Albert F. Wagner

Ab initio wavenumber accurate spectroscopy: 1CH2 and HCN vibrational levels on automatically generated IMLS potential energy surfaces.

We report here calculated J = 0 vibrational frequencies for (1)CH(2) and HCN with root-mean-square error relative to available measurements of 2.0 cm(-1) and 3.2 cm(-1), respectively. These results are obtained with DVR calculations with a dense grid on ab initio potential energy surfaces (PESs). The ab initio electronic structure calculations employed are Davidson-corrected MRCI calculations with double-, triple-, and quadruple-zeta basis sets extrapolated to the complete basis set (CBS) limit. In the (1)CH(2) case, Full CI tests of the Davidson correction at small basis set levels lead to a scaling of the correction with the bend angle that can be profitably applied at the CBS limit. Core-valence corrections are added derived from CCSD(T) calculations with and without frozen cores. Relativistic and non-Born-Oppenheimer corrections are available for HCN and were applied. CBS limit CCSD(T) and CASPT2 calculations with the same basis sets were also tried for HCN. The CCSD(T) results are noticeably less accurate than the MRCI results while the CASPT2 results are much poorer. The PESs were generated automatically using the local interpolative moving least-squares method (L-IMLS). A general triatomic code is described where the L-IMLS method is interfaced with several common electronic structure packages. All PESs were computed with this code running in parallel on eight processors. The L-IMLS method provides global and local fitting error measures important in automatically growing the PES from initial ab initio seed points. The reliability of this approach was tested for (1)CH(2) by comparing DVR-calculated vibrational levels on an L-IMLS ab initio surface with levels generated by an explicit ab initio calculation at each DVR grid point. For all levels ( approximately 200) below 20 000 cm(-1), the mean unsigned difference between the levels of these two calculations was 0.1 cm(-1), consistent with the L-IMLS estimated mean unsigned fitting error of 0.3 cm(-1). All L-IMLS PESs used in this work have comparable mean unsigned fitting errors, implying that fitting errors have a negligible role in the final errors of the computed vibrational levels with experiment. Less than 500 ab initio calculations of the energy and gradients are required to achieve this level of accuracy.

Quantum mechanical calculations of rotational‐vibrational scattering in homonuclear diatom‐atom collisions

Most calculations of the vibrational scattering of diatom-atom collisions use the breathing sphere approximation (BSA) of orientation averaging the intermolecular potential. The resulting angularly symmetric npotential can not cause rotational scattering. We determine the error introduced by the BSA into observables of the vibrational scattering of low-energy homonuclear diatom-atom collisions by comparing two quantum mechanical calculations, one with the BSA and the other with the full angularly asymmetric intermolecular potential. For ·reasons of economy the rotational scattering of the second calculation is restricted by the use of special incomplete channel sets in the expansion of the scattering wavefunction. nThree representative collision systems are studied: H_2-Ar, O_2-He, and I_2-He. From our calculations, we nreach two conclusions. First, the BSA can be used to analyze accurately experimental measurements of nvibrational scattering. Second, measurements most sensitive to the symmetric part of the intermolecular npotential are, in order, elastic cross sections, inelastic cross sections, and inelastic differential cross sections. nElastic differential cross sections are sensitive to the potential only if the collision is sticky, with scattering nover a wide range of angles; I_2-He is such a collision. Otherwise the potential sensitivity of elastic differential ncross sections is concentrated in the experimentally difficult region of very small angle scattering.

Improved multidimensional semiclassical tunneling theory.

We show that the analytic multidimensional semiclassical tunneling formula of Miller et al. [Miller, W. H.; Hernandez, R.; Handy, N. C.; Jayatilaka, D.; Willets, A. Chem. Phys. Lett. 1990, 172, 62] is qualitatively incorrect for deep tunneling at energies well below the top of the barrier. The origin of this deficiency is that the formula uses an effective barrier weakly related to the true energetics but correctly adjusted to reproduce the harmonic description and anharmonic corrections of the reaction path at the saddle point as determined by second order vibrational perturbation theory. We present an analytic improved semiclassical formula that correctly includes energetic information and allows a qualitatively correct representation of deep tunneling. This is done by constructing a three segment composite Eckart potential that is continuous everywhere in both value and derivative. This composite potential has an analytic barrier penetration integral from which the semiclassical action can be derived and then used to define the semiclassical tunneling probability. The middle segment of the composite potential by itself is superior to the original formula of Miller et al. because it incorporates the asymmetry of the reaction barrier produced by the known reaction exoergicity. Comparison of the semiclassical and exact quantum tunneling probability for the pure Eckart potential suggests a simple threshold multiplicative factor to the improved formula to account for quantum effects very near threshold not represented by semiclassical theory. The deep tunneling limitations of the original formula are echoed in semiclassical high-energy descriptions of bound vibrational states perpendicular to the reaction path at the saddle point. However, typically ab initio energetic information is not available to correct it. The Supporting Information contains a Fortran code, test input, and test output that implements the improved semiclassical tunneling formula.

Theoretical/experimental comparison of deep tunneling decay of quasi-bound H(D)OCO to H(D) + CO2

The measured H(D)OCO survival fractions of the photoelectron-photofragment coincidence experiments by the Continetti group are qualitatively reproduced by tunneling calculations to H(D) + CO2 on several recent ab initio potential energy surfaces for the HOCO system. The tunneling calculations involve effective one-dimensional barriers based on steepest descent paths computed on each potential energy surface. The resulting tunneling probabilities are converted into H(D)OCO survival fractions using a model developed by the Continetti group in which every oscillation of the H(D)-OCO stretch provides an opportunity to tunnel. Four different potential energy surfaces are examined with the best qualitative agreement with experiment occurring for the PIP-NN surface based on UCCSD(T)-F12a/AVTZ electronic structure calculations and also a partial surface constructed for this study based on CASPT2/AVDZ electronic structure calculations. These two surfaces differ in barrier height by 1.6 kcal/mol but when matched at the saddle point have an almost identical shape along their reaction paths. The PIP surface is a less accurate fit to a smaller ab initio data set than that used for PIP-NN and its computed survival fractions are somewhat inferior to PIP-NN. The LTSH potential energy surface is the oldest surface examined and is qualitatively incompatible with experiment. This surface also has a small discontinuity that is easily repaired. On each surface, four different approximate tunneling methods are compared but only the small curvature tunneling method and the improved semiclassical transition state method produce useful results on all four surfaces. The results of these two methods are generally comparable and in qualitative agreement with experiment on the PIP-NN and CASPT2 surfaces. The original semiclassical transition state theory method produces qualitatively incorrect tunneling probabilities on all surfaces except the PIP. The Eckart tunneling method uses the least amount of information about the reaction path and produces too high a tunneling probability on PIP-NN surface, leading to survival fractions that peak at half their measured values.

Effect of the potential well on vibrational scattering and the validity of SSH theory

The vibrational de‐excitation probability, P_(10), is calculated quantum mechanically over a large energy range for models of three collision systems: O_2-O_2,Cl_2-Cl_2, and Br_2-Br_2. The vibrational de‐excitation cross section, σ_(10), is similarly calculated for the Cl_2-Cl_2 model. P_(10) and σ_(10) are obtained for the Lennard‐Jones intermolecular potential and three other well‐less potentials designed to duplicate the scattering of the Lennard‐Jones potential. The results emphasize the adiabatic nature of potentials with wells and indicate that the acceleration approximation for the effect of the well is not valid. The curves of P_(10) and σ_(10) as a function of initial translational energy are used to obtain exact collision numbers. These numbers are compared to the results of SSH theory. SSH theory is found to predict collision numbers with reasonable accuracy except at low temperatures. SSH theory is also not suitable for analyzing experimental collision numbers for the well depth potential parameter.

A classical trajectory study of the intramolecular dynamics, isomerization, and unimolecular dissociation of HO2

The classical dynamics and rates of isomerization and dissociation of HO2 have been studied using two potential energy surfaces (PESs) based on interpolative fittings of ab initio data: An interpolative moving least-squares (IMLS) surface [A. Li, D. Xie, R. Dawes, A. W. Jasper, J. Ma, and H. Guo, J. Chem. Phys. 133, 144306 (2010)] and the cubic-spline-fitted PES reported by Xu, Xie, Zhang, Lin, and Guo (XXZLG) [J. Chem. Phys. 127, 024304 (2007)]. Both PESs are based on similar, though not identical, internally contracted multi-reference configuration interaction with Davidson correction (icMRCI+Q) electronic structure calculations; the IMLS PES includes complete basis set (CBS) extrapolation. The coordinate range of the IMLS PES is limited to non-reactive processes. Surfaces-of-section show similar generally regular phase space structures for the IMLS and XXZLG PESs with increasing energy. The intramolecular vibrational energy redistribution (IVR) at energies above and below the threshold of isomerization is slow, especially for O-O stretch excitations, consistent with the regularity in the surfaces-of-section. The slow IVR rates lead to mode-specific effects that are prominent for isomerization (on both the IMLS and XXZLG) and modest for unimolecular dissociation to H + O2 (accessible only on the XXZLG PES). Even with statistical distributions of initial energy, slow IVR rates result in double exponential decay for isomerization, with the slower rate correlated with slow IVR rates for O-O vibrational excitation. The IVR and isomerization rates computed for the IMLS and XXZLG PESs are quantitatively, but not qualitatively, different from one another with the largest differences ascribed to the ~2 kcal/mol difference in the isomerization barrier heights. The IMLS and XXZLG results are compared with those obtained using the global, semi-empirical double-many-body expansion DMBE-IV PES [M. R. Pastrana, L. A. M. Quintales, J. Brandão, and A. J. C. Varandas, J. Chem. Phys. 94, 8073 (1990)], for which the surfaces-of-section display more irregular phase space structure, much faster IVR rates, and significantly less mode-specific effects in isomerization and unimolecular dissociation. The calculated IVR results for all three PESs are reasonably well represented by an analytic, coupled three-mode energy transfer model.

A classical trajectory study of the dissociation and isomerization of C2H5.

Motivated by photodissociation experiments in which non-RRKM nanosecond lifetimes of the ethyl radical were reported, we have performed a classical trajectory study of the dissociation and isomerization of C2H5 over the energy range 100-150 kcal/mol. We used a customized version of the AIREBO semiempirical potential (Stuart, S. J.; et al. J. Chem. Phys. 2000, 112, 6472-6486) to more accurately describe the gas-phase decomposition of C2H5. This study constitutes one of the first gas-phase applications of this potential form. At each energy, 10,000 trajectories were run and all underwent dissociation in less than 100 ps. The calculated dissociation rate constants are consistent with RRKM models; no evidence was found for nanosecond lifetimes. An analytic kinetics model of isomerization/dissociation competition was developed that incorporated incomplete mode mixing through a postulated divided phase space. The fits of the model to the trajectory data are good and represent the trajectory results in detail through repeated isomerizations at all energies. The model correctly displays single exponential decay at lower energies, but at higher energies, multiexponential decay due to incomplete mode mixing becomes more apparent. At both ends of the energy range, we carried out similar trajectory studies on CD2CH3 to examine isotopic scrambling. The results largely support the assumption that a H or a D atom is equally likely to dissociate from the mixed-isotope methyl end of the molecule. The calculated fraction of products that have the D atom dissociation is ∼20%, twice the experimental value available at one energy within our range. The calculated degree of isotopic scrambling is non-monotonic with respect to energy due to a non-monotonic ratio of the isomerization to dissociation rate constants.

Fulminant Pseudomembranous Colitis

Pseudomembranous colitis is an increasingly frequent nosocomial infection; it commonly occurs after the use of broad-spectrum antibiotics, that allow overgrowth of Clostridium difficile. <