Dear users
I am trying to perform the group theory analysis for transition metal - vacancy complexes in diamond. I am using HSE06 level of theory to get the eigenvalues of defect related states in the band gap. I noticed that in many cases the symmetry of the defects does not correspond to the symmetry of pseudo wave functions i am looking for. As a result, the prediction of group theory does not match to the calculated eigenvalues. Any idea how to bring the symmetry of pseudo wave functions to the symmetry of defect in HSE06 using VASP engine ?
The second question is how to distinguish the defect states which fall below VBM ?
The second part of the question is easiest to answer, so let's start with that:
You can use the projected DOS (or LDOS) to see where the transition-metal states are located. If you want really detailed information, you can ask VASP to write the PROCAR file, and inspect it to find out which bands have a significant characteristic of your transition metal.
The first question I am not really sure this is even possible as
1) VASP uses the HKS-formalism, in which you are optimizing the density. The wavefunction VASP obtains is coming from the density, not the other way around, so I am not sure how well it behaves if you think of this wave-function as one normally does in QM.
2) I assume you are not working with a cluster but under periodic boundary conditions. Doesn't group theory assume you have a molecule/cluster, i.e. gives you the symmetry at the Gamma-point only?
Best,
Danny
Thanks for your response Danny ! At the beginning i wanna tell that i am conscious of the limitation of HKS-formalism, specifically when it comes to the pseudo wave functions. In principle, the group theory is mostly dedicated for molecules/ clusters but together with the crystal field theory it can be applied to the defects in wide band-gap semiconductors, where high degree of defect-related states localization may be expected. You can imagine such a complex, let's say Ti-V submerged in the spin bath of carbon sp3 electrons. I am using periodic DFT and large supercell with gamma point approximation only.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators