Machine learning powered CN-coordinated cobalt nanoparticles embedded cellulosic nanofibers to assess meat quality via clenbuterol monitoring.

ABSTRACT

The World Anti-Doping Agency (WADA) has prohibited the use of clenbuterol (CLN) because it induces anabolic muscle growth while potentially causing adverse effects such as palpitations, anxiety, and muscle tremors. Thus, it is vital to assess meat quality because, athletes might have positive test for CLN even after consuming very low quantity of CLN contaminated meat. Numerous materials applied for CLN monitoring faced potential challenges like sluggish ion transport, non-uniform ion/molecule movement, and inadequate electrode surface binding. To overcome these shortcomings, herein we engineered bimetallic zeolitic imidazole framework (BM-ZIF) derived N-doped porous carbon embedded Co nanoparticles (CN-CoNPs), dispersed on conductive cellulose acetate-polyaniline (CP) electrospun nanofibers for sensitive electrochemical monitoring of CLN. Interestingly, the smartly designed CN-CoNPs wrapped CP (CN-CoNPs-CP) electrospun nanofibers offers rapid diffusion of CLN molecules to the sensing interface through amine and imine groups of CP, thus minimizing the inhomogeneous ion transportation and inadequate electrode surface binding. Additionally, to synchronize experiments, machine learning (ML) algorithms were applied to optimize, predict, and validate voltametric current responses. The ML-trained sensor demonstrated high selectivity, even amidst interfering substances, with notable sensitivity (4.7527 µA/µM/cm2), a broad linear range (0.002-8 µM), and a low limit of detection (1.14 nM). Furthermore, the electrode exhibited robust stability, retaining 98.07% of its initial current over a 12-h period. This ML-powered sensing approach was successfully employed to evaluate meat quality in terms of CLN level. To the best of our knowledge, this is the first study of using ML powered system for electrochemical sensing of CLN.