Abstract  :

Process modelling and optimization of removal of biophile metals from Nigerian bitumen using a phosphonium-based ionic liquid was investigated. Bitumen samples were extracted from oil sands in a Soxhlet extractor using toluene. Bitumen demetallization was achieved with biphasic solvent extraction using trihexyltetradecylphosphonium chloride. Concentrations of Cu, Mn, Zn, Co, Fe, and Cr were measured by Atomic absorption spectrophotometer (Model, Alpha Star 4). Optimization of the demetallization process was done using Response surface methodology (RSM). Predicted to experimental values fitness was validated through diagnostic plots using Design Expert Stat-Ease software. Analysis of variance of the model (p<0.005) indicated significance of the model. Predicted regression error metric (R²) and Adjusted R² values were in agreement for the metals, indicating the fitness of the model to the experimental data set and integrity of the model. Low adequate precision predicted value (<10%) for the metals implied normal variation in the data set. Diagnostic plots revealed a close fit along a straight line, showing the accuracy of predicted data. Results of interactive effects of process parameters analysed with the aid of 2D contour and 3D response surface plots showed increased input variables and led to decreased metal concentrations. Optimization of the demetallization process indicated optimum conditions (temperature of 60°C, 60-minute reaction time and 3% w/v ionic liquid concentration) at which the process will yield the best result. Conclusively, RSM based on central composite design has proved to be accurate and reliable in modelling and optimizing the demetallization of bitumen using a phosphonium-based ionic liquid.