The mass spectrometric Knudsen effusion method was applied for the study of vaporization processes of mould powders. All measurements were performed in vacuum conditions in the temperature range between 100 and 1550°C, the standard rate of temperature elevation being 100°C/hour. Mould powders were selected from those used in continuous casting at Saarstahl AG (Völklingen, Germany) to represent the variety of types and compositions. In addition to SiO2, CaO, Al2O3 and MgO they contained 0.4-14.3 wt% Na2O, 0.3-1.0 wt% K2O, 0.1-3.6 wt% Fe2O3, 2.6-7.6 wt% fluorine and 3.4-9.1 wt% carbon.
With such complex mixtures as mould powders only a half-quantitative analysis of vaporization curves could be carried out. Mass spectra were interpreted with the help of isotopic abundance ratios and equilibrium constants for the gas phase components. Hence, they were not deciphered in full and the main results are given as time/temperature dependences of ion currents.
The results confirm that fluorides dominate in the gas phase. Vaporization (sublimation) occurs mainly in the form of NaF, Na2F2, KF, SiF4 and CaF2, while gaseous AlF3, MgF2, AlOF are generated in smaller amounts and depend on the slag composition. Yet, in the given experimental conditions, the gas phase was found to be richer in molecular species than found previously. Analysis of the NaF/Na2F2 ratio proved the existence of Na(g). Some components of the mass spectra indicate the presence of significant amounts of SiF2 and SiO in the vapour. The relative intensities of Mg+ and MgF+ ion currents suggest that Mg in vapour is present in the form of Mg(g) and its fluorides. The extra gaseous components observed in this study could be explained by the presence of carbon in the samples. This was confirmed by comparison with the vapour phase composition of the decarburised sample of a mould powder. CO could be registered in the vapour up to temperatures 1400-1500°C. In the temperature interval between 100 and 600°C, the gas phase was formed by H2O and CO2. In addition, numerous ion currents of small intensities were recorded between 100 and 450°C but their origin remained unclear. They were attributed to impurities, though some of them may have included components of the assay, such as K2O or Na2O. HF(g) could not be detected in the vapour.
The thesis is partly based on the work done within the European Coal and Steel Community Project concerning emissions of hazardous substances (the data related to pre-melted slags, fluorine free mixtures, Li2O substituted compositions).