Top Page > Browsing

Bad convergence for calculations with an electric field Date: 2020/01/13 18:19 Name: Chen Dear developers, I am experiencing bad convergence for calculations with an electric field. I am trying to look at how the structure responses to the external electric field. I understand that I need to a geometry relaxation under the electric field. The way I did is following. I gave a value like 5.0 0 0 to scf.Electric.Field(My system is a ribbon). Then switch on MD.Type. However, I got the convergence problem during the SCF calculation for the first step of geometry relaxation. And it never converged no matter how I modified the mixing parameters and the electronic steps for SCF. Any suggestion will be appreciated. Best regards, M. Chen

Page: [1]

Re: Bad convergence for calculations with an electric field ( No.1 ) Date: 2020/01/16 13:35 Name: Po-Hao  <chang.pohao@gmail.com> Hi I don't know what your system is but I think 5.0 (V/nm?) is way too large. Try start with smaller field like 0.2. Once it's converged, use it as initial state for the higher field. This should work to some extent.

Page: [1]