Hydration of an amphiphilic excipient, Gelucire 44/14
`Anna Svensson, Carole Neves, Bernard Cabane
`
`To cite this version:
`
`Anna Svensson, Carole Neves, Bernard Cabane. Hydration of an amphiphilic excipient, Gelu-
`cire 44/14. 2004. <hal-00015990>
`
`HAL Id: hal-00015990
`
`https://hal.archives-ouvertes.fr/hal-00015990
`
`Submitted on 15 Dec 2005
`
`HAL is a multi-disciplinary open access
`archive for the deposit and dissemination of sci-
`entific research documents, whether they are pub-
`lished or not. The documents may come from
`teaching and research institutions in France or
`abroad, or from public or private research centers.
`
`L’archive ouverte pluridisciplinaire HAL, est
`destin´ee au d´epˆot et `a la diffusion de documents
`scientifiques de niveau recherche, publi´es ou non,
`´emanant des ´etablissements d’enseignement et de
`recherche fran¸cais ou ´etrangers, des laboratoires
`publics ou priv´es.
`
`MYLAN Ex 1033, Page 1
`
`

`

`Hydration of an amphiphilic excipient, Gelucire 44/14
`
`A. Svensson#, C. Neves∗ and B. Cabane+
`
`# Department of Physical Chemistry 1, Chemical Center, University of Lund, POB 124,
`S221 00 Lund, Sweden
`
`∗ Aventis, Department of Pharmaceutical Sciences, Paris Research Center, 94400 Vitry-sur-
`Seine, France
`
`+ Laboratoire PMMH, CNRS UMR 7636, ESPCI, 10 rue Vauquelin, 75231 Paris Cedex 05,
`France
`
`Abstract
`
`The hydration behavior of an amphiphilic excipient, Gelucire 44/14, has been investigated. Two
`types of hydration processes were studied: one with increasing humidity to investigate the
`conditions during storage, and one with increasing water contents to study the behavior during
`dissolution. In addition, the main components of the excipient were investigated separately.
`These were polyethylene glycol (PEG), PEG monolaurate and PEG dilaurate (PEG esters),
`trilaurin (glyceride) and glycerol. The water uptake of Gelucire 44/14 at humidity ratios less than
`60%RH was very low (about 1 wt%), which was attributed to the dissolution of the most
`hydrophilic component, glycerol. The water uptake increased substantially above 70%RH as
`PEG started to dissolve, followed by the PEG esters. It was concluded that each component
`equilibrates separately with the aqueous solution, which itself is in equilibrium with the humid
`air. Hence, a liquid phase can form between the crystals with a chemical potential decided by the
`humidity ratio. The water uptake of Gelucire 44/14 could be described as a sum of the uptake of
`the individual components, weighted according to their relative amounts in the mixture. Phase
`maps of the Gelucire 44/14 and its components at different water contents were constructed. Dry
`Gelucire 44/14 contains lamellar crystals of mainly PEG and PEG esters which melt at 44 ºC.
`The crystals do not swell at increasing humidity, but dissolve above 75%RH at a water content of
`5 wt% in the excipient. At increasing water contents Gelucire 44/14 forms white gels composed
`of hexagonal and lamellar mesophases dispersed in a continuous liquid phase. These liquid
`crystalline phases dissolve at 35 ºC, i.e. below physiological temperatures. A dramatic viscosity
`maximum was observed in the lamellar region at 50 wt% water, which may be attributed to the
`formation of networks of PEG esters. The pure PEG esters were found to form cubic
`mesophases at 50 wt% water. The instruments used in this study were Dynamic Vapor Sorption
`(DVS), Thermal Gravimetric Analysis (TGA), Differential Scanning Calorimetry (DSC), Small-
`and Wide Angle X-ray Scattering (SWAXS) and Optical Microscopy.
`
`1
`
`MYLAN Ex 1033, Page 2
`
`

`

`Keywords
`
`Gelucire 44/14; phase diagram; hydration; DVS; SWAX
`
`1. Introduction
`
`Gelucires are a group of amphiphilic excipients which have been widely studied as controlled
`release matrices (Mouricout et al., 1990). The incorporation of drugs into Gelucires has been
`reported to increase the dissolution rate of poorly soluble drugs, often leading to improved drug
`bioavailability (Gines et al., 1995; Damian et al., 2000; Perissutti et al., 2000; Gupta et al., 2001).
`One compound from this group is Gelucire 44/14. This amphiphilic excipient has a
`Hydrophilic-Lipophilic Balance of 14 and a melting temperature of 44 °C, hence its name
`(Roussin et al., 1997).
`
`In pharmaceutical applications, it is important to know how the excipient interacts with the drug,
`and how the mixture behaves during manufacturing, storage as well as during administration.
`These behaviours depend on the effects of temperature and hydration on the physical state of the
`excipient. For instance, dissolution or dispersion of a drug in the excipient can be modified
`according to water content; conversely, the synthesized drug may be hydrated, and this water
`may alter the physical state of the excipient (Damian et al., 2002; Sutananta et al., 1994a;
`Jeanmaire-Wolf et al., 1990). The excipient-drug mixture may then be stored at a certain relative
`humidity, and moisture form the air may have an effect on its physical state. Finally, during
`administration, the mixture will be immersed in body fluids, and it is important to know what are
`the pathways for dissolution in this aqueous medium.
`
`There have been a few studies of the thermal behaviour of dry Gelucires (Craig et al., 1991;
`Sutananta et al., 1994b), but, to our knowledge, no systematic study of their hydration behaviour.
`The aim of this project was to investigate the behaviour of the Gelucire 44/14 when exposed to
`humidity and water, at ambient and physiological temperatures. We studied two hydration
`processes, corresponding respectively to the conditions encountered during storage and during
`dissolution of excipients. During storage, the excipients may be kept in equilibrium with an
`atmosphere of constant relative humidity (RH). Accordingly, we have measured the mass of
`water absorbed as a function of RH, through DVS experiments. During dissolution, the
`excipients are mixed with liquid water. Accordingly, we have also investigated the state of
`Gelucire 44/14 mixed with known amounts of water, or individual components mixed with
`known amounts of water. The nature of the phases (crystal, liquid crystal or isotropic solution)
`that are formed at selected compositions and different temperatures was determined, and their
`structural parameters were also measured.
`
`2
`
`MYLAN Ex 1033, Page 3
`
`

`

`Since Gelucire itself is a mixture, its hydration behaviour is determined by the behaviours of its
`components, and their interactions. Thus, in addition to Gelucire 44/14, we also investigated the
`pure components separately in order to distinguish their specific effects when mixed together.
`Gelucire 44/14 is composed of polyethylene glycol 33 (PEG 33), PEG mono- and diesters of
`fatty acids, glycerides and a small amount of glycerol. The most common fatty acid chain in the
`mixture is laurate. Consequently, we decided to investigate the following chemicals: PEG 33,
`PEG monolaurate, PEG dilaurate and Trilaurin (triglyceride with laurate chains). By using
`thermal analysis, microscopy and X-ray diffraction, we investigated the phase behaviour of
`Gelucire 44/14 and compared it with the phase behaviour of the simple components. In the end,
`we found that the hydration behaviour of Gelucire 44/14 can be explained by the behaviours of
`its components.
`
`2. Materials and methods
`
`Gelucire 44/14 (Gattefosse s.a) is produced by the reaction of hydrogenated palm kernel oil and
`polyethylene glycol, PEG 33 (1500 g/moL). It contains mostly fatty acids of the lauric type (i.e.
`C12 chains) of which more than 80 % are saturated. The final composition in the Gelucire 44/14
`is 72 wt% PEG esters, 20 wt% glycerides, 8 wt% pure PEG and 2% glycerol. The PEG esters
`are composed by PEG mono- and diesters and the glycerides by mono-, di or triglycerides.
`Laurate acid chains are the most common type with 40-50 %. The most common chemicals in
`the Gelucire 44/14 mixture are PEG, PEG mono- and dilaurate (the PEG esters), mono-, di and
`trilaurates (the glycerides). PEG 33 and trilaurin (triglyceride with laurate chains) were
`purchased from Sigma. PEG monolaurate and PEG dilaurate were produced by organic
`synthesis.
`
`PEG ester synthesis
`
`5 g PEG 33 (0,0035 mol) was dissolved in 25 ml ethanol free dichloromethane in a flask. 0,5 ml
`triethylamin (0,0035 mole) was added to buffer the reaction medium. The flask was put in an ice
`bath (0 °C). The reaction started by the addition of 0,8 ml lauryl chloride (0,0035 mole). After 2
`hours, the mixture contained the products PEG 33, PEG monolaurate and PEG dilaurate.
`Separation of the products was made by column chromatography on silica gel with a solvent
`containing dichloromethane:methanol with the proportions 9:1. Finally, the products were
`washed in cyclohexan to take away excess reactants. The structures of the PEG esters were
`confirmed by nuclear magnetic resonance, mass- and infrared spectrometry.
`
`Preparation of samples at different humidities
`
`Samples of Gelucire 44/14 were exposed to humidity ratios 0-97 % RH, in order to investigate
`the humidity effects on the properties of the excipient. The samples were equilibrated in
`
`3
`
`MYLAN Ex 1033, Page 4
`
`

`

`desiccators containing different saturated salt solutions: silica gel (0%RH), Mg(NO3)2
`(50%RH), CuCl2 (68%RH), NaCl (75%RH), KCl (84%RH) and K2SO4 (97%RH).
`
`Preparation of samples with high water contents
`
`To investigate the phase behaviour at high water contents, samples were prepared by mixing
`directly with water. Samples with weight ratios of component and water from 10/90 to 75/25
`were prepared. The component and water were weighed in small glass containers with a scale
`and sealed with a cap. The samples were mixed by hand and using an ultra sonic water bath.
`They were left to equilibrate for several weeks before investigation.
`
`Dynamic vapour sorption (DVS)
`
`The dynamic water sorption experiments were made with a DVS thermo-hygro-gravimetric
`balance under constant nitrogen sweeping and constant temperature (25 °C). The sample (assay
`sample of about 10 mg) was placed in a glass capsule. It was exposed to a series of relative
`humidity during a specific time interval at which the absorption of water is measured in weight.
`The steps are in 10 %RH. The maximum time at each relative humidity is six hours. If
`equilibrium is reached within this time limit, i.e. the derivative of weight vs. time is <0.02, the
`measurement automatically moves on to the next humidity.
`
`Small and Wide Angle X-ray Scattering (SWAXS)
`
`To determine the presence of structures in the samples, X-ray diffraction at small and wide
`angles was used. The SWAXS measurements were performed with a Kratky compact small and
`wide angle system equipped with a linear collimation system and two position sensitive detectors
`(Hecus M Braun, Austria). Each detector contains 1024 channels of width 54.0 µm. A
`monochromator with a nickel filter was used to select the Cu-K_ radiation (λ = 1.541 Å)
`provided by the generator. The generator, a Seiffert ID-3003 X-ray, was operating at 50 kV and
`40 mA. The sample was enclosed in a steel sample holder with mica windows. The distance
`between the sample and the detector was 279 nm. The measurements were made at room
`temperature (20°C) unless measurement with temperature increase (20-60°C) was performed.
`
`Thermal Gravimetric Analysis (TGA)
`
`A sample with a mass between 1.5 and 4 mg was deposited in an open 75-_L aluminum crucible.
`The sample was heated from 20°C to 250°C at a rate of 1°C/min. The analysis was carried out in
`a nitrogen stream.
`
`4
`
`MYLAN Ex 1033, Page 5
`
`

`

`3. Results
`
`In this section, the physical state and the hydration of Gelucire 44/14 are compared with those of
`the pure components. The aim is to find out how the excipient takes up water. Two alternative
`behaviours may take place : (a) the different components in the excipient are segregated, and they
`take up water independently ; (b) the components in the excipient are associated, forming solid
`or liquid solutions which take up water more readily (or less readily) than the segregated
`mixtures. It is also possible that the hydration process starts with the former behaviour
`(hydration of separated components) and then changes to the latter one through dissolution of 2
`of more components in the same aqueous solution.
`
`General view of the hydration processes
`
`A general overview of the hydration processes is given by the water sorption isotherms measured
`through DVS experiments (Figure 1). Each isotherm shows the relative change of mass
`according to RH valued. Of particular interest are RH values around 50 %, which correspond to
`usual storage conditions. At RH = 70 % and above, the samples did not reach equilibrium
`hydration within the time of the DVS experiments. In this range, the DVS experiments were
`supplemented by TGA experiments where the samples were equilibrated against water vapour
`from saturated salt solutions. The differences between both sets of results were found to be
`small, and well within experimental errors. Moreover, the hydrations measured through both
`experiments for PEG 33, up to RH = 90 %, are in good agreement with the values determined
`through osmotic pressure experiments by Rand and coworkers [yyy] (Figure 2). Since osmotic
`stress experiments yield true equilibrium hydrations, we may conclude that both DVS and TGA
`results gave equilibrium hydrations up to RH = 90 %.
`
`3
`
`2
`
`1
`
`0
`
`0
`
`10 20
`
`30
`
`40
`
`50
`
`60
`
`70
`
`80
`
`90
`
`100
`
`5
`
`MYLAN Ex 1033, Page 6
`
`

`

`Gelucire 44/14
`
`PEG 33
`
`Trilaurin
`
`PEG monolaurate
`
`PEG dilaurate
`
`Glycerol
`
`Calculated curve
`
`160
`
`140
`
`120
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Change in mass (%)
`
`0
`
`10
`
`20
`
`30
`
`40
`
`50
`
`60
`
`70
`
`80
`
`90
`
`100
`
`Target % RH
`
`Figure 1. Dynamic water sorption isotherms for Gelucire 44/14, PEG 33, PEG monolaurate,
`PEG dilaurate, trilaurin. and glycerol. The calculated curve gives the change in mass as a sum of
`the weighted masses of the components according to their relative amounts in Gelucire 44/14
`(see equation /1/ in the text).
`
`The water sorption isotherms of Gelucire 44/14 show a limited uptake of water at low relative
`humidities. For instance, at RH = 50 %, which corresponds to ambient humidity conditions, the
`uptake amounts to 1 wt% of the mass of Gelucire 44/14. This amount is also the maximum
`water uptake that is allowed for Gelucire 44/14 by the European Pharmacopia (Eur.
`Phramacopia, 2002). In this range Gelucire 44/14 is a white waxy solid.
`
`At higher relative humidities, the uptake of water is much larger. At RH = 75 %, TGA
`experiments indicate that Gelucire 44/14 contains 5 wt% water, but it still is a white waxy solid.
`At RH = 84 %, Gelucire 44/14 contains 11 wt% water, and it has turned into a white dispersion.
`Finally, at RH = 97 %, Gelucire 44/14 contains 40 wt% water, and it has turned into a very
`viscous dispersion, i.e. there is actually a viscosity increase with rising humidity.
`
`It is possible to interpret this hydration process as a sum of the hydrations of the individual
`components. For instance, Gelucire contains a small amount of glycerol, which adsorbs
`significant amounts of water, even at low RH (Figure 1). If it is assumed that glycerol constitutes
`3 wt% of the mass of Gelucire 44/14, and that it absorbs water independently from the other
`components, then the uptake of water into liquid glycerol at this RH would amount to 0.7 wt% of
`the Gelucire mass. Thus, at low RH, most of the hydration of Gelucire 44/14 may be caused by
`water absorption into liquid glycerol. Similar results are obtained up to RH values of 68-70%,
`
`6
`
`MYLAN Ex 1033, Page 7
`
`

`

`where most of the water uptake (2 wt% of the gelucire mass) may still be caused by water
`absorption into glycerol.
`
`Similarly, the faster rise of the water content of Gelucire 44/14 at RH values above 70 % is
`related to a surge in hydration of the other components, e.g. PEG 33 beyond RH = 70 % and the
`PEG esters beyond RH = 80 % (Figure 1). Over the whole range of relative humidities, the water
`sorption isotherm of Gelucire 44/14 can be reproduced by assuming that each component
`absorbed water independently. Thus, the water uptake by Gelucire, X, was calculated as the sum
`of contributions of the individual components weighted by their relative amounts [xxx] in
`Gelucire 44/14 :
`
`X = 0.03 Xglycerol + 0.07 XPEG33 + 0.71 (0.5 Xmonolaurate + 0.5 Xdilaurate) + 0.19 Xtrilaurin
`
`/1/
`
`This calculated isotherm is displayed in Figure 1. It is seen that, over the whole range of relative
`humidities, the calculated isotherm matches the measured water sorption isotherm of Gelucire
`44/14. Therefore the water uptake of Gelucire 44/14 can be reproduced by assuming that each
`component absorbs water independently.
`
`In the next sections, we try to find out whether the components actually take up water
`independently. For this purpose, we compare the physical states of the components at each stage
`of the hydration process with that of the excipient.
`
`Hydration of the individual components
`
`Glycerol
`
`At room temperature, glycerol is a very hygroscopic liquid (indeed, it is used in many topical
`formulations as a hydrating agent). The amount of water absorbed by glycerol, as a function of
`relative humidity, was measured by DVS (Figure 1). At 50 % RH, the water uptake is already 24
`wt% of the glycerol mass. At high RH, water sorption by glycerol still rises, but not as fast as
`that of the other components, and since the proportion of glycerol in the excipient is small this is
`no longer the main contribution to the total hydration.
`
`PEG 33
`
`PEG 33 is a crystalline solid at ambient conditions. WAXS experiments gave a sharp diffraction
`peak at a spacing of 4.6 Å, corresponding to the lateral distance between neighbouring chains
`and a small peak at 3.3 Å. SAXS experiments gave a weak diffraction peak at a real space repeat
`distance of 106.1 Å, corresponding to the repetition of successive lamellae (data not shown).
`This peak is weak because the region that separates successive lamellae has a small volume and a
`low contrast. A weak second order of that diffraction peak is also visible.
`
`7
`
`MYLAN Ex 1033, Page 8
`
`

`

`In equilibrium with a humid atmosphere, the PEG crystals do not swell : DVS experiments show
`that, for RH < 50 %, the water uptake is less than 0.2 wt% of the PEG mass. However, at RH =
`50 and 60 %, some water uptake is measured, up to 2 wt% of the PEG mass (Figure 1). In the
`TGA experiments, the crystal – solution equilibrium was observed at RH = 75 %. At higher
`relative humidities, up to RH = 90%, the hydrations measured through DVS and through TGA
`are in good agreement with the values measured through osmotic stress (Figure 2).
`
`osmo pressure
`
`TGA
`
`DVS
`
`fit1
`
`RH dissolution
`
`crystal
`
`solution
`
`200
`
`180
`
`160
`
`140
`
`120
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Uptake of water by PEG 33 (%)
`
`60
`
`70
`
`80
`
`90
`
`100
`
`Relative humidity
`
`Figure 2. Equilibrium hydration of PEG 33 according to relative humidity. Filled diamonds :
`Osmotic pressure measurements listed by Peter Rand [yyy]. Open triangles : this work, DVS
`experiments. Open squares : this work, TGA experiments.
`
`PEG esters
`
`The PEG esters also form lamellar crystals at ambient conditions. The WAXS spectra are shown
`in Figure 3. They are similar to the those obtained for PEG 33, with a sharp peak at 4.6 Å and a
`small peak at 3.3 Å, indicating that the PEG chains have the same organization. The SAXS
`spectra show strong diffraction peaks at real space repeat distances of 117.1 Å (for PEG 33
`monolaurate) and 113.7 Å (for PEG 33 dilaurate) (Figure 4). These peaks are much stronger
`than those of PEG 33. This indicates that the PEG lamella is sandwiched between layers that
`contain the alkyl chains ; the strong contrast of electronic density between the PEG regions and
`the alkyl regions is the cause of the high intensity of the diffraction peaks.
`
`8
`
`MYLAN Ex 1033, Page 9
`
`

`

`3
`
`3.5
`
`4
`
`4.5
`
`5
`
`5.5
`
`d (Angstroms)
`
`1800
`
`1600
`
`1400
`
`1200
`
`1000
`
`800
`
`600
`
`400
`
`200
`
`Intensity
`
`0
`
`2.5
`
`Figure 3. WAXS spectra of dry PEG dilaurate showing the peaks corresponding to the repetion
`of PEG chains inside a lamella.
`
`PEG monolaurathe
`
`PEG dilaurathe
`
`60000
`
`50000
`
`40000
`
`30000
`
`20000
`
`10000
`
`0
`
`Intensity
`
`7962
`
`257
`
`131
`
`87
`
`66
`
`53
`
`44
`
`d (Angstrom)
`
`Figure 4. SAXS spectra of dry PEG esters showing the successive diffraction orders
`corresponding to the repetition of lamellae in a crystal.
`
`9
`
`MYLAN Ex 1033, Page 10
`
`

`

`In equilibrium with a humid atmosphere, the crystals of PEG esters do not swell : DVS
`experiments show that, for all RH below 60 %, the water uptake is less than 0.3 wt% of the mass
`of PEG esters. Both esters begin to absorb water at RH above 80 %, and at RH = 90 % the
`strong water uptake indicates that dissolution is taking place. In TGA experiments performed
`with both esters at RH = 97 % we could see that all crystals had combined with water from the
`atmosphere to form a solution.
`
`Other experiments were performed by mixing PEG esters with liquid water. These samples were
`found to form either crystals, mesophases (liquid crystals) or isotropic liquids (solutions). The
`phase map of the PEG monolaurate/water system is shown in Figure 5. At 20 °C, samples
`containing less than 30 wt% water were mixtures of crystals with a saturated solution, as in the
`case of PEG 33. However, with increasing amounts of water, in the range 45-60 wt%, a cubic
`mesophase was formed. This phase appears as a transparent and hard gel, and does not show
`birefringence when observed between crossed polarizers, i.e. it is optically isotropic. SAXS
`experiments performed on this cubic phase gave a set of diffraction peaks with spacings in the
`ratios √4, √5, √10, √15. The unit cell size is a = 135 Å when the water content is 45 wt % (i.e.
`RH = 97 %) and swells with increasing water content to 143 Å at 50 wt % water.
`
`10
`
`MYLAN Ex 1033, Page 11
`
`

`

`50
`
`45
`
`40
`
`35
`
`30
`
`Cryst + L
`
`L
`
`Cub
`
` L
`
`25
`
`20
`
`0
`
`10
`
`20
`
`30
`
`40
`
`50
`
`60
`
`70
`
`80
`
`90
`
`100
`
`wt% water
`
`Temperature (°C)
`
`(a)
`
`(b)
`
`Cryst + Hex
`
`Hex
`
`Cub
`
`Cub + L
`
`L
`
`0
`
`10
`
`20
`
`30
`
`40
`
`50
`
`60
`
`70
`
`80
`
`90
`
`100
`
`wt% water
`
`50
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`Temperature (°C)
`
`Figure 5. Phase maps of the PEG monolaurate /water (a) and PEG dilaurate /water (b) systems
`at temperatures 20-50 °C. (Cryst. = lamellar crystal, L = liquid solution, Cub = cubic mesophase,
`Hex = hexagonal mesophase).
`
`11
`
`MYLAN Ex 1033, Page 12
`
`

`

`The phase map of the PEG dilaurate is also shown in Figure 5. At 20 °C, samples containing
`less than 25 wt% water were mixtures of crystals with a hexagonal mesophase. The hexagonal
`mesophase appears as a transparent and viscous gel, and show birefringence when observed
`between crossed polarizers, i.e. it is optically anisotropic. The pure hexagonal mesophase was
`found for samples containing 27 – 31 wt% water. SAXS experiments performed on this
`hexagonal phase gave a set of diffraction peaks with spacings in the ratios 1, √3, √4, and a unit
`cell size a = 62 Å. At higher water contents (40-60 wt%), a cubic mesophase was found, with a
`unit cell size similar to that of the monolaurate (a = 130 Å).
`
`Since Gelucire contains a mixture of the PEG monolaurate and dilaurate, we also investigated the
`phase behaviour of mixtures containing both esters and water. The ratio in the mixture was 1/3
`PEG monolaurate and 2/3 PEG dilaurate as a result of the conditions during the PEG ester
`synthesis. It was found that the extension of the mesophases was reduced compared with the
`pure esters : the hexagonal phase was not formed at all, and the cubic phase was found only
`between 43 and 47 wt% water.
`
`Trilaurin
`
`Trilaurin is also a crystalline solid at T = 20 °C. WAXS spectra of trilaurin gave the
`characteristic peaks of the β form (real spacings of 4.6, 3.85, and 3.70 Å), followed by some
`peaks of other unidentified polymorphs. SAXS spectra gave a main peak at a real space repeat
`distance of 32 Å, corresponding to the length of 2 laurate chains.
`
`Trilaurin is a very hydrophobic excipient. In equilibrium with a humid atmosphere, the crystals
`take up very little water, and this absorption is quite slow (slower than the time scales of the DVS
`experiments). In equilibrum with liquid water, the amount of water absorbed in the powder is
`only 2 wt% of the mass of trilaurin. Even the melted trilaurin does not mix with water to any
`significant extent.
`
`Hydration of Gelucire 44/14
`
`Dry state
`
`The dry Gelucire 44/14 is a white and waxy solid at 20 °C. According to melting enthalpies, its
`degree of crystallinity is 83 %. WAXS spectra show the same peaks as in PEG 33 and in the
`PEG esters, located at real space distances of 4.6 and 3.3 Å. Accordingly, the PEG chains in
`Gelucire are crystallized. SAXS experiments show a lamellar structure with a repeat distance of
`124.4 Å (Figure 6). This repeat distance is close to that of the PEG esters ; the small difference
`in repeat distance is presumably caused by a difference in the average molar mass of the PEG
`chains. The relative intensities of the peaks are also similar to those of the PEG esters, indicating
`that the content of the unit cell are quite similar. Accordingly, the crystal structure of the
`
`12
`
`MYLAN Ex 1033, Page 13
`
`

`

`excipient must be made of PEG lamellae separated by layers of fatty acid chains, as in the PEG
`esters.
`
`0% RH
`
`75% RH
`
`84% RH
`
`60000
`
`50000
`
`40000
`
`30000
`
`20000
`
`10000
`
`0
`
`Intensity
`
`0
`
`265
`
`133
`
`88
`
`66
`
`53
`
`44
`
`38
`
`33
`
`d (Angstroms)
`
`Figure 6. SAXS spectra of Gelucire 44/14 at different relative humidities. The diffraction peaks
`correspond to the repetition of lamellae consituted mainly of PEG esters.
`
`Hydration at ambient temperature
`
`At relative humidities up to RH = 75 %, the excipient remains crystallized : indeed, its SAXS
`spectra are still dominated by the peaks form the PEG ester crystals (Figure 6). This is
`consistent with the fact that the PEG esters do not dissolve until the relative humidity exceeds 84
`%. The excipient remains a waxy solid, and its water content reaches 5 wt% at RH = 75 %.
`
`At RH = 84 %, the SAXS experiment performed on the excipient shows a strong drop in the
`intensity of the PEG ester crystal peak. This is consistent wit the fact that the PEG 33 is fully
`dissolved, and PEG 33 monolaurate partly dissolved at this RH. The excipient becomes a white
`gel, and its water content reaches 11 wt%.
`
`Higher water contents were obtained by mixing the excipient with known amounts of water.
`These samples had the appearance of white gels (11 wt% water), white dispersions (25-31 wt%),
`white gels (50-61 wt%), and white dispersions again (71-75 wt%). At 90 wt% water, the sample
`separated into a white and a transparent liquid phase.
`
`13
`
`MYLAN Ex 1033, Page 14
`
`

`

`In this range of water contents, dramatic changes were observed in the WAXS and SAXS
`spectra. The WAXS peaks from the crystallyzed PEG esters had vanished, indicating that the
`PEG chains were in a liquid state. The SAXS spectra indicated the formation of mesophases
`(Figure 7). At 25 wt% and 31 wt% water, a hexagonal mesophase was found. The unit cell size
`of this mesophase was 72 Å, somewhat larger than that of the pure PEG diester. At 50 wt% and
`60 wt% water, a lamellar mesophase was found, with a unit cell size of 83 Å. This repeat
`distance is shorter than that of the crystallized lamellae (124.4 Å for Gelucire), which is expected
`since the PEG chains are now in a liquid state.
`
`11 wt% water
`
`25 wt% water
`
`50 wt% water
`
`90 wt% water
`
`12000
`
`10000
`
`8000
`
`6000
`
`4000
`
`2000
`
`0
`
`Intensity
`
`7962
`
`257
`
`131
`
`87
`
`66
`53
`d (Angstroms)
`
`44
`
`38
`
`33
`
`Figure 7. SAXS spectra of Gelucire 44/14 with different water contents.
`
`Inspection of the samples using optical microscopy with crossed polarizers revealed that they
`contained birefringent particles (the mesophases) dispersed in a non birefringent fluid, or in a
`weakly birefringent fluid (for the gels formed at 50-61 wt% water). Thus, all mesophases
`coexisted with another phase, which explains the white appearance of the samples.
`
`Hydration at elevated temperatures
`
`At higher temperatures, it was found that stronger water adsorption and dissolution take place.
`DSC experiments performed on samples with water content 2 % gave a curve that was quite
`close to that of the dry material, thereby confirming that the main components (PEG esters and
`glycerides) were still in a solid state up to T = 45 °C. However, experiments performed on
`samples with water content 5 % showed a strong endothermal event at 30 - 35 °C instead of the
`
`14
`
`MYLAN Ex 1033, Page 15
`
`

`

`fusion peak observed at 45 °C in the dry excipient. Accordingly, the combined effects of
`hydration (5 wt% water) and temperature (35 °C) cause the dissolution of the PEG esters.
`
`A phase map of Gelucire was constructed using the results of all experiments made by mixing
`Gelucire 44/14 with known amounts of water (Figure 8). At low temperatures, this map shows
`that the dissolution of Gelucire must proceed through the formation of a set of mesophases,
`some of which are quite viscous (the hexagonal phase and the gel found at 50 wt% water). Thus,
`dissolution must be a slow process. On the other hand, at physiological temperatures, this map
`shows that all mesophases have melted : thus the dissolution becomes a fast process.
`
`Cryst + L
`
`Hex + L
`
`Gel
`
`Lam + L
`
`L
`
`0
`
`10
`
`20
`
`30
`
`40
`
`50
`60
`wt% water
`
`70
`
`80
`
`90
`
`100
`
`50
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`Temperature (°C)
`
`Figure 8. Phase map of Gelucire 44/14 in water at temperatures 20-50 °C. (Cryst. = lamellar
`crystal, L = liquid solution, Hex = hexagonal mesophase, Lam = lamellar mesophase).
`
`15
`
`MYLAN Ex 1033, Page 16
`
`

`

`4. Discussion
`
`Dry state
`
`The WAXS and SAXS experiments show that Gelucire 44/14 is made mainly of lamellar
`crystals that contain the PEG esters and some pure PEG. In these crystals, the PEG chains form
`lamellae with a thickness that is determined by the length of a PEG chain in a helical
`configuration (bbb). These lamellae are separated by layers that contain the alkyl chains; the
`period of this assembly is 124 Å. In addition, there may be trilaurin crystals (but they were not
`resolved in our x-ray experiments on dry Gelucire), amorphous regions (less than 17 % of the
`total), and some liquid glycerol.
`
`Hydration through humid air (DVS experiments)
`
`The DVS and TGA experiments show that the hydration of Gelucire 44/14 follows a simple
`sequence. First, the most hydrophilic component, which is glycerol, absorbs water while all
`others remain crystalline and do not swell. The water uptake is about 1% over most of the range
`of RH which may be encountered during storage of the excipient (RH = 30 – 60 %). Then, at a
`critical RH (RH = 70 % according to DVS, see the insert in Figure 1), the water uptake speeds
`up dramatically. This may be related to the dissolution of the next most hydrophilic component,
`which is PEG 33 : indeed, crystals of pure PEG 33 dissolve when the RH reaches 75 %, and
`mixtures may dissolve at a slightly lower humidity. Water is then absorbed by a solution
`containing glycerol and PEG 33. The other components (PEG esters and glycerides) do not
`swell or dissolve until the RH is quite high (RH > 80 %).
`
`This sequence is exactly what would be expected for a system that contains a set of crystals in
`equilibrium with a humid atmosphere and with an aqueous solution : in this case, each
`
`component equilibrates separately with the aqueous solution, which is itself driven by the
`exchanges with the humid atmosphere. The thresholds for dissolution of the crystals may be
`shifted slightly with respect to those in a pure system, because the aqueous solution is a
`multicomponent mixture and not a 2 component solution : for instance, at RH = 75 %, the
`chemical potential of PEG 33 in the solution may not be exactly that in a binary solution,
`because the solution actually contains glycerol as well.
`
`Hydration through mixing with liquid water
`
`Small amounts of water
`
`For pharmaceutical uses, the excipient may be mixed with an active ingredient (AI). Some active
`ingredients contain a few water molecules per molecule of the AI. This water will then be
`equilibrated with all the components of the excipient, and the chemical potential of water will be
`
`16
`
`MYLAN Ex 1033, Page 17
`
`

`

`the same throughout the mixture. Thus, the same distribution of water will be achieved as that
`which is obtained when the mixture is exposed to humid air, which we have measured (indeed,
`relative humidity also measures the chemical potential of water).
`
`For instance, assume that the total quantity of water released by the AI amounts to 5 wt% of the
`excipient mass. This total hydration is also obtained when Gelucire is exposed to humid air at
`RH = 75 %. In these conditions, our TGA measurements indicate that glycerol and PEG 33 will
`form a solution with water, while the other components will take up very little water (see also the
`DVS results sh