The recently proposed Atomistic Structure Learning Algorithm (ASLA) builds on neural network enabled image recognition and reinforcement learning. It enables fully autonomous structure determination when used in combination with a first-principles total energy calculator, e.g. a density functional theory (DFT) program. To save on the computational requirements, ASLA utilizes the DFT program in a single-point mode, i.e. without allowing for relaxation of the structural candidates according to the force information at the DFT level. In this work, we augment ASLA to establish a surrogate energy model concurrently with its structure search. This enables approximative but computationally cheap relaxation of the structural candidates before the single-point energy evaluation with the computationally expensive DFT program. We demonstrate a significantly increased performance of ASLA for building benzene while utilizing a surrogate energy landscape. Further we apply this model-enhanced ASLA in a thorough investigation of the c(4x8) phase of the Ag(111) surface oxide. ASLA successfully identifies a surface reconstruction which has previously only been guessed on the basis of scanning tunnelling microscopy images.
Neural operator accelerated atomistic to continuum concurrent multiscale simulations of viscoelasticity
National Center for Computational Sciences, Oak Ridge National Laboratory（橡树岭国家实验室计算科学国家中心）；Department of Engineering, University of Cambridge（剑桥大学工程系）
From their birth in the manufacturing process, materials inherently contain defects that affect the mechanical behavior across multiple length and time-scales, including vacancies, dislocations, voids and cracks. Understanding, modeling, and real-time simulation of the underlying stochastic micro-structure defect evolution is therefore vital towards multi-scale coupling and propagating numerous sources of uncertainty from atomistic to eventually aging continuum mechanics. We develop a graph-based surrogate model of dislocation glide for computation of dislocation mobility. We model an edge dis…
Assessment and optimization of the fast inertial relaxation engine (FIRE) for energy minimization in atomistic simulations and its implementation in LAMMPS
Georgia Institute of Technology（佐治亚理工学院）；National Institute of Standards and Technology（美国国家标准与技术研究院）；Theiss Research（泰斯研究公司）
Machine learning (ML) based models have greatly enhanced the traditional materials discovery and design pipeline. Specifically, in recent years, surrogate ML models for material property prediction have demonstrated success in predicting discrete scalar-valued target properties to within reasonable accuracy of their DFT-computed values. However, accurate prediction of spectral targets such as the electron Density of States (DOS) poses a much more challenging problem due to the complexity of the target, and the limited amount of available training data. In this study, we present an extension of…
Tuning of the hole spin relaxation time in single self-assembled In$_{1-x}$Ga$_x$As/GaAs quantum dots by electric field
采用原子经验赝势方法研究了电场对单个自组装In $_{1-x}$ Ga $_{x}$ As/GaAs量子点中声子辅助空穴自旋弛豫的调控.我们发现，沿生长方向的电场可以调谐空穴自旋弛豫时间超过一个数量级。电场可以延长或缩短空穴的自旋寿命，并且这种调谐在电场方向上表现出不对称性。点越高，不对称性越明显。结果表明，电场是调控量子点自组装空穴自旋弛豫的有效手段。
The annealing treatment in the advanced manufacturing process, e.g., laser-assisted manufacturing, determines the final state of glasses which is critical to its thermal, electrical and mechanical properties. Energy barriers analysis based on the potential energy surface offers an effective way to study the microscopic evolution of the inherent structures during the annealing process in a broadening timescale range, i.e., from atomistic timescale ( ~ ps) to experimental timescale (~ s). Here, we find the distribution activation energy barriers in the potential energy surface can be divided int…
Experimentally Constrained Molecular Relaxation: The Case of Glassy GeSe2
无序材料的理想原子模型不应与实验相矛盾，并且还应与精确的力场（{ \it ab initio}或经验）相一致。我们取得了显着的进展，共同满足{ \it都}这些标准，使用混合逆向蒙特卡罗方法结合近似第一原理分子动力学。我们通过研究复杂的二元玻璃材料g-GeSe $_2$来说明该方法。通过对模型进行约束，使其与部分结构因子相一致，并利用{ \it ab initio}模拟，得到了一个与实验非常吻合的647原子模型，其中包含了静态结构因子中的第一个尖锐衍射峰。计算了电子态密度，并与光电子能谱学进行了比较。该方法具有通用性和灵活性。
While deeply supercooled liquids exhibit divergent viscosity and increasingly heterogeneous dynamics as the temperature drops, their structure shows only seemingly marginal changes. Understanding the nature of relaxation processes in this dramatic slowdown is key for understanding the glass transition. Here, we show by atomistic simulations that the heterogeneous dynamics of glass-forming liquids strongly correlate with the local residual plastic strengths along soft directions computed in the initial inherent structures. The correlation increases with decreasing temperature and is maximum in …
