The FMO RI-MP2 analytic nuclear gradient is derived and implemented in GAMESS. First, using the FMO approximation, a system is divided into small fragments to reduce dimension of the problem. The computational cost is further reduced by approximating four-centered two-electron repulsion integrals (4C2ERIs) with the V-type resolution-of-the-identity combined with matrix decomposition methods (RI-V). The latter approximation helps to reduce both the cost of integral transformation and memory storage. In the current research, the RI-MP2 analytic gradient is derived, discussed, and implemented in GAMESS. The FMO implementation is also discussed.

Implementation of analytic gradient for FMO RI-MP2 method

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.