Che Hang, Lu Tian, Cai Shumin, Li Minjie, Lu Wencong
Department of Chemistry, College of Sciences, Shanghai University, Shanghai 200444, China.
Shanghai Shuzhiwei Information Technology Co., Ltd., 668 ShangDa Road, Shanghai 200444, China.
Materials (Basel). 2024 Jul 22;17(14):3614. doi: 10.3390/ma17143614.
Ternary gold alloys (TGAs) are highly regarded for their excellent electrical properties. Electrical resistivity is a crucial indicator for evaluating the electrical performance of TGAs. To explore new promising TGAs with lower resistivity, we developed a reverse design approach integrating machine learning techniques and proactive searching progress (PSP) method. Compared with other models, the support vector regression (SVR) was determined to be the most optimal model for resistivity prediction. The training and test sets yielded R values of 0.73 and 0.77, respectively. The model interpretation indicated that lower electrical resistivity was associated with the following conditions: a van der Waals Radius () of 0, a (another van der Waals Radius) of less than 217, and a mass attenuation coefficient of MoKα () greater than 77.5 cmg. Applying the PSP method, we successfully identified eight candidates whose resistivity was lower than that of the sample with the lowest resistivity in the dataset by more than 53-60%, e.g., AuCuPt and AuPtIn. Finally, the candidates were validated to possess low resistivity through the pattern recognition method.
三元金合金（TGAs）因其优异的电学性能而备受关注。电阻率是评估TGAs电学性能的关键指标。为了探索具有更低电阻率的新型有前景的TGAs，我们开发了一种将机器学习技术与主动搜索过程（PSP）方法相结合的逆向设计方法。与其他模型相比，支持向量回归（SVR）被确定为电阻率预测的最优模型。训练集和测试集的R值分别为0.73和0.77。模型解释表明，较低的电阻率与以下条件相关：范德华半径（）为0，（另一个范德华半径）小于217，以及钼Kα的质量衰减系数（）大于77.5 cmg。应用PSP方法，我们成功识别出八个候选物，其电阻率比数据集中电阻率最低的样品低53 - 60%以上，例如AuCuPt和AuPtIn。最后，通过模式识别方法验证了这些候选物具有低电阻率。
