Author Topic: singlepoint with cc2 and adc(2) and natural transition orbitals  (Read 6122 times)

vspata0

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
singlepoint with cc2 and adc(2) and natural transition orbitals
« on: November 18, 2013, 08:34:50 PM »
Hi,
  I am trying to send singlepoint calculations with both adc(2) and cc2 simultaneously and afterwards I would like to calculate ntos.  How can I obtain the cc2 natural transition orbitals with cc2 if the eigenvectors are not symmetric.  Otherwise I already know how to obtain ntos for symmetric wavefunctions such as adc(2) with ricctools analyzing the right eigenvectors.  If anyone can help explain this to me I really appreciate it. 
Vince

Arnim

  • Developers
  • Sr. Member
  • *
  • Posts: 253
  • Karma: +0/-0
Re: singlepoint with cc2 and adc(2) and natural transition orbitals
« Reply #1 on: November 20, 2013, 01:16:30 PM »
Hi Vince,

you can't use adc(2) and cc2 simultaneously for this purpose. Both methods save their restart information on the same CCRE0* files. Just run it in different directories. The ricctools call is then the same in both cases.

Best regards,

Arnim

vspata0

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
Re: singlepoint with cc2 and adc(2) and natural transition orbitals
« Reply #2 on: November 20, 2013, 06:14:52 PM »
Thanks for your quick response.  There is something notable in the output though which I would like to ask about.  After the adc(2) calculations finish it says:
Info: copy ADC(2) right eigenvectors into left eigenvector files.
Do you know if these left eigenvector files remain untouched by the cc2 calculations?  Or is it possible to get the adc(2) ntos after all?
Vince

Arnim

  • Developers
  • Sr. Member
  • *
  • Posts: 253
  • Karma: +0/-0
Re: singlepoint with cc2 and adc(2) and natural transition orbitals
« Reply #3 on: November 21, 2013, 12:31:59 PM »
This Info comes up, when you calulate spectrum or exprop. But you don't need that for the ntos.

If you calculate spectrum or exprop with adc(2) and cc2 simultaneously, all the adc(2) steps are performed first and then the cc2 steps. This means the left eigenvetors will also be solved and overwritten in the cc2 steps.
The corresponding message from cc2 in the output is: "solve for left     eigenvectors needed for transition properties."

Arnim