[CP2K-user] [CP2K:16683] Re: Problems of external potential and meta-MD wall

Marcella Iannuzzi marci.akira at gmail.com
Wed Mar 9 09:01:21 UTC 2022

```Dear Bai

In the first approach that you propose, you are assuming that the
coordinates of atom 5 remain fixed at X0 Y0 and Z0 during the simulation.
Is this really the case? If not, the effect of the external potential is to
keep atom 9 within a certain distance from a fixed point in the space, with
Cartesian coordinates (X0,Y0,Z0).

The second approach could work, if you correctly define the collective
variable.
According to what you wrote, there is a wall acting on a coordination
number, not on a distance.
This means that it should prevent the coordination number to become larger
than 3, not the distance.
This  is not going to produce the effect you want.

Regards
Marcella

On Wednesday, March 9, 2022 at 5:06:41 AM UTC+1 lb90... at gmail.com wrote:

> Dear CP2K Users,
>
> I want to do a MD simulation about a reaction of two molecules, where the
> two molecules contain a O and a C respectively. I needed to prevent these
> two atoms (O and C in two molecules respectively) moving more than 3
> angstrom from each other. So I find two methods that might work, but I
> don't know which one is applicative or I need to connect both of the
> following methods together.
>
> 5 and 9 is the atom O and C. X0, Y0, Z0 in external potential section is
> the coordinates of atom 5.
> 1. Using External potential section.
>    &EXTERNAL_POTENTIAL
>      ATOMS_LIST   9
>      FUNCTION     A*(Z-Z0)^8
>      PARAMETERS   A Z0
>      VALUES       0.001  0.347
>      UNITS        eV angstrom
>    &END
>    &EXTERNAL_POTENTIAL
>      ATOMS_LIST   9
>      FUNCTION     A*(X-X0)^8
>      PARAMETERS   A X0
>      VALUES       0.001  1.091
>      UNITS        eV angstrom
>    &END
>    &EXTERNAL_POTENTIAL
>      ATOMS_LIST   9
>      FUNCTION     A*(Y-Y0)^8
>      PARAMETERS   A Y0
>      VALUES       0.001  0.062
>      UNITS        eV angstrom
>
> For the second methods, I just want to apply the Wall without doing
> Meta-MD, so I close the DO_HILLS
> 2. Using Meta-MD Wall section.
>       DO_HILLS .FALSE.
>       NT_HILLS 10000
>       &METAVAR
>         SCALE 0.2
>         COLVAR 1
>         &WALL
>           POSITION [angstrom] 3.0
>           TYPE QUARTIC
>           &QUARTIC
>             DIRECTION WALL_PLUS
>             K  [kcalmol] 40.0
>           &END QUARTIC
>         &END WALL
>       &END METAVAR
> ...........
>      &COLVAR
>        &COORDINATION
>          ATOMS_FROM  5
>          ATOMS_TO   9
>          R_0 [angstrom]  3.0
>          NN  8
>          ND  14
>        &END COORDINATION
>      &END COLVAR
>    &END SUBSYS
>
> If someone have time to give me some advices, i would appreciate it.
> Best regards
>
> Bai
>

--
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+unsubscribe at googlegroups.com.