ABSTRACT

This study examined implementation of Artificial Neural Network (ANN) for the prediction of Cd (II) adsorption from aqueous solution by Incinerated Rice Husk Carbon (IRHC). Batch adsorption tests showed that extent of Cd (II) adsorption depended on initial concentration, contact time and pH. Equilibrium adsorption was achieved in 60 min, while maximum Cd (II) adsorption occurred at pH 5. The Levenberg-Marquardt algorithm (LM) training algorithm was found to be the best among 8 backpropagation (BP) algorithms tested; lowest Mean Square Error (MSE) of 20.99 and highest R2 was 0.96. Langmuir constants Q° and b were 40 and 0.04, and Freundlich constants K f and 1/n were 2.15 and 0.69, respectively. Adsorption capacity of IRHC was compared with other adsorbents and activated carbons reported in the literature. Being a low-cost carbon, IRHC has potential to be used for the adsorption of Cd (II) from aqueous solution and wastewater in developing countries.