ANALYTICAL 2ND DERIVATIVES OF THE ENERGY IN MNDO METHODS

Analytical second derivatives of the energy are derived and efficientl y implemented for semiempirical MNDO-ty-pe methods including AM1, PM3, and MNDO/d. A new algorithm for the simultaneous solution of several CPHF equations is proposed in which the amount of memory required is i ndependent of the number of iterations. The analytical approach is fas ter than the numerical approach typically by a factor of 5 and exhibit s a reliable convergence over a wide range of molecules. The asymptoti c memory and secondary storage requirements of the reported procedure can be as low as O(N-2) without significant degradation of the perform ance. (C) 1996 by John Wiley & Sons, Inc.