Origins of shortcomings in recent realistic multiband Monte-Carlo studies for GaMnAs

TL;DR

This paper critically examines recent Monte Carlo studies of GaMnAs, revealing significant inaccuracies in their magnetic property predictions due to flawed parameter choices and simulation limitations.

Contribution

It identifies key shortcomings in recent multiband Monte Carlo simulations of GaMnAs, emphasizing the need for more accurate parameter selection and modeling approaches.

Findings

Zeeman-splitting underestimates first-principles results

RKKY oscillations are strongly present in couplings

Ferromagnetic stability region is narrower than previously thought

Abstract

Magnetic properties of Mn-doped GaAs are re-investigated within a realistic multiband description of the host valence bands. We explicitely demonstrate that the recent Monte Carlo (MC) simulations performed on a large scale supercomputer suffer from severe shortcomings. Indeed, it is shown, using identical parameters that (i) the calculated Zeeman-splitting largely underestimates that obtained from first principle studies, (ii) the couplings exhibit strong RKKY oscillations, (iii) the stability region for ferromagnetism is much narrower than obtained previously and (iv) the calculated Curie temperatures appear to be at least one order of magnitudes smaller. We show that the proposed choice of physical parameters cannot describe the physics in (Ga,Mn)As.