Nucleação transiente e estacionária em dissilicato de lítio

Abstract

Glasses and supercooled liquids tend to crystallize below the melting temperature. Crystallization is a phase transformation where the atomic structure ceases to be non-crystalline and becomes periodic. The focus of this work is on the process of emergence of one or more crystalline regions in a super-cooled liquid, known as crystal nucleation. The Classical Theory of Nucleation (CNT) is one of the theoretical bases that explain this process. Experimental results and previous analyzes show that the CNT is sufficient to describe the crystal nucleation rate as a function of temperature, but only for temperatures above the maximum nucleation temperature. Another issue is the way experimental data are obtained, generally using the double-treatment technique. This technique is known to overestimate the intrinsic induction time of nucleation. The objective of this project is to investigate these two CNT problems in a rigorous way to seek a resolution for both. Our results showed that the time of the nucleation treatment has a considerable influence on the values obtained for both the stationary nucleation rate and the induction time. Therefore, experiments performed for too short times may be the cause for the disagreement (theory / experiment) found below the maximum nucleation temperature. In this work, using longer nucleation times, the obtained nucleation rates were higher than those reported in the literature and the detachment between theoretical prediction and experimental values occurred at lower temperatures.