Temperature surges simulating calcite surface with water
Dear Lammps users,
I have been trying to simulate a surface of calcium carbonate surface in contact with water. Using the force field published by Raiteri et al in J. Phys Chem C 2015, 119, 24447-24458 DOI:10.1021/acs.jpcc.5b07532
The version of Lammps I am currently using for testing is the Ubuntu binary, obtained in ppa:gladky-anton/lammps
The structure data file was created using Molteplate. The dynamics computation details are the same as published in the above mentioned paper (as accurately as described in the article) The full input files are attached.
To facilitate the reading here are my main input parameters
units metal
atom_style full
boundary p p p
read_data system.data
include forcefield
variable Nrun equal 10000
variable Nf equal ${Nrun}/1000
variable Ne equal 10
variable Nr equal ${Nf}/${Ne}
variable Text equal 300.0
velocity all create 300 1234
velocity all scale 300.00
neighbor 2.0 bin
neigh_modify every 1 delay 0 check yes
timestep 0.001
fix integrate all nvt temp ${Text} ${Text} 0.1 tchain 5
compute T all temp
fix TempAve all ave/time ${Ne} ${Nr} ${Nf} c_T
thermo_style custom time temp f_TempAve etotal pe enthalpy cella
thermo_modify flush yes
thermo ${Nf}
dump trj all atom 100 sys.lammpstrj
dump_modify trj scale no sort id
run ${Nrun}
With this, the initial lines of the thermo output are:
Setting up Verlet run ...
Unit style : metal
Current step : 0
Time step : 0.001
Per MPI rank memory allocation (min/avg/max) = 38.53 | 38.53 | 38.53 Mbytes
Time Temp f_TempAve TotEng PotEng Enthalpy Cella
0 300 0 -31404.299 -31733.136 -31917.052 42.186
0.1 1029.5886 795.24135 -27208.552 -28337.111 -28067.012 42.186
0.2 758.85354 832.2868 -25304.993 -26136.793 -28280.761 42.186
0.3 790.00988 817.5759 -24610.583 -25476.533 -22370.756 42.186
0.4 752.93261 762.70952 -24623.035 -25448.344 -23777.292 42.186
0.5 645.94172 696.76808 -24416.257 -25124.291 -26312.212 42.186
I have been searching the mailing list, but finding working input files of surfaces is a bit difficult. I have also scanned the manual, changed the thermostat to Berendsen, that one controls the temperature very well, but similar system was published before with Nosé-Hoover. The paper cited uses npt, I have done so as well with the same results, but I have tried to reduce the problem
I would like some advice regarding the jump in the temperature from 300 K to over 1000K. It will gradually return to 300K but that initial surge causes huge relaxations that are not expected nor seeing with other codes.
I have tried other systems Ca, CO3 ions in solution. The temperature behaves very well in that case, but as a calcite surface is being published before, made me think I am missing something important in my input file.
Please note the run time used here is for testing purposes only.
Best regards,
Sergio