Lithium adsorption on 2D transition metal dichalcogenides: towards a descriptor for machine learned materials design
Abstract
The adsorption properties of alkali metal atoms on two dimensional (2D) materials play an important role in the performance of batteries, catalysts, and sensors. In view of designing materials for such applications, we provide a comparative study of the adsorption of alkali metal atoms on different single-layer 2D transition metal dichalcogenides (TMDCs). Based on quantum mechanical calculations, we find that the adsorption energies of alkali metal atoms on 2D TMDCs are strongly and linearly correlated with the energy of the lowest unoccupied states of the materials. This relation is associated with the strong ionization of alkali metal atoms. We propose and demonstrate the ELUS of TMDCs as a descriptor for predicting adsorption energies. As long as the adsorption is controlled by charge transfer providing a phenomenological model for connecting the electronic details of a material to its interaction with other atomic/molecular species this predictor is expected to be transferable. Using a simple regression model, we could predict the adsorption properties of other alkali atoms on a variety of 2D TMDCs from three groups of the periodic table. Our results strongly support the use of the energy of the lowest unoccupied states as a novel efficient descriptor for data-driven materials design.