The choice of solvent has a significant impact on crystal growth habit. It is therefore one of the most important design decisions in crystallization and its effect is one of the most desirable for a crystal growth model to predict.
Our group has developed a mechanistic crystal growth model, accounting for both spiral and two-dimensional nucleation & growth mechanisms [1,2]. Critical quantities in the model are kink and edge energies on the crystal surface, which affect the predicted relative face growth rates under these layered mechanisms. These energies are modified by solvation of the crystal surface, which must be effectively accounted for. However, the determination of this solvent-induced energetic modification must not compromise the ability to obtain fast growth shape predictions, in order for our model to remain a practical engineering tool.
Interfacial energy models based on the bulk interface approximation are suitable and their application to our mechanistic model is the focus of this work. The energy of an interface is expressed through a penalty term of each phase’s cohesive energy and an adhesive reward. For the work of adhesion, the van Oss, Chaudhury & Good (vOCG) model  provides a functional form that appropriately separates interactions according to how they match across an interface. In applying this to our model, solid-state crystal interactions are determined using a force field and solvent interactions are determined using solubility parameters and appropriate correlations to interfacial energies. These cohesive energies can be used to determine the required parameters for the vOCG model and although there is some uncertainty in the required partitioning, shape predictions are often not sensitive to this. Heuristic rules are developed for the correct application of this interfacial energy model to different types of systems.
To test the success of this within our mechanistic growth model, predictions for vapor and solution grown crystal shapes of various systems have been compared to experimental results, which isolates the effect of the solvent and has indicated that this approach is successful.
- Snyder, R. C. and Doherty, M. F. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Science 465(2104), 1145–1171 (2009).
- Lovette, M. A. and Doherty, M. F. Crystal Growth & Design 12(2), 656–669 (2012).
- Van Oss, C., Chaudhury, M., and Good, R. Advances in colloid and interface science 28, 35–64 (1987).
See more of this Group/Topical: Separations Division