Of DOS projected on localized molecular orbital

[yellowice]

Dear all:
I'm wondering if it's possible to calculate DOS projected on certain localized molecular orbital using ABINIT,
say,the sigma_g bonding orbital of H2 when it's adsorbed on Au(111) surface. Any comment is highly appreciated!

[jzwanzig]

The projected DOS feature currently projects angular momentum components around atoms, not groups of atoms. It may be though that by combining these projections with plots of the wavefunctions for specific bands viewed at specific atoms, the information you're looking for will be clear--

[yellowice]

The projected DOS feature currently projects angular momentum components around atoms, not groups of atoms. It may be though that by combining these projections with plots of the wavefunctions for specific bands viewed at specific atoms, the information you're looking for will be clear--

Thanks for your reply. I think your proposal is valuable but not workable since the "specific bands"
is quite vague due to the continuity of energy eigenvalues at least in the case H2/metal-surface.
Anyway, i thought the localized MO is just a combination of atomic orbitals, and there is no reason
that the coding for DOS projected on localized molecular orbital is much more complicated than
DOS projected on angular momentum components around atoms.

[jzwanzig]

You are probably right that it would be do-able to write a routine to project the DOS over specified linear combinations of atom-projected angular momentum states--why not give it a try?