An Investigation into the Role of Polymeric Carriers ... - ACS Publications

Feb 18, 2015 - Yiwei TianDavid S. JonesConor DonnellyTimothy BranniganShu LiGavin P. Andrews. Molecular Pharmaceutics 2018 15 (4), 1379-1391...
0 downloads 0 Views 3MB Size
Subscriber access provided by George Washington University Libraries

Article

An Investigation into the Role of Polymeric Carriers on Crystal Growth within Amorphous Solid Dispersion Systems. Yiwei Tian, David S. Jones, and Gavin P. Andrews Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/mp500702s • Publication Date (Web): 18 Feb 2015 Downloaded from http://pubs.acs.org on February 23, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

An Investigation into the Role of Polymeric Carriers on Crystal Growth within Amorphous Solid Dispersion Systems.

Yiwei Tian, David S. Jones, and Gavin P. Andrews*

The Drug Delivery and Biomaterials Group, School of Pharmacy, Medical Biology Centre, Queen's University Belfast, 97 Lisburn Road, Belfast, BT9 7BL, Northern Ireland, United Kingdom

*Correspondence to: Gavin P. Andrews (Tel: +44(0) 28 9097 2646; Fax: +44 (0)289024 7794; E-mail: [email protected])

1 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Using phase diagrams derived from Flory-Huggins theory, the thermodynamic state of amorphous felodipine within three different polymeric carriers has been defined. Variation in the solubility and miscibility of felodipine within different polymeric materials (using F-H theory) has been identified and used to select the most suitable polymeric carriers for the production of amorphous drug polymer solid dispersions. Using this information, amorphous felodipine solid dispersions were manufactured using three different polymeric materials (HPMCAS-HF, Soluplus® and PVP K15) at pre-defined drug loadings, and the crystal growth rates of felodipine from these solid dispersions were investigated. Crystallization of amorphous felodipine was studied using Raman spectral imaging and polarised light microscopy. Using this data we examined the correlation between several characteristics of solid dispersions to the crystal growth rate of felodipine. An exponential relationship was found to exist between drug loading and crystal growth rate. Moreover, crystal growth within all selected amorphous drug polymer solid dispersion systems were viscosity dependent, (η-ξ). The exponent, ξ, was estimated to be 1.36 at a temperature of 80°C. Values of ξ exceeding 1, may indicate strong viscosity dependent crystal growth in the amorphous drug polymer solid dispersion systems. We argue that, the elevated exponent value (ξ > 1) is a result of drug-polymer mixing which leads to less fragile amorphous drugpolymer solid dispersion system. All systems investigated displayed an upper critical solution temperature and the solid-liquid boundary was always higher than the spinodal decomposition curve. Furthermore, for PVP-FD amorphous dispersions at drug loadings exceeding 0.6 volume ratio, the mechanism of phase separation within the metastable zone was found to be driven by nucleation and growth rather than liquid-liquid separation. Key words Felodipine, Amorphous, Solid Dispersions, Flory-Huggins, Nucleation and Growth, Crystal Growth Kinetics, Viscosity, Raman Spectral Mapping, PVP, Soluplus, HPMCAS.

2 ACS Paragon Plus Environment

Page 2 of 49

Page 3 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

INTRODUCTION Organic amorphous systems and supercooled liquids are of interest in a number of fields including, polymer, food, metallurgy and pharmaceuticals. Over the last two decades, there has been increased interest in using the amorphous state to solve the limited water solubility of many new chemical entities being developed within the pharmaceutical industry (1, 2). Amorphous drugs exist in a higher free energy state and their tendency to revert to their most stable form via recrystallization is inevitable. In order to retain the advantages associated with the amorphous form, recrystallization must be avoided, over a pharmaceutically relevant shelf life. It is widely accepted that polymeric additives can improve the physical stability of the amorphous solid. One strategy to delay or prevent recrystallization is to add polymeric carriers and to form a solid dispersion. There are a number of techniques, such as micro-precipitation, ball milling, spray drying or hot-melt extrusion that could be used to achieve this (3, 4). Although, the amorphous drug generated during this process may be solubilised and/or dispersed within the polymeric matrix, it will still be unstable and phase separation and/or crystallisation may occur even at a low temperatures (5-7). It also has been reported that homogenous amorphous solid dispersions may rapidly precipitate into their crystalline form during dissolution and the amorphous solubility advantage dramatically reduced (8). The complex behaviour of drug-polymer amorphous solid dispersions during storage and dissolution has led to a significant barrier to commercial use of such systems. Currently, only a limited number of formulations have been successfully commercialised. The number of products does not directly correlate to the extensive research and development within this field (4, 9). Clearly, efficient development of stable amorphous drug polymer solid dispersions is required if these systems are to become commercially viable. This would require logical and rational methods in the selection of the most suitable polymeric carrier for individual active pharmaceutical ingredients and a definition of the maximum drug solubility and miscibility within the carrier (10). Moreover, an improved understanding of the phase separation of binary amorphous systems is also of paramount importance. A thermodynamic model, originally developed by Flory and Huggins, principally allows one to describe amorphous-amorphous phase behaviour of a polymer within a given solvent, or alternatively another polymer or small 3 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

molecule. This model has also been used in computational modelling to predict the stability of small molecule-small molecule systems (11). In previous publications, we have proposed a small scale screening method using differential scanning calorimetry (DSC) in combination with this theory to construct drug-polymer temperaturecomposition binary phase diagrams (10). Informatively, such phase diagrams reveal important parameters for solid drug dispersion systems, including the level of drugpolymer interaction (interaction parameter χ), the temperature dependence of the liquid-solid curve, amorphous drug-polymer miscibility and the level of drug supersaturation within a defined polymeric material. This valuable information aids our understanding of the phase behaviour of amorphous drug-amorphous polymer systems and may facilitate better understanding of separation during storage. However, understanding the thermodynamic miscibility of solid dispersion systems is only part of the problem. Crystallisation kinetics of amorphous drug and the contribution of previously discussed thermodynamic parameters are still poorly understood. Establishing links between thermodynamic and kinetic aspects of these systems is at the forefront of research in this field. Although certain factors have been related to physical stability such as drug-polymer intermolecular interactions, the Tg of pure polymer, the Tg of drug-polymer solid dispersion, and molecular mobility of the system (12-16), there are limited studies linking thermodynamic state and amorphous drug recrystallization (17, 18), especially in the presence of polymer (19). Lauer et al., (2011) have attempted to address amorphous-amorphous phase separation based upon atomic force microscopy (AFM) (20). The appearance of an amorphous continuous phase was observed during moisture induced storage studies. The surface enhanced drug and polymer amorphous phase separation was successfully observed through light modified AFM. In this study (Lauer et al., 2011), amorphousamorphous phase separation occurred, prior to recrystallization of the drug. However, the solid dispersions prepared in the study were limited to a 1:1 w/w ratio without considering the effect of the level of drug supersaturation in different polymeric carriers. Moreover, the impact of moisture during accelerated storage stability (45°C, 75%RH) was undoubtedly a dramatic factor contributing to poor physical stability. It is well known that miscibility and/or degree of supersaturation of drug within polymeric carrier may be dramatically reduced due to the ingress of water (21, 22).

4 ACS Paragon Plus Environment

Page 4 of 49

Page 5 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

The purpose of this study was to construct phase diagrams for respective drugpolymer combinations and use these to define drug supersaturation within each polymer material. A direct comparison of the crystallisation inhibition effects of the polymer candidates on a model poorly water-soluble compound (felodipine) was subsequently studied by measuring the bulk crystal growth rate of felodipine within the drug-polymer solid dispersions.

5 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 49

MATERIALS Felodipine (FD), with a purity of 99.9% was a gift from AstraZeneca (Macclesfield, UK). Hydroxypropyl methylcellulose acetate succinate HF (HPMCAS) was a kind gift

provided

by

Shin-Etsu

Chemical

Co.

(brand

name

AQOAT®).

Polyvinylcaprolactam-polyvinyl acetate-polyethylene glycol graft co-polymer (brand name Soluplus®) was a generous gift from BASF Chemical Co. (Ludwigshafen, Germany). Polyvinylpyrrolidone (PVP grade K15) was purchased from SigmaAldrich (Gillingham, England). The true densities of the drug and polymers were measured to be 1.28 g/cm3 for felodipine, 1.28 g/cm3 for HPMCAS, 0.99 g/cm3 for Soluplus and 1.26 g/cm3 for PVPK15 using an AccPyc 1330 helium pycnometer (Micromeritics®, Norcross, USA). METHODS Phase Separation Studies using Raman Spectral Imaging Phase separation of amorphous FD-polymer solid dispersions was studied using Raman Spectral Imaging using a Raman Microscope 300 (Perkin Elmer, UK). Spectral data processing software Grams/AI version 7.02 from Thermo GalacticTM (Waltham, MA) was used in analysis of spectral data. Raman analyses were conducted over a range from 200-3200 cm-1 using an acquisition time of 4 x 2.5 seconds at resolution of 2 cm-1. Felodipine/polymer solid dispersions were prepared as previously described with the exception that an aluminium disc was used as a casting platform due to significant noise from glass. Prepared samples were stored at 0% RH/80°C. In all phase separation studies, Raman spectral imaging was conducted using a x20 objective with laser diameter (D2) equal to 50 µm. Spacing between each respective Raman spectral image (distance between the centre of two neighbouring laser spots) was set at 25 µm. A typical image with scanning points is shown in Figure 1, where the colour map was constructed using a compare/correlation function (PE SpectrumIMAGE©, Version R1.6.4). Coloured Raman images were generated using specific spectral range (1750 ~1450 cm-1 drug peaks dominant) from formulations and by performing a “compare-correlation” against a standard reference (either pure amorphous or crystalline drug). The Compare function is a PE patented algorithm that calculates the similarity between spectra, putting an emphasis on features in the

6 ACS Paragon Plus Environment

Page 7 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

spectrum that relate to the chemical composition of the sample, and ignoring, or reducing the influence of, those features that have other causes. The highest correlation to each respective reference spectrum (amorphous or crystalline felodipine) was assigned to white, whereas lower correlations were assigned to colours from a rainbow-cubic lookup table. Prior to the generation of spectral images (maps), individual spectra were normalised and baseline corrected. Crystal Growth Rate Measurement In order to gain a better understanding of constructed phase diagrams in terms of phase separation and crystal growth within the prepared amorphous solid dispersions, quench-cooled melts of homogenised ball milled samples were used in this study. Mixtures of felodipine and either HPMCAS, Soluplus® or PVP at specified compositions (Table 1) were prepared by ball milling using a Retsch MM200 (Germany). In a typical process, 500 mg of drug and polymer mixture was ball milled at a frequency 30 Hz for at least 10 cycles. Each cycle of milling was 3 minutes, followed by a 3 minute cooling period. Long milling times were utilised to achieve maximum drug-polymer distribution. To examine crystal growth of amorphous solid dispersions using light microscopy, samples were prepared by melting 5-10 mg of ball milled mixture at 150°C (5°C higher than melting point of felodipine) for 3-5 minutes on a glass slip, pre-cleaned with methanol and acetone. The liquid drug was covered with another pre-heated glass slip and held at this temperature for another 2 minutes and then cooled to room temperature by contact with a cold steel block. The top glass slip was later removed to yield a film sample, in the case of surface free samples, whereas for bulk crystal growth, samples had the top glass slip retained during storage stability studies. A precut aluminium ring (50µm thick) was sandwiched between the upper and lower glass slips during preparation to control film thickness. Immediately prepared samples were confirmed to be crystal free by observation in the dark field of a polarised light microscope (PLM, Olympus BX50, Japan). Prepared samples were stored at 0% relative humidity (P2O5) at 80°C, which is 20-30°C higher than the Tg of solid dispersions. During measurement, samples were removed from the controlled temperature/humidity environment and placed onto the hot-stage PLM held at a temperature of 80°C to minimise temperature fluctuation. Samples were placed on the 7 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

stage for a maximum of 10 minutes during visual examination. Bulk crystal growth (ub) was measured with samples sandwiched between two glass slips for the entirety of the storage period (Figure 2). To observe the crystal growth at the free surface of samples, the top glass slip was carefully removed from the pre-formed covered sample to allow surface exposure to normal storage conditions (Figure 2). The growth rate was measured by tracking the edge of felodipine crystal form I. Each reported growth rate was the average of several directional measurements. Thermal Analysis Thermal analysis was used to determine the glass transition of quench-cooled melt samples prepared for crystallisation studies. Experiments were performed on a Perkin Elmer DSC8000 equipped with a refrigerated cooling accessory (Perkin Elmer, UK). Helium, 40 mL/min was used as purge gas. The instrument was calibrated using a heating rate of 200°C/min using high purity indium and zinc to standardise temperature and heat flow signal. The glass transition temperature (Tg) of amorphous solid dispersions were determined by heating the prepared quench-cooled melt sample from -50 to 200 °C at 200 °C/min. The midpoint of glass transition temperature was used to define Tg and values reported are the average of three independently prepared samples. Statistical Analysis The effect of polymeric carrier (Soluplus®/HPMCAS/PVP), polymer Tg and Tg of solid dispersion on the felodipine crystal growth rate from amorphous solid dispersions were statistically analysed using analysis of variance (SigmaPlot, version 11.0). Individual differences in each treatment group were identified using Tukey’s post-hoc test with P Soluplus > HPMCAS-HF. In contrast, when we correlate these parameters to the felodipine crystal growth rates (ucrystal) from different drug-polymer combinations, strong relationships existed in certain correlation factors as shown in Figure 9. The first linear regression obtained in these correlations was the logarithm of crystal growth rate versus drug loading with accuracy of the fit (r2) close to 0.74 (Figure 9a). This strong drug loading effect on the crystal growth of FD from solid dispersion was more evident when we plotted the crystal growth rate versus the supersaturation degree (Figure 9c), wherein a 90% or more drug loading resulted a massive 7-fold crystal growth rate comparing to lower drug loading solid dispersions. This occurred despite the supersaturation degree of felodipine in certain lower drug loading samples being higher than the 90% loaded solid dispersion; for instance the supersaturation degree of 77%FD/HPMCAS-HF is 227+ versus 90%FD/PVPK15 and 92%FD/Soluplus that are 2+ and 38+ respectively (Table 4). This also suggests that, comparing the degree of supersaturation of felodipine in the solid dispersion, the percentage of drug loading (the chance of drug molecule meets another drug molecule) has become a more dominant effect with regard to the crystal growth rate, hence controlling the mechanism of thermodynamic crystallisation for high drug loaded samples. Furthermore, we also plotted the correlation of crystal growth rate versus viscosity of the solid dispersion system using a double-logarithmic transformation. The viscosities of the amorphous solid dispersion were obtained across all felodipine-polymer samples at the same temperature as crystallisation studies were performed (Table 4). Surprisingly, the viscosities measured at 1.2 Hz for these combinations are in the same rank order as drug-polymer miscibility predicted using Flory-Huggins phase diagram. The dynamic viscosity value of FD/PVPK15 is higher than FD/Soluplus and FD/HPMCAS-HF at each comparable drug loading, which indicates a strong miscibility dependent viscosity effect. On the other hand, despite the different 17 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

chemical structures of the three polymers used in this study, a strong viscosity dependent crystal growth rate was obtained in the log-log scale with r2 equal to 0.83 and exponent ξ >1 (ξ≈1.63 from Equation 3, Figure 9b). It is also noted that our study was carried out at temperature 80°C which is much lower than 0.9Tm, where the kinetic control of crystal growth is dominant (38); and also at Temperature exceeding 1.2Tg where the decoupling between diffusion and viscosity may be eliminated (37). It is perhaps more interesting that, if the exponent ξ is close to 1 for single component supercooled liquid as proposed by Ediger et al. (37), then a pure supercooled felodipine should present the exponent ξ ≤ 1 indicating a fragile glass. However, in our system, among different drug loadings and drug polymer combinations, the value of ξ has yielded a value of 1.63 (>1), indicating a less fragile system may be formed through use of an amorphous solid dispersion; hence an increased exponent ξ was obtained. If a blend of amorphous drug in an amorphous polymer would subsequently reduce the liquid fragility of the amorphous drug, the possibility of these two components remaining in one phase would be the key to keeping the system stable at low drug loadings, hence the need to probe phase behaviour using Flory-Huggins theory becomes even more critical. Moreover, through examination of equation 2, which was used to describe the importance of configuration free energy (∆) in crystallisation; the consideration of entropy (∆ ) is also critical for crystallisation of the system. However, equation 2 is based on one single component, wherein only the entropy of melting was used in this relationship. In binary mixtures, the distributions of drug molecules within the amorphous polymeric matrices also increase the entropy term, thus, it should be taken into consideration. The importance of overall entropy of a mixture in terms of physical stability has been demonstrated elsewhere (39, 40). A system with a low level of disorder within the drug or uneven drug distribution within the polymer would lower the overall entropic contribution to the system and weaken the physical stability; hence it is not ideal for the stability of amorphous solid dispersion systems. Thus, a more sophisticated model should be proposed to further understand the phase separation and crystallisation of this amorphous solid dispersion system. Surface Crystal Upward Growth and Surface to Inner Bulk Growth

18 ACS Paragon Plus Environment

Page 18 of 49

Page 19 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Upward crystal growth has been reported on the surface of organic glasses, nifedipine and indomethacin (31, 41). Basically, this type of upward growth mechanism is featured when surface drug molecules move towards the existing crystal and the crystal layer grows with time, which may be significantly reduced by a thin coating on the surface of solid dispersion. In this study, for high drug loaded solid dispersions (90%+ FD loading), nucleation and crystal growth occurred spontaneously on the surface of solid dispersions, extending onto a thin layer across the surface. This thin layer of crystal on the surface acts as bulk crystallisation sites and grows as a function of time. For solid dispersions containing lower drug loadings such as 75% FD/Soluplus®, 80%-60%FD/PVPK15 and 77%, 56% FD/HPMCAS-HF, a different crystallisation mechanism was observed as shown in Figure 10. In PVPK15 and Soluplus®, the crystal was growing upward on the surface of the solid dispersion. It was also evident that the formed crystal was surrounded by a group of small dropletlike materials indicating the potential movement of drug molecules towards the existing crystal on the surface of the solid dispersion. The upward-layer growth of crystals on the surface was attributed to the fact that fast surface transport of drug molecules (velocity of surface molecules) to sustain the size and thickness growth of crystal on the surface (41). In contrast, for FD/HPMCAS-HF solid dispersion, a fast surface crystallisation was initiated at an early stage of the crystallisation, followed by an increase of both surface crystal and inner bulk crystal shown in Figure 10 and 11a. A slower bulk crystal growth rate was observed which is similar to the crystal growth rate obtained in the glass covered solid dispersion samples. Therefore, we may conclude that, although the movement of the drug molecules on the surface is faster than inner bulk in 56%FD/HPMCAS-HF system, bulk crystal growth underneath also occurred during the annealing study. To further understand the differences of crystal growth from amorphous solid dispersions containing different polymeric carriers, we also used a 100% pure polymer film (thickness ≈ 200 µm) to carefully coat the surface of the solid dispersion after initial formation of the crystals on the surface. The further storage of these coated samples at the same conditions suggested that coating the surface decreased crystal growth in all FD/PVPK15 and FD/Soluplus® samples. In contrast, for the FD/HPMCAS-HF sample with surface coating, the crystallisation behaviour has completely changed and an accelerated bulk crystallisation was observed. The number 19 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of bulk crystals observed in the same FD/HPMCAS-HF sample (56% drug loading) was dramatically increased after a simple surface film coating, despite the retardation of the surface crystal growth. Nucleation and Growth or Spinodal Decomposition in ‘Metastable’ Amorphous Solid Dispersions The crystal growth rate of three amorphous solid dispersions (∆G > 0), 35%FD/HPMCAS-HF, 44%FD/Soluplus® and 52%FD/Soluplus®, could not be calculated during the six month study. The photomicrographs of these samples during free surface studies are shown in Figure 12 (a, b and e). For 35% FD/HPMCAS-HF sample, a spinodal decomposition pattern was recorded with amorphous domains as separated and yet continuous, generally regarded as coarsening. These samples remained PLM and Raman crystal free (Figure 6). The size of the domains is believed to increase as function of time, which will eventually lead to crystal growth (20, 42). Whilst for Soluplus®, solid dispersions containing 44% FD and 52% FD, a localised phase separations were observed after 6 months in storage and the samples were remained amorphous (PLM and Raman). Such spinodal decomposition observed during storage has confirmed our phase diagram prediction, that the phase separation mechanism in these selected FD/HPMCAS-HF and FD/Soluplus samples at lower drug loadings would follow a phase separation controlled crystallisation. In contrast, a typical nucleation and growth type of phase separation was observed in both 50% and 60% FD/PVPK15 metastable system during the storage (Figure 12), which was also in agreement with our Raman mapping (Figure 6) and phase diagrams (Figure 3). These PLM images were recorded with the polariser and analyser setting to 45° in order to observe both droplets and crystals. There were also some droplets that coalesced in to bigger particles and small crystals were also observed within these domains (circled red). We believe that the small droplets are the growth front of amorphous to crystalline phase transition. A similar nucleation and growth mechanism has been well documented in polymer-polymer blends due to the fact that the scale of phase separation in polymer-polymer systems is large and the long length of polymer chain and/or long range interaction in polymers will create a nonequilibrium phenomena that can be easily observed with light microscopy (32, 33, 43, 44). Such hypothesis of phase separation dominated crystallisation may be crucial for

20 ACS Paragon Plus Environment

Page 20 of 49

Page 21 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

understanding the stability of these amorphous solids, especially for systems that contain low drug loadings (< 0.5 volume ratio) in which crystal may not be visible within short periods of storage. CONCLUSIONS In order to understand the role of thermodynamic solubility/miscibility of felodipine within polymeric carriers PVPK15, Soluplus® and HPMCAS-HF, phase separation and crystal growth rates of felodipine from the prepared homogenous amorphous solid dispersions were studied. The Flory-Huggins theory based phase diagram was used to guide the selection of storage conditions and drug-polymer compositions for this study. The phase separation and crystallisation were successfully monitored using Raman mapping and polarised light microscopy techniques. Felodipine crystalline form II was occasionally found in the unseeded 56%FD/HPMCAS-HF solid dispersion sample with a distinctive aliphatic bending peak at 1500 cm-1. No correlation was found between the Tg of polymer or Tg of felodipine-polymer solid dispersion and the extent of crystal growth rate. PVPK15 and Soluplus® exhibited significant efficacy in stabilising amorphous felodipine under dry conditions relative to HPMCAS-HF as predicted by F-H phase diagrams. The differences in terms of nucleation and crystal growth for amorphous solid dispersion systems were also discussed such that the systems FD and PVPK15 or Soluplus® appeared to be an upward surface crystal growth mechanism whereas FD/HPMCAS-HF exhibited a surface enhanced and surface to bulk penetration type crystal growth. Due to the high immiscibility of FD/HPMCAS system, the surface to bulk type crystal growth was not retarded by a thin layer of pure polymer coating. In contrast, a significant retardation of crystal growth was observed in FD/PVPK15 and FD/Soluplus® systems with surface coating. The correlations between the characteristics of solid dispersions and crystal growth rate have revealed that, at lower drug loadings, the thermodynamic miscibility and viscosity of the system dictate the speed of phase separation and crystal growth, respectively. The level of solubility/miscibility between drug and polymer appears to be in close agreement with the crystal growth rate results; PVPK15 > Soluplus® > HPMCAS-HF. The surface crystal growth rate of felodipine from the amorphous solid dispersion was also found to be strongly viscosity dependent with exponent (ξ)

21 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

estimated to be 1.36. We conclude that a surface enhanced crystallisation within the amorphous solid dispersion might be highly attributed to the molecular mobility difference between the surface and bulk of the sample, i.e. more viscosity dependent crystallisation. We also note that the inner bulk crystallisation might be attributed to the thermodynamic miscibility due to the fact that, at low drug loadings, phase separation (de-mixing) is the precursor for crystallisation. Funding source The authors would like to acknowledge financial support for this work from the Royal Society in the form of a Royal Society Industry Fellowship awarded to Dr Gavin P Andrews. References 1. Hancock, B. and Zograf, G. Characteristics and significance of the amorphous state in pharmaceutical systems. J. Pharm. Sci. 1997, 86, 1-12. 2. Hancock, B. and Parks, M. What is the true solubility advantage for amorphous pharmaceuticals? Pharm. Res. 2000, 17, 397-404. 3. Bikiaris, D.N. Solid dispersions, Part I: recent evolutions and future opportunities in manufacturing methods for dissolution rate enhancement of poorly water-soluble drugs. Expert Opin. Drug Deliv. 2011, 8, 1501-1519. 4. Bikiaris, D.N. Solid dispersions, Part II: new strategies in manufacturing methods for dissolution rate enhancement of poorly water-soluble drugs. Expert Opin. Drug Deliv. 2011, 8, 1663-1680. 5. Yoshioka, M.; Hancock, B.; Zografi, G. Crystallization of Indomethacin from the Amorphous State Below and Above its Glass-Transition Temperature. J. Pharm. Sci. 1994, 83, 1700-1705. 6. Hikima, T.; Hanaya, M.; Oguni, M. Microscopic observation of a peculiar crystallization in the glass transition region and beta-process as potentially controlling the growth rate in triphenylethylene. J. Mol. Struct. 1999, 479, 245-250. 7. Hikima, T.; Adachi, Y.; Hanaya, M.; Oguni, M. Determination of Potentially Homogeneous-Nucleation-Based Crystallization in O-Terphernyl and an Interpretation of the Nucleation-Enhancement Mechanism. Phys. Rev. B 1995, 52, 3900-3908. 8. Alonzo, D.E.; Zhang, G.G.Z.; Zhou, D.; Gao, Y.; Taylor, L.S. Understanding the Behavior of Amorphous Pharmaceutical Systems during Dissolution. Pharm. Res. 2010, 27, 608-618. 22 ACS Paragon Plus Environment

Page 22 of 49

Page 23 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

9. Bikiaris, D.N. Solid dispersions, Part I: recent evolutions and future opportunities in manufacturing methods for dissolution rate enhancement of poorly water-soluble drugs. Expert opinion on drug delivery 2011, 8, 1501-1519. 10. Tian, Y.; Booth, J.; Meehan, E.; Jones, D.S.; Li, S.; Andrews, G.P. Construction of Drug-Polymer Thermodynamic Phase Diagrams Using Flory-Huggins Interaction Theory: Identifying the Relevance of Temperature and Drug Weight Fraction to Phase Separation within Solid Dispersions. Mol. Pharm. 2013, 10, 236-248. 11. Pajula, K.; Taskinen, M.; Lehto, V.; Ketolainen, J.; Korhonen, O. Predicting the Formation and Stability of Amorphous Small Molecule Binary Mixtures from Computationally Determined Flory-Huggins Interaction Parameter and Phase Diagram. Mol. Pharm. 2010, 7, 795-804. 12. Khougaz, K. and Clas, S. Crystallization inhibition in solid dispersions of MK0591 and poly(vinylpyrrolidone) polymers. J. Pharm. Sci. 2000, 89, 1325-1334. 13. Weuts, I.; Kempen, D.; Six, K.; Peeters, J.; Verreck, G.; Brewster, M.; Van den Mooter, G. Evaluation of different calorimetric methods to determine the glass transition temperature and molecular mobility below T-g for amorphous drugs. Int. J. Pharm. 2003, 259, 17-25. 14. Weuts, I.; Kempen, D.; Decorte, A.; Verreck, G.; Peeters, J.; Brewster, M.; Van den Mooter, G. Physical stability of the amorphous state of loperamide and two fragment molecules in solid dispersions with the polymers PVP-K30 and PVP-VA64. Eur. J. Pharm. Sci. 2005, 25, 313-320. 15. Bhugra, C.; Shmeis, R.; Krill, S.L.; Pikal, M.J. Predictions of onset of crystallization from experimental relaxation times I-correlation of molecular mobility from temperatures above the glass transition to temperatures below the glass transition. Pharm. Res. 2006, 23, 2277-2290. 16. Caron, V.; Bhugra, C.; Pikal, M.J. Prediction of Onset of Crystallization in Amorphous Pharmaceutical Systems: Phenobarbital, Nifedipine/PVP, and Phenobarbital/PVP. J. Pharm. Sci. 2010, 99, 3887-3900. 17. Graeser, K.A.; Patterson, J.E.; Zeitler, J.A.; Gordon, K.C.; Rades, T. Correlating thermodynamic and kinetic parameters with amorphous stability. Eur. J. Pharm. Sci. 2009, 37, 492-498. 18. Zhou, D.; Law, D.; Schmitt, E. Thermodynamics, Molecular Mobility and Crystallization Kinetics of Amorphous Griseofulvin. Molecular pharmaceutics 2008, 5, 927-936. 19. Kestur, U.S. and Taylor, L.S. Role of polymer chemistry in influencing crystal growth rates from amorphous felodipine. Crystengcomm 2010, 12, 2390-2397. 20. Lauer, M.E.; Grassmann, O.; Siam, M.; Tardio, J.; Jacob, L.; Page, S.; Kindt, J.H.; Engel, A.; Alsenz, J. Atomic Force Microscopy-Based Screening of Drug-Excipient Miscibility and Stability of Solid Dispersions. Pharm. Res. 2011, 28, 572-584. 23 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

21. Vasanthavada, M.; Tong, W.; Joshi, Y.; Kislalioglu, M. Phase Behavior of amorphous molecular dispersions I: Determination of the degree and mechanism of solid solubility. Pharm. Res. 2004, 21, 1598-1606. 22. Vasanthavada, M.; Tong, W.; Joshi, Y.; Kislalioglu, M. Phase behavior of amorphous molecular dispersions - II: Role of hydrogen bonding in solid solubility and phase separation kinetics. Pharm. Res. 2005, 22, 440-448. 23. Tian, Y.; Caron, V.; Jones, D.S.; Healy, A.; Andrews, G.P. Using Flory-Huggins phase diagrams as a pre-formulation tool for the production of amorphous solid dispersions: a comparison between hot-melt extrusion and spray drying. J. Pharm. Pharmacol. 2013, n/a-n/a. 24. Abu Diak, O.; Jones, D.; Andrews, G. Understanding the performance of meltextruded poly(ethylene oxide)-bicalutamide solid dispersions: Characterisation of microstructural properties using thermal, spectroscopic and drug release methods. J. Pharm. Sci. 2011, 101, 200-213. 25. Karavas, E.; Georgarakis, M.; Docoslis, A.; Bikiaris, D. Combining SEM, TEM, and micro-Raman techniques to differentiate between the amorphous molecular level dispersions and nanodispersions of a poorly water-soluble drug within a polymer matrix. Int. J. Pharm. 2007, 340, 76-83. 26. Shah, B.; Kakumanu, V.K.; Bansal, A.K. Analytical techniques for quantification of amorphous/crystalline phases in pharmaceutical solids. J. Pharm. Sci. 2006, 95, 1641-1665. 27. Rollinger, J. and Burger, A. Polymorphism of racemic felodipine and the unusual series of solid solutions in the binary system of its enantiomers. J. Pharm. Sci. 2001, 90, 949-959. 28. Rumondor, A.C.F.; Jackson, M.J.; Taylor, L.S. Effects of Moisture on the Growth Rate of Felodipine Crystals in the Presence and Absence of Polymers. Cryst. Growth Des. 2010, 10, 747-753. 29. Donnelly, C. Tian, Y. Potter, C. Jones, D.S. Andrews, G.A. Probing the effects of experimental conditions on the character of drug-polymer phase diagram constructed using Flory-Huggins theory. Pharmaceutical Research 2014, 30. Wu, T.; Sun, Y.; Li, N.; de Villiers, M.M.; Yu, L. Inhibiting surface crystallization of amorphous indomethacin by nanocoating. Langmuir 2007, 23, 5148-5153. 31. Cai, T.; Zhu, L.; Yu, L. Crystallization of Organic Glasses: Effects of Polymer Additives on Bulk and Surface Crystal Growth in Amorphous Nifedipine. Pharm. Res. 2011, 28, 2458-2466. 32. Tanaka, H. and Nishi, T. New Types of Phase-Separation Behavior during the Crystallization Process in Polymer Blends with Phase-Diagram. Phys. Rev. Lett. 1985, 55, 1102-1105.

24 ACS Paragon Plus Environment

Page 24 of 49

Page 25 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

33. Tanaka, H. and Nishi, T. Local Phase-Separation at the Growth Front of a Polymer Spherulite during Crystallization and Nonlinear Spherulitic Growth in a Polymer Mixture with a Phase-Diagram. Physical Review a 1989, 39, 783-794. 34. Bhugra, C. and Pikal, M. Role of thermodynamic, molecular, and kinetic factors in crystallization from the amorphous state. J. Pharm. Sci. 2008, 97, 1329-1349. 35. Turnbull, D. and Cohen, M. Crystallization kinetics and glass formation, In Modern Aspects of the Vitreous State, MacKenzie, S., Ed.; Butterworths: London, 1960; 36. Kestur, U.S.; Van Eerdenbrugh, B.; Taylor, L.S. Influence of polymer chemistry on crystal growth inhibition of two chemically diverse organic molecules. Crystengcomm 2011, 13, 6712-6718. 37. Ediger, M.D.; Harrowell, P.; Yu, L. Crystal growth kinetics exhibit a fragilitydependent decoupling from viscosity. J. Chem. Phys. 2008, 128, 034709. 38. Sun, Y.; Xi, H.; Ediger, M.D.; Yu, L. Diffusionless crystal growth from glass has precursor in equilibrium liquid. J Phys Chem B 2008, 112, 661-664. 39. Qian, F.; Huang, J.; Zhu, Q.; Haddadin, R.; Gawel, J.; Garmise, R.; Hussain, M. Is a distinctive single Tg a reliable indicator for the homogeneity of amorphous solid dispersion? Int. J. Pharm. 2010, 395, 232-235. 40. Qian, F.; Huang, J.; Hussain, M. Drug-Polymer Solubility and Miscibility: Stability Consideration and Practical Challenges in Amorphous Solid Dispersion Development. J. Pharm. Sci. 2010, 99, 2941-2947. 41. Sun, Y.; Zhu, L.; Kearns, K.L.; Ediger, M.D.; Yu, L. Glasses crystallize rapidly at free surfaces by growing crystals upward. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 5990-5995. 42. Zhang, R.F. and Veprek, S. Metastable phases and spinodal decomposition in Ti(1-x)Al(x)N system studied by ab initio and thermodynamic modeling, a comparison with the TiN-Si(3)N(4) system. Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process. 2007, 448, 111-119. 43. Cheng, S. and Keller, A. The role of metastable states in polymer phase transitions: Concepts, principles, and experimental observations. Annual review of materials science 1998, 28, 533-562. 44. Wang, S.; Wu, C.; Ren, M.; Van Horn, R.M.; Graham, M.J.; Han, C.C.; Chen, E.; Cheng, S.Z.D. Liquid-liquid phase separation in a polyethylene blend monitored by crystallization kinetics and crystal-decorated phase morphologies. Polymer 2009, 50, 1025-1033.

25 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 49

Table 1. Volume fraction and weight fraction of examined points for FD-HPMCAS, FD-Soluplus® and FD-PVPK15 systems. FD-HPMCAS

Drug Fractions examined at of 80°C

0.90 0.77 0.56 0.35

Drug fraction (w/w) 0.90 0.77 0.56 0.35

-

-

Drug fraction (φ)

FD-Soluplus®

0.90 0.71 0.48 0.38

Drug fraction (w/w) 0.92 0.75 0.52 0.44

-

-

Drug fraction (φ)

26 ACS Paragon Plus Environment

FD-PVPK15 Drug fraction (φ) 0.89 0.79 0.69 0.59 0.49

Drug fraction (w/w) 0.90 0.80 0.70 0.60 0.50

Page 27 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Table 2. Mid-point of the glass transition temperature of FD-polymer solid dispersions measured at scanning speed of 200 °C/min, result shown is the average of three independent samples ± standard deviation FD-PVP

FD-HPMCAS

FD-Soluplus®

w/w of drug

Tg /°C

GT prediction

w/w of drug

Tg /°C

GT prediction

w/w of drug

Tg /°C

GT prediction

0.9

53.9±1.3

56.2

0.9

52.9±0.1

53.2

0.92

50.6±0.4

54.9

0.8

58.2±1.1

62.4

0.77

54.5±0.1

64.3

0.75

51.6±0.5

62.8

0.7

64.0±0.5

68.8

0.56

57.8±0.5

78.4

0.52

55.8±0.2

70.3

0.6

68.7±0.6

75.4

0.35

68.5±0.2

93.7

0.44

61.2±0.1

72.3

0.5

74.1±1.0

82.3

-

-

-

-

-

-

27 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 49

Table 3. The crystallisation onset time of unseeded amorphous solid dispersions

measured using polarised light and Raman microscopy at different temperatures. Results are the average value ± standard deviation of three independent samples. Sample at 80°C 100% aFD 90%FD/ HPMCAS 92%FD/ Soluplus 90%FD/ PVPK15 77%FD/ HPMCAS 75%FD/ Soluplus 80%FD/ PVPK15

Crystallisation onset time (hours) free ≤ 24±3 surface free ≤ 36±3 surface covered ≤ 36±1 free ≤ 36±3 surface covered ≤ 36±2 free ≤ 72±7 surface covered ≤ 264±22 free ≤ 120±5 surface covered ≤ 464±20 free ≤ 924±4 surface covered ≤ 1108±10 free ≤ 336±9 surface covered ≤ 1678±19

Form

Sample at 80°C

Crystallisation onset time (hours)

Form

I I I I I I I I I I I I I

56%FD/ HPMCAS 52%FD/ Soluplus 70%FD/ PVPK15 35%FD/ HPMCAS 44%FD/ Soluplus 60%FD/ PVPK15

free surface covered free surface covered free surface covered free surface covered free surface covered Free surface covered

≤ 528±15

I / II

≤ 528±16

I

≤ 696±32

I

> 4400

*

≤ 384±12

I

≤ 1424±26

I

> 4400

*

> 4400

*

≤ 3000±0

I

> 4400

*

≤ 4200±200 > 4440

I *

* indicates the crystal form was only detected by PLM but was not detectable using HyperDSC™ and Raman spectroscopy due small quantity and particle size present.

28 ACS Paragon Plus Environment

Page 29 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Table 4. Parameters characterising the free surface crystal growth rate (u) of felodipine within solid dispersions with polymers PVPK15, HPMCAS-HF and Soluplus® at 80°C; Viscosity of PVP and HPMCAS could not be measured at 80°C since polymers were in glassy state. Drug Samples

Weight

∆Gmix/RT

C/Cs

Fraction

η'

ufree

(Pa•s)

(µm/s)

FD+Soluplus

0.44

0.04

18.49

3.23E+06

2.78E-07

FD+Soluplus

0.52

0.08

21.86

3.49E+05

1.11E-05

FD+Soluplus

0.75

0.14

31.53

1.54E+05

1.21E-04

FD+Soluplus

0.92

0.08

38.67

1.18E+04

1.21E-02

FD+HPMCAS

0.35

0.40

103.34

2.01E+07

1.39E-07

FD+HPMCAS

0.56

0.51

165.35

1.08E+05

4.06E-05

FD+HPMCAS

0.77

0.40

227.35

6.19E+04

8.59E-04

FD+HPMCAS

0.9

0.20

265.74

1.78E+04

1.27E-02

FD+PVPK15

0.8

-0.19

1.76

8.11E+04

2.92E-06

FD+PVPK15

0.6

-0.32

1.31

2.31E+06

2.50E-06

FD+PVPK15

0.7

-0.26

1.55

4.90E+05

7.72E-07

FD+PVPK15

0.9

-0.10

2.00

2.56E+04

9.71E-03

Felodipine

1

--

--

3.47E+03

2.43E-02

29 ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Schematic of Raman mapping experiments demonstrating spot and space 361x214mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 30 of 49

Page 31 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 2. Schematic of crystal growth study on amorphous felodipine/polymer solid dispersions using Raman and/or Polarised Light Microscope at T = 80°C: (a) a covered surface crystal growth; (b) free surface crystal growth 281x211mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3 (a) Calculated free energy of mixing (∆Gmix) of FD/polymer system at a temperature of 80°C, the examined points are marked as (O). 282x190mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 32 of 49

Page 33 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 3b Examined points in the phase diagram of FD-PVPK15. 282x200mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3c. Examined points in the phase diagram of individual systems of FD-HPMCAS 282x183mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 34 of 49

Page 35 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 3d Examined points in the phase diagram of individual systems of FD-Soluplus 282x195mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. Raman spectra of crystalline, amorphous felodipine and high drug loaded (90% w/w) polymer solid dispersions. 190x90mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 36 of 49

Page 37 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 5. Raman and photomicrograph of crystal felodipine form I & II grown on the same surface of 56%FD/HPMCAS solid dispersion after annealed at 80 °C, 0% RH, size indicator is 50 µm. The stripes on the background are from the surface of aluminium disc 256x211mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6 Examples of Raman chemical maps of highest and lowest drug loadings of felodipine polymer solid dispersions annealed at 80 °C, 0% RH at various time points 190x300mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 38 of 49

Page 39 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 7. Variation in crystal growth of felodipine within 90%FD/Soluplus® free surface and covered surface solid dispersions. Scale bars represent 200 µm 264x211mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8a. Size of crystal as function of time at 80°C, 0% RH for samples (a) 100% aFD, 90%FD/PVPK15, 90%FD/HPMCAS-HF and 92%FD/Soluplus®. Data are presented as the average of at least three replicates ± standard deviation. 282x191mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 40 of 49

Page 41 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 8b. Size of crystal as function of time at 80°C, 0% RH for samples (b) 80%FD/PVPK15, 77%FD/HPMCAS-HF, 75%FD/Soluplus®, 70%FD/PVPK15 and 56%FD/HPMCAS–HF. Data are presented as the average of at least three replicates ± standard deviation. 278x211mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 9a. Correlation plots of felodipine crystal growth rate from solid dispersions versus drug loading 282x188mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 42 of 49

Page 43 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 9b. Correlation plots of felodipine crystal growth rate from solid dispersions versus viscosity 282x197mm (72 x 72 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 9c. Correlation plots of felodipine crystal growth rate from solid dispersions versus supersaturation ratio 138x79mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 44 of 49

Page 45 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 10. Photomicrographs of crystallised solid dispersion samples (a) 75%FD/Soluplus®, lense was focused on the surface of the film (b) 75%FD/Soluplus®, lense was focused on the surface of the crystal (c) 56% FD/HPMCAS-HF, lense was focused on the surface of the crystal (d) 56%FD/HPMCAS-HF, lense was focused on the bulk crystal underneath the surface crystal. Scale bars represent 200 µm. 238x190mm (150 x 150 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 11a. Photomicrographs of crystallisation within 56%FD/HPMCAS-HF amorphous solid dispersion free surface crystallisation. Scale bars represent 50 µm. 940x753mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 46 of 49

Page 47 of 49

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Figure 11b. Photomicrographs of crystallisation within 56%FD/HPMCAS-HF amorphous solid dispersion surface coated crystallisation; the coating was applied after the initiation of surface crystal growth. Scale bars represent 50 µm. 450x357mm (150 x 150 DPI)

ACS Paragon Plus Environment

Molecular Pharmaceutics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

534x438mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 48 of 49