From: SMTP%"hubbardcr@ma160.ms.ornl.gov" 12-NOV-1995 16:51:05.81
Reply to: Snyder RE: This is the first round of a discussion on
the Rietveld mail server that will be of interest to us at ICDD. Bottom
from LeBail and top is my RE - many others will probably be heard from. I
will collect the discussion and pass it on.
My quick thoughts while I should be taking care of my Group's funding and
operation are:
1. The problem of use of Rietveld data in the file is really a problem of
the search engine and match algorithm. Also there is the need for education.
Nowhere within the database is there a problem with the data storage. R
quality marks could be OK - certainly better than the C for calculated as we
have used C to mean simulation of the powder diffraction trace and subsequent
analysis of it like it were an experimental trace. (Should C be renamed S for
simulated pattern when R is introduced for Rietveld calculated?)
2. Many search engine algorithms on computers today can handle the full
detail of the Rietveld d/I lists. In particular, the so called third
generation codes handle it beautifully. They could even potentially spit out
info on texture as well. Clearly if we tell the authors of computer
algorithms which search the PDF, the intent to include such R type data then
they will write code to handle it or they will be dropped by users wanting to
get the correct answers for more modern codes. In either case the community
benefits.
3. Manual methods (Hanawalt index) also have rules (e.g. one strong line
within a certain window) that makes these patterns work for hand searches.
Perhaps better rules are needed, but what is in place work OK for some
patterns with resolution much higher than the average user will get during
phase ID data collection. The trained user of the file can handle the match
part - For those that can't, there you have a topic for ICDD workshops and
training efforts.
4. To me it is truly disappointing that Rietveld and simulated patterns are
not being saved in PDF with full detail for use by ICDD customers. Clearly,
a disgrace for the ICDD that this has lingered on for over a decade since a
number of us recognized and proposed the need.
Camden Hubbard
Oak Ridge National Lab