Constraining the OH covalent bond distance in liquid water

[Giuseppe Cassone]

Dear CP2K developers,

I was wondering whether it exists a simple method in CP2K (v. 2025.1) 

to constraint the maximum value assumed by O-H covalent bonds (let say

at 1.3 Angstrom) in ab initio molecular dynamics simulations of bulk liquid 

water without specifying a single COLVAR for each O-H covalent bond distance,

a not practical circumstance implying defining 2*N COLVARS and 2*N CONSTRAINTS.

Many thanks in advance and best wishes,

Giuseppe 

[Quentin Pessemesse]

Hello,

Not a dev ofc, if you have to use only CP2k I would suggest you write a bash/python script to add the 2N colvars and constraints, it is not so hard to do even if you are not a coding expert (some googling or LLM would help). but I do not believe multiple colvars can be defined in one go in the input.

If you can use the plumed software with CP2k, you can do this using the Multicovlar DISTANCES and the manyrestraint UWALLS. Plumed is a bit more user friendly when it comes to adding many different walls to a simulation.

Best,

Quentin

[Giuseppe Cassone]

Dear Quentin,

this was exactly what I did in the last days: I used multiple COLVARS and put external UWALLS with Plumed.

I expected that the &HBONDS keyword of CP2K alone was capable of doing something similar

but, apparently, it's much more convenient to follow the Plumed strategy, which doesn't require any

specific topology of the molecular system.

Best wishes,

Giuseppe

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/25773374-1d6e-411b-9f8b-87076f2d3db8n%40googlegroups.com.