Search

In classical and quasiclassical trajectory chemical dynamics simulations, the atomistic dynamics of collisions, chemical reactions, and energy transfer are studied by solving the classical equations of motion. These equations require the potential energy and its gradient for the chemical system under study, and they may be obtained directly from an electronic structure theory. This article reviews such direct dynamics simulations. The accuracy of classical chemical dynamics is considered, with simulations highlighted for the F− + CH3OOH reaction and of energy transfer in collisions of CO2 with a perfluorinated self-assembled monolayer (F-SAM) surface. Procedures for interfacing chemical dynamics and electronic structure theory computer codes are discussed. A Hessian-based predictor–corrector algorithm and high-accuracy Hessian updating algorithm, for enhancing the efficiency of direct dynamics simulations, are described. In these simulations, an ensemble of trajectories is calculated which represents the experimental and chemical system under study. Algorithms are described for selecting the appropriate initial conditions for bimolecular and unimolecular reactions, gas-surface collisions, and initializing trajectories at transition states and conical intersections. Illustrative direct dynamics simulations are presented for the Cl− + CH3I SN2 reaction, unimolecular decomposition of the epoxy resin constituent CH3NHCHCHCH3 versus temperature, collisions and reactions of N-protonated diglycine with a F-SAM surface that has a reactive head group, and the product energy partitioning for the post-transition state dynamics of C2H5F HF + C2H4 dissociation. © 2012 John Wiley & Sons, Ltd.

The direct dynamics technique, based on an AMI semiempirical Hamiltonian modified with specific reaction path parameters, has been employed to study the constant energy and thermal isomerisation of S,S-trans-cyclopropane-1,2-d 2 and cyclopropane-1,2,3-d3. At a classical energy of 48.25 kcal mol -1 above the conrotatory saddle point, the dynamics is mode specific. This raises the possibility of stereochemical control in this energy regime. The sensitivity of the reaction mechanism to the character of the excited normal mode eigenvector is also apparent under thermal conditions at 695 K. Similar product branching ratios are obtained for the d 2 and d 3 species at this temperature, when the trajectories are initialised at the conrotatory saddle point.

In classical and quasiclassical trajectory chemical dynamics simulations, the atomistic dynamics of collisions, chemical reactions, and energy transfer are studied by solving the classical equations of motion. These equations require the potential energy and its gradient for the chemical system under study, and they may be obtained directly from an electronic structure theory. This article reviews such direct dynamics simulations. The accuracy of classical chemical dynamics is considered, with simulations highlighted for the F− + CH3OOH reaction and of energy transfer in collisions of CO2 with a perfluorinated self-assembled monolayer (F-SAM) surface. Procedures for interfacing chemical dynamics and electronic structure theory computer codes are discussed. A Hessian-based predictor–corrector algorithm and high-accuracy Hessian updating algorithm, for enhancing the efficiency of direct dynamics simulations, are described. In these simulations, an ensemble of trajectories is calculated which represents the experimental and chemical system under study. Algorithms are described for selecting the appropriate initial conditions for bimolecular and unimolecular reactions, gas-surface collisions, and initializing trajectories at transition states and conical intersections. Illustrative direct dynamics simulations are presented for the Cl− + CH3I SN2 reaction, unimolecular decomposition of the epoxy resin constituent CH3NHCHCHCH3 versus temperature, collisions and reactions of N-protonated diglycine with a F-SAM surface that has a reactive head group, and the product energy partitioning for the post-transition state dynamics of C2H5F HF + C2H4 dissociation. © 2012 John Wiley & Sons, Ltd.

An analytic function for the ground-electronic-state HCO {yields} H + CO potential energy surface is presented. Both experimental and ab initio data are used to derive parameters for the potential energy surface. Bound-state eigenvalues are obtained, and the results show good agreement with reported experimental and theoretical values. The analytic function is quite flexible and can be easily adjusted to determine how specific potential energy surface properties affect state-specific unimolecular decomposition.

Calculations of phase space theory and molecular dynamics rate constants and product energy and angular momentum distributions for Al6 and Al13 dissociation are reviewed. The molecular dynamics simulations indicate the unimolecular dynamics of these clusters are ergodic and agree with the random lifetime assumption of Rice-Ramsperger-Kassel-Marcus (RRKM) theory. With anharmonicity effects properly included, orbiting transition state theory (OTS/PST) gives rate constants and product energy and angular momentum distributions in very good agreement with those obtained from the molecular dynamics simulations.

The decomposition rate of chemically activated ethyltrimethylgermane from the reaction 1CH2 + (CH3)4Ge, where 1CH2 was produced from diazomethane photolysis at 3660 A, is 8.6 × 105 sec−1. This result combined with RRKM theory and critical energy estimates yields an Arrhenius A factor of log[A (sec−1)/methyl] = 14.7 ± 0.8 for methyl rupture from germanium. Log A values for methyl rupture from carbon, silicon, and germanium linearly correlate with the vibrational-rotational entropies of the corresponding tetramethyls. Extrapolation predicts log[A (sec−1)/methyl] = 14.4 and 14.3 for methyl rupture from tin and lead, respectively.

The specific decomposition rates of chemically activated methylcyclobutane produced from CH2(1A1) reaction with cyclobutane have been determined. CH2(1A1)was produced from ketene photolyses at 3340 and 3130 A and from diazomethane photolyses at 4358 and 3660 A. Comparisons of the excitation energies of the methylcyclobutane, determined by RRKM theory calculations, and the experimental results for the ketene systems, with thermochemically predicted maximum excitation energies, favor an Arrhenius A factor in the range of 5 × 1015 to 1 × 1016 sec−1 for methylcyclobutane. This result is consistent with (1) the comparison of RRKM theory calculations and the experimental unimolecular falloff for methylcyclobutane, (2) the comparison of experimental A factors for cyclobutane and other alkylcyclobutane decompositions, and (3) two out of three reported experimental A factors for methylcyclobutane. An analysis of these and previous results leads to a value of the CH2(1A1) ↔ CH2(3B1) energy splitting of 9±3 kcal/mole.

The classical trajectory method is used to investigate the unimolecular dynamics of ethyl radical dissociation. It is found that chaotic trajectories need not be backward integrable to yield accurate lifetime, and product energy and angular momenta distributions. This allows the use of large numerical integration step sizes in trajectory calculations. The product energy and angular momenta distributions are independent of the ethyl radical lifetime, and are obtained after only 50 dissociation events. Differences between classical and quantal unimolecular dynamics are discussed, and a prognosis for future trajectory studies of large-molecule unimolecular decompositions is given.

The total decomposition rates of the chemically activated alkanes n-butane, n-pentane, isopentane, and neohexane were measured using an internal comparison technique. Chemical activation was by the CH insertion reaction of excited singlet-state methylene radicals. A total of ten rate constants ranging from 4.6 × 105 to 2.3 × 107 sec−1 were measured for these alkanes at different excitation energies. These rates correlate via RRKM theory calculations with thermal A-factors in the range of 1016.1 to 1017.1 sec−1 for free rotoractivated complex models and in the range of 1016.4 to 1017.8 sec−1 for vibrator-activated complex models. It was found that high critical energies for decomposition, “tight” radical models, and activated complex models with free internal rotations were required to correlate the decomposition rates of these alkanes with estimated alkyl radical recombination rates. The correlation is just barely possible even for these favorable extremes, indicating that there may be a basic discrepancy between the recombination rate and decomposition rate data for alkanes.

Chemically activated ethane, with an excitation energy of 114.9 ± 2 kcal/mole, was formed by reaction with methane of excited singlet methylene radicals produced by the 4358 A photolysis of diazomethane. A decomposition rate constant of (4.6 ± 1.2) × 109 sec−1 was measured for the chemically activated ethane. This result agrees, via RRKM theory, with most other chemically activated ethane data, and the result predicts, via RRKM and absolute rate theory for E0 = 85.8 kcal/mole, E* = 114.9 kcal/mole, and kE = 4.6 × 101 sec−1, a thermal A-factor at 600°K of 1016.6±0.2 sec−1, in approximate agreement with the more recent experimental values. Combining 2 kcal/mole uncertainties in E0 and E* with the uncertainty in our rate constant yields an A-factor range of 1016.6±0.7 sec−1. It is emphasized that this large uncertainty in the A-factor results from an improbable combination of uncertainty limits for the various parameters. These decomposition results predict, via absolute rate theory (with E0(recombination) = 0) and statistical thermodynamic equilibrium constants, methyl radical recombination rates at 25°C of between 4.4 × 108 to 3.1 × 109 l.-mole−1-sec−1, which are 60 to 8 times lower, respectively, than the apparently quite reliable experimental value. A value of E0(recombination) greater than zero offers no improvement, and a value less than zero would be quite unusual. Activated complexes consistent with the experimental recombination rate and E0(recombination) = 0 greatly overestimate the experimental chemical activation and high pressure thermal decomposition rate data. Absolute rate theory as it is applied here in a straightforward way has failed in this case, or a significant amount of internally consistent data are in serious error. Some corrections to our previous calculations for higher alkanes are discussed in Appendix II.

Monte Carlo rate constants and lifetime distributions for H–C≡C–H, H–C≡C–Cl, and Cl–C≡C–Cl at 200 kcal/mole were determined by classical trajectories. The Monte Carlo rate constants agree to within 30% of the RRKM prediction. Energy randomization is faster in Cl–C≡C–Cl than in H–C≡C–H and H–C≡C–Cl. An analysis of nonrandom lifetime distributions indicates that energy randomization is complete by 10−12 sec. for each of the molecules. Comparisons are made between our results and other unimolecular studies.

Quasiclassical trajectory calculations have been performed to study intramolecular vibrational energy redistribution (IVR) from CH overtone states in benzene. The rate and extent of this redistribution is sensitive to details of the potential energy surface. A particularly important potential energy surface property is attenuation of the HCC bending frequency upon CH stretch excitation. This property gives rise to non‐irreversible IVR when the CH stretch is highly excited. An analog to quantum beats is observed in phase averaged quasiclassical trajectories.