Correlation Length, Timestep, Sample Interval for Thermal Conductivity

Hello all!

Do you know some basic references about correlation lenght, timestep and sample interval to calculate thermal conductivity of liquid argon?

The results of my calculations are far from experimental data…

I’ve already tried to increase simulation box, number of atoms and simulation time. Nothing works…

Why p = 200, dt = 4.0, s = 10 in lammps’ manual? (http://lammps.sandia.gov/doc/compute_heat_flux.html)

I don’t understand these values …

I’m using real units.

Thanks al lot

R. Callef