Artificial neural network (ANN) based prediction and modeling of Fe(II) adsorption   from contaminated groundwater using Deccan Hemp stem-derived activated carbon

Iron is a vital element for humans to survive, yet consuming too much of it can be detrimental to one's health. Iron contamination in groundwater is mainly in the form of ferrous iron Fe(II). Adsorption is highly suggested as a commonly utilized, cost-effective, and convenient method for Fe(II) removal. This study uses activated carbon derived from the stem of deccan hemp (ACDH) as adsorbent for removing Fe(II) from the groundwater. Fourier transform infrared, scanning electron microscopy, energy dispersive X-ray, and Brunauer-Emmett-Teller analysis were used to characterize the adsorbent. The Langmuir, Freundlich, Redlich-Peterson and Temkin isotherm models were used to describe the adsorption process and are best fitted to the Langmuir isotherm, with a maximum adsorption capacity of 10.989 mg/g Fe(II) at a 4.0 g/L ACDH dose. The adsorbent ACDH exhibits second- order adsorption kinetics for the adsorption of Fe(II). The thermodynamic study verified that the adsorption process is both endothermic and spontaneous, leading to an increasing degree of adsorption. The adsorption of Fe(II) was simulated using the Artificial Neural Network (ANN) technique and accurately predicted the batch process through the use of the Levenberg-Marquardt algorithm. The comparison between the experimental data and model outputs showed a strong correlation, indicating that the initial information was in close accord with the data predicted by ANN. The high Fe(II) adsorption capacity makes ACDH an efficient and economical adsorbent material for removing Fe(II) from groundwater.