(549h) Droplet Traffic in a Microfluidic Loop

We study the dynamics of a train of confined droplets passing through a fluidic loop, where a main channel bifurcates into two branches and recombines. Prior works have shown that even though droplets enter the loop at uniform time intervals, they can exit the loop at either periodic or aperiodic time intervals. Despite a growing body of literature on this problem, there have been little direct comparisons between experiments and models, and the mechanisms that cause transitions between periodic and aperiodic behaviors remain unknown.

We conduct experiments in an easily assembled millifluidic loop device and record the dynamics of several hundreds of drops as a function of droplet spacing. In the given duration of our experiments, we observe three distinct dynamics – sustained periodic behaviors, intermittent behaviors (i.e., system switches between a finite number of periodic states) and aperiodic behaviors (i.e, the system shows a large number of periods). We compare the experimental data with simulations in which drops are modeled as mobile fluidic resistors. Using Poincare maps and period lengths as determined by a spacing quantization rule, we find good agreement with experimental data for certain inlet droplet spacings, particularly when period 2 behaviors and other stable periodic behaviors are exhibited.  In some cases, however, our simulations match experiments only after adjusting certain parameters (e.g. droplet resistance and inlet drop spacing) from our predicted values. We will discuss possible experimental non-idealities that need to be implemented in simulation models to better capture the experimentally observed behaviors.

Furthermore, using both simulations and experiments, we find that simultaneous droplet entrance and exit events are the main source of transitions from periodic to intermittent and aperiodic behaviors.  We also observe that the more distant droplet entrance and exit events are from each other, the more stable and resilient the resulting behavior will be to fluctuations in inlet drop spacing. Our results impact the design of microfluidic devices in which fluidic networks are used to autonomously control droplets for applications in lab-on-chip technologies.