I am trying to run tension and compression simulations on an amorphous cell model that I constructed using Material Studio. The purpose is to determine the Young’s modulus for pure polyamide and polyamide reinforced with Silica nanoparticles.
However, before running the tension and compression tests, I have to minimize the structure in LAMMPS. I have attached the input file for this minimization process as well as the file containing the amorphous cell parameters. Upon running the process on 1 processor, the following error occurs:
ERROR on proc 0: Bond atoms 4545 4548 missing on proc 0 at step 1880
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
rank 0 in job 1 ubuntu_33789 caused collective abort of all ranks
exit status of rank 0: killed by signal 9
If I run the same process on 8 processor in parallel, something similar to the following shows up 8 times with different numbers each time:
rank 0 in job 1 ubuntu_33789 caused collective abort of all ranks
exit status of rank 0: killed by signal 9
Is this a problem with the model itself or am I not running it correctly? I am a newbie here and would appreciate any kind of help in this matter. If it’s a problem with the model itself, can I construct the model in LAMMPS itself rather than using Material Studio?
Thanks
Regards,
Saaket
P.S: I’m using the 5-Jun 10 version of LAMMPS and have been able to run processes on multiple processors in the past, indicating that MPICH is working fine.
This error is b/c a bond is stretching so far the atoms
on either end can't be found on the same processor. You need
to figure out why your dynamics is allowing that to happen - e.g.
overlapping atoms, pulling too far, too fast, etc.
But I am getting the same error when I’m running the process on a single processor too. If its running on 1 processor, shouldn’t it be able to find the bond on that processor?
Also, I tried a number of different permutations for my dynamics code (which is basically just the minimize command). I also tried using communicate and min_modify to get around this problem. However, nothing seems to be working and no matter what values I chose for these commands, the process always failed at the 160 time step. I then restarted the machine, and now no matter what values I choose, the process fails at the 108 time step. If possible, could someone run the code on their machine? I am attaching the input files again. I have been stuck at this for a number of weeks now and have tried everything that I could think of.
I was wondering if using LAMMPS, is it possible to coat a layer of surfactant (Silane) to a Silica particle and then enclose it in a polyamide cell such that there is chemical bonding between the Silica-Silane and the Silane-Nylon? Right now, I am using Materials Studio to construct the initial model and then exporting it to LAMMPS to run the dynamics. Normally, the interaction between the Silica and Nylon is just due to Van der Waal's forces so I want to add Silane between the two in order to increase the bonding. If possible, can you tell me how to go about doing this? I have tried looking it up online but could not find much information on it.
That sounds like a Matl Studio question. From the LAMMPS
perspective, it's just input data. However you create the
model, with some builder, is fine. Maybe others on the mail
list want to comment.
Thanks for your response. So is it possible to achieve that if I construct the model in LAMMPS itself and forget about Materials Studio? Or does LAMMPS need an input file from some other builder?
So is it possible to achieve that if I construct
the model in LAMMPS itself and forget about Materials Studio? Or does LAMMPS
need an input file from some other builder?