 4651 1.3559 11.29 
72.0 38.55 1. 4700 1.3622 11. 34 
73.0 39.05 1.4725 1.3686 11.39 
74.0 39.54 1. 4749 1. 3750 11. 45 
75.0 40.03 1. 4774 1. 3814 11.50 
76.0 40.53 1. 4799 1. 3879 11.55 
77.0 41. 01 1.4825 1.3944 11. 61 
78.0 41.50 1.4850 1.4010 11. 66 
79.0 41. 99 1.4876 1. 4076 11. 72 
80.0 42.47 1. 4901 1. 4142 11.77 
Source: The Manufacturing Confectioner. Vol. 70, No.7, July 1970. 
needed from granular sugar or purchase liquid sugar directly from their 
sugar refiners. 
Corn Syrup 
Corn syrups are produced by either acid, enzyme, or acid-enzyme combination 
hydrolysis of cornstarch and are generally available in several 
grades, varying in degree of conversion [dextrose equivalent (DE)] and 
solids content (degrees Baume) [4]. 
The manufacture of all corn sweeteners begins with the hydrolysis 
of cornstarch, a process involving the splitting of the starch molecules 
by Chemical reaction with water. During the process, a thoroughly 
agitated slurry of purified starch granules containing the required amount of 
dilute acid is brought to the desired temperature by the injection of steam. 
A variety of acids will affect the conversion, but in the United States hydrochloric 
acid is used almost exclusively. Time and temperature are varied depending 
on the type of corn sweetener to be manufactured [6]. 
As the reaction progresses, the gelatinized starch is converted first to 
other polysaccharides and subsequently to sugars, mostly maltose and dextrose. 
The sugar content increases and viscosity decreases as the conversion 
proceeds. Complete hydrolysis produces dextrose.

Medicated Lozenges 423 
The hydrolysis of the starch is halted when partially complete-to produce 
corn syrup, the exact degree depending on the type of syrup being 
made. Partial hydrolysis of starch converts part of the starch completely 
to dextrose; the remainder, which is not completely hydrolyzed to dextrose, 
consists of maltose and higher saccharides. The proportions of saccharides 
vary, depending on the extent and method of hydrolysis. 
Two methods of hydrolysis are in commercial use for the production of 
corn syrup-the acid process and the acid-enzyme process. In the latter, 
acid hydrolysis is followed by conversion with an amylolytic enzyme, resulting 
in a syrup with a higher proportion of maltose than can be obtained 
by acid hydrolysis alone. The dextrose/maltose ratio can be varied, within 
certain limits, depending on the type of enzyme used and on the extent of 
t he preliminary acid conversion. 
In the acid hydrolysis process, the hydrolysis is stopped when the reaction 
has reached the desired DE range. by transferring the contents of 
the converter into a neutralizing tank where the pH is raised to the level 
necessary to stop the reaction. The acid acts as a catalyst and does not 
combine chemically with the starch. The acidified product is partially neutralized 
by adding a calculated quantity of sodium carbonate to the solution. 
Fatty substances which rise to the surface are skimmed and then removed 
in centrifuges or by preeoated filters. Suspended solid matter is 
removed by filtering the hydrolyzate in vacuum filters. The filtrate is 
then evaporated to a density of about 60% dry substance. 
After this initial evaporation, the hydrolyzate is passed through either 
bone char or other carbon filters, which causes further clarification and 
decolorization so that the resulting syrup is clear and practically colorless. 
This process partially removes soluble mineral substances, which also can 
be removed by an ion exchange process. 
After final filtration, evaporation is carried out in vacuum pans at relatively 
low temperature to avoid damage to the syrup. The syrup is cooled 
and can be stored or loaded directly in tank cars, tank trucks I steel drums, 
or cans. 
In the production of high-conversion acid-enzyme or dual-conversion 
syrups, acid hydrolysis is carried to a level of 48- 55 DE. The syrup 
then is neutralized, clarified, and partially concentrated, and the enzyme 
added. In other products the acid hydrolysis may be stopped at a level 
as low as 15 DE. When the enzyme hydrolysis has progressed to the desired 
degree, the enzyme is inactivated. Adjustment of the pH, further 
refining, and final evaporation follow as in the production of acid conversion 
syrup. A summary of the corn-refining process is described in 
Figure 3. 
Dextrose Equivalent 
Dextrose equivalent is a measure of the reducing-sugar content of a 
product calculated as dextrose and expressed as a percentage of the total 
dry substance [7,8]. Essentially, the dextrose equivalent is the percentage 
of pure dextrose that gives the same analytical effect as is given by 
the corn syrup. Certain sugars, such as dextrose, maltose, lactose, and 
levulose, are called reducing sugars because when a copper hydroxide solution 
(Fehling's solution) is warmed with these sugars, they react with 
cupric hydroxide to form cuprous oxide. Sucrose is not a reducing sugar; 
thus it does not react with Fehling's solution. Generally, dextrose equivalent 
indicates the degree of conversion in corn syrup. The higher the

"""__ot......... _ ............. 
=~.:;.=':"'..:.-:l~:: ---",.---.-.....-- =::-otC;.."":'a:o"',",=.: ...Il _"'_r.::Slm "" """_..........""."...... 
:"~-::"..:=.:-:.::-.r:: -..,- 
v IU 
~=== ~-- ..- 
1 "r.~ :;;- =~ Ili -- " 
- I-l" ~::;., _._J__~. ~ 
~ -_. I __ 9> _ 
11~ t:.~  -- ..- , i I t r i _.._ .... _ .a_ _ ---- ..- "- --- 
/' (;0001ATf!
Figure 3 The corn refining process. (Corn Refiners Association, Inc., '''ashington, 
(l) .... (l) 

Medicated Lozen gB8 425 
dextrose equivalent, the further the conversion has been carried out, resulting 
in less of the higher sugars (maltotriose and maltotetrose). 
The classes of corn syrups categorized as to degree of conversion [8] 
Low-eonversion corn syrup 
Regular conversion corn syrup 
Intermediate-conversion corn syrup 
High-conversion corn syrup 
Extra high-conversion corn syrup 
20-38 DE 
38-48 DE 
48-58 DE 
58-68 DE 
68-99 DE 
100 DE 
A typical analysis of corn syrup with representative carbohydrate composition 
and physical and chemical characteristics is included in Table 2. 
Physical Characteristics 
Corn syrups with 42- 43 DE are called normal corn syrups; those with 
37- 38 DE, low-dextrose-equivalent corn syrups; and those with 58-62 DE, 
high-dextrose-equivalent corn syrups. Regular- or low-conversion dextrose 
equivalent corn syrups are widely used in hard candy. For caramels, lowdextrose-
equivalent syrup is preferred because it prevents the product 
from "flowing" in the cold state because of the high viscosity that low-dextrose-
equivalent corn syrups impart to products to which they are added. 
The high viscosity prevents the caramel from losing its shape when the 
product is stored at elevated temperature or high-humidity conditions. 
High-dextrose-equivalent corn syrups are generally used for filling where 
a low-viscosity and higher sweetness medium is required. Since the introduction 
of enzyme conversion, corn syrups can be varied to best suit their 
application. The properties and functional applications of corn syrups 
based on degree of conversion may be described as follows [6]. 
Browning reaction. The typical brown color that candy base may develop 
during cooking results from a reaction between reducing sugars and 
proteins (Maillard reaction). As the corn syrup conversion continues. more 
reducing sugars are produced. The higher dextrose equivalent syrups are 
more prone to darkening. Some reducing sugars are more active than 
others. For example, dextrose is more reactive than maltose. Therefore, 
the more highly converted products containing maltose are selected in preference 
to the dextrose-containing syrups. Fructose reacts more readily 
than dextrose and will give a greater amount of browning than dextrose 
at the same solids level. 
Fermentability. Yeast-raised goods, particularly bread, require fermentable 
sugars to serve as food for the yeast, and also some residual 
sugars to give good crust color and add a mild sweetness to the finished 
product. Because fermentable sugars increase with dextrose equivalent 
level, the high-DE, dextrose-rich corn syrups are always utilized in making 
yeast-raised products with crystalline dextrose as the ultimate ingredient. 
Foam stabilizer. Because the lower dextrose equivalent syrups have 
a greater ability to retain incorporated air, they are always chosen as the 
best foam stabilizer.

426 Peters 
Table 2 Typical Analysis of Various Corn Syrup Grades 
Representative carbohydrate composition 
Degree of conversion Very low Regular Regular 
Type of conversion Acid- Acid Acidenzyme 
Dextrose equivalent (%) 26 35 43 
Fermentable extract (%) 23 32 42 
Dextrose (monosaccharides) (%) 5 14 20 
Maltose (disaccharides) (%) 14 12 14 
Maltotriose (trisaccharides) (%) 14 11 12 
Higher saccharides (%) 67 63 54 
Representative chemical and physical data 
Baume at 100F (degrees 42 43 43 
Total solids (%) 77.5 79.9 80.3 
Moisture (%) 22.5 20.1 19.7 
pH 5 5 5 
Acidity as HCI (%) 0.015 0.015 0.015 
Viscosity (poises at 100F) 220 220 125 
Boiling point (OF) 222 226 227 
Weight (lb gal at 100F) 11. 70 11. 81 11.81 
Percentage ash (sulfated) of resin-refined corn syrup I less than 0.02%. 
Percentage ash of vegetable-carbon refined corn syrup I 0.3% 
Source: A. E. Staley Manufacturing Co. I Decatur, Illinois (Tech. Data 
Sheet No. 110). 
Freezing point depressio/ and osmotic pressure. Because freezing 
point depression and osmotic pressure are directly related to the number 
of molecules present, the highest dextrose equivalent products give the 
greatest freezing point depression and the highest osmotic pressure. 
Hygroscopicity. The more highly converted syrups have the greatest 
ability to take up water and the low-conversion products the least. If a 
base product for preparing a dry powder with low hygroscopicity is desired, 
then the lowest dextrose equivalent products are used, sometimes 
extending below the 20-DE range into the maltodextrins.

Medicated Lozenges 427 
Regular Intermediate High High Very High 
Acid- Acid Acid- Acid- Acidenzyme 
enzyme enzyme enzyme 
42 54 64 64 68 
58 54 76 76 79 
7 30 39 39 40 
34 18 33 33 39 
27 13 12 12 4 
32 39 16 16 17 
43 43 43 44 43 
80.5 81. 0 81. 8 83.8 82.0 
19.5 19.0 18.2 16.2 18.0 
5 5 5 5 5 
0.015 0.015 0.015 0.015 0.015 
125 75 55 155 55 
227 229 233 234 233 
11.81 11.81 11. 81 11. 93 11. 81 
Nutritive solids. Since the caloric value of starch hydrolyzates is 
based primarily on carbon content, there is no significant difference among 
the various corn syrups when nutritive value is based on solids content. 
If a controlled rate of assimilation is required for specialty applications, 
such as infant foods, the lower converted products with lower rates of 
assimilation are used. In a special application, there could be preference 
for a corn syrup containing dextrose, maltose. or fructose. 
Control of sugar crystallization. In the preparation of hard candies. 
control of the number and size of sugar crystals is required. The higher

428 Peters 
polysaccharides of the low converted corn syrups are effective agents for 
this purpose. By selecting syrups with the correct higher polysaccharide 
content and distribution, control of crystallization can be obtained. 
Sweetness. Fructose is sweeter than dextrose, which is sweeter than 
maltose, which is sweeter than higher polysaccharides. Since the sugars, 
fructose, dextrose, and maltose are all reducing sugars, the higher dextrose 
eq uivalent corn syrups are generally sweeter than the lower dextrose 
equivalent products. However. at any dextrose equivalent level, the corn 
syrup containing a given amount of fructose will be sweeter than a syrup 
containing an equal quantity of dextrose or maltose. Where sweetness is 
the major functional property desired, the high-dextrose-equivalent corn 
syrups, especially those containing fructose, should be selected. 
Viscosity. This property is basically dependent on the average molecular 
size. The most viscous syrups are the lowest dextrose equivalent 
Miscellaneous. Corn syrup is transported from the manufacturer to 
customers or to distribution points in rail tankers as a thick, viscous, 
water-white syrup. The tankers are usually insulated to maintain the temperature 
of the syrup at 90-1400F. depending on the type of syrup being 
shipped. A summary of the physical characteristics available with various 
corn syrups appears in Figure 4 [61. 
Degrees Baume 
Corn syrups are sold on a Baumd basis, which is a measure of specific 
gravity or dry substance content [8J. Since corn syrups are viscous at 
room temperature, Baume determination is made at 140F (60C) with an 
arbitrary correction of 1. 000 Baume added to the observed reading to correct 
the value, which would be reported at 100F (37. 7C) . This is called 
commercial Bawne [9J. Specific gravity is an important consideration when 
choosing a grade of corn syrup (43 Baume corn syrup having about 20% 
water, 45 Baume about 15% water, and 37 Baume about 30% water). For 
transport by tank cars, a corn syrup of 43 Baume is preferred over one 
of 45 Baume because of its superior flow characteristics. Forty-three 
degree Baume corn syrup, even with improved flow vs. 45 Baume syrup, 
still must be heated to 100F to effect acceptable flow. Use of 41 Baume 
corn syrup (77% solids) eliminates the heating of corn syrup during storage. 
This requires longer heating during candy base preparation, thus 
resulting in longer cooking time and possibly more browning [10J. The 
overall advantages of 43 Baume corn syrup make this the syrup of choice 
in the preparation of hard candy lozenges. 
The primary functions of corn syrup in hard candy base are (a) to 
control crystallization; (b) to add body; (c) to supply solids at a reduced 
cost; (d) to adjust sweetness level. Control of sugar crystallization is a 
primary application of corn syrup in hard candy. Since sugar is readily 
crystallized when the water of sugar solutions is boiled off. the presence 
of the noncrystallizable corn syrup is necessary to inhibit the graining or 
recrystallization of the sucrose. This inhibition of sugar recrystallization 
is accomplished by surrounding each molecule of sucrose with a film of

Medicated Lozenges 429 
paOHlIY OR FUNCtiONAl US! lOW. IIG INTU . "10". "10M. 
4...IPM....nIC...u 11 CO"l'" CONY CONY CONY CON'" 
ViSCOSITY  Figure 4 Properties and functional uses of corn syrup. (Corn Refiners 
Association, Inc., Washington, D. C. ) 
uncrystallizable corn syrup. Hard candy, in essence, may be characterized 
as a supersaturated sugar solution in corn syrup [4]. The sugar molecules 
are dissolved and separated in the c()rn syrup, and because of the high viscosity 
of the corn syrup solution, movement of sugar molecules in the corn 
syrup is slowed. Eventually, though, molecules of sugar meet and combine, 
causing the formation of larger sugar crystals or the phenomenon described 
as graining [4]. 
The viscosity of the internal solution (determined by the grade of corn 
syrup and the moisture content of the finished candy base after cooking) 
and the storage conditions under which the finished lozenges are SUbjected 
(e.g . protection from moisture) determine the product's shelf life and rate 
of crystallization that can be expected [l1J. All hard candies do eventually 
grain. but the speed at which this phenomenon occurs depends on the aforementioned 
grade of corn syrup (viscosity), mositure content of the cooked 
base, and storage conditions. Modification of the ratio of sugar solids to 
corn syrup solids in candy base will also affect the rate of graining in the 
finished product. 
Incorporation of corn syrup solids at greater than 50% decreases graining 
tendencies because of the lower percentage of sugar solids dissolved in

430 Peters 
the syrup, but this increases moisture absorption, thus resulting in an increase 
in product stickiness and in the interactions of medicaments. Higher 
percentages of corn syrup reduce lozenge sweetness but allow longer processing 
time because of the slowed rate of candy base hardening. Addition of 
greater than 70% sucrose solids to candy base increases graining tendencies 
due to the high solids content in the oom syrup. Candy base crystallizes 
rapidly, thus decreasing mixing time, and increasing opaclty and the brittleness 
of the final lozenge. Candy base formulations containing 55- 65% 
sugar and 45- 35% corn syrup solids offer the best compromise among these 
factors: resistance to grsining, reduction of moisture absorption, and a 
realistic processing time period during manufacture. 
Invert Sugar 
Invert sugar is a mixture of two sugars (levulose and dextrose) in equal 
parts, produced by hydrolizing (inverting) sucrose. Molecules of sucrose 
combine with water to form smaller molecules during the cooking of the 
candy base [12]. 
Invert sugar has the power to absorb moisture from the air and at the 
same time retard crystallization. Controlling candy base cooking time will 
reduce the quantity of invert sugar. A standardized cooking time will result 
in the formation of uniform quantities of invert. 
Reducing Sugars 
The quantity of reducing sugar present in the corn syrup plus the quantity 
of reducing sugar formed during the cooking cycle determines the quantity of 
total reducing sugars in the final candy base. Controlling the total reducing 
sugars will determine how resistant the candy will be to graining and moisure 
Production of hard candy base containing greater than 20% reducing 
sugars slows the rate of product graining by lengthening crystallization 
time. This attribute is advantageous during manufacturing since the candy 
base will harden at a slower rate. The result is a base that can be mixed 
longer to assure a complete distribution of medicament while entrapping less 
air. This allows formation of a piece of hard candy with a greater degree 
of clarity. Increased crystallization time also produces a candy base that 
is more pliable during the lozenge-forming operation. This reduces the 
number of rejects formed because of lozenges breaking due to candy base 
brittleness. The incidence of sugar dusting is also lowered, resulting in 
a cleaner product and a more sanitary operation. 
Preparation of candy base with reducing sugar content below 14% leads 
to the formation of brittle candy that is susceptible to breakage, dusting, and 
formation of high quantities (greater than 20%) of lozenge rejects. This is 
the direct result of manufacturing difficulties caused from candy base hardening 
through rapid crystallization. The resultant lozenges, while possessing 
less hygroscopicity than product prepared with higher reducing sugars, are 
more susceptible to graining when exposed to moist conditions. 
A final reducing sugar content in the 16-18% range brings to the formulator 
many of the advantages cited for low and high reducing sugar content 
while minimizing the disadvantages. Crystallization time is slow enough 
to assure proper incorporation of medicaments, but sufficient candy base 
plasticity is available for the forming and molding operation. The resultant

Medicated Lozenges 431 
lozenges are not brittle, resist dusting during the packaging operation, 
and resist both graining and excessive moisture absorption. 
When selecting a grade of corn syrup suitable for lozenge manufacture, 
the formulator should consider a corn syrup prepared at a regular conversion 
level (41-44 DE), dual-converted (acid-enzyme) to a high maltose 
content (above 42%). The regular conversion imparts the proper internal 
viscosity to control graining, while the high-maltose-containing syrup is 
designed for use in products where a sweetener with minimum dextrose 
(less than 10%) and a resultant decrease in lozenge hygroscopicity is desired. 
The reduced dextrose content imparts better color stability, expecially 
during heating and storage, when higher dextrose contents would 
cause darkening. 
Lozenges containing high-maltose corn syrup have increased internal 
viscosity. This retards sugar movement and aids in controlling crystallization 
of sucrose, whUe the lower water-pickup tendency improves and extends 
the lozenge shelf life- both from the chemical aspect of reducing 
drug decomposition and from the physical aspects of reducing graining and 
High-maltose syrups were originally developed for use in hard candy, 
the theory being that a manufacturer using 40-50% regular conversion corn 
syrup (dry basis) could go to 50- 60% with high -maltose syrup [13,14]. 
While noticeable improvements resulted in the winter months, stickiness is 
still a problem in the summer. Many processors who ventured to the 60% 
level gradually cut back to the 40-50% level. The use of high percentages 
(above 50%) of high-maltose corn syrup produced lozenges that exhibited 
increased breaking or stress cracking becuase of the high viscosity imparted 
by the corn syrup. 
Most lozenges manufactured today possess a sugar-to-corn syrup ratio 
in the range of 50:50 to 70:30, with the greatest number of medicated 
lozenges produced with a ratio of 55- 65 parts sugar to 45- 35 parts corn 
syrup. This ratio produces lozenges with adequate sweetness, resistance 
to moisture pickup (with resultant stickiness), graining, and reactivity 
with medicinal components [15]. 
Acidulents are generally added to candy base as fortifiers to strengthen 
the flavor characteristics of the finished product. Acids commonly used 
include citric, tartaric, fumaric, and malic; of these, citric is by far the 
most common. 
A second use for acidulents in candy base is to control pH in order to 
preserve the stability of selected medicaments. Since hard candy base is 
considered a supersaturated solution of sugar in a corn syrup medium. and 
because of the presence of water in the medicated lozenge base, pH is an 
important factor in maintaining the stability of medicaments affected by an 
acid or alkaline medium. The reactivity of the corn syrup and reducing 
sugars, the presence of moisture in the candy base, and the presence of 
flavors and acidulents increase the reactivity of medicament in the vehicleto 
the extent that the kinetics of drug decomposition is related to liquid 
(as opposed to solid) dosage forms. 
Regualr hard candy base has a pH of 5.0-6. O. Addition of acidulents 
for flavor enhancement will lower the pH to 2.5- 3. O. At this pH many

432 Peters 
medicaments exhibit acceptable chemical stability, while others are subjected 
to rapid decomposition. A determination of the stability profiles of the 
medicaments intended for incorporation into the lozenge base should be 
carried out at various pH levels to determine that which is optimum. This 
determination may preclude the use of acidulents and the flavors with which 
they are most compatible. 
In some special applications, addition of selected ingredients (calcium 
carbonate, sodium bicarbonate, magnesium trisilicate) to raise the lozenge 
pH to 7.5- 8. 5 will be necessary to effect the desired stability profiles. 
Method of Addition 
Addition of acidulents to candy base is not a random procedure. Acidulent 
addition should be performed under controlled conditions since, even 
under the best circumstances, the acidulent will react with the candy base. 
Addition of acid to sugar (sucrose) causes inversion. which yields by hydrolysis 
glucose and fructose (dextrose and levulose). As the percentage 
of invert sugar in the candy base increases, the internal viscosity of the 
lozenge decreases, and the moisture absorption characteristics increase. 
Both phenomena increase tendencies for lozenges to grain. absorb moisture J 
and become sticky [16]. 
A certain quantity of invert sugar is produced during the cooking 
cycle. The faster the cooking cycle. the lower the quantity of cook invert 
formed. Addition of aeidulents to candy base during the cooking cycleor 
the failure to neutralize excess acid in any salvage that may be incorporated-
acts as a sugar doctor or inverting agent. This so-called doctor 
will markedly increase the quantities of invert sugar formed, negating the 
advantages of a low moisture content in the base preparation or the use 
of high-maltose corn syrup. The aeldulents should be added at the completion 
of the cooking cycle at temperatures not exceeding 120C. Final invert 
sugar levels in candy base should not exceed 2.0- 2.5%. 
The presence of acidulents in the completed lozenge will shorten the 
shelf life of the final product. since even at room temperature the aeidulent 
will continue to invert the sugar. Thus, the rate of graining and 
degree of stickiness will be higher than in lozenges prepared at pH 5-6. 
Another drawback of acidulents in lozenges OCCurS with elevated temperature 
and humidity. Under these conditions, a localized discoloration or 
burning of the candy will OCCUr. Use of finely powdered acids helps to 
reduce this problem but will not eliminate it. 
Incorporation of the acidulents to the vehicle as a controlled procedure 
helps minimize the disadvantages acidulents can represent in reducing the 
extended shelf life of the products to which they are added. The acidulent 
should be added to candy base at the lowest possible workable temperature 
of the candy mass (100-1100C). At the same time, the acidulent should 
be added at the lowest effective concentration (0.1- 0.5%) in a manner that 
will prevent direct contact of the acid with the mass. Incorporation of 
the acidulent as a mixture with dry, ground salvage and the flavor-ant will 
lessen contact of the acidulent with the base. and at the same time help 
distribute it uniformly throughout the mass. This uniform incorporation 
prevents reaction during the addition procedure and reduces the degree of 
localized discoloration Or burning during storage. The use of granular 
acidulent instead of finely powdered material will result in localized discoloration 
if the lozenges are exposed to prolonged heating Or high humidity

Medicated Lozenges 433 
during storage. The reactivity of acidulent with candy base during product 
manufacture is reduced because of lower overall particle surface area. 
The advantages that acidulents bring to lozenge formulations through 
pH oontrol and flavor enhancement usually exceed the disadvantages of 
discoloration and sugar inversion during storage, if the degree of inversion 
can be controlled during lozenge manufacture by proper addition techniques. 
Incorporation of powdered or micronized dyes is not practical because of 
the low moisture content (less than 1. 5%) and the high viscosity of the 
cooked candy base. Not all the dye will dissolve in the base. resulting 
in a nonuniform and nonreproducible colored product containing particles 
of undissolved dye. A method used to circumvent this difficulty involves 
the incorporation of colors into hard candy base as pastes. in mixtures of 
sugar, dextrose, corn syrup, dextrin, and glycerin; as aqueous solutions; 
or as commercially prepared color cubes (Figure 5) [17}. When adding 
colors as aqueous solutions, no more than 30.0 g of water should be added 
per 100 lb of candy base. More than this quantity will result in localized 
sticking and lumping during the mixing cycle. if more liquid is required. 
combinations with glycerin or propylene glycol should be used. 
The formulator, during product development. should investigate the 
compatibility of the colorants- both at ambient temperature and at llO115
C. the temperature at which the colors will be added to the products-esince 
many dye systems are altered when added at the elevated temperature. 
A second factor that should be considered is the product pH. Addition of 
acidulents to candy base at elevated temperature along with. or shortly 
after, color addition can result in a noticeable change in the final product 
color as well as color differences between batches. Stability of colors in 
Figure 5 Colors may be added to candy base as pastes, as aqueous solutions, 
or as commercially prepared color cubes.

434 Peters 
the final product (effects of moisture. sunlight, pH, and medicaments) is 
also a matter of concern since changes in product appearance with time 
are not uncommon. 
Many in-process color changes result when colored liquid salvage is incorporated 
in the candy base. This color modification may occur because 
of the pH of the salvage solution before COoking or may be because of a 
color change effected during the candy base cooking cycle. Color changes 
that result from pH may be remedied by a change in salvage pH. The 
salvage solution pH may be adjusted anywhere in the range of 4.5-7.5. If 
a pH in this range can produce a stable color solution, then color change 
problems can be avoided. Color change problems caused by the cooking 
temperature of the candy base cannot be alleviated. If this problem occurs, 
the salvage solution may have to be decolorized before use [69]. Modification 
of the candy base color back to the original shade can be effected by 
the addition of more color to the cooked candy base. This is practical 
only if a uniformly color-modified product is produced each time the colored 
salvage solution is manufactured. Candy base colors that prove to be stable 
when added to candy base during the cooking cycle may be added to 
salvage solutions before the cooking cycle instead of to the cooked candy 
base during the mixing cycle. 
The addition of flavors to cooked candy base can pose a variety of problems 
to the fonnulator. These include flavor losses during processing, 
flavor incorporation difficulties, flavor and candy base interactions, and 
flavor-medicament interactions. The specific flavor-related difficulty must 
be determined. and remedial actions taken, if a stable and reproducible 
product is to result. 
Addition of flavors to candy base usually takes place at temperatures 
from 120 to 135C. At these temperatures, flash-off is the primary problem. 
Addition of flavors to