Correlation in the transition metal based Heusler compounds Co$_2$MnSi and Co$_2$FeSi
Hem Chandra Kandpal, Gerhard H. Fecher, and Claudia Felser, and Gerd, Sch\"onhense

TL;DR
This paper investigates the role of electronic correlations in Co-based Heusler compounds, showing that including on-site Coulomb interactions explains experimental magnetic moments and half-metallicity, which standard calculations fail to reproduce.
Contribution
It demonstrates that on-site electron correlation effects are crucial for accurately modeling magnetic properties of Co$_2$MnSi and Co$_2$FeSi Heusler compounds, resolving discrepancies between theory and experiment.
Findings
LDA calculations underestimate magnetic moments without lattice expansion.
LDA+U with U_eff between 2-5 eV reproduces experimental moments.
Strong correlations can destroy half-metallicity in these compounds.
Abstract
Half-metallic ferromagnets like the full Heusler compounds with formula XYZ are supposed to show an integer value of the spin magnetic moment. Calculations reveal in certain cases of X = Co based compounds non-integer values, in contrast to experiments. In order to explain deviations of the magnetic moment calculated for such compounds, the dependency of the electronic structure on the lattice parameter was studied theoretically. In local density approximation (LDA), the minimum total energy of CoFeSi is found for the experimental lattice parameter, but the calculated magnetic moment is about 12% too low. Half-metallic ferromagnetism and a magnetic moment equal to the experimental value of are found, however, only after increasing the lattice parameter by more than 6%. To overcome this discrepancy, the LDA scheme was used to respect on-site electron correlation in…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
