An algorithm for geometry optimization without analytical gradients

A numerical algorithm for locating both minima and transition states designed for use in the ab initio program package GAUSSIAN 82 is presented. It is based on the RFO method of Simons and coworkers and is effectively the numerical version of an analytical algorithm (OPT = EF) previously published in this journal. The algorithm is designed to make maximum use of external second derivative information obtained from prior optimizations at lower levels of theory. It can be used with any wave function for which an energy can be calculated and is about two to three times faster than the default DFP algorithm (OPT = FP) supplied with GAUSSIAN 82.