Adaptive Neuro-Fuzzy Inference System Modeling of Reactive Red 250 Dyestuff Sorption to Cucurbita moschata Rind in a Fixed-Bed Column

摘要

There are two main objectives of this study: the first was to examine the performance of Cucurbita moschata as sorptive media for the removal of Reactive Red 250 dye compound as a function of contact time, concentration, pH, particle size, flow rate, and bed depth; the second was to demonstrate that Adaptive Neuro-Fuzzy Inference System can be effectively used as a tool to model breakthrough profiles in fixed-bed column systems. The breakthrough curves of Reactive Red 250 were investigated in a fixed-bed column under different conditions. Besides, the mechanism of Reactive Red 250 uptake was analyzed based on the results obtained from Fourier transform infrared spectroscopy. At different pHs, the adsorbents' electrophoretic mobilities were determined to get information about the adsorption process. The adsorbed quantity of Reactive Red 250 dyestuff decreased with the initial pH, particle size, and flow rate augmentation. The total pore volume and Brunauer, Emmett, and Teller (N2) surface area of the particles were defined to be 0.0014 cm3/g and 1.06 m2/g, respectively. The highest observed and predicted bed capacities for Reactive Red 250 were 8.319–8.322 mg/g. The coefficient of determination and mean squared error for the optimal model were obtained to be 0.999 and 1.5 × 10−5. These findings indicate that the model can estimate the breakthrough profiles of Reactive Red 250 with high accuracy.