Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn
You are not logged in.
Hello Prof Tian,
There are two different types of basis sets, cartesian and spherical type. If the basis set has only p-type functions then they are the same. However, for d-type and f-type the components of the basis set are different. The d-functions for spherical are 5D and for cartesian are 6D.
For correlated calculation (CCSD) one can obtain the natural orbitals and create the .wfn file, 1PDM and 2PDM (after solving the lambda equations).
I want to ask,
What are the changes
1. in the natural orbitals
2. in the 1PDM (both in MO and AO basis)
3. in the 2PDM (both in MO and AO basis)
when one use Cartesian (or Spherical) basis rather than Spherical (or Cartesian) ones?
Thanks,
Prasanta
Offline
Dear Prasanta,
When >=D basis functions are involved, in fact, calculations in Cartesian and spherical types of basis set correspond to slightly different basis sets. So, all information related to orbitals and density matrix (irrespective of MO or AO basis) is changed, .
Best,
Tian
Offline
Thank you Prof. Tian. The problem lies there actually for using >= D basis functions. I also think that for each basis set and for each atom (say aug-cc-pVTZ with S atom) the orbitals and density matrices will be different. Can you shed some light if I have the orbitals/density matrices in spherical gaussian, can I transform them to cartesian versions?
Thanks,
Prasanta
Offline
Dear Prasanta,
You can perform the transformation using the transformation relationship between 5D and 6D, see my blog article http://sobereva.com/97. Also you can check the subroutine of loading .fch file in Multiwfn source code package (search "subroutine readfch" in fileIO.f90), there are relevant codes.
Best,
Tian
Offline