Dear all,
I am simulating a nanotube under tension, the condition is NVT with constant temperature 300K, one end fixed ,and set the velocity of the other end, the aim is to get the force-strain curve of atomswhere velocity is applied,
but during the simulation,there is the error: lost atoms origin center ** current ***
after I added “thermo_modify lost ignore flush yes” to my input script, there is no warning, I can only see the nanotube is stretched at a little strain by using VMD, but I can’t see the nanotube fractured. I wonder why this happens?
here is some part of my input script:
region 1 block INF INF INF INF INF 0.5
region 2 block INF INF INF INF 99 INF
group low region 1
group up region 2
group mobile subtract all low up
timestep 0.00001
velocity up set 0 0 1
fix 1 mobile nvt 300.0 300.0 0.00005 drag 0.2
run 10000
fix 2 up nve
fix 3 up setforce NULL NULL 0
fix 4 low nve
fix 5 low setforce NULL NULL 0
minimize 0.0000000000000001 0.000000000001 100000 100000
output
thermo_style custom step temp etotal f_3[3] f_5[3]
thermo_modify lost ignore flush yes
thermo 100
dump 1 all atom 100 cnt.lammpstrj
dump_modify 1 format “%7d %3d %10.5e %10.5e %10.5e” scale yes
dump 2 up custom 100 dump.myposition tag z
temp controllers
run 10000
thank you very much and happy new year!
Li Ming
Dear all,
dear li ming,
I am simulating a nanotube under tension, the condition is NVT with
constant temperature 300K, one end fixed ,and set the velocity of
the other end, the aim is to get the force-strain curve of
atomswhere velocity is applied, but during the simulation,there is
the error: lost atoms origin center ** current *** after I added
"thermo_modify lost ignore flush yes" to my input script, there is
no warning, I can only see the nanotube is stretched at a little
strain by using VMD, but I can't see the nanotube fractured. I
wonder why this happens? here is some part of my input script:
this is most likely a VMD "problem", or more precisely a "you not
using the proper representation in VMD" kind of problem.
i assume that you are using the default representation in VMD
(Lines). this will assign bonds based on either a provided topology
file or - if that is not available - guess bonds from distances
and element radii. VMD was originally written for biomolecular system
being modelled with classical force fields. those cannot break bonds,
so for efficiency reasons, the bonding information is computed/read
only once and reused for the whole trajectory. you have to use a
"Dynamic Bonds" representation in combination with a suitable distance
parameter to see bond breaking. for more details have a look at the
VMD user's guide and the various available tutorials. e.g.,
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-vmd/part2.html#chap4_sect5
cheers,
axel.
If LAMMPS says you are losing atoms, this is not a viz issue.
You need to figure out why. Something about where the atoms
are moving and the boundary conditions on your simulation box.
Steve