[lammps-users] increasing the density and amount of particles in the simulation box

was really looking to increase the density of the atoms in a simulation. say i have a box 10 x 10 x 1 i want to keep it that size and just increase the density of the atoms inside that box. I've tried using the lattice command but the smaller or larger i scale it seems to make no difference in the total amount of atoms. Is there a way to change that without reading in positions from a data file?


Maybe a naive suggestion, but what about running it in a barostat (NPT ensemble) at high pressure? That should certainly increase the density (by making the volume smaller)... then you could take a configuration at the end of the high pressure run and do whatever you want with it, run other sims etc. It will be easier to do this than it will be to insert more atoms into an already full box, IMO...

Try the following code snippet:

lattice fcc 1
region mybox block 0 10 0 10 0 1 units box
create_box 1 mybox
create_atoms 1 box

Changing the density (the numerical parameter in the lattice command)
changes the number of atoms that are placed in the simulation box.


Hi all,

I have a simple question but maybe useful to the entire community.

I was wondering if there are any ways to dump files after the minimization step (not during the minimization process) so that the dump file would not be so huge and not missing the important snapshots since we cannot predict which step the minimizor would stop. Any suggestions?


From the manual: “Dumps are performed on timesteps that are a multiple of N (including timestep 0) and on the last timestep of a minimization if the minimization converges.”

Have you tried setting the dump frequency to be the same as your maximum allowed iterations for minimization? That should give you at most two frames in your dump file.



Thanks for your reply. Yes, it’s amazing that it works. I do appreciate your help.


Using the lattice and create_box and create_atoms commands
you can create a box of any size with as many atoms as you
want. If you want a periodic system it's up to you to
insure that the box length is a multiple of the lattice spacing.
Note that the region command used to create the region give
to create_box has a units option (box or lattice) to allow you
to specify the box size in multiples of the lattice constant or


Not so amazing :slight_smile: From the doc page for dump:

Dumps are performed on timesteps that are a multiple of N (including
timestep 0) and on the last timestep of a minimization if the
minimization converges.


2008/10/30 AJING CAO <[email protected]>: