[sdpd] How hot are the "special methods" for SDPD ?

[Armel Le Bail <armel...@fluo.univ-lemans.fr>]

Hi,

We are all waiting for that promised new revolution in powder 
diffractometry.

The latest was the Rietveld method revolution. In fact, it very slowly
changed our habits, but now, after 30 years, there are more than 500 
papers using and citing that method, per year.

Was SDPD a revolution in the eighties ? Maybe, but we are still at 
50-60 papers per year (I mean solving real cases, because writing 
about SDPD - methods - review papers - redetermining known 
structures - leads probably to more than 100 papers per year). 

But the real revolution, which we are all expecting, will occur if
SDPP is applied to organic compound at a large scale. What would be a
large scale ? Well, 500 SDPD papers per year would be fine. We are 
far from this number.

Almost all ingredients for that revolution are already here or in 
preparation :
- Thousands of powder diffractometers in laboratories, plus
    a few synchrotron sources.
- Inclusion of >200000 calculated powder patterns of organic and
    organo-metallic compounds in the JCPDS-ICDD : this will
    allow identification, first SDPD step (or second if you count
    synthesis as the first).
- Multitude (really amazing !) of "special methods" allowing 
     molecule location by Monte Carlo / simulated annealing, genetic 
     algorithm, maximum likelihood, maximum entropy, structure 
     prediction, grid search (etc, etc). The subject being so promising
     (pharmaceutical companies invloved...) that commercial 
      software are already available (some very expensive) like
      PowderSolve, Endeavour, DASH, etc, as well as freeware 
    (PSSP, OCTOPUS, ROTSEARCH, ESPOIR, etc...), some of 
    them being not easily obtainable after 6 years of existence (this
    could explain why you find always the same author names in the
    publication lists on these subjects).

See how hot are the "special methods" in a citation study at :
http://www.cristal.org/iniref/most-cited-5.html

In fact, my personal view is that the subject is made artificially
hot (lot of noise, few experimental studies, almost same number
of review papers, redeterminations of already known structures
  and "new" method papers), but is warming, so that 
you may consider to contribute more (give access to your software 
for instance) in order to accelerate the revolution come in... 
We are still at a 1/4 ratio of organic/inorganic compounds in the 
SDPD world, compared to 4/1 in the single crystal world. The 
revolution will be there when both worlds will coincide at the 
same 4/1 ratio.

Best,

Armel Le Bail
http://www.cristal.org/course/


PS - New clue about the revolution coming : The long awaited
         program DASH (known since 1998 from the SDPD Round 
         Robin, but still unavailable ?), its manual is cited in an abstract 
         at APD-III :
http://bragg.ing.unitn.it/apd3/INVITED/Shankland.doc
http://bragg.ing.unitn.it/apd3/Technical_programme/technical_programme.html

PS2 - If the revolution comes, it will not change a lot the current
SDPD approach to inorganic compounds ;-).

PS3 - ESPOIR 3.5 for Windows was downloaded exactly 250 times since
one year. What can be said about that :
a) if it was sold at the same cost as PowderSolve, I would be rich ;
b) if it was sold, how many would have bought it ? I would probably
not be rich ;
c) not seen a lot of structures solved by using ESPOIR... but it is too
soon for concluding.


 

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