can atoms be fixed in NEB calculations in cp2k?

[anyu...@gmail.com]

Dear all,

I want to fix some atoms in CI-NEB calculations with cp2k by using "CONSTRAINT FIXED_ATOMS", however, the results show that the atoms I want to fix changed there positions a lot. For example, in my input file, I want to fix two O atoms, but the results shown they are not fixed at all. So far, I hadn't find the solution.

Could anyone please help to see where are the errors? Thanks a lot in advance.

Best,
Yun

[Maxime Van den Bossche]

Dear Yun,

I'm just a regular CP2K user, but I have some experience running
such calculations and can just give you these suggestions:

- Have you searched for similar topics in this group? E.g. this one:
https://groups.google.com/forum/#!topic/cp2k/WE5T3HgGSAc

- Could you turn your fairly convoluted input and structure into
a more minimal example, e.g. just H2 dissociation with standard DFT,
with one H atom fixed? That would make it easier for other people
to understand and help out with your issue, and you may even already
find a solution just by doing this.

- I'm actually using the ASE interface to CP2K to run NEB calculations
(see e.g. https://www.cp2k.org/tools:ase). Using the FixAtoms constraints
in ASE, I have no problems keeping atoms fixed during a NEB run. So
this is an alternative in case the "pure CP2K approach" turns out to
be problematic.

Best,
Maxime