Uncovering the puzzle of complex magnetism in Fe16N2: a first-principles based study

TL;DR

This study uses first-principles calculations to explore the electronic structure and magnetic properties of Fe16N2, revealing insights into its high Curie temperature, magnetic moments, and potential as a spin injector for semiconductors.

Contribution

It provides a detailed first-principles analysis of Fe16N2's magnetism, including Curie temperature estimation and spin injection potential, which were not comprehensively studied before.

Findings

Curie temperature from MFA exceeds experimental values, highlighting thermal fluctuation effects.

Fe16N2 exhibits high spin polarization, making it suitable for spin injection.

Magnetic susceptibility shows sharp peak at TC, indicating local moment behavior.

Abstract

The electronic structure and magnetic exchange interactions in pure and V-doped Fe16N2 are studied within the framework of density functional theory. The Curie temperatures were obtained with both mean field approximation (MFA) as well as Monte Carlo (MC) calculations. The Curie temperature (TC) for pure Fe16N2 obtained within MFA are significantly larger than the experimental value, suggesting the importance of thermal fluctuations in these systems, and has a resemblance of a lower dimensional spin system. We also briefly discuss about the various possible factors which may lead to a large magnetic moment in this material. The calculated magnetic susceptibility at zero field shows sharp peak at T=TC which resemble a local moment system. From the nature of exchange interactions we try to figure out the nature of the Fesites which might contain localized d-states. Finally, we point out that Fe16N2 can also act as a good spin injector for the III-V semiconductors in addition to its well promised application as permanent magnet since it has a very high spin polarization (larger compared to elemental ferromagnets) as well as quite smaller lattice mismatch (compared to half-metallic Heusler alloys) with the conventional III-V semiconductors such as GaAs or InGaAs. We further demonstrate this through our calculations for Fe16N2(001)/InGaAs(001) heterostructures which shows the non-negligible spin polarization in the semiconductor (InGaAs) region implying a long spin diffusion length.