Dear support, I need to do a simulations with ReaxFF and another one with Martini FF; there are some project about implementation of these force fields?
Regards.
N.V.
dear nicola,
Dear support, I need to do a simulations with ReaxFF and another one with
Martini FF; there are some project about implementation of these force
fields?
AFAIK, martini just uses regular lennard jones and coulomb.
that should already be implemented in LAMMPS. all you'd have
to do is to generate proper data files for your system.
the implementation of ReaxFF has been discussed on this list
multiple times and is supposedly nearing completion.
cheers,
axel.
Observation(1 of 2) I am constructing Aluminum single crystal lattice box using 4.05 Angstrom lattice constant at 0 K, oriented [1-1 0] along X and [1 1 1] along Y axis with periodicity in all the directions.
However when I construct it and let it run a couple of steps using NVE the temp and pressure increases to ~ 150 K and ~ 9000 bars (from 0 K and 540 bars) and keeps fluctuating by large amounts (+- 1000 bar pressure).
Here is the log o/p:
--------------start log-----------------------------------
LAMMPS (21 May 2008)
Lattice spacing in x,y,z = 5.72756 7.01481 6.61362
Created orthogonal box = (0 0 0) to (85.9135 105.222 66.1362)
4 by 1 by 4 processor grid
Created 36000 atoms
36000 atoms in group mobile
WARNING: Temperature for thermo pressure is not for group all
Setting up run ...
Memory usage per processor = 2.53796 Mbytes
Step Temp E_pair E_mol TotEng Press Volume
0 0 -118921.61 0 -118921.61 504.62579 597871.12
100 178.58236 -119753.41 0 -118922.42 12845.038 597871.12
200 170.22992 -119714.74 0 -118922.62 12046.226 597871.12
300 234.66014 -120014.4 0 -118922.47 9193.535 597871.12
400 177.97712 -119750.66 0 -118922.49 10413.375 597871.12
500 163.06517 -119681.2 0 -118922.42 10314.166 597871.12
600 182.0137 -119769.41 0 -118922.45 9092.424 597871.12
700 168.39333 -119706.06 0 -118922.49 10199.216 597871.12
800 193.39162 -119822.57 0 -118922.68 11253.99 597871.12
900 186.50156 -119790.45 0 -118922.62 8614.744 597871.12
1000 156.72693 -119651.5 0 -118922.21 9245.482 597871.12
Loop time of 37.3914 on 16 procs for 1000 steps with 36000 atoms
----------------end log of observation 1 of2 -------------------
Observations (2 of 2) When I construct using [001]X, and [010]Y or another orientation particularly [-1-1 2]X and [1 1 1]Y the box size seems to be doing good in fitting the lattice and the temp and pressure are almost the same when I run NVE steps.
So it seems the box size might be off when constructing the initial rotated lattices in SOME cases. Has anybody experienced this (or you can point out if I am doing something incorrect?); if yes then is there any way to take steps so that the initial configuration built would be stable for NVE, instead of trying to relax using NPT ?
Thanks
Nitin P. Daphalapurkar
If you orient the lattice in non-coord directions, you need to
insure the box size and periodicity is what you exepect. The
lattice doc page talks about this. The lattice command prints
out what it's repeat distance is, which may not be what you expect.
Try doing a small system and viz it. And study the command
output and doc page.
Steve