I want to do a steered molecular dynamics simulation on adsorption of a molecule onto a layer.
I use this command:
fix pmfsmd layer smd cvel 300 0.0000001 couple moli NULL NULL 1 0
However, upon simulation, the molecule “suddenly” falls onto the surface of the layer. By sudden, I mean 500 fs, for example.
I used different parameters for K vel (cvel values), but the same sudden phenomenon happens.
Can anybody please help me what I am using wrong about this command?
Thank you.
Hello Xiaolin that sounds like a pbc issue, perhaps?
Giacomo
I want to do a steered molecular dynamics simulation on adsorption of a
molecule onto a layer.I use this command:
fix pmfsmd layer smd cvel 300 0.0000001 couple moli NULL NULL 1 0
However, upon simulation, the molecule "suddenly" falls onto the surface of
the layer. By sudden, I mean 500 fs, for example.
I used different parameters for K vel (cvel values), but the same sudden
phenomenon happens.
there seems to be a bug in fix smd for your kind of setup. at least
that is what it looks like, when setting up a quick test.
a possible workaround seems to be to use "auto" instead of 1.0 for the
coupling direction. at least that seems to generate the expected
force.
also, it seems that your coupling velocity has the wrong sign. if you
want the groups to move toward each other, the velocity should be
negative.
the more common case for using couple mode is to push two groups apart.
since your purpose is to steer a molecule toward adsorption, i would
expect that doing the calculation in tether mode with the target
location given would be the cleaner approach, and that is what i would
set up. in my quick test, that worked without a problem.
fix smd has been largely unmaintained (and obviously not fully tested)
for quite a while, so you may want to consider using the colvars
package instead. which is actively maintained, far more flexible, and
also offers steered MD.
axel.