Modeling of quaternary dyes adsorption onto ZnO–NR–AC artificial neural network: Analysis by derivative spectrophotometry

• Dyes adsorption onto ZnO–NR–AC was described.
• Removal efficiency was modeled by various models.
• Efficiency was optimized by artificial neural network (ANN) and experimental design.

The novel adsorbent i.e. ZnO–NR–AC was synthesized and used for the rapid removal of the quaternary dyes from the aqueous solution. The ANN model was used for the optimization and modeling of sonication time, amount of sorbent and dyes concentrations to study their simultaneous adsorption based on achievement of minimum mean squared error as criterion. The optimized parameters was found to be 4 min sonication time, 0.022 g of ZnO–NR–AC; MB, EY, CV and AO concentrations were 8.0, 9.7, 8.0 and 10.6 mg L−1possible to achieve the removal percentage of 99.89, 99.2, 99.68 and 99.45% for MB, EY, CV and AO, respectively. The analysis of variance (ANOVA) support the high suitability of achieved equation for the efficient prediction of understudy adsorption system behavior that proofed by the presence of good agreement among the predicted and experimental data. The Langmuir isotherm model with maximum adsorption capacities were 89.29, 93.46, 87.52 and 88.5 mg g−1 correspond to MB, CV, EY and AO, respectively.

Figure optionsDownload as PowerPoint slide