Effect of liquid phase separation on crystal nucleation in glass-formers. Case closed

The objective of this critical review paper is to shed light on a longstanding, relevant problem in materials science: what is the possible effect of liquid phase separation (LPS) on crystal nucleation? Several decades after the first published report on this subject, the widely discussed (none? Interfacial or compositional?) role of LPS on crystal nucleation is still a matter of debate and controversy. Here, the most relevant results of a research project on the kinetic analyses of LPS and simultaneous crystallization of BaO-SiO2 and Li2O-SiO2 glasses are reviewed, considering and substantially complemented by recent findings. Crystal nucleation kinetics in glasses having compositions inside and outside the miscibility gaps in both systems were determined by microscopy methods at different temperatures and complemented by small angle X-ray scattering and transmission electron microscopy analyses to probe the LPS kinetics. The effect of the liquid droplet interfaces was found to be negligible - their interfacial energy, approximately 0.004 J/m(2), is much smaller than that of the crystal/nucleus surface energy, 0.130-0.200 J/m(2) - and the number of vitreous droplets is many orders of magnitude higher than the number of crystals. The old and new results and an improved analysis demonstrate that the crystal nucleation rates in the stoichiometric glasses are higher than in their (alkali-poorer) phase separating glasses of the same system. These combined results clearly show that primary role of LPS is to shift the composition of the glass matrix towards that of the stoichiometric crystal phase, which leads to enhanced crystal nucleation. These findings settle an old-standing enigma in glass science.