Reverse Monte Carlo Analysis of Powder Patterns

1996 ◽  
Vol 29 (3) ◽  
pp. 285-290 ◽  
Author(s):  
W. Montfrooij ◽  
R. L. McGreevy ◽  
R. Hadfield ◽  
N. H. Andersen

A Monte Carlo algorithm for the analysis of powder diffraction patterns is presented. One aim of this algorithm, which can be used as a supplement to the regular Rietveld refinement, is to provide a self-consistent determination of the thermal displacements of the atoms. This is achieved by modelling the total scattered intensity, comprising both the Bragg peak intensity and the diffuse contribution to the spectrum from the scattering density of an assembly of atoms. This assembly, which is constructed by the reverse Monte Carlo technique so as to yield a best fit with the data, is then used to calculate the average atomic thermal displacements. This allows for a refinement that, in principle, no longer requires angle-dependent background parameters, and that is well suited for dealing with highly anisotropic Debye–Waller factors and split atomic sites.

Langmuir ◽  
2018 ◽  
Vol 34 (50) ◽  
pp. 15403-15415 ◽  
Author(s):  
Marian Cors ◽  
Lars Wiehemeier ◽  
Yvonne Hertle ◽  
Artem Feoktystov ◽  
Fabrice Cousin ◽  
...  

2005 ◽  
Vol 17 (17) ◽  
pp. 2605-2616 ◽  
Author(s):  
G Opletal ◽  
T C Petersen ◽  
D G McCulloch ◽  
I K Snook ◽  
I Yarovsky

2020 ◽  
Vol 53 (6) ◽  
pp. 1509-1518
Author(s):  
Yuanpeng Zhang ◽  
Maksim Eremenko ◽  
Victor Krayzman ◽  
Matthew G. Tucker ◽  
Igor Levin

Reported here are the development and application of new capabilities in the RMCProfile software for structural refinements using the reverse Monte Carlo (RMC) method. An algorithm has been implemented to enable the use of arbitrary peak-shape functions in the modeling of Bragg diffraction patterns and instrumental resolution effects on total-scattering data. This capability eliminates the dependence of RMCProfile on preset functions, which are inadequate for data produced by some total-scattering instruments, e.g. NOMAD at the Spallation Neutron Source (SNS) at Oak Ridge, Tennessee, USA. The recently developed procedure for the instrument-resolution correction has been modified to improve its accuracy, which is critical for recovering nanoscale structure. The ability to measure fine details of local and nanoscale structures with high fidelity is required because such features are increasingly exploited in the design of materials with enhanced functional properties. The new methodology has been tested via RMC refinements of large-scale atomic configurations (distances up to 8 nm) for SrTiO3 using neutron total-scattering data collected on the Polaris and NOMAD time-of-flight powder diffractometers at the ISIS facility (Didcot, Oxfordshire, UK) and SNS, respectively. While the Polaris instrument is known to provide the high-quality data needed for RMC analysis, the similar and sound atomic configurations obtained from both instruments confirmed that the NOMAD data are also suitable for RMC refinements over a broad distance range.


Author(s):  
Th. Proffen ◽  
T. R. Welberry

AbstractAn improved method for the analysis of single crystal diffuse scattering using the Reverse Monte Carlo (RMC) simulation technique is presented. Previous RMC studies showed that with respect to the size of the model crystal used for the simulation there are two conflicting requirements. A large crystal size gives relatively noise-free calculated diffraction patterns, but it is found that the ‘fit’ that is obtained has been achieved by adjusting the large number of high-order correlations in the structure rather than the relatively few low-order correlations of interest. On the other hand a small crystal size necessarily gives a defect structure characterised by short-range correlations, but gives a diffraction pattern so noisy that it cannot meaningfully be fitted to the observed data. The procedure described here overcomes this problem by using a quite different way of calculating the diffuse scattering intensity. Rather than computing the Fourier transform of the complete model crystal, the intensity is taken as the average of scattering intensities of many small crystal volumes (lots) chosen at random. This produces high quality diffraction patterns and at the same time restricts the effect of the RMC refinement to correlation vectors no greater than the lot size.The viability of this modified RMC method compared to the ‘normal’ RMC simulation technique is investigated using as input the calculated diffuse intensity of simple two dimensional (2D) model examples with known disorder properties.


Sign in / Sign up

Export Citation Format

Share Document