(219d) Computer-Aided Solvent Selection to Control Nucleation and Resulting Product Quality Attributes in API Crystallization

As development timelines for new small molecule therapies become increasingly compressed, attention has turned to the creation of new tools and methodologies to streamline the laboratory process development effort. Computational approaches for the prediction of solubility are now widely used in the design of industrial crystallization processes to aid solvent selection and reduce development timeframes and material needs.¹ However, despite the significant influence of solvents on the nucleation and growth kinetics and thus particle size outcomes, there have been limited examples of computational methods being used to guide solvent selection with consideration of the particle size goals of the process.

A relationship between nucleation difficulty and solvent-solute interaction strength has been predicted computationally and demonstrated experimentally for a range of different organic molecules, strongly suggesting that the desolvation of a solute molecule is one of the key elements governing crystal nucleation kinetics.² In contrast to current computational approaches which require both specialized knowledge and supercomputing capabilities, we demonstrate that activity coefficients (which can be easily computed with COSMO-RS software) can be used to estimate the strength of the solvent-solute interaction in solution. Those activity coefficients can serve as in silico descriptors, enabling a priori prediction of the relative nucleation ease/difficulty of a solute crystallized out of a set of different solvents and the preparation of a ranking list to guide solvent selection.

In this contribution, we demonstrate how this original workflow can be practically applied as part of a high-throughput in silico solvent screening to guide the selection of a preferred solvent to facilitate ease of nucleation out of a broad list of solvent possibilities. The insights from this approach can allow crystallization practitioners to assess how different solvents may alter the balance of nucleation and crystal growth occurring in the process to support decision making on what solvents best enable the design goals of the crystallization (e.g., particle size, uniformity, and size distribution).

References

[1] Lovette, M.A.; Albrecht, J.; Ananthula, R.S.; Ricci, F.; Sangodkar, R.; Shah, M.S.; Tomasi, S. Cryst. Growth Des. 2022, 22, 5239–5263.

[2] Khamar, D.; Zeglinski, J.; Mealey, D.; Rasmuson, Å.C. J. Am. Chem. Soc. 2014, 136, 11664–11673.

Acknowledgement

This research was partially supported by the European Union Horizon 2020 MSCA RISE programme, "ORBIS" grant agreement No. 778051