The Oxygen Steelmaking process is used to refine pig iron produced in the blast furnace to produce liquid steel for further refining in secondary steelmaking processes. The main advantages of the process are its autogenous nature, wherein the heat is generated through the refining reactions itself, and the refining is completed in a relatively short time (typically 15-25 mins). Achieving the desired end-point composition of refined steel is essential to avoid re-blows, which lead to delays in downstream processes and an increase in steel production costs. Improving process control through regular monitoring and a better understanding of the process is thus very critical. Multiple reaction interfaces are formed between various phases (slag, metal, gas), at extremely high temperatures and this makes the monitoring of the process through sampling and observation difficult and expensive. Consequently, mathematical modelling has been used as a tool to improve the understanding of the process and propose developments in operation. Numerous models have been developed in the past; however, these models do not address several open questions regarding the detailed reaction mechanisms and the contributions from different reaction zones inside the Basic Oxygen Furnace. The current work aimed to fill this gap. In this work, four prominent reaction zones, namely; impact, slag-metal bulk, cavity, and emulsion zones were identified. A more mechanistic approach involving process variables has been used to decrease the level of empiricism. With regards to the impact and the slag-metal bulk zones, the velocity of flow of metal (or surface-renewal) at the interfaces of these zones are calculated by taking into the momentum induced by the top-jets and bottom-stirring plumes. This study found that these zones contribute negligibly to overall refining in the oxygen steelmaking process. In the case of the emulsion zone, a very rigorous description of all aspects (external and internal decarburization, bloating behavior, and trajectory) pertaining to the life cycle of a single metal droplet in slag has been achieved. The emulsion zone is found to contribute 5 to 75 % of decarburization during various times of blow. The cavity zone model represents the first reported effort to predict the refining behavior of metal droplets that are exposed to oxygen jets within the lance cavities. The model incorporated the mass transfer, reaction equilibria, and kinetics of the reactions. It is predicted that this zone plays a critical role in the removal of silicon and FeO formation in the early part of the blow and removal of carbon throughout the blow. Several significant insights with regards to improvement in the operation of the oxygen steelmaking process are derived from each sub-models. The integration of these models will guide the steelmaker to improve their practices so that they can achieve better consistency in the end-point composition of refined steel and reduce re-blows. / Thesis / Doctor of Philosophy (PhD)