[gpaw-users] Dipole correction with implicit solvent

[Ask Hjorth Larsen]

Hi

2017-02-05 0:57 GMT+01:00 Georg Kastlunger <georg.kastlunger at univie.ac.at>:
> Hi Ask,
>
> thank you for the fast reply, where you confirm my understanding of the
> routine.
>
> I suppose the subtraction and solvation of the poisson equation(accounting
> for the FDPoissonsolver boundary conditions), and addition (giving two
> distinct vacuum levels) are performed via these two lines:
>
> iters = self.poissonsolver.solve(vHt_g, rhot_g + drhot_g, **kwargs)
> vHt_g += dvHt_g
>
> However, I do currently not understand why dvHt_g  is subtracted from the
> electrostatic potential beforehand. Is it done in order to create a good
> starting guess for the poisson solver? What puzzles me is that, although

Exactly!  I remember adding that line because the Poisson solver was
using a lot of iterations.  The thing is, the initial guess for the
*internal solver* should be a dipole-free charge distribution.
Normally the Poisson solver is happy with the vHt_g from last SCF
step, but in this case the vHt_g already contains a dipole and is
therefore a very bad guess for the dipole-free calculation of the
internal solver.

> dvHt_g is constant at the outer parts of the unit cell the resulting
> potential is sometimes not flat near the boundary, when I scale it and keep
> the correcting charge density drhot_g untouched.

Sorry, I don't quite understand.  Could you explain in more detail please?

Best regards
Ask

>
> When you have more time, I would be happy to here your expertise on that.
> However, don't feel pushed, since I was just wondering how this can happen.
>
> Best wishes,
> Georg
>
>
> On 02/04/2017 09:54 AM, Ask Hjorth Larsen wrote:
>
> Hi! Since the equation is linear, the correction works by adding something
> to remove the dipole, then solving, then re-add the necessary compensation
> to the solution. I am a bit busy but can provide more details later.
>
> Best regards
> Ask
>
>
> El 4 feb. 2017 15:06, "Georg Kastlunger via gpaw-users"
> <gpaw-users at listserv.fysik.dtu.dk> escribió:
>>
>>
>> Dear gpaw community,
>>
>> I am currently working on a dipole correction combined with the provided
>> implicit solvent model in GPAW. This has proven to be quite a challenge
>> in terms of electrostatics and I am afraid I am not able to capture the
>> physics correctly, unless I can understand the routine for the dipole
>> correction completely.
>>
>> Therefore, I was wondering if there is any literature reference which
>> explains the reasoning behind the dipole correction in fd mode.
>>
>> One question upfront:  Why is the potential representing the dipole
>> contribution (dvHt_g) subtracted before the poisson equation is solved?
>> Does that mean that the starting guess for the potential is changed?  I
>> noticed that if I change dvHt_g it changes the slope at the outer parts
>> of the resulting electrostatic potential, even though dvHt_g is always
>> constant in those regions.
>>
>> Best wishes,
>> Georg
>>
>> _______________________________________________
>> gpaw-users mailing list
>> gpaw-users at listserv.fysik.dtu.dk
>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
>