Evaluation of glass-stability criteria for chalcogenide glasses: Effect of experimental conditions

•Performance of the Hrubý, Weinberg and Lu-Liu glass-stability criteria was tested.•Their dependence on heating rate and particle size was evaluated for nine chalcogenide glasses.•Bulk samples (as opposed to powders) were found to be most suitable for evaluation of glass stability.•A novel performance index evaluating invariability of the glass-stability criteria was suggested.•Hrubý criterion exhibited the lowest normalized variability due to the experimental conditions.

The Hrubý, Weinberg and Lu-Liu glass-stability criteria for chalcogenide materials were for the first time demonstrated to be significantly dependent on the experimental conditions (heating rate and sample form). Origins of the heating rate and sample form dependencies were found to subsist in the magnitude of activation energies of involved thermokinetic phenomena and amount of stress/mechanically induced defects acting as crystallization centers, respectively. As-prepared bulk samples were found to be most convenient for determination of glass stability, both from reproducibility and simplicity points of view. Good empirical correlation of the glass stability and glass-forming ability was found for the three criteria and nine tested chalcogenide materials. A novel performance index evaluating invariability of the glass-stability criteria with respect to experimental conditions was introduced. It was found that the Hrubý criterion exhibited the lowest normalized variability due to the experimental conditions, which confirmed the superiority of this criterion when dealing with chalcogenide glassy materials.