Rietveld Refinement

Things you will need

• Data
  • Collect your data in constant theta steps.
  • Convert the spec data file into a GSAS input file.
  • See GSAS notes on the group webpage.
• Instrument Parameter File
  • You should at least be familiar with the format of this file,so read page 177 of the gsas manual.
  • There are a bunch of these in CMP_TOOLS_ROOT:[GSAS.AXP.PARM] on DIRAC,so you can take one and edit it.
  • Check sure that the wavelength and target parameters are correct.
  • There are two files in [xxxx] which you can use for data taken on the Huber with a Cu. tube and a Mo. tube.
• Standard Sample
  • If the alignmemt of the diffractometer has been changed at all you need to run a scan on a standard sample and perform a Reitveld Refinement on it
• Lattice and Symmetry Information
  • Wyckoff,(3 vol's),and the International Tables for Crystallography,(vol's A&C in lab),have most of what you'll need.

  Refining Your Data

  Follow the steps as in the Nickel Example keeping the following in mind.

  Run EXPEDT and goto Least Squares Refinement Setup,(LSR), and edit Least Squares Controls,(LSC),setting maximum number of cycles to 10.

  Switch on the following flags,in the following order,always running POWPREF and GENLES each time.

  • ZERO in Diffractometer Constants,(DC),in Overall Parameters,(OP).
  • Increase the number of background coefficients,(in OP),by one,until you get a decent fit to the background.Choose the background function wisely,read about the options on page 119.
  • POLA,the polarization flag,in DC.
  • GW,in Profile Coefficients,(PC).Choose profile function #2,(see footnote).
  • Lattice Parameters.
  • UISO,the thermal factor in atom parameters,(AP).(se footnote)
  • GW in PC,OP.
  • X,atom position,in AP.
  • asym in PC,OP.

  notes

  1. If the diffractometer has been realligned,then refine data from a standard.There is a Lanthanium Hexaboride sample in the lab.Use the values obtained for ZERO and KRATIO thus obtained when doing further refinements.
  2. Leave IPOL = 1.There are 3 types of Polarization,corrosponding to IPLO = 0,1,2.See pages 125 and 177.
  3. Do not accept the default value for UISO when entering the atom data,it's too high.Try UISO = 0.015.
  4. Do not refine lattice parameters for the standard sample,(they're well known,that's why it's been chosen as a standard).
  5. If you need to change the symmetry information,do so in "expedt data setup" menu,choosing "powder data preperation",then "Phases - lattice & sp. group".