Query about the superlattice

LAMMPS LAMMPS Mailing List Mirror
1

Dear :

I am simulating a thermal resistance between the Si and Ge.The MP method is used in my model.I created the initiating structure in lammps .The input deck is below. In the relaxation process,first I ran the system in NPT 500000 steps. Second I ran the system langevin+nve 500000 steps .At last I ran the nve 100000 steps .However，I found the temperature continues increasing .Why ? Is my initiating structure wrong? Thank you very much !

I 'm very grateful!

…
units metal
boundary p p p
atom_style atomic
newton on
neighbor 2 bin
neigh_modify every 1 delay 0 check yes

lattice diamond 5.657
region md block -100 100 -12.5 12.5 -12.5 12.5 units box
create_box 2 md
region Geregion block -50 50 INF INF INF INF units box
create_atoms 2 region Geregion

lattice none
lattice diamond 5.431
region Siregion1 block INF -51 INF INF INF INF units box
region Siregion2 block 51 INF INF INF INF INF units box
region Siregion union 2 Siregion1 Siregion2
create_atoms 1 region Siregion

pair_style tersoff
pair_coeff * * SiCGe.tersoff Si(D) Ge

mass 1 28.0855
mass 2 72.64
…

2

2011/10/14 王丹 <[email protected]...>:

Dear :

           I am simulating a thermal resistance between the Si and Ge.The
MP method is used in my model.I created the initiating structure in lammps
.The input deck is below. In the relaxation process,first I ran the system
in NPT 500000 steps. Second I ran the system langevin+nve 500000 steps
.At last I ran the nve 100000 steps .However，I found the temperature
continues increasing .Why ? Is my initiating structure wrong? Thank you
very much !

have you checked, whether your time step is adequate?

axel.

3

Dear axel：
Hello,Thank you for your help.I have ran 5000000 steps and timestep=0.5fs.However,the result is same . I am very puzzled.

dan

在 2011年10月17日 下午10:36，Axel Kohlmeyer <[email protected]>写道：