Solubility Model to Guide Solvent Selection in Synthetic Process Development

Solvent selection to obtain adequate yield and impurity rejection is a critical activity in the development of crystallization processes for synthetic chemicals. This activity requires the determination of solubilities for a solute (moreover, a polymorph of a solute) in a variety of solvents and solvent combinations. A model was developed to facilitate the exploration solubility for single solvent systems. This model is based on the dynamics of incorporation events at discrete interfaces that are most applicable at temperatures far from the melt. Parameters within the model were trained, hyper-parameters tuned, and model performance cross-validated using a set of ∼500 solubility measurements across a variety of solutes and solvents. Cross-validation predictions appear unbiased and accurate, with errors approximated by a log10-normal distribution described by a mean and standard deviation of 0.011 and 0.197, respectively.