The dependence of homogeneous nucleation rate on supersaturation

The claim that classical nucleation theory (CNT) correctly predicts the dependence on supersaturation of the steady-state rate of homogeneous nucleation is reexamined in light of recent experimental studies of nucleation of a range of substances, including water, argon, nitrogen, and several 1-alcohols. Based on these studies (which include, for water, a compilation of nine different studies), it is concluded that the dependence of nucleation rate on supersaturation is not correctly predicted by CNT. It is shown that CNT's incorrect prediction of the supersaturation dependence of nucleation rate is due to its incorrect prediction of the Gibbs free energy change associated with formation of small clusters from the monomer vapor, evaluated at the substance's equilibrium vapor pressure, even though that free energy change is itself a function only of temperature.