Optimization of solar photocatalytic biodegradability of seawater using statistical modelling

The performance of zinc oxide (ZnO) as a photocatalyst was evaluated for the treatment of pollutants present in seawater. Batch experimental studies were carried out by varying the dosage of photocatalyst (1–4 ?g/L). The effect of reaction time, pH and the dosage of photocatalyst was evaluated with the percentage removal efficiencies of chemical oxygen demand (COD), biological oxygen demand (BOD), total organic carbon (TOC) and the biodegradability (BOD/COD) of the seawater. Response surface methodology-central composite design (RSM-CCD) and artificial neural network-Levenberg Marquardt (ANN-LM) statistical models were employed to optimize the photocatalytic biodegradability (BOD/COD). A quadratic polynomial statistical model was obtained to predict the percentage removal efficiencies of COD, TOC, BOD and biodegradability. For the experimental runs, the maximum percentage removal efficiencies for COD, TOC, BOD was found to be 62.3, 40.1, and 18.8%, respectively. Whereas, the maximum biodegradability was 0.036. As per RSM-CCD and ANN-LM statistical model method the maximum percentage removal efficiencies were found to be COD ?= ?58.14, 60.39%, TOC ?= ?33.74, 40.09%, BOD ?= ?18.47, 18.7% and Biodegradability ?= ?0.0315, 0.0360, respectively. The predicted values from statistical models were well correlated with experimental values. ANN modelling predicted better values for the responses with an average of R² ?= ?0.99697 than RSM modelling with average R² ?= ?0.8948.