"strategies to relax local energy spikes during dynamic topology changes"

LAMMPS LAMMPS Beginners
1

Dear Team,

i’m new to LAMMPS. I’m running REACTER simulations on LAMMPS for PE based polymers. Befor I successfully completed the REACTER for 1 ns and but when I’m running the longer simulation 10 ns , they are failing with errors like

ERROR on proc 44: Angle atoms 26984 25243 24185 missing on proc 44 at step 3712138
For more information see https://docs.lammps.org/err0005 (src/ntopo_angle_all.cpp:62)
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 44 in communicator MPI_COMM_WORLD
  Proc: [[7979,1],44]
  Errorcode: 1

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
prterun has exited due to process rank 44 with PID 2373136 on node c0711a-s3 calling
"abort". This may have caused other processes in the application to be
terminated by signals sent by prterun (as reported here).
--------------------------------------------------------------------------
�

this is the input script that I’m using for the simulation

units           real
atom_style      full
pair_style      lj/cut/coul/long 12.0  12.0
boundary        p p p
kspace_style    pppm 1e-4
bond_style      harmonic
angle_style     harmonic
dihedral_style  fourier
improper_style  cvff
special_bonds   amber
pair_modify     mix arithmetic

################################################################################
dielectric 1.0
neighbor 3.0 bin
neigh_modify delay 0 every 1 check yes
timestep 1.0
################################################################################

read_restart    nvt.restart
reset_timestep  0           

variable	runsteps equal 10000000
group OH_wt molecule > 4

molecule mol1 unreacted_template
molecule mol2 reacted_template

thermo 1000

fix myrxns all bond/react stabilization yes statted_grp .03 reset_mol_ids molmap &
  react rxn1 all 1 0.0 1.5 mol1 mol2 map02 rate_limit 1 1

fix 1 statted_grp_REACT nvt temp 300 300 100

thermo_style custom step temp pe etotal press density f_myrxns[1] 

dump mydmp1 OH_wt custom 1000 linear.lammpstrj id mol type x y z xu yu zu
run ${runsteps}
write_restart restart.run
write_data reacter_done.lmps
write_data reacter_done.lmps

I think the crash is happening because of the frequent topology changes. So i want to ask about the strategies to relax local energy spikes during dynamic topology changes

2

Based on what you provided, your settings look reasonable. The two parameters to tune are the ‘xmax’ and ‘stabilize_steps’ keywords. For example, you could trying increasing ‘stabilize_steps’ to 100 (default is 60 steps). You can also modify the behavior of reacting atoms. For example, you can add the following line after ‘fix bond/react’ to rescale the velocities of reacting atoms on every step:

fix 4 bond_react_MASTER_group temp/rescale 1 300 300 10 1

It sounds like your simulation is becoming unstable only rarely. Sometimes, you can restart from a restart file to get past a rare, problematic numerical event and the simulation will continue successfully after that.

3

I also suggest that you confirm that your simulation runs stably for long non-reactive simulations, and double-check the force field parameters, especially those of the angle that is misbehaving.

4

Thank you Dr.Jacob for replying.

i started using the restart files.