QCT and QM calculations of the Cl(2P) + NH3 reaction: influence of the reactant well on the dynamics

Accurate ab initio based potential energy surface and kinetics of the Cl + NH3 → HCl + NH2 reaction

The gas-phase reaction Cl + NH3 → HCl + NH2 is a prototypical hydrogen abstraction reaction, whose minimum energy path involves several intermediate complexes. In this work, a full-dimensional, spin-orbit corrected potential energy surface (SOC PES) is constructed for the ground electronic state of the Cl + NH3 reaction. About 52 000 energy points are sampled and calculated at the UCCSD(T)-F12a/aug-cc-pVTZ level, in which the data points located in the entrance channel are spin-orbit corrected. The spin-orbit corrections are predicted by a fitted three-dimensional energy surface from about 7520 energy points in the entrance channel at the level of CASSCF (15e, 11o)/aug-cc-pVTZ. The fundamental-invariant neural network method is utilized to fit the SOC PES, resulting in a total root mean square error of 0.12 kcal mol-1. The calculated thermal rate constants of the Cl + NH3 → HCl + NH2 reaction on the SOC PES with the soft-zero-point energy constraint agree reasonably well with the available experimental values.

The CN(X 2Σ+) + C2H6 reaction: Dynamics study based on an analytical full-dimensional potential energy surface

The hydrogen abstraction reaction of the cyano radical with molecules of ethane presents some interesting points in the chemistry from ultra-cold to combustion environments especially with regard to HCN(v) product vibrational distribution. In order to understand its dynamics, a new analytical full-dimensional potential energy surface was developed, named PES-2023. It uses a combination of valence bond and mechanic molecular terms as the functional form, fitted to high-level ab initio calculations at the explicitly correlated CCSD(T)-F12/aug-cc-pVTZ level on a reduced and selected number of points describing the reactive process. The new surface showed a continuous and smooth behavior, describing reasonably the topology of the reaction: high exothermicity, low barrier, and presence of intermediate complexes in the entrance and exit channels. Using quasi-classical trajectory calculations (QCT) on the new PES-2023, a dynamics study was performed at room temperature with special emphasis on the HCN(v1,v2,v3) product stretching and bending vibrational excitations, and the results were compared with the experimental evidence, which presented discrepancies in the bending excitation. The available energy was mostly deposited as HCN(v) vibrational energy with the vibrational population inverted in the CH stretching mode and not inverted in the CN stretching and bending modes, thus simulating the experimental evidence. Other dynamics properties at room temperature were also analyzed; cold rotational energy distribution was found, associated with a linear and soft transition state, and backward scattering distribution was found, associated with a rebound mechanism.

Theoretical study of the Cl(2P) + SiH4 reaction: global potential energy surface and product pair-correlated distributions. Comparison with experiment

For the theoretical study of the title reaction, an analytical full-dimensional potential energy surface named PES-2021 was developed for the first time, by fitting high-level explicitly-correlated ab initio data. This reaction presented high exothermicity, (298 K) = -11.6 kcal mol^-1, reproducing the experimental evidence; it is a barrierless reaction and no intermediate complexes were found. PES-2021 is a continuous and smooth potential energy surface, it includes intuitive concepts in its development and fitting, such as stretching and bending nuclear motions, and it presents analytical first energy derivatives. Based on PES-2021, kinetics and dynamics studies were carried out using quasi-classical trajectory calculations. In the kinetics study, over the temperature range 300-450 K, we observed that rate constants were practically independent of temperature, with an almost zero activation energy, as compared to 0.0 and -0.48 kcal mol^-1 experimentally reported. In this kinetics study the role of the spin-orbit effect on reactivity was analysed. In the dynamics study, different product pair-correlated dynamics properties were compared with the only experimental evidence: product energy partition, product vibrational distribution, product angular distribution and product speed distribution. We observed two mechanisms of reaction, a stripping mechanism associated with large impact parameters and forward scattering, and an indirect mechanism associated with sideways-backward scattering related with "nearly-trapped" trajectories due to the product rotation. In general, theoretical results reasonably simulate the experimental measurements when they consider some rotational and vibrational constraints as well as binning techniques to mimic a quantum-mechanical behaviour. Although the agreement is not quantitative, the present results shed light on the mechanism of this difficult polyatomic reactive system.

Quasi-Classical Trajectory Study of the CN + NH3 Reaction Based on a Global Potential Energy Surface

Based on a combination of valence-bond and molecular mechanics functions which were fitted to high-level ab initio calculations, we constructed an analytical full-dimensional potential energy surface, named PES-2020, for the hydrogen abstraction title reaction for the first time. This surface is symmetrical with respect to the permutation of the three hydrogens in ammonia, it presents numerical gradients and it improves the description presented by previous theoretical studies. In order to analyze its quality and accuracy, stringent tests were performed, exhaustive kinetics and dynamics studies were carried out using quasi-classical trajectory calculations, and the results were compared with the available experimental evidence. Firstly, the properties (geometry, vibrational frequency and energy) of all stationary points were found to reasonably reproduce the ab initio information used as input; due to the complicated topology with deep wells in the entrance and exit channels and a “submerged” transition state, the description of the intermediate complexes was poorer, although it was adequate to reasonably simulate the kinetics and dynamics of the title reaction. Secondly, in the kinetics study, the rate constants simulated the experimental data in the wide temperature range of 25–700 K, improving the description presented by previous theoretical studies. In addition, while previous studies failed in the description of the kinetic isotope effects, our results reproduced the experimental information. Finally, in the dynamics study, we analyzed the role of the vibrational and rotational excitation of the CN(v,j) reactant and product angular scattering distribution. We found that vibrational excitation by one quantum slightly increased reactivity, thus reproducing the only experimental measurement, while rotational excitation strongly decreased reactivity. The scattering distribution presented a forward-backward shape, associated with the presence of deep wells along the reaction path. These last two findings await experimental confirmation.

Quasi-Classical Trajectory Dynamics Study of the Cl(2P) + C2H6 → HCl(v,j) + C2H5 Reaction. Comparison with Experiment

To understand and simulate the dynamics behavior of the title reaction, QCT calculations were performed on a recently developed global analytical potential energy surface, PES-2017. These calculations combine the classical description of the dynamics with pseudoquantization in the reactants and products to perform a theoretical/experimental comparison on the same footing. Thus, in the products a series of constraints are included to analyze the HCl(v = 0,j) product, which is experimentally detected. At collision energies of 5.5 and 6.7 kcal mol^-1 the largest fraction of available energy is deposited as translation, 67%, while the ethyl radical shows significant internal energy, 27%, and so it does not act as a spectator of the reaction, thus reproducing recent experimental evidence. The HCl(v=0, j) rotational distribution is cold, peaking at j = 2, only one unit hotter than experiment, which represents an error of 0.12 kcal mol^-1. At a collision energy of 5.5 kcal mol^-1 product translational distribution is slightly hotter than experiment, but at 6.7 kcal mol^-1 agreement with recent experiments is practically quantitative, suggesting that the first experiments should be revised. In addition, we observe that the HCl(v=0, j) scattering distribution shifts from isotropic at low values of j to backward at high values of j, which is in agreement with experimental data. Finally, no evidence was found for the "chattering" mechanism suggested to explain the low translational energy of the HCl product in the backward scattering region. In sum, agreement with experiments of a series of sensible dynamic properties permits us to be optimistic on the quality and accuracy of the theoretical tools used in the present work, QCT and PES-2017.

QCT dynamics study of OH/OD + GeH4 reactions. The problem of water bending excitation

The difficulties in the description of the water bending population could be related to the zero-point violation problem in QCT calculations.

Dynamics study of the OH + NH3 hydrogen abstraction reaction using QCT calculations based on an analytical potential energy surface

To understand the reactivity and mechanism of the OH + NH3 → H2O + NH2 gas-phase reaction, which evolves through wells in the entrance and exit channels, a detailed dynamics study was carried out using quasi-classical trajectory calculations. The calculations were performed on an analytical potential energy surface (PES) recently developed by our group, PES-2012 [Monge-Palacios et al. J. Chem. Phys. 138, 084305 (2013)]. Most of the available energy appeared as H2O product vibrational energy (54%), reproducing the only experimental evidence, while only the 21% of this energy appeared as NH2 co-product vibrational energy. Both products appeared with cold and broad rotational distributions. The excitation function (constant collision energy in the range 1.0-14.0 kcal mol(-1)) increases smoothly with energy, contrasting with the only theoretical information (reduced-dimensional quantum scattering calculations based on a simplified PES), which presented a peak at low collision energies, related to quantized states. Analysis of the individual reactive trajectories showed that different mechanisms operate depending on the collision energy. Thus, while at high energies (E(coll) ≥ 6 kcal mol(-1)) all trajectories are direct, at low energies about 20%-30% of trajectories are indirect, i.e., with the mediation of a trapping complex, mainly in the product well. Finally, the effect of the zero-point energy constraint on the dynamics properties was analyzed.

Ab initio based potential energy surface and kinetics study of the OH + NH3 hydrogen abstraction reaction

A full-dimensional analytical potential energy surface (PES) for the OH + NH3 → H2O + NH2 gas-phase reaction was developed based exclusively on high-level ab initio calculations. This reaction presents a very complicated shape with wells along the reaction path. Using a wide spectrum of properties of the reactive system (equilibrium geometries, vibrational frequencies, and relative energies of the stationary points, topology of the reaction path, and points on the reaction swath) as reference, the resulting analytical PES reproduces reasonably well the input ab initio information obtained at the coupled-cluster single double triple (CCSD(T)) = FULL/aug-cc-pVTZ//CCSD(T) = FC/cc-pVTZ single point level, which represents a severe test of the new surface. As a first application, on this analytical PES we perform an extensive kinetics study using variational transition-state theory with semiclassical transmission coefficients over a wide temperature range, 200-2000 K. The forward rate constants reproduce the experimental measurements, while the reverse ones are slightly underestimated. However, the detailed analysis of the experimental equilibrium constants (from which the reverse rate constants are obtained) permits us to conclude that the experimental reverse rate constants must be re-evaluated. Another severe test of the new surface is the analysis of the kinetic isotope effects (KIEs), which were not included in the fitting procedure. The KIEs reproduce the values obtained from ab initio calculations in the common temperature range, although unfortunately no experimental information is available for comparison.