[snip]
I must admit I have been luckier.
On the occasions I have struck phases not in the ICDD database,
having access to the ICSD or Cambridge databases would not have
helped as they had not been seen before in the literature.
(Submission to the ICDD was not applicable in many occasions
due to time constraints and/or commercial and in confidence aspects
of much work - It would be nice to be able to have a type of "easy"
submission for new data to the ICDD over the internet. Perhaps in
the same manner of the CIF/abstract submission for the IUCr -
in that you are emailed back any errors or problems?)
----
Of more concern is the effect on people doing phase ID who have
no or limited training or background knowledge, and are effectively
using phase ID as a peak matching operation and nothing more.
I think you can survive doing phase ID without any basics of
X-ray diffraction or crystallography. Not saying this is a good
thing - but if training is not an option - it can be done.
If you don't know what you don't know, such things like spurious
peaks due to contaminating radiation, sample effects, et al
can catch you unawares for quite some time. However, the
word reference database implies something to be trusted.
The more expensive it is, then the more it is considered
to be trustworthy?
-----
In terms of the overall quality of the ICDD database,
I should note that is very advisable to use it to help
checking structures added from the literature into
rietveld input files - or at least that's the way I was
taught. Calculate a pattern from the Rietveld structure,
then compare it to ICDD intensities. If variation, then it
is usually a problem with the structure - not the ICDD
database.
Though an interesting paper to read is:
"Reliability of Commonly Used Versions of the Powder Diffraction
File PDF-2" A. Drummond and D.G. Hay. Powder Diffraction, Vol 5. No 3,
September 1990.
Cheers,
Lachlan.
PS: While not directly on this topic - but more on the idea of
doing black box science and analysis - it would be interesting to
get opinions on the idea that much of inorganic crystallography
and crystal chemistry could become a lost or dying art when the
generation of crystallographers who were rigorously trained
(before the age of powerful desktop computers) depart the
scene. Leaving the possibility that only structures that are
easily solvable by automated computer based means will be
attempted or successful?
Which institutions now teach a rigorous crystallography
course of the old style - where you learn to recite space
groups, transformations and extinctions backwards in your sleep?
As with no longer teaching Latin in public schools, no longer
having the rigor of old style crystallography (and relying on
computers to make up for this) will diminish the available
skills base to attempt these thorny crystallography problems.
---
People may think the above a bit inconsistent with saying how desirableit would be to have "expert systems" in Quantitative Rietveld analysisto make the decisions of a trained diffractionist/crystallographer.Though I don't believe that "routine", "accurate" quantitative analysis should be the pain in the backside it probably is at the moment. Or considered anything special other than peak fitting of known crystal structures (or hkl files) to raw data - with rigorousmethods for identifying problems, determining (or overcoming) present sources of error such as microabsorption. Though creating such an expert system is obviously not a 5 minute project.
Lachlan Cranswick - Melbourne, Australia _--_|\ Phone/Fax : (613) 9455-1345 / \E-mail : lachlan@melbpc.org.au \_.--._/Mobile Phone/Voice Mail : 0412-1141-31 vCrystallographic WWW : http://www.unige.ch/crystal/stxnews/stx/volnteer.htm