Abstract
The fourth and fifth virial coefficients of an additive hard sphere mixture, predicted from our recent formulae, are compared with the values calculated by Saija et al. and Enciso et al. Our formulae are supported by recent simulation and theoretical results. The Saija et al. and Enciso et al. values are calculated from what are purported to be exact formulae but differ drastically. In particular, Saija et al. claim that the fifth virial coefficient can be negative whereas Enciso et al. question this result. Our estimates agree closely with the results of Enciso et al. and the estimates of Boublik and Mansouri et al. This does not mean that the Boublik and Mansoori et al. equation of state is accurate under all circumstances, as our formulae lead to rather different results, including demixing, when the large spheres are present in small concentrations.