[CP2K-user] NEB replicas?

[Ernst Larsson]

Dear all,

I'm slightly confused about the "NUMBER_OF_REPLICA" keyword when performing 
a NEB-calculations.
I've optimized the start and end points separately, which due to symmetry 
have the same energies.

When I run the NEB-calculation, I provide three intermediate starting 
guesses which I've interpolated myself (I know that CP2K in principle can 
do this for me) and set "NUMBER_OF_REPLICA = 5".

My confusion comes when I look at the energies and files:
1) I get six files of the type '...-BANDX.out' rather than five.
2) The ENERGIES section in the NEB-output give me five energies, which is 
fine, but only the start point matches what I got from the geometry 
optimisation. The last replica has an energy which is larger than what it 
should be by 0.011205 a.u., (when it should be identical to the first). If 
I check "...-pos-Replica_nr_5-1.xyz" which I presume is for the endpoint, 
it is not identical to what I specified as my end point in the input.
3) The energy in this extra '...-BANDX.out'-file does not match the correct 
end-point either.

Does anyone have any good idea on what may be going wrong?
I've attached my NEB-input script.

Regards,
Ernst
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210326/37c6e508/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MgO_100.inp
Type: chemical/x-gamess-input
Size: 1654 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210326/37c6e508/attachment.inp>