Abstract:Catalytic-materials design requires predictive modeling of the interaction between catalyst and reactants. This is challenging due to the complexity and diversity of structure-property relationships across the chemical space. Here, we report a strategy for a rational design of catalytic materials using the artifcial intelligence approach (AI) subgroup discovery. We identify catalyst \textit{genes} (features) that correlate with mechanisms that trigger, facilitate, or hinder the activation of carbon dioxide (CO$_2$) towards a chemical conversion. The AI model is trained on frst-principles data for a broad family of oxides. We demonstrate that surfaces of experimentally identifed good catalysts consistently exhibit combinations of \textit{genes} resulting in a strong elongation of a C-O bond. The same combinations of \textit{genes} also minimize the OCO-angle, the previously proposed indicator of activation, albeit under the constraint that the Sabatier principle is satisfed. Based on these fndings, we propose a set of new promising catalyst materials for CO$_2$ conversion.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve a key challenge in the design of catalytic materials, that is, how to predict the interaction between the catalyst and the reactants, especially for semiconductor oxide catalysts for carbon dioxide (CO₂) activation and conversion. Specifically, researchers attempt to identify "catalyst genes" (features) that can trigger, promote or hinder CO₂ activation through artificial intelligence (AI) - driven methods. These genes are certain properties of the materials, and their combinations can significantly affect the chemical conversion of CO₂ molecules. #### Main problems: 1. **Complexity and diversity**: The structure - property relationships of catalytic materials are very complex and diverse in the chemical space, which makes it difficult to predict their performance. 2. **CO₂ activation mechanism**: Understand which material properties determine the activation of CO₂ molecules and how these properties work together to achieve the best catalytic performance. 3. **Rational design of catalysts**: By identifying and optimizing these "catalyst genes", find new and potential catalyst materials, thereby accelerating the chemical conversion process of CO₂. #### Solutions: Researchers combined first - principles calculations and an artificial intelligence method of Subgroup Discovery (SGD) to identify material properties that can optimize CO₂ activation indicators. They pay special attention to oxide materials because such materials have structural and compositional stability at actual temperatures and are relatively low - cost. #### Specific goals: - **Identify catalyst genes**: Find out material properties related to the CO₂ activation mechanism (such as C - O bond length, OCO angle, etc.), and determine how these properties affect catalytic performance. - **Meet the Sabatier principle**: Ensure that the adsorption energy is neither too strong nor too weak, so as to achieve the best catalytic effect. - **Propose new catalyst materials**: Based on the discovered catalyst genes, propose a series of new and potential catalyst materials for CO₂ conversion. Through this method, researchers hope to provide a clear design path for CO₂ conversion and promote the further development of related fields.

Artifcial-intelligence-driven discovery of catalyst \textit{genes} with application to CO2 activation on semiconductor oxides

Materials Genes of CO2 Hydrogenation on Supported Cobalt Catalysts: an AI Approach Integrating Theoretical and Experimental Data

Materials Genes of CO 2 Hydrogenation on Supported Cobalt Catalysts: An Artificial Intelligence Approach Integrating Theoretical and Experimental Data

Rules Describing CO2 Activation on Single-Atom Alloys from DFT-meta-GGA Calculations and Artificial Intelligence

Materials genes of heterogeneous catalysis from clean experiments and artificial intelligence

Machine-learning-accelerated Discovery of Single-Atom Catalysts Based on Bidirectional Activation Mechanism

Data-Centric Heterogeneous Catalysis: Identifying Rules and Materials Genes of Alkane Selective Oxidation

Machine-Learning-Driven High-Entropy Alloy Catalyst Discovery to Circumvent the Scaling Relation for CO2 Reduction Reaction

Carbon-Mediated Catalysis: Oxidative Dehydrogenation on Graphitic Carbon

Artificial Intelligence Pathway Search to Resolve Catalytic Glycerol Hydrogenolysis Selectivity

Identifying Key Descriptors for the Single-Atom Catalyzed CO Oxidation

An Artificial Intelligence (AI) workflow for catalyst design and optimization

Catalyst design using actively learned machine with non-ab initio input features towards CO2 reduction reactions

Single-Atom Alloy Catalysts Designed by First-Principles Calculations and Artificial Intelligence

Data-driven discovery of electrocatalysts for CO2 reduction using active motifs-based machine learning

CO Oxidation Catalyzed by Perovskites: The Role of Crystallographic Distortions Highlighted by Systematic Experiments and AI

Data‐driven Design of Enhanced In‐based Catalyst for CO2 to Methanol Reaction

Revisiting Electrocatalyst Design by a Knowledge Graph of Cu-Based Catalysts for CO 2 Reduction

Data-driven design of new catalytic materials in methane oxidation based on a site isolation concept

Accelerated prediction of Cu-based single-atom alloy catalysts for CO2 reduction by machine learning

Effective Screening Descriptors of Metal-Organic Framework-Supported Single-Atom Catalysts for Electrochemical CO2 Reduction Reactions: A Computational Study