Response to “Comment on ‘Theoretical prediction of crystallization kinetics of a supercooled Lennard-Jones fluid’” [J.Chem.Phys. 151, 017101 (2019)]
The Classical Nucleation Theory (CNT) describes the Gibbs free energy cost to create a crystallite of N atoms out of a metastable phase as follows:
ΔG = −N|Δμ| + γA. (1)
The first term gives the Gibbs free energy cost to create a crystallite of N atoms in its bulk phase. The term Δμ = μc − μl is the thermodynamic driving force, where μc and μl are the chemical potentials of bulk crystal and liquid phases. The second part is the contribution from the solid-liquid interface, where γ is the solid-liquid interfacial free energy and A is the area of the interface. The driving force is estimated with bulk properties of liquid and crystal phases. The interfacial free energy γ = γ0 is often estimated from its planar interface value γ0, the capillarity approximation. These independently estimated quantities lead to a nucleation profile, where the critical nucleus locates at the maximum of the profile and the resulting nucleation barrier can be used to estimate the nucleation rate. It has been a long standing goal of the classical nucleation theory to be able to predict accurate nucleation rate from these independently estimated thermodynamical properties.