Quantum Chemistry by Random Walk

James B. Anderson教授获得了理论与计算化学项目的资助，致力于探索格林函数量子蒙特卡罗技术用于研究化学系统。安德森教授一直在使用大规模并行计算机，如CM-2连接机，对简单的三原子系统（如氢原子和氢分子之间的反应）进行格林函数蒙特卡罗计算。在他最近对完整的四体氢分子计算的计算中，他能够在65K处理器的CM-2计算机上实现1.5千兆次的持续速度，总体精度为0.006千卡/摩尔。安德森将使用相同的方法研究氢交换反应的反应表面，期望在整个势能表面上的精度优于0.01千卡/摩尔。他还期望研究氟原子与氢分子反应的势能面。其他要研究的系统包括半导体晶体中的激子，由小多原子分子组成的扩展分子系统，以及氦原子簇。在这个项目中进行的计算是很重要的，因为它们形成了一个精确的基准，其他电子结构理论模型可以据此进行测试。

Professor James B. Anderson is supported by a grant from the Theoretical and Computational Chemistry Program to explore Green's Function Quantum Monte Carlo techniques for studying chemical systems. Professor Anderson has been using massively parallel computers such as the CM-2 connection machine to perform Green's Function Monte Carlo calculations on simple triatomic systems such as the reaction between hydrogen atoms and hydrogen molecules. In his recent calculations on the complete 4-body hydrogen molecule calculation, he was able to achieve a sustained speed of 1.5 gigaflop on a 65K processor CM-2 computer, with an overall accuracy of 0.006 Kcal/mole. Anderson will study the reaction surfaces for the hydrogen exchange reaction using the same approach, with the expectation of accuracies better than 0.01 Kcal/mole over the entire potential energy surface. He also expects to study the potential energy surface for the flourine atom plus hydrogen molecule reaction. Other systems to be studied include excitons in semiconductor crystallites, extended molecular systems consisting of small polyatomic molecules, and clusters of helium atoms. %%% The calculations which will be done in this project are important because they form an exact benchmark against which other electronic structure theoretical models can be tested.