Predicting the Capability of Carboxylated Cellulose Nanowhiskers for the Remediation of Copper from Wastewater Effluent Using Statistical Approach

This study focused on remediation of spiked Cu(II) from wastewater effluent obtained from a real wastewater treatment plants (WWTPs) using oxidised cellulose nanowhisker (CNW) adsorbents. Response surface methodology (RSM) and artificial neural network (ANN) were used to develop an approach for the remediation of spiked Cu(II) from wastewater effluent. As remediation processes from wastewater are often complicated due to the variation in wastewater compositions, results obtained from the benchmark experiments are included as one of the independent variables for ANN modelling. This novel approach and the outcomes are allowed for the first time, since most studies do not consider matrix variability and its impact when evaluating the efficiency of an adsorbent. Moreover, to confirm the model suitability, additional 10 unseen experiments, which were not used in developing both models, were chosen to represent the system of conditions both inside and outside the system. This study found that the ANN model accounting for wastewater variability was superior to the RSM model and to the ANN model not including wastewater variability, in terms of the coefficient of determination (R2), the absolute average deviation (AAD) and root mean squared error (RMSE) when predicting the efficiency of Cu(II) removal from the wastewater matrix.