The objective of this project was to treat petroleum refinery wastewater using a combination of anaerobic and aerobic processes, namely an Up-flow Anaerobic Sludge Blanket (UASB) reactor paired with a Rotation Algae Biofilm Reactor (RABR), respectively, to produce a treated effluent. The treatment method developed needed to produce a cost-effective and efficient way to decrease nitrogen, phosphorous, total suspended solids (TSS), and COD concentrations to below State of Utah limitations. It was demonstrated that RABR treatment was capable of reducing effluent concentrations of nitrogen, phosphorus, and TSS to State of Utah limitations. RABR treatment did not significantly reduce COD from the wastewater. The COD reduction requirement, however, was met through anaerobic digestion of the wastewater. Therefore, our system proved effectual at the treatment of the wastewater and met all design criteria.

M.Tech. (Department of Chemical Engineering, Faculty of Engineering and Technology), Vaal University of Technology / Petroleum wastewater is highly contaminated with toxic organic pollutants that are harmful to the environment. The heterogeneous photocatalytic oxidation (HPO) process has shown the ability to remove these pollutants through the application of a fluidized bed reactor (FBR). The purpose of the study was to apply response surface modelling (RSM) and computational fluid dynamics (CFD) to optimize the operating conditions for the photodegradation process in an FBR. This was done by investigating the hydrodynamics, photodegradation efficiency and reaction kinetics; that gave a holistic view on the performance of the FBR. The hydrodynamic study focused on modelling the axial liquid velocity, gas hold-up and turbulence quantities due to their substantial impact on the design and performance of the FBR. This was done by implementing the Eulerian-Eulerian approach which solves the continuity and momentum equations for each phase. In addition, the standard k-ε turbulence model was used to capture the turbulent characteristics in the liquid phase. A numerical optimization technique (desirability) was used to determine the optimal simulation setting methods; that were found to be a fine grid size (500 000 cells), 2nd Order Upwind discretization scheme and a small time step size (0.001) and gave the best desirability (0.985). The axial liquid velocity was maximal towards the centre of the reactor and decreased towards the wall. The same trend was seen with the local gas hold-up, where it was high towards the centre and low near the wall region. This was an indication that the bubbles tended to gather towards the central region as they move up. Furthermore, the bubbles had a spherical–like shape due to the low superficial gas velocity and operating within the homogeneous regime. The turbulent kinetic energy increased at distances away from the distributor region, due to the bubbles accelerating, and it balanced well with the energy introduced by the bubbles. Central composite design (CCD), which is a type of response surface modelling technique, was used to investigate and optimize the photodegradation operating parameters. The maximal degradation efficiency in the current study was found to be 65.9%, which was relatively low when compared to literature (80.84%). This was attributed to the increase in the catalyst particle size from nanometer to micrometer. Furthermore, the second-order empirical model that was developed, using the analysis of variance (ANOVA), presented a sufficient correlation to the photodegradation experimental data. The optimal photodegradation operating conditions were found to be: superficial gas velocity of 17.32 mm/s, composite catalyst loading of 1.0 g/L, initial pH level of 3.5 and reaction time being 210 min. Using the Langmuir-Hinshelwood model, it was found that the photocatalytic degradation of petroleum wastewater follows pseudo first-order reaction kinetics. Since the photocatalytic degradation mechanism of phenol follows three stages whereby the second stage is the photocatalytic degradation on the surface of the catalyst to form by-products. This is the rate dominant stage and follows the pseudo firstorder reaction kinetics.