[Jmol-users] Bond lengths & valence sum calculation with Jmol?

[Alan H. <he...@il...>]

Jmol already calculates bond-lengths to decide which atoms are bonded
according to distance criteria. It would be very nice to be able to output
a list of these bond-lengths, sorted according to central bonded sites.
For example, the output obtained from
http://icsd.ill.fr/icsd/index.php?action=Bonds&id[]=72989

Then of course it would be an extra bonus to be able to output a list of
the bond angles as well, as can be done from the above URL.

Finally :-) if you have a list of bond lengths, it would be very useful
for inorganic/ionic structures to be able to calculate the bond-valence
sum around the different cations. This is simply the sum of the individual
bond valences = exp((Ro-R)/B) where R is the bond length and Ro and B are
parameters for given cation-anion oxidation states obtained from a look-up
table such as http://icsd.ill.fr/icsd/help/valence.html

The valence states would be extracted from the CIF file, as used for
drawing ionic radii. (Actually in my calculation I have a self-consistency
loop, which assumes intial valence states, calculates the valence sum for
comparison, and then if it is significantly different, uses this new value
as the supposed valence state to select new Ro and B parameters).

The bond valence sum is a very important check on the correctness of a
structure. If it unreasonable then almost certainly something is wrong
(atom coordinates, disorder, stoichiometry etc...)

Alan.
_____________________________________________________________
Dr Alan Hewat, ILL Grenoble, FRANCE <he...@il...>fax+33.476.20.76.48
+33.476.20.72.13 (.26 Mme Guillermet) http://www.ill.fr/dif/people/hewat/
_____________________________________________________________