Polymer Grafted Nanoparticles

Dear all

I’m trying to calculate spherical nanoparticles with a single grafted linear polymer.
Spherical nanoparticle diameter is treated as LJ spheres with diameter of 3.0 sigma.
Grafted polymer are treated as beads-spring of Kremer-Grest with standard parameters (bead diameter is set to 1.0 sigma and the number of bead per a chain is 10.).
I use langevin thermostat and nve. At first, to relax grafted polymers, I set LJ as purely repulsive interaction and run simulation. However, energy becomes huge and calculation stops very fast. And this warning displayed.

WARNING: FENE bond too long: 10904999 441 442 1.54137 (…/bond_fene.cpp:90)

Strangely, if in case that the number if grafted polymers is larger than 2, it work.

In only case of single grafted polymer system, this error occur.

In my data file, I define three types as follows;

type 1: bead of grafted polymer which grafted on the surface of nanoparticles.
type 2: nanoparticles
type 3: grafted polymers except type 1.

I treat type 1 and type 3 as rigid moleucle and exclude its interaction.

Here, I show my input file below.
///////////////////////////////////////////////////////////////////////////////////////////////

units lj
atom_style full
special_bonds fene

read_data PGNPs.np
neighbor 1.0 bin
neigh_modify every 1 delay 0 check yes page 100000

bond_style fene
bond_coeff 1 30.0 1.5 1.0 1.0

group PGNP type 1 2
group Chains type 3
neigh_modify exclude molecule PGNP

pair_style lj/cut 3.36
pair_modify shift yes
pair_coeff 1 1 0.0 1.0 1.12
pair_coeff 1 2 0.0 2.0 2.24
pair_coeff 1 3 0.0 1.0 1.12
pair_coeff 2 2 0.0 3.0 3.36
pair_coeff 2 3 0.0 2.0 2.24
pair_coeff 3 3 0.0 1.0 1.12

fix 1 Chains nve
fix 2 Chains langevin 1.0 1.0 1.0 904297
fix 3 PGNP rigid molecule langevin 1.0 1.0 1.0 904297
dump 1 all xyz 20000 dump_i.RC_Rod_AS5_AL03_N20_F1

thermo 20000
thermo_style custom step temp ke etotal press vol
timestep 0.005
restart 200000 restart_1i_temp restart_2i_temp
run 2000000
write_restart restart_i

///////////////////////////////////////////////////////////////////////////////////////////////

I am very confused. Please help me.

Best wishes,
Makoto

You already got some answers, but the error itself

is very simple. The FENE potential blows up

if the 2 bonded atoms get too far apart. LAMMPS

warns if that is about to happen, then gives an

error (and stops) if it does happen. Whenever you

get that warning/error, it simply means you have

bad dynamics. You have to figure out why that is.

I suggest you simply your script as much as possible,

track forces on the 2 particles in the offending bond,

and see if you can get stable dynamics with NVE

only. If you can’t do that, you have an ill-formed

model.

Steve