Machine Learning to Predict Quasicrystals from Chemical Compositions
Machine Learning to Predict Quasicrystals from Chemical Compositions
复制标题
DOI：
10.1002/adma.202102507
复制
发表时间：
2021-07-19
影响因子：
29.4
通讯作者：
Yoshida, Ryo
中科院分区：
文献类型：
--
作者：
Liu, Chang;Fujita, Erina;Yoshida, Ryo
准晶已经成为第三类固态材料，区别于周期性晶体和无定形固体，其具有长程有序，而没有周期性，表现出在大多数情况下不允许周期性晶体的旋转对称性。迄今为止，已经报道了一百多个稳定的准晶，导致了许多新的和令人兴奋的现象的发现。然而，近年来发现新准晶的速度有所下降，主要是由于缺乏合成新准晶的明确指导原则。在这里，它表明，新的准晶的发现可以通过一个简单的机器学习工作流程来加速。利用已知稳定准晶、近似晶体和普通晶体的化学组成列表，训练预测模型以解决三级分类任务，并评估其与三元铝系统的观察到的相图相比的可预测性。验证实验有力地支持了机器学习的上级预测能力，相位预测任务的总体预测精度达到约0.728。此外，分析的输入-输出关系黑盒到模型中，非平凡的经验方程解释的人类，描述稳定的准晶形成所需的条件被确定。
Quasicrystals have emerged as the third class of solid-state materials, distinguished from periodic crystals and amorphous solids, which have long-range order without periodicity exhibiting rotational symmetries that are disallowed for periodic crystals in most cases. To date, more than one hundred stable quasicrystals have been reported, leading to the discovery of many new and exciting phenomena. However, the pace of the discovery of new quasicrystals has lowered in recent years, largely owing to the lack of clear guiding principles for the synthesis of new quasicrystals. Here, it is shown that the discovery of new quasicrystals can be accelerated with a simple machine-learning workflow. With a list of the chemical compositions of known stable quasicrystals, approximant crystals, and ordinary crystals, a prediction model is trained to solve the three-class classification task and its predictability compared to the observed phase diagrams of ternary aluminum systems is evaluated. The validation experiments strongly support the superior predictive power of machine learning, with the overall prediction accuracy of the phase prediction task reaching approximate to 0.728. Furthermore, analyzing the input-output relationships black-boxed into the model, nontrivial empirical equations interpretable by humans that describe conditions necessary for stable quasicrystal formation are identified.