[CP2K-user] [CP2K:15456] Re: Local potential for potential alignment in charged systems

[Matt W]

Would depend what you need to apply your chosen scheme. 
Pretty much by definition you can't get the potential accurately on a 
regular grid near all-electron cores.

On Tuesday, May 25, 2021 at 2:32:57 AM UTC+1 n... at berkeley.edu wrote:

> A developer might be able to comment more thoroughly, but I believe 
> you have to post-process the electron density in order to do this. It is 
> very rare to do defect calculations with all-electron calculations, so 
> unless you have a good reason, consider switching back to GPW.
>
> On Mon, May 24, 2021 at 5:36 PM ma... at gmail.com <ma... at gmail.com> 
> wrote:
>
>> Hi Nick,
>>
>> Yes, this is very helpful. But another issue is that V_HARTREE_CUBE is 
>> only valid for QS with GPW which is mentioned in the manual... I'm 
>> wondering is there any approach to print the elctrostatic potential for 
>> ALL_ELECTRON basis sets with GAPW?
>>
>> Regards,
>> Hongyang
>>
>> 在2021年5月25日星期二 UTC+10 上午10:29:33<n... at berkeley.edu> 写道：
>>
>>> For potential alignment corrections, one should use the electrostatic 
>>> potential in the V_HARTREE cube file. 
>>>
>>> Many potential alignment correction schemes exist, but I am partial to 
>>> the one by Freysoldt (https://doi.org/10.1103/PhysRevLett.102.016402) 
>>> and its extension to anisotropic systems by Kumagi (
>>> https://doi.org/10.1103/PhysRevB.89.195205), which are the two I use 
>>> for charged defect calculations.
>>>
>>> Is this what you need to know, Hongyang?
>>>
>>> -Nick
>>>
>>> On Monday, May 24, 2021 at 4:22:35 PM UTC-7 ma... at gmail.com wrote:
>>>
>>>> Hi ub,
>>>>
>>>> I'm wondering have you figured out the method to solve the potential 
>>>> alignment correction?
>>>>
>>>> Regards,
>>>> Hongyang
>>>>
>>>> 在2020年2月7日星期五 UTC+11 下午7:35:46<ub> 写道：
>>>>
>>>>> Thanks for the article Matt. I'll go through it to see any 
>>>>> similarities to the issues faced.
>>>>>
>>>>> Regards,
>>>>> ub
>>>>>
>>>>>
>>>>> On Thursday, 6 February 2020 21:11:29 UTC+1, Matt W wrote:
>>>>>>
>>>>>> <self promotion>
>>>>>> Relation between image charge and potential alignment corrections for 
>>>>>> charged defects in periodic boundary conditions
>>>>>> TR Durrant, ST Murphy, MB Watkins, AL Shluger
>>>>>> The Journal of chemical physics 149 (2), 024103 (2019)
>>>>>> </self promotion>
>>>>>>
>>>>>> might be useful. You could contact Tom Durrant 
>>>>>> <https://www.ucl.ac.uk/physics-astronomy/people/iris-profile-thomas-durrant> 
>>>>>> about what he has been doing. There is a load of other stuff in the 
>>>>>> literature, of course.
>>>>>>
>>>>>> Matt
>>>>>>
>>>>>> On Thursday, February 6, 2020 at 6:17:54 PM UTC, Travis wrote:
>>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> To your last question - the authors report the plane-averaged 
>>>>>>> potential (refer to the caption of Fig 2 in your linked paper), not the 
>>>>>>> actual potential. The actual potential should be noisy as you described.
>>>>>>>
>>>>>>> -T
>>>>>>>
>>>>>>> On Thursday, February 6, 2020 at 2:00:32 PM UTC-4, Udit wrote:
>>>>>>>>
>>>>>>>> Hello world,
>>>>>>>>
>>>>>>>> As part of my research, I'm performing simulation of charged 
>>>>>>>> dopants in Silicon for calculation of the charge transition levels. One 
>>>>>>>> crucial step in the formation energy estimation is the determination of an 
>>>>>>>> alignment constant which is determined by comparing the local potential 
>>>>>>>> profiles of the defective and pristine supercells. 
>>>>>>>> https://journals.aps.org/prl/pdf/10.1103/PhysRevLett.102.016402 (Fig. 
>>>>>>>> 2)
>>>>>>>>
>>>>>>>> My question is, for the local potential profiles is V_HARTREE_CUBE 
>>>>>>>> sufficient? A lot of python codes designed for such calculations do so by 
>>>>>>>> evaluating the LOCPOT obtained from VASP calculations. Is V_HARTREE_CUBE 
>>>>>>>> similar to the LOCPOT inVASP? 
>>>>>>>>
>>>>>>>> Also, a lot of these publications during alignment show smoothly 
>>>>>>>> varying local potential profiles. However, I find that the Hartree 
>>>>>>>> potential profile of the defective supercell varies a lot from that of the 
>>>>>>>> pristine supercell and contains a lot of oscilattions mainly around the 
>>>>>>>> defect due to the relaxation of atomic positions, which would potentially 
>>>>>>>> make the alignment procedure impractical. Any comments or experiences with 
>>>>>>>> Charge Transition Level calculations on condensed matter systems would be 
>>>>>>>> greatly appreciated!
>>>>>>>>
>>>>>>>> Thanks.
>>>>>>>>
>>>>>>> -- 
>> 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/ff485f75-3d2d-4173-a9a7-15dbf9e4d2dbn%40googlegroups.com 
>> <https://groups.google.com/d/msgid/cp2k/ff485f75-3d2d-4173-a9a7-15dbf9e4d2dbn%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210525/6a5f40dc/attachment.htm>