Re: [sdpd] size and strain analysis

[Armel Le Bail <alb...@cristal.org>]

>f i want to ?standardise?
>grain size and strain determination for
>non-cristallographer, what are the steps to do ¿?

First step : transform the non-crystallographers into
crystallographers. No other way.

>1-instrumental resolution
>
>the pseudo-voigt function used for resolution function
>determination on standard compounds do not allow to
>fit the asymetry observed at low angle. Does it matter
>really if i do not take into account this asymetry ¿?
>Is one side of the peak more significant than the
>other ¿?

There are many size/microstrain approaches. See the current
Size/Strain Round Robin at :
    http://www.boulder.nist.gov/div853/balzar/s-s_rr.htm

The best approach for size/strain is Fourier analysis.
With that method, you do not fit but you use the peaks
exactly like they are, asymmetry included. The problem
with this method is that you need at least 2 harmonics
for every family of reflections. This method contains
already a lot of approximations and assumptions, but
all the other methods contain more assumptions and
approximations up to various degrees. So that, you may
not take account of the peak asymmetry, however, be
aware that doing so, you place yourself into one of the
highest degree of assumptions and approximations.

>2- powder analysis
>
>how to subtract the resolution function from the
>fitted pseudo-voigt function since it is a mixture of
>gaussian and lorentzian.

See one approach detailed by Leonid Solovyov at the
Rietveld mailing list archives :
http://www.mail-archive.com/rietveld_l...@ill.fr/msg01544.html

Deconvolution is the mathematical term for it : obtaining
f from the h = f * g convolution product, knowing g (instrumental)
and h (experimental). However, the Rietveld approach uses
size/strain models for building f. The fit is made directly on
h by convoluting the f model and g.

>Does the williamson-hall method involves the FWHM or
>the integral width ¿?

None of them : it involves Integral breadth.
See for instance the J.I. Langford paper at APD-II :
NIST Special Publication 846 (1992) 110-126.

Armel Le Bail
http://www.cristal.org/course/


 

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