NEMD simulation by Langevin thermostat

LAMMPS
1

Dear all,
I am reading the example file (lammps/examples/KAPPA/in.langevin) in the github.

nemd
nemd831×332 306 KB

The image above shows a typical NEMD schematic.
Two Langevin thermostats, each at a different temperature, are applied to specific regions to create a temperature gradient.

From the example file:

region          hot block INF INF INF INF 0 1
region          cold block  INF INF INF INF 10 11

fix             1 all nve
fix             hot all langevin ${thi} ${thi} 1.0 59804 tally yes
fix             cold all langevin ${tlo} ${tlo} 1.0 287859 tally yes

The regions for the hot and cold Langevin thermostats are defined. However, I noticed that the Langevin thermostat is applied to “all” atoms, regardless of whether they are in the hot or cold region.

Wouldn’t it make more sense to define groups for the hot and cold regions and apply the thermostats specifically to those groups, as shown below?

region          hot block INF INF INF INF 0 1
region          cold block  INF INF INF INF 10 11
group           g_hot block hot
group           g_cold block cold

fix             1 all nve
fix             hot g_hot langevin ${thi} ${thi} 1.0 59804 tally yes
fix             cold g_cold langevin ${tlo} ${tlo} 1.0 287859 tally yes

Could anyone clarify if there’s a specific reason why the thermostats are applied to “all” atoms in the original script?

2

Hello,

The fix_modify commands are important here:

region hot block INF INF INF INF 0 1
region cold block INF INF INF INF 10 11
compute Thot all temp/region hot
compute Tcold all temp/region cold

(…)

fix hot all langevin {thi} {thi} 1.0 59804 tally yes
fix cold all langevin {tlo} {tlo} 1.0 287859 tally yes
fix_modify hot temp Thot
fix_modify cold temp Tcold

They allow the Lanvegin thermostats to target different regions. In fact, if you run the script, you should see that the temperature gradient is generated as expected.

Best,

Simon

3

Thanks a lot simongravelle.