During optimization, sometimes we obtained the warning message given below:
Warning: There are no empty bands for at least one k-point and spin; this may slow the convergence and/or lead to an inaccurate groundstate. If this warning persists, you should consider increasing nextra_bands and/or reducing smearing_width in the param file. Recommend using nextra_bands of 7 to 15.
How can we overcome this problem although finally shows: BFGS : Geometry optimization completed successfully?
I've experienced also this type of problems during optimization. I was working on strongly correlated electronic system. To solve the problems, I took a few steps. I changed the pseudopotential to Norm conserving and electronic minimizer to density mixing. Finally, it was optimized succesfully. The U parameter with ultrasoft pseudopotential was taken for some optimization.
For correlated electronic system, the more realistic result like very close to experimental values was attained through employing Hubbard U parameter.
Thank you very much.
Thank you Ali Hossain for addressing the issue. Stay safe by the grace of almighty Allah.
Best wishes,
M.A. Hadi
Dear M. A. Hadi Sir,
It's my pleasure. Thank you too.
Best Regards
Ali Hossain