Search

Engineering the inclusion content during secondary steelmaking is essent ial in controlling the mechanical properties of the cast steel product. The type and morphology of the inclusions depend on the local thermodyna mic equilibrium in the ladle, hence on the temporal and spatial evolutio n of the concentration in the melt. When adding cold alloy particles to liquid steel, a layer of bath material freezes around them. Only after t he sample is heated and this layer has molten, the alloy will be release d in the melt. An experimental setup was built to study this behavior at laboratory scale, by continuously and indirectly measuring the thicknes s of dissolving cylindrical Ti and ferro-Ti (70 wt% Ti) samples. Microst ructural analysis of Ti samples quenched after different immersion times shows that a liquid reaction zone forms between the addition and shell. This reaction zone develops in three distinct stages: instantaneously u pon immersion a liquid phase with eutectic composition (31 wt% Fe) forms , followed by the formation of a second liquid phase with eutectic compo sition (86 wt% Fe) seconds thereafter, and the mixing of both liquid pha ses when the intermetallic phases in between them have molten. Whereas o nly a part of the Ti dissolves in this reaction zone, ferro-Ti 70 additi ons, which have a lower liquidus temperature, can melt entirely within t he shell. This results in a significant reduction of the total dissoluti on time of ferro-Ti 70 compared to Ti, even though the shell period is s lightly longer for the former. In the Ti experiments, the shell melting time averages 45±8 % of the total dissolution time and increa ses in importance for slower shell melting. The thickness measurements w ere used to quantitatively evaluate the time to melt the shell and the f raction of liquefied alloy within the shell for varying bath temperature s and sample sizes. This data was used to validate a conservative 1D sha rp interface model which describes the coupled heat and mass transport i n the addition-shell composite, and which considers explicitly all phase s present in the reaction zone. The model is coupled to thermodynamic an d kinetic databases and shows good qualitative and quantitative agreemen t with the experimental dissolution data, and illustrates how the reacti on zone development reduces the shell melting time, as well as the remai ning fraction of Ti and ferro-Ti 70 at the end of the shell period. More over, it is concluded that the reaction zone size is not directly influe nced by the flow behavior in the surrounding melt, but instead is almost solely governed by the duration of the shell period. In addition, corre lations between the shell melting time and the addition size and the mag nitude of the heat transfer coefficient at the shell-melt interface have been determined for typical steelmaking conditions, which can be used t o estimate the spatial and temporal evolution of the melt concentration. The dissolution experiments as well as the model have been genericallyconceived to allow the methodology to be applied to other alloying agent s orin other application domains, like soldering or thermal galvanizing. nrpages: 229 status: published