Our letter entitled Curvature corrections remove the inconsistencies of binary classical nucleation theory has been accepted for publication in Phys. Rev. Lett.
In the letter we have studied the condensation process in vapors of supersatured alcohol—water mixtures. The nanoscopic droplets forming in such vapors have highly curved surfaces, making their surface tension very different from that of a macroscopic, planar surface. We developed a method to estimate curvature effects on surface tension, and found that it has a dramatic impact on model predictions for the rate of formation of such droplets. Current models, that do not account for curvature effects, yield rate predictions that deviate many orders of magnitude from experiments, and are moreover marred by physical inconsistencies. A key inconsistency is that classical nucleation theory (CNT) can give a negative number of particles of some of the components in the critical cluster. We show in the letter that incorporating curvature effects removes these inconsistencies and restores near-quantitative agreement with experiments. The figure below shows the characteristic “hump” in the activity plot from CNT for the highly surface active water-propanol mixture, which due to the first nucleation theorem means that one of the components has a negative number of particles in the critical cluster. The figure further shows that curvature corrected nucleation theory removes this peak and restores good agreement with experiments. We also show that this is true for the water-methanol and water-ethanol mixtures.