fix langevin in two dimension

Dear All,

I tried to use fix langevin along with fix NVE. I get this ERROR: atoms lost. However when i use only fix NPT, I do not see any error.
So I would like know if anyone has used fix langevin to achieve thermal equilibrium in two dimensional system.

Thank you,

Chances are your initial config is not good (atoms too close) or
perhaps you are not applying the Langevin thermostat properly. Did you
check that indeed your random forces only have 2d character?
In addition, note that NVE + Langevin is not identical to NPT. One is
constant V the other one constant P.
Carlos

Like Carlos said, you need to make sure that your temperature is only
computed in two dimensions. For x and y, this can be done with something
like

compute newtemp all temp/partial 1 1 0

and then with a fix_modify for your langevin fix (here 1)

fix_modify 1 temp newtemp #remove temperature component in z direction

Good luck,
Mierk

If you are using fix langevin with fix nve in a 2d simulation, it should
just work. You do need to use fix enforce2d with a 2d model.

Steve