(744a) Linear Models for Prediction of Ibuprofen Crystal Morphology Based on Hydrogen Bonding Propensities

Solvents have a significant impact on the final crystal form of organic solids during solution crystallization. The use of polarity scales such as Hildebrand solubility parameter and dielectric constant for solvent selection often proves too generalized and do not provide enough insights into the solvent-solute intermolecular interactions directly affecting crystal growth and morphology. This paper addresses the challenging task of selecting an appropriate single component solvent property index that most accurately and sufficiently characterizes crystal morphology. Cooling crystallization experiments were carried out in a wide range of solvents using ibuprofen as a model pharmaceutical compound. Subsequently optical microscope images were used for quantitative characterization of morphology. Finally, linear models that correlate ibuprofen crystal morphology with pure solvent properties were developed. Our results show that, in general, there is a negative linear correlation between crystal aspect ratio (morphology) and a given solvent index. Some correlations revealed significant deviations which can be explained. Where appropriate, infrared spectroscopic measurements were used to analyze the deviations. The ?acceptance number? has been identified as an index that significantly captures the ibuprofen-solvent hydrogen bonding intermolecular interactions. Predictions, using model based on acceptance number, were found to compare very well with experimentally determined aspect ratio data from the open literature. Finally, based on insights gained from this work, a flowchart to serve as a useful solvent selection guideline for crystallization of ibuprofen is proposed.