In the field of computational chemistry, simulation of several to several thousand atoms is possible both in terms of predictive simulation and simulation verifying the consistency with experimentally-measured values, but complex-system simulations on the order of several hundred thousand atoms by the same method used have been difficult due to insufficient computer performance.

By using “dynamic bond-type large-scale molecular dynamic methods” we are able to perform large-scale molecular simulations that even include chemical reactions.

As it is possible to simulate proton transport and the SiC/SiO2 interface in the solid electrolyte of a fuel cell, this can be used in the search for efficient materials for clean technologies. The present method allows the development of structural models of large-scale crystals, which means we can use this as a precondition for smaller-scale, higher-accuracy simulations.

Because any material is viable, simulations are possible of organic materials such as amino acids and inorganic materials (regardless of whether they are single or multi-crystal) such as silicon and other semiconductors, metals and oxide interfaces. As we can elucidate the physical mechanisms of material properties that have previously been used heuristically, it is possible to use not only in searching for novel materials but even for extending the properties of existing materials.

Sponsord research, Collaboration research, Technical consultation