Spreading out of molecules in a simulation of a droplet over a substrate

I am trying out a simulation of a droplet of a Lennard-Jones fluid over a horizontal surface.
But instead of assuming a near-spherical shape over the surface, the fluid particles appear
to spread out.

I first create a rectangular box of fluid particles over a horizontal surface using
the create_atoms command. And then, I define random lj/cut potentials for the
interactions between different pairs. Temperature is kept at 300K and rescaled
every 100 timesteps.

#LJ potentials
pair_coeff 1 1 0.1553 3.166 9.8

pair_coeff 1 2 0.1045 3.19 9.8

1: fluid particles
2: substrate particles

Any kind of help will be appreciated.

I am trying out a simulation of a droplet of a Lennard-Jones fluid over a
horizontal surface.
But instead of assuming a near-spherical shape over the surface, the fluid
particles appear
to spread out.

please explain, why should the droplet remain spherical.

I first create a rectangular box of fluid particles over a horizontal
surface using
the create_atoms command. And then, I define random lj/cut potentials for
the

what do you mean by "random potentials"?

interactions between different pairs. Temperature is kept at 300K and
rescaled
every 100 timesteps.

it has been explained here many times that temperature rescaling is a
bad thing for several reasons. more importantly, it has no real
purpose for simulation like you describe since there is no bulk that
your droplet couples to.

axel.