Fingerprint of a Phosphorus Producing Submerged Arc Furnace B ? The Promise of CFD Modelling

The process under investigation is the large-scale production of phosphorus. Gravity delivers the feed, consisting of pelletised apatite, coke and silica (in the form of gravel), to a submerged arc furnace through ten, evenly distributed feed chutes ensuring constant packed bed volume. Within this packed bed, solid-gas reactions are dominant in generating the gaseous product. These reactions, as well as the packed bed itself, were the main focus in the development of a fully three-dimensional CFD model in which fluid flow and energy distribution are represented as functions of chemical reactions, heating and melting of material and heat losses. The associated complex process interactions necessitated the creation of two process-specific, user-developed models that integrates accurate, thermodynamical data with computational fluid dynamics calculations, as well as reaction kinetics. The boundary conditions are provided by industrial measurements, and the key model parameters of reaction kinetics were obtained by laboratory experiments for different feed mixtures. Thirteen temperature probes installed inside the refractory lining of the furnace especially for this project allowed parameter fitting and validation to be guided within the constraints of real measurements. Simulation results reveal a thin, dome-shaped, reductive zone where conditions are optimum for maximum phosphorus recovery, as well as evidence of a thermal reserve zone similar to an ironmaking blast furnace located above this reduction zone. When compared to actual operating data, the CFD model can successfully predict the specific power consumption, the phosphorus in the slag and gaseous product flowrate under given operating conditions. This CFD model forms part of a larger project that aims to create a Dynamic-CFD hybrid control model.