Optimization and genetic programming modeling of humic acid adsorption onto prepared activated carbon and modified by multi-wall carbon nanotubes

Milk vetch species (MV) as activated carbon (AC) precursors were prepared and modified by multi-wall carbon nanotubes (CNTs) in order to remove humic acid (HA) removal. Scanning electron microscopy (SEM), diffuse reflectance infrared Fourier transform spectroscopy (FTIR) and Brunauer Emmett Teller (BET) techniques were applied to characterized MVAC/CNTs. The HA adsorption presented the highest value (22.57 mg/g) at pH 3, while the equilibrium state was reached after 30 min. By increasing the adsorbent's dosage (0.4-0.5 g/L), the removal efficiency was also improved from 90.3 to 97.6. The kinetics of adsorption were fitted to the pseudo-second-order model (R-2 = 0.9991). The Langmuir isotherm model was well fitted to the equilibrium data revealing monolayer adsorption model of HA onto the MVAC/CNTs with the maximum capacity of 73.29 mg/g (R-2 = 0.9967). The obtained values of thermodynamic parameters indicated that the HA adsorption was spontaneous, endothermic and physical in nature. Also, genetic programming was used to obtain an appropriate equation revealing the relationship between the variables. The results showed that the introduced equation can successfully predict the HA removal. All experimental results suggested that MVAC/CNTs are promising adsorbents for the contaminated (with HA) water. (C) 2020 Published by Elsevier Ltd