Siesta

TranSIESTA: Bulk Electrode Repetition in x and y

Asked by Tristan Zaborniak

Based on what I can gather, Transiesta requires the use of bulk electrodes. If this is the case, is it necessary in defining the unit cells of the scattering region and electrodes that they be of such dimensions that in being repeated in x and y the electrode regions are 'continuous' - i.e. that the spacing between repetitions is the same as in the bulk of the electrode?

If I've not made myself clear, the following will hopefully serve to illustrate. Suppose I am working with Au (1 1 1) electrodes. Should I make sure their unit cells such that if they are stacked in x and y I would see an unbroken Au (1 1 1) crystal? Would I see different results if I made the unit cell far bigger in x and y, so that there would be large spaces between the repeated electrode structure?

And, finally, if I'm wrong about Transiesta requiring bulk electrodes, is it true that the unit cell dimensions in x and y are arbitrary, so long as they contain fully the electrode and scattering regions?

Thanks!

Question information

Language:
English Edit question
Status:
Solved
For:
Siesta Edit question
Assignee:
No assignee Edit question
Solved by:
Nick Papior
Solved:
Last query:
Last reply:
Revision history for this message
Best Nick Papior (nickpapior) said : 		#1

The *ONLY* requirements the electrodes have are:

1) They should be bulk-like along their respective semi-infinite directions.
2) they should only couple to their first nearest neighbour cells.
3) IFF your electrode is bulk transverse to its semi-infinite direction, then the scattering region calculation must have the same periodicity (bulk property) in the transverse direction as well.

Chains (which are not bulk in x or y) are of course viable electrodes.

If you want bulk electrodes in x + y directions then there should be no vacuum in those directions.
And YES (see 3), electrodes should share the same electronic structure in the scattering region where the electrode enters.

Revision history for this message
Tristan Zaborniak (tristanz) said : 		#2

Thanks Nick Papior, that solved my question.