[CP2K-user] [CP2K:17576] Re: dftmd with b3lyp hybrid

[Alex]

Hi Marcella,

Thanks for responding! There is a giant zoo of methods people seem to use 
for a system like this, from PBE to relatively exotic stuff and everyone 
states that their method is "optimal." That said, I do agree that I should 
have probably started with something that is known to reproduce the 
electronic structure of graphene rather than what works for small organics. 
:)

Again, thanks.

Alex

On Monday, August 29, 2022 at 6:10:57 AM UTC-6 Marcella Iannuzzi wrote:

>
>
> Well
> Let's say that you should have a good reason to use all-electrons and 
> hybrid-functionals with no ADMM for a BOMD simulation of a condensed matter 
> system. 
> Moreover one should consider that graphene is semi-metallic and the 
> convergence might be strongly dependent on the system size as well as other 
> parameters/conditions. 
> In the literature you can find many DFT studies of water/graphene 
> interfaces, I would start from there to figure out which are the optimal 
> settings to correctly reproduce electronic, structural and dynamic 
> properties. 
> Regards
> Marcella
>
>
> On Monday, August 29, 2022 at 9:37:05 AM UTC+2 Alex wrote:
>
>> Hi everyone,
>>
>> I am trying set up a dftmd simulation (room-temp NVT, rectangular box 
>> periodic in XYZ) of a tiny solvated sheet of graphene (~1700 electrons 
>> total). I am using the following as my starting point for the force 
>> calculation: 
>> https://github.com/misteliy/cp2k/blob/master/tests/QS/regtest-hybrid/H2O-hybrid-b3lyp.inp 
>> -- the only two major changes being POISSON set to be periodic in xyz and 
>> I'm also adding D3 corrections. 
>> The initial structure comes from classical MD and SCF is refusing to 
>> converge with that dreaded huge oscillation. Should I abandon this idea and 
>> use something much more familiar to me (say, PBE + GTH potentials) or am I 
>> setting it up incorrectly? 
>> Thank you,
>>
>> Alex
>>
>

-- 
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/346a9665-9286-43c1-8d45-3abc2d06ddfen%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220829/70e83e3f/attachment-0001.htm>