Abstract
Recent careful studies showed that the phase separation of a system of hard disks with strongly positive nonadditivity predicted by Tenne and Bergmann [J. Chem. Phys. 70, 1952 (1979)] using scaled particle theory is dubious. As exact computer calculation results are still lacking for these systems, we demonstrate here the phase separation by molecular-dynamics calculations with hard soft-sphere mixture (HSS) potentials (the softness of the potential is so small that it can be regarded as a hard potential). The HSS potential represents the hard-sphere (HS) potential with regard to computation of fluid properties to a very good approximation. HSS systems with extremely positive and negative nonadditivity are considered. While the former separate already at a density of only 3.5% of the closed-packed value, the latter remain homogeneous up to very high densities, as expected. Apparently the numerical method used by Tenne and Bergmann fails to account correctly for the phase separation in strongly (positive) nonadditive HS systems, while it gives good results for the Widom-Rowlinson model recently investigated by computer calculations.