Rotational drift of CNT atoms ( # rigid rotation of whole CNT )

LAMMPS LAMMPS Mailing List Mirror
1

Hello All,

I have been trying to equilibrate a (10,10) CNT of length 50 A/ 100 A using modified Tersoff potential. After relaxation of the structure, when the structure is thermostatted to evolve under canonical ensemble (NVT), the CNT seems to gradually gain rigid rotation. In particular, if the atoms of the CNT are monitored, it shows rotational drift. Is it physical or an artifact of the potential or errors in running the simulation. My Lammps script goes like

clear
units metal
dimension 3
boundary p p p
atom_style atomic
atom_modify map array

CNT setup

read_data ${infile}
group sol type 1

------------------- FORCE FIELDS----------------------------

pair_style hybrid tersoff lj/sf 3.0 lj/sf 3.0
pair_coeff * * tersoff C.tersoff C NULL # C-C
pair_coeff 2 2 lj/sf 1 0.0 0.0 # Ar-Ar
pair_coeff 1 2 lj/sf 2 0.0 0.0 # C-Ar

#-------------------- NEIGHBOR UPDATING CRITERIA -------------
neighbor 3.0 bin
neigh_modify delay 5

#-------------------- MINIMIZATION ---------------------------
fix fi1 sol box/relax z 0.0
minimize 1e-25 1e-25 5000 10000
thermo 1
thermo_style custom step v_torz v_comx v_comy v_comz
unfix fi1

fix fi0 sol recenter {comxs} {comys} NULL units box
thermo 1
thermo_style custom step v_lxs v_lys v_lzs v_torz v_comx v_comy v_comz
run 5
unfix fi0

reset_timestep 0
timestep ${t_stp1}

#gaussian velocity distribution to each atom
velocity sol create {beg_temp} {velseed} mom yes rot yes dist gaussian
velocity sol zero angular

#constant volume time integration
fix fi3 sol npt temp {beg_temp} {end_temp} 20.0 z 0.0 0.0 1000.0
fix fi4 sol recenter {comxs} {comys} NULL

#Set thermo output
thermo ${thermo_freq1}
thermo_style custom step v_lxs v_lys v_lzs press pe ke temp v_torz v_omeg

#print
fix p1 id1 print 1 “{vx1} {vy1}” file id400.cor screen no
fix p2 sol print 1 “{torx} {tory} {torz}" file CNT.tor screen no fix p3 sol print 1 "{omeg} {comx} {comy} ${comz}” file CNT.omeg screen no

run {n_step1} write_restart restart.{outfile}

unfix fi3
unfix fi4

Any suggestions will be much appreciated

Thanks

De

-Subhadeep De
Graduate student
CMN Group

2

Hello All,

I have been trying to equilibrate a (10,10) CNT of length 50 A/ 100 A using
modified Tersoff potential. After relaxation of the structure, when the
structure is thermostatted to evolve under canonical ensemble (NVT), the CNT
seems to gradually gain rigid rotation. In particular, if the atoms of the
CNT are monitored, it shows rotational drift. Is it physical or an artifact

since you are using fix recenter, you cannot know whether you have a
translational drift as well.

of the potential or errors in running the simulation. My Lammps script goes

there are no obvious grave errors in the input, but issues can arise
from all kinds of sources and it is part of the process of learning
how to do simulations better, that you study, which changes have an
impact or not. if it was easy to tell this just from looking at an
input file, a lot of simulations would not need to be done. in
addition, quite a bit of important information is not provided here.

the general nature of your inquiry makes this more of a topic that you
should discuss with your adviser and senior colleagues in your group
or in the groups of collaborators. a mailing list is not a good place
to provide this kind of advice.

axel.