A few topics concerning nucleation and crystallization in glasses

Abstract

Herein a review is presented of some of the work we have performed on the subject of crystallization of glass. The various topics fit into two catagories: crystal nucleation in inorganic glasses and formal theory of phase transformations (i.e., Johnson–Mehl–Avrami–Kolmogorov theory denoted as JMAK theory). Our nucleation studies have focussed primarily upon testing the applicability of classical nucleation theory (CNT) to crystal nucleation experiments in simple, pseudo-one-component systems. The general conclusion of the work to date is that this theory does not provide an accurate description of crystal nucleation in common glasses. One of the key unknown parameters in this theory is the liquid–crystal surface tension. A description is given of one method of determining the latter quantity without attempting to fit the surface tension to obtain accord between the results of steady-state nucleation rate experiments and theory. This new method utilizes transient nucleation data and the theory of transient nucleation formulated by Shneidman. The work in the formal theory of phase transformations has concentrated upon providing extensions of JMAK theory for certain isothermal transformation processes. For example, we have examined the changes which are needed in JMAK theory when one accounts for inhomogeneous nucleation (non-random) and finite sample size effects. The results of this study have been used to describe surface crystallization rates and combined bulk and surface crystallization rates. Also, we have formulated some of the theory which deals with crystallization processes in which non-spherical, anisotropic particles are formed. The 1-D versions of our equations have been compared with computer simulation results and have proven highly accurate. Finally, we have presented a correction to the JMAK t 4 law in the case of transient nucleation effects which is self-consistent in that it takes into account equally significant size dependent crystal growth effects. The results of the theory are in excellent agreement with computer solutions of the nucleation equations.