problem in energy minimization

Dear All
I am trying to simulate SAM monolayer of allkylthiol. Initially I build one thiol molecule using Gaussian and did energy calculation in Gaussian. Using vmd tcl script I create SAM monolayer.
When I am trying ti minimize the SAM I am getting the following error.

Step Temp E_pair E_mol TotEng Press
0 0 29841.789 0.93942272 29842.729 22952534
ERROR on proc 0: Angle atoms 1 2 3 missing on proc 0 at step 12 (neigh_bond.cpp:131)

data.hexan_lj_dih (28.3 KB)

Dear All
I am trying to simulate SAM monolayer of allkylthiol. Initially I build one
thiol molecule using Gaussian and did energy calculation in Gaussian. Using
vmd tcl script I create SAM monolayer.
When I am trying ti minimize the SAM I am getting the following error.

have you tried increasing your box in z direction?

axel.

Dear Axel
Yes already I tried with increasing box size in z direction but still the problem was exist.

Thanks

Section_errors.html of the doc pages explain
what this error means (and all others). It means
your dynamics is bad and one atom in the angle
is moving so far, that a processor can't find it.
That should not normally happen if dynamics are good.
If you print thermo output every step, are things well
behaved before the error appears?

Steve

Dear Steve
Thanks.
I increased neighbor value, Now I am getting the same error after 132 step.
I am attaching my log file.

log.lammps (52.8 KB)

Can you also post the data file? I am
suprised that such a small system can
not find an atom when running a minimize.
How many processors are you running on?
Are you certain that all your bonds are between
nearby atoms, i.e. your molecule topology
is correct?

Steve

Dear Steve
I attached my data file.
I am running on single processor.
I build one thiol molecule in Gaussian and did energy calculation in Gaussian. Created SAM using tcl script in VMD. Using topotool in vmd I created lammpsdata file. So I think my molecular topology is correct. You think that the parameters which I mentioned in the data file may give incorrect topology?
Thanks
Mathi

data.hexan_lj_dih (28.3 KB)

Dear Steve
I attached my data file.
I am running on single processor.
I build one thiol molecule in Gaussian and did energy calculation in
Gaussian. Created SAM using tcl script in VMD. Using topotool in vmd I
created lammpsdata file. So I think my molecular topology is correct. You
think that the parameters which I mentioned in the data file may give
incorrect topology?

you seem to have overlapping atoms due to periodic boundaries.
just enlarge your box, e.g. by increasing xlo, ylo, zlo by 1 angstrom
and the input works.

axel.

If you put these lines in your input script you
will see the problem:

variable z1 equal z[1]
variable z2 equal z[2]
variable z3 equal z[3]
thermo_style custom step v_z1 v_z2 v_z3

One of your atoms in angle 1 2 3 is moving
far outside the box in z, which you have
specified as type f, which means you will
lose the atom. That only happens
on a reneighbor step. The first time it
reneighbors is step 132, so it is saying
it can't find the atom anymore (b/c it was
deleted b/c it was outside the box).

Axel said to increase the box size in his
first response, which works. Try zlo = -5.
Or use a shrink-wrap boundary condition.

Steve

Dear Axel and Steve
Thanks. Problem is solved.