Re: [sdpd] X-Ray for measuring stresses

[Jonathan WRIGHT <wright...@esrf.fr>]

Dear Rogelio,

>1. An approach for determining the unstressed lattice
>spacing, considering that I am working with Ti6Al4V
>which has an hexagonal system (hcp).

Measure a "powder" sample, or carefully cut up your sample into little 
pieces, as the stress at a surface must be zero. There are other methods, 
but cutting up the sample must remove stresses and will leave strains (due 
to precipates for example) unchanged.

>2. An "effective" method for indexing the peaks of the
>I Vs. (2-theta) graphics of an hexagonal system, and
>if there is a software for indexing those systems.

Crysfire from www.ccp14.ac.uk collects together a suite of indexing 
programs for determining unknown unit cells, in all symmetries, but needs 
you to extract the peak positions first. Many people doing "strain 
scanning" seem to fit the individual peak positions without any unit cell 
constraint and go on to process them separately. In a single powder diagram 
the scattering vector for each peak corresponds to a different direction in 
the sample, so you'd expect the strain(=peakshift) for each peak to be 
different, so that's why the cell constraint is not used. So far as I know, 
no one has yet incorporated this into a multipattern Rietveld code for 
getting directly at the strain tensor. There are many different programs 
for peak fitting, each with their own strengths and weaknesses.

Having said all that, these are only the views of someone who has 
occasionally been involved with collecting this kind of data, and not 
analysing it. There is a new facility which has been set up on the ILL/ESRF 
site, specialising in just this field:
http://www.esrf.fr/news/FaME38/press-release.html
I believe you can contact them at FaME38...@esrf.fr, and I'm sure they will be 
glad to assist.

Also useful to you might be a copy of Rene Martins thesis, who has recently 
completed a PhD in this area.There is a copy online at:
http://www.esrf.fr/exp_facilities/ID11/handbook/staff/rene/rene.html

You might get more help on the Rietveld mailing list (rietveld_l...@ill.fr), 
compared to this one, which is mainly aimed at solving unknown crystal 
structures from powder data.

Best wishes,

Jon





 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/