Hi,
I am new to LAMMPS and I am trying to equilibrate a Cu sample at 300 K,
with free surfaces in Y and Z directions and periodic in X direction. I
was wondering what ensemble or command to use for relaxation under these
BCs?
Any help would be deeply appreciated.
Thanks a lot,
Yashashree
Use fix nve for integration and the boundary command to
set your BC in xyz.
Steve
Hi Steve,
Thank you very much for your response. I have been trying the commands and
fix that you suggested and I have a couple of questions.
I tried the fix nve without and with temp/rescale to adjust the
temperature at 300 K. In both cases, the Ly dimension keeps growing. Also,
with fix nve, Lx, being periodic, does not change to accommodate the
thermal expansion. So, I specified the lattice constant at 300 K in the
lattice command.
I also tried the npt fix using NULL for the non-periodic directions. It
gave similar results.
For some reason, when I group together a few layers at the bottom
(Z-direction) and set their velocity to zero and use setforce to freeze
them, the Ly dimension stabilizes around the initial value. But I am not
sure if this is the right way of equilibrating the sample.
Please find below the input file that I used. It would be very helpful if
you could advise me on these issues.
Thanks again,
Yashashree
units metal
boundary p s s
atom_style atomic
# define simulation cell
lattice fcc 3.6357
region box block 0 9.797959 0 14.14213 0 86.6025404
create_box 5 box
lattice fcc 3.6357 orient x 1 -1 -2 orient y 1 1 0 orient z 1 -1 1
spacing 1.22475 1.41421 1.73205
create_atoms 1 box
# specify material and atom types
mass 1 63.55
pair_style eam
pair_coeff * * Cu_u3.eam
neighbor 2.0 bin
neigh_modify delay 5
# define boundary
region lower block INF INF INF INF INF 1.5
region upper block INF INF INF INF 48.5 INF
group lower region lower
group upper region upper
group boundary union lower upper
group mobile subtract all boundary
compute mobile_temp mobile temp
compute mobile_stress mobile stress/atom
compute stress mobile reduce sum c_mobile_stress[1]
c_mobile_stress[2] c_mobile_stress[3]
variable sxx equal c_stress[1]/vol
variable syy equal c_stress[2]/vol
variable szz equal c_stress[3]/vol
# equilibrate
velocity all create 600.0 5713354
fix 1 all nve
thermo 500
thermo_style custom step temp press lx ly lz v_sxx v_syy v_szz
thermo_modify lost warn norm yes
run 200000
Use fix nve for integration and the boundary command to
set your BC in xyz.Steve
Hi,
I am new to LAMMPS and I am trying to equilibrate a Cu sample at 300 K,
with free surfaces in Y and Z directions and periodic in X direction. I
was wondering what ensemble or command to use for relaxation under these
BCs?
Any help would be deeply appreciated.
Thanks a lot,
Yashashree
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to
help
Using fix npt on the periodic dimension that you want to
expand thermally is the right way to do it. If the free
surfaces expand when you heat the sample, then that is
what they want to do. Unless you apply some other constraint,
like periodic (or fix npt) boundaries, then the atoms go where
they want to go.
Steve
Hi Steve,
Thanks again for your reply. Sorry to bother you again with this. I
understand that the free surfaces would also want to expand at finite
temperature. But in my runs, the free surfaces don't stop expanding. That
is, even if I increase the equilibration time from 20 to 50 ps, the
non-periodic dimensions keep growing. For instance, Ly (~5nm initially)
grows almost by 10 angstrom in 20 ps and does not stabilize. So how does
one know when the system has equilibriated?
Regards,
Yashashree
Using fix npt on the periodic dimension that you want to
expand thermally is the right way to do it. If the free
surfaces expand when you heat the sample, then that is
what they want to do. Unless you apply some other constraint,
like periodic (or fix npt) boundaries, then the atoms go where
they want to go.Steve
Hi Steve,
Thank you very much for your response. I have been trying the commands and
fix that you suggested and I have a couple of questions.
I tried the fix nve without and with temp/rescale to adjust the
temperature at 300 K. In both cases, the Ly dimension keeps growing.
Also,
with fix nve, Lx, being periodic, does not change to accommodate the
thermal expansion. So, I specified the lattice constant at 300 K in the
lattice command.
I also tried the npt fix using NULL for the non-periodic directions. It
gave similar results.
For some reason, when I group together a few layers at the bottom
(Z-direction) and set their velocity to zero and use setforce to freeze
them, the Ly dimension stabilizes around the initial value. But I am
not
sure if this is the right way of equilibrating the sample.
Please find below the input file that I used. It would be very helpful if
you could advise me on these issues.
Thanks again,
Yashashree
units metal
boundary p s s
atom_style atomic
# define simulation cell
lattice fcc 3.6357
region box block 0 9.797959 0 14.14213 0 86.6025404
create_box 5 box
lattice fcc 3.6357 orient x 1 -1 -2 orient y 1 1 0 orient z 1
-1
1
spacing 1.22475 1.41421 1.73205
create_atoms 1 box
# specify material and atom types
mass 1 63.55
pair_style eam
pair_coeff * * Cu_u3.eam
neighbor 2.0 bin
neigh_modify delay 5
# define boundary
region lower block INF INF INF INF INF 1.5
region upper block INF INF INF INF 48.5 INF
group lower region lower
group upper region upper
group boundary union lower upper
group mobile subtract all boundary
compute mobile_temp mobile temp
compute mobile_stress mobile stress/atom
compute stress mobile reduce sum c_mobile_stress[1]
c_mobile_stress[2] c_mobile_stress[3]
variable sxx equal c_stress[1]/vol
variable syy equal c_stress[2]/vol
variable szz equal c_stress[3]/vol
# equilibrate
velocity all create 600.0 5713354
fix 1 all nve
thermo 500
thermo_style custom step temp press lx ly lz v_sxx v_syy v_szz
thermo_modify lost warn norm yes
run 200000
Use fix nve for integration and the boundary command to
set your BC in xyz.
SteveHi,
I am new to LAMMPS and I am trying to equilibrate a Cu sample at 300 K,with free surfaces in Y and Z directions and periodic in X direction. I
was wondering what ensemble or command to use for relaxation under
these
BCs?
Any help would be deeply appreciated.
Thanks a lot,
Yashashree
------------------------------------------------------------------------------SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas,
Nevada.
If you have a good lattice constant and geometry
and are running at a reasonable Temp, then free
surfaces should not expand dramatically. If you
run with periodic boundaries is the pressure near
0 (averaged over fluctuations)? If so, it shouldn't
expand much with non-PBC.
Steve