[GSAS-II] old GSAS peak shape question

[Bell, Anthony]

Dear all

A few months ago I emailed about a problem that I have been having trying to use GSAS-II. The problem is that the structure which I am trying to refine is a silicate framework structure, there are four restraints for each tetrahedrally coordinated site but for some reason GSAS-II doesn't include all of them (see below).

So I used the old GSAS instead I have got a reasonable fit, with all my tetrahedrally coordinated site restraints in place. However, the 3 strongest peaks are not well fitted, there is clearly something not quite right with the peak shape, something is wrong with asymmetry of the peak shape.

I am using profile function 3 in the old GSAS, this uses S/L and H/L to define the asymmetry. I determined these from a Rietveld refinement of a silicon standard and then I have kept these silicon S/L and H/L terms fixed in the refinement of my silicate framework structure.

My problem is that my structure is low symmetry (P21/c monoclinic, with a b c all about 13 angstroms, beta about 93 degrees. My I11 data from the postal service has a wavelength of 0.8259 angstroms. Peak overlap is a big problem with such a short wavelength, which is the only wavelength that diamond can offer for their postal service. I am finding that I am getting a much better fit with my PANalytical X'Pert machine with Cu K-alpha than with one of the best synchrotron powder beamlines in the world!

Has anyone got any advice on how to better model the peak shape in the old GSAS, or how to get all of my restraints included into the new GSAS?

I am sure I could publish the structure from the X'Pert data, but as an old synchrotron scientist it would be good to include the synchrotron data in my paper.

Best wishes


Dr. Tony Bell
Senior X-ray Technician
Materials and Research Engineering Institute
Sheffield Hallam University
City Campus
Howard Street
Sheffield
S1 1WB

Anthony.Bell at shu.ac.uk<mailto:Anthony.Bell at shu.ac.uk>
0114 225 3401



From: gsas-ii-bounces at aps.anl.gov [mailto:gsas-ii-bounces at aps.anl.gov] On Behalf Of Bell, Anthony
Sent: 18 July 2016 15:49
To: 'gsas-ii at mailman.aps.anl.gov'
Subject: [GSAS-II] GSAS-II questions

Dear all

Thanks to the replies to my emails last week. I am still struggling up the learning curve with GSAS-II.

I have done two of the tutorials with GSAS-II so I am now returning to some of my own data, recently collected for me on I11 at diamond light source. I had done a reasonable refinement using laboratory XRD data here in Sheffield, I am now trying this refinement with synchrotron data to get a better refined crystal structure. My FULLPROF refinement of these data was reasonably good but not great, as I wasn't sure of some of the peak width terms to use for I11. I knew with GSAS I could include an instrument parameter file which would hopefully give me a better starting model for the peak width terms.

Two questions.

Is it possible to refine some peak asymmetry terms in the peak width parameters in GSAS-II?

My synchrotron data is on a synthetic mineral sample with tetrahedrally coordinated silicate framework structure. I wish to put in Si-O bond restraints. I know that there should be 4 Si-O bonds per tetrahedrally coordinates silicon, each bond should be about 1.61 angstroms in length. However, when I add restraints not all of the silicon atoms have 4 Si-O bond restraints calculated, some only have 3 or 2 Si-O bond restraints? Is there some way of manually adding restraints? My previous FULLPROF refinement had 4 Si-O bond length restraints per Si, is it possible to import restraints from a FULLPROF input file?

Thanks in advance

Dr. Tony Bell
Senior X-ray Technician
Materials and Research Engineering Institute
Sheffield Hallam University
City Campus
Howard Street
Sheffield
S1 1WB

Anthony.Bell at shu.ac.uk<mailto:Anthony.Bell at shu.ac.uk>
0114 225 3401

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20161010/b044b172/attachment.html>