UT theoretical chemistry code forum

Hi All,

Recently, I have been trying to converge an NEB calculation that involves the rotation of a molecule over a surface. Because the NEB calculation was slow to converge (I think it is because the reaction coordinate was rather long--about 12 angstroms), I tried using the dimer method to optimize the highest energy image from this (not-yet-finished) NEB run.

I was able to successfully optimize the structure and I am currently calculating the frequencies to ensure that the optimized structure is indeed at the saddle point. However, I was wondering if there was a way in vasp to double check to make sure that this TS is part of the intrinsic reaction coordinate between the initial and final images of the NEB run. I know in gaussian, there is a way to do this but I'm not so sure in vasp.

Thank you!

Yes, what you can do is make a small displacement along both directions of the unstable mode and minimize those structures. We have a small script, dimmins.pl which should set up those minimization calculations from a converged dimer calculation.