[Pw_forum] NEB procedure

[Ettore Baldini-Neto]

Dear users,

I have some questions regarding a NEB calculation.

As input to a neb file I should use, at least, the initial and final positions (both well previously relaxed).
What I want is to calculated the energy barrier for the transition of a metal from one stable position (let's say a hollow position on a C ring) to another one (also stable but with slightly higher total energy than first, let's say the Top position). 

When I use just the initial and final positions and perform the calculations, asking the code to automatically divide the grid, I get an ascendent curve form the first to the last as expected.
But at a position just before the Top position I also expect the energy barrier to be a little higher than the energy at this final site (Top) because the metal is at a unstable position.

So here are is my question, a very basic one since I have no previous experience with this kind of calculation.

I can put more intermediate images, automatically or by hand, in order to better mapping the region around this position (Top). 
In the second case, should these intermediate images be previously optimized or the neb does it for me??

Thanks in advance,


***************************************************************************************
Ettore Baldini-Neto
Researcher at W. von Braun Center for Advanced Research, Campinas, Brazil
***************************************************************************************