[molpro-user] CCSD Electron Density Calculations?

Gerald Knizia knizia at theochem.uni-stuttgart.de
Wed Mar 4 09:48:48 GMT 2009


On Wednesday 04 March 2009 02:17, Terry Frankcombe wrote:
> Am I missing something here?  You have available the natural (or some
> other) orbitals and occupations, right?  Surely it's a simple matter to
> write about ten lines of Fortran to read these in and calculate the
> total electron density at any point?

Yes, you are missing something:
  1) It would be /at least/ several 100 lines and multiple days, because you 
would also need to read in the basis declarations, match them to the 
orbitals, implement the spatial symmetry adaption and calculate the basis 
functions themselves (which you also would need to program). Certainly not a 
task you would do if there was any way to avoid it! (The minimum line- and 
time estimates are for the case in which you know exactly what you are doing 
beforehand...)
  2) To obtain natural orbitals and occupation numbers for CCSD you first need 
to have a density matrix to extract them from. And for open-shell CCSD, 
Molpro cannot calculate them. And if they are available, Molpro can also 
evaluate the densities itself (no need to code that up).
  3) Seriously, don't do scripting in Fortran.
-- 
Gerald Knizia



More information about the Molpro-user mailing list