).Bentonite, which is typically clay, is widely used for liner material in the barrier system. It has local availability and a low hydraulic conductivity. However, leakage can result from shrinkage cracking if only bentonite is used [11]. For this reason, a suitable sand-bentonite mixture to determine the minimum percentage of bentonite necessary to fulfil the given requirements is the main task [12]. Previous studies showed that quantities higher than 15% of bentonite as an amendment in a mixture do not lead to a significant decrease in hydraulic conductivity, while strength properties and mechanical behaviour of the mixture may be adversely affected by the clay [13, 14].The artificial neural network (ANN) is a system of data processing based on the structure of a biological neural system.
The prediction with ANN is made by learning of the experimentally generated data or using validated models [15]. Because of their reliable, robust, and salient characteristics in capturing the nonlinear relationships existing between variables (multiinput/output) in complex systems, numerous applications of ANN have been successfully conducted to solve environmental problems [16�C18]. In the literature, there are few studies relating to operation problems for landfilling processes based on ANNs. In the present work, heavy metal removal during landfilling of industrial waste is investigated. The effects of various liner materials, such as bentonite, natural zeolite, expanded vermiculite, and pumice on the removal of Cu(II) and Zn(II) are examined.
On the basis of batch adsorption experiments, a three-layer ANN model to predict heavy metal removal efficiency of composite used as a liner material is applied in this work. Removal of heavy metal from landfilling process is optimized to determine the optimal network structure. Finally, outputs obtained from the models are compared with the experimental data, and advantages and the further developments are also discussed.2. Material and Methods2.1. MaterialsThe three natural materials and the commercially available bentonite were investigated as a liner material in this study. Among them, natural zeolitee was obtained from the Rota Mining Industry (G?rdes, Manisa, Turkey); expandable vermiculite was obtained from the Fitar Agricultural Industry (Antalya, Turkey); pumice was obtained from the Soylu Mining Industry (Nev?ehir, Turkey); illite was obtained from Sud Chemie Mining Industry and Trade Co.
Ltd. (Ordu, Turkey); kaolinite was obtained from the Kale Mining Industry and Trade Co. Ltd. (?anakkale, Turkey); and bentonite was obtained from the Bensan Activated Bentonite Brefeldin_A Company (Enez, Edirne, Turkey). The chemical composition of the materials is presented in Table 1. Samples were crushed and then milled resulting in small particles with a size of about 0.5mm.