GSASIIRefineFitPeaks v1¶

Summary¶

Perform Rietveld or Pawley refinement of lattice parameters on a diffraction spectrum using GSAS-II scriptable API

Properties¶

Description¶

Uses GSAS-II [TobyVonDreele2013] as external software to fit peaks to a powder / engineering diffraction pattern. Here the process of peak fitting is in the context of Rietveld / Pawley / Le Bail analysis [LeBail2005]

The algorithm supports two refinement methods: Pawley refinement and Rietveld refinement. The use of this algorithm is very close to the examples described in these two GSAS-II tutorials: Rietveld fitting / CW Neutron Powder fit for Yttrium-Iron Garnet, and Getting started / Fitting individual peaks & autoindexing, The functionality of this algorithm is based on the powder calculation module and the structure routines of GSAS-II.

The refinement methods of this algorithm are equivalent to the function “Calculate / Refine” from the main menu of the GSAS-II GUI.

The main inputs required are histogram data, an instrument definition parameter (in GSAS format, readable by GSAS-II), phase information and various parameters for the fitting/refinement process.

The phase information must be provided in CIF format (Crystallographic Information File). When phase information is available the algorithm will output the lattice parameters in a table workspace. The values are given for the the full set of lattice parameters (three lattice constants, three angles, and volume in this sequence: a, b, c, alpha, beta, gamma, volume). The a,b, and c values are given in Angstroms (). The angles are given in degrees, and the volume in .

The algorithm provides goodness-of-fit estimates in the outputs GoF and Rwp or weighted profile R-factor [Toby2008]. The Rwp is given as a percentage value.

When Pawley refinement is selected as refinement method the flag for histogram scale factor refinement is disabled, as recommended in the GSAS-II documentation, as this cannot be refined simultaneously with the Pawley reflection intensities.

The GSAS-II Rietveld/Pawley refinement process writes lattice parameters and extensive additional information in an output file with the same name as the output GSAS-II project file but with extension ”.lst”. This is noted in a log message that specifies where the file has been written (next to the output project file).

Installing GSAS-II¶

On Windows, run either scripts\GSAS-II\install_gsas_latest.bat or scripts\GSAS-II\install_gsas_vetted.bat. Use the former for the latest version of GSAS-II and the latter for the most recent version to have been manually verified to work with GSASIIRefineFitPeaks. This will install GSAS-II to your current working drive (usually C) in a directory called g2conda.

On Linux, from scripts\GSAS-II run python install_gsas_proxy.py. Optionally, you can supply a desired revision number (defaults to the latest) with the -v flag and a directory (defaults to /) to install GSAS-II to with the -d flag.

References:

Usage¶

Example - Pawley refinement of lattice parameters from a diffraction spectrum

# You would normally generate the focused file using the Engg GUI or,
# alternatively, with commands like these:
#
# ws_to_focus = Load('ENGINX00256663-256675')
# wks_ceria = Load('ENGINX00255924')
# wks_vana = Load('ENGINX00254854')
# # Using default expected peaks for Ceria
# difa, difc, tzero, peaks_tbl = EnggCalibrate(InputWorkspace=wks_ceria, VanadiumWorkspace=wks_vana, Bank='North')
# wks_focused = EnggFocus(InputWorkspace=wks, VanadiumWorkspace=wks_vana, Bank='North')
# SaveNexus(InputWorkspace=wks_focused, Filename='focused_bank1_ENGINX00256663.nxs')
#
wks = Load('focused_bank1_ENGINX00256663.nxs')
peaks, lattice_params, rwp, sigma, gamma = \
    GSASIIRefineFitPeaks(InputWorkspace=wks,
                         RefinementMethod="Pawley refinement",
                         InstrumentFile='template_ENGINX_241391_236516_North_bank.prm',
                         PhaseInfoFiles='Fe-alpha.cif,Fe-gamma.cif',
                         PathToGSASII='/home/user/g2conda/GSASII',
                         RefineSigma=True,
                         RefineGamma=True,
                         SaveGSASIIProjectFile='example_gsas2_project.gpx',
                         OutputWorkspace="FittedPeaks")
print("Weighted profile R-factor (Rwp): {0:.5f}".format(rwp))
print("Lattice parameters, a: {a}, b: {b}, c: {c}, alpha: {alpha}, beta: {beta}, gamma: {gamma}, "
      "Volume: {volume:.3f}".format(**lattice_params.row(0)))
print("Sigma={}, Gamma={}".format(sigma, gamma))

Output:

Weighted profile R-factor (Rwp): 77.75515
Lattice parameters, a: 2.8665, b: 2.8665, c: 2.8665, alpha: 90.0, beta: 90.0, gamma: 90.0, Volume: 23.554
Sigma=81.0939, Gamma=0.1855

Example - Rietveld refinement of lattice parameters from a diffraction spectrum

wks=Load('focused_bank1_ENGINX00256663.nxs')
peaks, lattice_params, rwp, sigma, gamma = \
    GSASIIRefineFitPeaks(InputWorkspace=wks,
                         RefinementMethod='Rietveld refinement',
                         InstrumentFile='template_ENGINX_241391_236516_North_bank.prm',
                         PhaseInfoFiles='Fe-alpha.cif,Fe-gamma.cif',
                         PathToGSASII='/home/user/g2conda/GSASII',
                         SaveGSASIIProjectFile='example_gsas2_project.gpx',
print("Weighted profile R-factor (Rwp): {0:.5f}".format(rwp))
print("Lattice parameters, a: {a}, b: {b}, c: {c}, alpha: {alpha}, beta: {beta}, gamma: {gamma}, "
      "Volume: {volume:.3f}".format(**lattice_params.row(0)))
print("Sigma={}, Gamma={}".format(sigma, gamma))

Output:

Weighted profile R-factor (Rwp): 77.75499
Lattice parameters, a: 2.8665, b: 2.8665, c: 2.8665, alpha: 90.0, beta: 90.0, gamma: 90.0, Volume: 23.554
Sigma=81.0939, Gamma=0.1855

Categories: Algorithm Index | Diffraction\Engineering | Diffraction\Fitting

Source¶

Python: GSASIIRefineFitPeaks.py (last modified: 2018-10-05)