[molpro-user] Individual atom basis functions

[Nicholas Marshall]

Dear Molpro users,

I am performing computations on a system with significantly different 
partial charges on nonequivalent atoms of the same element. (Some 
carbons in my organic compound are partially negative, according to 
Mulliken population analysis: most are not.) I want to include diffuse 
functions on those atoms, but not the rest of the atoms of the same 
type in my compound. (The geometry optimization gets hopelessly 
confused if I augment everything: see a previous post on that topic.) I 
have tried to say

basis
   default=cc-pVTZ
     c1=aug-cc-pVTZ
     c6=aug-cc-pVTZ
     n=aug-cc-pVTZ
     o=aug-cc-pVTZ
   end

according to the numbering in the Z-matrix, but this does not work.

How can I use a different basis for atoms of the same type?

Thanks,

Nicholas Marshall
2nd year graduate student
Center for Computational Chemistry
University of Georgia