We present a procedure for locating stationary points on molecular potential energy surfaces based on a constrained Newton-Raphson algorithm. This procedure is based on work by Simons et al. and by Cerjan and Miller but which has not previously been applied to systems with more than a few degrees of freedom. Attention is paid to practical aspects of efficiency and stability in finding both transition states and minima. For cyclooctane, using the AMBER empirical potential energy function, we can "walk" from the global (boat-chair) minimum to find three other low-energy conformers and nine low-energy transition states that connect them. We also report stationary points for N-methylglycylacetamide ("glycyl dipeptide") and N-methylalanylacetamide ("alanyl dipeptide"). The results indicate that it is possible to find multiple conformational minima in a nearly automatic fashion for molecules of this size. Prospects for using these techniques on larger systems are discussed.
All Science Journal Classification (ASJC) codes
- Physical and Theoretical Chemistry