CI-NEB in DFT+D2

[kai]

Dear all,

I am wondering if anyone has ever tried to run CI-NEB calculations with the DFT-D2 method (Grimme's vdW method);

It appears that i am struggling with this method to run an CI-NEB to a very satisfying way; the force could only get down to 0.1 eV/A. And all the tricks i had based on standard DFT to run NEB is not working really.

My question:

Is there anything else I need to pay attention to when i switch on the LVDW= .TRUE.?
Is there any test has been done to see what algorithem/opitimizer is the more efficient one?

[kai]

a quick post note: My favourite optimizer in standard DFT, FIRE is behaving rather strange with the DFT+D2; the austin versioned quick min seems to work reasonably