When a metastable state relaxes to a more stable state, the initial stage is often accompanied by nucleation. Formation of cloud and fog in the atmosphere and sudden bubbling of water superheated in a microwave are familiar examples of nucleation. 

Nucleation plays an important role in various contexts ranging from atmospheric science to material processing. However, fundamental understanding of the process is still lacking and nucleation theory has found only a limited utility in many industrial processes. This is due primarily to the fact that, while nucleation behavior is dictated by statistical properties of aggregates . 

Naturally, computer simulation is expected to play an important role. A simulation study itself, however, poses great difficulties since nucleation is a rare event and a direct simulation requires a very large system to be simulated for a prohibitively long time. The difficulty is particularly acute for complex molecules because time scales of processes involving them are much longer than those for simple systems. 

Our current research effort is aimed at developing efficient algorithms for simulating nucleation of complex molecules. More phenomenological and less computationally demanding approaches are also developed. Our ongoing research projects include a theoretical account of micelle-assisted cavitation, computer simulation of crystallization of molecular fluids, and a development of a force field to describe aromatic hydrocarbon-water clusters.