Adsorption of phenol on granular activated carbon from nutrient medium: equilibrium and kinetic study

This paper presents the adsorption of phenol on granular activated carbon [GAC] from nutrient medium suitable for microorganisms' growth and also the subsequent biodegradation. Two parameter Langmuir and Freundlich adsorption isotherm models were studied using large range of phenol concentration [50- 1000 mg/L]. in low range of phenol concentration [50-300 mg/L], correlation coefficient, normalized deviation "g% and separation factor were 0.9989, 2.18% and 0.38-0.78 respectively, while for higher concentration range [400-1000 mg/L], the corresponding values were 0.9719, 1.9% and 0.45-0.67. Freundlich isotherm gave correlation coefficient of 0.9984, l/n. value of 0.7269 and normalized deviation of 4.55%. Comparison based on R[2], adjusted R[2], normalized deviation and root mean square deviation [RMSD] showed that the Redke-Prausnitz isotherm model gives better prediction compared to other models. Adsorption of phenol follows pseudo second order kinetics with correlation coefficient closer to one. Biodegradation study using immobilized cells of Nocardia hydrocarbonoxydans on GAC showed that, biodegradation begins well before GAC reaches the saturation period

Prediction of water quality indices by regression analysis and artificial neural networks

The quality of wastewater generated in any process industry is generally indicated by performance indices namely BOD, COD and TOC, expressed in mg/L. The use of TOC as an analytica parameter has become more common in recent years especially for the treatment of industrial wastewater. In this study, several empirical relationships were established between BOD and COD with TOC using regression analysis, so that TOC can be used to estimate the accompanying BOD or COD. A new, the use of Artificial Neural Networks has been explored in this study to predict the concentrations of BOD and COD, well in advance using some easily measurable water quality indices. The total data points obtained from a refinery wastewater [143] were divided into a training set consisting of 103 data points, while the remaining 40 were used as the test data. A total of 12 different models [A1-A12] were tested using different combinations of network architecture. These models were evaluated using the% Average Relative Error values of the test set. It was observed that three models gave accurate and reliable results, indicating the versatility of the developed models