Premium
Reverse Monte Carlo refinements of nanoscale atomic correlations using powder and single‐crystal diffraction data
Author(s) -
Krayzman Victor,
Levin Igor
Publication year - 2012
Publication title -
journal of applied crystallography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.429
H-Index - 162
ISSN - 1600-5767
DOI - 10.1107/s0021889812000398
Subject(s) - reverse monte carlo , neutron diffraction , scattering , diffraction , materials science , single crystal , bragg's law , neutron scattering , powder diffraction , monte carlo method , extended x ray absorption fine structure , computational physics , crystallography , chemistry , optics , physics , absorption spectroscopy , mathematics , statistics
Combined refinements of local atomic structure that involve simultaneous fitting of powder‐averaged and single‐crystal data were implemented as an extension to the publicly available RMCProfile software. The refinements employ the reverse Monte Carlo method to fit neutron total scattering data, the neutron Bragg profile, extended X‐ray absorption fine structure (EXAFS) and structured diffuse scattering in electron diffraction. The procedure was tested using simulated data generated for a realistic model of perovskite‐like KNbO 3 , which exhibits strongly correlated Nb and O displacements. The results indicated that fitting the powder data ( i.e. total scattering or total scattering plus EXAFS) alone fails to reproduce the nanoscale range of displacement correlations. Simultaneous fitting of the powder data and the patterns of diffuse intensity in single‐crystal electron diffraction patterns returns significantly more accurate correlation characteristics. The approach and the computer software described in this contribution are not limited to electron diffraction patterns as other types of single‐crystal diffraction data ( i.e. X‐ray or neutron) can be included in the fit.