shows the dependence on the measured onset
temperature with the heating rate for all the platforms.
In the range of dT/dt used (0.5 to 8 °C.min1
) the thermal
lag varied from ~4.5 to ~11 °C. No significant
changes are seen between the different platform lengths
for any given heating rate. In a previous work [6] in was
found that the temperature along the sample length
varied, the regions near the furnace wall being hotter and
the region in the middle of the sample surface cooler.
Therefore, it should be expected that, on average, the
water in the sample would melt at lower measured temperatures
for greater platform lengths. This seems not to
be the case in a real sample, where the mean temperature
is not influenced by the sample length. A possible explanation
for this behaviour could be the high contribution
to the thermal lag due to the thermal resistance of the
sample.