structural relaxation under finite bias using transiesta

Asked by Lingling Tao on 2019-02-09

Hi all,

I want to perform structural relaxation under finite bias using transiesta. The published work ACS Nano, 6, 1473 (2012) shows that the structural relaxation under finite bias can indeed be done by siesta/transiesta. However, I can not find any clues from the siesta manual. Can anyone tell me how to perform structural relaxation under finite bias using transiesta?

I appreciate any response and suggestions.

Question information

Language:
English Edit question
Status:
Solved
For:
Siesta Edit question
Assignee:
No assignee Edit question
Solved by:
Lingling Tao
Solved:
2019-02-14
Last query:
2019-02-14
Last reply:
2019-02-11
Maria (maria11) said : #1

For the pseudopotentiels in the website of siesta (no relativistic), We can do a spinpolarised calculation with a non relativistic pseudopotentiel?

Nick Papior (nickpapior) said : #2

In principle there isn't anything special you need to do. Simply use the MD.TypeOfRun tag for the MD algorithm, and SolutionMethod transiesta for TS.

However, performing relaxation using TranSiesta is about as hard as it can get. One problem is that the "Fermi-level" will change when one shifts the atomic positions. Essentially you need to ensure that the amount of charge is "steady" along the relaxation path (in 4.1 the dq column shows this number).

To stabilize such a calculation you may add more electrode layers to create a longer screening length, this also stabilizes the above problem.

All-in-all, you need to perform such relaxations with extreme care. Good luck!

Lingling Tao (lltao) said : #4

Thank you Nick! This solved my problem.