[Wien] About isplit!!

[Peter Blaha]

> The problem has been discussed previously. Thank all for the reply.
> However, the case is that the central Co atom has an octahedral surrounding with the neighboring six N ligands, but the x, y, z axes DO NOT point towards these ligands as reminded by Prof.P.Blaha.
> Now could I plot t2g orbital as a sum of dxy,dyz,dxz orbitals and eg orbital as a sum of dx2y2 and dz2 orbitals?

Two possibilities: a) Use the program qtl. In the input you can define any
linear combination of Ylm's. However, this might still be more difficult
than

b) If the point symmetry in case.outputs is 1 (or -1) you can introduce
your own "local rotation matrix". Just introduce a local coordinate system
pointing towards the ligands as you like it.
If you have highet symmetry, you still can use a local rotation matrix,
but the must use more (all) LM values and a "full" k-mesh, i.e. one
generated with just the identity operation.

                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------