ENVIRONMENTAL-SEM (ESEM) INVESTIGATION OF ENAMELS DURING THERMAL CYCLES: “IN SITU” OBSERVATION OF BUBBLE GROWING AND SOFTENING
Giovanni Baldi¹, Alessio Antonini¹, Luciano Pellicci², Fabrizio Bruscoli³, Gabriele Pasqualetti³ and Franco Bruni^3
1Advanced Research Laboratory, Colorobbia Group, Via Pietramarina 123 Sovigliana, (FI) Italy
²Chemical Laboratory, Via Pietramarina 123 Sovigliana, (FI) Italy
³Porcelain Enamel Division, Via Pietramarina 123 Sovigliana (FI) Italy

Introduction
In this work a new environmental microscopic technique has been applied to the study of the behaviour of enamels at high temperature.
The use of porcelain enamels in coating hot water tanks has grown during these last years: in comparison with different coating technologies, porcelain enamels show a better resistance to corrosion and lower environmental costs.
In order to improve the resistance of the enamel coating, bubble structure has to be taken into consideration.
Bubble growth, especially in electrostatic powder application, is a complex phenomena and takes place when gas develops from metal substrate in correspondence to the melting stage of the coating.
We have analysed tree kind of steel plates with different carbon contents and thickness. These have been enamelled with a formulation having a very high resistance to hot water and steam and then submitted to a thermal cycle simulating the industrial firing process.
ESEM observation has been finally acquired in a video to understand coating behaviour during this thermal cycle.

Experimental procedures
Three sample plates analysed in this work have different carbon content and thickness, respectively 1.7 mm thick with 0.36 % C, 1.7 mm thick with 0.024 % C and 2.1 mm thick with 0.026 % C.
The enamel was made by a mixture of three different frit named A, B and C, the chemical analysis can be found in Table 1.
Each steel plate has been cut in 3 mm. squares and adapted to measure with the ESEM hot chamber.
Viscosity measurements (Vezas, High Temperature Viscosimeter) and observations with the heating optical microscope (Expert System, Misura) of each enamelled component have been also performed in order to show different fusibility behaviours at high temperatures.
Each thermal measurement has been observed and acquired in a video clip using Philips Environmental Scanning Electron Microscope equipped with a hot chamber Anton Paar operating at low vacuum conditions. The thermal cycle was studied to best simulate industrial firing process: a ramp from room temperature to 550 °C with a slope of 150 °C/min and from 550 °C to 850 °C with a slope of 20 °C/min.
Due to some instrumental limitations of ESEM, pressure conditions were kept 2.9-3 torr. above ambient.

Fig. 1 Temperature 650 °C.

Fig. 2 Temperature 680 °C.

Fig. 3 Temperature 710 °C.

Fig. 4 Temperature 730 °C.

Fig. 5 Temperature 750 °C.

Fig. 6 Temperature 755°C.

Fig. 7 Temperature 760 °C.

Fig. 8 Temperature 770 °C.

Fig. 9 Temperature 800 °C.

Fig. 10 Temperature 850 °C.

In Fig.1 it is possible to distinguish three different frits during the first stage of melting, where frit A and B are starting to melt, frit C is going to soften and melt. This process goes forward until the first bubbles appear.
Bubbling continues until temperature reaches 850 °C and during the early stages of cooling phase at 840 °C, as showed in Fig. 11.
It is remarkable that all three plates with different carbon contents show a similar behaviour during the thermal cycle.

The curve of viscosity vs. temperature was taken with a high temperature viscosimeter VEZAS in a range 1000 – 700°C for the first two frits, and 1300 – 900 °C for the third frit, confirming the differences in melting temperature observed in ESEM experiment.

Table 2

At higher temperature the dynamic of bubble growing and their escape from the surface were observed and recorded in a video clip.
We followed the thermal cycle to the maximum temperature and during the cooling phase until 800 oC. A complex process of grow and escape of bubbles in the enamel and on the surface was observed even during the first phases of cooling.

Fig. 11 Bubble forming during cooling phase at 840 °C.

Conclusions
The dynamics of bubble forming in the enamel during a firing cycle do not seem to be strictly connected to the content of carbon in metal substrates, even if normally carbon free steel shows an improved bubble structure. The different viscosity of frits constituting the enamel appears correlate to the growth and escape of bubbles at certain temperature: a possible mechanism could be found in preferential paths formed in the enamel at the first stage of melting; whereas, during the cooling phase, bubbles remain entrapped into enamel due to the rapid increasing of viscosity of the melt.
In this work we use a new powerful technique in the study of defects: the possibility to see the emerging of the bubble structure in real time and the behaviour of the melt permit us to understand the complex mechanism beneath the process of enamelling.
This powerful technique has been used even in the complex technology of “two coat one firing”, where the bubble structure of the Ground coat has a big influence on the final result.