Peculiar application of NEB

[cbbasak]

1. I have a known parent (initial) structure and a known product (final) structure; i.e. lattice parameters and atom positions are known for both.

2. Energy of the final structure is always slightly higher than the energy of the initial structure.

3. During the transition:
i. Initial and final structure has same lattice parameters but different atomic coordiantes (all are known).
ii. All intermediate structure/image has same lattice volume as that of initial and final structure (volume is known).
iii. Any given intermediate structure/ image has a pre-defined (i.e. known) lattice parameters but UNKNOWN atomic positions.

4. At this point one may think that this is a straightforward case of internal parameter (i.e. atom position) optimization (since lattice parameters are known for all structures). But the point is that I want my initial structure to go to the pre-defined end-structure which has a higher energy. So, a straightforward atom position optimization would end up with a structure exactly same as that of the initial structure having lower energy (since both have the same lattice parameters).

In a nutshell, is it possible to apply NEB only in terms of atom positions ?

It will be of great help for me if you could comment on this problem if such condition can be tackled with NEB or its variations.

[graeme]

This is a standard case for the NEB; there is no problem that the products are lower in energy than the reactant. The initial and final structures are not allowed to relax during the NEB optimization. The images can relax, but they will not relax to the product state because they are restrained to be equally spaced along the path.

[cbbasak]

Dear Graeme... thanks for the reply. But my question, in a nutshell, was is it possible to relax the intermediate structures only in terms of atomic positions with a predefined/known lattice parameters (even for the intermediate steps).

[graeme]

Yes, this is the standard way to run the NEB.