Search

We present global ground-state potential energy surfaces for the quintet and singlet spin states of the O 4 system that are suitable for treating high-energy vibrational-rotational energy transfer and collision-induced dissociation in electronically adiabatic, spin-conserving O 2 -O 2 collisions. The surfaces are based on MS-CASPT2/maug-cc-pVTZ electronic structure calculations with scaled external correlation. The active space has 16 electrons in 12 orbitals. The calculations cover nine kinds of geometrical arrangements corresponding to dissociative diatom-diatom collisions of O 2 , geometries corresponding to O 3 -O, geometries identified by running trajectories, and geometries along linear synchronous transit paths. The global ground-state potential energy surfaces were obtained by a many-body approach with an accurate O-O pairwise interaction and a fit of the many-body interaction to 12 684 electronic structure data points for the singlet and 10 543 electronic structure data points for the quintet. The many-body fit is based on permutationally invariant polynomials in terms of bond-order functions of the six interatomic distances; the bond-order functions are mixed exponential-Gaussian functions.

Accurate modeling of high-temperature hypersonic flows in the atmosphere requires consideration of collision-induced dissociation of molecular species and energy transfer between the translational and internal modes of the gas molecules. Here, we describe a study of the N2 + N2⟶N2 + 2N and N2 + N2⟶4N nitrogen dissociation reactions using the quasiclassical trajectory (QCT) method. The simulations used a new potential energy surface for the N4 system; the surface is an improved version of one that was presented previously. In the QCT calculations, initial conditions were determined based on a two-temperature model that approximately separates the translational-rotational temperature from the vibrational temperature of the N2 diatoms. Five values from 8000 K to 30,000 K were considered for each of the two temperatures. Over 2.4 × 10(9) trajectories were calculated. We present results for ensemble-averaged dissociation rate constants as functions of the translational-rotational temperature T and the vibrational temperature T(v). The rate constant depends more strongly on T when T(v) is low, and it depends more strongly on T(v) when T is low. Quasibound reactant states contribute significantly to the rate constants, as do exchange processes at higher temperatures. We discuss two sets of runs in detail: an equilibrium test set in which T = T(v) and a nonequilibrium test set in which T(v) < T. In the equilibrium test set, high-v and moderately-low-j molecules contribute most significantly to the overall dissociation rate, and this state specificity becomes stronger as the temperature decreases. Dissociating trajectories tend to result in a major loss of vibrational energy and a minor loss of rotational energy. In the nonequilibrium test set, as T(v) decreases while T is fixed, higher-j molecules contribute more significantly to the dissociation rate, dissociating trajectories tend to result in a greater rotational energy loss, and the dissociation probability's dependence on v weakens. In this way, as T(v) decreases, rotational energy appears to compensate for the decline in average vibrational energy in promoting dissociation. In both the equilibrium and nonequilibrium test sets, in every case, the average total internal energy loss in the dissociating trajectories is between 10.2 and 11.0 eV, slightly larger than the equilibrium potential energy change of N2 dissociation.

Fitting potential energy surfaces to analytic forms is an important first step for efficient molecular dynamics simulations. Here, we present an improved version of the local interpolating moving least squares method (L-IMLS) for such fitting. Our method has three key improvements. First, pairwise interactions are modeled separately from many-body interactions. Second, permutational invariance is incorporated in the basis functions, using permutationally invariant polynomials in Morse variables, and in the weight functions. Third, computational cost is reduced by statistical localization, in which we statistically correlate the cutoff radius with data point density. We motivate our discussion in this paper with a review of global and local least-squares-based fitting methods in one dimension. Then, we develop our method in six dimensions, and we note that it allows the analytic evaluation of gradients, a feature that is important for molecular dynamics. The approach, which we call statistically localized, permutationally invariant, local interpolating moving least squares fitting of the many-body potential (SL-PI-L-IMLS-MP, or, more simply, L-IMLS-G2), is used to fit a potential energy surface to an electronic structure dataset for N4. We discuss its performance on the dataset and give directions for further research, including applications to trajectory calculations.