Calculation of macroscopic growth rates from nucleation data

The atomic interfacial mobilities for nucleation and crystal growth are often taken to be the same, in principle allowing the cyrstal growth rate, u, to be determined from nucleation date. The validity of this assumption can be questioned, however, since nucleation and growth are generally widely separated in temperature, and the nature of the clusters involved in the two processes are probably quite different. To investigate this, temperature-dependent macroscopic growth rate calculated from nucleation data in lithium disilicate glass, assuming the classical theory of nucleation, is compared with experimental measurements. The transient time for nucleation, τ, is used to determine the interfacial mobility and three methods are used to compute u. Two assume that the temperature dependence of τ follows that of the bulk diffusion coefficient (or viscosity); the third method directly extrapolates the temperature dependence of τ into the growth regime. Good agreement between the experimental and calculated values of u is found for the Turnbull-Fisher model of nucleation assuming the temperature dependence for bulk diffusion.