[Pw_forum] "Centering" Requirement of ESM

[Minoru Otani]

Dear Robert,

Let me start from the definition of the lattice vector.As defined in the manual:
http://www.quantum-espresso.org/wp-content/uploads/Doc/INPUT_PW.html#idm140629871958288
The crystal lattice vector is defined by the following expression:
 v1(1) 	 v1(2) 	 v1(3) 	... 1st lattice vector
 v2(1) 	 v2(2) 	 v2(3) 	... 2nd lattice vector
 v3(1) 	 v3(2) 	 v3(3).      ... 3rd lattice vector

> Requires cell with a_3 lattice vector along z, normal to the xy plane, with the slab centered around z=0


This sentence means that the inner product both v1 dot v3 and v2 dot v3 must be zero. Current implementation of the ESM does not allow to have finite values in v1(3), v2(3), v3(1), and v3(2).

>  Also requires symmetry checking to be disabled along z, either by setting nosym = .TRUE.

>  or by very slight displacement (i.e., 5e-4 a.u.) of the slab along z.

This requirement is necessary for bc2 and bc3. When we apply a bias potential to the slab, we break the symmetry along z-direction. However, if we put a symmetric (along z) slab at z=0, the pw.x find a symmetry operation in z-direction and try to symmetrize wave functions and potentials along z. You can not get a converged result with this symmetrize process. Thus I proposed two ways to avoid the symmetrize process along z:
The first one is to set “nosym = .TRUE.”. Note that this kills the symmetrize process not only the z-direction but also xy-direction.
The second one is to give a very slight displacement to the slab along z-direction. By doing this pw.x can not find any symmetry operation along z-direction but can find a symmetry operation along xy-direction.

If you want to calculate non-symmetric (along z) slab or you put a slab at z!=0, you can ignore this.

Best regards,
Minoru
--------------------------------------------------------------------------------------------------------
Minoru Otani
National Institute of Advanced Industrial Science and Technology
Research Centre for Computational Design of Advanced Functional Materials
email : minoru.otani at aist.go.jp <mailto:minoru.otani at aist.go.jp>
---------------------------------------------------------------------------------------------------------

> On Jul 29, 2017, at 21:13, Robert Molt <r.molt.chemical.physics at gmail.com> wrote:
> 
> I appreciate the help. If ESM does not require this symmetry, can you explain what is meant by the manual's statement of:
> :
> "Requires cell with a_3 lattice vector along z,
>      normal to the xy plane, with the slab centered
>      around z=0
> . Also requires symmetry checking to be
>      disabled along z, either by setting 
> nosym
>  = .TRUE.
>      or by very slight displacement (i.e., 5e-4 a.u.)
>      of the slab along z."
> 
> 
> 
> I want to make sure I am in compliance with the manual's specifications, and I clearly must not understand what that means if I misinterpreted it to mean it requires symmetry along the z axis.
> 
> Separately, given that we have no symmetry in xy, is it logically necessary to "cap" boundary atoms with hydrogens (or the like) to avoid artifacts at the boundary?
> 
> On 7/28/17 6:41 PM, 大谷実 wrote:
>> Dear Robert,
>> 
>> Besides Lorenzo’s message, I can add one more thing. ESM does not require symmetry along z-axis, so you do not need to put an additional adsorbate molecule on the bottom. 
>> 
>>> Also requires symmetry checking to be
>>>      disabled along z, either by setting 
>>> nosym
>>>  = .TRUE.
>>>      or by very slight displacement (i.e., 5e-4 a.u.)
>>>      of the slab along z.
>>> 
>> 
>> 
>> Best regard,
>> Minoru
>> --------------------------------------------------------------------------------------------------------
>> Minoru Otani
>> National Institute of Advanced Industrial Science and Technology
>> Research Centre for Computational Design of Advanced Functional Materials
>> email : minoru.otani at aist.go.jp
>> ---------------------------------------------------------------------------------------------------------
>> 
>>> On Jul 29, 2017, at 5:15, Lorenzo Paulatto <lorenzo.paulatto at impmc.upmc.fr> wrote:
>>> 
>>> It does not matter, it is only need to be roughly centered 
>>> 
>>> -- 
>>> Lorenzo Paulatto
>>> Written on a virtual keyboard with real fingers
>>> 
>>> On 28 Jul 2017 9:36 p.m., "Dr. Robert Molt Jr." <r.molt.chemical.physics at gmail.com> wrote:
>>> Good evening,
>>> I am confused by the "centering" requirements of ESM. It is stated that:
>>> Requires cell with a_3 lattice vector along z,
>>>      normal to the xy plane, with the slab centered
>>>      around z=0. Also requires symmetry checking to be
>>>      disabled along z, either by setting 
>>> nosym
>>>  = .TRUE.
>>>      or by very slight displacement (i.e., 5e-4 a.u.)
>>>      of the slab along z.
>>> 
>>> I have constructed a slab which is symmetric about the xy plane and centered around z=0 contour. This is fine and good.
>>> However, once I add the adsorbate molecule, I cannot have symmetry...unless it is intended that I add the same adsorbate molecule on the "bottom" of the simulation, too? I am unclear what is meant by the slab being "centered." Center of mass? Center of symmetry?
>>> 
>>> 
>>> _______________________________________________
>>> Pw_forum mailing list
>>> Pw_forum at pwscf.org
>>> http://pwscf.org/mailman/listinfo/pw_forum
>>> _______________________________________________
>>> Pw_forum mailing list
>>> Pw_forum at pwscf.org
>>> http://pwscf.org/mailman/listinfo/pw_forum
>> 
>> 
>> 
>> _______________________________________________
>> Pw_forum mailing list
>> 
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170729/a4d28502/attachment.html>