Efficient encapsulation with plug-triggered drop formation

Monodisperse microscale drops formed with microfluidic devices are useful for encapsulating cells, microgel particles, or even additional drops. These techniques are thus useful for applications ranging from high-throughput biology to monodisperse particle and capsule synthesis, which require encapsulation of such objects. However, it is challenging to efficiently encapsulate the objects in all drops; often, the objects are encapsulated inefficiently, resulting in many improperly filled, unusable drops. Here, we describe a phenomenon that allows very efficient encapsulation. We use the inflow of the object to plug the drop maker nozzle; the continued injection of the outer phase pinches off a drop, thereby encapsulating the object; this yields precisely one object encapsulated per drop.

Figure 1

Schematic of triggered drop formation device. Drops or microgels are introduced into the inner phase inlet of the device and the continuous phase from two side inlets; when the objects enter the nozzle, they plug it, triggering drop formation. The objects should be slightly deformable, allowing them to plug the nozzle without clogging it.

Figure 2

Image sequence of a small drop triggering the formation of a large drop. For the small drop to fit in the nozzle, it must distort from a spherical shape to a sausage shape; this allows it to plug the nozzle, restricting the path of the continuous phase, causing a pressure rise upstream that induces pinch off of a large drop. The channel width is 25 μm. Movies of the process are available in the Supplemental Material [24].

Figure 3

Dependence of triggering on inner drop size. (a) Schematic of valve-based flow focusing device for making double emulsions; using the valves, the inner drop size can be varied while maintaining constant flow rates in the triggering junction. (b) Images of double emulsions formed with different inner drop sizes. (c) The number of inner drops encapsulated per outer drop and (d) the size of the entire double emulsion as a function of the inner drop diameter divided by the channel width. When the inner drop is smaller than the nozzle, triggering does not occur; the number encapsulated is inversely proportional to inner drop size and the double emulsion diameter is roughly constant. By contrast, when the inner drop is larger than the nozzle, drop formation is triggered; every outer drop contains exactly one inner drop, and the size of the double emulsion scales with that of the inner drops.

Figure 4

Comparison of our data with three potentially applicable drop formation scenarios. The plots show the measured drop lengths divided by the channel width, as a function of control parameters relevant to the scenarios. (a) Shear-driven drop formation predicts scaling with the inverse of the Ca of the continuous phase. (b) Plugging drop formation predicts scaling with the flow rate ratio of the inner and middle phases, with respect to the continuous phase. (c) Plug-triggered drop formation predicts scaling with the triggering period times the middle phase flow rate. While our data neither collapse nor scale as predicted when compared with shear-driven or plugging drop formation, they do when compared with plug-triggered drop formation.

Figure 5

Scaling of middle-phase drop volumes with the triggering control parameter $T{Q}_m$. Each drop is composed of a volume of middle phase $V_{\mathrm{mid}}$ in which an object is encapsulated, so that the total volume $V_{\mathrm{drop}}$ = $V_{\mathrm{mid}}$ + $V_{ob}$, where $V_{ob}$ is the object volume. Increasing the middle phase flow rate $Q_m$ or the triggering period T, both lead to a proportional increase in the $V_{\mathrm{mid}}$. The solid line shows the scaling expected for triggering, in which $V_{\mathrm{mid}}$ = $T{Q}_m$.

Figure 6

Analysis of when triggering fails. (a) Image sequence of microgel-triggered drop formation; the green dashed line and red dotted line correspond to the locations at which the microgel and drop intervals, respectively, are measured. (b) The drop formation interval as a function of the microgel interval; for triggered drop formation, these intervals must be equal, so that there is a linear dependence between them with a slope of unity, which is what we observe for a majority of drops. However, occasionally microgels enter too quickly so that two are encapsulated, or too slowly so that empty drops are formed. For drop formation to be successfully triggered, the object frequency must be within ∼50% of the natural drop formation frequency of the device.