Bond Valence constraints in RMCProfile
Andrew
Stefan Norberg
Dear Andrew,
Sorry about the lack of information in the RMCProfile manual, I ought to do something about that...
Anyway, the BVS constraint is easy to set up. What you need is to create a small .bvs file (using your overall common filename but with file extension bvs).
The content of the file is like this:
== == == == This is not part of the file == == == ==
Y 3 // atom #1
Zr 4 // atom #2
O -2 // atom #3
0.055 0.055 0.140 // chi^2 weights (for each type)
0 0 0 // Y-Zr (Rij, B, cut-off distance)
2.019 0.37 3.2 // Y-O
1.928 0.37 3.2 // Zr-O
400000 // "intermediate save"
200000 // "neighbour list update"
== == == == File ends above here == == == ==
The first set of rows defines the type of atoms and the oxidation state you expect them to have. I think 6 rows of atom information is the current limit.
After that you have a line defining how constraining the BVS constraint should be for each type of atom. It could for example say "0.015 0.015 0.015 0.015" if you have defined 4 atoms above.
Next is a set of instructions concerning BVS parameters, given in the following sequence:
"for 3 atoms"
"for 4 atoms"
etc.
The final two rows tell how often you want a short BVS status update, as well as how often the neighbour list should be recalculated. The later is a list that tells which atoms are near each others and it needs to be regularly updated to adjust for atom moves in your config file.
It is possible (and often useful) to use the BVS constraint to monitor the BVS distribution for each type of atom even if you dont want it to impact on the RMC modelling. That can be done by setting all the chi^2 weights to 100 instead of a small number like 0.025. The smaller the more constraining.
I use "Brese N E and O'Keeffe M, Bond valence parameters for solids, (1991) Acta Cryst B47, 192-197" to obtain BVS parameters.
Please cite "Norberg S T, Tucker M G and Hull S, Bond valence sum: a new soft chemical constraint for RMCProfile, (2009) J. Appl. Cryst. 42, 179-184" for use of the BVS constraint.
You can e-mail me at s...@chalmers.se if you have further questions about the .bvs file.
Best regards,
Stefan
Stefan T Norberg
Environmental Inorganic Chemistry
Sent: Tuesday, January 22, 2013 21:42
To: rmcpr...@googlegroups.com
Subject: [RMCProfile] Bond Valence constraints in RMCProfile
You received this message because you are subscribed to the Google Groups "RMCProfile" group.
To post to this group, send email to rmcpr...@googlegroups.com.
To unsubscribe from this group, send email to rmcprofile+...@googlegroups.com.