James C. Ianni

A comparison of the Bader-Deuflhard and the Cash-Karp Runge-Kutta integrators for the GRI-MECH 3.0 model based on the chemical kinetics code Kintecus

The chemical kinetics and associated thermodynamics of the GRI-MECH 3.0 model is described and modeled with the Bader–Deuflhard and the Cash–Karp Runge–Kutta Integrators as implemented by W.T. Vetterling et al. in Kintecus V3.0. Comparisons are made to verify the mathematical correctness of the chemical kinetic model against those already published models and the experimental data in the literature. For the simple H2–O2 combustion model, the Cash–Karp Runge–Kutta integrator outperforms the Bader–Deuflhard method in speed and accuracy, but for combustions involving the GRI-MECH model the situation is reversed. The Bader–Deuflhard integrator is selected for the GRI-MECH runs and the selected published GRI-MECH runs in the literature. The comparisons of the Bader–Deuflhard integrator’s results and the published results from the GRI-MECH model and experimental results are in excellent agreement.

QMQSAR: Utilization of a semiempirical probe potential in a field‐based QSAR method

A semiempirical quantum mechanical approach is described for the creation of molecular field‐based QSAR models from a set of aligned ligand structures. Each ligand is characterized by a set of probe interaction energy (PIE) values computed at various grid points located near the surface of the ligand. Single‐point PM3 calculations afford these PIE values, which represents a pool of independent variables from which multilinear regression models of activity are built. The best n‐variable fit is determined by constructing an initial regression using standard forward stepwise selection, followed by refinement using a simulated annealing technique. The resulting fit provides an easily interpreted 3D physical model of ligand binding affinity. Validation against three literature datasets demonstrates the ability of the semiempirical potential to model critical binding interactions in diverse systems.