(702d) Growing Macroscopic Hydrates Using a Bioinspired Approach and Investigating Dehydration Induced Polymorphism

Growing of macroscopic single crystals of organic materials is not necessarily a straight forward procedure. Carbamazepine dihydrate is one of these cases. Traditional approaches of crystallization of this, particular, compound using mixtures of alcohols with water lead to crystals with high aspect ratios (needle shaped or high aspect ratio plates)¹. These crystals are not suitable for wettability or XPS measurements on individual facets². In this work, a bioinspired crystallization method is employed for the growth of macroscopic carbamazepine dihydrate crystals.

In a work on the interactions of surfactants with carbamazepine in aqueous solutions, investigators have reported the formation of prismatic carbamazepine dihydrate crystals in the micron scale³. From the surfactants employed, it seems that the carbonyl groups of sodium taurocholate influence the crystal habit. Unfortunately, sodium taurocholate proved not to be adequate for the growth of crystals of larger sizes, that can be used as seeds for the growth of macroscopic crystals. Also, other studies suggest that the growth of large protein crystals is favored by biomimetic solutions, imitating the design of the digestive vacuole of malarian parasites such as Plasmodium falciparum, comprising of an aqueous and an organic phase^4,5.

A two phase system with cyclohexanone in the organic phase and water in the other phase was developed. An adequate amount of carbamazepine was added in the solution under stirring and the mixture was heated using a water bath. When all the material was dissolved, stirring stopped and a temperature gradient of 1^oC*d^-1 was applied. After a few days, depending on the amount of carbamazepine added, crystals started to appear inside precursor droplets formed on the interface between the two phases. When the temperature of the water bath reached 4^oC the crystals were harvested. They were quite large with the (0 2 0) facet to be the dominant one, having sizes up to 2x1 cm. The harvested crystals were examined using XRD and found to be carbamazepine dihydrate crystals. Similar experiments were performed with heavy alcohols and butanone, but the crystals obtained were not that promising, even though they were in all cases dihydrate crystals, in terms of crystal habit. The shift from the needle like shape to a more prismatic one can be thus attributed to the carbonyl chemistry.

Molecular dynamics simulations suggest that in such systems droplets of the aqueous phase are formed in the organic phase⁶. This droplets were tracked using Dynamic Light Scattering and found to be about 500 nm in diameter. Nonetheless, these droplets do not seem to act as nucleation sites. Only the main water cyclohexanone interface operates as a nucleation site. This preferential nucleation can be explained on the ground of heterogeneous nucleation. The shape factor for nucleation on an interface should result to a much higher decrease on the Gibbs free energy of nucleation rather than the shape factor for nucleation on a droplet^7,8. Furthermore, molecular dynamics simulations propose the formation of pockets of enhanced water concentration on the interface. These pockets is speculated to promote even further nucleation.

Drying experiments performed on these macroscopic crystals. From SEM images obtained, it was shown that dehydration is driven by well-ordered cracks formed on the surface of the dominant facet. Two types of cracks appear on the (0 2 0) facet: 1) macroscopic cracks vertical to the (0 2 0) facet with crack spacing a few microns big and 2) cracks with smaller spacing inclined, both clockwise and counterclockwise, with respect to the primary cracks of type 1. These two types of cracks correspond to the following main cleavage planes: (h 0 0), (-1 ±1 1) and (0 ±1 1). Furthermore, macroscopic crystals were sliced with razor blade and the examination of their core revealed an additional cleavage plane, (0 k 0). These findings are important, as previous studies reported only the (h 0 0) cleavage plane. It should, also, be mentioned that the crack appearance is independent of the application of vacuum upon drying.

The anhydrous form nucleates, during dehydration, on the surface of the dihydrate crystals in the form of small islands. The dehydration mechanism for macroscopic crystals results solely to the triclinic polymorph of anhydrous carbamazepine. The triclinic polymorph was depicted both with XRD and SEM. The SEM images are quite interesting showing the needle shaped triclinic crystals growing vertically on the surface of the dihydrate. Comparing the results of the dehydration induced polymorphism for macroscopic dihydrate crystals and micron-sized crystals it can be deduced that the crystal size and habit influence the final polymorph. It seems that for cases where the need for heat dissipation is greater (i.e. in macroscopic crystals), the triclinic polymorph nucleating during drying is higher. This is because the needle-shaped triclinic crystals have a large surface area to volume ratio providing better heat dissipation. This concomitant nucleation scheme has been reported computationally via Monte Carlo simulations^9,10. However, the suggested drying mechanism can be better understood, mechanistically, on the same ground Machon described phase transformations on nanoparticles using the one dimensional Ginzburg-Landau equation^{11, 12}.

This study expands the current knowledge on nucleation of organic materials using bioinspired approaches. It is quite important to notice the nucleation of dihydrate and the absence of any other polymorphs. This shows that for such systems, the chemical potential associated with the aqueous phase and with the small amount of water dissolved in cyclohexanone drive the nucleation. It is, also, of particular importance that the drying mechanism is revealed; suggesting that the heat dissipation mechanism drives the dehydration induced polymorphism. The image analysis revealed also the nucleation mechanism of the anhydrous polymorphs resembling to the anisotropic nucleation model proposed in the past^13,14.

Disclosure:

Imperial College London and AbbVie jointly participated in study design, research, data collection, analysis and interpretation of data, writing, reviewing, and approving the publication. Eftychios Hadjittofis is a graduate student at Imperial College London; Jerry Y. Y. Heng is a professor at Imperial College London. They all have no additional conflicts of interest to report. Geoff G. Z. Zhang is an employee of AbbVie and may own AbbVie stock.

References:

1. Khoo JY, Shah UV, Schaepertoens M, Williams DR, Heng JYY. Process-induced phase transformation of carbamazepine dihydrate to its polymorphic anhydrates. Powder Technol. 2013;236:114-121.

2. Ho R, Hinder SJ, Watts JF, Dilworth SE, Williams DR, Heng JYY. Determination of surface heterogeneity of d-mannitol by sessile drop contact angle and finite concentration inverse gas chromatography. Int J Pharm. 2010;387(1â??2):79-86.

3. Rodriguez-Hornedo N, Murphy D. Surfactant-facilitated crystallization of dihydrate carbamazepine during dissolution of anhydrous polymorph. Journal of Pharmaceutical Sciences. 2004;93(2):449-460.

4. Olafson KN, Rimer JD, Vekilov PG. Growth of large hematin crystals in biomimetic solutions. Crystal Growth and Design. 2014;14(5):2123-2127.

5. Vekilov PG, Rimer JD, Olafson KN, Ketchum MA. Lipid or aqueous medium for hematin crystallization? Crystal Engineering Communications. 2015;17:7790-7780.

6. Wick CD, Chang TM. Computational observation of pockets of enhanced water concentration at the 1â??Octanol/water interface. The Journal of Physical Chemistry B. 2014;118(7785-7791).

7. Fletcher NH. Size effect in heterogeneous nucleation. Journal of Chemical Physics. 1958;29:572-576.

8. Johnson WC, White CL, Marth PE, et al. Influence of crystallography on aspects of solid-solid nucleation theory. Metallurgical transactions A. 1975;6(4):911-919.

9. Sanders DP, Larralde H, Leyraz F. Competitive nucleation and the ostwald rule in a generalized potts model with multiple metastable phases. Physical Review B. 2007;75:132101.

10. Hedges LO, Whitelam S. Limit of validity of ostwald's rule of stages in a statistical mechanical model of crystallization. Journal of Chemical Physics. 2011;135:164902.

11. Machon D, Piot L, Hapiuk D, et al. Thermodynamics of nanoparticles: Experimental protocol based on a comprehensive ginzburg-landau interpretation. Nano Letters. 2014;14(1):269-276.

12. Machon D, Melinion P. Size-dependent pressure-induced amorphization: A thermodynamic panorama. Physical Chemistry Chemical Physics. 2014;17(2):903

13. Bianchet MA, Rigotti G, Blesa MA. Surface nucleation and anisotropic growth model for solid state reactions. Solid State Ionics. 1990;42(1):21-27.

14. Bianchet MA, Rigotti G, Blesa MA. Surface nucleation and anisotropic growth model for the dehydration of 11-aminoundecanoic acid dihydrate. Solid State Ionics. 1995;78(1â??2):49-53.