Electrical Control of Individual Droplet Breaking and Droplet Contents

Feb 21, 2011 - ... Cited-by Linking service. For a more comprehensive list of citations to this article, users are encouraged to perform a search inSc...
9 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/ac

Electrical Control of Individual Droplet Breaking and Droplet Contents Extraction Shaojiang Zeng, Xiaoyan Pan, Qingquan Zhang, Bingcheng Lin,* and Jianhua Qin* Department of Biotechnology, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian, China

bS Supporting Information ABSTRACT: The controllable extraction of the contents from an individual droplet is very important for the analysis and further processing of droplets. Most of the available methods lack the control and flexibility over the extraction of droplet contents. Here, we present a novel electrical-based method that can selectively break a droplet and extract its contents into aqueous buffer in a controllable manner on a microfluidic device. The device consisted of two layers, in which the top layer was modified to be hydrophobic for droplet generation and the bottom layer modified to be hydrophilic for maintaining aqueous buffer. A stable oil/water interface was formed at the intersection of the two layers where the oil and aqueous buffer met. When a droplet flew through the oil/water interface, a voltage was applied in the aqueous buffer. The generated electro-osmotic flow in the aqueous buffer and the electric field facilitated the breaking of the water/oil interface between the droplet and aqueous buffer, promoting the breaking of the droplet and a transient coalescence of the droplet with aqueous buffer. During the transient coalescence, droplet contents could be extracted into aqueous buffer. The amount of the droplet contents extracted into the aqueous buffer could be controlled by varying the strength of the electric field, and the droplet still existed and could be further processed after complete extraction of the droplet contents. This method could be used for droplet analysis and may provide a way to perform complex fluid handling in a droplet.

D

roplets in microfluidic systems have recently received intense interests as microreactors to perform a diverse range of reactions and assays.1-8 The droplet has a number of excellent properties that hold huge promise in chemistry and biology. The small size of droplets can significantly reduce reagent consumption, and when in combination with fast droplet generation, it is especially suitable for high-throughput screening applications.9-11 Rapid mixing within droplets by chaotic advection is particularly useful to study reaction kinetics.12 Encapsulation of reagents by immiscible oil can eliminate dispersion effect and prevent cross contamination between droplets as well as the evaporation of solution. These unique properties have been exploited for a number of applications including protein crystallization,9 directed evolution of proteins,11 enzyme kinetics study,12 micro- and nanoparticle synthesis,13,14 and single cell analysis.15,16 Up until now, numerous methods and techniques, including generation,17,18 sorting,19-21 fusion,22,23 splitting,24 and storage25 of droplets, have been developed and employed to manipulate and process droplets to fully exploit the potential of droplets as microreactors. However, because of the isolated nature of droplets by immiscible oil and the tiny size of droplets, it is very difficult to extract the contents in a droplet after the reagents and samples have been encapsulated into the droplet and reaction takes place. This problem prevents multistep processing of the droplet, which is usually required for many assays. One approach to the problem is to collect the droplets, break them outside the device, extract the target products, and then perform further analysis or r 2011 American Chemical Society

other processes. Polymerase chain reaction of single copy DNA has been carried out in a continuous flow of droplets, and the reaction product was analyzed and processed in this way.26 Thus, if the extraction of the contents of a droplet could be performed online and connected to other analysis methods, the applications of droplet as microreactor could be expanded to a large extent.27,28 Very recently, several methods, utilizing local surface modification,29-31 electric field,32,33 or surface tension,34 have been developed to extract droplets into a continuous aqueous stream followed by electrophoresis or mass analysis. However, it still remains a challenge to controllably extract the contents from a droplet. Here, we describe an electrical-based method that can controllably extract the contents of a droplet into aqueous buffer. This method used a two-layer microfluidic device, with oil and droplets flowing in the top layer and aqueous buffer in the bottom layer. A stable oil/water interface was formed in the device. When a droplet flew over the oil/water interface, a voltage was applied in the aqueous buffer. The generated electro-osmotic flow in aqueous buffer and the electric field facilitated the breaking of the droplet and promoted a transient coalescence of the droplet with aqueous buffer. During the transient coalescence, the droplet contents could be extracted into aqueous buffer. The amount of droplet contents extracted into aqueous buffer could be controlled Received: November 3, 2010 Accepted: January 6, 2011 Published: February 21, 2011 2083

dx.doi.org/10.1021/ac1028775 | Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry

Figure 1. Schematic illustration of the droplet microfluidic device; inlet showed the formation of a stable oil/water interface in the device; dye solution was used as aqueous buffer for demonstration.

by varying the strength of the electric field. With optimum droplet size and appropriate voltage, the contents of a droplet could be completely extracted, which may find applications for analysis of individual droplets.

’ EXPERIMENTAL SECTION Chip Fabrication. The two-layered microfluidic device (Figure 1) was fabricated using PDMS according to the standard soft lithography procedure. Briefly, Su-8 3035 photoresist (MicroChem Corp.) was spin-coated onto cleaned glass wafers, followed by prebaking, pattern transfer, postbaking, development, and hardbaking. The top layer took a flow-focusing structure for droplet generation. The width of the flow-focusing part of the channel was 35 μm, and other parts of the channel had a width of 80 μm; the height was 55 μm unless specified. The bottom channel had a general width of 35 μm and narrowed to 20 μm at the intersection with the top channel. The height of this channel was 40 μm and the distance between the two inlets was 2.5 cm. Filter structures were added right after the inlets to prevent the blockage of the narrowed part in the bottom channel due to dust in the oil or aqueous solution. After casting PDMS (Dow Corning Corp.) over the molds, followed by curing, peeling off, and the drilling of inlet holes, the two layers were bonded using oxygen plasma treatment. Immediately after the oxygen plasma treatment, a few drops of methanol were added between the two PDMS slabs to allow the alignment of the two layers under a microscope. The two-layered PDMS slab was heated at 80 °C overnight. Surface Modification. A two-step treatment procedure was used to make the two layers hydrophobic and hydrophilic, respectively. The hydrophilic modification was carried out first and it was based on covalent grafting on a PDMS surface.35 Briefly, the bottom channel was first flushed with benzophenone in acetone (10%, w/w) for 10 min by adding benzophenone solution into the two inlets of the bottom channel and applying vacuum at the waste port of the bottom channel. Later, the PDMS device was put in an 80 °C oven for 10 min to remove the acetone in PDMS and dry the PDMS channel. Then a water solution containing 10% (w/w) acrylamide monomer, 0.5% (w/w) acrylic acid, 0.5 mM NaIO4, and 0.5% (w/w) benzyl alcohol was infused into the bottom channel through one of the inlets of the channel. Because of the hydrophobic nature of PDMS and the surface tension, and with slow infusing, the solution would not enter the

ARTICLE

top channel when flowing through the intersection. After exposure to 12 J/cm2 of UV light (365 nm), the device was thoroughly washed with deionized water and the hydrophilic modification was completed. Before carrying out the hydrophobic modification, the device was washed with anhydrous ethanol followed by heating at 80 °C for half an hour to dry the channel completely. Then 50 μL of 5% (v/v) 1H,1H,2H,2H-perfluorooctyltrichlorosilane in FC-40 was added into the device when it was still hot and the device was immediately put into a desiccator for 30 min, followed by washing with anhydrous ethanol. Octadecyltrichlorosilane in anhydrous hexadecane could also be used, but more attention should be paid to prevent the moisture during the procedure. Since the polyacrylamide and carboxyl did not react with silanization reagent, it generally did not affect the hydrophilic modification. Sometimes, glycerol was added into the bottom channel before adding silanization reagent to protect the hydrophilic modification. After surface modification, two platinum electrodes were inserted into the inlets of the bottom channel from the side of the device and sealed with PDMS. Instruments. Syringe pumps (Longerpump, LSP02-1B, Hebei, China) were used to deliver the oil and aqueous phases into the device. Oil was carried in a 1 mL plastic syringe and ink solution in a 100 μL glass syringe. An inverted microscope (Olympus IX 71, Japan) with a CCD (MicroPublisher RTV 5.0, QImaging) was used to capture the images of the droplets. The integrated optical density (IOD) data of droplets was obtained using image analysis software Image-Pro (Media Cybernetics) by carefully selecting the droplet region and subtracting the background of the same region. Droplet Generation. For the generation of droplets, mineral oil containing 0.3% (w/w) Span 80 was used as the oil phase. To test the effect of surfactant, mineral oil containing various concentrations of Span 80 (0.6%, 0.1%, 0.01%, and 0.001%, w/w) and FC-40 containing 2% (w/w) EA surfactant were also used as the oil phase; 20 mg/mL amaranth in 10 mM sodium borate buffer (pH 9) was used as the aqueous phase to demonstrate the droplet coalescence and extraction process; ink solution in 10 mM sodium borate buffer (pH 9) was used as the aqueous phase for the demonstration of the complete extraction of droplet contents.

’ RESULTS AND DISCUSSION Formation of Stable Oil/Water Interface. The formation of a stable oil/water interface was critical for the controllable extraction of droplet contents. It has been previously shown that a stable oil/water interface could be formed using local surface modification29-31 or by flowing oil and aqueous solution side by side.32,33 Here we used a two-layer microfluidic device in combination with local surface modification to form a stable oil/water interface. The top channel was treated with silanization reagent to improve the hydrophobicity of the channel surface. The bottom channel was modified to be hydrophilic to maintain the aqueous buffer inside. A stable oil/water interface could be well formed at the intersection of the two layers where oil and the aqueous buffer met. The inlet in Figure 1 showed a stable oil/water interface. The local surface modification described above could also prevent the encroaching of oil into the bottom channel or aqueous buffer into the top channel after the formation of the oil/water interface. During the operation, the droplet flow was pumped by syringe pumps; the aqueous buffer was delivered by adjusting the height of the liquid level of the aqueous buffer (Figure 1). 2084

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry The advantage of supplying aqueous buffer in this way was that it could provide a stable driving force to the aqueous buffer at the oil/water interface and this was critical for the stabilization of the oil/water interface. Initially, the height of the liquid level of the aqueous buffer was set at a higher level to prevent the oil from entering the buffer channel. After the stabilization of droplets generation, the liquid level was lowered slowly and a stable oil/ water interface was formed. Controlled Breaking of a Droplet. Conventionally, the extraction of the contents of a droplet could be achieved by isolating the droplet from oil into an aqueous stream or by transiently fusing the droplet with an aqueous stream and transferring part of its contents into the aqueous stream.27-34 All these methods involved the breaking of a droplet and the coalescence (complete or transient) of the droplet with an aqueous stream. Here we employed a novel method to facilitate the breaking of a droplet and promote a transient coalescence of the droplet with aqueous buffer. This method depended on the use of an electric field in aqueous buffer to facilitate the breaking process. Figure 2 showed the processes of a droplet flowing through the oil/water interface with and without the application of an electric field. In the absence of an electric field, as shown in Figure 2A, a droplet would flow through the oil/water interface without coalescing with the aqueous buffer. On the contrary, with an electric field applied in aqueous buffer, as shown in Figure 2B, a droplet would break and a transient coalescence occurred between the droplet and aqueous buffer. The coalescence efficiency was almost 100%. During the transient coalescence, mass exchange occurred between the droplet and aqueous buffer. The mass exchange would be discussed in more detail in the Extraction of Droplet Contents section later. In Figure 2B, part of the dye molecules in the droplet transferred into aqueous buffer and the transferred dye molecules immediately underwent electrophoresis and migrated upward under the electric field. We then characterized the effect of the oil flow rate on the coalescence of the droplet with aqueous buffer. However, results showed that the oil flow rate had little effect on droplet coalescence as long as the electric field was high enough and we have observed droplet coalescence and extraction at an oil flow rate up to 800 nL/min with the application of 600 V (data not shown). In order that the whole process of droplet coalescence and extraction could be captured, the droplet generation was set to be ∼1 Hz and the oil flow rate was generally in the range of 80-200 nL/min. The above results clearly demonstrated that the application of the electric field would facilitate the breaking of the droplet and promoting a transient coalescence between the droplet and the aqueous buffer. The process of the coalescence of a droplet with a wetting substrate has been extensively characterized in “chemistrode” by Chen.36 According to Chen’s work, two factors largely dictate the coalescence of droplets with the wetting surface: (1) the rate to drain the carrier oil to bring the droplets and wetting surface to a critical distance and (2) surfactant dynamics. The process described here was similar to that in “chemistrode”. When a droplet flew over the oil/water interface, there was a thin oil film between the droplet and the aqueous buffer. We attributed the observed promotion effect largely to the accelerated drainage of the oil film. Normally, when no electric field was applied in aqueous buffer, the oil film was drained away as a result of the external flow induced by a moving droplet.37 In the case where an electric field was applied in the aqueous buffer, an electro-osmotic flow was generated. This electro-osmotic flow, as well as the external flow induced by moving droplets, would

ARTICLE

Figure 2. Pictures showing the processes of a droplet flowing through the oil/water interface in the absence (A) and presence (B) of voltage. (A1) A droplet would flow through the oil/water interface without coalescing with the aqueous buffer in the absence of an electric field. (A2) The schematic of the process in a side view highlighting the droplet flowing through the oil/water interface. (B1) A droplet coalesced with aqueous buffer when flowing through the oil/water interface under a voltage of 500 V. (B2) The schematic of the process in a side view. Oil, mineral oil with 0.3% Span 80, 180 nL/min in part A and 80 nL/min in part B; aqueous solution, 20 mg/mL amaranth in 10 mM sodium borate buffer (pH 9), 20 nL/min in parts A and B. The height of the liquid level of aqueous buffer: 32 cm in part A and 35 cm in part B.

facilitate the drainage of the oil film between the droplet and aqueous buffer and thus to break the droplet and promote a transient coalescence of the droplet with the aqueous buffer. To validate this assumption, we investigated the effect of the strength of electric field on the coalescence of droplets with aqueous buffer. For the ease of vision, we adjusted the height of the liquid level of the aqueous buffer such that during the transient coalescence, aqueous buffer would flow into the droplet. Images right before and after coalescence were shown. As shown in Figure 3A, with the voltage increased from 200 to 1200 V, the relative position in a droplet where coalescence occurred moved gradually from the end half of the droplet to the front half (relative to the droplet moving direction and more data are shown in Figure S1 in the Supporting Information). The time interval between two adjacent frames of the CCD was about 20-25 ms. Under the experiment conditions, it would take 4-5 frames for a droplet to flow through the oil/water interface. Thus, the coalescence time (including the time for oil film drainage and oil film breaking) could be calculated to be about 100 ms under a low electric field (200 V) and to be about 20-25 ms under a high electric field (1200 V). Taking into consideration the time for aqueous buffer to flow into a droplet, the actual coalescence time was shorter than calculated above. In the process of coalescence, 2085

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry

ARTICLE

Figure 3. The effect of voltage on the coalescence of a droplet with aqueous buffer. Images right before (the upper row) and after (the bottom row) the coalescence were shown. (Ai-Aiii) As voltage increased, the position in a droplet where the droplet fused with aqueous buffer moved from the end to the front of the droplet (relative to the droplet flow direction). The coalescence of a droplet with aqueous buffer was marked by the increased length of the droplet. Oil, mineral oil with 0.3% Span 80, w/w, 180 nL/min; aqueous solution, 20 mg/mL amaranth in 10 mM sodium borate buffer, pH 9, 20 nL/min. (Bi-Biii) The coalescence of droplets with aqueous buffer in the presence of a high concentration of stabilizing surfactant. Oil, FC-40 with 2% EA, w/w, 150 nL/min; aqueous solution, ink solution diluted 3-fold with 10 mM sodium borate buffer, 30 nL/min. The height of the liquid level of aqueous buffer: 35 cm in part A and 50 cm in part B. Because of the different surface tension as a consequence of mineral oil and FC-40 as the carrier oil, the heights of the liquid level of aqueous buffer differed significantly in parts A and B.

the time for drainage of the oil film was much longer than that required for oil film breaking,37 so the promoted drainage of the oil film should play a major role in decreasing the coalescence time. The coalescence time here was longer than that in “chemistrode”, and this might be due to the higher viscosity of mineral oil (14.2-17 cs of kinetic viscosity at 40 °C and 2.2 cs for FC-40 at 25 °C). A second contribution to the facilitated breaking of a droplet might be the effect of the electro-osmotic flow on the surfactant at the oil/water interface. In the above discussion, mineral oil containing 0.3% (w/w) Span 80 was used as the carrier fluid. We then tried mineral oil with various concentrations of Span 80 (0.001%, 0.01%, 0.1%, 0.6%, w/w), and all showed similar results (data not shown). It is well-known that the presence of surfactant at the oil/water interface would prolong the time for oil film breaking and stabilize droplets, and the higher the concentration of surfactant, the more stable the droplets. However, it seemed that the concentration of Span 80 in mineral oil had little effect on stabilizing droplets under an electric field based on the above results. It has been reported that it would take some time for surfactant to accumulate at the oil/water interface, and the dynamics of the adsorption of surfactant at the oil/water interface played a key role for droplets stabilization.38 One possible explanation to the above results was that the electro-osmotic flow would decrease the effective concentration of surfactant at the oil/water interface and thus the stability of droplets. Given that mineral oil with Span 80 was not very good for stabilizing droplets, we then used FC-40 with 2% (w/w) EA surfactant (Raindance Technologies) as carrier fluid, which had been reported to perfectly stabilize droplets.39 To our surprise, it still exhibited similar results: the higher the electric fields, the shorter the coalescence time, see Figure 3B. This promotion of coalescence should be due to the effect of electro-osmotic flow on the dynamics of surfactant at the oil/water interface. Another possible mechanism might be that the electric field would induce rearrangement of the surfactant at the oil/water interface and decrease the stability of droplets. These results showed that this method could also be used in circumstances where stabilizing surfactants are required. Extraction of Droplet Contents. Besides the promotion effect on droplet breaking, one more important advantage of using an

electric field was that it could be used to control the extraction of the contents of an individual droplet. During the transient coalescence, mass exchange occurred between the droplet and aqueous buffer. The mass exchange could be mainly divided into two situations: the droplet contents transferring into aqueous buffer and aqueous buffer flowing into droplets. (1) When the height of the liquid level of aqueous buffer was below a critical value, the pressure of the aqueous buffer near the oil/water interface was lower than the pressure inside a droplet (depending on experimental conditions, including flow rates of oil and dye solution, the channel length downstream the oil/ water interface) and part of the droplet contents (here dye molecules) would transfer into aqueous buffer. Figure 4A showed the transfer of dye solution in droplets into aqueous buffer under different heights of the liquid level of the aqueous buffer (the change in droplet volume before and after the coalescence could be seen in Figure S2 in the Supporting Information), and Figure 4B showed the corresponding integrated optical density (IOD) of the transferred dye molecules. The IOD data was obtained by integrating the optical density of the area containing the extracted dye molecules and subtracting the optical density of the same area containing only aqueous buffer. It demonstrated that as the height of the liquid level of aqueous buffer increased, the amount of dye molecules released into aqueous buffer decreased gradually. The error bar indicated the standard deviation of the IOD of the extracted dye molecules for 17 droplets (n = 17). This dependence of the transfer of droplet contents on the height of the liquid level of the aqueous buffer, when in combination with electrically facilitated breaking of a droplet, provided a reliable way to control the release of droplet content and might find applications for droplet analysis. (2) When the height of the liquid level of the aqueous buffer was above the critical value, the pressure of the aqueous buffer near the oil/water interface was higher than the pressure inside a droplet and the buffer solution would flow into droplets. In this situation, the transfer of dye molecules into aqueous buffer would not happen. However, the unique advantage of using an electric field still made it possible to extract the dye molecules from the droplets. As has been shown in Figure 2B, the released dye molecules underwent electrophoresis immediately after transferring into aqueous buffer. This electrophoresis of the dye molecules could counteract the flow of aqueous buffer into the droplet. Since the velocity of electrophoresis was linearly related to the strength of the electric field, there should be a critical value of the electric field 2086

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry

ARTICLE

Figure 4. Factors that affect the extraction of droplet contents. (A,B) Effect of the height of the liquid level of aqueous buffer on the amount of droplet contents released into the aqueous buffer. Flow rates: oil, 80 nL/min; aqueous solution, 20 nL/min. Voltage: 500 V. (C,D). Effect of electric field strength on the amount of extracted droplet contents. Flow rates: oil, 200 nL/min, aqueous solution, 30 nL/min. Oil, mineral oil with 0.3% Span 80, w/ w; aqueous solution, 20 mg/mL amaranth in 10 mM sodium borate buffer. The height of the liquid level of aqueous buffer was 30 cm. The data in parts A and C were obtained from two devices, and the height of the liquid level could not be compared directly.

(depending on the mobility of the dye molecules and the difference between the pressure inside a droplet and the pressure of the aqueous buffer near the oil/water interface) beyond which the dye molecules could be extracted into aqueous buffer even in the presence of a net flow of aqueous buffer into the droplet. Figure 4C showed the effect of the voltage on the extraction of dye molecules (more data and the change in droplet volume before and after the extraction could be seen in Figure S3 in the Supporting Information). As expected, there was a critical value of voltage (∼700 V under the experimental conditions). The integrated optical density of the extracted droplet contents, shown in Figure 4D, indicated that beyond the critical voltage, the amount of extracted dye molecules increased almost linearly with the voltage (n = 15). On the basis of the above observation and analysis, we then tried to extract the contents of a droplet completely using an electric field. Increasing the electric field was a choice but it was not practically feasible. It could be seen in Figure 4C that only a small fraction of the dye molecules had been extracted into the aqueous buffer. Decreasing the size of a droplet might be another choice. In order to estimate the maximum droplet that could be completely extracted and given the potential applications of this technique in bioassays, we arbitrarily assumed the mobility of the molecules to be extracted from a droplet was on the scale of 1  10-4 cm2 V-1 s-1, which agreed with the mobility of many biomolecules. Also under an electric field of 600 V cm-1(1500 V), in the period of 100 ms (the time for a droplet to flow through the oil/water interface under the experimental conditions and the coalescence time was neglected), the molecules with the assumed mobility would migrate a distance of 60 μm under electrophoresis. During the transient coalescence with aqueous buffer, the droplet could be regarded as a semisphere and taking into consideration the distribution of electric field inside the droplet (Figure S4 in the Supporting Information), the maximum diameter

of the droplet would be 60 μm  2/π ≈ 38 μm. This was only the theoretical maximum under the given conditions. The actual size depended on the molecules to be extracted and their surrounding medium. Here, we used ink solution diluted 3-fold with sodium borate buffer (pH 9) to test the feasibility. We decreased the height of the droplet generation channel to 30 μm and generated droplets with ∼35 μm diameter. Figure 5A showed the process of a droplet containing ink solution flowing through the oil/water interface. It clearly demonstrated that the droplet became almost blank in color after extraction. Integrated optical density data (not shown) also indicated that the dye molecules in the droplet have indeed been completely extracted. It was noted that the droplet became larger after extraction due to the flow of aqueous buffer into the droplet. In order to make it more convincing that the lack in color after extraction was the result of extraction rather than dilution by aqueous buffer, we diluted the droplet ∼10-fold (based on area of droplets images) with aqueous buffer (see Figure S5 in the Supporting Information), and from the result we could conclude that the lack in color was indeed the result of extraction. We then added magnetic beads into the ink solution to test whether the magnetic beads could be maintained in the droplets under a magnetic field during the extraction process. Results showed that with a sufficient magnetic field, the magnetic beads did remain in the droplets after extraction, which was shown in Figure 5B. This operation, when in combination with the droplet fusion technique, might provide a way to perform the washing step for magnetic bead-based assays in a droplet. During the experiment, we noted that droplets would slow down due to surface tension immediately after coalescing with aqueous buffer, and the actual time for extraction was longer than estimated (see Figure 5). The prolonged extraction time was helpful for complete extraction of the dye molecules in the droplet, but it would also lead to unwanted coalescence with the 2087

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry

ARTICLE

controlled extraction of the droplet contents may find use for online droplet analysis. The complete extraction of the contents of a droplet containing magnetic beads, when in combination with the droplet fusion technique, may provide a way to perform multistep operations in a droplet. We expect this method could improve the manipulation of droplets in microfluidic systems and extend the applicability of droplet-based microfluidics.

’ ASSOCIATED CONTENT

bS

Supporting Information. Pictures showing the detail of the coalescence of droplets with aqueous buffer, the extraction of droplet contents, the distribution of the electric field in a droplet after the coalescence of the droplet with aqueous buffer and the effect of droplet size on droplet coalescence, videos recording the processes of a droplet flowing through the oil/water interface without and with the application of voltage, and the process of droplet extraction. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Fax: (þ86) 411-84379650. E-mail: [email protected] (B.L.); [email protected] (J.Q.).

Figure 5. Pictures showing the process of complete extraction of droplet contents. (A) Complete extraction of the dye molecules in a droplet after passing through the oil/water interface. Oil flow rate, 150 nL/min; dye solution flow rate, 40 nL/min; voltage, 1500 V. (B) Extraction of the dye molecules in a droplet containing magnetic beads and the magnetic beads remained in the droplet after the extraction. Oil flow rate, 100 nL/min; dye solution flow rate, 60 nL/min; voltage, 1200 V. The height of the liquid level of aqueous buffer: 45 cm in parts A and B.

following droplets. This could be prevented by increasing the oil flow rate and adjusting the rate of droplet generation. Another issue was the ratio of droplets that have achieved complete extraction. In a total number of 72 droplets (without magnetic beads), based on IOD data, 16 droplets were completely extracted, 25 droplets partly extracted (ranged from 10% to 80%), and 31 droplets did not coalesce with aqueous buffer. This might be due to the decrease in the size of the droplets. As the droplets became smaller, it would be more difficult to drain the oil between the droplets and aqueous buffer (Figure S6 in Supporting Information). In some cases, the diameter of the droplet was smaller than the height of the channel, and the droplet would not touch the oil/water interface, leading to the failure of coalescence.

’ CONCLUSIONS We presented a novel method that used an electric field to facilitate the breaking of a droplet and promote a transient coalescence of the droplet with aqueous buffer. Beyond a critical strength of the electric field, a droplet would coalesce with aqueous buffer on a millisecond time scale even in the presence of a high concentration of stabilizing surfactant. During the transient coalescence, the contents of a droplet could be extracted. The amount of extracted droplet contents could be controlled by varying the strength of the electric field. With proper droplet size and electric field, the droplet contents could be completely extracted. This

’ ACKNOWLEDGMENT This research was supported by the National Nature Science Foundation of China (Grant 90713014), Chinese National Programs for High Technology Research and Development (863 Program, Grant 2006AA020201), Key Project of Chinese National Programs for Fundamental Research and Development (973 Program, Grants 2007CB714505 and 2007CB714507), and Knowledge Innovation Program of the Chinese Academy of Sciences (Grant KJCX2-YW-H18). We thank RainDance Technologies (MA, USA) and 3M (MN, USA) for the supply of surfactant EA and FC-40. ’ REFERENCES (1) Song, H.; Chen, D. L.; Ismagilov, R. F. Angew. Chem., Int. Ed. 2006, 45, 7336–7356. (2) Teh, S. Y.; Lin, R.; Hung, L. H.; Lee, A. P. Lab Chip 2008, 8, 198– 220. (3) Huebner, A.; Sharma, S.; Srisa-Art, M.; Hollfelder, F.; Edel, J. B.; Demello, A. J. Lab Chip 2008, 8, 1244–1254. (4) Chiu, D. T.; Lorenz, R. M. Acc. Chem. Res. 2009, 42, 649–658. (5) Chiu, D. T.; Lorenz, R. M.; Jeffries, G. D. M. Anal. Chem. 2009, 81, 5111–5118. (6) Theberge, A. B.; Courtois, F.; Schaerli, Y.; Fischlechner, M.; Abell, C.; Hollfelder, F.; Huck, W. T. S. Angew. Chem., Int. Ed. 2010, 49, 5846–5868. (7) Tewhey, R.; Warner, J. B.; Nakano, M.; Libby, B.; Medkova, M.; David, P. H.; Kotsopoulos, S. K.; Samuels, M. L.; Hutchison, J. B.; Larson, J. W.; Topol, E. J.; Weiner, M. P.; Harismendy, O.; Olson, J.; Link, D. R.; Frazer, K. A. Nat. Biotechnol. 2009, 27, 1025–U1094. (8) Dittrich, P. S.; Jahnz, M.; Schwille, P. ChemBioChem 2005, 6, 811. (9) Zheng, B.; Roach, L. S.; Ismagilov, R. F. J. Am. Chem. Soc. 2003, 125, 11170–11171. (10) Hatakeyama, T.; Chen, D. L.; Ismagilov, R. F. J. Am. Chem. Soc. 2006, 128, 2518–2519. (11) Agresti, J. J.; Antipov, E.; Abate, A. R.; Ahn, K.; Rowat, A. C.; Baret, J. C.; Marquez, M.; Klibanov, A. M.; Griffiths, A. D.; Weitz, D. A. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 4404–4409. 2088

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089

Analytical Chemistry

ARTICLE

(12) Song, H.; Ismagilov, R. F. J. Am. Chem. Soc. 2003, 125, 14613– 14619. (13) Shestopalov, I.; Tice, J. D.; Ismagilov, R. F. Lab Chip 2004, 4, 316–321. (14) Xu, S. Q.; Nie, Z. H.; Seo, M.; Lewis, P.; Kumacheva, E.; Stone, H. A.; Garstecki, P.; Weibel, D. B.; Gitlin, I.; Whitesides, G. M. Angew. Chem., Int. Ed. 2005, 44, 724–728. (15) Zhan, Y. H.; Wang, J.; Bao, N.; Lu, C. Anal. Chem. 2009, 81, 2027–2031. (16) Shim, J. U.; Olguin, L. F.; Whyte, G.; Scott, D.; Babtie, A.; Abell, C.; Huck, W. T. S.; Hollfelder, F. J. Am. Chem. Soc. 2009, 131, 15251– 15256. (17) Song, H.; Tice, J. D.; Ismagilov, R. F. Angew. Chem., Int. Ed. 2003, 42, 768–772. (18) Anna, S. L.; Bontoux, N.; Stone, H. A. Appl. Phys. Lett. 2003, 82, 364–366. (19) Ahn, K.; Kerbage, C.; Hunt, T. P.; Westervelt, R. M.; Link, D. R.; Weitz, D. A. Appl. Phys. Lett. 2006, 88, 024104. (20) Baret, J. C.; Miller, O. J.; Taly, V.; Ryckelynck, M.; El-Harrak, A.; Frenz, L.; Rick, C.; Samuels, M. L.; Hutchison, J. B.; Agresti, J. J.; Link, D. R.; Weitz, D. A.; Griffiths, A. D. Lab Chip 2009, 9, 1850–1858. (21) Franke, T.; Abate, A. R.; Weitz, D. A.; Wixforth, A. Lab Chip 2009, 9, 2625–2627. (22) Link, D. R.; Grasland-Mongrain, E.; Duri, A.; Sarrazin, F.; Cheng, Z. D.; Cristobal, G.; Marquez, M.; Weitz, D. A. Angew. Chem., Int. Ed. 2006, 45, 2556–2560. (23) Ahn, K.; Agresti, J.; Chong, H.; Marquez, M.; Weitz, D. A. Appl. Phys. Lett. 2006, 88, 264105. (24) Link, D. R.; Anna, S. L.; Weitz, D. A.; Stone, H. A. Phys. Rev. Lett. 2004, 92, 054503. (25) Shi, W. W.; Qin, J. H.; Ye, N. N.; Lin, B. C. Lab Chip 2008, 8, 1432–1435. (26) Schaerli, Y.; Wootton, R. C.; Robinson, T.; Stein, V.; Dunsby, C.; Neil, M. A. A.; French, P. M. W.; deMello, A. J.; Abell, C.; Hollfelder, F. Anal. Chem. 2009, 81, 302–306. (27) Edgar, J. S.; Pabbati, C. P.; Lorenz, R. M.; He, M. Y.; Fiorini, G. S.; Chiu, D. T. Anal. Chem. 2006, 78, 6948–6954. (28) Niu, X. Z.; Zhang, B.; Marszalek, R. T.; Ces, O.; Edel, J. B.; Klug, D. R.; Demello, A. J. Chem. Commun. 2009, 6159–6161. (29) Roman, G. T.; Wang, M.; Shultz, K. N.; Jennings, C.; Kennedy, R. T. Anal. Chem. 2008, 80, 8231–8238. (30) Wang, M.; Roman, G. T.; Perry, M. L.; Kennedy, R. T. Anal. Chem. 2009, 81, 9072–9078. (31) Zhu, Y.; Fang, Q. Anal. Chem. 2010, 82, 8361–8366. (32) Fidalgo, L. M.; Whyte, G.; Bratton, D.; Kaminski, C. F.; Abell, C.; Huck, W. T. S. Angew. Chem., Int. Ed. 2008, 47, 2042–2045. (33) Fidalgo, L. M.; Whyte, G.; Ruotolo, B. T.; Benesch, J. L. P.; Stengel, F.; Abell, C.; Robinson, C. V.; Huck, W. T. S. Angew. Chem., Int. Ed. 2009, 48, 3665–3668. (34) Kelly, R. T.; Page, J. S.; Marginean, I.; Tang, K. Q.; Smith, R. D. Angew. Chem., Int. Ed. 2009, 48, 6832–6835. (35) Hu, S. W.; Ren, X. Q.; Bachman, M.; Sims, C. E.; Li, G. P.; Allbritton, N. L. Anal. Chem. 2004, 76, 1865–1870. (36) Chen, D.; Du, W. B.; Liu, Y.; Liu, W. S.; Kuznetsov, A.; Mendez, F. E.; Philipson, L. H.; Ismagilov, R. F. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 16843–16848. (37) Liu, Y.; Ismagilov, R. F. Langmuir 2009, 25, 2854–2859. (38) Baret, J. C.; Kleinschmidt, F.; El Harrak, A.; Griffiths, A. D. Langmuir 2009, 25, 6088–6093. (39) Holtze, C.; Rowat, A. C.; Agresti, J. J.; Hutchison, J. B.; Angile, F. E.; Schmitz, C. H. J.; Koster, S.; Duan, H.; Humphry, K. J.; Scanga, R. A.; Johnson, J. S.; Pisignano, D.; Weitz, D. A. Lab Chip 2008, 8, 1632– 1639.

2089

dx.doi.org/10.1021/ac1028775 |Anal. Chem. 2011, 83, 2083–2089