A Study of the Adsorption Properties of Individual Atoms on the Graphene Surface: Density Functional Theory Calculations Assisted by Machine Learning Techniques.

In this research, the adsorption performance of individual atoms on the surface of monolayer graphene surface was systematically investigated using machine learning methods to accelerate density functional theory. The adsorption behaviors of over thirty different atoms on the graphene surface were computationally analyzed. The adsorption energy and distance were extracted as the research targets, and the basic information of atoms (such as atomic radius, ionic radius, etc.) were used as the feature values to establish the dataset. Through feature engineering selection, the corresponding input feature values for the input-output relationship were determined. By comparing different models on the dataset using five-fold cross-validation, the mathematical model that best fits the dataset was identified. The optimal model was further fine-tuned by adjusting of the best mathematical ML model. Subsequently, we verified the accuracy of the established machine learning model. Finally, the precision of the machine learning model forecasts was verified by the method of comparing and contrasting machine learning results with density functional theory. The results suggest that elements such as Zr, Ti, Sc, and Si possess some potential in controlling the interfacial reaction of graphene/aluminum composites. By using machine learning to accelerate first-principles calculations, we have further expanded our choice of research methods and accelerated the pace of studying element-graphene interactions.

Exploration of Solid Solutions and the Strengthening of Aluminum Substrates by Alloying Atoms: Machine Learning Accelerated Density Functional Theory Calculations.

In this paper, we studied the effects of a series of alloying atoms on the stability and micromechanical properties of aluminum alloy using a machine learning accelerated first-principles approach. In our preliminary work, high-throughput first-principles calculations were explored and the solution energy and theoretical stress of atomically doped aluminum substrates were extracted as basic data. By comparing five different algorithms, we found that the Catboost model had the lowest RMSE (0.24) and lowest MAPE (6.34), and this was used as the final prediction model to predict the solid solution strengthening of the aluminum matrix by the elements. Calculations show that alloying atoms such as K, Na, Y and Tl are difficult to dissolve in the aluminum matrix, whereas alloy atoms like Sc, Cu, B, Zr, Ni, Ti, Nb, V, Cr, Mn, Mo, and W exerted a strengthening influence. Theoretical studies on solid solutions and the strengthening effect of various alloy atoms in an aluminum matrix can offer theoretical guidance for the subsequent selection of suitable alloy elements. The theoretical investigation of alloy atoms in an aluminum matrix unveils the fundamental aspects of the solution strengthening effect, contributing significantly to the expedited development of new aluminum alloys.