[QE-developers] nonlinear core correction in phonon codes

[Jae-Mo Lihm]

Dear Dr. Hyungjun Lee and All,

I also had the same question while working on the interpolation of phonon
potential. And I think I found an answer, at least partially.

The contribution of the nonlinear core correction is added to the variable
dvscf in the following part of solve_linter.f90:
        if (imode0+ipert > 0) then
           call addcore (imode0+ipert, drhoc)
        else
           drhoc(:) = (0.0_DP,0.0_DP)
        endif
        !
        ! Compute the response HXC potential
        call dv_of_drho (dvscfout(1,1,ipert), .true., drhoc)

So, in most parts of ph.x, multiplication of psi with dvscf gives the
induced part *plus *the NLCC contribution, while dvqpsi_us gives the bare
part *except *NLCC.
In contrast, in zstar_eu, only the bare potential perturbation for atomic
displacement is needed. Hence, the NLCC contribution is added by dvqpsi_us.

I hope this answers your question.

Though, I would still like to know why the ph.x code was designed this way.

Regards,
Jae-Mo Lihm
Seoul National University

2020년 3월 28일 (토) 오전 12:50, H. Lee <hjunlee at gmail.com>님이 작성:

> Dear Prof. Paolo Giannozzi:
>
> Thank you for your reply.
>
> Sincerely,
>
> Hyungjun Lee
>
> On Fri, Mar 27, 2020 at 3:44 AM Paolo Giannozzi <p.giannozzi at gmail.com>
> wrote:
>
>> Sorry for the sloppy last sentence: "I didn't succeed".
>>
>> On Fri, Mar 27, 2020 at 9:42 AM Paolo Giannozzi <p.giannozzi at gmail.com>
>> wrote:
>>
>>> On Thu, Mar 26, 2020 at 4:36 PM H. Lee <hjunlee at gmail.com> wrote:
>>>
>>>> Dear all:
>>>>
>>>> I have one simple and silly question on the nonlinear core correction
>>>> in phonon codes.
>>>>
>>>
>>> your question is simple but not silly at all. Some years ago I tried to
>>> figure out where the missing core correction term was added. I didn't
>>> manage succeed.
>>>
>>> Paolo
>>>
>>>
>>>> The subroutine of dvqpsi_us takes care of the contribution of the local
>>>> part of pseudopotential to dvscf.
>>>>
>>>> However, in most parts except for zstar_eu.f90, when calling dvqpsi_us
>>>> the contribution due to the nonlinear core correction is not included; the
>>>> third argument of dvqpsi_us controls the inclusion of this correction.
>>>>
>>>> elphon.f90:           CALL *dvqpsi_us* (ik, u (1, mode), .FALSE.,
>>>> becp1, alphap)
>>>>
>>>> ep_matrix_element_wannier.f90:           CALL *dvqpsi_us* (ik, u (1,
>>>> mode), .FALSE., becp1, alphap)
>>>>
>>>> raman_mat.f90:        call *dvqpsi_us* (ik, uact (1, imod),.false.,
>>>> becp1, alphap)
>>>>
>>>> raman_mat.f90:           call *dvqpsi_us* (ik, uact (1, imod),.false.,
>>>> becp1, alphap )
>>>>
>>>> solve_linter.f90:                    call *dvqpsi_us* (ik, u (1,
>>>> mode),.false., becp1, alphap )
>>>>
>>>> solve_linter.f90:                    call *dvqpsi_us* (ik, u (1,
>>>> mode),.false., becpt, alphapt)
>>>>
>>>> zstar_eu.f90:           call *dvqpsi_us* (ik, u (1, mode), .not.okvan,
>>>> becp1, alphap)
>>>>
>>>> Is there a specific reason to not include this correction?
>>>> Or is this correction added in other parts of the code?
>>>>
>>>> Sincerely,
>>>>
>>>> Hyungjun Lee
>>>> _______________________________________________
>>>> developers mailing list
>>>> developers at lists.quantum-espresso.org
>>>> https://lists.quantum-espresso.org/mailman/listinfo/developers
>>>>
>>>
>>>
>>> --
>>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>>> Phone +39-0432-558216, fax +39-0432-558222
>>>
>>>
>>
>> --
>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>> Phone +39-0432-558216, fax +39-0432-558222
>>
>> _______________________________________________
>> developers mailing list
>> developers at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/developers
>>
> _______________________________________________
> developers mailing list
> developers at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20200330/00156114/attachment.html>