Degassing Kinetics of Molten Metals in Vacuum

A brief review of the theoretical and experimental work on the kinetics of the vacuum degassing of liquid metals is given. This review is limited to the case usually encountered where the transport in the melt alone is rated determining, i.e., where desorption processes at the melt surface and the transport in the gas phase may be neglected. Different models describing the mass transfer from the interior of the melt to the interface are also discussed. The course of degassing of a melt pool without bubble formation, based on the mathematical treatments of Kraus (see Refs. 1 and 2) and Machlin (see Ref. 3) is described. In the model of Kraus—which may be applied, for instance, to ladle degassing—convection currents in the interior of the melt develop by density differences caused by the heat loss of the freely radiating melt surface. The calculations of Machlin refer to inductively stirred melts. Formulas for the mass transfer coefficients for these two cases are given. If layers of surface active elements or slags accumulate at a melt surface, the rate of degassing may be largely reduced. There is also an examination of how far the model of Kraus is applicable to degassing in electron-beam and vacuum-arc melting with consumable electrodes. Most technical degassing processes rely on the formation of bubbles, for only with them are the high rates required for economical degassing obtained. Also shown are the conditions under which bubble nuclei may exist and growth of the bubbles in a melt pool or in stream degassing is possible. Reference is made to the calculations of the gas pick-up of ascending CO bubbles in a steel melt made by Kraus (see Refs. 1 and 2), which also permit the calculation of the mass transfer coefficient if, in this case, a surplus of nuclei in the melt exists. If the gas content is too low for spontaneous bubble nucleation, the use of a neutral scavenging gas is advantageous. Computations of its purging effect have been made by Bradshaw and Richardson (see Ref. 4) as well as by Lange et al. (see Ref. 5). Stream degassing is not accessible to an exact mathematical treatment. However, an estimation of the degassing rates of droplets and experimental results show that the removal of hydrogen probably takes place mainly by diffusion from the external surfaces of metal droplets and bubbles, whereas nitrogen and oxygen removal depends more on the bubble formation in the metal stream and melt pool.