UC Berkeley

Sweat is an emerging biofluid that is non-invasively accessible and rich in biochemical information. It has traditionally been used for diagnosis of medical conditions, including cystic fibrosis, autonomic and small fiber neuropathy. Historically, sweat testing has been limited to clinical and research settings due to the need of specialized equipment with trained personnel. In this regard, wearable sensors open up opportunities for expanding sweat testing beyond the specialized environment and the established diagnosis of medical conditions. Major challenges in wearable sweat sensing research include (1) obtaining reliable sensors which can accurately provide the target’s signals and do not require repeated calibration for the wear duration, (2) designing an integrated device that mitigates the rapid evaporation, mixing, and contamination of freshly secreted sweat, and (3) interpreting potential sweat biomarkers to realize its role in health management. In this thesis, I demonstrated different wearable sweat sensors to address these challenges. Specifically, the first part of the thesis focuses on electrochemical sensors which allow multiplexed detection of target chemicals through modulation of sensing surfaces with suitable chemicals and materials on flexible plastic substrates. The second part includes integration of microfluidics with electrochemical and electrical sensors to deliver a device that can guide sweat flow and achieve ultra-low sweat measurement for sedentary individuals with and without active sweat stimulation. Lastly, a method for rapid and mass production of sweat sensors via roll-to-roll process is demonstrated, along with subject studies to explore the significance of sweat in relation to health.