temperature increase in SiC with Tersoff potential in NVE calculations

Dear lammps users,

I run nve calculations in SiC with tersoff potential with time step 1 fs,
however at about 200000 rises from 300 K to 400K and continue to increase.
Thank you for any helps.
German

Below the input script:

units metal

boundary p p p
atom_style atomic
#lattice sc 4.382
lattice sc 4.329
region box block 0 1 0 1 0 1
create_box 2 box
create_atoms 1 single 0 0 0
create_atoms 1 single 0.5 0.5 0
create_atoms 1 single 0.5 0 0.5
create_atoms 1 single 0 0.5 0.5
create_atoms 2 single 0.25 0.25 0.25
create_atoms 2 single 0.75 0.75 0.25
create_atoms 2 single 0.75 0.25 0.75
create_atoms 2 single 0.25 0.75 0.75
mass 1 28.086
mass 2 12.011
replicate 4 4 4

velocity all create 600.0 87287 loop geom

timestep 0.001

thermo 1

pair_style tersoff
pair_coeff * * SiC.tersoff Si C

# --------------- Equilibration in nve -----------------

fix NVE all nve
run 800000

Tersoff potentials are usually ran with < 0.4 fs timesteps; try with
smaller timesteps.

Best,
Ray

Dear Ray,

I tied to do simple NVE calculation with SiC.tersoff potential and
timestep 0.04 fs and have the same problem
temperature, total energy and MSD increases.

By the way, I can not find any support for you your statement about
time step < 0.4 fs.
For example, authors in Journal of Nuclear Materials 385 (2009) 572,
used timestep 1 fs for thermal equilibration.
The same 1 fs time step was used in other cascade modeling publications.

I presented results of calculation for T=1500 K, but the similar
effect takes place at 300 K also.

So I think that problem exists.

I used version: lammps-14Apr11.

Below the input I used and I attached time dependencies of E, T and MSD:

units metal

boundary p p p
atom_style atomic
#lattice sc 4.382
lattice sc 4.329
lattice sc 4.397
region box block 0 1 0 1 0 1
create_box 2 box
create_atoms 1 single 0 0 0
create_atoms 1 single 0.5 0.5 0
create_atoms 1 single 0.5 0 0.5
create_atoms 1 single 0 0.5 0.5
create_atoms 2 single 0.25 0.25 0.25
create_atoms 2 single 0.75 0.75 0.25
create_atoms 2 single 0.75 0.25 0.75
create_atoms 2 single 0.25 0.75 0.75
mass 1 28.086
mass 2 12.011
replicate 4 4 4

group Si type 1
group C type 2

velocity all create 3000.0 87287 loop geom

timestep 0.0004

thermo 100

pair_style tersoff/zbl
pair_coeff * * SiC.tersoff.zbl Si C

fix NVE all nve

compute 1 all msd
compute 2 Si msd
compute 3 C msd

thermo_style custom step temp press etotal pe vol c_1[4] c_2[4] c_3[4]

run 10000000

Best,

German

e.jpg

Dear German,

I suggest that you try running for longer time. It might saturate at some temperature. I had the same problem initially for Si nanowires that I was simulating. But I had to run ~15 ns to make sure that temperature gets stable. Please see supporting info of our recently accepted paper

http://xlink.rsc.org/?doi=C2NR30602F

I hope this helps.

Best Regards,
Vikas

P.S. if your temperture goes above 600, there is something wrong. Also, I used SiCGe.tersoff parameters from potentials directories for my study.

Hi German,

Energy conservation depends on the following:

1. time step size,
2. implementation of potential, and
3. appropriateness of the potential for the system.

All of the above are necessary conditions to achieve energy conservation.

I trust (2), so it really comes down to (1) and (3). You can try a
series of conservation tests, and plot the RMS of energy fluctuation
w.r.t. time step. Check out Allen and Tildesley for more info.

Maybe the lattice parameter is not at the equilibrium of the
SiC.tersoff parameterization?