In this work, a waste material (coal fly ash) containing a large amount of muscovite was used to synthesize a mesoporous Fe-ZSM-5 nano-zeolite by hydrothermal approach. Batch scale studies under optimal circumstances were used to investigate the adsorptive behavior of the Fe-ZSM-5 nano-zeolite in removing benzene, toluene, and m-xylene from aqueous solutions. The maximum adsorption capacity of Fe-ZSM-5 towards the toxic elements attained from the Langmuir isotherm model (non-linear type; R2 0.97) was qmax-B=10.55 mg/g, qmax-T=10.09 mg/g, and qmax-m-X=10.02 mg/g for benzene, toluene, and m-xylene, respectively. Kinetic and isotherm assessments demonstrated that pseudo-second-order and Freundlich models described the equilibrium data well. Evaluation of thermodynamic parameters revealed that the aromatic hydrocarbons uptake onto Fe-ZSM-5 nano-zeolite was exothermic, feasible, and spontaneous at lower temperatures. Moreover, the conventional and meta-heuristic neuro-fuzzy systems were designed to predict the behavior of Fe-ZSM-5 in the adsorption of toxic hydrocarbon residues in aqueous solutions. Statistical evaluations of the adsorption data disclosed the superiority of classical ANFIS-FCM models (0.984<R2<0.996) and ANFIS-GA meta-heuristic models (0.965<R2<0.975) with a minor error compared to other hybrid models. Based on the empirical and predicted results, the Fe-ZSM-5 nano-zeolite can be considered a potential adsorbent for removing hazardous hydrocarbons from aqueous media.
