LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] Error on proc 0: Out of range atoms - cannot compute PPPM
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] Error on proc 0: Out of range atoms - cannot compute PPPM

From: Axel Kohlmeyer <akohlmey@...24...>
Date: Wed, 29 Nov 2017 11:14:14 -0500

On Wed, Nov 29, 2017 at 9:35 AM, bahman daneshian <bahmanpbamp@...24...> wrote:
Dear Lammpser, 

I am permanently facing with this error. please let me know you can help me or not. Actually, I am sure about lattice and pair_style commands as I selected their parameters from text books.

​i just ran your input and - as expected - there are lots of issues, that can be easily detected by carefully examining the output and applying elementary testing/debugging steps. please find comments inline.

dimension 3
units    real
atom_style     charge
boundary       p p p

#simulation box
region box block 0 10 0 10 0 10 units lattice

​so this​ will use the default lattice (simple cubic with 1.0 1.0 1.0 spacing, i.e. it will create a 10x10x10 \AA box. not likely want you want).
create_box      2 box    

​and now create a periodic box for that.​


lattice custom 1 a1 3.786  0.00000  0.00000   a2  0.0000   3.786  0.00000  a3 1.8925  1.8925  4.7570  &
basis 0.1250 0.8750 0.7500 &
basis 0.1250 0.1250 0.2500 &
basis 0.0816 0.3316 0.8368 &
basis 0.9184 0.6684 0.1632 &
basis 0.3316 0.0816 0.3368 &
basis 0.6684 0.9184 0.6632
mass 1 47.86
mass 2 16.00

​lattice spacing will be 3.786 3.786 4.7570​

region          substrate block 0 37.86 0 37.86 0 47.570 units lattice

​this now creates a region using the lattice spacing from above while you obviously don't want that as the resulting region should have dimensions that - by far - exceed the size of the box (even with units box).
create_atoms 2 region substrate &
basis 1 1 &
basis 2 1 &
basis 3 2 &
basis 4 2 &
basis 5 2 &
basis 6 2
set type 1 charge 2.196 
set type 2  charge -1.098
group substrate region substrate

#--Phase 2----------------------------------------Buckingham Potential-----------------------------------------------
# Potential

pair_style buck/coul/long 10.0
pair_coeff   1 1   717647.40 0.154 121.067
pair_coeff   1 2   391049.10 0.194 290.331
pair_coeff   2 2   271716.30 0.234 696.888

kspace_style pppm 0.0001

timestep 5

​this looks like a rather aggressive choice of time step for this kind of system. specifically if you are having problems, it is usually a good idea to reduce it.

if you want to confirm, that you have create a proper geometry it is always a good idea to write out a data or dump file at this point and a) visualize them and check for correctness, especially with periodicity added and b) manually check a few coordinates for correctness.

next step should be to run a minimization. if your geometry is good, then there should not be significant changes. however, your system changes *massively* and starts with a gigantic positive pressure that turns into a gigantic negative pressure after minimization. that is a strong indication that something is seriously wrong here.
you should also see that your system is not neutral. that doesn't make sense for entering a crystal structure. it *should* be neutral.

velocity all create 300 12345 mom yes rot no
fix 1 all nvt temp 300.0 300.0 0.01

thermo 100

thermo_style custom   step  temp   

​turning off useful diagonostics like energies and pressure is a bad idea. you are essentially running your simulation "blind". ignoring bad news does not make it go away.​

dump  1 all image 500 image.*.png  type  type  adiam 1.0 axes yes 0.8 0.02 zoom 1.5 view 80 -30
dump_modify 1 backcolor white boxcolor white acolor 1*2 blue/red/orange/green/cyan/aqua/yellow/red
run 10000

​at this point having the pppm error is inevitable. the GI-GO​ principle applies. unless you feed LAMMPS meaningful input, it will not be able to compute meaningful results.

all of the above should be standard protocol when setting up a new system. claiming, that you took data from a textbook is no proof that you are setting up your system correctly. as shown above, there are many obvious issues. you need to improve your skills of how to confirm that commands are actually producing the correct effect.


unfix 1

Best regards,

Bahman Daneshian

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
lammps-users mailing list

Dr. Axel Kohlmeyer  akohlmey@...12...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.