Author Topic: Cannot converge S1 CC/cc-pVTZ with S0 MP2/cc-pVTZ geometry  (Read 2749 times)

farmaceut

  • Newbie
  • *
  • Posts: 2
  • Karma: +0/-0
Hello everyone! :)

I have geometries of structures I optimized as a ground states (S0) using MP2/cc-pVTZ . Now, I need to recalculate them as S1 using CC2/cc-pVTZ. Nevertheless, it cannot converge them anyhow – 200 iterations seems to be not enough what seems to be quite weird...

I cannot attach files directly, so here are links to Pastebin:
- control https://pastebin.com/529HNwnd
- job.last https://pastebin.com/9gLFyv8Y

Thank you

uwe

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 558
  • Karma: +0/-0
Re: Cannot converge S1 CC/cc-pVTZ with S0 MP2/cc-pVTZ geometry
« Reply #1 on: July 01, 2020, 12:35:06 PM »
Hello

from what I can see from your output, the shortest distance between two atoms is an O-H bond of 182 pm. To me this looks like a mismatch of units and the coordinates are probably in Angstroem and not in atomic units?

Uwe

farmaceut

  • Newbie
  • *
  • Posts: 2
  • Karma: +0/-0
Re: Cannot converge S1 CC/cc-pVTZ with S0 MP2/cc-pVTZ geometry
« Reply #2 on: July 01, 2020, 08:18:01 PM »
Hello Uwe,

I found out what was wrong. I prepared input using TMolex and coord file from the S0 geometry. I did it two times:
1)  importing coordinates directly from coordinates file into new input
2) generating .xyz from coordinates file and then importing it.
Both ways caused presented error... because after imporiting they completely differs from the original file! The solution I found was quite simple – I have duplicated coord file from S0 stage and moved it to the S1 job folder. Then run... and it goes on as for now :) Hope it will finish!