Electroreduction of CO2 is one of the most potential ways to realize CO2 recycle and energy regeneration. The key to promoting this technology is the development of high-performance electrocatalysts. Generally, high-throughput computational screening contributes a lot to materials innovation, but still consumes much time and resource. To achieve efficient exploration of electrocatalysts for CO2 reduction, we created a machine learning model based on an extreme gradient boosting regression (XGBR) algorithm and simple features. Our screening model successfully and rapidly predicted the Gibbs free energy change of CO adsorption (ΔG CO) of 1060 atomically dispersed metal–nonmetal codoped graphene systems, and greatly reduced the research cost. The competitive reaction, the hydrogen evolution reaction (HER), is also discussed with respect to such a screening model. This work demonstrates the potential of machine learning methods and provides a convenient approach for the effective theoretical design of electrocatalysts for CO2 reduction.

Fe–N–C electrocatalysts have emerged as promising substitutes for Pt-based catalysts for the oxygen reduction reaction (ORR). However, their real catalytic active site is still under debate. The underlying roles of different types of coordinating N including pyridinic and pyrrolic N in catalytic performance require thorough clarification. In addition, how to understand the pH-dependent activity of Fe–N–C catalysts is another urgent issue. Herein, we comprehensively studied 13 different N-coordinated FeN x C configurations and their corresponding ORR activity through simulations which mimic the realistic electrocatalytic environment on the basis of constant-potential implicit solvent models. We demonstrate that coordinating pyrrolic N contributes to a higher activity than pyridinic N, and pyrrolic FeN4C exhibits the highest activity in acidic media. Meanwhile, the in situ active site transformation to *O-FeN4C and *OH-FeN4C clarifies the origin of the higher activity of Fe–N–C in alkaline media. These findings can provide indispensable guidelines for rational design of better durable Fe–N–C catalysts.

Due to the high cost and insufficient resource of lithium, sodium-ion batteries are widely investigated for large-scale applications. Typically, insertion-type materials possess better cyclic stability than alloy-type and conversion-type ones. Therefore, in this work, we proposed a facile and effective method to screen sodium-based layered materials based on Materials Project database as potential candidate insertion-type materials for sodium ion batteries. The obtained Na-based layered materials contains 38 kinds of space group, which reveals that the credibility of our screening approach would not be affected by the space group. Then, some important indexes of the representative materials, including the average voltage, volume change and sodium ion mobility, were further studied by means of density functional theory computations. Some materials with extremely low volume changes and Na diffusion barriers are promising candidates for sodium ion batteries. We believe that our classification algorithm could also be used to search for other alkali and multivalent ion-based layered materials, to accelerate the development of battery materials.

No abstract

Batteries based on multivalent ion (such as Al 3+ , Ca 2+ , and Mg 2+ ) intercalation materials have attracted extensive research interest due to their impressive capacity improvement and cost reduction compared with Li-ion batteries. However, the materials for state-of-the-art multivalent ion batteries still suffer from drawbacks such as sluggish ion mobility, poor rate performance, and low cyclic stability, bringing challenges for the design and investigation of new materials. Layered cathode materials are widely applied in current commercial batteries due to their outstanding ionic conductivity and structural stability, which may also hold the key for the cathodes of multivalent batteries. Therefore, combining database screening and density functional theory computations, we evaluated the layered compounds in Materials Project database by theoretical capacity, thermodynamic stability, experimental availability, voltage, volume variation, electronic conductivity, and ionic migration barrier and achieved over 20 kinds of layered cathode materials for multivalent batteries. Through Mg ion substitution for Ca sites, we further achieved several kinds of cathode materials for Mg-ion batteries with ideal stability, voltage, and ion diffusion barriers. We hope the methodology and screened materials could promote the development of multivalent ion batteries.