Abstract
This Ph.D. thesis focuses on the development and characterization of a flexible, three-electrode based biosensing system for the electrochemical detection of interleukin 6 (IL-6) and tumor necrosis factor-alpha (TNF-α) in sweat. The research encompasses sensor design, fabrication, biofunctionalization, and evaluation under physiologically relevant conditions. The work begins with a comprehensive review of state-of-the-art electrochemical methods for cytokine detection in wearable applications. It summarizes fundamental principles, fabrication techniques, biofunctionalization strategies, and key challenges, providing the context and motivation for the experimental work. Building on this overview, the methodology outlines the workflow for developing and evaluating gold nanoparticle (AuNP)-aptamer-functionalized sensors for wearable IL-6 and TNF-α detection. It includes sensor fabrication, microfluidic system design, sensor characterization, iontophoresis studies using skin analogs, and data analysis. Based on this framework, the results are summarized, with detailed analyses of each experimental phase provided in the subsequent sections, in line with the paper-based dissertation format. A complete list of journal publications and conference proceedings forming the core of this research is included in the appendices. The experimental work progressed through five interconnected phases, each building on the out comes of the previous one. The first part focused on the fabrication and optimization of the transducing platform, with particular attention to electrode geometry and process standardization. Four flexible, polyethylene terephthalate (PET)-based three-electrode designs were systematically characterized using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS), supported by multiphysics simulations. Refinement of the electrodes optimized the balance between compactness for wearable integration and electrochemical performance, with a larger active electrode area and reduced interelectrode spacing enhancing current response and lowering charge-transfer resistance. Building on the optimized platform, the second phase refined the fabrication protocol for AuNP modified sensors and their biofunctionalization with commercially available IL-6-specific ap tamers. EIS characterization in artificial sweat (AS) demonstrated a limit of detection (LOD) of 0.2 pg/mL across a range of 0.2–200 pg/mL, fully covering physiological IL-6 concentrations in sweat (5–15 pg/mL). While traditional CV showed limitations in differentiating subtle con centration changes, machine learning (ML)-enhanced CV analysis using a k-Nearest Neighbors (kNN) model distinguished physiological (20 pg/mL) IL-6 concentrations with 96.7% accuracy. The third phase of the research extended the approach to TNF-α detection. Employing TNF-α iv specific aptamers, real-time measurements under continuous AS flow in a preliminary microfluidic system demonstrated robust performance across physiological concentrations with minimal interference from nonspecific analytes. The system achieved an LOD of 3.2 pg/mL over a range of 0.2–1000 pg/mL, fully covering the physiological TNF-α range in sweat (9–362 pg/mL). Building on that, the fourth phase expanded detection to multiplexed monitoring of IL-6 and TNF-α within a single platform by employing a sensor array. The array was integrated with a simulation-guided microfluidic system to optimize flow, minimize recirculation, and ensure efficient sweat transport while maintaining full hand flexibility. Performance was validated using human sweat samples collected under ethical approval, as detailed in the appendices. The sensor array enabled multiplexed detection of IL-6 (LOD: 5.11 pg/mL, range 0.5–200 pg/mL) and TNF-α (LOD:2.64 pg/mL, range 0.5–1000 pg/mL) under competitive conditions in human sweat. Finally, the fifth phase investigated controlled sweat induction using a lab-scale iontophoretic platform on skin analogs. Applied currents facilitated ionic transport across the analogs, with total transferred charge increasing proportionally with iontophoretic stimulation time and in tensity, demonstrating cumulative ion migration below human sensory thresholds. This proof of-concept highlights the feasibility of on-demand sweat extraction using wearable-compatible electrodes suitable for on-body applications. Overall, this research establishes a comprehensive framework for the design and optimization of a flexible biosensing platform for real-time cytokine monitoring in sweat. By integrating simulation-driven electrochemical optimization, wearable-compatible biofunctionalization, ML-assisted detection, and microfluidic system design, it provides a robust foundation for next-generation wearable health monitoring technologies.