DROPS ON RAILS
R. Dangla, S. Lee, and C.N. Baroud* LadHyX and Department of Mechanics, Ecole Polytechnique, CNRS, FRANCE ABSTRACT We present a method to control the motion of drops in a wide and thin microchannel, by etching fine patterns into the top surface. Such control is possible for drops that are squeezed by the channel roof, by allowing them to reduce their surface energy as they enter into a local depression. The resulting reduction in surface energy pulls a drop into the groove such that localized holes can be used to anchor drops, while linear patterns can be used as rails to guide them along complex trajectories. An anchored drop remains stationary as long as the velocity of the surrounding oil is below a critical value which depends namely on the drop size and channel geometry. A rail guides drops whose size is below a critical radius which can be used to separate drops based on size or on their physical properties. KEYWORDS: Droplet microfluidics, droplet arrays, digital microfluidics, rails. INTRODUCTION Several groups have recently begun to explore two-dimensional (2D) arraying of micro-drops in closed microchannels [1], in order to reproduce some of the major advantages of digital microfluidics within the well-controlled environment of droplet microchannel transport. Through these techniques, nano-liter drops can be held stationary while the carrier fluid continues to flow, allowing long measurements to be performed and reactions to take place. However, all of the work on two-dimensional droplet arrays has relied so far on random or quasi-random placing of drops in the traps. While this is acceptable for samples with one or two types of droplets, this approach is not scalable when a multitude of drop contents needs to be observed. THEORY We have recently developed a method for manipulating drops in a two-dimensional field in the absence of lateral walls, by using forces due to the surface energy of the drops. We begin with drops that are strongly confined in one direction (pancake drops). A “rail” that is etched in one of the microchannel surfaces, as sketched in Fig. 1, allows the drop to reduce its surface energy by partially entering it. In this way, when a drop is pushed by the outer fluid, it will follow the etched grooves even for thin rails, as shown in the experimental images of Fig. 2. The drops all follow the rails, although they could in principle go anywhere in the wide microchannel. Since these rails are etched using lithography techniques, any geometry can be designed and the drops will follow the easiest path (Fig. 2b).
Figure 1: Working principle of the rails: By providing a way for the squeezed drop to expand into the rail, the surface energy is reduced. In this way, the drops will follow the rail as they are pushed by the external fluid. EXPERIMENTAL The microchannels are fabricated using simple soft-lithography techniques. Two-layer micro-fabrication is performed by exposing the base layer (microchannel) and and upper layer (the rails) separately in a solid resin. PDMS is then poured onto the master in order to obtain the desired three-dimensional structure in the microchanels. The PDMS is bonded on a microscope slide using an oxygen plasma treatement. The experimentes are performed after the channels are made hydrophobic by passing Aquapel through them. The current experiments show water drops in FC40, a fluorinated oil.
978-0-9798064-3-8/µTAS 2010/$20©2010 CBMS
2059
14th International Conference on Miniaturized Systems for Chemistry and Life Sciences 3 - 7 October 2010, Groningen, The Netherlands
Figure 2: A train of drops follows the rails in a wide microchannel (width=3 mm), regardless of the complexity of the network. Drops can be much larger than the rail width (part a) or smaller (part b).
RESULTS AND DISCUSSION This principle can be extended to simpler circular holes. In this case, the energy reduction is localized in both directions and a drop which reaches the hole will be trapped. This is true as long as the force due to the external flow, which tends to push the drop away, is smaller than the force necessary to keep the drop in place. Again, the small scales associated with microfluidics favour surface forces and we find the surprising effect that drops can be trapped by holes measuring a fraction of their size, even for high flow velocities, as shown in Fig. 3a. Indeed, the drop will remain trapped as long as the velocity of the carrier phase is below a threshold value, which scales as the within the inverse of the square of the radius (R) of the trapped droplet. This theoretical result has been confirmed experimentally, as shown in Fig. 3b for a particular trap geometry and oil viscosity. Holes can therefore be positioned in geometric patterns to produce arrays of stationary droplets.
Figure 3: (a) A drop that is trapped in a drop hole. (b) Threshold flow velocity necessary to dislodge drop from a trap. CONCLUSION While different in function, anchors and rails are both based on the principle that a confined microdroplet prefers to lower its surface energy by moving to a location with less confinement. More specifically, anchors introduce an axisymmetric increase in channel height and have shown to hold droplets in place for a range of external flow velocities below the critical value. This critical velocity is predicted theoretically to scale as R−2, where R is the droplet radius, and has been confirmed experimentally as well. Furthermore, the strength of a particular anchor appears to be only a function of the anchor geometry and not of the droplet size. 2060
Rails, on the other hand, provide a linear indentation in the channel height that can guide the droplet to different positions inside the channel. Performing a similar analysis as for holes, the critical flow rate at which the droplet moves with the external flow, rather than the rail, is predicted to vary as R−1. The detailed calculations and experimental confirmation are still under investigation. Furthermore, combining these two simple geometrical variations promises to be an effective way to control droplets in two-dimensional microfludics. For instance, rails can be used to guide the droplets and deposit them over holes, where the droplets can remain stationary and monitored over time. Such different configurations of anchors and rails are also being actively investigated. The strength of this method is to combine the anchors with the rails, in order to position several populations of drops at pre-determined locations. This can involve bringing drops of variable content near each other, or alternatively separating initially similar drops. When used in conjunction with the laser-control technique of droplets, developed in our lab [2], this approach can provide completely controlled 2D droplet microfluidics within microchannels, which can bridge the gap between micro-channel droplet control and surface droplet manipulation using embedded fabrication. Finally, an application of this technique to the observation of sickling of red blood cells is demonstrated in MicroTAS 2010 paper number 0659 [3]. ACKNOWLEDGEMENTS Surfactants for fluorinated oils were graciously provided by Abdessalam el Harrak and Jean-Christophe Baret. The authors also acknowledge useful discussions with Paul Abbyad and Antigoni Alexandrou. S.Lee is partially funded by a Chateaubriand fellowship and from the DRE at Ecole Polytechnique. REFERENCES [1] A. Huebner, D. Bratton, G. Whyte, M. Yang, A.J. deMello, C. Abell, and F. Hollfelder. Static microdroplet arrays: a microfluidic device for droplet trapping, incubation and release for enzymatic and cell-based assays. Lab Chip, 9(5):692–698, 2009. [2] C.N. Baroud, M.R. de Saint Vincent, and J-P. Delville. An optical toolbox for total control of droplet microfluidics. Lab Chip, 7:1029–1033, July 2007. [3] P. Abbyad, R. Dangla, P.-L. Tharaux, A. Alexandrou and C. N. Baroud. Sickling red blood cells in droplet arrays. MicroTAS 2010, Paper number 0659. CONTACT *Charles Baroud, tel: +33.1.69.33.52.61;
[email protected]
2061
