A new, simple, computational technique allows molecular electron densities to be divided into atoms which have intuitively correct shapes and charges.

A simple computational technique is introduced for generating atomic electron densities using an iterated stockholder procedure. It is proven that the procedure is always convergent and leads to unique atomic densities. The resulting atomic densities are shown to have chemically intuitive and reasonable charges, and the method is used to analyze the hydrogen bonding in the minimum energy configuration of the water dimer and charge transfer in the borazane molecule.

The structures and vibrational spectra of small methanol clusters from dimer to decamer have been calculated using a newly developed intermolecular potential which is essentially based on monomer wave functions. Special care has been taken for the description of the electrostatic interaction using a distributed multipole representation and including a penetration term. In addition, the potential model consists of repulsion, dispersion, and induction terms. Based on this potential model cluster structures have been calculated. The lowest energy dimer configuration is linear, while from trimer to decamer for the most stable structures ring configurations were found. Tetramer, hexamer, and octamer have S4-, S6-, and S8-symmetry, respectively. Vibrational spectra of the CO stretch and the OH stretch mode have been determined in the harmonic and in the anharmonic approximation using perturbation theory and variational calculations. Up to the tetramer the experimental spectra of the CO stretch mode are well reproduced, for larger clusters an increasing blueshift with respect to the experimental evidence is found. The experimental data for the OH stretch mode of the dimer are fairly well reproduced in all approximations, however, the spectrum of the trimer can only be reproduced using the variational calculation which includes Darling–Dennison resonance terms.

We construct a rigid-body (five-dimensional) potential-energy surface for the water-hydrogen complex using scaled perturbation theory (SPT). An analytic fit of this surface is obtained, and, using this, two minima are found. The global minimum has C2v symmetry, with the hydrogen molecule acting as a proton donor to the oxygen atom on water. A local minimum with Cs symmetry has the hydrogen molecule acting as a proton acceptor to one of the hydrogen atoms on water, where the OH bond and H2 are in a T-shaped configuration. The SPT global minimum is bound by 1097 microEh (Eh approximately 4.359744 x 10(-18) J). Our best estimate of the binding energy, from a complete basis set extrapolation of coupled-cluster calculations, is 1076.1 microEh. The fitted surface is used to calculate the second cross virial coefficient over a wide temperature range (100-3000 K). Three complementary methods are used to quantify quantum statistical mechanical effects that become significant at low temperatures. We compare our results with experimental data, which are available over a smaller temperature range (230-700 K). Generally good agreement is found, but the experimental data are subject to larger uncertainties.

No abstract

We construct potential-energy surfaces for the water–neon and water–argon complexes from scaled perturbation theory, and calibrate them using accurate supermolecule data. Our best estimates of the binding energies for these two systems are 66.9 and 142.7 cm−1, respectively, where the latter value is in good agreement with the spectroscopically determined AW2 potential. We calculate second virial coefficients, B12(T), and the related property φ12=B12−T(dB12/dT), and compare our results with experimental data for water–argon. The perturbation theory and AW2 B12(T) results are consistent, and demonstrate that current theoretical approaches yield more precise second virial coefficient data than any in the literature. Our φ12 calculations are in good agreement with experimental results derived from enthalpy-of-mixing data, though our estimated uncertainties are significantly smaller.

A virial expansion of fluid pressure in powers of the density can be used to calculate a wealth of thermodynamic information, but the Nth virial coefficient, which multiplies the Nth power of the density in the expansion, becomes rapidly more complicated with increasing N. This Letter shows that the Nth virial coefficient can be calculated using a method that scales exponentially with N in computer time and memory. This is orders of magnitude more efficient than any existing method for large N, and the method is simple and general. New results are presented for N = 11 and 12 for hard spheres, and N = 9 and 10 for soft spheres.

Three active learning schemes are used to generate training data for Gaussian process interpolation of intermolecular potential energy surfaces. These schemes aim to achieve the lowest predictive error using the fewest points and therefore act as an alternative to the status quo methods involving grid-based sampling or space-filling designs like Latin hypercubes (LHC). Results are presented for three molecular systems: CO 2 −Ne, CO 2 −H 2 and Ar 3. For each system, two of the active learning schemes proposed notably outperform LHC designs of comparable size, and in two of the systems, produce an error value an order of magnitude lower than the one produced by the LHC method. The procedures can be used to select a subset of points from a large pre-existing data set, to select points to generate data de-novo, or to supplement an existing data set to improve accuracy.