Dear LAMMPS users,
Hello, I am having an issue with minimizing my energy. I don't think the issue is specifically with my minimization command, though. I can successfully minimize energy only if I'm using a small number of molecules: When I create a box of 27 (3X3X3) molecules (Tetrolic acid: http://www.chemspider.com/Chemical-Structure.61810.html), the minimization step completes just fine, minimizing the energy from -178.857129425 to -262.013061474 Kcal/mol (I am using real units).
However, this simulation is unstable. Particularly the hydrogens on the OH groups begin flying around at exceptional speeds, quickly raising the temperature to almost 2000K before one of them leaves and cannot be found. It may be worth noting that, even though I run 100 timesteps with random velocities before minimization to get rid of symmetry, the minimized configuration (as seen in VMD) is very symmetrical, with all molecules ending up in the same orientation with equal spacing. Before failing, several atoms begin overlapping (but they approach each other very slowly until they are nearly in the same position). I'm not sure if this has anything to do with my problems minimizing larger simulations.
In simulations with 4X4X4 molecules and larger, the energy begins at some sort of reasonable value. Afterwards, however, the total energy is on the order of 10^16 or 10^17 Kcal/mol, at which point LAMMPS obviously refuses to run the simulation. I had this problem once before, but fixed it and the cause must be different this time. The first time I ran into this, the issue was that my bond coefficients, pair coefficients, etc. were in a different order in my input script than they were in my data file. That is no longer the case and I am fairly positive that all those values are correct and in the right order at this time. If anybody else has had this issue, do you have any idea what my mistake might be? In case you would like to run my simulation, I've posted some text files online. Copy and paste them into a text editor and make sure to name the data files 'data.TTA' so that the input script can call it correctly.
data file with 27 TTA molecules: (This one should get through the minimization step just fine, but fail before completing the simulation (5000 timesteps to start)
http://justpaste.it/j3rv
data file with 64 TTA molecules: (This one should not minimize correctly. Energy before minimization: ~ -440 kcal/mol. After: ~ 3.567e17 kcal/mol
http://justpaste.it/j3rz
Input file:
http://justpaste.it/j3ru
*NOTE: depending on your processor, you may need to change or remove the first line. My computer has an i7 (which has 4 cores, 8 threads) so I tried to make use of all of them, but you could just remove that line. The simulation does not take very long.
If you need any more information, let me know. Thanks very much in advance for the help!
N.H.