Maria,
an objective evaluation of the performance of a method should be done
independently from the methodic preferences of any author. Thus, only a
comparative study like the Reynolds Cup (mineral analysis) can give a
suitable picture.
McCarty, D.K. (2002) Mineral analysis of clay-bearing mixtures: The
Reynolds Cup contest. International Union of Crystallography Commission
on Powder Diffraction Newsletter, 27, 12-15.
Omotoso, O., McCarty, D.K., Hillier, S., Kleeberg, R. (2006) Some
successful approaches to quantitative mineral analysis as revealed by
the 3^rd Reynolds Cup contest. Clays and Clay Minerals, 54 (6), 751-763.
Kleeberg, R. (2004) Results of the second Reynolds Cup contest in
quantitative mineral analysis. Commission of Powder Diffraction,
International Union of Crystallography CPD Newslett. 30: 22-24.
From these publications you will see that the Rietveld method is in
fashion, but it is not (yet?) the only and always most successful method
in QPA of complex samples.
Regarding the accuracy of the Rietveld method in QPA one must take into
account the problem of qualitative analysis (in my personal experience
the most important issue in practice of complex samples). But even in
qualitatively known and "easy" systems, absolute errors of 1-2% are an
optimistic estimation for the state-of-the-art of a broad community, see:
Madsen, I.C, Scarlett, N.V.Y., Cranswick, L.M.D. and Lwin, T., 2001,
Outcomes of the International Union of Crystallography Commission on
Powder Diffraction round robin on quantitative phase analysis: samples
1a to 1h. J. Appl. Cryst: 34, 409-426.
Scarlett, N.V.Y., Madsen, I.C., Cranswick, L. M. D., Lwin, T., Groleau,
E., Stephenson, G., Aylmore M. & Agron-Olshina N. 2002. Outcomes of the
International Union of Crystallography Commission on Powder Diffraction
round robin on quantitative phase analysis: samples 2, 3, 4, synthetic
bauxite, natural granodiorite and pharmaceuticals. J. Appl. Cryst. 35:
383-400.
Do not trust the accuracies as demonstrated in some publications for
known mixtures or standards: It is easy to "press" a Rietveld QPA result
towards the known true concentrations. The accuracy for real unknown
systems can be expected to be much lower, see above.
Regarding refinement of atomic position in QPA: No, because of the
availability of good structural daa from single crystal measurements,
and because of the limited influence of slightly wrong atomic positions
on the total diffracted intensity of a phase.
Refinement of elemental substitutions: Can be necessary if elements
having different scattering power are substituting and significantly
biasing the total diffracted intensity as well as the phase density.
Constraints, for example to lattice parameters, and limitation of the
parameter space can be helpful.
Reinhard Kleeberg
Maria Chrysochoou schrieb:
To the best of my knowledge, Rietveld is the only statistically
reliable method to obtain a quantitative distribution of mineral
phases in complex systems, such as soils, cements etc. I know that
there are some thermal methods (TGA, DTA) that require pure standards
for quantitative analysis, which are not always available for complex
materials. Please correct me if I am wrong, I deal with waste and
heavy-metal contaminated soils and my experience with analytical
techniques is limited to XRD, SEM, EPMA, XANES, FTIR and optical
microscopy, so I am not be up-to-date with more sophisticated equipment.
Can anybody provide me with a good reference to support such a
statement? I apply Rietveld quantitative analysis to assess the
mineralogy of contaminated media, and I want to support the statement
in a journal publication that this is the more statistically reliable
of the available techniques, if one needs to look at a bulk sample
(not focus on the nm scale).
I also have a question on the accuracy of the method. I follow the
discussions on the list and I would like to ask where you draw the
line for a “good” and a “bad” refinement. Do you make a distinction
between different types of materials and applications? My take is that
I cannot expect to have an accuracy of 0.1% when I am dealing with
samples that contain 10 phases and are bound to be subject to
significant variability (such as waste and soils). I am happy if I
have a 1-2% absolute error in the quantification and I do not even
attempt to refine atomic positions or elemental substitutions. Can you
give me your two cents on that?
Maria Chrysochoou, Ph.D.
W.M. Keck Geoenvironmental Laboratory
Stevens Institute of Technology
Tel.: +1 201 216 8773
Fax: +1 201 216 8212
E-mail: [EMAIL PROTECTED]
begin:vcard
fn:Reinhard Kleeberg
n:Kleeberg;Reinhard
org;quoted-printable:TU Bergakademie Freiberg;Institut f=C3=BCr Mineralogie
adr:;;Brennhausgasse 14;Freiberg;Sachsen;D-09596;Germany
email;internet:[EMAIL PROTECTED]
title:Dr.
tel;work:(+49) (0)3731 393244
tel;fax:(+49)(0)3731 393129
url:http://www.mineral.tu-freiberg.de/mineralogie/roelabor/
version:2.1
end:vcard
