Atomic thermal fluctuation reduction method for robust local lattice structure identification in finite-temperature molecular dynamics

Author:

Hirano AtsuoORCID,Tsunemoto YosukeORCID,Takahashi AkiyukiORCID

Abstract

Abstract Classical molecular dynamics (MD) is extensively employed to explore the properties, deformations, and fractures of materials at the atomic scale. Identifying local structures is crucial for understanding the mechanisms behind material deformation and fracture. Nevertheless, analyzing the local lattice structure at high temperatures poses challenges due to atomic thermal fluctuations, which act as noise and potentially lead to misjudgment of the local lattice structure. To date, various strategies have been implemented to circumvent this issue. However, they cannot be a solution because it is unable to reproduce phenomena unique to high temperatures, whereas others require significant computational resources. This paper introduces an innovative method to reduce atomic thermal fluctuations using a straightforward algorithm, thereby facilitating accurate identification of local lattice structures even at high temperatures. Our approach incorporates novel degrees of freedom, termed ‘Markers,’ that are linked to atoms. By reducing the thermal fluctuation of these Markers, precise analysis of the local lattice structure becomes feasible. The efficacy of this method is validated through its thermal reducibility and Markers trackabilities to atoms. Utilizing common neighbor analysis, the error rate for structure identification with our method is nearly 0% at temperatures up to 1200 K in Fe, in contrast to approximately 5% without it. Furthermore, the average distance between atoms and Markers remains below 0.1 Å. Applying our method to phase transformations, we successfully observed the transition from face-centered cubic to body-centered cubic structure in Fe at 1200 K. This method holds promise for expanding the capabilities of MD simulations at high temperatures.

Publisher

IOP Publishing

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3