WOLFGANG QUAPP, MICHAEL HIRSCH, OLAF IMIG and DIETMAR H El DR ICH. Mathematisches Institut, and Institut fur Physikalische und Theoretische Chemie, Unwersitdt Leipzig. Augustus-Platz, D-04109 Leipzig, Germany.
The old coordinate <łriving procedurę1 to follow the reaction path and to find the transition structure in Chemical systems is revisited. The well-known gradient criterion, V£(x) = 0, which defines the stationary points of the potential energy surface (PES), is reduced l>y one equation corrcsponding to one search direction. In this manner, instead of minimization orthogonally to the reaction coordinate, abstract curves can be defined connecting stationary points of the PES. Starting at a given minimum, one follows a well-selected direction to reach the saddle of interest. Usually, but not nccessarily, this coordinate will be related to the reaction progress like a reaction path.2 The reduced gradient search locally bas an explicit analytical definitiou. We present a prcdictor-corrector method for tracing such curves. It uscs the gradient and the Hessian matrix at every curve point. We show that the method works well by using a simple update instead of the cxact Hessian.
The 6-dimensional PES of formaldehyde, H2CO, is explored. We carry out ab-initio calculations using the reduced gradient method. 49 minima and saddle points of different index were found. At least seven stationary points representing bonded structures are detected for the first timc, in addition to those located using another search algorithm. Further examples are the loc&lization of the saddle for the HCNr^CNH isomerization (used for stcplcngtb tests) and for the ring closure of azidoazomethine to lH-tetrazole. The results show that following the reduced gradient fonns a serious alternative to locate saddle points in comparison to other rnethods used in quantum chemistry at this time. Additionally, it is possible to determine so-called valley-ridge inflcction points. These points iudicate bifurcations of reaction paths, and they should be of interest also in spectroscopy.
1. I. H. Williams and G. M. Maggiora, J. Mol. Struct. (Theochem) 89, 365 (1982).
2. cf. W. Quapp, O. Imig, and D. Heidrich. in: The Reaction Path in Chemistry: Current Approachcs and Perepectwcs, cd. D. Heidrich (Kluwer, Dordrecht, 1995) p.137.