The effect of cooling rate on the specific volume of glassy phases

The specific volume of glasses and ceramics depends on the cooling rate applied during the transition from a viscous liquid to a solid glass. A slow cooling rate allows molecules to reorganize from the disordered liquid structure to the partially ordered glassy phase, with a dramatic reduction in the coefficient of thermal expansion and a smaller specific volume. Faster cooling rates, on the opposite, will not give the time for such rearrangement, producing larger final specific volumes: the faster, the larger. This behaviour is characteristic of most inorganic glassy phases, and particularly so for silica-based.

If the cooling rate is faster than the transmission of heat through the material, large rate differences can develop between the edges (faster) and the bulk of a sample (slower). These cooling rate differentials translate in specific volume inhomogeneities: the edges and external surface can have significantly higher specific volume than the bulk, leading to residual tensions and increased fragility of the whole sample.

The ELS-MDF Optical Dilatometer directly measures the effect of cooling rate on the specific volume of glasses, ceramics and glass-ceramics as produced in the real industrial process. For this purpose, a specimen is cropped form a piece obtained from the factory, and tested with a thermal cycle which rises above its glass transition (but largely below its softening point) and slowly cools down to room temperature. If the industrial process was faster than the test cooling rate, the test result will be a shorter specimen, because the slower thermal cycle allowed the glassy phase to completely relax to its equilibrium volume.

We call this a Volume Variation test, and we quantify the volume difference between the heating segment and the cooling segment of the curve at a fixed temperature, e.g. 100°C. The following graph shows a slow cooling test (-10°C/min) performed on a specimen from a ceramic tile fired in an industrial single-channel roller kiln. In this case, the volume difference at 100°C is 0.0521%.

The ELS-MDF Optical Dilatometer can simulate a complete industrial firing cycle from a raw sample, thanks to its fast heating rate (up to 80°C/min) combined with the new Flash Cooling instrument option. The Flash Cooling realizes the abrupt cooling experienced by the material, for example in a roller kiln, by opening the motorized furnace at the end of the heating cycle.

The next plot shows a sample being fired at 66°C/min and then flash-cooled:

The specific volume excess caused by the flash cooling can be assessed by repeating a measurement on the fired sample, where it is heated up past its glass transition and then cooled down slowly (-10°C/min). The resulting volume difference at 100°C would be 0.077%.

By comparison, if the same raw sample is fired and then allowed to slowly cool, the volume variation test will produce a volume difference of 0.042%.

By slowly cooling the sample after firing, the volume variation test gives a volume shrinking which is 45% smaller with respect to a flash-cooled sample.

The volume variation test, combined with the flash cooling, is a valuable tool for the material designer looking to minimize residual stresses in traditional ceramics, glasses and glassy materials in general.