机器学习用于材料设计最具吸引力的方面之一是其对化学空间的高通量探索,但要达到机器学习辅助探索的上限,仅靠当前的模型架构和处理算法是不够的。近来,图神经网络(GNN)等新架构已获得大量研究投入。对于多相催化,定义基底分子内键以及吸附物/基底分子间键是一个耗时且具有挑战性的过程。在应用模型之前,必须考虑数据集预处理、节点/键描述符设计以及特定的模型约束。在本工作中,提出了一个旨在解决这些问题的框架,其形式为自动图表示算法(AGRA)工具,用于提取金属表面吸附位点的局部化学环境。该工具能够收集由不同体系组成的多个吸附几何数据集,并将它们组合到单一模型中。为展示 AGRA 相较于其他图表示方法的优异可迁移性和更低的计算成本,将其应用于5个不同的催化反应数据集,并与 Open Catalyst Projects(OCP)图表示方法进行基准比较。两个含 O/OH 吸附物的 ORR 数据集在合并后获得了0.053 eV RMSD,而三个含 CHO/CO/COOH 的 CO2RR 数据集获得了0.088 eV RMSD 的平均性能。为进一步展示该算法的多功能性和外推能力,在全部5个数据集的子集组合上训练了一个模型,其 RMSD 为0.105 eV。随后,该通用模型被用于预测广泛的吸附能以及一个全新的 ORR 催化剂体系,并通过密度泛函理论计算进行了验证。
英文摘要:
One of the most appealing aspects of machine learning for material design is its high throughput exploration of chemical spaces, but to reach the ceiling of ML-aided exploration, more than current model architectures and processing algorithms are required. New architectures such as Graph Neural Networks (GNNs) have seen significant research investments recently. For heterogeneous catalysis, defining substrate intramolecular bonds and adsorbate/substrate intermolecular bonds is a time-consuming and challenging process. Before applying a model, dataset pre-processing, node/bond descriptor design, and specific model constraints have to be considered. In this work, a framework designed to solve these issues is presented in the form of an automatic graph representation algorithm (AGRA) tool to extract the local chemical environment of metallic surface adsorption sites is presented. This tool is able to gather multiple adsorption geometry datasets composed of different systems and combine them into a single model. To show AGRA's excellent transferability and reduced computational cost compared to other graph representation methods, it was applied to 5 different catalytic reaction datasets and benchmarked against the Open Catalyst Projects (OCP) graph representation method. The two ORR datasets with O/OH adsorbates obtained 0.053 eV RMSD when combined together, whereas the three CO2RR datasets with CHO/CO/COOH obtained an average performance of 0.088 eV RMSD. To further display the algorithm's versatility and extrapolation ability, a model was trained on a subset combination of all 5 datasets with an RMSD of 0.105 eV. This universal model was then used to predict a wide range of adsorption energies and an entirely new ORR catalyst system and then verified through Density Functional Theory calculations
AdsMind: A Physics-Grounded Multi-Agent System for Self-Correcting Discovery of Adsorption Configurations on Heterogeneous Catalyst Surfaces
Department of Computer Science;Engineering, Hong Kong University of Science;Department of Chemistry, Hong Kong University of Science;Laboratory of Artificial Chemical Intelligence (LIAC), EPFL, Lausanne, Switzerland;Platform Laboratory for Science \& Technology, Asahi Kasei Corporation, Tokyo, Japan;IAS Center for AI for Scientific Discoveries, Hong Kong University of Science