Electrode for two parallel wires

[hellboy]

Dear Anders Blom,

I want to pass current through two nanotubes that are kept  in parallel. Will it be ok if i use a two probe system to define left and right electrodes for one nanotube and then put another nanotube in parallel between the same electrodes? Will the dangling-bonds of carbon atom provide an electrode surface connection? OR do i need to edit the script and add the Carbon atoms to the left and right surfaces and also modify central region?

Your valuable comments/suggestions are required.

[Anders Blom]

I guess it depends a bit on what you actually are looking for. If you simple have 2 nanotubes, which are not interacting, then you only need one anyway, as they will be equal and equivalent

And even if you are looking for some interaction between then, because they are very close in distance (in X/Y), then you just make a small vacuum cell around the tube, but still calculate only one.

The case where you will need to put two tubes in the simulation cell is of course more interesting, and indicates that the tubes are not equivalent, for some reason. In that case, it depends on whether you want both tubes to carry a current caused by a bias between the reservoirs (the electrodes), or you want to induce some charge redistribution in a nanotube stump placed in the proximity of another current-carrying tube.

Only the last case corresponds to the idea of having just an extra tube piece in the central region. Otherwise you will need to extend the tube into the electrodes.

In the end, the system you calculate should correspond to the reality. If the "second tube" is finite, you model it as a finite piece. If not (i.e. if it's a least much longer than the central region), you must make it infinite from ATK's perspective by extending it into the electrode.

[hellboy]

Dear Anders Blom,

Thanks for your expert suggestions.

I find that my case is, when the two tubes are "equal in length" and "both tubes carry a current" caused by a bias between the electrodes. The two tubes are placed so close that there is an "interaction (induced charges) between them".

How do i define electrodes for the second nanotube in this situation?

waiting for your expert comments... 

[Anders Blom]

There are no separate electrodes for the other tube, so you just have to make a repeated supercell, basically, with 2 tubes, then add some vacuum outside them.
´

[hellboy]

Can you say how do we add vacuum outside the tubes?
Do we have vacuum inside the nanotube?

[hellboy]

Dear Dr Blom,

I am unable to find a way to add vacuum outside the nanotubes. I searched the tutorials and the forum but couldn't get the solution. Kindly help me out with this situation. Also does ATK asumes vacuum inside the nanotube while doing two probe calculations?

[Anders Blom]

"Vacuum" is perhaps the wrong word to use, sorry if it caused confusion. What I meant was to extend the unit cell to create some space between the repeated copies (periodic boundary conditions normally apply in X and Y).

You only need to put in the atoms in ATK, that's all you need to worry about. The middle of tube and everywhere else will be treated correctly as long as the atoms are in their right places (but sure, you can think of it as vacuum where there are no atoms).

[hellboy]

Thank you Dr Blom.

[hellboy]

Dear Dr Blom

I have a small query..

For a system with a 5,5 nanotube parallel to 4,4 nanotube I obtained left and right surfaces, electrodes and central region. I started with 5,5 tube then cleaved and dropped 5,5 again then i made a 4,4 nanotube in parallel to 5,5 and gave some vacuum between them.

The probelem with the script is that whole (4,4) nanotube appears in the central region which means 4,4 tube is not in direct contact with electrodes  . While 5,5 is correctly having electrodes, surfaces and central region.

For this parallel nanotube (4,4) to conduct current, can I use the atoms and their cordinates of left and right surfaces/electrodes of 4,4 tube and add them to the extracted surfaces/electrodes of the whole system? And also modify the central region?

Will it work?


witing for your expert comments.

[Anders Blom]

Picture, please

[hellboy]

Dear Sir,

Please see the picture and script attached.

[hellboy]

I also found that "Equivalent atoms" need to be defined for setting the two probe geometry since i am using ATK 2008.10.
That has also been discussed in this forum http://quantumwise.com/forum/index.php?topic=691.0 .
Is there an easy way to find equivalent atoms?

[Anders Blom]

Ok, right! In principle, if you've done it all right, you can leave them at default. That is, if your atoms are ordered in the same way in the central region and the electrode, then the general rule applies: the first atom in the left electrode is equivalent to the first in the central region, and the last in the central region is equivalent to the last in the right electrode. That is,

Code

equivalent_atoms = [[0,0],[-1,-1]]

which is the default.

[hellboy]

Ok, I understood.

But I am not sure about the situation presented in my previous post where i have also attached picture and script.

Can you please look at that and comment?

Also in case of a single SWCNT equivalent atoms are [0,0] [1,56].
When i use repetition along 'A' equivalent atoms are [0,0] [2,48]
when i use repetition along both 'A' and 'B' Equivalent atoms are [0,0],[4,96]

I cannot understand, please clarify if possible.