Tip4 water - Running into problems

LAMMPS LAMMPS Mailing List Mirror
1

Hello Everyone,

I am having problem running a simple water simulation using TIP4 water model. The problem I am facing is with minimization and fix/shake. I tried different minimization criteria with a range of parameters but I am still getting the same “Stopping criterion = linesearch alpha is zero” that results in unrealistic high value of energies. Because minimization is not working, I think I am getting NAN as result for fix/shake and the program stops with a SEGMENTATION FAULT.

I am attaching my input script and data file with this message. Please help me to sort out this problem. I have also added the log file that

I am ruuning this simulation on the latest version of LAMMPS on a linux machine.

I would really appreciate if anybody can help me sort out this problem. Thanks for the help.

Rahul

log.lammps (6.69 KB)

in.watertip4 (1.31 KB)

waterbox.dat (7.86 MB)

2

Hello Everyone,

    I am having problem running a simple water simulation using TIP4 water
model. The problem I am facing is with minimization and fix/shake. I tried
different minimization criteria with a range of parameters but I am still
getting the same "Stopping criterion = linesearch alpha is zero" that
results in unrealistic high value of energies. Because minimization is not
working, I think I am getting NAN as result for fix/shake and the program
stops with a SEGMENTATION FAULT.

shake doesn't work during minimization and your force constants for OH
bonds and HOH angles are zero. this will lead to a totally unrealistic
geometry during minimization.

create a (much) smaller waterbox and visualize what is going one and
improve your input until it works well.

axel.

3

Hello Everyone,

I wanted to know the procedure of starting up a water vapor calculation in LAMMPS. I want to calculate properties of water vapor at a higher temperature than 100 Celsius. Currently I am setting up the simulation same as liquid water but with the density of 0.001gm/cm3. So, initially my box has 1000 times less molecules than that of liquid water and it has 800 atoms in 200x200x200 A3 box. The simulation runs perfectly till 267 K but after that simulation stops mid-way giving me errors.

I would appreciate if anybody can help me.

Thanks,
Rahul

INPUT SCRIPT

clear

units real

dimension 3

boundary p p p

atom_style full

read_restart 374K40box.1Bar.restart.150000

create groups

group ox type 3

group hy type 1 2

set charges - beside manually

set group ox charge -0.830

set group hy charge 0.415

TIP3P Potential Parameters

pair_style lj/cut/coul/long 10.0

pair_coeff 1 1*2 0.000 0.000

pair_coeff 1 3 0.000 0.000

pair_coeff 2 2 0.000 0.000

pair_coeff 2 3 0.000 0.000

pair_coeff 3 3 0.102 3.188

bond_style harmonic

bond_coeff 1 1050 0.9572

angle_style harmonic

angle_coeff 1 550 104.52

kspace_style pppm 1.0e-5 #final npt relaxation

fix 1 all shake 1e-6 200 0 b 1 a 1

velocity all create 1.0 1234546 dist gaussian

fix 2 all npt temp 374.0 374.0 100.0 iso 1.0 1.0 1000.0

#Compute RDF and save to file

This will compute rdf between oxygen-oxygen, oxygen-hydrogen, and hydrogen-hydrogen

compute myRDF all rdf 500 1 1 1 2 2 2

fix 3 all ave/time 2 100 5000 c_myRDF file tip3p_afterboiling40.rdf mode vector

neighbor 10.0 bin

neigh_modify delay 10 every 2 check yes page 100000

thermo 100

thermo_style custom step temp press vol ke pe etotal enthalpy

thermo_modify norm no flush yes

#dumps

dump 1 all custom 1000 traj_npt_after373.lammpstrj id mol type x y z ix iy iz

restart 50000 374K40box.1Bar.restart

#run variables

timestep 0.0001

run 60000000

4

Hello Everyone,

I wanted to know the procedure of starting up a water vapor calculation in
LAMMPS. I want to calculate properties of water vapor at a higher
temperature than 100 Celsius. Currently I am setting up the simulation same
as liquid water but with the density of 0.001gm/cm3. So, initially my box
has 1000 times less molecules than that of liquid water and it has 800 atoms
in 200x200x200 A3 box. The simulation runs perfectly till 267 K but after
that simulation stops mid-way giving me errors.

... and we should use our mind reading abilities to figure out *what*
those errors are, or what?

I would appreciate if anybody can help me.

impossible without more specific information.

what stands out in the script below are the rather unusual settings
for neighbor list skin and updates. what is the reason for those?
don't you get dangerous builds with this?

also, why on earth do you run with fix npt when you want to run with a
fixed density?

in any case, have you visualized your trajectories? how long can you
keep your system in the (obviously) meta-stable vapor phase? if things
go wrong, the first thing to check is always whether the system
evolves as expected.

axel.