UT theoretical chemistry code forum

Hi guys,

I am calculating transition state using vtst code in VASP. How many images should I set up for CI-NEB calculation?

At first I tried 3 images, but it didn't converge. Then I tried only 1 image, it converged but there was not an energy barrier, e.g. it gives a barrierless pathway. How accountable is this result? Can I further add image between the initail state and the negative "transition state" and go on calculating to get more accurate results?

Thank you in advance for any help you can provide!

yat

I think this is a very general question. One way is to see the distance between your initial and final ones and then see how many images can be considered as proper. I usually use 4, 6 or 8 depending on the nature of my calculations.

Cheers