Grant for Exploratory Research: Monte Carlo-Gaussian Importance Sampling Evaluation of Real-Time Feynman Path Integrals

化学系的理论和计算化学项目正在支持爱荷华州立大学的大卫·霍夫曼教授关于量子动力学计算的新方法的研究。这是一个与休斯顿大学的唐纳德·库里教授合作的项目，他的努力也得到了该司的支持。用Feynman路径积分方法取代了传统的计算矩阵乘积、特征值和逆的方法，这些计算随着系统规模的增加而迅速扩大。高斯重要性抽样用于在蒙特卡罗积分中产生可管理的方差。其目标是使对比以前可能的更大尺寸的散射系统进行完全量子力学处理成为可能。过去，用于分析化学反应的精确的数值量子力学方法仅限于包含不超过三个原子的体系，其中至少一个必须是氢。该项目正在测试一种新的计算技术，该技术有望在更大的系统上进行此类计算，更广泛地说，可以用于研究其他量子力学碰撞过程。

The Theoretical and Computational Chemistry Program in the Chemistry Division is supporting research by Prof. David Hoffman at Iowa State University on a new approach to quantum dynamics calculations. This is a collaborative project with Professor Donald Kouri of the University of Houston whose effort is also being supported by the Division. The Feynman path integral method is used to replace the customary calculation of matrix products, eigenvalues, and inverses which scale very rapidly with system size. Gaussian importance sampling is used to produce a manageable variance in the Monte Carlo integral. The goal is to make possible fully quantum mechanical treatment of scattering systems of a larger size than previously possible. %%% Numerically exact quantum mechanical methods for analyzing chemical reactions have in the past been limited to systems containing no more than three atoms, at least one of which must be hydrogen. This project is testing a new computational technique which offers promise of permitting such calculations on larger systems and, more generally, can be used in studying other quantum mechanical collisional processes.