International Journal of Pharmaceutics, 62 (1990) 87-95 Elsevier IJP 02152
`
`Review Article
`
`The molecular basis of moisture effects on the physical and chemical stability of drugs in the solid state
`
`Claes Al&reck
`
`* and George
`
`‘Zografi
`
`School of Pharmacy, University of ~~scons~~-~~~son, Madison, WI 53706 (U.X.4.J (Received 29 December 1989) (Modified version received 2 April 1990) (Accepted 13 April 1990)
`
`transition;
`Key words: Moisture sorption; Glass transition; Crystal defect; Solid-state
`Solid-state degradation; Molecular mobility; Drug-excipient
`interaction; Water sorption
`
`Introduetiou
`
`is well recognized that residual water associ-
`ated with drugs in the solid state can have signifi-
`cant effects on a variety of physical and chemical
`properties, such as chemical degradation, dissolu-
`tion rate, flow and compactibility.
`Such residual
`water exists because of prolonged exposure to an
`atmosphere containing water vapor, or as a result
`of processing
`that involves the use of water, e.g.
`lyophilization, spray drying, aqueous film coating,
`wet granulation or recrystallization. To develop
`strategies
`that can be used to deal with such
`physical
`and chemical
`changes, or more
`im-
`portantly,
`to anticipate
`them, requires an under-
`standing of the molecular events underlying such
`solid-state phenomena and of the means by which
`water molecules can influence these events. Conse-
`quently, in this brief review we have attempted to
`bring together a body of literature and concepts
`that we believe can provide the basis for address-
`ing these important pharmaceutical problems
`less
`
`Corr~~ondence: G. Zografi, School of Pharmacy, University of Wi~ons~-Mason, Madison, WI 53706, U.S.A. * On leave from the Department of Pharmaceutics, Uppsala University, S-751 23, Uppsala, Sweden.
`
`empirically and with greater understanding. Em-
`phasis will be placed on the role of water
`in
`affecting drug entities
`that are believed
`to exist
`predo~nantly
`in the crystalline state, in the ab-
`sence and presence of excipients and other drugs
`in the formulation.
`
`Mechanisms of Water-Solid Interactions
`
`to think of water as being able
`It is convenient
`to interact with crystalline
`solids in three major
`ways: adsorption of water vapor to the solid-air
`interface; crystal hydrate formation; and deliques-
`cence. For solids cont~ng microvoid spaces it is
`also possible
`for capillary condensation
`to occur
`at fairly
`low relative humidities,
`leading
`to oc-
`cluded water (El-Sabaawi
`and Pei, 1977; Carsten-
`sen et al., 1980). Two of these processes, deliques-
`cence and capillary condensation,
`lead to the for-
`mation of condensed or bulk water, capable of
`dissolving water-soluble
`components. Crystal hy-
`drates are characterized
`by
`the penetration of
`water molecules into the crystal lattice, most often,
`but not always, in a well-defined molecular posi-
`tion within the unit cell, and hydrogen bonded to
`certain groups with a specific stoichiometry
`(Bym,
`1982). The nature of the stoichiometry, position of
`
`IPR2018-00126
`
`Page 1 of 9
`
`I-MAK 1019
`
`87
`It
`0378-5173/90/$03.50 0 1990 Elsevier Science Publishers B.V. (Biomedical Division)
`

`

`88 the water molecules and the strength of the inter- action determine the extent to which such water can enter or leave the crystal unit cell under a given set of conditions. Water molecules adsorbed to the surface of the solid generally exist, as ex- pected for physically adsorbed monolayers, with a first layer hydrogen bonded to the solid, and at most 2-3 additional molecular layers formed at the higher relative humidities, Such adsorption generally is readily reversed by small increases in the temperature or by small decreases in the rela- tive humidity (Thiel and Madey, 1987). Recognition of the various mechanisms by which water can interact with water-soluble drugs in the solid state has led to some important per- ceptions and misperceptions of how water affects the properties of such solids. Given, for instance, that condensed water is produced during deliques- cence and that such water continues to dissolve the solid as long as a sufficiently high relative hu~dity is maintained, i.e. a relative hu~~ty in excess of that for a saturated solution of the material, there is a general perception that small amounts of water below this point, somewhere in excess of a monolayer also can cause small amounts of ‘surface dissolution’ of the solid, which in turn, can trigger a variety of more subtle and slow physical and chemical changes. This, for ex- ample, is the basis for the model of Leeson and Mattocks (1958) for drug degradation in the solid state, where the drug is visualized to exist as a saturated solution around the solid particle and where the rate of degradation is determined by the aqueous solubility of the drug and the first-order solution rate constant. From a thermodynamic perspective it can be shown that dissolution of a crystalline water-soluble solid should not occur in any water presenf until the chemical potential of the water is equal to that of a saturated solution (Van Campen et al., 1983). Indeed, it is possible repeatedly to adsorb and desorb water vapor on freshly crystallized sodium chloride up to its criti- cal relative humidity of 76% and observe no changes in the amounts of adsorbed water and no physical changes in the solid (Barraclough and Hall, 1974; Kontny et al., 1987). Related to this issue also is the common per- ception that a certain proportion of water mole- cules associated with a solid is unavailable to ‘dissolve’ the solid and to cause changes in the system. Water, for example, present in crystal hydrates or adsorbed in the first surface layer in direct contact with the solid is often thought of as ‘ tightly bound’ and unavailable for ‘dissolution’ or ‘interaction’, while more nonspecifically surface bound water, in excess of a monolayer, as dis- cussed above, is thought of as being available to act as a solvent for dissolution and other changes in the solid. It is important in this context, how- ever, to recognize that the degree of hydrogen bonding and the strength of hydrogen bonds be- tween water and solid, under ambient conditions, can be very variable and that this va~abi~ty can lead to many situations where under different conditions water molecules directly bound to the solid can be shown to be quite mobile and free to move around on a long time scale, either within the crystal lattice or along the solid surface (Jelin- ski et al., 1983; Zografi, 1988). Thus, one cannot assume that such water is absolutely ‘frozen’ into a static tightly bound state with no possible role to play in affecting solid properties. In the context of these observations, therefore, how do we explain the fact that adsorbed water associated with crys- talline solids, in relatively low amounts, appears to promote chemical degradation or other types of physical changes, but that below a certain level of water these phenomena do not occur?
`
`To put the issues raised so far for water and crystalline solids into a broader context, it will be helpful to review briefly a few concepts concern- ing the ~teractions of water with amo~hous solids, where considerably more water is taken up, relative to the crystalline form of the same chem- ical entity (Nakai et al., 1977; Pikal et al., 1978). Here, because of the disordered state of the solid it is possible for water to dim&e in the solid. Thus, in contrast to adsorption, where the amount of water taken up depends on the available surface area, uptake by amorphous solids is predomi- nantly determined by the total mass of amorphous solid. Critical to our interest in the effects of water
`
`Water and Amorphous Solids
`
`IPR2018-00126
`
`Page 2 of 9
`
`I-MAK 1019
`
`

`

`Tm To loo x water 0 Fig. 1. Solute-water state diagram which illustrates the effect of water plasticization and its effect on Ts. on solid properties is the fact that water dissolved in an amorphous solid, can act as a plasticizer to greatly increase the free volume of the solid by reducing hydrogen bonding between adjoining molecules of the solid, with a corresponding re- duction in its glass transition temperature, Tg (Franks, 1982; Levine and Slade, 1987,1988; Slade and Levine, 1988). In Fig. 1 is given a schematic representation of this change for a typical amorphous solid having very high water solubility and a high Tg in the dry state. Thus, water with a very low Tg , - 134 o C, increasingly and continu- ally reduces the Tg of the solid system as its concentration in the solid increases. As with any completely miscible mixture, at low concentration of water, we can think of the system as a solution of water in an amorphous solid, whereas as the amount of water in the system increases we can think of this more as a solution of the solid in amorphous water. What is important here, is that the change in free volume occurring as one in- creases the temperature, T, above Tg (or by de- creasing Tg below T), has a profound effect on a number of properties related to it. Of particular interest to us are the significant changes which take place in the viscoelasticity of the solid as one passes from the glassy state below Tg to the rub- bery state above Tg. For example, as shown from the WLF-equation (Williams-Landel-Ferry, 1955) for the viscosity of an amorphous rubber solid, going just 20 o C above T5 will cause the viscosity to change from about 10 P at Tg to lo8 P, with a very significant increase in the molecular mobility 89 of the solid and water. Thus, the mixture of the two amorphous components, if kept below the T,-line in Fig. 1, will remain as an extremely viscous immobilized glassy solution where water molecules in this highly immobilized state behave as if they were in a tightly bound state. If, how- ever, the temperature of the system is allowed to go above the T,-line a significantly less viscous rubbery state will be formed with greatly en- hanced molecular mobility of both the solid and water. The implications of this in terms of product stability are shown in Fig. 2 where a plot of Tg vs relative humidity exposed to the solid is given for the amorphous polymer poly(vinylpyrrolidone), PVP (Oksanen, 1989; Oksanen and Zografi, 1990). Here, we can see, for example, that a sample of PVP stored at 25” C and 80% relative humidity would have its Tg decreased to approx. 10 o C and hence would have been converted from a glassy to a rubbery state. At 40 o C such a conversion only would require storage at about 65% relative humidity. The increased mobility which occurs as Tg is reduced to values near and below the operat- ing temperature has been shown to be sufficient to allow amorphous solids to readily undergo solid- state chemical reactions (Pikal et al., 1977) and to support the recrystallization of small molecules rendered amorphous through various types of processing, e.g. lyophilization, mechanical grind- ing or rapid precipitation (Makower, 1956; Palmer, E 150 Rubber
`
`20 40 60 60 100 Relative Humidity (X) Fig. 2. Effect of relative humidity on the glass transition temperature of PVP K30. The box illustrates conditions nor- mally in use during accelerated storage testing. (Data compiled from Oksanen (1989)).
`
`01
`0
`
`IPR2018-00126
`
`Page 3 of 9
`
`I-MAK 1019
`
`

`

`1956; Otsuka and Kaneniwa, 1983; Fukuoka et
`al., 1986, 1989). Thus,
`in these cases
`it is not
`surprising
`that water
`is often suggested,
`incor-
`rectly, to have attained ‘solvent-like’ or ‘unbound
`properties when it produces such chemical and
`physical changes.
`
`Molecular Disorder in Crystalline Solids
`
`It is well recognized that the regular and repeat-
`ing arrangements of atoms and molecules
`in a
`crystalline
`state are most often altered by
`the
`presence of defects,
`imperfections or regions of
`amorphous structure. Such defects or imperfec-
`tions, which give rise to local regions of molecular
`disorder, relative to that of the crystal structure,
`can arise
`from a variety of processes used
`in
`pharmaceutical development,
`including mechani-
`cal grinding,
`lyophilization,
`and other processes
`where rapid drying and recrystallization
`occur.
`Since
`the molecules
`located
`in such regions of
`local disorder can exhibit greater chemical reactiv-
`ity (Htittenrauch,
`1978, 1983, 1988; Hersey and
`Krycer, 1980; Htittenrauch et al., 1985) and ‘solu-
`bility’ (Waltersson and Lundgren 1985), they are
`often said
`to be
`in an
`‘activated
`state’. Such
`‘activation’ arises from a combination of greater
`molecular mobility and the exposure of more reac-
`tive chemical groups. Taken
`to the extreme of a
`complete
`lack of
`long-range order
`in the solid
`state, the system will be completely amorphous, as
`described above. As in the case of amorphous
`solids,
`the regions of greater
`local disorder and
`reactivity,
`should exhibit an ability
`to take up
`more water than would ordinarily be adsorbed on
`the surface of the crystalline portions of the solid.
`If the amount of water taken up is sufficient
`to
`plasticize the local region to a point where
`molecular mobility can be high enough to support
`enhanced dissolution rates (Fukuoka et al., 1986)
`and chemical reactivity. Direct evidence
`for this
`plasticizing effect of water on alkali halide crystals
`comes from studies of surface electrical conduc-
`tance as a function of exposure to various relative
`humidities
`(Asselmeyer and Zott, 1965; Knacke
`and Neuschutz, 1970). Here,
`it was shown that
`significant
`increases
`in electrical conductance oc-
`
`relative
`the critical
`cur near to, but still below,
`humidity, RH,,
`for deliquescence, at a level well
`below that of a saturated salt solution. The fact
`that the conductance near RH,
`increases signifi-
`cantly but does not equal
`that of a saturated
`solution is consistent with an increasing
`ionic mo-
`bility due to plasticization as water dissolves in a
`metastable manner into the defects and other dis-
`ordered regions of the crystal. In this regard, also,
`it is interesting
`to note that, whereas the maxi-
`mum number of water molecules adsorbed on
`crystalline surfaces usually amounts to about three
`equivalent monolayers
`(Barraclough
`and Hall,
`1974; Thiel and Madey, 1987; Kontny et al., 1987)
`in one study where NaCl crystals were subjected
`to mild comminution,
`as much as much as five
`equivalent monolayers were observed
`just below
`RH,
`(Walter, 1971).
`‘hot spots’ on the particle
`The role of energetic
`surfaces or thin amorphous layers surrounding the
`particles
`in acting as sites for chemical degrada-
`tion of solids particularly at higher temperatures
`has been extensively discussed (Prout and Tomp-
`kins, 1944; Ng, 1975; Hasegawa et al., 1975).
`What has not been considered
`is the fact
`that
`relatively small amounts of moisture absorbed into
`these hot spots can produce significant
`increases
`in molecular mobility at much lower temperatures
`by means of its plasticizing properties. Thus, rather
`than think of water as having solubilized the drug
`in order to influence chemical degradation rates,
`we can think of water as dissolving
`in the local
`regions of disordered molecules to produce enough
`mobility to support chemical reactivity. Therefore,
`in most chemical
`reactions
`in the solid state of
`pharmaceutical
`interest, e.g. hydrolysis or oxida-
`tion, water can act as both a plasticizer and a
`chemical reactant dissolved in the drug.
`the
`Visualizing
`the role of water, in affecting
`properties of ‘crystalline’
`solids,
`in this manner,
`also helps to explain a number of other observa-
`tions related to the physical stability of drugs. For
`example,
`it has been shown that many crystalline
`water-soluble solids after grinding or compaction,
`insufficient
`to produce measurable
`amorphous
`structure using powder X-ray diffraction, and ex-
`posure to relative humidities below the deliques-
`cence point exhibit significant changes in specific
`
`IPR2018-00126
`
`Page 4 of 9
`
`I-MAK 1019
`
`90
`Tg < T,
`

`

`area, due to what appears to be surface sintering
`and recrystallization
`(Chikazawa et al., 1972ab,
`1976; Kaiho et al., 1973, 1974; Kontny et al.,
`1987; Ahlneck
`and Alderborn,
`1989b).
`Such
`sintering,
`furthermore, has been shown to most
`likely be the cause of increased
`tablet hardness
`upon exposure
`to various
`relative humidities,
`(Lordi and Shiromani,
`1983, 1984; Down and
`McMullen, 1985; Ahlneck and Alderbom, 1989ab)
`as well as the cause of decreased dissolution rates
`after storage at elevated temperatures and relative
`humidities
`(Danjo and Otsuka, 1988).
`Indepen-
`dent measurements
`of water uptake by such
`activated solids, indeed, reveal that they sorb more
`water per unit area than an untreated crystalline
`sample at low relative humidity
`(Kontny et al.,
`1987), and that above a certain point,
`local re-
`crystallization or surface sintering
`is induced by
`the plasticizing effects of the absorbed water, just
`as described above for amorphous so!ids when
`is brought down below
`To put this discussion on a more quantitative
`basis, it would be useful to carry out some model
`calculations
`showing the extent to which a given
`total moisture content
`in a sample might cause
`physical or chemical changes to occur in a crystal-
`line material. In Table 1 are given values of par-
`ticle size and specific surface area of solid spheres
`along with the number of layers of water mole-
`cules requires to cover these surfaces if 0.1% (0.001
`g of water per g of solid) is adsorbed. Such a level
`of moisture is quite common in crystalline drugs.
`A cross-sectional area of 0.125 nm2 per molecule
`of water is assumed for the purposes of this calcu-
`lation. As seen in Table 1 for solids having par-
`ticle sizes normally encountered
`in solid dosage
`
`TABLE I
`
`Specific surface area of sucrose spheres and theoretical number of water layers surrounding the spheres if 0.1% (w/w) moisture is a&orbed
`
`Particle size
`(pm)
`
`SW
`(cm2 g-‘)
`
`Number of water layers
`
`10
`38
`100
`
`38000
`3800
`1000
`380
`
`1.1
`11
`42
`110
`
`91
`
`TABLE 2
`
`Moisture content in the amorphow portion of sucrose and the glass transition temperature if a total of 0.1 or 0.5% moisture is taken up
`
`Amount of Amount of Moisture content Glass
`moisture
`amorphous
`in amorphous
`transition
`material
`temperature ’
`material
`(W
`(mg f-f@/
`(“C)
`100 mg
`solid)
`
`(W
`
`0.1
`
`0.5
`
`0.5
`1
`2.5
`5
`
`0.5
`1
`2.5
`5
`
`20
`10
`4
`2
`
`100
`50
`20
`10
`
`9
`21
`45
`49
`
`-13
`-36
`9
`2-l
`
`a Compiled from Slade and Levine (1988), assuming that the
`glass transition temperature of purely amorphous sucrose is
`52OC.
`
`pm, the number of adsorbed
`i.e. lo-100
`forms,
`layers of water is quite extensive and far beyond
`what one normally would expect for surface ad-
`sorption on such particles.
`Indeed,
`true surface
`adsorption of water at 25°C on well-defined solid
`surfaces has been shown to be orders-of-magni-
`tude less than such a value (Kontny et al., 1987).
`What appears to be more likely is that, in addition
`to any occluded water that might be present in the
`crystal, a significant portion of this water is taken
`up and dissolved into the ‘disordered regions’, i.e.
`water is concentrated
`in these regions.
`In Table 2, we have calculated
`the amount of
`water absorbed
`into the ‘disordered’ amorphous-
`like regions of sucrose
`for 0.1 and 0.5%
`total
`moisture content, assuming various percentages of
`amorphous structure between 0.5 and 5% of the
`total solid, and further assuming that essentially
`all of the water is preferentially
`taken up in these
`regions. As expected, depending on the amount of
`amorphous material present,
`there can be a con-
`siderable concentration
`of water for a given sys-
`tem, particularly
`as the fraction of amorphous
`material becomes quite small. To examine how
`such water
`contents might affect
`the
`of
`amorphous sucrose
`in these samples, and hence,
`
`IPR2018-00126
`
`Page 5 of 9
`
`I-MAK 1019
`
`Tg
`T.
`1
`Tg
`

`

`the molecular mobility of sucrose molecules, we
`have estimated
`for sucrose as a function of
`water content from the data of Slade and Levine
`(1988), as also shown in Table 2. Here, we can see
`that only 0.1% total moisture,
`if concentrated
`in
`the 1% of the mass of the solid that is amorphous,
`can lower the
`of sucrose to a value approaching
`room temperature, while 0.5% total moisture (not
`uncommon
`in many pharmaceutical
`systems) can
`produce much lower values of
`and therefore,
`regions of very high molecular mobility at room
`temperature.
`We would conclude from this analysis that the
`effects of water on the solid-state properties of
`drugs are directly linked to the extent to which the
`solid contains
`regions of higher energy, higher
`molecular disorder and higher molecular mobility.
`Water absorbed
`into these regions can plasticize
`the solid and further promote the molecular mo-
`bility needed to support chemical degradation, as
`well as solid-state phase changes, such as recrys-
`tallization. Attention
`to the possible activation of
`molecules, when crystalline solids undergo stress-
`ful processes such as milling or lyophilization,
`in
`this regard, is considered essential
`for an under-
`standing of why they undergo the chemical-physi-
`cal instabilities so often observed at relatively
`low
`moisture contents. Since moisture and tempera-
`ture both play an
`important
`role
`in affecting
`molecular mobility,
`it is particularly
`important
`to
`keep these principles
`in mind when carrying out
`accelerated
`stability studies at elevated
`tempera-
`tures and relative humidities.
`
`Drug-Excipient Interactions
`
`that many drugs in the
`It is well established
`solid state undergo significant physical chemical
`change in the presence of certain solid excipients.
`Of most significance are: (1) an increased rate of
`chemical degradation
`(Carstensen,
`1974, 1988;
`Akers, 1976; Ahlneck and Lundgren, 1985);
`(2) a
`reduction
`in the degree of crystallinity
`(Nakai et
`al., 1978; Cotton et al., 1988; Ishizaka et al., 1988,
`1989); and (3) the formation of molecular com-
`plexes (Kararli et al., 1989).
`
`two ways that water can be
`There are basically
`involved in drug-excipient
`interactions. First, water
`brought
`into
`the product by the excipient
`can
`redistribute via the vapor phase and become asso-
`ciated with the drug by means of adsorption or
`absorption. Second, sorbed water located at the
`points of physical contact between drug and ex-
`cipient can facilitate an interaction between
`the
`drug and the excipient. Such interactions might
`help to catalyze chemical degradation rates or to
`stabilize amorphous or activated structures of the
`drug against recrystallization
`to lower energy solid
`forms.
`In the first case
`it is not necessary
`for
`direct contact
`to occur since the total water pre-
`sent in the system will redistribute via the vapor
`phase,
`as predicted
`from
`the water vapor
`sorption-desorption
`isotherm of each
`ingredient
`(Zografi et al., 1988). Thus, the effect of the ex-
`cipient only depends on the amount of excipient
`present and, hence
`the amount of moisture
`it
`brings into the closed system, as well as the rela-
`tive ability of each solid to take up and retain
`water at a particular
`temperature
`and relative
`humidity. Such behavior most likely will assume
`importance when
`the excipients
`contain
`large
`amounts of water which can escape into the head
`space of a closed container
`to produce significant
`relative humidities,
`i.e. with crystal hydrates and
`amorphous and partially amorphous polar poly-
`mers, having water contents of approx. 2-20%
`(e.g. Zografi and Kontny, 1986). A good example
`of such behavior would be that of the redistribu-
`tion of water from gelatin capsules
`into a solid
`capsule formulation and the change caused in the
`mechanical
`integrity of the capsule (Kontny and
`Mu&i,
`1989). In all such cases the excipient sim-
`ply acts as a source of relative humidity which, in
`turn, causes water vapor to associate with the drug
`to produce
`its effects. The mechanism of these
`effects would be the same as discussed earlier,
`where water acts as a plasticizer of locally dis-
`ordered regions in the crystal.
`In the second case, the effects of excipients on
`the properties of drugs are directly
`linked to the
`interfacial area of contact between drug and ex-
`cipient,
`in addition to the total amount of excipi-
`ent present
`(Jain et al., 1982; Ahlneck and Al-
`derborn, 1988). Generally,
`therefore as the ratio of
`
`IPR2018-00126
`
`Page 6 of 9
`
`I-MAK 1019
`
`92
`Tg
`Tg
`Tg
`

`

`93 drug to excipient decreases, and the true area of contact increases, the effects of the excipient will in&r-ease to a m~um (Cotton et al., f988)* This maximum generally appears to occur somewhere between a ratio of 1: 3 and 1: 10. A very common perception of the molecular events occuring dur- ing many drug-excipient interactions of this type would have the moisture present dissolving the drug, and perhaps also the
`
`extipien% m a&m the
`
`two to mix
`
`and interact. When the excipient is not one which is soluble in water, it has been sug- gested (Carstensen, 1969) that drug molecules dis- solved in the water might be adsorbed to the excipient surface> forming a layer of molecules, now more suspeetible to chemical and physical change. Thus, in such a case the excipient would simply act as a means of supporting the less- ordered adsorbed layer. in the earlier discussion concerning drug mole- cules and water, it was s~~est~ tha& gisen the small boots of water needed to support physi- cal and chemical change of crystalline drug mole- cules, dissolution of the solid into the sorbed water is highly unlikely. Rather, it would appear that water, by being preferentially taken up by disordered regiorrs of the solid, acts as a pfasti&zer to produce a significant increase in molecular mo- bility. It seems reasonable therefore, to expect that water located at the interface between drug and excipient, likewise, could plasticize both the ex- cipient and the drug to facilitate any interactions or phase changes that might occur. If
`
`chemical reaction or, simply, stabilize metastable solid forms, If the excipient or drug exhibit acid- base properties, dissociation and pH changes could easily occur if plasticization provided sufficient molsular mobility_ In situations where chemical d~~adatio~ occurs it is also important to recog- nize that degradation products accumulating at the interface could further plasticize the system, leading to even greater rates of chemical change. Since increasing the temperature also facilitates molecular mobility, we would expect high temper- ature, in combination with high relative humidity, as used in most accelerated stability testing of solid dosage forms, to greatly influence any tend- encies for such drug-excipient interactions to OC- cur. This means, therefore, that caution must be taken to notice if a glass-to-rubber transition takes place during these tests. Otherwise, this can lead to ~~nte~re~~o~s of test data. Conclusions Central to an ~de~st~ding of the effect of water on the physical chemical properties of drugs in solid dosage forms is an understanding of how water behaves in such systems as a plasticizer. These effects are clearly operating when amorphous drugs and excipients are present. However, they are most likely also operating when so-called crystalline drugs have been activated by v&arts processes such as grinding, lyophilization, wet granulation or recrystallization to produce regions of partial amorphous structure or local disorder. In the activated state such molecules are more suspectible to physical and chemical change in both the absence and presence of various types of pharmaceutical excipients. Thus, in considering the effects of small or residual amounts of water on drugs and drug products, rather than speaking of dissolution in water, or of bound and unbound water, it is preferabie to think in terms of plasti- cized disordered regions ~ont~~i~g water, drug and, possibly excipients, all with varying degrees of molecular mobility and hence varying degrees of physical and chemical reactivity. The authors wish to thank Rorer Group, Inc. and the Squibb Institute for Medical Research, for their support of this research. We also would like to thank Dr. Felix Franks, of PAFRA, Ltd, Rio- preservation Division, for valuable discussions. C,A, gratefully acknowledges the Swedish Academy of Pharmaceutical Sciences, Pharmacia AR, the Swedish
`
`enough
`
`plasticization takes place, it is even possible for drug molecules to actually move into the structure of a plasticized polymer to form a ternary drug excipient-water ‘solution’ in the amorphous solid state. In this situation the polymer would provide an approp~ate ~~ro~e~t which could catalyze
`
`a
`
`fnstitute,
`
`IPR2018-00126
`
`Page 7 of 9
`
`I-MAK 1019
`
`IF : s St&else, the Wen-
`

`

`94 nergren-Center and the
`
`Faculty of
`Pharmacy, Up- psala University for supporting his stay in Ma&
`XXI.
`
`Refwtmces
`
`Ahlneck, C. and Iundgren, P., Methods for the evaluation of solid state stability and compatibility between drug and excipient. Acru Pfiarm Suet., 22 (I985) 303-314. Ahlrreck, C. and Alderborn, G., Solid state stability of acetyl- salicyclic acid in binary mixtures with microcrystalline and microfine cellulose. A&r PZazrm. Suce., 25 (1938) 41-52. Ablneck, C. and Alderborn, G., Moisture adsorption and tabletting. I. Effect on volume reduction Properties and tablet strength for some crystalline materiaIs. Znt. J. Phurm., 54 (1989a) 131-141. Ahlneck, C. and Alderborn, G., Moisture adsorption and t&letting. II. The effect on tensile strength and air permea- bility of the relative humidity during storage af tablets of three crystalline materials. Znt. S. Pharm., 56 (1989b) 143- 150. Akers, M.J., Preformulation testing of solid oral dosage form drugs. Methodology, management and evaluation. Can J. Pharm Sci., 1 (1976) l-10. Asselmeyer, F. and Zott, H., Sorption von wasserdampf an NaCl-oberflacben Z. Angerv. P&r., 19 (IWS) 168-175. Barraclough, P.B. and Hall, PG., The adsorption of water vapour by lithium Euoride, sodium fluoride and sodium chloride, Surjuce Sci., 46 (1979) 393-417. Bym, S,,
`
`Salid State Chemistry of
`
`J.T.,
`
`Drugs, Academic Press, New York, 1982, pp 7-10. Carstensen,
`Stability of solids and solid dosage forms. J. Pharm. Sri., 63 (1974) l-14. Carstensen, J.T., Effects of moisture on the stability of solid
`
`dosage forms.
`Carstenseq
`J.T., Osadca,
`
`14 (1988) 1927-1969.
`
`M. and Rubin, S.H., Degradation mechanisms for water-soluble drugs in solid dosage forms. .Z. PAann. sci., 58 (169) 549-553. Carsteesen, J.T., Tome’, P., Van Campen, L. and Zografi, G., Hygroscopicity of poorly soluble porous substances. J. Phann. sci., 69 (1980) 742-744. Chikazawa, M., Kaiho, M. and Kanazawa, T., An X-ray dif- fractometric study of hygroscopic process of alkali halides.
`
`Powder Techno/.,
`
`42 (1985) 169-174. El-Sabaawi, M. and Pei,
`
`D.C.T., Moisture isotherms of hygro-
`scopic parous solids. Znd.
`
`Pharm. ML, 34 (1986)
`
`Pharm. Bull., 37 (1989) 1047-1050.
`
`J. Pharm. Technol. & Prod.
`
`Munuf., 1 (1980) 18-2X
`
`l-10.
`Hiittenrauch, R., Fricke, S. and Zielke, P., Mechanical activa-
`tion of pharmaceutical
`systems.
`
`Pkurm. Phnmrard.,
`
`Nippon Kagaku K&&i,
`(1973) 914-917. Kaibo, M., Chikazawa, M. and Kanazawq
`
`T.,
`
`Relationship between adsorption of water vapor on ground sodium ebloride and caking of the salt. Nippon Kugaku Kaishi, (1974) 233-238.
`
`(X976) 410-414. Cotton, ML., Wu, D.W. and Vadas, EB., Drug-excipient interaction study of enalapril maleate using thermal analy- sis and acanning electron microscopy. Znnt. J. Pharm., 40 (1987) 129-142. Danjo, K. and Otsuka, A., The effect of temperature on diametral compression strength of S-phenylbutazone and barbital (form II) tablets. Chem. Pharm. Bull., 36 (1988) 763-768. Down, G.R.B. and McMullen, J.N., The effect of interpar- ticulate friction and moisture on the crushing strength of sodium chloride compacts.
`
`Eng. Chem. Fund., 16 (1977) 321-326. Franks, F., Water - A Comprehensive Treatise, Vol. 7, Plenum, New York, 1982, pp. 215-338. Fukuoka, E., Makita, M. and Yamamura~ S., Some physicochemical properties of glassy indomethacin. Ckem.
`4314-4321. Fukuoka, E., Makita, M. and Yamamura, S., Glassy state of pharmaceuticals. III. Thermal properties and stability of glassy pharmaceuticals and their binary glass systems. Chem.
`Hasegawa, J., Hanano, M. and Awazu, S., Decomposition of acetylsahcyclic acid and its derivatives in solid state. Chem. Phorm. BUN., 23 (1975) 86-97. Hersey, J.