The main goal of this research is to get a better understanding of the manganese refining path in the steelmaking converter process. This study focuses primarly on manganese refining due to manganese mass transfer, between metal containing droplets and oxidized slag in the metal-slag emulsion zone, in the top part of the steelmaking converter. Manganese and iron content in the slag heavily influences the slag formation path. Controlling the slag viscosity is essential for controlling the efficiency of the steelmaking process and to preserve the quality of the final steel product. Manganese refining in the steelmaking process occurs in three phases, the primary oxidation phase, the reversion phase and the secondary oxidation phase. This manganese reversion phase is hard to predict and influences the final amount of manganese in the end steel product. In order to raise understanding of the reversion behaviour of manganese in the converter, unique data gathered from in situ measurements done in the EU funded IMPHOS project is utilised. The following research questions were formulated and answered in this study: 1. Are thermodynamic models for manganese distribution calculation applicable for the prediction of the three different behavioral phases of manganese in IMPHOS? 2. What is the influence of the slag-metal emulsion droplets size on the manganese refining behaviour? 3. What is the influence of metal-slag data input obtained from 2 different heights in IMPHOS, on the final resulting manganese behaviour in the kinetic model made in python? 4. Were there any chemical deviations in different zones of the hot metal bath during IMPHOS project sampling? IMPHOS heat S1836, S1841, S1844, S1845 were selected and analyzed on their theoretical thermodynamic equilibrium behaviour for manganese. This analysis was done with Suito No. 3 and Morales and Fruehan equilibrium partition model. Furthermore, the heats slag composition data, hot metal data and converter process variables were used as input in a kinetic model. The model is simulating the micro and macro kinetic behaviour of manganese in the converter emulsion zone with a fixed time step of 1s. The mass transfer in the model is based on Fick's first law of diffusion and uses penetration theory by Higbie, to describe mass transfer along metal droplet boundaries. Parameters such as the manganese mass transfer and the predicted manganese content in the hot metal bath were evaluated to validate the model with the measured concentration reported in the IMPHOS study. Bulk chemical analysis is done on physical samples from heat S1836 cup height 2. This is done to check the inhomogeneity in the hot metal bath and the possibility of different reaction zones in the converter process. The thermodynamic evaluation performed on IMPHOS data showed the difficulty of thermodynamic models derived from laboratory tests, to describe the manganese equilibrium behaviour within the converter. Suito no 3 model gave a better prediction of the manganese distribution than the Morales and Fruehan model. The results from the kinetic model showed a slower primary oxidation and reversion behaviour of manganese compared to the original data. This is mainly due to the manganese equilibrium behaviour of the thermodynamic model and the negligence of the manganese refining in the jet impact zone and metal-slag interface. The model gave a better prediction when a small size droplet range was considered, which implies that the smaller size droplets less than 100 micron are significantly contributing to the manganese transfer from metal to slag and vice versa. The present study reveals that droplet size play an important role and thus need further investigation to accurately describe the size fraction for the accuracy of the kinetic model. The difference between metal-slag data input from the different IMPHOS cup height 4 and 5 on the outcome of the thermodynamic evaluation and kinetic model results, proved to be small enough to be considered negligible. The results from the sample analysis showed deviations for carbon and phosphorus in the new examined cup height 2. These measurements were taken at a high position in the hot metal bath and were different from the measured concentration in the bottom of the hot metal bath. This can indicate deviations in chemical content in the hot metal bath during the original IMPHOS research. More heat data of height 2 , close to the bath surface, is required to confirm this finding.