[GSAS-II] Can Parameters be restricted?

[Andrzej Falenty]

Dear Brian,

if I may to drop the comment. The enquiry is not entirely without merit. 
It is not infrequent that one combines information from many sources to 
build a model and constrains help to fix deficiency of a single method. 
I guess that one case where a hard limit could be useful is a case of 
vacancies in non-stoichiometric compounds. As you say it is easy to 
constrain a site shared by multiple elements if they add up to 1.  Yet, 
there has been always a problem, also for GSAS I, to deal with a sum of 
vacancy + element on a site. From this what I observed the refinement 
was trying to compensate vacancies with artificiality increased thermal 
displacement parameters if the data at higher 2-Theta were not "perfect" 
(actually only a case of good TOF with neutrons or an exceptional 
quality X-ray diffraction).  If I know that the occupancy cannot be 
larger than e.g. 1 for a site from other sources (e.g. spectroscopy or 
other x-ray neutron diffraction experiments) then it should be possible 
to set a hard limit on the site. Of course such decision must be well 
substantiated. Otherwise one risks producing some rubbish results...

Best wishes

Andrzej



Am 12.09.2016 um 11:42 schrieb Toby, Brian H.:
> No.
>
> To be honest I am not sure why one would want to do this. When a 
> parameter refines "out of range” it tells you something about your 
> data or model. With a restraint, you would never know that for 
> example, a Uiso value is refining negative by > 2 sigma or a occupancy 
> is >>1. It makes sense to me to define a model so that the sum of 
> occupancies is fixed — a well defined assumption. Defining a range for 
> a parameter scares me as I could never know that my model is highly 
> flawed is some way.
>
> Brian
>
>> On Sep 12, 2016, at 10:27 AM, Ivo Alxneit <ivo.alxneit at psi.ch 
>> <mailto:ivo.alxneit at psi.ch>> wrote:
>>
>> Dear all
>>
>> is there a way to restrict fitted parameters to a certain range? For
>> example restrict site occupancies to positive numbers or to a range such
>> as 0.9-1.0?
>>
>> thanks for any help
>> -- 
>> Dr. Ivo Alxneit
>> Catalysis for Energy Group
>> Bioenergy and Catalysis Laboratory        phone: +41 56 310 4092
>> Paul Scherrer Institute                     fax: +41 56 310 2688
>> CH-5232 Villigen                      gnupg key: 0x515E30C7
>> Switzerland
>> https://www.psi.ch/ceg/catalysis-for-energy-group/
>>
>> _______________________________________________
>> GSAS-II mailing list
>> GSAS-II at mailman.aps.anl.gov
>> https://mailman.aps.anl.gov/mailman/listinfo/gsas-ii
>
> ********************************************************************
> Brian H. Toby, Ph.D.                          office: 630-252-5488
> Advanced Photon Source Chief Computational Scientist
> Senior Physicist/Computational X-ray Science Group Leader
> 9700 S. Cass Ave, Bldg. 401/B4192       cell: 630-327-8426
> Argonne National Laboratory
> Argonne, IL 60439-4856        e-mail: brian dot toby at anl dot gov
> ********************************************************************
> "We will restore science to its rightful place, and wield technology's 
> wonders... We will harness the sun and the winds and the soil to fuel 
> our cars and run our factories...  All this we can do. All this we 
> will do."
>
>
>
> _______________________________________________
> GSAS-II mailing list
> GSAS-II at mailman.aps.anl.gov
> https://mailman.aps.anl.gov/mailman/listinfo/gsas-ii


-- 
--
Dr. Andrzej Falenty <afalent at uni-goettingen.de>
Abt. Kristallographie
GZG, Georg-August-Universität
Goldschmidtstraße 1
D-37077 Göttingen

Tel. 0049 551 3912546
http://kristall.uni-mki.gwdg.de/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20160913/b41aaad1/attachment.html>