Efficient Calculation of Statistical and Dynamical Reaction Rates for Large Dimensional Molecular Systems


Knowledge of how quickly chemical reactions occur is an essential ingredient in the rigorous modelling of combustion, industrial and atmospheric reaction systems. This project focuses on the development of new methods for the accurate prediction of rate constants for chemical reaction. Crucial to this development is a more complete understanding of how and on what timescale energy moves about a molecule (IVR) and between molecules (collisional energy transfer).

The current focus is in two areas. Firstly, on the role of the intermolecular potential energy and quantum effects on collisional energy transfer in highly excited molecules. Secondly, on the modeling of complex chemical reactions in the gas phase, which involves describing the molecular species and their transformation by reaction at a detailed level.


Principal Investigator

Harold Schranz
Physical and Theoretical Chemistry
RSC
ANU

Project

s10

Facilities Used

VPP, PC, MDSS

Co-Investigator

Terry Frankcombe
Chemistry
Molecular & Microbial Sciences
University of Queensland

RFCD Codes

250101, 250601, 250602, 250603


Significant Achievements, Anticipated Outcomes and Future Work

A. Effect of Potential Energy on Collisional Energy Transfer: A quantitative measure of the average steepness of an intermolecular potential energy surface between a polyatomic molecule and a monotonic gas atom is proposed. Classical trajectory studies of collisions between rovibrationally excited carbon dioxide molecules and argon atoms selected from a thermal bath show that there is a strong correlation between this average steepness and the first moment of intermolecular energy transfer over a wide range of intermolecular potentials describing this CO2Ar interaction. The strong correlation, as illustrated, has the potential to allow insight into the effect that more subtle features of the intermolecular potential energy surface have on collisional energy transfer.

Figure 1 in pdf format

Fig. 1. Correlation of average energy transfer per collision with a intemolecular potential energy steepness parameter for highly excited CO2 with internal energy E=75 kcal/mol colliding with Ar at T=300K.

B. Prediction of Absolute Rate Coefficients and Product Branching Ratios: Complex chemical reactions in the gas phase can be decomposed into a network of elementary (e.g. unimolecular and bimolecular) steps which may involve multiple reactant channels, multiple intermediates, and multiple products. Here we focus on a detailed modeling of the C(3P)+Allene (C3H4) reaction, for which molecular beam experiments and theoretical calculations have previously been performed.

Figure 2 in pdf format

Fig. 2. The potential energy surface for C(3P)+Allene (C3H4) at a separation of 4 based on ab initio calculations generated with Gaussian 98.

We predict absolute unimolecular rate coefficients and branching ratios using microcanonical variational transition state theory (-VTST) with full energy and angular momentum resolution. Our calculation of the initial capture rate is facilitated by systematic ab initio potential energy surface calculations that describe the interaction potential between carbon and allene as a function of the angle of attack. Furthermore, the chemical kinetic scheme is enhanced to explicitly treat the entrance channels in terms of a predicted overall input flux and also to allow for the possibility of redissociation via the entrance channels. Thus, the computation of overall capture rates and partial capture rates is now possible.

Figure 3 in pdf format

Fig. 3. Temperature dependence of the capture rate constant k(T) predicted by this work (solid curve) compared to experimental results (points) for C(3P) + allene.

 

Computational Techniques Used

Quantum chemical ab initio packages (GAUSSIAN) and density functional theory packages (ADF) were used for calculating points for the intermolecular and intramolecular potential energy of the molecular systems under study.

For the collisional energy transfer studies, a model of the colliding system (e.g. an excited target molecule and a thermal projectile molecule) was described by a global potential energy surface including intra- and intermolecular parts, and the classical equations of motion were solved for a specified ensemble of initial conditions before the collision, thereby ending up with a set of final states after the collision is over. Home grown efficient vectorised trajectory codes were employed for this portion of the project and production runs were performed using the VPP and SGI Power Challenge supercomputers. A variety of detailed and averaged quantities were extracted from such studies.

For the calculation of overall capture rates and partial capture rates for complex chemical reactions, a statistical treatment was employed which involved a combination of Monte Carlo sampling over configuration space and convolution techniques.