[GSAS-II] instrument parameters

Wed Feb 28 11:16:35 CST 2018

Thank you to ALL who have replied,
I forward here the main answers, so that some future GSAS-II user browsing the mail archive can find some info and does not need to bother you again for the same things.

Instead, I will probably bother you again for different things,

Kind regards,

Arianna Lanza

From: Nick Weadock [mailto:nweadock at gmail.com]
Sent: martedì 27 febbraio 2018 19:39
To: gsas-ii at aps.anl.gov; Arianna Lanza <Arianna.Lanza at iit.it>
Subject: Re: GSAS-II Digest, Vol 162, Issue 1

Hi Arianna,

The instrument broadening parameters are unique to each instrument. There was a paper published that looks at a few common instruments (Kaduk, J. and Reid, J. Powder Diffraction 26 (1) 2011, 88-93) and reports U,V,W,X,Y, asym, S/L, H/L, etc. These may be a good guideline, but your instrument should really be calibrated with a SRM.

I had a similar result to you on my instrument when refining the instrument parameters using LaB6 SRM. I achieved better fits refining first X and Y, then UVW. I think if you get large positive or negative values for these it may indicate that there is not much broadening in the Caglioti type equation, and the parameters diverge in the fit. You could play around with some dummy instrument parameters and make simulated histograms to see the difference.

If you do not have a standard reference, you really can't deconvolute the instrument and sample effects.

As for the difference between old and new GSAS, I would take a look in the source code, it is well commented as to what models have been incorporated and their references.

Hope this helps,
Nick

From: Von Dreele, Robert B. [mailto:vondreele at aps.anl.gov]
Sent: martedì 27 febbraio 2018 17:40
To: Arianna Lanza <Arianna.Lanza at iit.it>
Cc: Toby, Brian H. <toby at anl.gov>
Subject: RE: instrument parameters

Dear Arianna,
‘Z’ is a constant Lorentzian broadening term (i.e. independent of 2-theta or TOF) recently added as it seems to be occasionally useful for some instrument characterization. As for the values, for conventional (especially neutron) diffractometers U & W are > 0 and V < 0 so the curve is a parabola with a minimum somewhere near 2-theta for the monochromator (see the garnet neutron tutorial example in GSAS-II). Lab x-ray diffractometers will be parabolic but usually fairly flat thus being a bit difficult to refine. Typical older ones have U~2, V~-2 and W~5. Newer ones with focusing optics will be different & smaller, however the Gaussian & Lorentzian resolution curves should be never negative.  GSAS-II will show the plots of the resolution curves for Gaussian, Lorentzian and their combination (pseudo-Voigt) as calculated and fitted – see Instrument Parameters. As Brian Toby noted, for GSAS-II the object is to calibrate your instrument with a standard (Si, LaB6, etc.) to get good values of U, V, W  (& X, Y – careful with these as sample broadening shows up in them which is NOT instrument broadening). Then never refine them again for the same instrument setup (same slits, etc.). Instead refine the sample broadening for each phase (see the Phase/Data tab for these) as mustrain & size.
Bob

From: Toby, Brian H. [mailto:toby at anl.gov]
Sent: martedì 27 febbraio 2018 17:29
To: Arianna Lanza <Arianna.Lanza at iit.it>
Cc: Von Dreele, Robert B. <vondreele at aps.anl.gov>
Subject: Re: [GSAS-II] instrument parameters

Arianna,

   We need to tabulate the equations we use for FWHM as a function of the profile variables somewhere (you could read the source code), but a quick answer is that you can see the resulting FWHM curves under the instrument parameters. Also, the new parameters (Z plus a bunch of TOF terms) are generally not needed for unless one is doing unusually careful fitting of an instrument.

    First of all, I am assuming that you are fitting a standard such as NIST LaB6 that has minimal broadening. If not, you should do that before starting with a “real” sample. You don’t say but I will also assume you are using a lab instrument. A good place to start is to fit individual peaks across the range of your pattern so you can see how the Lorentzian (gamma) and Gaussian (sigma) widths vary without constraints. You can also see how well your UVW & XY curves reproduce this.

Brian

On Tue, Feb 27, 2018 at 10:00 AM, <gsas-ii-request at aps.anl.gov<mailto:gsas-ii-request at aps.anl.gov>> wrote:
Send GSAS-II mailing list submissions to
        gsas-ii at mailman.aps.anl.gov<mailto:gsas-ii at mailman.aps.anl.gov>

To subscribe or unsubscribe via the World Wide Web, visit
        https://mailman.aps.anl.gov/mailman/listinfo/gsas-ii
or, via email, send a message with subject or body 'help' to
        gsas-ii-request at mailman.aps.anl.gov<mailto:gsas-ii-request at mailman.aps.anl.gov>

You can reach the person managing the list at
        gsas-ii-owner at mailman.aps.anl.gov<mailto:gsas-ii-owner at mailman.aps.anl.gov>

When replying, please edit your Subject line so it is more specific
than "Re: Contents of GSAS-II digest..."

Today's Topics:

   1. instrument parameters (Arianna Lanza)

----------------------------------------------------------------------

Message: 1
Date: Tue, 27 Feb 2018 16:03:36 +0000
From: Arianna Lanza <Arianna.Lanza at iit.it<mailto:Arianna.Lanza at iit.it>>
To: "gsas-ii at mailman.aps.anl.gov<mailto:gsas-ii at mailman.aps.anl.gov>" <gsas-ii at mailman.aps.anl.gov<mailto:gsas-ii at mailman.aps.anl.gov>>
Subject: [GSAS-II] instrument parameters
Message-ID: <a06a772caf334efaaa50f35772a46e32 at iit.it<mailto:a06a772caf334efaaa50f35772a46e32 at iit.it>>
Content-Type: text/plain; charset="iso-8859-1"

Dear all,

I am new to GSAS II and have no experience with old-GSAS, so forgive me for the basic level of the questions.

I don't fully understand the meaning of the parameters describing the instrumental profile.

?         What is Z? I could not find it in the old-GSAS manual.

?         How can I judge if the U V W X Y Z are reasonable?  (in case I don't have a reference profile from a standard) And how should a well behaved peak width plot look like?

?         I read that U W X Y are constrained to be >= 0, while V should be <= 0. However I have a not so badly-fitted plot where U refined to very negative, V to very positive, W is small and negative, X is positive and Y negative. So (i) those parameters have no constraint by default!Why not? Can I do it with equations?; (ii) even from my na?ve understanding it looks like a nonsense. Most likely it is because I have broad peaks and the interplay between strain/size/etc and instrumental parameters converged to this artificial mess. Is it possible to deconvolute the instrument contribution from the sample contribution, in absence of a standard?

?         SH/L is one unique parameter as opposed to 2 separate parameters in old-GSAS, what is its meaning now, the units and/or reasonable range?

If you can help with even partial answers or point out useful sources of information, I will be grateful!

Kind regards,

Arianna

,_,.-'???'-.,_,.-'???'-.,_,.-'???'-.,_,,.-'???'-.,_,.-'???'-.,_,

Arianna Lanza, PhD

Center for Nanotechnology Innovation at NEST
Istituto Italiano di Tecnologia
Piazza San Silvestro, 12, 56127, PISA, Italy

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20180227/7df88e57/attachment-0001.html>

------------------------------

_______________________________________________
GSAS-II mailing list
GSAS-II at mailman.aps.anl.gov<mailto:GSAS-II at mailman.aps.anl.gov>
https://mailman.aps.anl.gov/mailman/listinfo/gsas-ii

End of GSAS-II Digest, Vol 162, Issue 1
***************************************

--
Nick Weadock
PhD Candidate, Materials Science
California Institute of Technology
1200 E California Blvd, Pasadena, CA 91125
MC 138-78
x2520
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20180228/700c8e14/attachment-0001.html>