Melting temperature prediction via first principles and deep learning
Melting temperature prediction via first principles and deep learning
复制标题
通过第一原理和深度学习预测熔化温度
DOI：
10.1016/j.commatsci.2022.111684
复制
发表时间：
2022
影响因子：
3.3
通讯作者：
Hong, Qi-Jun
中科院分区：
文献类型：
--
作者：
Hong, Qi-Jun
熔化是一个高温过程，需要对组态空间进行广泛的采样，这使得熔化温度预测的计算非常昂贵和具有挑战性。在过去的几年里，我已经建立了两种方法来应对这一挑战，一种是通过直接密度泛函理论(DFT)分子动力学(MD)模拟，另一种是通过深度学习图神经网络。DFT方法是基于对小尺寸固液共存分子动力学模拟的统计分析。它消除了快速加热方法中亚稳态过热固体的风险，同时相对于传统的大规模共存方法，也显著降低了计算机成本。它既准确又高效(以每种材料数天的速度计算)，被认为是直接计算DFT熔化温度的最佳方法之一。深度学习方法基于图神经网络，有效地处理了化学式中的排列不变性，大大提高了效率，降低了成本。在每种材质的毫秒速度下，该模型非常快，同时也具有中等精度，特别是在由数据集扩展的合成空间内。我已经将这两种方法都实现到自动计算机代码包中，使它们公开可用并可免费下载。DFT和深度学习方法具有很强的互补性，因此它们有可能很好地集成到熔化温度预测框架中。我展示了将这些方法应用于材料设计和高熔点材料发现的例子。
Melting is a high temperature process that requires extensive sampling of configuration space, thus making melting temperature prediction computationally very expensive and challenging. Over the past few years, I have built two methods to address this challenge, one via direct density functional theory (DFT) molecular dynamics (MD) simulations and the other via deep learning graph neural networks. The DFT approach is based on statistical analysis of small-size solid–liquid coexistence MD simulations. It eliminates the risk of metastable superheated solid in the fast-heating method, while also significantly reducing the computer cost relative to the traditional large-scale coexistence method. Being both accurate and efficient (at the speed of several days per material), it is considered as one of the best methods for direct DFT melting temperature calculation. The deep learning method is based on graph neural networks that effectively handles permutation invariance in chemical formula, which drastically improves efficiency and reduces cost. At the speed of milliseconds per material, the model is extremely fast, while being moderately accurate, especially within the composition space expanded by the dataset. I have implemented both methods into automated computer code packages, making them publicly available and free to download. The DFT and deep learning methods are highly complementary to each other, and hence they can be potentially well integrated into a framework for melting temperature prediction. I demonstrated examples of applying the methods to materials design and discovery of high-melting-point materials.
登录
查看更多内容
DOI：
10.1016/j.calphad.2015.12.003
发表时间：
2016-03-01
影响因子：
2.4
作者：
Hong, Qi-Jun;van de Walle, Axel
通讯作者：
van de Walle, Axel
DOI：
10.1016/j.calphad.2015.08.005
发表时间：
2015-12
影响因子：
2.4
作者：
L. Miljacic;S. Demers;Qi-Jun Hong;A. Walle
通讯作者：
L. Miljacic;S. Demers;Qi-Jun Hong;A. Walle
影响因子：
4.4
作者：
Alfe, D.;Cazorla, C.;Gillan, M. J.
通讯作者：
Gillan, M. J.
影响因子：
3.7
作者：
Qi-Jun Hong;A. van de Walle
通讯作者：
Qi-Jun Hong;A. van de Walle
影响因子：
4.4
作者：
Hong, Qi-Jun;van de Walle, Axel
通讯作者：
van de Walle, Axel