[CP2K-user] [CP2K:15030] Bulk Silicon Tddft

[Laurens]

Many thanks for the suggestions. 
The lowest excitation energies indeed look more reasonable when not using 
smearing.

Laurens



On Monday, March 29, 2021 at 5:42:04 PM UTC+2 Matt W wrote:

> Yes, this is likely the problem. I don't think TDDFT with smearing has 
> been tested / considered.
>
> On Monday, March 29, 2021 at 4:33:38 PM UTC+1 Lucas Lodeiro wrote:
>
>> Hi,
>>
>> I am not an experto on TDDFT, but seeing your output there is something 
>> suspicious. For TD1 you have 4 states occupied (thinking in the 0K 
>> electronic structure), and correctly the TDDFT transitions occur from these 
>> 4 occupied states to the virtual states in the "Gamma Point". But when you 
>> double (or higher) the cell, the excitations occur from states over the 
>> occupied states. For example in TD2: 32 occupied states, and the lower 
>> citations occur from 38... the same occur for TD3, TD4, TD5.
>> It seems that the smearing pumps fractions of electrons to virtual 
>> states, and the program assumes that virtual-low populated states are 
>> occupied states.
>> If this is the problem, you can solve it by removing the smearing... Si 
>> is a semiconductor, and without smearing is possible to compute it 
>> electronic structure.
>>
>> Regards
>>
>> El lun, 29 mar 2021 a las 10:10, Laurens (<lau... at msn.com>) escribió:
>>
>>> Dear all,
>>>
>>> I am trying to do tddft simulations of bulk silicon (please see attached 
>>> band structure and in/output files).
>>>
>>> Since tddft is not implemented for k-points, I have tried using 
>>> supercells of sizes up to and including 5x5x5 times the primitive unit cell.
>>> This changes the lowest excitation energy, but unfortunately the lowest 
>>> excitation energy does not seem to converge to a value close to the 
>>> calculated bandgap of 0.7 eV: the lowest excitation energy decreases from 
>>> an energy of 2.2 eV for the single unit cell, to an energy between 0.28 eV 
>>> (4x4x4 cell) and 0.85 eV (3x3x3 cell).
>>>
>>> Does anybody have any suggestions on tddft simulations of bulk solids?
>>>
>>> Best regards,
>>>
>>> Laurens
>>>
>>> -- 
>>> 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 cp... at googlegroups.com.
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/cp2k/7c817d8a-23c9-4f6c-95af-48cdb4bdaef2n%40googlegroups.com 
>>> <https://groups.google.com/d/msgid/cp2k/7c817d8a-23c9-4f6c-95af-48cdb4bdaef2n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210331/15c36b73/attachment.htm>