`
`COMPRESS
`
`ADJUST
`WEIGHT
`
`DIE
`FILL
`
`Argentum EX1013
`Page 1
`
`
`
`'I
`
`' ,,
`
`I
`
`I !
`I I ; ,
`
`\
`i
`' 1.
`•' I
`I
`C ,,
`i
`
`PHARMACEUTICAL
`,, DOSAGE FORMS
`
`Tablets
`
`SECOND EDITION, REVISED AND EXPANDED
`
`In Three Volumes
`VOLUMEl
`
`EDITED BY
`
`Herbert A. Lieberman
`H.H. Lieberman Associates, Inc.
`Consultant Services
`Livingston , New Jersey
`
`Leon Lachman
`Lachman Consultant Services
`Westbury, New York
`
`Joseph B. Schwartz
`Philadelphia College of Pharmacy and Science
`Philadelphia, Pennsylvania
`
`MARCEL DEKKER, INC.
`
`New York and Basel
`
`Page 2
`
`
`
`Library of Congress Cataloging-in-Publication
`
`Data
`
`/ edited by Herbert A. Lieberman,
`and expanded.
`rev.
`- - 2nd ed.,
`
`forms--tablets
`dosage
`Pharmaceutical
`Joseph B. Schwartz.
`Leon Lachman,
`p.
`cm.
`index.
`Includes
`(v. 1 : alk. paper)
`ISBN 0-8247-8044-2
`I. Lieberman,
`forms.
`2. Drugs--Dosage
`(Medicine)
`1. Tablets
`Joseph B.
`Schwartz,
`III.
`II. Lachman, Leon.
`Herbert A.
`& dosage. QV
`2. Drugs--administration
`1. Dosage Forms.
`[DNLM:
`785 P535]
`RS201. T2P46
`615'.191--dc19
`DNLM/DLC
`for Library
`
`1989
`
`of Congress
`
`89-1629
`CIP
`
`Copyright © 1989 by MARCEL DEKKER, INC. All Rights Reserved
`
`in
`or transmitted
`this book nor any part may be reproduced
`Neither
`including photo(cid:173)
`any form or by any means, electronic or mechanical,
`storage
`or by any information
`copying, microfilming , and recording,
`from the publisher.
`in writing
`system, without permission
`and retrieval
`
`MARCEL DEKKER, INC.
`270 Madison A venue, New York, New York 10016
`
`(last digit):
`printing
`Current
`10 9 8 7 6 5 4 3 2 1
`
`PRINTED IN THE UNITED STATES OF AMERICA
`
`Page 3
`
`
`
`Contents
`
`Preface
`Contributors
`of Pharmaceutic al Dosage Forms:
`Contents
`~~~s2~d3
`of Pharmaceutical Dosage Forms: Parenteral Medications,
`~~mnl~d2
`of Pharmaceutical Dosage Forms: Disperse Systems,
`Volumes 1 and 2
`
`Contents
`
`Contents
`
`Tab lets, Second Edition
`
`Chapter
`
`1.
`
`and
`
`Studies
`
`Chapter
`
`and Design
`2. Tablet Formulation
`Garnet E. Peck, George J. Baley, Vincent E. Mccurdy,
`and Gilbert S. Banker
`
`Introduction
`I.
`Studies
`Preformulation
`II.
`and Modern Approach
`III. A Systematic
`Tablet Product Design
`and Additives
`IV. Tablet Components
`for Excipients
`V. Regulatory Requirements
`the United States
`VI. References
`
`to
`
`in
`
`Testing
`Preformulation
`Deodatt A, Wadke, Abu T. M. Serajuddin,
`Harold Jacobson
`I.
`II.
`III.
`IV.
`v.
`VI.
`VII.
`VIII.
`IX.
`x.
`XI.
`
`Properties
`
`and Surface Area
`
`Introduction
`Organoleptic
`Purity
`Particle Size, Shape,
`Solubility
`Dissolution
`Parameters Affecting Absorption
`and Polymorphism
`Crystal Properties
`Stability
`Miscellaneous Properties
`Examples of Preformulation
`References
`
`iii
`xi
`
`~i
`
`n
`
`xxx
`
`1
`
`1
`2
`3
`5
`12
`18
`24
`34
`42
`53
`57
`69
`
`75
`
`75
`77
`
`79
`88
`
`121
`128
`
`vii
`
`Page 4
`
`
`
`1
`Pref ormulation Testing
`
`Deodatt A. Wadke, Abu T. M. Serajuddin,
`New Jersey
`E. R. Squibb & Sons, New Brunswick,
`
`and Harold Jacobson
`
`I.
`
`INTRODUCTION
`
`in the rational development of dosage
`is the first step
`testing
`Preformulation
`of physi(cid:173)
`It can be defined as an investigation
`forms of a drug substance.
`alone and when combined
`of a drug substance
`cal and chemical properties
`is to gen(cid:173)
`testing
`The overall objective of preformulation
`with excipients.
`stable and bio(cid:173)
`in developing
`to the formulator
`information useful
`erate
`type of
`the
`Obviously,
`forms that can be mass produced.
`aveilable dosage
`This
`form to be developed.
`information needed will depend on the dosage
`the de(cid:173)
`to support
`program needed
`chapter will describe a preformulation
`forms .
`as dosage
`velopment of tablets and granulations
`the synthetic
`the early development of a new drug substance,
`During
`(includ(cid:173)
`in other disciplines
`alone or in cooperation with specialists
`chemist,
`con(cid:173)
`some data that can be appropriately
`may record
`ing preformulation),
`such
`This early data collection may include
`data.
`as preformulation
`sidered
`chroma(cid:173)
`analysis,
`infrared
`size, melting point,
`as gross particle
`information
`laboratory
`of different
`and other such characterizations
`purity,
`tographic
`and becoming part of,
`in guiding,
`These data are useful
`scale batches.
`the responsible
`between
`Interactions
`the main body of preformulation work.
`at the very
`and pharmacologist
`scientist, medicinal chemist,
`preformulation
`and must also focus
`are to be encouraged
`early stages of drug development
`on the biological data. Review of such data for a series of compounds when
`of the compounds
`chemical properties
`and review of the physical
`available,
`if necessary, would help in the early
`studies
`with some additional probing
`for
`and chemical form of the drug entity
`selection of the correct physical
`further development.
`should start at the point after biologi(cid:173)
`study
`The formal preformulation
`is made for further development of the com(cid:173)
`cal screening, when a decision
`the pre(cid:173)
`Before embarking on a formal program,
`in clinical trials.
`pound
`the following:
`scientist must consider
`formulation
`
`Page 5
`
`
`
`2
`
`Wadke, Serajuddin,
`
`and Jacobson
`
`data
`
`(including
`
`chemical structure,
`
`different
`
`Available physicochemical
`salts available)
`dose
`Anticipated
`(i.e.,
`schedule
`and development
`situation
`Supply
`assay
`of stability -indicating
`Availability
`s hould have or would like
`the formulator
`information
`Nature of the
`to have
`some
`scientist
`the preformulation
`The above considerations will offer
`that need at (cid:173)
`of studies
`the urgency
`and
`types
`the
`in deciding
`quidance
`the success of the preformulation
`to
`is very critical
`Selectivity
`tention.
`for every
`are determined
`parameters
`Not all the preformulation
`program.
`to decide
`Data, as they are generated, must be reviewed
`new compound.
`inves(cid:173)
`a detailed
`For example,
`studies must be undertaken.
`what additional
`On
`soluble compound.
`for a very
`is not warranted
`tigation of dissolution
`the means of
`area, dissolution , and
`surface
`size,
`the other hand, particle
`in the preformu(cid:173)
`considerations
`8l'e important
`rate of dissolution
`enhancing
`soluble drug.
`lation evaluation of a sparingly
`
`time available)
`
`II. ORGANOLEPTIC PROPERTIES
`of the
`the description
`should begin with
`program
`A typical preformulation
`taste of the new drug must be re(cid:173)
`and
`The color, odor,
`substance.
`a stand(cid:173)
`drug
`to establish
`It is important
`terminology.
`corded using descriptive
`to avoid confusion
`in order
`these properties
`to describe
`terminology
`ard
`the same property.
`to describe
`terms
`using different
`among scientists
`the most commonly encountered
`to describe
`terms
`list of some descriptive
`in Table 1.
`is provided
`powders
`and odors of pharmaceutical
`tastes,
`colors,
`The color of all the early batches of the new drug must be recorded
`A record of co:lor of the early batches
`terminology.
`the descriptive
`for later production.
`using
`specifications
`appropriate
`in establishing
`is very useful
`of a
`incorporation
`or variable,
`are undesirable
`When the color attributes
`could be recommended.
`final product
`in the body or coating of the
`dye
`
`A
`
`to Describe
`Suggested Terminology
`Table 1
`of Pharmaceutical
`Properties
`Organoleptic
`Powders
`
`Color
`
`Off-white
`Cream yellow
`Tan
`Shiny
`
`Odor
`
`Pungent
`Sulfurous
`Fruity
`Aromatic
`Odorless
`
`Taste
`
`Acidic
`Bitter
`Bland
`Intense
`Sweet
`Tasteless
`
`Page 6
`
`
`
`s-
`
`l-
`
`1.
`
`s
`ion.
`
`Preformulation Testing
`
`3
`
`In
`odors and tastes.
`have characteristic
`in general,
`Drug substances,
`is considered
`If taste
`due caution must be exerted.
`the new drug,
`tasting
`to the use of a less soluble
`to be given
`ought
`consideration
`as unpalatable,
`the
`of course,
`if one is available-provided,
`form of the drug,
`chemical
`The odor and taste may
`compromised.
`is not unacceptably
`bioavailability
`or by coating
`flavors and excipients
`by using appropriate
`be suppressed
`to
`selected
`and other excipients
`dyes,
`The flavors,
`final product.
`the
`odor
`color, and unpleasant
`or variable
`the problems of unsightly
`alleviate
`and bio(cid:173)
`influence on the stability
`for their
`taste must be screened
`and
`of the active drug .
`availability
`Such
`to skin or sternutatory.
`are irritating
`Many drug substances
`the course of pre(cid:173)
`information may already be available or developed during
`information must be highlighted
`studies, Where available , this
`formulation
`pro(cid:173)
`and personnel
`for material handling
`procedures
`that appropriate
`such
`can be developed.
`tection
`
`Ill.
`
`PURITY
`
`of the purity of a
`scientists must have some perception
`The preformulation
`to rigor (cid:173)
`responsibility
`primary
`individual's
`It is not this
`substance.
`drug
`is an
`this
`that
`(Notwithstanding
`the purity
`and investigate
`ously establish
`in an analytical
`are most often performed
`Such studies
`subject).
`important
`is necessary
`But some early knowledge
`group.
`and development
`research
`and /or early safety and clinical studies
`preformulation
`subsequent
`so that
`is not to mean necessarily
`This
`are not compromised as to their validity.
`some impurity be
`inhomogeneous material or material showing
`relatively
`that
`be
`It does mean that such properties
`studies.
`for preformulation
`rejected
`that allows
`control parameter
`It is another
`and be acceptable.
`recognized
`batches.
`for comparison with subsequent
`can af(cid:173)
`an impurity
`Occasionally
`concerns.
`There are also more direct
`at the level of a few parts per million
`fect stability. Metal contamination
`are
`in which certain classes of compounds
`common example
`is a relatively
`impurity
`area where a slight
`is another
`Appearance
`affected.
`deleteriously
`can be(cid:173)
`Off-color materials upon recystallization
`can have a large effect.
`circumspec(cid:173)
`require
`some impurities
`Further,
`come white in many instances.
`of aromatic amines,
`The presence
`toxic.
`they are potentially
`tion because
`discus(cid:173)
`instances,
`In these
`is an example.
`of being carcinogenic,
`suspected
`so that re(cid:173)
`the material
`sions must be initiated with the chemist preparing
`Very often a problem batch can be made satis(cid:173)
`medial action can be taken.
`factory by a simple recrystallization.
`the purity of a
`for characterizing
`used
`the techniques
`Fortunately,
`in a preformulation
`for other purposes
`drug are the same as those used
`in greater
`study. Most of the techniques mentioned below are described
`the solid state,
`to characterize
`and are used
`in the chapter
`detail elsewhere
`studies.
`or solubility
`tool in stability
`or as an analytical
`liquid chromato(cid:173)
`(TLC) and high-pressure
`chromatography
`Thin-layer
`tools
`and are excellent
`applicability
`(HPLC) are of very wide-ranging
`graphy
`the chemical homogeneity of very many types of materials.
`for characterizing
`in the deter(cid:173)
`are also useful
`and gas chromatography
`Paper chromatography
`mination of chemical homogeneity.
`can be designed
`techniques
`All of these
`index
`such as impurity
`of purity. Measures
`
`to give a quantitative
`(II) and homogeneity
`
`estimate
`index
`
`Page 7
`
`
`
`Wadke, Serajuddin,
`
`and Jacobson
`
`4
`
`53.00
`
`43.61
`
`§ ... 38.22
`w
`0
`:::,
`I-
`:.i 30.83
`a.
`~
`~
`
`23.44
`
`16.05
`0.00
`
`1.50
`
`9.00
`7.50
`6.00
`4.50
`3.00
`RETENTION TIME IN MINUTES
`
`10.50
`
`12.00
`
`Figure 1 HPLC chromatograph
`
`of a typical batch of an experimental
`
`drug.
`
`In
`
`from the HPLC chromato(cid:173)
`especially
`to calculate,
`and easy
`(HI) are useful
`to
`due
`ratio of all responses
`as the
`is defined
`The II of a batch
`graphs.
`the
`Typically,
`response.
`total
`the main one to the
`than
`other
`components
`proce(cid:173)
`in the chromatographic
`as area measurements
`are obtained
`responses
`due
`ratio of the response
`is the
`The obverse of the II is HI, which
`dure.
`Figure 1, which shows an
`response.
`total
`to the
`to the main component
`determination
`illustrates
`drug,
`for an experimental
`HPLC chromatograph
`using a UV detector.
`The chromatograph was generated
`of II and HI.
`time of
`at a retention
`to the main component occurs
`1, peak due
`Figure
`are due
`of 4620. The seven other minor peaks
`4. 39 with an area response
`II
`of 251. Thus,
`response
`impurities with total area
`to the UV-absorbing
`0.0515 =
`and the HI is 1 -
`is 251/(4620 + 251) = 0.0515,
`in this case
`o. 9485.
`a related procedure
`(USP) has proposed
`The United States Pharmacopeia
`In
`using TLC .
`impurities
`that estimates
`test
`impurities
`called ordinary
`due to impurities
`as a ratio of responses
`is defined
`index
`impurity
`test
`this
`of the main
`of a standard
`due to a defined concentration
`response
`to that
`[ 1, 2] •
`limit of 2% impurities
`a general
`The USP is proposing
`component.
`by the USP are not
`as proposed
`index
`impurity
`ll, HI, and the
`The
`(i.e., molecular
`response
`the specific
`since
`of impurity
`absolute measures
`to
`is assumed
`impurity
`to each
`due
`coefficients)
`or extinction
`absorbances
`re(cid:173)
`analysis
`A more accurate
`be the same as that of the main component.
`followed by preparation
`impurity
`of each individual
`identification
`the
`quires
`is almost always un(cid:173)
`information
`Such
`for each one of them.
`of standards
`at the early stages of development.
`available
`and
`are differential
`of purity
`in the assessment
`tools useful
`Other
`a qualit ative
`often provide
`techniques
`These
`thermal analyses.
`gravime tric
`of sol (cid:173)
`evidence of the presence
`and also give direct
`of homogeneity
`picture
`the
`in characterizing
`these methods are simple and are used
`Since
`vates.
`The appearance
`is incidental.
`information
`their use for purity
`material,
`can often be indicative
`of an endotherm
`peaks or the acuteness
`of several
`by
`information may sometimes also be generated
`Similar
`of the purity.
`
`Page 8
`
`
`
`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.
`
`IV. PARTICLE SIZE, SHAPE, AND SURFACE AREA
`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
`biopharmaceutical
`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 1•ate-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 diffi(cid:173)
`cult 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 elect1·ostatic effects
`and other surface active p.roperties
`causing undue stickiness
`and Jack 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 inter(cid:173)
`acting exipients
`than coarse materials. Weng and Parrott
`[ 6] investigated
`influence of particle size of sulfacetamide
`on its reaction with phthalic an(cid:173)
`hyd1.•ide in 1: 2 molar compacts after 3 hr at 95°C. Their data, presented
`in Table 2, clarly demonstrate
`greater
`reactivity of sulfacetamide with de·
`creasing particle
`size.
`Because of these significant
`roles,
`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 µm
`in diameter.
`If the material consists of particles primarily 30 µm 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
`l'ed uce coarse material
`to, preferably,
`the 10-
`to
`
`Page 9
`
`
`
`Wadke, Serajuddin,
`
`and Jacobson
`
`6
`
`of Part i cle Size
`Influence
`Table 2
`of Sulfa cetamide
`on Conversion
`
`size of
`Particle
`sulfacetamide
`(µm)
`
`% Conversion
`± SD
`
`21.54 ± 2.74
`19. 43 ± 3.25
`17.25 ± 2.88
`15. 69 ± 7.90
`9.34 ± 4.41
`
`128
`164
`214
`302
`387
`Source: Modifie d from Weng, H. ,
`J. Pharm . Sci.,
`E. L.,
`and Parrott,
`73: 1059 ( 1984) . Reproduced with
`owne r .
`of copyright
`the permission
`
`ion
`is
`
`te s ting can be per (cid:173)
`controlled
`is accomplished,
`this
`40- µm ra nge . Once
`for in - depth p1•eformulat
`and
`in vivo studies
`for s u bsequent
`formed b oth
`that grin ding
`it may become apparent
`proceed,
`. As the studies
`it
`studies
`time,
`are acceptab l e. At that
`that coar ser materials
`and
`information
`not required
`the
`jeopardizing
`step without
`to omit that
`simpler
`is conceptually
`so that
`the material
`to stage
`is
`concept
`de velo ped . Th e governing
`already
`are maximized.
`challenges
`to grindi n g that may make it inadvisable.
`are s everal drawb acks
`losses when
`There
`are material
`there
`For example,
`Some are of lesse r importanc e.
`the
`buildup occurs, making
`electricity
`is d one . Sometim es a static
`if it exists, may
`g·rinding
`this problem,
`to handle . Often , however,
`material difficult
`to grinding.
`such as lac t os e prior
`by mixing with excipients
`be circumvented
`to aggre(cid:173)
`to too small a dimens ion ofte n leads
`of the pru•ticle size
`the
`Reduction
`lowering
`possibly
`in hydrophobicity,
`in crease
`g ation and an apparent
`are
`troubleso me. When materials
`and making handling more
`rate
`disso lu tion
`size
`in the particle
`for changes
`should be monitored not only
`they
`ground,
`polymorphi c or chemical
`inadvertent
`but also for any
`area,
`change
`and surface
`thereby
`and
`solvates
`. Undue g rindin g can destory
`tr-ans formations
`can
`Some materials
`stics of a substance.
`characteri
`important
`some of the
`reaction.
`a chemical
`also undergo
`
`Particle Size
`
`fo r Determining
`
`A. General Techniques
`The most
`the parti cle size.
`to monitor
`tools are commonly e mployed
`is microscopy . Microsc opy,
`Several
`appraisal
`allowin g for a quick
`rapi d technique
`countin g of a lar g e number of particles when quantitative
`it requires
`(cid:173)
`size determina
`s i nce
`quantitative
`rapid,
`for
`in fo1•mation i s desired , is not suited
`the
`range of sizes and
`the
`in estimating
`, it is very useful
`tions . However
`i s
`if grinding
`to determine
`then be used
`data can
`. The preliminary
`taken botl1 before and after grinding
`shapes
`A phot o microgr aph s hould be
`is fr om abou t 1 µ m upward.
`needed.
`by microscopy
`range of sizes observable
`The
`
`.
`
`Page 10
`
`
`
`Preformulation
`
`Testing
`
`7
`
`it
`
`by suspending
`is best observed
`the material
`For optical microscopy,
`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 amor(cid:173)
`phous
`state after grinding.
`that
`analysis of materials
`size distribution
`For a quantitative
`particle
`al(cid:173)
`is appropriate,
`from about 50 µm, sieving or screening
`range upward
`though
`shape has a strong
`influence
`on the results. Most pharmaceutical
`powders,
`however,
`range
`in size from 1 to 120 µm. To encompass
`these
`ranges,
`a variety of instrumentation
`has been developed.
`There are in -
`struments
`based on lasers
`(Malvern),
`light scattering
`(Royea),
`light block(cid:173)
`age (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 prop(cid:173)
`erly dispersed
`in a suspending medium.
`The concentration
`of sample sus(cid:173)
`pension
`should be such
`that only a single particle
`is presented
`to the sen(cid:173)
`sor in unit
`time,
`thus avoiding coincidence
`counting.
`are also
`Other
`techniques
`based on centrifugation
`and air suspension
`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 techniques
`useful
`for measurement
`of different
`size ranges
`[ 7].
`There are many mathematical expressions
`that can be used
`to charac(cid:173)
`terize an average
`size.
`These
`refer
`to average volumes or weights,
`geo(cid:173)
`metric mean diameters,
`and relationships
`reflecting
`shapes,
`such as the
`ratio of an area
`to a volume or weight
`factor
`[ 8) .
`
`Table 3 Common Techniques
`Measuring Fine Particles of
`Various Sizes
`
`for
`
`Technique
`
`Microscopic
`
`Sieve
`
`Sedimentation
`
`Elutriation
`
`Centrifugal
`
`Permeability
`
`light scattering
`
`size
`Particle
`(µm)
`
`1-100
`
`>50
`
`>1
`
`1-50
`
`<50
`
`>1
`
`0.5-50
`
`,. ,·
`
`Parrott, E. L., Pharm. Mfg .,
`Source:
`4: 31 ( 1985). Reproduced with the
`Permission of. copyright
`owner.
`
`Page 11
`
`
`
`Wadke, Serajuddin,
`
`and Jacobson
`
`Cumulative Weight Percentage at the Indicated Size
`10 !I 2 I O.!I
`99 98 96 90 8070 ~ 30
`99.9
`40
`60
`99.8
`
`20
`
`8
`
`100
`80
`60
`
`40
`30
`
`20 -E
`
`::l.
`w 10
`!:::! 8
`6
`
`(/)
`
`4
`3
`
`2
`
`Log probability
`Figure 2
`amcinolone acetonide.
`
`plot of the size distribution
`
`of a sample of tri(cid:173)
`
`to con(cid:173)
`is
`size distribution
`a particle
`A convenient way to characterize
`is commercially
`graph paper
`Log probability
`plot.
`a log probability
`operation
`struct
`from a grinding
`resulting
`size distributions
`and particle
`available,
`is illus(cid:173)
`An example
`with no cut being discarded will give a linear plot.
`The
`acetonide.
`sample of triamcinolone
`2 for a powder
`in Figure
`trated
`in Table 4.
`2 are presented
`of Figure
`in the construction
`data used
`fractions
`into weight
`in Table 4 are converted
`of particles
`The numbers
`the volume of a single
`by
`and multiplying
`to be spheres
`them
`by assuming
`from the geomeh--ic relationship:
`calculated
`(particle)
`sphere
`
`is
`is
`
`the average
`(using
`diameter
`the volume and d the particle
`where V is
`column of Table 4) . The result
`first
`the
`in
`range given
`value of the
`and
`in each of the size ranges
`total volume occupied by particles
`to
`related
`the
`is directly
`The volume
`table.
`column of the
`third
`the
`the density
`in
`given
`s.ince
`However,
`of the density.
`reciprocal
`the
`term by
`and is rarely known accurate(cid:173)
`a mass
`species
`for all particl,es of a single
`is constant
`the weight percent(cid:173)
`to calculate
`terms
`the volume
`to use
`is sufficient
`total volume of all the particles
`it
`ly,
`the
`range by dividing
`in each size
`If densities were
`ages
`(column 4 of Table 4).
`in each range
`the volumes
`The
`into
`in this calculation.
`they would cancel out
`that
`is obvious
`last column.
`it
`used,
`in the
`is shown
`range
`in each size
`cumulative weight percentage
`most often give a measure of
`of distributions
`descriptions
`Statistical
`in
`are skewed
`the distributions
`However, with powders
`tendency.
`can be described
`central
`type of distribution
`This
`size.
`of increasing
`the direction
`equation:
`by the Hatch-Choate
`
`Page 12
`
`
`
`Preformulation Testing
`
`En
`f = _
`v'2-rr ln o
`g
`
`exp
`
`[
`
`-
`
`2
`
`]
`
`(ln d -
`2 ln
`
`ln M)
`2
`cr
`g
`
`9
`
`( 1)
`
`and
`f is the frequency with which a particle of diameter d occurs,
`where
`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 crg. Equation
`( 1)
`is succi nctly discussed by Orr and Dalla Valle [ 9].
`and
`a distribution,
`The two measures M and erg uniquely
`characterize
`plot in which cumu(cid:173)
`are readily obtained graphically
`from a log probability
`size (Fig. 2) . The
`lative weight percentage
`is plotted against
`the particle
`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
`0g = 50% size
`
`50% size
`= -
`----,-----,--
`15. 87% size
`
`the values are 8. 2 and 1. 5 µm for the geometric mean
`For the example,
`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
`
`in -
`areas of powders has been getting
`of the surface
`The determination
`in recent years.
`The techniques
`employed are relatively
`creasing
`attention
`simple and convenient
`to use, and the data obtained
`reflect
`the particle
`
`Table 4
`Acetonide
`
`Particle Size Distribution
`
`of a Ground Sample of Triamcinolone
`
`Size range
`(µm)
`
`22.5-26.5
`
`18.6-22. 0
`
`14. 9-18. 6
`
`11. 8-14. 9
`
`9. 4-11. 8
`7.4-9.4
`
`5. 9-7. 4
`
`4. 7-5. 9
`3,7-4.7
`
`No. of
`particles
`
`5
`54
`488
`
`2072
`
`5376
`
`9632
`12,544
`12,928
`13,568
`
`Volume of
`particles
`3
`-3
`X 10
`(µm )
`
`Weight
`percent
`in range
`
`Cumulative
`weight
`percent
`
`38
`
`237
`
`1212
`2552
`
`3352
`
`2989
`1888
`
`1008
`
`526
`
`0.2
`
`1.7
`8.8
`
`18.5
`24.3
`
`21.7
`
`13.7
`
`7.3
`
`3.8
`
`100.0
`99.8
`98.1
`89.3
`70.8
`
`46.5
`24.8
`
`11.1
`3.8
`
`Page 13
`
`
`
`10
`
`Wadke, Serajuddin,
`
`and Jacobson
`
`one,
`is an inverse
`to an in(cid:173)
`size leads
`
`in
`
`two par a.meters
`the
`the particle
`reduces
`
`between
`The relationship
`size.
`that
`operation
`that a grinding
`area.
`in the surface
`crease
`is based
`area
`the surface
`for determining
`The most common approach
`An excellent
`of adsorption.
`theory
`(BET)
`on the Brunauer-Emmett-Teller
`involved has been given by
`techniques
`and
`of the principles
`discussion
`will
`that most substances
`states
`theory
`the
`[ 10). Briefly,
`Gregg and Sing
`of partial
`conditions
`certain
`layer of a gas under
`a monomolecular
`adsorb
`capacity
`the monolayer
`Knowing
`(of the g·as) and temperature.
`pressure
`that can be accommodated
`of adsorbate
`(i.e . , the quantity
`of an adsorbent
`the area of
`and
`as a monolayer on the surface of a solid , the adsorbent)
`be calculated.
`in principle,
`area can,
`the surface
`the adsorbate molecule,
`at a specific partial
`is used as the adsorbate
`nitrogen
`Most commonly
`The
`helium.
`typically
`it with an inert gas,
`by mixing
`established
`pressure
`( - 195°C) .
`temperatures
`is carried out at liquid nitrogen
`process
`adsorption
`attainab le
`of nitrogen
`pressure
`at a partial
`that,
`It has been demonstrated
`it i s in a 30% mixture with an inert gas and at - 195°C, a monolayer
`when
`the
`these conditions
`under
`Apparently,
`onto most solids.
`is adsorbed
`be (cid:173)
`for van der Waals forces of attraction
`is sufficient
`of nitrogen
`polarity
`energy
`The kinetic
`to be manifest.
`tile adsorbents
`and
`the adsorbate
`tween
`be (cid:173)
`attraction
`intermolecular
`the
`overwhelms
`these conditions
`under
`present
`the bonding
`to break
`it i s not sufficient
`atoms . However,
`tween nitrogen
`are most often more
`latter
`The
`atom s.
`and dissimilar
`the nitrogen
`between
`The nitrog·en mole(cid:173)
`to van der Waals forces of attracti on.
`polar and prone
`its binding
`and thus
`into chemical combinations,
`cule does not readily enter
`con(cid:173)
`adsorption);
`into a physical
`(i.e . , it enters
`nature
`is of a nonspecific
`role.
`for this
`is well suited
`, the nitrogen molecule
`seq uently
`is
`The BET equation
`
`(2)
`
`Am the value of
`per gram of adsorbent,
`the grams of adsm·bate
`whei•e A is
`gas, Po
`of the absorbate
`pressure
`for a monolayer, P the partial
`ratio
`that
`The
`gas, and C a constant.
`adsorbate
`of the pure
`the vapor pressure
`mea (cid:173)
`as ore P and Po; consequently,
`- dependent,
`constant C i s temperature
`that of
`is
`The eq uation
`conditions.
`isothermal
`are made under
`surements
`1) /
`[ (C -
`the slope
`of the sum of both
`inverse
`and the
`line,
`a straig ht
`it is neces (cid:173)
`In an experiment
`(1/XmC) gives Am·
`the y intercept
`AmCl and
`litera (cid:173)
`from the
`at various values of P; Po can be obtained
`to measure
`sary
`the constant C
`calculated . Often
`readily
`are then
`The other values
`ture.
`to:
`simplifies
`( 2) then
`large and Equation
`is
`
`then
`is
`( e. e. , using only one value of P)
`A single - point determination
`( Am) in a monolayer,
`the specific weight of adsorbate
`Knowing
`possible.
`(SSA) of the sample
`area
`surface
`the specific
`to calculate
`it is possible
`following equation:
`the
`using
`
`Page 14
`
`
`
`Preformulation Testing
`
`SAA=
`
`A NAN
`2
`m
`
`MN
`2
`
`11
`
`where N is the Avogadro number, AN 2 the area of the adsorbate molecule
`16.2 x 10- 20 m2 per molecule),
`(generally
`taken
`to be,
`for nitrogen,
`and
`MN2 the molecular weight of the adsorbate.
`and
`rapidity
`that enable
`Several
`experimental
`approaches
`are available
`rep(cid:173)
`Volumetric
`techniques
`convenience
`as well as accuracy
`and precision.
`instrumentation
`available has
`resent
`the classic approach,
`and the modern
`and dynamic methods are also
`made the procedure
`convenient.
`Gravimetric
`available.
`The latter methods measure
`the adsorption
`process
`by monitor(cid:173)
`ing
`the gas streams,
`using devices
`such as thermal conductivity
`detectors
`and
`transducers.
`in the data given
`is illustrated
`An example using a dynamic method
`to the instrument,
`In addition
`Table 5 for a sample of sodium epicillin.
`the requirements
`are a supply of liquid nitrogen,
`several gas compositions,
`a barometer,
`and several gas-tight
`syringes.
`Briefly,
`the procedure
`en(cid:173)
`tails passing
`the gas over an accurately weighed
`sample contained
`in an
`appropriate
`container
`immersed
`in liquid nitrogen,
`removing
`the
`liquid ni(cid:173)
`trogen 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 practi