Search

Classical trajectory simulations are used to determine the energy transfer distribution function P ( E f ) for Ne-atoms scattering off the n -hexylthiolate self-assembled monolayer (SAM)/Au{1 1 1} surface. The form of P ( E f ) depends on the angles at which it is measured. Though trapping desorption is insignificant for this system, backward scattering and scattering normal to the surface give rise to a unimodal P ( E f ) in excellent agreement with the Boltzmann distribution. Forward scattering at large polar angles is bimodal with a Boltzmann-like component. These simulation results show that a Boltzmann component in P ( E f ) does not necessarily correspond to a trapping desorption intermediate.

Ab initio direct dynamics was used to calculate product energy distributions for the F + C 2 H 4 → C 2 H 3 F + H reaction. A broad product translational energy distribution, similar to that observed experimentally, is found when the trajectories are initialized with a statistical vibrational energy distribution at the exit channel barrier. The trajectories show that, on average, orbital angular momentum is conserved in going from the exit channel barrier to products, and a model which incorporates this dynamical constraint reproduces the ensemble averaged trajectory results.

Classical trajectories for H 2 CO→H 2 +CO dissociation have been calculated directly from ab initio molecular orbital computations at the HF/3-21G and HF/6-31G** levels of theory, without constructing a global potential energy surface. The classical equations of motion were integrated on the local fifth-order polynomial surfaces fitted to the energies, gradients and hessians computed at the beginning and end points of each step along the trajectory. The calculated vibrational and rotational energy distributions and average impact parameter of the products are in very good agreement with experiment. The relative translational energy is higher than experiment because the barrier is overestimated at both levels of theory.

Both adiabatic and active treatments of the rotational quantum number K are considered in evaluating the RRKM unimolecular rate constant k ( E, J ). If K is treated adiabatically, k ( E, J ) is a weighted average of individual k ( E, J, K ) unimolecular rate constants. For non-variational transition states, this k ( E, J ) is identical to the one which results by treating K as an active degree of freedom with the proper limits placed on K . Another model for treating K as active is to let energy exchange statistically between the vibrational modes and the K -dependent term of the rotational energy. If a separable rigid rotor/harmonic oscillator model is used to calculate the sum and density of states for the transition state and reactant, respectively, the latter K active model is found to give a substantially smaller k ( E, J ).

The photochemistry of ketene has been studied at 3130, 3340 and 3660 A. A chemical trapping technique was used to determine the relative singlet methylene quantum yields. At a pressure of ≈ 120 torr the 3130:3340:3660 A CH 2 ( 1 A 1 ) quantum yield ratio is 1.0:0.16

A quasi-classical trajectory calculation is performed to simulate formaldehyde decomposition following S 1 → S o internal conversion. No mode specificity is observed in the unimolecular decay rates. Each of the trajectories exhibited chaotic motion. These findings suggest that the S 1 → S o internal conversion step does not prepare long-lived vibrational states in ground-state formaldehyde.

Using a potential-energy surface obtained in part from ab initio calculations, the H + CH 3 → CH 4 bimolecular rate constant at T = 300 K is determined from a Monte Carlo classical trajectory study. Representing the CH stretching potential with a standard Morse function instead ofthe ab initio curve increases the calculated rate constant by an order of magnitude. The experimental recombination rate constant is intermediate of the rate constants calculated with the Morse and ab initio stretching potentials. Two properties of the H + CH 3 α CH 4 potential-energy surface which significantly affect the recombination rate constant are the shape of the CH stretching potential and the attenuation of the H 3 CH bending frequencies. Ab initio calculations with a hierarchy of basis sets and treatment of electron correlation indicate the latter is properly described [13]. The exact shape of the CH stretching potential is not delineated by the ab initio calculations, since the ab initio calculations are not converged for bond lengths of 2.0–3.0 A [12]. However, the form of this stretching potential deduced from the highest-level ab initio calculations, and fit analytically by eq. (2), is significantly different from a Morse function. The experimental recombination rate constant is intermediate of the rate constants calculated with the Morse and ab initio CH stretching potentials. This indicates that the actual CH potential energy curve lies between the Morse and ab initio curves. This is consistent with the finding that potential energy curves for diatomics are not well described by a Morse function [12].

Ab initio calculations are used to study the attenuation of HCC bending force constants in ethylene and propylene upon HC bond extension. The attenuations found for these two molecules and for CH 4 and SiH 2 in previous work are nearly the same. To a good approximation, attenuation functions and parameters are transferable between these molecules. In contrast to these similar bend attenuations, a decidedly more rapid attenuation is found for the O 2 ·H bend in HO 2 .

A simple semi-empirical model is found to explain numerous trends observed in canonical variational transition state theory calculations for association reactions. The trends include the relationship of canonical transition state structures to potential energy surface properties, temperature, and rotational angular momentum. The model gives near quantitative results for H + CH3→CH4 association.

Two efficient methods are presented for performing exact microcanonical sampling of harmonic oscillators. It is shown that orthant sampling, which has previously been used to approximately sample microcanonical ensembles of anharmonic oscillators, is an exact sampling technique when applied to harmonic oscillators.

A benzene fragment HC3/DC3 is used as a model for benzene in a quasiclassical trajectory study of the intramolecular decay of the n=3 overtone. The calculated linewidth (fwhm) for the CH overtone is ≈ 4 cm−1 and that for CD

It is pointed out that the proper comparison to make between theoretical and experimental results on collisional energy transfer in highly excited molecules involves the rate of energy transfer (or its moments) rather than the average energy transfer per collision. This rate can be computed from trajectory studies without the need to define a “collision” or to omit arbitrary defined elastic trajectories. Illustrative calculations for CH3CH2Cl/Ne collisions are presented.

Relative translational energies are computed for H C C Cl decomposition using classical trajectories. It is found that excited molecules with non-RRKM lifetime distributions dissociate to products with RRKM translational energy distributions. This result suggests that statistical relative translational energy distributions do not necessarily imply complete intramolecular vibrational energy redistribution.

The semiclassical eigenvalues of a realistic three-dimensional model Hamiltonian are computed with a method proposed recently by Miller that makes use of a single arbitrary trajectory. The calculated results are compared with the earlier results obtained using the Sorbie—Handy method and are found to be in good agreement.

Preliminary results are reported of a Monte Carlo trajectory study of intramolecular vibrational energy relaxation in a model of benzene. Very little relaxation is found to occur on a picosecond timescale. The rate of intramolecular vibrational energy relaxation appears to be dependent upon the nature of the initial energy distribution. Deuteration causes a pronounced increase in the relaxation.

The sensitivity of vibrational overtone excitation experiments to non-RRKM behavior is considered. It is proposed that the experimental results are not inconsistent with intrinsically non-RRKM unimolecular lifetime distribution.

CF 4 and C 8 F 18 collisionally remove excess energy, presumably vibrational, from CH 2 ( 1 A 1 ). Results from increasing proportions of C 8 F 18 with CH 2 N 2 and CH 2 CO precursors indicated that Δ H fo 0 [ 1 CH 2 ( 1 A 1 )] ≈ 101 kcal/mole. A simplified model indicates that 80–800 CF 4 and 10–100 C 8 F 18 collisions remove one bending vibrational quantum from CH 2 ( 1 A 1 ).