PHARMACEUTICAL
`DOSAGE FORMS
`Disperse Systems
`In Three Volumes
`VOLUME1
`Second Edition, Revised and Expanded
`
`EDITED BY
`Herbert A. Lieberman
`H. H. Lieberman Associates, Inc.
`Livingston, New Jersey
`Martin M. Rieger
`M. & A. Rieger Associates
`Morris Plains, New Jersey
`Gilbert S. Banker
`University of Iowa
`Iowa City, Iowa
`
`Marcel Dekker, Inc.
`
`New York• Basel• Hong Kong
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`Library of Congress Cataloging-in-Publication Data
`
`Pharmaceutical dosage forms-disperse systems I edited by Herbert A. Lieberman,
`Martin M. Rieger, Gilbert S. Banker. -2nd ed.
`p. em ..
`Includes bibliographical references and index.
`ISBN 0-8247-9387-0 (v. 1 :hardcover: alk. paper)
`1. Drugs-Dosage forms. 2. Dispersing agents. I. Lieberman, Herbert A.
`II. Rieger, Martin M. III. Banker, GilbertS.
`[DNLM: 1. Dosage Forms.
`2. Chemistry, Pharmaceutical. 3. Emulsions. 4. Suspensions. QV 785 P5349
`1996]
`RS200.P42 1996
`615'.19-dc20
`DNLM/DLC
`for Library of Congress
`
`96-15604
`CIP
`
`The publisher offers discounts on this book when ordered in bulk quantities. For more
`information, write to Special Sales/Professional Marketing at the address below.
`This book is printed on acid-free paper.
`Copyright C9 1996 by Marcel Dekker, Inc. All Rights Reserved.
`Neither this book nor any part may be reproduced or transmitted in any form or by any
`means, electronic or mechanical, including photocopying, microfilming, and recording,
`or by any information storage and retrieval system, without permission in writing from
`the publisher.
`Marcel Dekker, Inc.
`270 Madison Avenue, New York, New York 10016
`Current printing (last digit):
`10 9 8 7 6 5 4 3 2
`PRINTED IN THE UNITED STATES OF AMERICA
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`7
`Viscosity-Imparting Agents in Disperse
`Systems
`
`Joel L. Zatz
`Rutgers-The State University of New Jersey, Piscataway, New Jersey
`
`Joseph J. Berry
`Bristol-Myers Squibb Company, New Brunswick, New Jersey
`
`Daniel A. Alderman
`The Dow Chemical Company, Midland, Michigan
`
`INTRODUCTION
`I.
`Polymers are used in suspensions, emulsions, and other dispersions, primarily to mini-
`mize or control sedimentation. The rheological character given to disperse systems also
`plays a role in maintaining pharmaceutical preparations at their application site. For
`example, highly fluid skin lotions may run, whereas viscous preparations tend to remain
`in place for longer time periods. A related application is in ophthalmic preparations, for
`which polymers are used to enhance drug retention.
`In addition to their effect on dispersion rheology, polymers may also play a role
`in determining the flocculation state of suspended particles. By virtue of their surface
`activity, some polymers can directly improve emulsion stability; the ability of acacia to
`function as an emulsifier is well known.
`Various substances have been used over the years to build viscosity in aqueous drug
`systems: Included are such familiar compounds as sucrose and other sugars, and polyols
`such as glycerin. These materials suffer from two major disadvantages. They are needed
`in high concentration to product significant viscosity changes, and their aqueous solu-
`tions are newtonian in nature (see Chap. 5).
`On the other hand, only small amounts of many polymers (depending on chemis-
`try and molecular weight) are needed to bring the viscosity of an aqueous preparation
`to almost any desired value. Furthermore, most polymer solutions or dispersions are
`nonnewtonian; in addition to being pseudoplastic, they may exhibit a yield point or
`thixotropy. These properties are advantageous in combining sedimentation resistance with
`processing ease. This point is explained further in Section III.
`
`287
`
`--
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`288
`
`Zatz et at.
`
`The nonnewtonian nature of polymer solutions makes it difficult to compare the
`properties of different polymers. Viscosity is a function of shear rate and, quite often,
`shear history, so that the numerical value of viscosity that is measured is a function of
`the method and conditions of measurement. Typically, manufacturers quote viscosity
`figures obtained from a single measurement at relatively high shear; this is insufficient
`to characterize a nonnewtonian material, particularly since its application to sedimen-
`tation control will take place under quiescent (low-shear) conditions.
`As a consequence, the viscosity data provided by raw material suppliers is useful
`in only a very general way. Such data can show, for example, that certain polymer
`grades yield more viscous solutions than other grades made by the same manufacturer.
`However, it is extremely difficult to compare data on different polymers supplied by
`different manufacturers. It is not uncommon to find that two polymers, the solutions of
`which have nearly the same quoted viscosity value, affect a disperse system in mark-
`edly different ways.
`Within a polymer family, an increase in molecular weight results in an increase in
`molecular asymmetry; hence, in viscosity. Different viscosity grades, based on a dif-
`ference in average molecular weight, are described in several ways. With methylcellu-
`lose, the viscosity of a 2% aqueous solution measured in a standard manner is provided.
`Polyethylene glycols are described in terms of average molecular weight, whereas des-
`ignations, such as low, medium, and high, are used in connection with viscosity grades
`of carboxymethylcellulose.· All aqueous systems containing polymers require a preser-
`vative. Many polymers of natural origin are attacked directly by microorganisms. Cel-
`lulose derivatives are degraded by cellulases, enzymes that may be produced by microbial
`agents. Even if the polymer chosen is totally resistant to bacteria and molds, the aque-
`ous medium may allow growth, and a preservative is still necessary.
`Certain inorganic agents are also used as viscosity builders. Examples are colloi-
`dal magnesium aluminum silicate (Veegum) and microcrystalline silica. These substances
`do not support bacterial or mold growth and are relatively inert from a physiological
`standpoint.
`
`II. POLYMER SOLUTION RHEOLOGY
`Typically, polymer solutions are nonnewtonian. The three most commonly observed
`behaviors for polymer solutions are plastic, pseudoplastic, and thixotropic. Plastic sys-
`tems flow only after a critical shear stress is exceeded (yield value). In pseudoplastic
`or shear-thinning systems, the viscosity decreases with increasing rates of shear. Thix-
`otropy is the case in which a plastic or pseudoplastic system exhibits a time-dependent
`recovery, resulting in a hysteresis loop if shear stress is alternatively increased and
`decreased.
`The type of rheological behavior, as well as the magnitude of the viscosity, is a
`critical factor determining the usefulness of a particular polymer for each potential ap-
`plication. For example, pseudoplasticity, the existence of a yield point, and some de-
`gree of thixotropy are useful characteristics for a polymer used as a suspending agent.
`Thus, xanthan gum, which has a yield value and is highly pseudoplastic, was a more
`effective retardant of creaming in mineral oil-in-water emulsions than either methylcel-
`h.ilose or carboxymethylcellulose [1], despite that comparisons were made at concentra-
`tions yielding the same range of measured viscosity values. Thus, it is not possible to
`evaluate polymer usefulness on the basis of a single viscosity value measured under
`arbitrary conditions.
`
`i
`, 'I
`i
`I
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`Viscosity-Imparting Agents in Disperse Systems
`
`289
`
`The situation is complicated because rheological characteristics of polymer solutions
`may vary, depending on the concentration and degree of substitution. For example,
`solutions of medium- and high-viscosity grades of carboxymethylcellulose that are not
`highly substituted tend to exhibit thixotropic behavior, whereas more highly substituted
`grades are pseudoplastic.
`It is sometimes advantageous to combine viscosity builders with different proper-
`ties. The addition of xanthan gum to dispersions of magnesium aluminum silicate re-
`duced the extent to which the viscosity of the latter increased over time [2]. Magnesium
`aluminum silicate is highly thixotropic in dispersions by itself; this was reduced by the
`gum. In addition, steady-shear measurements suggested synergy between the two ma-
`terials in both viscosity and yield value.
`Viscoelastic properties of the same materials were investigated by oscillatory shear
`[3]. The storage modulus G' was essentially independent of frequency for 1 and 3% clay
`dispersions containing no gum. The addition of gum shifted the behavior, and the data
`for the combined materials contained some of the rheological characteristics of each of
`the pure substances. The results were interpreted in terms of a reduction in structural
`·
`rigidity of the clay and an increase in flexibility.
`A recent study evaluated combinations of magnesium aluminum silicate with three
`carbomers [4]. The data suggested enhancement of the structure (yield value) in com-
`parison with the properties of the two substances taken separately.
`Specialized applications require specific rheological characteristics. For example,
`viscoelastic substances are used in eye surgery to prevent mechanical damage to sensi-
`tive tissues and avoid adhesions [5]. Polymer solutions are used in cataract, corneal, and
`glaucoma surgery. Among the agents employed are sodium hyaluronate, hydroxypropyl
`methylcellulose and chondroitin sulfate.
`Power law relations are frequently used to describe the behavior of pseudoplastic
`polymer solutions. As part of a research program on natural polymer properties, the
`effects of concentration and temperature on the behavior of guar gum dispersions were
`evaluated [6]. A plot of the logarithm of shear rate was a linear function of the loga-
`rithm of shear stress and the following equation was applied to each flow curve:
`v = aab
`where
`v = shear rate
`cr = shear stress
`a and b are constants
`The power constant b was directly proportional to gum concentration and inversely
`related to temperature. From the data, the authors were able to formulate a single
`empirical equation that permitted calculation of shear rate from shear stress, concentra-
`tion, and temperature over a relatively wide range of values.
`
`Ill. SEDIMENTATION CONTROL IN DISPERSE SYSTEMS
`The control of sedimentation is of primary importance in maintaining the integrity of a
`disperse system. Stokes' law [7] defines the sedimentation rate of a sphere in a fluid
`as
`
`V = 2 r2
`
`( d5 - dL ) g
`9Tj
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`290
`
`Zatz eta/.
`
`where
`V = sedimentation rate
`r = particle size
`ds = density of sphere
`dL = density of liquid
`g = gravitational constant
`11 = viscosity of continuous phase
`Although most drugs in suspension are not perfect spheres and the suspensions are
`not dilute enough to follow Stokes' law, the equation is still useful. From the equation,
`three methods of controlling sedimentation present themselves: (a) particle size reduc-
`tion, (b) density matching, and (c) viscosity building.
`Since particle radius is raised to the second power, a modest change in this param-
`eter translates into a much larger change in sedimentation rate. In practice, it is diffi-
`cult to achieve particle reduction into the submicron range.
`If the densities of the medium and the suspended particle are the same, no sedimen-
`tation will occur. An exact match is usually difficult to obtain, but it is occasionally
`possible to add ingredients that will bring the density of the continuous phase closer to
`that of the dispersed phase. Salts, if appropriate to the formulation, sugar, and other
`polyols may be used in aqueous suspensions for this purpose. Often the amount of
`density-increasing agents required to make the densities of the two phases equal would
`be too great to be practical.
`A common method of controlling sedimentation is through use of viscosity-build-
`ing agents, alone or in combination with one or both of the aforementioned approaches.
`The rheological characteristics of the polymer solutions used to stabilize disperse sys-
`tems are very important. Systems that exhibit shear thinning are useful. They allow the
`dispersion to have a high "resting" viscosity and also enable redispersion of any settled
`material. Furthermore, it is possible to pour material from the container, even though
`the viscosity of the dispersion at rest may be considerable. It is not always feasible, or
`even necessary, to completely arrest the sedimentation of suspended particles. However,
`for suspensions that do settle, it is important that they be easily resuspendible (see Sec.
`V).
`
`Zatz [8] has reviewed the merits of several thickeners, as well as how rheological
`behavior affects sedimentation stability of disperse systems. Relatively small increases
`in the concentration of xanthan gum dramatically reduced the sedimentation rate of
`sulfamerazine suspensions. The logarithm of the initial settling rate was a linear func-
`tion of gum concentration.
`
`IV. VISCOSITY CHANGES DURING PRODUCT AGING
`The shelf-life of a dispersion depends on the chemical stability of its ingredients, as well
`as the physical stability of the system as a whole. Because of the importance of viscos-
`ity in terms of stability and certain use characteristics, major changes in viscosity over
`a time period are cause for concern.
`Several factors may be responsible for changes in dispersion viscosity over time.
`Some are obviously due to alterations in the viscosity-building agent or its interaction
`with the rest of the system. Other factors, such as particle growth, may be independent
`of polymer content, although the polymers present may reduce the rate of change of par-
`ticle size.
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`Viscosity-Imparting Agents in Disperse Systems
`
`291
`
`Depolymerization results in a decrease in average molecular weight; hence, a de-
`crease in viscosity. Although processing under high shear may result in depolymeriza-
`tion, further viscosity changes would not be expected after manufacture unless chemi-
`cal degradation of the polymer were to take place. Degradation of cellulosic derivatives
`by cellulase is an example of such a process. Cellulases may be introduced by micro-
`organisms, or may be present as contaminants in other raw materials. Other chemical
`changes, such as hydrolysis of polymers at low pH values, are also possible.
`Chemical changes in the system over time, producing a drift in pH or generating
`ionic products, may alter viscosity by virtue of the effect of these environmental alter-
`ations. Polymers may introduce a second-order effect on viscosity by acting as either
`flocculating or deflocculating agents in certain instances (see following section).
`Time-dependent hydration of polymers or other viscosity-building agents can also
`result in a change in measured viscosity over time. In contrast with many of the other
`influences, continuing hydration results in an increase in viscosity after manufacture.
`Typically, viscosity reaches a plateau value after 1 or 2 weeks.
`Because of the variety of factors that can alter viscosity over time, some of which
`increase viscosity, whereas others have the opposite effect, it is often difficult to pin-
`point the exact cause in a particular situation. Small drifts in apparent viscosity are often
`encountered and are usually considered acceptable. However, substantial changes are
`cause for concern because of changes in the resistance to sedimentation and, also, be-
`cause they suggest that chemical or physical changes of some kind are taking place. In
`other words, they are a sign. that chemical or physical stability are not all that they should
`be.
`
`In a study of polymer stabilization of 10% mineral oil emulsions, the apparent vis-
`cosity at 25°C was measured over time [1]. In emulsions containing 1% emulsifier and
`1% carbxoymethylcellulose (CMC), high-viscosity grade, the apparent viscosity dropped
`from about 780 mPa s 1 week after manufacture to about one-tenth that value after 448
`days. Apparent viscosity of emulsions containing other CMC concentrations also de-
`creased over time. The effect of storage on apparent viscosity of another anionic poly-
`mer, xanthan gum, depended on gum concentration. Apparent viscosity decreased when
`the gum concentration was 0.1%, remained essentially constant at a 0.2% gum level,
`and increased over time at higher concentrations. Although several factors may be in-
`volved in these changes, the patterns observed suggest that small amounts of ionic prod-
`ucts were produced during storage [1,9].
`Elevated storage temperature can have an adverse effect on polymer stability that
`will result in a viscosity change over time. In a study of methylhydroxyethylcellulose,
`single-point measurements were used to track stability of several polymer grades at
`different temperatures over time [10]. There was little change in viscosity following
`storage for 2 months at room or refrigerator temperature. Storage at 40°C, however,
`usually caused losses of 15% or more. Storage temperatures of this magnitude may be
`encountered in tropical areas. Another concern is that high temperatures produced during
`manufacture, even for a short period., may adversely affect viscosity.
`
`V. FLOCCULATION AND DEFLOCCULATION PHENOMENA
`Flocculation is a process in which particles are allowed to come together and form loose
`agglomerates. Unlike coalescence, the total surface area is not reduced during the floc-
`culation process. Deflocculation is the opposite, that is, breakdown of clusters into in-
`dividual particles. Table 1 lists the properties of both types of suspensions.
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`' ii
`I
`
`292
`
`Zatz et at.
`
`Table 1 Properties of Flocculated and Deflocculated Suspensions
`De flocculated
`
`Flocculated
`
`Property
`
`Sedimentation rate
`Supernatant
`Sediment
`Redispersibility
`
`Rapid
`Clear
`Voluminous
`Easy
`
`Slow
`Cloudy
`Compact
`Difficult
`
`It can be seen from the table that the chief advantage of a flocculated suspension
`is its redispersibility. Deflocculated suspensions settle more slowly and look more el-
`egant while settling, but typically the sediment is close-packed and cannot be redispersed
`by shaking the container. The goal of controlled flocculation is to produce reasonably
`sized aggregates or floes. In this way redispersibility is maintained while sedimentation
`rate is kept within reasonable bounds.
`The difference in sediment density can be used as a means for characterizing the
`flocculation state. After allowing a group of suspensions containing the same active solid
`at the same concentration to settle completely, it is possible to compare the volume of
`sediment to identify those that are flocculated. In practice, the parameter measured is
`the sedimentation volume, the ratio of the volume of sediment to total suspension vol-
`ume, often abbreviated as F [7]. In a series of suspensions, those that are deflocculated
`have the smallest values of F. Flocculated suspensions generally have F values at least
`twice as large as those for the deflocculated systems. The extent of flocculation is some-
`times assumed to be proportional to the F value of a flocculated system.
`Flocculation can be achieved by various means. The use of a polymeric agent is one
`common method. Polymers that contain chemical groups that interact with the suspended
`particles can be added to the continuous phase. In such a case, polymer segments can
`then attach to individual particles to form a polymer-particle complex. As a polymer
`links two or more particles, a floc can be formed. This flocculation process is called
`polymer bridging and is often concentration-dependent. When the concentration of poly-
`mer is high, particles can be completely surrounded by polymer and, thus, little floc
`formation can occur. Felmeister et a!. [11] observed this concentration-dependent be-
`havior in sulfaguanidine suspensions that were flocculated with an ionic polysaccharide.
`An increase in sedimentation volume was noted as polymer concentration was increased,
`but as concentration was increased still further, sedimentation volume decreased.
`Law and Keyes [12] have shown that multilayer absorption of certain water-soluble
`cellulose derivatives can lead to deflocculation. The authors examined various concen-
`trations of polymers. At high polymer concentrations deflocculation was usually ob-
`served. This effect was believed to be caused by steric repulsion.
`A second method of flocculation is based on electrical charge. Suspended particles
`often have an associated charge. These charged particles will repel each other and,
`thereby, resist forming floes. Reduction of surface charge may be accomplished by added
`electrolytes or surfactants of opposite charge. Reduction of particle repulsion permits the
`particles to become close enough to allow the attractive van der Waals forces to domi-
`nate. The complex nature of these particle interactions are explained by DLVO theory,
`·
`which is covered by Hiemenz [13].
`Components other than the suspended drug can influence flocculation by polymers.
`In a study of sulfamerazine suspensions, two surfactants, one anionic and the other non-
`
`I
`
`I
`I
`I j,
`
`i l II
`
`1l1 :1
`.
`
`'.1111,
`I', 1
`r .·.
`i! 1
`': I ~ I
`1 '11
`
`; [
`'1'1.
`'It'·.
`'~ : ! ii
`i 1111
`1:,,: I II
`
`1'
`
`'1,[1
`I,
`I'
`; l'l,li
`·!:hi
`!
`I.
`
`I
`I i
`
`,
`
`'1.
`
`1.: I'
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`Viscosity-Imparting Agents in Disperse Systems
`
`293
`
`ionic, were employed as wetting agents [14]. Primafloc C3, a cationic polymer used in
`water purification, was added at different concentrations that spanned a wide range.
`Some of the results for suspensions containing the anionic wetting agent, docusate so-
`dium, are shown in Fig. 1, in which sedimentation volume F is plotted as a function
`of the logarithm of polymer concentration. At very low concentrations, the suspensions
`were deflocculated and settled to form a nondispersible cake; variations in polymer con-
`centration had no significant effect. But, at a concentration of about 0.1% polymer in
`suspensions containing 0.2% surfactant, the suspension volume was increased signifi-
`cantly, signaling a change from a de flocculated to a flocculated state (see curve 1, Fig.
`1). The addition of more polymer had little effect until its concentration reached 1%,
`at which point the suspension was once again deflocculated.
`A similar pattern was observed in suspensions containing 1 % docusate sodium but
`flocculation occurred at higher polymer concentration (see curve 2, Fig. 1). The choice
`of wetting agent influenced the results obtained. When a nonionic surfactant, polysor-
`bate 40, was used in place of docusate sodium, addition of the polymer resulted in a
`minimal increase in sedimentation volume. Combinations of the two surfactants at the
`same total concentration were employed in some experiments. The polymer concentra-
`tion at the flocculation peak was a function of the weight fraction of docusate sodium
`in the surfactant mixture.
`Figure 2 contains diagrams that suggest the mechanisms involved in determining
`flocculation state of the suspensions. In the absence of polymer, or at low polymer
`concentrations (see Fig. 2a), the negatively charged particles repel each other, preventing
`flocculation. At higher polymer concentrations (see Fig. 2b), electrostatic attraction
`results in simultaneous polymer adsorption to more than one particle, leading to floc-
`culation. At much higher concentrations, the ratio of polymer to particle area· is such
`that adsorption tends to cover each particle with a positively charged polymer layer, once
`again producing a deflocculated system (see Fig. 2c).
`The same surfactants were employed in a study of flocculation of magnesium car-
`bonate suspensions by xanthan gum [15]. The gum carries a negative charge, whereas
`the particle surfaces are positively charged in aqueous dispersion. Aqueous magnesium
`carbonate dispersions were flocculated in the absence of additives. The addition of the
`gum increased the sedimentation volume (enhanced flocculation), whereas the opposite
`effect resulted from incorporation of docusate sodium, an anionic surfactant. When both
`
`o~~--~---~1---L--~
`LOG C
`Fig. 1 Sedimentation volume (F) of sulfamerazine suspensions containing docusate sodium as
`a function of polymer concentration (C). Curve 1, 0.2% docusate sodium; curve 2, 1.0% docusate
`sodium. (From Ref. 11.)
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`i[li ~ .. ·
`: I
`'I !j
`
`294
`
`0 0
`
`Zatz eta!.
`
`(c)
`
`Fig. 2 Schematic representation of sulfamerazine particles in suspension. Spheres are particles;
`rods are polymer molecules. (a) Deflocculation at low polymer concentration; (b) flocculation by
`bridging; (c) deflocculation at high polymer concentration. (From Ref. 11.)
`
`the gum and docusate sodium were present, the net result in terms of flocculation state
`depended on the relative concentrations of the two materials. This was rationalized in
`terms of competitive adsorption between the surfactant and polymer on the particle
`surface. A bridging mechanism implies adsorption of the polymer; if this is blocked by
`surfactant adsorption, flocculation enhancement by the polymer is reduced or does not
`occur. On the other hand, the nonionic surfactant polysorbate 40 did not itself influence
`flocculation of the particles or block the effect of the gum. Apparently, the lack of charge
`on this surfactant reduced its adsorption, thereby, allowing molecules of the gum to
`attach to the particle surface.
`
`VI. EXCIPIENT COMPATIBILITY
`Suspension dosage forms and other disperse systems may contain a large number of
`additives. Among them typically are surfactants, polymers, electrolytes, and polyols.
`There are many possibilities for complex interactions among these agents and between
`them and the drug [ 16].
`As a general guide, precipitation reactions can be anticipated when charged poly-
`mers are mixed with other polymers or surfactants of opposite chemical type. Thus,
`anionic polymers tend to be compatible with other anionic polymers or nonionic poly-
`mers, or with nonionic or anionic surfactants, but not with cationic polymers or surfac-
`tants. Occasionally, precipitation can. be avoided if one of the interacting species is
`present in much larger concentration than the other. Added salts sometimes block the
`reaction.
`A troublesome interaction, although it does not usually lead to precipitation, is the
`binding (complexation) of phenolic and carboxylic preservatives by nonionic polymers
`and surfactants. Complexation may result in a shortage of sufficient free preservative
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`Viscosity-Imparting Agents in Disperse Systems
`
`295
`
`to effectively protect against microorganisms. If it is not possible to substitute another,
`less interactive, preservative, or to remove the offending adjuvant, one must resort to
`an increase in preservative concentration.
`Some caution should be exercised when charged polymers are combined with op-
`positely charged particles. The adsorption that results may lead to flocculation, which
`can be desirable (see previous section). However, excessive flocculation should be
`avoided, or the suspensions may turn into an inelegant, yogurtlike mass.
`
`VII. POLYMER SELECTION
`A number of decisions have to be made in selecting a viscosity-building agent for a
`formulation. General rheological behavior and possible interactions with the particles and
`other components of the dispersion are major factors. The following considerations help
`provide general guidance, but the selection process may not always follow a step-by-
`step procedure. After eliminating those materials that are obviously not suitable and
`selecting likely candidates from those that remain, empirical testing is necessary to iden-
`tify the best candidates.
`
`A. Biological Compatibility
`It is obvious that certain polymeric agents will be removed from consideration, depending
`on the route of administration or application. With the possible exception of novel poly-
`mers with truly unique properties, only those that have been approved for a particular
`use should be tested.
`
`B. Physicochemical Compatibility
`Iri general, the chemical type of the polymers and surfactants in a disperse system must
`be mutually compatible. A knowledge of the charge on the particle surface is helpful
`in anticipating changes in flocculation character. These considerations were discussed
`more fully in the section on flocculation-deflocculation phenomena.
`
`C. Natural Versus Synthetic
`To obtain a polymer derived from a natural source, it is necessary to go to that source.
`If the desired material is obtained from a plant exudate, consider the site where the plant
`grows. Political changes, climatic changes, or local disasters of one kind or another can
`influence availability of a natural raw material as well as the quality of what is avail-
`able. Thus, given recent events in Iran and Afghanistan, traditional sources for gum
`tragacanth, it is not surprising that the price of this substance has skyrocketed and that
`high-grade material is currently difficult to obtain. Natural products produced domesti-
`cally by fermentation, such as xanthan gum, do not suffer from the same limitation.
`Another consideration is the degree of chemical uniformity. In general, synthetic
`polymers and modified natural polymers (such as cellulose and chitin derivatives) might
`be expected to have more reproducible characteristics than plant exudates, provided that
`suitable care is taken in raw material selection and manufacture. Certainly, the ability
`to procure pharmaceutical dispersion adjuvants with predictable, consistent rheological
`properties is an important factor in selection.
`The vulnerability of materials of natural origin to microbiological attack is often
`cited; however, natural raw materials do differ in susceptibility. Furthermore, many
`
`LUYE1022
`IPR of Patent No. 6,667,061
`
`

`
`.I
`
`i;
`i
`I
`
`296
`
`Zatz et at.
`
`synthetic and semisynthetic polymers may also undergo degradation. Consequently,
`preservatives are required in all aqueous systems.
`
`D. Viscosity Type and Concentration
`Major considerations that come into play here involve effectiveness in retarding sedi-
`mentation (low-shear rheology), patient handling (medium-shear rheology), and process-
`ing during manufacture (high-shear rheology). A complete rheological description at
`various concentrations, therefore, is of great help, but empirical testing is likely to be
`necessary in fine-tuning the selection process.
`
`VIII. VISCOSITY-BUILDING SUBSTANCES
`In this section, the most frequently employed viscosity-building agents are reviewed, and
`specific information useful to the formulator is provided. Water-soluble polymers are
`emphasized, but thickeners for nonaqueous and partially aqueous systems are also avail-
`able. Many of the same substances can be used to produce gels. Many polymer prop-
`erties are summarized in Table 2. Additional information on substances useful in topi-
`cal formulations is also available [ 17].
`Synthetic and semisynthetic polymers are generally available in several grades,
`varying in chemistry and molecular weight and, therefore, in rheological properties. Most
`of the cellulose derivatives, for example, can be obtained in a variety of grades that have
`different degrees of substitution and molecular weight. In the natural polymers, func-
`tional differences are introduced by a process of selection or by chemical treatment.
`Thus, several alginates, differing in molecular weight and calcium content (and, con-
`sequently, viscosity) are offered commercially.
`Most of the specific agents, the descriptions of which follow, are generally recog-
`nized as safe (GRAS) for use in foods. Many are listed in the United States Pharam-
`copeia/National Formulatory (USP/NF). In some instances, specific grades have been
`approved for use in foods and pharmaceuticals, whereas othe'rs have not. Before using
`any of these substances, its current status should be verified.
`
`A. Natural Polymers
`Water-soluble polymers are widely found in nature; they are derived from plant exu-
`dates, seaweed extracts, seed extracts, and fermentation processes of certain microor-
`ganisms. These polymers may be either nonionic or anionic. Different structures, branch-
`ing, and mol