Ferromagnetism of transition metals and screened exchange interactions

TL;DR

This study uses first-principles calculations to analyze the magnetic properties of Fe, Co, and Ni, focusing on the effects of screened exchange interactions on finite-temperature magnetism.

Contribution

It evaluates the impact of screened exchange interactions on magnetic properties and Curie temperatures, revealing their varying effectiveness across different transition metals.

Findings

Screened J improves predictions for Co but not for Ni.

Calculated Curie temperatures for Fe and Ni show overestimation and good agreement respectively.

Effective Bohr magneton numbers match experimental values.

Abstract

We have investigated the magnetic properties of Fe, Co, and Ni at finite temperatures on the basis of the first-principles dynamical coherent potential approximation (CPA) in order to clarify the role of the exchange interaction energy ($J$) screened by {\it sp} electrons and its applicability to finite-temperature magnetism. With use of the atomic $J$, we obtained the Curie temperatures ($T_{\rm C}$) 1930 K for Fe and 2550 K for fcc Co, which are overestimated by a factor of 1.8 as compared with the experimental values, while we obtained $T_{\rm C}=620$ K for Ni being in good agreement with the experiment. Calculated effective Bohr magneton numbers also quantitatively agree with the experimental values. By comparing the results with those obtained by the screened $J$ and by comparing them with the experiments, we found that the screened $J$, which are reduced by 30% as compared with the atomic ones, improve the ground-state magnetizations and densities of states at low temperatures in Fe and fcc Co, as well as $T_{\rm C}$ in fcc Co. But the screened $J$ yield worse results for finite temperature properties of Fe and underestimate both the ground-state magnetization and $T_{\rm C}$ in case of Ni. We discuss possible origins for these inconsistencies.