Enhanced Glucose Sensing Performance through Controlled Pulsed Laser Deposition of Au Clusters on Anodized Glassy Carbon Electrodes

Diabetes, a global health problem, necessitates precise blood glucose measurement for effective management. Despite their prevalence, enzymatic glucose sensors suffer from sensitivity to temperature and pH and limited lifetime of the enzyme, thus requiring replacement every two weeks (Teymourian et al., 2020). In contrast, non-enzymatic sensors offer enhanced durability, superior sensitivity, and ease of miniaturization, making them promising for the so-called "fourth generation" glucose sensors (Aun et al., 2021). Noble metals like Pt (Kim et al., 2012), Au (Shen et al., 2022), Pd (Elkholy et al., 2019), and Ir (Dong et al., 2018) have emerged as prospective candidates for glucose sensing due to their ability to oxidize glucose at a physiological pH of 7.4. Among these, Au has gained attention due to its remarkable electroactivity for glucose electro-oxidation (Vassilyev et al., 1985), despite its weak chemisorptive properties arising from filled d-orbitals (Hammer & Nørskov, 1995). Various methods have been employed to fabricate nanostructured gold electrodes, encompassing direct electrostatic assembly, covalent linking, polymer entrapment or co-mixing, sol-gel processes, electrochemical etching, and electrodeposition (Dahan et al., 2023). Nevertheless, many of these fabrication approaches entail the use of hazardous chemicals or involve cumbersome procedures, de facto limiting their practical application. As a result, scalable solutions to catalyst fabrication and deposition are needed to seamlessly integrate with the cleanroom process flows to build non-enzymatic glucose sensors. In this study, 7x7mm glassy carbon electrodes (GCs) served as working electrodes in a three-electrode system, with a reversible hydrogen electrode employed as a reference and a Pt mesh as a counter electrode. The GCs underwent meticulous polishing using alumina slurry of varying sizes (1μm, 0.3 μm, and 0.01 μm). Subsequently, they were anodized in 1M KOH at 1.8V for 2 hours in ambient air conditions. The electroactive area of the GCs was determined by measuring the capacitance with EIS before and after anodization, and by later dividing by the specific capacitance (13 μF/cm 2 ). A remarkable tenfold increase in surface area was observed after anodization. With a custom-built cluster deposition apparatus (Figure 1a), four monolayers of gold clusters were deposited on three bare GCs (BareGC) and three anodized ones (AnGC). Field Emission Scanning Electron Microscopy (FE-SEM) images captured the morphology of BareGC and AnGC, both with and without clusters (Figure 1b). Atomic force microscopy was also utilized to complement the morphology study of the electrodes' surface. Grazing Incidence X-ray Diffraction (XRD) at 0.5 degrees revealed a peak corresponding to the Au(111) plane in Au+BareGC (Figure 1c). Rutherford backscattering was employed to quantify the gold loading on the samples. To facilitate the generation of active Au-OH species on the nanoclusters, the electrodes underwent cycling between 0.43V and 1.53V in 0.1M PBS, devoid of chlorides. Subsequently, the sensing performance towards glucose was evaluated via chronoamperometric methods, maintaining the potential at 0.884V under stirring conditions (250 rpm). Gradual increases in glucose concentration were applied in steps of 0.25 mM and later in 1 mM increments up to 7 mM (Figure 1d). Both Au+BareGC and Au+AnGC electrodes exhibited an extremely linear response (R²=0.99) within the 0.25 mM to 7 mM linear range. After calibration, the electrodes underwent cycling in 0.5 M H 2 SO 4 between 0.015V and 1.715V ( vs RHE). The gold oxide reduction peak was integrated (Figure 1e), allowing estimation of the electroactive surface area (ESA) using the formula ESA=(Integrated charge)/(scan rate*0.386 mC/cm -2 ). A comparative analysis with similar nanostructured Au samples (Table 1) revealed that Au+AnGC displayed a four times higher sensitivity and substantially lower detection limits.The impact of chlorides was explored by repeating the chronoamperometric calibration in a PBS+0.1M KCl buffer. Chloride adsorption led to complete catalyst deactivation and eventually to gold loss from both samples. Notably, Au+BareGC displayed higher sensitivity to Cl etching, experiencing a 50% reduction in gold ESA compared to a 30% drop in Au+AnGC. This work has proved that the deposited nanoclusters possess an extremely high density of active sites towards glucose electro-oxidation. In addition, the deposition on highly porous electrodes decreases the poisoning effect of chlorides. These findings show significant promise as a scalable fabrication strategy to glucose sensor development. References in Table 1: 1 https://www.sciencedirect.com/science/article/pii/S001346861400766X#bib0060 2 https://pubs.acs.org/doi/10.1021/jp810235v 3 https://www.sciencedirect.com/science/article/pii/S0925400515007133?via%3Dihub 4 https://www.sciencedirect.com/science/article/pii/S0039914012008971 5 https://pubs.acs.org/doi/10.1021/ac035143t Figure 1