Molcas Forum

Addiw7

Member

Registered: 2016-05-24

Posts: 11

MP2 natural orbitals

Dear Molcas Users,

Is it possible to generate something like *Scf.Orb file but containing the MP2 natural orbitals from MBPT2 for instance?

Also when I run my following input

> EXPORT MOLCAS_MOLDEN=ON
> EXPORT MOLCAS_PRINT=verbose
> EXPORT MOLCASMEM=3000

&Gateway
 Coord
  eten.xyz
 Basis set
  H.6-31G*, C.6-31G*
 group
  nosym
 RICD
&Seward
 cholesky
&scf
 iterations
  100
 Charge
  0
 End of Input

&mbpt2
 Frozen
  0
 End of Input

I do not get *mp2.molden file although the manual says it should be generated by default.

Best wishes,
Dawid

PS Now I know I shouldn't use both cholesky and RICD but this is an old input

Last edited by Addiw7 (2017-10-26 08:52:54)

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: MP2 natural orbitals

Try adding the PRPT keyword to &MBPT2.

Addiw7

Member

Registered: 2016-05-24

Posts: 11

Re: MP2 natural orbitals

Right, thank you. I read about this option but I did not relate it to generating MOs files. So I get my *mp2.molden and *MP2Orb files in the latter, it says that these are canonical MP2 orbitals. Can I still treat the occupation numbers as if these were natural orbitals?

Last edited by Addiw7 (2017-10-26 11:14:44)

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: MP2 natural orbitals

It seems to me the label is wrong, these look like natural orbitals: obtained from density matrix, with defined occupations and not defined energies.

moev96

Member

Registered: 2020-11-02

Posts: 13

Re: MP2 natural orbitals

Hello everyone,

would be great if someone could help me. I try to use the PRPT in MBPT2 to get the orbital file.
However this seems to fail with chol in seward. Without this it runs and has no problems. For a simple structure this would not be a problem, however I want to extend this to a system where I really need the cholesky decomposition.

Here is my example input file:

&GATEWAY
Basis set
 C.cc-pVDZ
 * C positions  
 C1     0.00000000     0.70551969     0.00000000 /Angstrom
*C2     0.00000000    -0.70551969     0.00000000 /Angstrom
 C3     1.24440473     1.39490430     0.00000000 /Angstrom
*C4    -1.24440473    -1.39490430     0.00000000 /Angstrom
*C5    -1.24440473     1.39490430     0.00000000 /Angstrom
*C6     1.24440473    -1.39490430     0.00000000 /Angstrom
 C7     2.42020458     0.70939025     0.00000000 /Angstrom
*C8    -2.42020458    -0.70939025     0.00000000 /Angstrom
*C9    -2.42020458     0.70939025     0.00000000 /Angstrom
*C10    2.42020458    -0.70939025     0.00000000 /Angstrom
 Cartesian(all)
end of basis

Basis set
 H.cc-pVDZ
 * H positions
 H3     1.24242935     2.47717260     0.00000000 /Angstrom
*H4    -1.24242935    -2.47717260     0.00000000 /Angstrom
*H5    -1.24242935     2.47717260     0.00000000 /Angstrom
*H6     1.24242935    -2.47717260     0.00000000 /Angstrom
 H7     3.36119277     1.24299298     0.00000000 /Angstrom
*H8    -3.36119277    -1.24299298     0.00000000 /Angstrom
*H9    -3.36119277     1.24299298     0.00000000 /Angstrom
*H10    3.36119277    -1.24299298     0.00000000 /Angstrom
 Cartesian(all)      
end of basis
Symmetry
 XY XZ XYZ

&SEWARD 
 chol
&SCF    
 iterations 
  200
 charge 
  0
  Frozen 
 0 0 0 0 0 0 0 0
&MBPT2
 PRPT

The end of the log file gives:

 ***************WARNING************************
 
 Too many iterations, this is what you get after:
                  100  Iterations
 The residual is  0.19310849754510317      and not   1.0000000000000000E-008
 **********************************************

  Input file to MOLDEN was generated!


***** Cholesky MP2 energy completed *****
Total CPU  time:       0 hours  0 minutes   0.23 seconds
Total wall time:       0 hours  0 minutes   0.26 seconds


***** Cholesky MP2 completed *****
Total CPU  time:       0 hours  0 minutes  10.71 seconds
Total wall time:       0 hours  0 minutes  17.80 seconds

--- Stop Module:  mbpt2 at Fri Jan 29 12:44:24 2021 /rc=-1 (Unknown) ---
--- Module mbpt2 spent 19 seconds 
 Aborting..

Would be great if anyone could help me, I am a little bit lost at this point.

Best,

Marvin