[gpaw-users] rt-tddft + kpoint

[Ask Hjorth Larsen]

You can modify the wavefunctions any way you want, but remember that
with the PAW corrections, it is not so easy to be certain that you are
modifying them in a physically sensible way (for example normalization
is not trivial).

The best way of doing it might be to implement it directly in the
calculation script using numpy.  It could be troublesome in parallel
of course, so you could load, modify, and save as a new gpw file in
serial, again making sure in a simple case that it works.  This is
tricky business.

Best regards
Ask

2016-01-18 17:47 GMT+01:00 Ask Hjorth Larsen <asklarsen at gmail.com>:
> Dear Luca
>
> There is a 'laser' module, gpaw.tddft.laser, with a CWField class for
> which implements a time-dependent potential.  It would be
> straightforward to implement other external potentials.
>
> I cannot promise that this mechanism is properly implemented and in
> working condition, so maybe someone else can confirm, or you can test
> it and verify that it behaves as it should in a simple case.
>
> Best regards
> Ask
>
> 2016-01-18 15:47 GMT+01:00 Luca Daumann <luca.daumann at gmail.com>:
>> Dear Ask
>>
>> could you please give me an example of how to use  time-dependent external
>> field ?
>> Is there a way to write the wfns in ASCII format ? In this way one could
>> modify them and than restart from them in ASCII format.
>>
>> Luca
>>
>> 2016-01-18 14:34 GMT+01:00 Ask Hjorth Larsen <asklarsen at gmail.com>:
>>>
>>> Maybe use a time-dependent external field?  (I don't know so much
>>> about the TDDFT code)
>>>
>>> 2016-01-18 14:19 GMT+01:00 Luca Daumann <luca.daumann at gmail.com>:
>>> > Dear Tuomas
>>> >
>>> > thank you for you kind answer. Anyway to get a spectra one should apply
>>> > a
>>> > kick in a certain direction to the molecular system. The problem is that
>>> > the
>>> > td_calc.absorption_kick does not work with periodic boundary conditions.
>>> > Is
>>> > there another way to deform the initial wfns ?
>>> >
>>> > Thank you
>>> >
>>> > Luca
>>> >
>>> > 2016-01-18 12:53 GMT+01:00 Tuomas Rossi <tuomas.rossi at aalto.fi>:
>>> >>
>>> >> Dear Luca,
>>> >>
>>> >> Yes, real-time propagation with k-points should work out. (I don't have
>>> >> personal experience on such calculations though.)
>>> >>
>>> >> Best regards,
>>> >> Tuomas
>>> >>
>>> >>
>>> >>
>>> >> On 17.01.2016 18:30, Luca Daumann wrote:
>>> >>>
>>> >>> Dear Developers
>>> >>>
>>> >>> is GPAW able to perform  Real-Time-TDDFT calculations in presence of
>>> >>> K-point grids ?
>>> >>>
>>> >>>
>>> >>> Thank you in advance
>>> >>> Best Regards
>>> >>>
>>> >
>>> >
>>> >
>>> >
>>> >
>>> > _______________________________________________
>>> > gpaw-users mailing list
>>> > gpaw-users at listserv.fysik.dtu.dk
>>> > https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>>
>>
>>
>> _______________________________________________
>> gpaw-users mailing list
>> gpaw-users at listserv.fysik.dtu.dk
>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users