Modeling and Control Challenges in the development of Discrete Microfluidic Devices

Droplet-based microfluidics is an emerging area that has important applications in chemical engineering and several other disciplines. For example, these devices have applications in the areas of bubble computers, cancer diagnostics and development of micro-factories on a chip. Some of the challenges that need to be overcome to realize the potential of this area are related to experimental uncertainties and development of devices that offer multiple functionalities, thereby reducing the footprint and aiding in miniaturization. Achieving precise functionalities in microfluidics is challenging because droplets exhibit complex dynamic behavior in these devices due to hydrodynamic interactions and discontinuities that are a result of discrete decision-making at junctions. Remarkably, transitions from periodic to aperiodic/chaotic behavior based on input conditions can be witnessed even in a simple loop device. Hence rational design frameworks that handle this complexity are required to make significant progress. Further, passive designs that achieve such functionalities might entail large footprint. As a result, active control might become a necessity for developing efficient device designs. The main purpose of this paper is the identification of the challenges and opportunities for the PSE community in this area of emerging importance. Further, our work in the area model predictive control (MPC) for active control in a microfluidic loop device will also be briefly described.