Representations of molecular force fields. I. Ethane: A b i n i t i o and model, harmonic and anharmonic

Abstract

The quadratic and selected cubic force constants for ethane have been computed, using single determinant molecular orbital wavefunctions at the 4‐31G level, with a view to testing and extending model consistent force fields (CFF) for ’’molecular mechanics’’ calculations. Results agree semiquantitatively with experiment, but experimental force constants of sufficient reliability to provide a definitive comparison are not yet available. In a comparison with the most rational general CFF available, that of Ermer and Lifson, the most significant discrepancies found to occur are those for certain stretch–bend couplings assumed to be zero in the CFF but shown to be appreciable by quantum calculation. It is observed that these couplings, but not the stretch–stretch couplings, are well accounted for by a steric interaction model. The ab initio cubic constants examined display the same pattern of conformity with a steric model. Bend–bend–bend and bend–bend–stretch but not all stretch–stretch–stretch interactions agree with those of the steric model. The partial success of the steric model shows that it is possible to represent a large number of interaction constants, quadratic and higher order, by a small number of parameters in molecular mechanics. The failure of the steric model to account for predominantly stretching interactions confirms that ’’classical’’ nonbonded interactions as embodied in conventional Urey–Bradley fields are not the only major contributors to off‐diagonal force constants. An alternative model, the anharmonic model of Warshel, as modified by Kirtman et al., was found to account well for pure stretches but not for bends or stretch–bend interactions.