During the magnesium-lime co-injection process for hot metal desulphurisation, graphite can precipitate as a result of carbon oversaturation. The formed graphite is known to form a layer between the slag and the hot metal. This potentially blocks the sulphides, that are formed during the desulphurisation process, to reach the slag phase thus hampering the desulphurisation efficiency. In this research it was aimed to obtain experimental evidence for the postulated hampering effect of graphite on the hot metal desulphurisation efficiency at an industrial process. In 2018 at Tata Steel in IJmuiden, the Netherlands, the carbon concentration in the hot metal was measured after the reagent injection, instead of being calculated assuming carbon saturation of the hot metal. This provides the opportunity to predict graphite formation during the process. Although a correlation is found between graphite formation and specific magnesium consumption, which is a measure for desulphurisation efficiency, the effect could not be directly proven, as too many other parameters could have influenced the desulphurisation process. Data analysis does show that the observed correlation could not be attributed to temperature and initial sulphur concentration of the hot metal, which are the two most important factors for the desulphurisation efficiency. The observed correlation between graphite formation and desulphurisation efficiency is significant at low initial sulphur concentrations (<225 ppm) and insignificant at high initial sulphur concentrations (>225 ppm). Key Words Hot metal desulphurisation, magnesium consumption, graphite precipitation, kish@en

Carbon may precipitate during the hot metal desulfurization (HMD) process as a result of carbon oversaturation because of temperature decrease. The precipitated carbon flakes form a layer between hot metal and slag. It is postulated that this carbon layer hampers desulfurization with magnesium by preventing MgS particles from reaching the slag phase. At Tata Steel in IJmuiden, the Netherlands, carbon in hot metal is measured in 657 heats after reagent injection. With this data, it can be determined whether the hampering effect of precipitated carbon on MgS flotation has a significant effect on the performance of the industrial HMD process. Plant data show a correlation between the precipitated carbon and the specific magnesium consumption for hot metal with a low initial sulfur concentration (below 225 ppm). This correlation cannot be found for hot metal with a higher initial sulfur concentration (above 275 ppm). Furthermore, a sulfur mass balance is made over the converter process, that shows no effect of carbon precipitation during HMD on resulfurization in the converter. The limited experimental accuracy of the plant data prevents a quantitative description of the hampering effect. The measurements do suggest that the effect is small.

@en

Ladle metallurgical processes constitute a portion of the total production chain of steel from iron ore. With these batch processes, the hot metal or steel transfer ladle is being used as a reactor vessel and a reagent is often injected in order to bring the composition of the hot metal or steel to the specification of the final product. To control and further improve these processes, often use is made of models that predict the course of the processes. Models derived from first principles of mass and energy transport have the advantage over empirical descriptions that predictions outside the established window of operation can be made. The establishment of such a model, however, requires deeper knowledge of the underlying thermo-chemical processes. The purpose of this work is to provide a uniform method for the development of a model of injection processes in the ladle metallurgy. This will give direction to the development of new models, and will clarify blind spots in the existing knowledge for which further research is required. It is chosen to study two ladle treatments and to develop a reactor model of these, namely the desulphurisation of hot metal by the injection of magnesium and lime, and the modification of inclusions in aluminium killed steel by the injection of calcium. The hot metal desulphurisation has been studied by microscopic analysis of hot metal samples taken during different heats. The top layer of the bath, where the hot metal is in contact with the slag layer, has also been studied. From these analyses, it follows that during the injection of magnesium, magnesium sulphide particles are formed which continue to be present for some time in the hot metal, grow and later on rise out of the bath. This corresponds to the mechanism as has been proposed by G. Irons and R. Guthrie, based on their experiments on a laboratory scale. Due to the differences in scale, however, the accumulation of magnesium sulphide particles in the hot metal plays a significant role in the explanation of sulphur levels observed in the industrial desulphurisation process. The measurements furthermore show that the rise of the particles to the slag layer is partially obstructed by graphite and Ti (C, N) particles that accumulate in the colder surface layer of the hot metal. This creates a layer with a high concentration of MgS-particles that remain unnoticed but can lead to undesirable sulphur pick up in the converter process. It also appears that the MgS particles that rise to the slag layer react with co-injected lime to form MgO and CaS. In order to prevent sulphur reversal by oxidation of MgS in the ambient air, lime should always be injected in a slight excess. During the injection process, iron droplets are thrown up from the spout area. These droplets slowly sink through the slag layer and do not make it back to the hot metal before the end of the injection process. Based on this it can be explained how the hot metal loss depends on the amount of injected magnesium, the hot metal temperature, and the hot metal titanium content. Because these droplets are entrained with the sulphide containing slag during deslagging, this presents a major cost in the form of loss of hot metal. Based on these findings, a reactor model of the hot metal desulphurisation has been developed The mixing in the hot metal bath has been described as an ideally mixed tank reactor, wherein the residence time of injected magnesium and lime has been described by a generic model of the bubble plume. An important aspect of the model is that the total sulphur content is formed by the sulphur that is dissolved in the hot metal and the sulphur which is bound in the MgS particles suspended in the hot metal. The specific surface area of the MgS particles is derived from the microscopic observations and the flotation of the MgS particles is described by a first order rate equation. The predicted development of the concentrations of magnesium and sulphur during the duration of the injection is in good agreement with the measurements which have been made with two treatments. The calcium treatment of aluminium killed steel has been extensively studied by W. Tiekink. Based on the measurements carried out by him on laboratory and industrial scale, the steel bath in the ladle is divided into two reaction zones: a zone plume in which the steel is saturated with calcium and a bulk zone in which initially no calcium is present. In the plume zone CaO and CaS are deposited on the Al2O3 particles. Induced by the bubble plume resulting from the calcium injection, a circulating current flows between these zones. Each zone is modelled by a continuously stirred, ideally mixed tank reactor (CISTR). Carried along with this circulating current, the particles are alternately exposed to high and low levels of calcium activity which ultimately results in a variation in size and composition of the particles. This is modelled by taking a population balance of the particles for each zone containing terms for convection and growth by deposition of CaO and CaS. The results of the model are consistent with the observations when it is assumed that the solubility of calcium in steel amounts to 1 ppm. There is a good prediction of particle sizes and composition, but the CaS content of the particles is slightly overestimated by the model. This is explained by the pick up of oxygen at the surface of the steel which is not yet taken into account in the model. The final conclusion of this work is that the chemical conversion of the hot metal and steel during ladle metallurgical injection processes can be well modelled with a detailed description of nucleation, growth and rise of the product particles in conjunction with a relatively simple macroscopic description of the flow field in the ladle. This can best be achieved by providing a population balance of the product particles in which, to the extent appropriate, terms for nucleation, growth, convection and flotation are included. The flow field in the ladle then is modelled using three CISTR's coupled by circulating currents of liquid metal. These CISTR's represent the three zones that can be distinguished in a ladle stirred by a bubble plume; the plume zone, the recirculation zone and the stagnant zone above the bottom of the ladle. @en