Estimated need confine an atomic solvent between two walls fixed at different temperatures to see the density profile when the walls are fixed.
There is the option of using an explicit wall (made of atoms), limit the movement of the atoms of the wall and apply a thermostat to maintain its temperature. However, I do not want to use the walls above.
My intention is to use an ideal((http://lammps.sandia.gov/doc/fix_wall.html) ) or reflective wall to impose conditions fixed edges. I am not clear as to approximate a real wall at a given temperature using the model of ideal wall though. One option that I can think of is to use a thermostat in a thin region around each wall.
Another option is to implement stochastic walls, which consist of defining a region to which solvent atoms can enter or leave (in the same direction) but with different velocities obtained speeds distributions different. Not if this option can be implemented with lammps.
I wonder if someone made something simulate and see if there is any way to implement the type of wall (Thermal wall) that mention using lammps.
I’m having a hard time to understand your message. You want your solvent confined between two “flat” walls also experiencing a given temp? Do you want your wall to be impenetrable? I don’t think a “thin” thermostated region is a good idea as there is always the possibility that after time integrating a solvent particle for one step the particle could move from one side of the region to the other one thus completely skipping the action of the thermostat. This leads to the need of defining “thin” unambiguously. Perhaps you should explain better what is the exact model of interactions between the solvent and wall that you are after. Mind you somebody who does this type of simulations
frequently may know better what you are asking for.
Carlos
Estimated need confine an atomic solvent between two walls *fixed* at
different temperatures to see the density profile when the walls are fixed.
There is the option of using an explicit wall (made of atoms), limit the
movement of the atoms of the wall and apply a thermostat to maintain its
temperature. However, I do not want to use the walls above.
My intention is to use an ideal(( http://lammps.sandia.gov/doc/fix_wall.html) ) or reflective wall to
impose conditions fixed edges. I am not clear as to approximate a real wall
at a given temperature using the model of ideal wall though. One option
that I can think of is to use a thermostat in a thin region around each
wall.
Another option is to implement stochastic walls, which consist of defining
a region to which solvent atoms can enter or leave (in the same direction)
but with different velocities obtained speeds distributions different. Not
if this option can be implemented with lammps.
I wonder if someone made something simulate and see if there is any way to
implement the type of wall (Thermal wall) that mention using lammps.
considering that computer time is rather cheap these days, i would not use
walls at all, but rather use a couple of layers of atoms that are
immobilized and then put a sufficiently thick layer of atoms that have
(weak) position restraints via fix spring/self or something equivalent, and
thermalize this section with a dissipative thermostat (temp/csvr,
temp/csld, langevin). no hacks, no worries.
I'm having a hard time to understand your message. You want your solvent
confined between two "flat" walls also experiencing a given temp? Do you
want your wall to be impenetrable?
*Yes, I would induce a temperature gradient in the solvent and the walls
are impenetrable.*
I don't think a "thin" thermostated region is a good idea as there is
always the possibility that after time integrating a solvent particle for
one step the particle could move from one side of the region to the other
one thus completely skipping the action of the thermostat. This leads to
the need of defining "thin" unambiguously. Perhaps you should explain
better what is the exact model of interactions between the solvent and wall
that you are after. Mind you somebody who does this type of simulations
frequently may know better what you are asking for.
*You are right sir, the entry and exit of particles can become a problem
for the statistics. Also particles in and out of the thin area where the
thermostat applies imply that the number of particles is not constant ,
therefore , not be sampling from a canonical ensemble strictly speaking .
After establishing a steady state desire to incorporate a large particle ,
whose interaction with the solvent and the wall is established with
pair_style colloid that has lammps .*
Estimated need confine an atomic solvent between two walls *fixed* at
different temperatures to see the density profile when the walls are fixed.
There is the option of using an explicit wall (made of atoms), limit the
movement of the atoms of the wall and apply a thermostat to maintain its
temperature. However, I do not want to use the walls above.
My intention is to use an ideal(( http://lammps.sandia.gov/doc/fix_wall.html) ) or reflective wall to
impose conditions fixed edges. I am not clear as to approximate a real wall
at a given temperature using the model of ideal wall though. One option
that I can think of is to use a thermostat in a thin region around each
wall.
Another option is to implement stochastic walls, which consist of
defining a region to which solvent atoms can enter or leave (in the same
direction) but with different velocities obtained speeds distributions
different. Not if this option can be implemented with lammps.
I wonder if someone made something simulate and see if there is any way
to implement the type of wall (Thermal wall) that mention using lammps.
considering that computer time is rather cheap these days, i would not
use walls at all, but rather use a couple of layers of atoms that are
immobilized and then put a sufficiently thick layer of atoms that have
(weak) position restraints via fix spring/self or something equivalent, and
thermalize this section with a dissipative thermostat (temp/csvr,
temp/csld, langevin). no hacks, no worries.
*I think it's a good idea. *
*I have another question about the type of wall you mention :For example if
we consider a colloidal particle immersed in the solvent , which colloid
-solvent interactions are modeled with pair_style colloid ( Hamaker ) . To
model the interaction between these large particles and the wall of atoms ¿
I can use the colloid pair_style ?, considering a similar colloid -solvent
interaction.At first glance it seems that if , as would be sufficient to
establish the appropriate Hamaker constant for this case. However , I'm not
quite sure , maybe there is a better approach.*
considering that computer time is rather cheap these days, i would not
use walls at all, but rather use a couple of layers of atoms that are
immobilized and then put a sufficiently thick layer of atoms that have
(weak) position restraints via fix spring/self or something equivalent, and
thermalize this section with a dissipative thermostat (temp/csvr,
temp/csld, langevin). no hacks, no worries.
*I think it's a good idea. *
*I have another question about the type of wall you mention :For example
if we consider a colloidal particle immersed in the solvent , which colloid
-solvent interactions are modeled with pair_style colloid ( Hamaker ) . To
model the interaction between these large particles and the wall of atoms ¿
I can use the colloid pair_style ?, considering a similar colloid -solvent
interaction.At first glance it seems that if , as would be sufficient to
establish the appropriate Hamaker constant for this case. However , I'm not
quite sure , maybe there is a better approach.*
with this kind of setup there is not particular "wall". everything is set
up the same way as the rest of the system, only the particles near the edge
cannot move away, and the fully immobilized particles are needed to reduce
surface effects on the restrained particles. now, what is the best model
and set of parameters for your particular project, is something that you
will have to figure out for yourself.
Yes, it was the first thing that occurred to me when I thought about the interaction between colloid and a wall . However, as you could use the fix wall / colloid to describe the interaction between the colloid and the wall at a desired temperature ?