MTD CV: &REACTION_PATH

marci marc... at pci.uzh.ch
Fri Oct 28 10:39:44 UTC 2011


Dear Ross,

The first thing to do is to decide how your path is defined.
You can use frames, in this case two are not sufficient, or you can
use CV and parameterized them through function of a variable T that
spans a given RANGE of values.
See examples in the regtests.

In 1-D metadynamics you have one collective variable. The only way to
condensate the information from two variables into one, is to define a
sort of combination of the two.
Whether this  can be of any use in the case of s and z is not that
clear to me.

best
marcella

On Oct 27, 6:49 pm, "Ross, Sun" <sunshou... at yahoo.com.cn> wrote:
> Hi everyone:
> I run 1D MTD with &REACTION_PATH    (PCV). The input file as follows:
>
>      &REACTION_PATH
>        &FRAME
>          COORD_FILE_NAME a.xyz
>        &END
>        &FRAME
>          COORD_FILE_NAME b.xyz
>        &END
>        SUBSET_TYPE LIST
>        ATOMS 2 3 4 5 6 7 8
>        ALIGN_FRAMES  T
> #       DISTANCES_RMSD T
>        LAMBDA 5.0
>        RANGE 0.01 10.0
>        RMSD T
>        STEP_SIZE 0.01
>     &END REACTION_PATH
>
> I don't know whether it is correct or not. Could anyone tell me?
> Still, there is another question, in 1D-metadynamics how can I get the
> two function s(R) and z(R) as listed (1-2) in J. Chem. Phys. 125.
> 054103 (2007)
>
> Thank you in advance and looking forward your reply.


More information about the CP2K-user mailing list