LCALCPOL won't work for low band gap materials
Posted: Tue Mar 27, 2012 2:52 pm
Hi all,
I am having trouble with LCALCPOL calculating the macroscopic polarization for low (although non-zero) band gap materials. The strange thing is that it won't even print an error in the OUTCAR.
One odd thing is that the older routine set with LBERRY will indeed calculate the polarization for some of those materials for which LCALCPOL does not do anything, or print an error if it can - i.e. if LDA is predicting a metallic state.
Therefore, I have been using LBERRY for the calculation of the polarization in LDA. However, with HF-type calculations LBERRY does not work: VASP's execution stops when IBZKPT_HF tries to generate the kmesh for the exchange kernel.
So I have two questions:
(1) Is it possible to force LCALCPOL to calculate the polarization for low band gap materials or at least print an error?
(2) Is there a way to use the LBERRY option with hybrid-type calculations?
I am particularly interested in (2).
Thanks
I am having trouble with LCALCPOL calculating the macroscopic polarization for low (although non-zero) band gap materials. The strange thing is that it won't even print an error in the OUTCAR.
One odd thing is that the older routine set with LBERRY will indeed calculate the polarization for some of those materials for which LCALCPOL does not do anything, or print an error if it can - i.e. if LDA is predicting a metallic state.
Therefore, I have been using LBERRY for the calculation of the polarization in LDA. However, with HF-type calculations LBERRY does not work: VASP's execution stops when IBZKPT_HF tries to generate the kmesh for the exchange kernel.
So I have two questions:
(1) Is it possible to force LCALCPOL to calculate the polarization for low band gap materials or at least print an error?
(2) Is there a way to use the LBERRY option with hybrid-type calculations?
I am particularly interested in (2).
Thanks