I try and avoid this, but as the scenario Leo describes is identical to those 
that I deal with, I'll self-cite.  Whitfield et al, Solid State Ionics, v175 
(2005), 463-471 describes a relatively simple (R-3m) application using lab, 
neutron, off-edge and resonant patterns simultaneously.  A paper dealing with a 
more complex system is coming up in the proceedings of EPDIC10.
 
I used Topas for my analysis using the approach that Joubert described for 
setting up the constraints in GSAS, but it works equally well in other 
programs.  It can be a bit fiddly but it does work.  I wholeheartedly agree 
that knowing the chemical composition (or even just some ratios) helps alot.  I 
would also agree that this kind of analysis is not as easy as alot of people 
seem to think, and routine it most definitely isn't.  Correlations in the 
least-squares normal matrix can be a serious issue as the minimum can be quite 
shallow, and keeping an eye on -log(conditioning number) in GSAS/Topas should 
help (there's another self citation for the conditioning number but you can 
look that one up if interested).
 
Pam
 
 

________________________________

From: Leopoldo Suescun [mailto:[EMAIL PROTECTED]
Sent: Sun 25/02/2007 1:03 AM
To: [email protected]
Subject: Re: Occupancy Constraint...



Dear Geetika
This will not answer your question, buy maybe allows you change it.

As has been discussed several times in the Rietveld list, the only way
to refine the occupancy of x cations in certain site is to have x or
more independent determinations of the scattering power of that site.
In your case it would be extremely difficult (if not impossible) to get
four independent determinations (1 neutrons and 3 x-rays, 2 of them
close to the absorption edge of two of you cations, at least).

The result of one x-ray or neutron pattern is the average scattering
power of the site, no matter how many cations you have spread in the
same site. With only one average scattering power you can determine
only one compositional variable for the site. So if you have cations A
and B you can determine the partial occupancies of both only
considering that they add up to one, so you are determining just one
compositional variable with the constrain SOF(A)=1-SOF(B).

Additionally, if you have 4 cations in the same site they are probably
very similar in size, so also probably similar in scattering power, so
x-rays will not allow you to determine the relative occupancies with
accuracy.

If you still want to determine the partial occupancy of the four
cations in one site and don't have the four independent observations I
suggest you to do a good chemical analysis that allows you deduce the
site composition to a maximum of one variable.

Hope this helps.
Best regards
Leo



Dr. Leopoldo Suescun
Argonne National Laboratory
Materials Science Division Bldg 223
9700 S. Cass Ave. Argonne IL 60439
Phone: 1 (630) 252-9760
Fax: 1 (630) 252-7777

----- Original Message -----
From: [EMAIL PROTECTED]
Date: Saturday, February 24, 2007 1:20 am
Subject: Occupancy Constraint...

> Dear All,
>
> I am using FULLPROF 2006 for the analysis of my system. I am
> having one
> problem in defining the constraint. Actually I have 4 cations in
> my system
> and want to put a constraint as w+x+y+z=1. Is it possible to
> define such
> constraints using FULLPROF.
> 
>
> --
> "A successful man is one who can lay a firm foundation with the
> bricks
> others have thrown at him"
>
> Regards,
> Geetika
> Research Scholar,
> C/o Prof. A. M. Umarji,
> Materials Research Centre,
> Indian Institute of Science,
> Bangalore,Karnataka.
> Pin Code-560012
>
>


<<winmail.dat>>

Reply via email to