This thread is locked.Only browsing is available.
Top Page > Browsing
PAOs of lead atoms can overlap only to the next nearest region
Date: 2016/07/01 21:36
Name: Vipin Kumar

Dear openmx users,

I am running left lead calculation for 2D ribbon, I am getting the following error:


TRAN_Check_Region_Lead()

The length between atomA=5 and atomB=2 is too short for the transport calculation.
distance=12.661164 rcutA=7.000000 rcutB=7.000000

PAOs of lead atoms can overlap only to the next nearest region.

I did not understand what does it mean, can anyone help me. Your suggestions may be helpful.

Thanking you.

Regards,

Vipin
メンテ
Page: [1]

Re: PAOs of lead atoms can overlap only to the next nearest region ( No.1 )
Date: 2016/07/04 10:15
Name: T. Ozaki

Hi,

Inverting matrices in NEGF requires a sparse structure of matrices, which
is determined by cutoff radii of basis functions. The requirement is explained at
http://www.openmx-square.org/openmx_man3.8/node116.html

Regards,

TO
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.2 )
Date: 2016/07/05 20:23
Name: Vipin Kumar

Dear sir,

I am new user of Openmx and trying to do some transport calculations for 2D materials. Before going on new work using this code I want to reproduce the results of your paper (PHYSICAL REVIEW B 81, 075422 2010) using code Openmx. I am going through your paper where you have shown the transport calculations of 8-ZGNR using openmx code. I also have confusion in the definition of 8-zgnr. When defining the zgnr we count only the outer atoms on the zz edges as given in (PHYSICAL REVIEW B, VOLUME 54, NUMBER 24, 17954)and as also given on wikipedia page of graphene nanoribbon in the animated structures. I want to ask that these calculations are really done in case of ribbon or considering the patch of ribbon because you are passivating the dangling carbon bonds on zigzag edges so it seems like a patch of ribbon. I have taken a look on the input files of 'negf-example' in the directory 'work'. I have regenerated the structure taking the co-ordinates of 'left, right and central lead'. In 'negf' file the central region has only 10 atoms, however, in the structure FIG. 1(a) in the above your paper the central region has more than 10 atoms. So how are you defining the left, right lead and the central region. The facility of sending the pictures is not given with this forum so the concerned structures I am sending in your email. It is my humble request you to please explain my doubts as I am greatly depending on these because we want to use this code in my group for transport calculations.

Thanking you.

Regards,

Vipin
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.3 )
Date: 2016/07/06 19:26
Name: Artem Pulkin

Take a look at the first figure of

https://www.researchgate.net/publication/234919032_Graphene-based_spin_logic_gates

There, all parts of NEGF setup are highlighted. The lead is defined as a smallest periodic unit (armchair) of what is on the left or on the right. It contains ~18 atoms, vacuum in vertical direction and vacuum perpendicular to the plane.

However, if you receive messages like the first one in this thread, you probably have to consider a supercell of your lead in horizontal direction with number of atoms doubled. This has nothing to do with physics, it is just a limitation of OpenMX (and any other NEGF code).
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.4 )
Date: 2016/07/11 21:48
Name: Vipin Kumar

Dear Pulkin,

Thanks for your reply. I have one more doubt. I have gone through several papers on transport properties. In all cases everybody is discussing nanoribbon with hydogen passivation at either end. I want to ask that if the hydrogen passivation is done on either ends then how it is ribbon rather it looks like a patch of ribbon, and what is the reason behind hydrogen passivation.

Thank you in advance.

Regards,

Vipin
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.5 )
Date: 2016/07/11 22:04
Name: Artem Pulkin

Nanoribbon is a 1D object, infinite along its edge. Patch is a 0D object, it has a finite size. Passivation makes something chemically passive: in this case dangling bonds of carbon on TWO PARALLEL edges are passivated by hydrogen. Otherwise reactive systems present challenge for those who make devices on a chip.
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.6 )
Date: 2016/07/11 23:52
Name: Vipin Kumar

Dear Pulkin,

Thanks for your quick reply. Can you please explain me how it is ribbon because in calculation we take finite length of left, right and central region. I am very highly confused with this.

メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.7 )
Date: 2016/07/12 18:37
Name: Artem Pulkin

In the calculation we involve the scattering (device) region and UNIT CELLs of the leads involved. The leads are infinite but the unit cell is finite (i.e. 18 atoms). Similarly to calculation of the solid we include a UNIT CELL of the solid. Keywords: translation invariance, periodic boundary conditions, Bloch theorem, tight binding band structure.
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.8 )
Date: 2016/07/13 13:27
Name: Vipin Kumar

Dear sir,

Thank you to give me clarification. Why are we doing edge passivation only by Hydrogen. Can we use Hydrogen like atoms having single valance electron or other periodic elements.


メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.9 )
Date: 2016/07/13 18:46
Name: Artem Pulkin

I guess you can do that, look for the literature.
メンテ
Re: PAOs of lead atoms can overlap only to the next nearest region ( No.10 )
Date: 2016/07/18 20:18
Name: Vipin Kumar

Dear Sir,

Is there any text limit for posting the message in the forum. Whenever I am trying to post on forum by creating a new theme, it is showing ERROR. What is the problem I am not able to figure out. I am trying it since two-three days, and getting the ERROR message. I need your suggestions on it.

メンテ

Page: [1]