Ab initio molecular orbital analysis of dimers of cis-formic acid. Implications for condensed phases

The gas phase structure of cis-formic acid dimers is investigated by high-level correlated ab initio molecular orbital methods using large basis sets augmented with both polarization and diffuse functions (up to MP2/D95++(d,p) level). Seven stable dimer structures were located on the potential energy hypersurface of the dimer configurational space with two H-bonding interactions. No stable dimer with only one H-bond was found. The heat of dimerization (10.7-11.3 kcal/mol) of the most stable, C-2h cyclic dimer is in excellent agreement with more recent experimental measurements. Beside the dimer with two equivalent O-H ... O=C H-bonds, there exist two local minima with significant stabilization from C-H ... O interactions. The three weakest of the seven complexes contain exclusively C-H ... O H-bonding interactions. The dimers can interconvert to each other by rotation, disrupting one of their H-bonds. The saddle points and the local minima are anticipated not to play important roles in the gas phase but can have dominant influence on liquid dynamics. The interaction energies of the complexes allow us to assess the relative importance and approximate energetic contributions of the individual H-bonds to the overall stability of the, dimers. It is illustrated that, in addition to the inferred stabilization of two separate O-H ... O=C H-bonds, the most stable C-2h complex is stabilized by about 0.4-0.6 kcal/mol internal cooperative effect, less than in the similar acetic acid dimer. The C-H H-bonding dimers display contraction of the C-H bond lengths and positive frequency shift of the v(C-H) stretching modes relative to the noninteracting monomer. We also show that the effect of BSSE on the intermolecular potential surface of the dimers and, in particular, on the location of the true potential minimum only negligibly influences the interaction energy, but significantly distorts the intermolecular equilibrium geometry.