[GSAS-II] Clarification Request (MuStrain, Size, LGmix)

[Toby, Brian H.]

For the mailing list, Bob’s answer is below. Just to make clear, LGmix determines the relative proportions of Gaussian vs. Lorentzian (Cauchy) broadening from size and microstrain, where 0.0 gives 100% Gaussian broadening and 1.0 is 100% Cauchy. There is a term for each type of broadening. Values of LGmix less than 1 are uncommon.

While on the subject, the values used in GSAS-II for size broadening are those of the average crystallite size, with units of micrometers (microns), and the microstrain values are expressed in terms of the amount of lattice spread, as delta-Q/Q x 10^6 (or equivalently, delta-d/d x 10^6) and thus have no units.

Brian

Dear Audrey,
Far and away the most common powder peak broadening by the sample is Lorentzian mustrain especially for samples prepared by grinding (to reduce “grain” size). Size broadening is also almost invariably Lorentzian. In order for either to be Gaussian, they must have very narrow distributions. This is rarely achieved except in specially prepared materials. The LGmix parameter allows mixing of Lorentzian & Gaussian broadening. The default (LGmix=1) is for pure Lorentzian. NB: it takes very carefully prepared & collected high resolution powder data to determine LGmix.
For a simulation use LGmix=1, mustrain=1000 & size=1. That should suffice.
Bob


On Mar 5, 2021, at 8:03 AM, Audrey Mayrat <audrey.mayrat at etu.unilim.fr<mailto:audrey.mayrat at etu.unilim.fr>> wrote:

Ok, thank you very much. It answered  my questions.

However, I have another question about the broadening type used (Gaussian or Lauretzian).
We  already have all the instrument parameters defined (with the background). Though, how is defined  the broadening type parameter (LGmix) by default ?

When set by the user in setSampleProfile, LGmix is a float between 0 and 1 (Gaussian and Lorentzian) but if the value is not defined this parameter is ignored. Thus, how is it set naturally ? Is it a constant, peak dependant, broadening dependant (size or mustrain) or maybe phase dependant. What define the default values for the broadening type in a simulation ?

Thank you very much

Best Regards

AM

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20210305/d1a13e41/attachment-0001.html>