Problem with extremely small time step

LAMMPS LAMMPS Mailing List Mirror
1

Dear axel,

Again i’ve created with graphene single layer sheet(atomic style full) using VMD nanobuilder plugin. But, with that data file also atoms are lost.
Is their any way to correct geometry of a sheet? Can I manually change the boundaries in data file? When I tried I’m not getting good boundary conditions.
Now My new Input file is

units metal
atom_style full

boundary m p m
newton on
processors 1 1 1

dimension 3

read_data data.graphene

pair_style airebo 3.0
pair_coeff * * CH.airebo C

timestep 0.0005
#------------------- RELAXING-------------------

fix 1 all npt temp 300.0 300.0 0.05 y 0 0 0.5
thermo 2000
compute 1 all stress/atom NULL
compute 2 all pe/atom pair bond
compute 3 all reduce sum c_1[1] c_1[2] c_1[3]
thermo_style custom step temp pe ke etotal press lx ly pxx pyy c_3[1] c_3[2] c_3[3]
run 60000

unfix 1
-----------------deformation-----------------

fix 1 all nvt temp 300 300 0.05
fix 2 all ave/atom 1 1000 1000 c_1[1] c_1[2] c_1[3] c_2 fx fy fz
dump 1 all custom 1000 dump.new.* id type x y z vx vy vz c_1[1] c_1[2] c_1[3] c_2 f_2[1] f_2[2] f_2[3] f_2[4] f_2[5] f_2[6] f_2[7]
dump 2 all xyz 10000 dump.graphene.*.xyz
variable srate equal 1.0e9
variable srate1 equal “v_srate/1.0e12”

fix 3 all deform 1 y erate 0.005 units box remap x
run 600000
-> I’m first relaxing the sytem for 30ps (0.0005*60000) and then applying strain in y direction.
I’ve tried boundary conditions m and p, but the potential energy is increasing in relaxing time with fluctuating temperature and potential energy is constant when applied with strain.
Even When I try to visualize the system it is very unusual.Please have a look at the image attached.
Thanks for your co-operation Axel and have a good day ahead.

graphene.png

data.graphene (57.1 KB)

input.graphene (856 Bytes)

2

Dear axel,
      Again i've created with graphene single layer sheet(atomic style full)
using VMD nanobuilder plugin. But, with that data file also atoms are lost.
Is their any way to correct geometry of a sheet? Can I manually change the
boundaries in data file? When I tried I'm not getting good boundary

sure you can. you can also use change_box.

conditions.
Now My new Input file is

units metal
atom_style full

atom_style full???

your data file has bonds and other topology data. this is very, *very*
bad. in AIREBO bonds, angles, dihedrals and so are implicit. if you
add them explicitly, you have a problem. as you can see, your
structure is not stable. this is because with defined bonds, all kinds
of 1-2, 1-3 pairs are excluded from the non-bonded calculation, but
the airebo pair style requires those and thus all your forces and
interactions are completely bogus.

axel.

3

... as a quick workaround you can use

special_bonds lj/coul 1.0 1.0 1.0

this can run with 1fs.

so in conclusion, it *was* the input geometry that was wrong.

axel.

4

Thank you axel,

  1. Data file when used with airebo potential should not have any bonds, angles, dihedrals.

  2. I’m using VMD nanotube builder to build graphene sheet. Since I convert that into lammps data file i’m getting full atomic style graphene. Can please suggest me another way to do build graphene with good geometry.

Thank you very much for your help axel.

5

Thank you axel,
           1) Data file when used with airebo potential should not have any
bonds, angles, dihedrals.
           2) I'm using VMD nanotube builder to build graphene sheet. Since
I convert that into lammps data file i'm getting full atomic style graphene.

please RTFM!

6

Sorry for disturbing you sir. thanks for your reply.