diffpy.mpdf: open-source software for magnetic pair distribution function analysis

Abstract

The open-source Python package diffpy.mpdf, part of the DiffPy suite for diffraction and pair distribution function analysis, provides a user-friendly approach for performing magnetic pair distribution function (mPDF) analysis. The package builds on existing libraries in the DiffPy suite to allow users to create models of magnetic structures and calculate corresponding one- and three-dimensional mPDF patterns. diffpy.mpdf can be used to perform fits to mPDF data either in isolation or in combination with atomic pair distribution function data for joint refinement of the atomic and magnetic structure. Examples are given using MnO and MnTe as representative antiferromagnetic compounds and MnSb as a representative ferromagnet.

Spinteract: a program to refine magnetic interactions to diffuse scattering data

“…Computational software packages, such as the ALPS libraries [26], are typically optimised for direct simulations of pre-defined interaction models, rather than fitting models to experimental data. Recently, programs to analyse magnetic diffuse-scattering data have been developed that fit local spin arrangements directly to experimental data, using either 'big box' methods such as reverse Monte Carlo refinement [27][28][29][30][31][32][33][34], or 'small box' methods such as magnetic pairdistribution function analysis [35][36][37][38][39]. These approaches aim to describe the correlations between spin pairs using methods familiar from crystal-structure refinement, and have provided important insights into the physics of disordered magnetic states, from magnetic nanoparticles [37] to emergent partial magnetic ordering [40].…”

Spinteract: A Program to Refine Magnetic Interactions to Diffuse Scattering Data

“…Traditionally, diffuse-scattering analysis approaches were developed specifically for the problem at hand, because general-purpose software was not available. Recently, programs to analyse magnetic diffuse-scattering data have been developed that fit local spin arrangements directly to experimental data, using either "big box" methods such as reverse Monte Carlo refinement [26][27][28][29][30][31][32][33], or "small box" methods such as magnetic pair-distribution function analysis [34][35][36][37][38]. These approaches aim to describe the correlations between spin pairs using methods familiar from crystalstructure refinement, and have provided important insights into the physics of disordered magnetic states, from magnetic nanoparticles [36] to emergent partial magnetic ordering [39].…”