Re: [sdpd] Recent Congress and comiing ones about SDPD

[Jon Wright <wright...@esrf.fr>]

 > PS - About "realizing the obvious", some obvious things that could be 
realized :

How about using texture or thermal expansion for the indexing problem as 
well? For the former case systematic variations in intensity relate 
lines to each other and give some indication about directions in 
reciprocal space. For the latter case systematic shifts in peak 
positions relate lines together and generate similar information. 
D'Yvoire indexed FeAsO4 by doing cation/anion substitutions and 
generating systematic lineshifts more than 30 years ago (C. R. Acad. Sc. 
Paris, 1972 275C 949) and temperature can do the same thing where the 
chemistry is more resticted. I wonder how many of the new crop of 
indexing programs can use information like this, and even try to figure 
out which lines are impurities based on their behaviour in multiple 
patterns? An input file would only need to accept more than one position 
for each line, the set of temperatures, and calculate the thermal 
expansion for each proposed unit cell. Those which are right ought to 
get a much better "multipattern" M20/F20/Rwp/etc.

Celeste Riess had a nice example at the SSPD meeting where the data was 
collected for a very thin flat plate sample and also a capillary. The 
strong texture seemed to help with the indexing as at least one axis was 
unambiguous. Presumably McMaille could exploit this kind of information 
as it is using the intensity information? Someone would have to get it 
to remember the intensities from each pattern and fit some kind of 
preffered orientation function, but it ought to help select the good 
solution from the crop of possibilities, even if the texture function is 
only a lousy approximation. The input file would only need to accept 
more than one intensity for each Bragg peak and maybe a hint about which 
ones are the most "texture free".

Any indexing program which can generate lots of possible unit cells 
should be able to benefit from more information in the input data. Can 
all this be done using laboratory diffractometers? Sadly I don't have 
one, but Zachariasen was using anisotropic thermal expansion (Acta Cryst 
1963 16 369) long before big facilities were available. You can even 
collect data for differently textured samples at multiple temperatures, 
all in the home lab. Perhaps the diffractometer manufacturers can one 
day produce a "single crystal" instrument which accepts powdered samples...?

Cheers,

Jon

PS: Something being obvious (=someone did it before I was born ;-) 
doesn't make it any less useful!


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/UIYolB/TM
---------------------------------------------------------------------~->

 

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