An Electrochemical Detection of Malathion Pesticide Using Cu Electrode and Enhanced by Machine Learning

Abstract The present work demonstrates the development of an economical and user‐friendly “copper rods” sensor for detecting malathion. Differential pulse voltammetry (DPV) was performed to observe the inhibition ratio at various concentrations of malathion, which increases with an increase in malathion concentration. The parameters like pH and accumulation time were optimized at 4 pH and 18 min, respectively, corresponding to the maximum inhibition ratio (Δ I / I 0 ). The electrochemical sensor had a relative standard deviation (RSD) of up to 7.05 % ( n = 3), which indicated reproducible results. The regression line showed linearity over a range of 25–200 parts per billion (ppb), and the limit of quantification (LOQ) was as low as 25 ppb (75.67 nM). The developed sensor was sensitive and selective, with a limit of detection (LOD) as low as 1 ppb (3.03 nM). The selectivity of the sensor was also studied by adding Pb(NO 3 ) 2 , Zn(NO 3 ) 2 , and NiCl 2 to a solution of fixed malathion concentration, and minimal interference was observed. The sensor's functionality was validated using an unknown concentration of malathion with 96 % and 106 % recovery, respectively. The sensitivity of this proposed sensor was 0.0165 µA ppb −1 . Quantification of malathion was also facilitated using partial least squares (PLS) algorithms utilizing the sensory measurements of the malathion‐contaminated samples. PLS is a statistical machine learning algorithm that has been used here to develop a predictor for unknown malathion concentration using the DPV current signatures of the contaminated solution with a nominal error of 5.0 %.