Hu Qin, Weng Mouyi, Chen Xin, Li Shucheng, Pan Feng, Wang Lin-Wang
School of Advanced Materials , Peking University Shenzhen Graduate School , Shenzhen 518055 , China.
Materials Sciences Division , Lawrence Berkeley National Laboratory , Berkeley , California 94720 , United States.
J Phys Chem Lett. 2020 Feb 20;11(4):1364-1369. doi: 10.1021/acs.jpclett.9b03780. Epub 2020 Feb 4.
A method using machine learning (ML) is proposed to describe metal growth for simulations, which retains the accuracy of ab initio density functional theory (DFT) and results in a thousands-fold reduction in the computational time. This method is based on atomic energy decomposition from DFT calculations. Compared with other ML methods, our energy decomposition approach can yield much more information with the same DFT calculations. This approach is employed for the amorphous sodium system, where only 1000 DFT molecular dynamics images are enough for training an accurate model. The DFT and neural network potential (NNP) are compared for the dynamics to show that similar structural properties are generated. Finally, metal growth experiments from liquid to solid in a small and larger system are carried out to demonstrate the ability of using NNP to simulate the real growth process.
提出了一种使用机器学习（ML）的方法来描述用于模拟的金属生长，该方法保留了从头算密度泛函理论（DFT）的准确性，并使计算时间减少数千倍。此方法基于DFT计算的原子能量分解。与其他ML方法相比，我们的能量分解方法在相同的DFT计算下可以产生更多信息。该方法应用于非晶态钠系统，在该系统中，仅1000个DFT分子动力学图像就足以训练一个精确的模型。比较了DFT和神经网络势（NNP）的动力学，以表明产生了相似的结构特性。最后，在小型和大型系统中进行了从液体到固体的金属生长实验，以证明使用NNP模拟实际生长过程的能力。
