QuantumATK Forum

QuantumATK => Scripts, Tutorials and Applications => Topic started by: xanh on October 17, 2010, 17:53

Title: How to charge a system NiO with oxygen vacancy
Post by: xanh on October 17, 2010, 17:53

I have a question which would like to get suggest from all you.
I follow the guide from tutorial for electronic structure of NiO using LDA . I have a NiO cell , i want take out one oxygen ion (O-2) therefore the NiO system will be charge +2 . However , I dont find out the way to charge the system to simulate the bandgap .

Could any one can help me in this issue? thank you for your help

Title: Re: How to charge a system NiO with oxygen vacancy
Post by: Anders Blom on October 20, 2010, 14:39

From a purely technical standpoint, this is obtained by adding the keyword "charge" to the LCAOCalculator(). Like so:

Code

calculator = LCAOCalculator(charge=-1)

This cannot, currently, not be done from within VNL, you have to edit the script manually.

You will need to set "multipole" boundary conditions to handle a charged system.

Numerically, it is often much harder to converge charged systems than neutral ones, so expect to use a lot of k-points, and perhaps a high temperature, at least to start with.

As for the physics, one has to be aware that the model you compute is in reality a crystal of ions, since we don't add any background compensating charge. Therefore the best idea is probably to make a large supercell, so that the net charge is small compared to the total amount of electrons.

Good luck!