[CP2K-user] [CP2K:16724] Re: MP2 dynamics with multiple time step scheme

['Frederick Stein' via cp2k]

>From CP2K 8.1 on, the MP2 section should look like this
&WF_CORRELATION
  &RI_MP2
# It is not necessary to set this section but its options determine 
accuracy and performance of your calculation
    &CPHF
# As tight as necessary, as high as possible
      EPS_CONV 1.0E-4
# Large enough to converge your structures, default is usually good enough
      MAX_ITER 30
    &END CPHF
  &END
  &INTEGRALS
    &WFC_GPW
# Check the convergence of this cutoff (forces)
# Cutoffs of 150-200 Ry are usually sufficient and cause much lower costs
      CUTOFF 300
      REL_CUTOFF 50
      EPS_FILTER 1.0E-12
      EPS_GRID 1.0E-8
    &END
  &END
  MEMORY 1200
  NUMBER_PROC 1
&END 

But I cannot help with the velocities.

Try some force calculations first to check all parameters (RUN_TYPE 
ENERGY_FORCE)

Thorren Kirschbaum schrieb am Mittwoch, 16. März 2022 um 15:04:57 UTC+1:

> I have now set up an input for mp2-AIMD with the current input format, 
> does this look correct? It confuses me a bit thet the MP2 method is a 
> subsection of DFT...
> Which unit does CP2K expect for the velocities? Can I set the units 
> somewhere?
> Many thanks and best
> Thorren
>
> Thorren Kirschbaum schrieb am Mittwoch, 16. März 2022 um 13:49:11 UTC+1:
>
>> Thank you!
>>
>> Frederick Stein schrieb am Mittwoch, 16. März 2022 um 12:54:55 UTC+1:
>>
>>> Dear Thorren,
>>>
>>> At least for the WF_CORRELATION section, I can tell you, it will not 
>>> work anymore. For reference, check our manual (
>>> https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/XC/WF_CORRELATION.html). 
>>> The major changes are
>>> - remove the METHOD keyword
>>> - move the CPHF section into the RI_MP2 section
>>> - wrap the WFC_GPW section into an additional INTEGRALS section
>>>
>>> Furthermore, remove the RI_AUX_BASIS_SET keywords in the KIND sections 
>>> and use something like 'BASIS_SET RI_AUX RI_TZ'.
>>> If you use the CP2K trunk, replace the LIBXC section with functional 
>>> <your-libxc-functional> by a section of name <your-libxc-functional>. The 
>>> manual will tell you all available functionals and their parameters.
>>>
>>> HTH
>>>
>>> Frederick
>>>
>>> Thorren Kirschbaum schrieb am Mittwoch, 16. März 2022 um 12:14:47 UTC+1:
>>>
>>>> Hi everyone, 
>>>>
>>>> I found this input for running MP2 dynamics using a multiple timestep 
>>>> scheme in the SI of this publication from 2015: 
>>>> https://doi.org/10.1063/1.4927325 
>>>> Will this input still work when I use it with the current version of 
>>>> CP2K?
>>>>
>>>> Many thanks and best
>>>> Thorren Kirschbaum
>>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/7d0712e3-2554-4528-ba29-421eb93223a5n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220316/874eb87d/attachment.htm>