(296b) Computer Simulation of Nucleation in Small Organic Crystals: Homogeneous Nucleation in Benzene | AIChE

(296b) Computer Simulation of Nucleation in Small Organic Crystals: Homogeneous Nucleation in Benzene

Authors 

Shah, M. R. - Presenter, Massachusetts Institute of Technology
Santiso, E. - Presenter, Massachusetts Institute of Technology
Trout, B. L. - Presenter, Massachusetts Institute of Technology


Crystallization is an important purification step in pharmaceutical and food industries to obtain high purity chemicals. Most of these industrially relevant molecules can exist in different crystal forms, known as polymorphs, which have different physical properties, which determine their suitability for different applications. Many factors such as the nature of the solvent and degree of under-cooling determine which crystal structure would be stable.1 Nucleation is the key step by which the molecules aggregate together and grow to form the final crystal structure. For rational synthesis of a given polymorph, it is imperative to understand the molecular mechanism of nucleation in small molecular organic crystals. In this research, we study the nucleation mechanism in small organic crystals by employing molecular simulations. Since nucleation is an activated step involving a free energy barrier, we need sophisticated simulation techniques that can study rare events in chemical systems. We employ a technique known as aimless shooting2, which is based on transition path sampling3 to generate reactive trajectories between two phases of our interest. Together with likelihood maximization algorithm2, aimless shooting can provide insights into activation free energy; the nucleation mechanism and the reaction coordinate for the crystallization reaction. We apply these methodologies to study homogeneous crystallization in the melt of benzene. Reactive trajectories between liquid and solid (Form 1 crystal) basins are obtained. A novel set of order parameters which characterize the degree of order in molecular crystals, are screened using likelihood maximization in order to obtain an approximate reaction coordinate. We are able to provide mechanistic insights into the nucleation mechanism and identify the key collective variables for a good reaction coordinate.

1 A.Y. Lee and. A. S. Myerson, MRS Bulletin, 31, 881 (2006). 2 B. Peters and B.L. Trout, J. Chem. Phys., 125, 054108 (2006). 3 P.G. Bolhuis, D. Chandler, C. Dellago and P.L. Geissler, Annu. Rev. Phys. Chem., 54, 291 (2002).