I would also like to see Rietveld data on ICDD cards. But I am going to put
my stick in to further stir things up.
If one has a set of Rietveld ICDD card data, one will have to massage it to
put it into a form suitable for use in the current standard search/match
methods. If we have to do that, why not go one step further and store L/p
and absorption corrected data? Looking at the diffraction line intensity
equation, |Fhkl|^2 is multiplied by L/p(theta), which includes both
monochromator & polarization corrections (polarization corrections are not
always required) and by A(theta), which varies by instrument/specimen
geometry. The result would be a value that is instrument independent, except
for a change in radiation wavelength.
Yeah, it will take a bit longer to calculate values for search/match. But
Moore's law says processor speeds are *doubling* every 18 months.
Maybe we shouldn't think of Rietveld ICDD card data as just useable in
Rietveld and search/match programs. We may also want to consider
whole-pattern fitting methodologies. With current and near future (3 years?)
processor speeds whole-pattern, rather than profile, fitting will probably
become the de-facto pattern handling procedure.
Recording values proportional to |Fhkl|^2, along with a scaling factor,
would also make Rietveld-based quant methods using these data a minor
procedure. Preferred orientation and thin-film corrections could also be
easily handled. (They are the next best thing to crystal structure data.)
Scott A. Howard, Ph.D. | SSSS A H H
howard@RollaNet.com | S A A H H
http://www.RollaNet.org/~howard | SSSS AAAAA HHHH
Phone USA: (314)364-0851 | S A A H H
See one, do one, teach one! | SSSS A A H H