Graphene Oxide as an Adsorbent in the Removal of Methylene Blue from Water: Optimization, Isotherms, Kinetics, and Mechanistic Insights

This study synthesized graphene oxide (GO) from graphite flakes and evaluated its performance in removing methylene blue (MB) from water using modeling. Characterization was done using FTIR, FESEM, HRTEM, XRD, Raman spectroscopy, and BET. The analysis showed an amorphous nature with micro- and mesopores and the presence of aromatic rings, carbonyl, epoxy, and hydroxyl groups. Box–Behnken design (BBD) and artificial neural networks (ANNs) were used to optimize process parameters, including concentration (10–30 mg/L), pH (3–6), time (10–30 min), and dosage (0.05–0.15 g). BBD yielded a high correlation (R2 = 0.9988), while ANN demonstrated superior predictive accuracy (R2 = 0.9998). Isotherm models of Langmuir, Temkin, Sips, Freundlich, and Dubinin–Radushkevich had high R2 values, and they were used to understand the mode. Sips best described the process (R2 = 0.9953), yielding a theoretical maximum adsorption capacity of 350.4 mg/g. The experimentally observed capacity at the highest tested concentration (400 mg/L) was approximately 160 mg/g, indicating that the Sips model provides a reasonable fit but extrapolates beyond the measured concentration range. The model thus suggests that both monolayer and multilayer adsorption occur simultaneously. The kinetic models of Elovich, pseudo-second-order, intraparticle diffusion, and pseudo-first-order provided evidence for a mechanism that occurs either by chemisorption or physisorption. Both PFO (R2 = 0.9978) and PSO (R2 = 0.9989) confirmed both physisorption and chemisorption mechanisms. Lastly, the predominant forces in the removal of MB were due to π-π interactions, and electrostatic attractions.