作者: Dyutiman Das , Richard M. Martin , Malvin H. Kalos
DOI: 10.1103/PHYSREVE.73.046702
关键词: Monte Carlo molecular modeling 、 Physics 、 Hybrid Monte Carlo 、 Monte Carlo integration 、 Quasi-Monte Carlo method 、 Monte Carlo method in statistical physics 、 Dynamic Monte Carlo method 、 Quantum Monte Carlo 、 Statistical physics 、 Monte Carlo method
摘要: The quantum Monte Carlo (QMC) technique is an extremely powerful method to treat many-body systems. Usually the has been applied in cases where interaction potential a simple analytic form, like 1/r Coulomb potential. However, complicated environment as semiconductor heterostructure, evaluation of itself becomes nontrivial problem. Obtaining from any grid-based finite-difference for every walker and step infeasible. We demonstrate alternative approach solving Poisson equation by classical calculation within overall scheme. have developed modified "walk on spheres" algorithm using Green's function techniques, which can efficiently account energy configurations, typical algorithms. This stochastically obtained be easily incorporated with variational, diffusion, other techniques. validity this studying problem, polarization helium atom electric field infinite capacitor.