In a solution crystallization, final crystal shape is usually a function of process parameters, such as concentration, temperature, and supersaturation, and the type of solvent used. Interestingly, the effect of solvent on particle morphology can be either strong or it can be small or negligible, and it appears to depend both on the process solvent and the structural features of the active pharmaceutical ingredient (API) molecule and its crystal form.
In this contribution, by combining different computational approaches, including morphology prediction, attachment energy analysis, and the thermodynamic descriptors of solvent-crystal facet interaction strength1,2 we present a computational workflow that allows users to: (1) assess how sensitive an API is to its morphology modulation by solvent environment, and (2) screen a range of different solvents in silico to produce a ranking of the solvents, from those that have the strongest influence on changing the morphology (e.g., aspect ratio) down to the solvents which are predicted to have a minimal or zero impact on morphology change during crystallization. The computational predictions are compared with the outcomes of relevant crystallization experiments to assess the applicability of the modelling approach for guiding crystallization development.
The proposed in silico screening tool may be used to support the selection of solvents to meet specific morphology requirements, e.g., to avoid the crystallization of undesired needle- or rod-like particles.
References
[1] Soper, E. M.; Penchev, R. Y.; Todd. S. M.; Eckert, F.; Meunier, M. Quantifying the effect of solvent on the morphology of organic crystals using a statistical thermodynamics approach. J. Cryst. Growth 2022, 591, 126712.
[2] Zeglinski, J.; Jakubowska, E.; Koczorowski, T.; Ukrainczyk, M.; Roche, B.; Morris, G.; Wood, B.; Glennon, B. Towards predicting nucleation difficulty of organic compounds crystallized from different solvents. Cryst. Growth Des 2025 – submitted.