17.7 Droplet Microfluidics Co-Designed with Real-Time CMOS Luminescence Sensing and Impedance Spectroscopy of 4nL Droplets at a 67mm/s Velocity

Hybrid bio-electronic systems use genetically engineered cell-based biosensors (CBBs) [1] augmented with CMOS integrated circuits (ICs) for biochemical sensing. CBBs achieve high specificity and sensitivity in harsh environments (e.g., human gut, wastewater, bioreactor, and soil), while CMOS ICs provide reliable operation and communication abilities. The hybrid bio-electronic systems represent a critical technology to help address societal challenges in healthcare (e.g., disease monitoring, closed-loop drug delivery, and disease biomarker discovery), environmental monitoring (e.g., water and soil quality), and sustainable manufacturing. To test and optimize biosensor performance for application deployment, traditional experimental protocols require manual pipetting of cells into well plates, growing cell cultures overnight, and quantifying optical CBB output (luminescence/fluorescence) and sample cell density via optical density measurement using a plate reader under different environmental conditions. Therefore, screening large libraries of biosensors (millions of combinations) under thousands of environmental and chemical conditions, e.g., temperature, pH, chemical composition, and inducer concentrations, would be near impossible. The creation of a modular droplet microfluidic $( \mu F)$ device utilizing aqueous droplets encapsulating cells emulsified in an oil-filled microfluidic channel with embedded CMOS electronic sensors (Fig. 17.7.1) will enable rapid screening of the biological design space for various environments, thus improving the biosensor performance to meet target sensitivity, specificity, and durability under application-specific conditions. Luminescence readout of CBBs is preferred for low-cost deployable systems since it delivers maximum volume- and power-efficient readout without requiring a light source or optical filters. However, as alluded to in previous droplet microfluidic works [2], when cells are processed using luminescence or fluorescence only, determining the generation of an observed signal by either a small number of cells with high protein expression or many cells with weak expression can be challenging without normalizing for cell density. To better characterize the performance of living biosensors, combining luminescence sensing with cell density measurement via impedance analysis is essential.