PC GAMESS/Firefly-related discussion club

Learn how to ask questions correctly

Andrei V Scherbinin
scherb@classic.chem.msu.su

>>>>I was thinking that perhaps symmetry groups could be used, but merely repeating the lattice over hundreds of times would not be a true simulation because it would not account for the interactions between unit cells. Am I wrong in thinking this?

>>>>>One more piece of information to explain why I don't just let GAMESS take care of the forces - the material I'm analyzing is doped, but irregularly, so the unit cell I'm using doesn't include all aspects of the compound's structure.


If your doped material is so irregular, may be this is not a good idea to look for models with periodic
boundary conditions, since a very huge supercell would be required. As an alternative, embedded cluster
models would be appropriate. To study localized effects (say, local electronic excitation or emission
at the impurity atom), it is sometimes sufficient to use a cluster that includes much less number of
QM atoms (say, the impurity atom and its nearest coordination sphere only) than a typical supercell does,
provided that the interface region and the environment of the cluster are described properly. For example,
an EFP technique can be used for this purpose, or some other QM/MM approach, etc.