ERROR: Out of range atoms - cannot compute PPPM

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I have been running simulations of a system of an interface of a Palmitic acid monolayer with water. I am using an OPLS force field. I was running with a timestep of 0.5 fs but I changed for a timestep of 2 fs and I constrained the C-H bonds and O-H bonds for the system to converge faster.

Here is some parameters:

pair_style lj/cut/tip4p/long 10 6 10 16 0.1546 8.5
pair_modify shift yes
bond_style harmonic
angle_style harmonic
dihedral_style opls
improper_style cvff
kspace_style pppm/tip4p 0.00001
special_bonds lj/coul 0.0 0.0 0.5

Everything worked fine but when I am trying to do the same with a system of Palmitate with water and Sodium ions (used as counterion). This error message appear after a certain time:

67210220 293.08927 -27320.486 -23285.127 339.14041 -515.10328 733.95842 798.56607 166.66195 -316.54275 -215.71227
67210240 300.78373 -27430.911 -23289.612 418.8136 -165.66854 440.11505 981.9943 248.66122 -160.85701 -285.41721
67210260 297.21773 -27373.697 -23281.496 -31.431348 -273.83507 -35.4121 214.95312 172.81895 -290.30868 -759.74581
67210280 352.2488 -26971.634 -22121.746 277.34245 172.24114 638.28525 21.500959 5.5087997 -171.32616 -228.80115
67210300 343.21135 -26809.168 -22083.71 65.124872 -289.03532 987.42832 -503.01839 212.1897 74.469053 300.96044
ERROR: Out of range atoms - cannot compute PPPM (…/pppm_tip4p.cpp:107)

So I read earlier posts about this kind of error and tried to play vith the kspace_style parameter and the neighbor parameter. But I still get the error message.

So my questions are:

  • Why does it work for the Palmitic acid but not for the Palmitate?

  • Why is the temperature (second column) raising just before crashing?

  • Is there something else I could try to fix this error without to have to reduce the timestep?

I hope that everything is clear and I gave enough information.

Thanks,

Raphael

2

Hello,

I have been running simulations of a system of an interface of a Palmitic acid monolayer with water. I am using an OPLS force field. I was running with a timestep of 0.5 fs but I changed for a timestep of 2 fs and I constrained the C-H bonds and O-H bonds for the system to converge faster.
Here is some parameters:

pair_style lj/cut/tip4p/long 10 6 10 16 0.1546 8.5
pair_modify shift yes
bond_style harmonic
angle_style harmonic
dihedral_style opls
improper_style cvff
kspace_style pppm/tip4p 0.00001
special_bonds lj/coul 0.0 0.0 0.5

Everything worked fine but when I am trying to do the same with a system of Palmitate with water and Sodium ions (used as counterion). This error message appear after a certain time:

67210220 293.08927 -27320.486 -23285.127 339.14041 -515.10328 733.95842 798.56607 166.66195 -316.54275 -215.71227
67210240 300.78373 -27430.911 -23289.612 418.8136 -165.66854 440.11505 981.9943 248.66122 -160.85701 -285.41721
67210260 297.21773 -27373.697 -23281.496 -31.431348 -273.83507 -35.4121 214.95312 172.81895 -290.30868 -759.74581
67210280 352.2488 -26971.634 -22121.746 277.34245 172.24114 638.28525 21.500959 5.5087997 -171.32616 -228.80115
67210300 343.21135 -26809.168 -22083.71 65.124872 -289.03532 987.42832 -503.01839 212.1897 74.469053 300.96044
ERROR: Out of range atoms - cannot compute PPPM (../pppm_tip4p.cpp:107)

So I read earlier posts about this kind of error and tried to play vith the kspace_style parameter and the neighbor parameter. But I still get the error message.

So my questions are:

- Why does it work for the Palmitic acid but not for the Palmitate?

perhaps, there is something not quite right in the topology, or force
field parameters are missing or incorrect, or you are missing bonds to
hydrogen atoms in your fix shake setting.

- Why is the temperature (second column) raising just before crashing?

this is the symptom of a problem with inadequate combination of
timestep and force field or topology. atoms get too close, then get
accelerated too much.
with a shorter time step, you may still have problems, e.g.
insufficient conservation of energy, but the larger the time step, the
closer atoms can get and thus the more likely is an unstable
trajectory.

- Is there something else I could try to fix this error without to have to reduce the timestep?

see above, you may need to verify all parts of your topology and force
field parameters. if you are using a software tool to build your data
file, perhaps there are known bugs or newer versions of the software.

please note, that an alternate approach to constraints to reduce the
computational cost would be to use a r-RESPA integrator and do bonded,
non-bonded and long-range coulomb interactions at different time
scales.

axel.