Simulating Phase Equilibria using Wang-Landau-Transition Matrix Monte Carlo

7th Brazilian Meeting on Simulational Physics

Maerzke, Katie A.^1 ; Gai, Lili^1 ; Cummings, Peter T.^1,2 ; McCabe, Clare^1,3

Department of Chemical and Biomolecular Engineering, Vanderbilt University, Nashville, TN 37235, United States^1

Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States^2

Department of Chemistry, Vanderbilt University, Nashville, TN 37235, United States^3

关键词: Convergence properties;    Density of state;    Initial estimate;    Molecular systems;    Monte carlo algorithms;    Transition matrices;    Transition matrix monte carlo;    Wang-Landau algorithm;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/487/1/012002/pdf
DOI  :  10.1088/1742-6596/487/1/012002

来源: IOP

【 摘 要 】

We will examine the strengths and weaknesses of the Wang-Landau and transition matrix Monte Carlo methods for simulating phase equilibria of continuous molecular systems alone and as a combined Wang-Landautransition matrix Monte Carlo algorithm. Although a combined Wang-Landau transition matrix Monte Carlo algorithm has been previously reported in the literature, the details of the method and a discussion of its performance for phase equilibria simulations has not been presented. The hybrid method combines the rapid initial estimate of the density of states from the Wang-Landau algorithm with the continual improvement in convergence of transition matrix Monte Carlo. The hybrid Wang-Landau-transition matrix (WL-TM) algorithm is found to be more efficient and has much better convergence properties than the Wang-Landau algorithm and is more robust than the transition matrix algorithm, enabling the simulations to reach relatively low reduced temperatures with ease.