Saikat,

Your issue here is that you are using the default value of:

MD.MaxForceTol        0.04 eV/Ang

because you are not using it in your fdf file. For what you are trying to
accomplish, I think a MD.MaxForceTol of about 0.001 eV/Ang is more
appropriate.This is a good target for small molecules. Try using far fewer
MD.NumCGsteps too, I usually use about 30-50 at a time so that if something
goes wrong you can trace it without waiting forever, a good practice if
you're moving from small molecules to molecules adsorbed onto a surface or
something. MaxSCFiterations is very high as well, should something
unfortunate happen with your convergence. It won't affect anything here,
though. You can also use:

UseSaveData             .true.

To restart simulations that go to the full # of CG steps using saved
density matrices and atomic positions, and recent CG history.

The reason why you were seeing negative frequencies is because your
geometry optimization is not finding the global minimum, so that when atoms
are displaced in the FC step by very small amounts, it might find a lower
energy state. This topic is discussed at length elsewhere on the list.

-- 
Abraham Hmiel
Katherine Belz Groves Fellow in Nanoscience
Xue Group, College of Nanoscale Science and Engineering at SUNY Albany

Responder a