fix spring/pull

Dear Axel and all,

I would like to pull a group of (thermostated) atoms with a constant
velocity. Fix smd would be perfect to do that, except that the tether
point will have to cross periodic boundaries during my simulation. I
saw on Axel webpage that lammps-icms now includes a fix spring/pull
that could do the trick. Where could I find information on the syntax
for this new fix?

Best regards,
Laurent

dear laurent,

Dear Axel and all,

I would like to pull a group of (thermostated) atoms with a constant
velocity. Fix smd would be perfect to do that, except that the tether
point will have to cross periodic boundaries during my simulation. I
saw on Axel webpage that lammps-icms now includes a fix spring/pull
that could do the trick. Where could I find information on the syntax
for this new fix?

the syntax can be inferred from the source. :wink:

fix ID group-ID spring/pull K xr yr zr xv yv zv r0

K is the force constant
xr yr zr are the coordinates of the tether point (can be NULL)
xv yv zv are the velocity components with which the tether point moves
r0 is the radial distance that the group is kept away from the tether
point (usually 0.0)

a few more comments.

i have not submitted this fix, since i felt it
was solving a one-time problem. i may reconsider
if people can produce credible use scenarios.

overall, fix smd has long been due for a complete
overhaul and the functionality of fix spring/pull
could simply be integrated as a third mode of
operation.

salut,
    axel.

fix ID group-ID spring/pull K xr yr zr xv yv zv r0

K is the force constant
xr yr zr are the coordinates of the tether point (can be NULL)
xv yv zv are the velocity components with which the tether point moves
r0 is the radial distance that the group is kept away from the tether
point (usually 0.0)

Thanks!

a few more comments.

i have not submitted this fix, since i felt it
was solving a one-time problem. i may reconsider
if people can produce credible use scenarios.

overall, fix smd has long been due for a complete
overhaul and the functionality of fix spring/pull
could simply be integrated as a third mode of
operation.

I'm trying to investigate the sliding friction between two graphene
layers. As I am usually working on liquid/solid friction and I am not
used to this kind of problems, here I simply try to reproduce the many
published results on the subject. It appears that people usually
impose a sliding velocity (in a more or less invasive way) between the
two graphene sheets and measure the resulting shear force.

I will try to modify slightly fix spring/pull in order to apply the
pull only in the sliding direction, and to use a small force constant
(and check that the results do not depend on the latter)...

Best,
Laurent

fix ID group-ID spring/pull K xr yr zr xv yv zv r0

K is the force constant
xr yr zr are the coordinates of the tether point (can be NULL)
xv yv zv are the velocity components with which the tether point moves
r0 is the radial distance that the group is kept away from the tether
point (usually 0.0)

Thanks!

a few more comments.

i have not submitted this fix, since i felt it
was solving a one-time problem. i may reconsider
if people can produce credible use scenarios.

overall, fix smd has long been due for a complete
overhaul and the functionality of fix spring/pull
could simply be integrated as a third mode of
operation.

I'm trying to investigate the sliding friction between two graphene
layers. As I am usually working on liquid/solid friction and I am not
used to this kind of problems, here I simply try to reproduce the many
published results on the subject. It appears that people usually
impose a sliding velocity (in a more or less invasive way) between the
two graphene sheets and measure the resulting shear force.

I will try to modify slightly fix spring/pull in order to apply the
pull only in the sliding direction,

I don't think there is a modification needed for that. Just set the
reference point coordinates in the undesired direction to NULL.

and to use a small force constant
(and check that the results do not depend on the latter)...

You don't want to have too small a force constant or else it will be
problematic to get the PMF.

Axel.