[LCRC Accounts] Yearly Allocation Request from HPCC
Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Amit Sharma Project Name: HPCC Division: CSE Project title: Ab-initio chemical kinetics study of high pressure combustion: A theoretical study of the radical-complex mechanism. Associated funding: Department of Energy Other Systems: Two small clusters (named: Pople and Linus) in the CSE fundamental interaction theory group. Science: Recent experimental studies of the falloff curve of the radical-radical association reaction show an unexpected additional rise of the rate constants at pressures above 300 bar in the 300 to 400 K temperature range. Contribution from the radical-complex (RC) mechanism, in addition to the usual energy-transfer (ET) mechanism, is believed to be the dominant factor for this phenomena. The radical-complex mechanism, which has not been theoretically studied and quantified, is the formation of the radical-bath gas (R-M) complex and the stabilization/complexation is related to the strength of the R-M bond in the van der Waals complexes. The decomposition of these R-M complex to stabilized reactants yields an increase in the net association rate. We propose to study the radical-complex mechanism using high level ab-initio kinetics approach and first principle electronic structure methods. The theoretical/computational approach involves development of a reduced dimensional, semi-global, potential energy surface (PES). This PES development is a novel approach due to high-dimensionality of the molecular system under study. For example, a simple oxidation reaction of CH3 with O2 in a monoatomic bath gas is a 15 dimensional system. The PES is developed by fitting high-level electronic structure energies for the radical-radical addition channel and R-M molecular system.The rate constant for the association reaction is calculated using variable reaction coordinate-transition state theory (VRC-TST) for different bath gas species. Project description: We extensively use single reference methods and multi reference electronic structure methods for evaluating the electronic structure energies for a given molecular configuration. In order to fit a full dimensional potential energy surface we need to compute thousands of such configuration spanning a multi-dimensional configuration space. These electronic structure methods are available in the software packages like Molpro, Gaussian, CFOUR, NWCHEM and similar quantum chemistry packages. The VRC-TST code for computing the sum of states at the transition states is developed within our group. This code is used for the calculation of reaction rates. The size of electronic structure calculations, typically, is about 40,000 to 50,000 electronic structure calculations for each system. The computational cost of each electronic calculation depend on the size of the basis set, electronic structure methods and other auxiliary routines such as integral calculations and i/o routines. This can vary anywhere between 0.5 hours to couple of hours for a single electronic structure calculation, for the system we propose to study. There will be 1 to 2 people working on this project who will be using the computational resources. About 30-40% of the allocation will use non-parallel computations. This year we would be making use of the MULTIMODE program developed in the group of Joel Bowman from Emory university. Multimode does Vibrational Self Consistent Field (VSCF) and several types of "CI" calculations for the rovibrational energies and wavefunctions of polyatomic molecules, transition states, clusters, etc. The code is based on the Watson Hamiltonian for non-linear molecules. The many-mode wavefunction is expanded in terms of the eigenstates (also known as the virtual states) of a given VSCF Hamiltonian. Usually the ground state VSCF Hamiltonian is used. This basis is orthonormal and results in a standard eigenvalue problem. The resulting eigenvalues and eigenfunctions are true variational upper bound approximations to the exact eigenvalues and eigenfunctions. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 220000 Q1: 55000 Q2: 55000 Q3: 55000 Q4: 55000 Justification: Thank You, The LCRC Accounts System
participants (1)
-
accounts@lcrc.anl.gov