Project Details
Projekt Print View

Stochastic approach to heterogeneous crystal nucleation in silicate glasses

Subject Area Thermodynamics and Kinetics as well as Properties of Phases and Microstructure of Materials
Term from 2017 to 2020
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 329439308
 
Final Report Year 2021

Final Report Abstract

Cooling of hot silicate melts (> 1000 °C) to room temperature free of crystals is an essential precondition for glass formation and for hot-forming of glass articles in industry. The undesired devit- rification during cooling the melt is generally triggered by crystal nucleation at external sites (so-called heterogeneous nucleation). Active external sites can be found at the three phase contact of the melt with the gas atmosphere (in industrial processes air is mostly present) and surfaces of materials used for forming or handling the hot melt. Crystal nucleation is a random (stochastic) event. Reliable predictions on the crystal nucleation rate in terms of probabilities can only be made from experiments that have a large number of trials. Against this background, the project seeks to establish a novel method to gain such a prediction from a high number of repetitive cooling-heating cycles (~ 300) of a small liquid volume (ml) of a glass-forming melt. As crystallization generates heat, a special device (differential scanning calorimeter DSC) was used to detect the time when crystal growth starts during cooling in each run. The survival analysis of the research approach brought us to the corresponding type of statistical experiments used in medicine and public health. Key point was to identify that the nucleation rate is related to the conditional probability of survivorship of the melt until the time of the nucleation event or later, which is known as the hazard rate. Thus, the development of the hazard rate expression for cooling the melt at a constant rate is the first important outcome of the project. With it, we were able to provide temperature-dependent nucleation rates for different kinds of settings. Among these were cooling paths under different oxygen levels of the gas atmosphere and usage of containers of different materials. We found that cooling in noble metal containers led to crystal nucleation at higher temperatures than carbon- based materials. The preference of nucleation at high-temperature in case of noble metal surfaces re- sulted also in the occurrence of an uninspected second type of the crystals, which were most likely nucleated at different sites. The second type of crystal nucleation became stronger for a lower level of oxygen of the gas atmosphere, which suggests that noble metals can react with oxygen at high-temper- atures ≥ 900 °C giving preference for the formation of active sites of the first (proper) type of crystals. This outcome is important in terms of the industrial practice since long-term contact of the melt with noble metals at nozzles (bushings) and slits for down-drawing of fibers and thin glasses products bear the risk of potential crystallization that can be possibly minimized by changes of the surrounding gas atmosphere and the type of the noble metal alloy.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung