A novel MOF-based electrochemical sensor for simultaneous quantification of nitrophenols in binary mixtures using artificial neural networks: challenges and opportunities

Abstract

This study introduces an artificial neural network (ANN)-assisted electrochemical strategy for the simultaneous quantification of 2-nitrophenol (2-NP) and 4-nitrophenol (4-NP), two environmentally hazardous pollutants with highly overlapping voltammetric responses. A glassy carbon electrode (GCE) was modified with a multilayer nanocomposite (Ni-MOF-74/Fe₃O₄/SiO₂/NH₂/β-CD) to enhance mechanical stability and facilitate electron transfer during the oxidation of 2-NP and 4-NP. For the first time, it was demonstrated that the optimal electrocatalytic parameters are concentration-dependent. Therefore, central composite design (CCD) was employed to determine common optimal oxidation conditions across the entire calibration range of both analytes. Despite severe peak overlap and matrix-induced non-linearities, ANN modeling successfully resolved the electrochemical data and provided accurate predictions, yielding high calibration accuracy (R² = 0.9302 for 2-NP and 0.9604 for 4-NP) with good reproducibility and broad dynamic range. Considering that the maximum allowable concentration of nitrophenols in environmental samples is 20 ppm, the proposed method offers sufficient sensitivity and a suitable linear range to allow reliable detection and quantification in real environmental matrices. Thus, the integration of ANN with a MOF-based electrochemical sensor provides a novel and robust approach for overcoming signal overlap and non-linear behavior in complex electrochemical systems, offering a promising analytical tool for environmental monitoring of nitrophenols.