Abstract
Layered intercalation compounds, in which atoms or molecules (intercalants) are intercalated into layered materials (hosts), are an attractive group of materials that have applications as active materials in batteries and exhibit unique physical properties such as superconductivity and topological electronic states, and have been the subject of various material investigations. However, there is no systematic understanding of which combinations of hosts and intercalants are stable, and it is not easy to design materials from a huge number of combinations. In order to clarify what factors determine the stability of the combinations, we performed system-atic first-principles calculations for materials with a structure in which a single atom is inserted into the host, changing the atom that serves as the intercalant. Then, the intercalation energy, which is the reaction energy between the host and the intercalant, was calculated to evaluate the stability of the interlayer compound. As a result, we found that the intercalation energy can be linearly regressed by a linear regression equation based on the hard and soft acids and bases principle by using two literature values, the standard free Gibbs energy of ion formation and the ion radius of the intercalant. This shows a new possibility that the concept of complex chemistry can be applied to the energetics of layered intercalation compounds and is expected to be applied to battery materials.
Supplementary materials
Title
Supporting Information
Description
The structures of layered intercalation compounds, the results of regression in other models, comparison of the vdW correction methods, the values of 〖〖∆G〗^°〗_(f,M^n ) and r_(M^n ) of intercalant, the conditions of random forest regression, and information about the .xlsx file of Supporting Information.
Actions
Title
Supporting Information database
Description
The value of Eint, Ef of all layered intercalation compounds, Adj. R2, α_(M"-" Host), β_(M"-" Host), and γ_(M"-" Host) of each layered intercalation compounds with same host and stacking, the value of descriptors for random forest regressions.
Actions