Feng Xilan, Gong Xiangrui, Liu Dapeng, Li Yang, Jiang Ying, Zhang Yu
Department of Automation Science and Electrical Engineering, Beihang University, Beijing, 100191, P. R. China.
Key Laboratory of Bio-inspired Smart Interfacial Science and Technology of Ministry of Education, School of Chemistry, Beihang University, Beijing, 100191, P. R. China.
Angew Chem Int Ed Engl. 2023 Nov 20;62(47):e202313068. doi: 10.1002/anie.202313068. Epub 2023 Oct 25.
Formula regulation of multi-component catalysts by manual search is undoubtedly a time-consuming task, which has severely impeded the development efficiency of high-performance catalysts. In this work, PtPd@CeZrO core-shell nanospheres, as a successful case study, is explicitly demonstrated how Bayesian optimization (BO) accelerates the discovery of methane combustion catalysts with the optimal formula ratio (the Pt/Pd mole ratio ranges from 1/2.33-1/9.09, and Ce/Zr from 1/0.22-1/0.35), which directly results in a lower conversion temperature (T approaching to 330 °C) than ones reported hitherto. Consequently, the best sample obtained could be efficiently developed after two rounds of iterations, containing only 18 experiments in all that is far less than the common human workload via the traditional trial-and-error search for optimal compositions. Further, this BO-based machine learning strategy can be straightforward extended to serve the autonomous discovery in multi-component material systems, for other desired properties, showing promising opportunities to practical applications in future.
通过人工搜索对多组分催化剂进行配方调控无疑是一项耗时的任务，这严重阻碍了高性能催化剂的开发效率。在这项工作中，作为一个成功的案例研究，明确展示了贝叶斯优化（BO）如何加速发现具有最佳配方比例（Pt/Pd摩尔比范围为1/2.33 - 1/9.09，Ce/Zr为1/0.22 - 1/0.35）的甲烷燃烧催化剂，这直接导致其转化温度（T接近330°C）低于迄今报道的催化剂。因此，经过两轮迭代就能高效开发出最佳样品，总共仅包含18次实验，这远远少于通过传统试错法搜索最佳组成所需的人力工作量。此外，这种基于BO的机器学习策略可以直接扩展到多组分材料系统中用于其他期望性能的自主发现，为未来的实际应用展现出广阔的前景。
