Enhanced detection of glucose with carbon quantum dot-modified copper oxide: Computational insight and machine learning modeling of electrochemical sensing

Abstract

Poor conductivity and surface passivation pose critical challenges in metal oxide structures during their application for non-enzymatic oxidation. To address this, we systematically employed in-situ deposition of carbon-quantum dots (C-dots) over copper oxide (CuO), enhancing its electrocatalytic properties for direct non-enzymatic glucose oxidation in alkaline media. The process involved the systematic deposition of varying wt.% of C-dots onto the CuO nanostructure. The electrode’s sensing capability was assessed through CV, DPV, and amperometric measurements, evaluating its suitability in high (0.1 to 0.85 mM) and low glucose concentration levels (15 to 225 nM) with a representative LOD of 1.4 nM (17142.86 µA mM−1 cm−2). Additionally, the CuO-Cdot-16.6 protective coating allowed for long-term working capability, with chronoamperometric measurement confirming a 99 % current retention ability compared to pristine CuO’s 39 % retention during 3500 s of continuous measurement. DFT calculations further confirmed the efficacy of CuO substrate as a scaffold for glucose adsorption. The stable CuO-glucose complex formed due to energetically favorable conditions further strengthens its potential as a sensor. Successful recoveries of spiked glucose serum samples validated the sensor’s practical usage in complex matrices. Moreover, Machine learning was also adopted to validate the accuracy of glucose detection, where artificial neural network (ANN) model emerged as a suitable model to interpret the DPV derived data relationships, adding in sensor working capability and promising its future application in precision/intelligent healthcare.