To quantify the film drop emission, we analysed aerosols collected during experiments with and without bubbling, using granulometric and gravimetric techniques. Without bubbling, the particles detected come from the vaporization of the steel charge which is clearly visible in the form of fumes inside the furnace.
Granulometric analyses of the experimental dust samples were performed by wet laser diffraction. The Coulter LS 130 granulometer used gives the volume distribution of a suspension of particles ranging from 0.04 μm to 2.103 μm. Dust collected on membranes is dispersed in pure ethyl alcohol and desagglomerated by ultrasonic and mechanic agitation following the procedure described in figure 18. Such a procedure leads to an optimal desagglomeration of the particles and a good reproducibility of the measurements.
Figure 18. Procedure for the preparation of the particle suspension and for the granulometric analysis
Figure 19 shows the typical results of a granulometric analysis of two kinds of dust samples (with and without bubbling). They confirm the existence of two particle populations: submicronic particles coming from the vaporization of the liquid steel, and film drops. Concerning the first population, we do not have yet a satisfactory explanation for the presence of three modes. The important feature in the figure is the apparition of the biggest particles in the sample collected with bubbling.
Figure 19. Size distribution of two dust samples collected with and without bubble bursting (bubble diameter: 7 mm)
From such granulometric results and knowing the amount of dust collected with and without bubbling (see below), it is possible to obtain the size distribution of film drops by “substracting” both types of results. Figure 20 presents the distributions obtained for three bubble sizes. It can be seen that the size spectrum broadens up when the bubble size increases, even if most of the film drops remain under 20 μm, as in EAF dust samples. This phenomenon is confirmed by the SEM observation of the samples.
Figure 20. Size distribution of film drops produced by bubble bursting.
Bubble diameter: a) 8.3 mm, b) 10.2 mm, c) 12.5 mm The mass of particles collected was determined by weighing of the glass fiber filters before and after each experiment. The difference between the results of the experiments with- and without bubbling gives the mass of film drops collected. The results are gathered in table 1; for an easier and more meaningful comparison, the amounts have been referred to one bubble burst (MB in table 1) and to the volume of injected gas (MG in table 1).
Table 1. Mass of film drops as a function of the parent bubble size
The results clearly show that the amount of film drops produced by one bubble burst greatly increases with the bubble size. For the smallest bubbles, under 4.6 mm in diameter, no film drop was detected. More precisely, the amount of film drops was so low that it could not be detected, the difference in weight between the samples with- and without bubbling being below the balance sensitivity (10-5 g). The mass of film drops for 1 m3 of injected gas also increases with the bubble size as long as this bubble size is lower than 11 mm. Above this size, we can notice a plateau or a slight decrease in the amount of film drops. It would be interesting to investigate further this phenomenon, as well as to relate the amount of film drops to the mass of the bubble cap prior to bursting.