Monte Carlo study of magnetic order at ferromagnetic and antiferromagnetic surfaces: Implications for spin-polarized photoelectron diffraction.

Abstract

We have used Monte Carlo simulations on simple-cubic Ising lattices with modified surface interaction parameters to model phenomenologically the temperature dependence of magnetic order near ferromagnetic and antiferromagnetic surfaces. These results are also discussed in connection with previous experiments suggesting surface-specific magnetic transition temperatures for semi-infinite systems, with special emphasis on spin-polarized photoelectron diffraction as a probe of short-range magnetic order. The calculated spin-spin correlation functions show no evidence of a high-temperature transition in short-range magnetic order. However, over a plausible range of choices for the surface interaction parameters, these correlation functions do show distinct surface transitions in long-range magnetic order that can be well above ${\mathit{T}}_{\mathit{N},\mathrm{b}\mathrm{u}\mathrm{l}\mathrm{k}}$ for antiferromagnets (both frustrated and nonfrustrated) and well above ${\mathit{T}}_{\mathit{C},\mathrm{b}\mathrm{u}\mathrm{l}\mathrm{k}}$ for ferromagnets. Thus, prior spin-polarized photoelectron data from antiferromagnetic ${\mathrm{KMnF}}_{3}$ and MnO may be explainable via such surface magnetic transitions, although further theoretical and experimental work are necessary to make this connection quantitative and definitive.