PREDICTING THE CRYSTAL-STRUCTURE OF ORGANIC MOLECULAR MATERIALS

This paper describes a novel method for predicting the crystal structu re of organic molecular materials which employs a series of successive approximations to focus on structures of high probability, without re sorting to a brute force search and energy minimization of all possibl e structures. The problem of multiple local minima is overcome by assu ming that the crystal structure is closely packed, thereby eliminating 217 of the 230 possible space groups. Configurations within the 13 re maining space groups are searched by rotating the reference molecule a bout Cartesian axes in rotational increments of 15 degrees. Initial en ergy minimization is performed using (6-12) Lennard-Jones pair potenti als to produce a set of closely packed structures. The structures are then refined with the introduction of a Coulombic potential calculated using molecular multipole moments. This method has successfully locat ed local minima which correspond to the observed crystal structures of several saturated and unsaturated hydrocarbons with no a priori infor mation provided. For large polycyclic aromatic hydrocarbons, additiona l refinements of the energy calculations are required to distinguish t he experimental structure from a small number of closely packed struct ures. Our methodology for a priori crystal structure prediction repres ents the most efficient algorithm presented to date, in a field where the first successes have only been described within the past year and have been few and far between. Since our algorithm is capable of locat ing a large number of reasonable structures with similar energy in a s hort period of time, and is more likely to locate a minimum correspond ing to the experimental structure, our program provides a superior fra mework to determine the level of theory required to calculate the inte rmolecular potential. For all but highly asymmetric hydrocarbons, howe ver, distinguishing the observed structure from a large number of high ly probable structures requires more rigorously calculated intermolecu lar interactions than pair potentials, plus an ad hoc electrostatic po tential, and is thus beyond the scope of this paper. All calculations were performed on the Ohio Supercomputer Center's Gray Y-MP.