(205c) Brownian Dynamics Simulations Predict the Nucleation Probability of Polymorphs during the Process of Crystallization
AIChE Annual Meeting
Monday, November 8, 2021 - 4:15pm to 4:36pm
Crystallization of organic molecules is a highly stochastic process involving a large number of molecules spanning long timescales. Few degrees of freedom arising due to the molecular structure can result in multiple possible metastable crystal structures, leading to polymorphism. Such metastable structures prevent the ability to predict and control the outcome of crystallization effectively. To connect the events that occur during the molecular motion to the outcome of crystallization, the calculation of the probability of events leading to desired polymorphs is necessary. In previous studies, we have identified the rapid solvent exchange in the solvation shell and configuration of the solvent molecules in the solvation shell as the critical aspects of crystallization that lead to the desired polymorph. To further obtain the probability of events leading to the desired polymorph, we use the Brownian Dynamics simulation approach. Here, we coarse-grain the molecular structure to reduce the number of degrees of freedom, and the effect of solvent interaction is captured by the noise term in the Langevin equation. This approach allows simulating the interaction of a large number of solute molecules and spans a longer timescale. The configuration of solute molecules at each time is used to predict the probability of events leading to a particular polymorph. The approach stated above is applied to the Glutamic acid molecule to effectively predict the polymorphic form obtained at the end of the crystallization process. Furthermore, results suggest that an increase in the local density due to high supersaturation allows the formation of metastable polymorph at a greater probability, and tuning the solvent environment to restrict such events can yield significant control on the process of crystallization.