Thermal properties
The results are in Figs. 1–5. Table 5 summarises the temperatures measured by DTA and heating microscopy. From these values, only one exothermic peak relating to the formation of crystalline is observed, exception made for frits B and D. In frit B there are two crystallization peaks (see Fig. 2) whereas in frit D there no peak (Fig. 4). Additionally, frit D is the only frit presenting a ball temperature (B).
From the analysis of given above it is possible to establish a relationship between the sintering behaviour and the crystallization process, for each frit. Indeed, considering Table 5, one can see that the glass transition temperature (Tg) always occurs before the sintering temperature (S) while the softening temperature (M) always happens after the crystallization event. To better visualize this correlation, we plotted the characteristic temperatures given in Table 5 on the graphic of sample height versus firing temperature. Figs. 6–8 show some examples of this representation for frits A–C.
From these figures it is also evident that sinterization finishes when the crystallization process occurs. That is, during crystallization the characteristic sample shrinkage, due to the sintering process stops. On the other hand, softening occurs only once the crystallization process is finished. Therefore, these results prove that crystal formation interferes with the sintering process of a crystalline frit. For frit D it was not possible to establish this type of correlation. In fact, it was explained before that this frit did not exhibit an exothermic reaction, probably because there was no devitrification of a crystalline phase. That is the reason why, for frit D, the softening temperature and the other temperatures measured by heating microscopy occurred earlier, the sintering behaviour showing the existence of a ball temperature.