Controlled production of size-tunable Janus droplets for submicron particle synthesis using an electrospray microfluidic chip

Abstract

Size-tunable Janus droplets of two miscible liquids were produced and further employed for synthesizing Janus particles with submicron size. An electrospray microfluidic chip was fabricated to form the Janus droplets from two-phase laminar flow, and to quickly regulate the droplet size by an electric field. With the aid of a direct-current electric field, we could obtain Janus droplets with various diameters in the range from 135 μm down to 3 μm. This regulation method was adaptive to the droplets made up of both aqueous solutions and organic polymers. The mechanism for the dependency of drop size on the high voltage was discussed comprehensively. Finally, submicron Janus particles were synthesized taking the minimized Janus droplets as templates.

---

1. Introduction

Janus droplets with two distinct hemispheres are frequently employed as templates in the synthesis of Janus particles.^1,2 Janus particles are believed to hold great potential in interface stabilizing, catalysis, electronic-paper, biomedicine, etc.^3,4 Therefore, better control over the production of Janus droplets is critical for the engineering of particle size and structure. In terms of bulk production, Janus droplets are usually formed by an emulsification process, during which two immiscible oils are dispersed in an aqueous phase under intense mixing.^5,6 This approach allows for large-scale production of drops, but there exists one main limitation, the relative wide distribution range of drop dimension. Besides, the two phases of one droplet can only be made up of immiscible or separable liquids, and this issue may lead to an arduous screening of materials and the necessary for particular surfactants. A promising alternative is microfluidic methods. In 2004, Nisisako et al.⁷ designed a microfluidic device combining a Y-shaped channel and a sheath-flow junction, in which hemispherically colored micro-droplets from miscible monomers were formed with a coefficient of variation less than 1%. This method was considered to have opened up a path to microfluidic production of Janus micro-droplets and particles. In the last decade, a large number of researches focused on microfluidic droplet-based synthesis of Janus microparticles with various components, morphologies and functions. In the microfluidic droplet-based synthesis, monomers,⁷ hydrogels,⁸ resins,⁹ or even nanoparticles¹⁰ were employed as materials of Janus particles. And the particles could be engineered to spheres,^7–10 discs,¹¹ raspberries,¹² moons,¹³ etc. Moreover, the microfluidic synthesis devices varied from flow-focusing chips¹⁴ to coaxial capillaries⁹ or centrifuging systems.¹⁵ However, the dimension of particles generated from microfluidic droplet templates is usually limited at tens to hundreds of micrometers, which is mainly due to the restriction on the size of droplets formed in microchannels. The relatively large dimension blocks the particles from nanoscale applications. Although size regulation of droplets can be achieved by altering the flow rate of each phase in microfluidic systems, it remains a challenge to produce droplets with a diameter less than 10 μm, particularly droplets containing viscous monomers which are frequently used in microfluidic fabrication of Janus particles. Furthermore, excessive difference between the flow rates of dispersed and continuous phases may lead to over-dilution of products. Efforts have been made on tiny droplet formation, including nanofluidics,¹⁶ tip-streaming,¹⁷ gas bubble condensation,¹⁸ and recently-reported volatile component evaporation method.¹⁹ These strategies are robust in generating ultra small droplets even down to sub-micrometer scale. But some seem not to be applicable in Janus droplet formation from miscible liquids, in that severe mixing may occur across the two phases at the nanoscale. And the time lag during phase transition and solvent evaporation may intensify the mixing in droplets. Electrospray strategy is a promising alternative for tiny drop formation. Kim and colleagues²⁰ carried out electrospray in a microfluidic emulsification device with a flow-focusing geometry. The droplet size was controlled by the electric field in microchannels, and droplets were less than 1 μm when produced by a Taylor cone. This method was further extended to produce tiny polymer solution droplets by Yeh et al.,²¹ thus exhibited the possibility of spraying droplets for particle synthesis.

In the present protocol, the electrospray technique was applied to produce Janus droplets in a microfluidic flow-focusing device. It has unique advantages in the formation of tiny Janus droplets: (i) specific structures, such as ultra narrowed orifice, are unnecessary under the active regulation mode; (ii) fast response can be achieved for the charged droplets in an electric field; (iii) tiny droplets would be facilely formed even for viscous dispersed phases. These advantages exactly allowed for producing small droplets on the premise of maintaining the biphasic morphology of Janus droplets. The dependency of drop size on the high voltage was quantified, and possible mechanism was discussed in detail. In addition, the dispersed phases from aqueous solutions to organic polymers were tested to show the wide applicability of the electric regulation. Taking the minimized Janus droplets as templates, submicron Janus particles were synthesized.

2. Experimental

2.1 Microfluidic device

The microfluidic device comprises two layers: a channel layer fabricated from polydimethylsiloxane (PDMS, RTV615, Dow Corning, USA), and an electrode layer made of indium tin oxide (ITO) glass. For the electrode layer, chemical wet etching was used to form the ITO electrodes after lithography. Detailed process of microchip fabrication was available in ESI. The whole structure of the electrospray microfluidic device is illustrated in Fig. 1. The microchannel network was designed into a cross-shaped flow-focusing geometry with Y-shaped dual inlets for introducing the dispersed phases, and the continuous phase was introduced from the side inlets. The channel widths are 50 μm, 100 μm and 200 μm for Y tributaries, cross tributaries and the main downstream channel, respectively. The channel depth is all 75 μm. The ITO was etched to a pair of parallel stripes. The one contacting the dispersed phases was connected to the high voltage supply, and the other located at the downstream channel was grounded. When the voltage supply was turned on, an electric field was formed between the electrodes. The width of the electrode stripes is 0.5 mm, and the electrodes are 1 mm apart.

Fig. 1 Schematic diagram of the electrospray microfluidic chip.

2.2 Reagents and apparatus

Deionized (DI) water was used throughout the experiments. Carbon ink was selected for colouration of Janus droplets. 2 wt% sodium alginate (Sigma, USA) solutions were prepared by dissolving the salt into water or ink-doped water. Poly(lactic-co-glycolic acid) (PLGA, Aldrich, USA) was dissolved into dimethyl sulphoxide (DMSO, Sinopharm Chemical Reagent Co., China) to a concentration of 1 wt%. Rhodamine B (Sinopharm Chemical Reagent Co., China) and fluorescein isothiocynate (FITC, Sigma, USA) were doped at 1 mg mL⁻¹ respectively to the two phases of Janus droplets for particle characterization. The continuous phase was mineral oil (M8410, Sigma, USA) with 5 wt% Span 80 (Sinopharm Chemical Reagent Co., China) as a surfactant.

For droplet formations, the flow rate of continuous phase was always set at 10 μL min⁻¹, and the flow rates of dispersed phases were especially illustrated for each experiment. All fluids were driven by syringe pumps (PHD2000, Harvard, USA) with glass syringes (Hamilton, Switzerland). A direct-current high voltage power supply (DW-P303-1ACCC, TianJinDongWen, China) was used for generating electric fields with various strengths. The fluids in the microfluidic device were observed under an inverted fluorescent microscope (Ti-U, Nikon, Japan) and recorded by a high speed camera (i-speed3, IX-Cameras, UK) at 2000 frames per s. Diameters of droplets and particles were measured by NanoMeasure software. The size distributions of the particles were mapped by measuring the diameters of the particle populations on the scanning electron microscope images.

3. Results and discussion

3.1 Size regulation of Janus droplets by an electric field

To test the microfluidic system integrated with the electric field, single-phase droplets were first formed using DI water, and the electrospray process was carried out. With the flow rates of continuous phase (10 μL min⁻¹) and dispersed phase (0.1 μL min⁻¹) fixed, the drop size decreased along with the elevated voltage. When the voltage reached 1600 V, a Taylor cone formed and tiny droplets were sprayed from the cone tip, as displayed in Fig. 2A (also available in movie S1†). The droplet size varied from 90 μm to 4 μm controlled by a voltage ranging from 0 to 1600 V (Fig. 3, black line). And an unstable cone was observed when the voltage exceeded 2000 V. For Janus droplets, we tested aqueous solutions by introducing water-soluble carbon ink to a hemisphere of the droplets for labelling. Ink solution and DI water were respectively fed into the arms of the Y-shaped channel at a flow rate of 0.5 μL min⁻¹. A laminar flow formed after two fluids converged. It was noteworthy that the channel region for the laminar flow should have a proper depth-to-width ratio to avoid severe diffusion. Setting the flow rate of oil phase at 10 μL min⁻¹, the biphasic laminar flow was sheared to Janus droplets at the flow-focusing cross junction (see movie S2†). These droplets still remained biphasic in that, (i) the flow rates of both phases were made equal; (ii) the viscosity of each hemisphere was nearly the same; (iii) the downstream channel was widened and any windings were avoided in order to alleviate mixing. These factors have been already proved to work well with the Janus droplet production elsewhere.^8,22 Under the stable dynamic condition, different phenomena were observed when the electric field was applied. The ink-containing Janus droplets deformed to spindle shape in the electric field (Fig. 2B), and recovered to spherical shape when flowing downstream past the grounding electrode. The drop size declined in virtue of the electric field (Fig. 3, red line). Throughout the formation–deformation–recovery process, the Janus droplets maintained obvious interface of the two phases. This is mainly due to the ultra fast regulation provided by the electric field, and the time lag in droplet shrinking process, during which mixing might occur, was avoided. In this work, shrinking of the droplets actuated by an electric field could be accomplished in less than a second according to the recorded videos, while it took tens of seconds to acquire the equivalent shrinkage by regulating the flow rates because it needed time for the fluids to reach new stable flow rates. Electrospray happened when the voltage climbed to 1600 V, and interestingly, tiny droplets were sprayed from the tip of a spindle-shaped mother droplet other than from the front surface of the dispersed phase (magnified image in Fig. 2B or see movie S3†). Compared with the formation of single-phase water droplets, slightly higher flow rates of the biphase were employed to prevent cross mixing of the two phases, and thus the spraying corn was stretched into a mother droplet under the electrospray mode. Spray droplets with a diameter down to 3 μm were successfully produced. And both mother drops and spray drops are in a narrow range of size distribution. The key parameters affecting droplet formation and size regulation, such as flow rate ratio and surfactant concentration, were carefully investigated. The effective regulations of droplet size by an electric field only took place when the ratio of flow rates of dispersed phase to continuous phase (V_d/V_c) was kept in a proper range, less than or equal to 1/20 in this work. And the amount of surfactant in oil phase had better be no less than 5 wt% for helping to reduce the interfacial tension. The effects of flow rate ratio and surfactant concentration on Janus droplets are similar to those on single phase droplets.^20,21

Fig. 2 Dependence of droplet diameter on voltage: DI water droplets (A) and ink labelled Janus droplets (B). Inserts are magnified images of spray droplets. Scale bars are 200 μm for main images and 50 μm for the magnified images.

Fig. 3 Effects of voltages on diameters of various droplets. Dashed lines represent the diameters of mother droplets in electrospray.

3.2 Mechanism discussion

Different points of view have been proposed about the mechanism for drop formation under an electric field.^19,23 For electrospray driven by a direct-current electric field, it was believed that the formation of a Taylor cone was necessary to generate very fine droplets.²⁰ In another study on drop size regulation by altering-current voltage, the dependence of droplet size on the applied voltage was explained by the introduction of an effective capillary number that took into account the Maxwell stress.²³ Herein, we attribute the variation of droplet size with the high voltage to the fact that the electrostatic force surmounts the interfacial tension, which can comprehensively explain the droplet diminishing process. The process is divided into two stages: droplet size decreasing, and tiny droplet producing through electrospray from a Taylor cone. During the first stage, the other factors that affect the droplet size in a shear-force-dominated microfluidic system should be also taken into account. Thorsen et al.²⁴ first reported droplet formation based on the shear forces in a T-junction microfluidic device, and the size of droplets could be predicted by equating the Laplace pressure with the shear force:²⁵

r ∼ σ/ηε (1)

where r is the droplet radius, σ is the interfacial tension, η is the viscosity of the continuous phase, and ε is the shear rate which is estimated as

ε ∼ 2v/y₀ (2)

where v is the linear velocity of the continuous phase at the junction, and y₀ is the channel radius. Thus r can be deduced from eqn (1) and (2):

r ∼ σy₀/2ηv (3)

In our experiments, an electric field was applied to the dispersed phase, and its front surface carried like charges under the electric field. Thus the front surface was subjected to the attracting force from the downstream electrode as well as electrostatic repulsive force, both of which were exactly in the opposite direction against the interfacial tension, as shown in Fig. 4. As a result, the interfacial tension was partially offset by the attracting force and electrostatic repulsion. Here we define the effective interfacial tension as Δσ, and eqn (3) can be evolved as:

r ∼ Δσy₀/2ηv (4)

Fig. 4 Schematic diagram of Janus droplet formation under an electric field. The magnified image shows the forces on the front surface of dispersed phase.

During the voltage climbing process, the electrostatic repulsion and the attracting force were increased, leading to the decrease of Δσ. When the channel radius, viscosity and velocity of the continuous phase are constant, droplet radius r will decrease with the increase of the voltage according to eqn (4). When the enhancing electrostatic repulsion and downstream attracting force exceeded the interfacial tension, electrospray of tiny droplets occurred. The mechanism of electrospray was believed to be associated with the Taylor cone formation.²⁰ The water–oil interface was charged under an electric field and acted as a capacitor. As the voltage was high enough, the interfacial tension was counteracted and the tip of the interface was stretched to a cone shape, which was called “Taylor cone”. Simultaneously, a thin thread was stretched from the Taylor cone, and the thread broke up into tiny-size individual droplets owing to the Rayleigh instability.²⁶ This stage is critical to reduce the droplet size by orders of magnitude.

3.3 Minimized polymer droplets generation and submicron Janus particle synthesis

For particle synthesis based on droplet templates, the droplet size regulation scheme must be versatile for a wide range of solutions to produce different particles. The conductivities of solutions which can be manipulated under an electric field must be within a definite range. It has been already reported that effective atomization could only be achieved if the conductivities of the liquids were within the limits of 10⁻⁶ to 10⁻⁸ Ω⁻¹ m⁻¹.²⁷ In our study, aqueous solutions and organic solutions of polymers were tested respectively. Alginate was first tested for it is hydrophilic and is frequently used to produce microspherical bio-carriers. Water-soluble ink was doped into an alginate aqueous solution, and Janus alginate droplets, one hemisphere doped with ink and the other undoped, were formed in the microfluidic chip. With the dispersed phase flow rate of 0.3 μL min⁻¹, the alginate droplets behaved similarly under a high voltage as the ink-containing Janus droplets discussed in 3.1. Fig. 5A exhibits the droplet diminishing process and a magnified image of the spray droplets. A biphasic Taylor corn and a thin thread are visible in the magnified image, presenting typical characterization of the electrospray. The droplet size dropped from 119 μm to less than 3 μm with the voltage increasing from 0 to 1600 V (Fig. 3, blue line), in spite of the much higher viscosity of the alginate solution (332 mPa s, 25 °C).

Fig. 5 Dependence of droplet diameter on voltage: (A) alginate Janus droplets and (B) PLGA Janus droplets. Magnified images are spray (upper right) and minimized droplets (lower right). Scale bars are 200 μm for main images and 50 μm for the magnified images.

PLGA was then used for producing Janus particles in the electric field. PLGA is the drug excipient permitted by U.S. FDA, which is popular in the investigations on drug-loading microspheres. In our work, DMSO was used as the solvent, for PLGA is insoluble in water. We used an organic dye, rhodamine B, to tag one of the PLGA solutions, and set the flow rate of both PLGA solutions at 0.2 μL min⁻¹ (the viscosity of the PLGA solution, 2.19 mPa s, 25 °C). As presented in Fig. 5B, Janus droplets formed with clear interfaces at different voltages, and their size decreased when the voltage was turned up (Fig. 3, green line). However, electrospray did not occur, even with the voltage as high as 2000 V, where a jet formed rather than spray droplets. It might be due to the low conductivity of the PLGA solution. According to our test, the conductivity of PLGA solution is even lower than a tenth of the conductivity of DI water, and this might lead to different behavior under an electric field. Despite the absence of electrospray, the PLGA droplet diameter was reduced to half of its original diameter under electrical regulation. Finally, we synthesized Janus particles using the minimized biphasic droplets of PLGA. Janus particles produced in microfluidic systems suffer from the relative large size, and here we combined minimized droplets with a nanoprecipitation method to generate submicron Janus particles. The nanoprecipitation method refers to the formation of polymer nanoparticles by means of solvent displacement based on solubility difference. In the droplet-template method, one particle is formed from one droplet. The microfluidic systems tend to produce droplets with diameters of tens to hundreds of micrometers, from which large particles are likely to be generated. Therefore, minimized droplets can help reducing the particle size. To collect the droplets for nanoprecipitation, the outlet of the microfluidic chip was inserted into the wall of a slightly modified centrifuge tube. DI water with 0.1 wt% polyvinyl alcohol was filled into the tube serving as the non-solvent, and the droplets outflowing from the chip immediately entered the non-solvent. Thus the biphasic morphology of the droplets could be maintained in the tube. Droplets then separated from oil spontaneously due to the density difference, and nanoprecipitation occurred as the water-soluble DMSO diffused to the non-solvent, giving rise to the formation of submicron Janus particles. Here we prepared two groups of PLGA Janus particles using droplets produced at the same flow rate without an electric field and with an electric field (the voltage was set at 1600 V) respectively. The SEM images of the particles are shown in Fig. 6A and B, and the average diameters of particles prepared without an electric field and with an electric field were 600 nm and 145 nm respectively. This result indicated that the method using an electric field was capable of generating submicron particles, and the particle size could be dramatically decreased by taking the diminished droplets as templates. To confirm the biphasic feature of the particles, two fluorescent dyes, rhodamine B and FITC, with different emission wavelengths were doped into the two hemispheres of the Janus droplets respectively. We intentionally used large droplets to synthesize the fluorescent particles sized approximate 500 nm to ease the visibility of the particles under a confocal laser scanning microscope. In Fig. 6C, red and green fluorescent is observed respectively from different hemispheres of a particle, confirming that submicron Janus particles were successfully synthesized. Fig. 6C and D depict the diameter distribution of the particles prepared without an electric field and with an electric field, respectively. Of the particles produced without an electric field, 80% are in the diameter range of 550–650 nm, and of the particles produced with an electric field, over 90% are approximately 150 nm, indicating the uniformity of the prepared particles. And a high production rate up to 4600 particles per minute could be achieved. The applications of the submicron Janus particles in drug and gene delivery, cell and tissue imaging and molecule sensing would have great potential. In these research fields, Janus particles with submicron dimension would be more favorable compared with micron particles. They may enter cells and bring about improved performances when used for therapy agent delivery and cell imaging. And they may be preferred in sensing of molecules because faster diffusion of smaller particles would shorten response time.

Fig. 6 SEM images of PLGA particles prepared without an electric field (A) and with an electric field (B), and confocal laser scanning microscope images of the particles to show asymmetry (C). The insert in (C) is a magnified image of a Janus particle. Size distribution of the particles prepared without an electric field (D) and with an electric field (E). Scale bars are 5 μm in (A), 500 nm in (B) and 3 μm in (C).

4. Conclusions

We developed an electrospray microfluidic method for preparing uniform Janus droplets, and the size of the Janus droplets could be rapidly regulated by an electrical field. The response time of droplet regulation was dramatically shortened by using an electric field compared with by altering the flow rates. The dimension of droplets was scalable from more than 100 μm to less than 5 μm. And Janus droplets and particles were produced from miscible fluids, which could hardly be realized by conventional methods. The electric field regulation was proved to be also applicable to the droplets of polymer solutions, including aqueous alginate solutions and organic PLGA solutions. Janus particles with submicron dimension were successfully synthesized by employing the minimized PLGA droplets as templates. The mechanism of the dependency of droplet size on the voltage was expounded, and we attributed the dependency to the counteracting against the interfacial tension from the electrostatic force. The present method is feasible for generation of Janus droplets from miscible liquids and regulation of droplet size, which provides a new path for tiny microparticle synthesis using droplet-template methods.

Acknowledgements

Financial support from the Natural Science Foundation of China (21375012 and 21305010), and the Fundamental Research Funds for the Central Universities (N110805001 and N110705002) is gratefully acknowledged.

References

1. J. T. Wang, J. Wang and J. J. Han, Small, 2011, 7, 1728.
2. S. Lone and I. W. Cheong, RSC Adv., 2014, 4, 13322.
3. J. Hu, S. Zhou, Y. Sun, X. Fang and L. Wu, Chem. Soc. Rev., 2012, 41, 4356.
4. A. Walther and A. H. E. Müller, Chem. Rev., 2013, 113, 5194.
5. L. Ge, S. Lu, J. Han and R. Guo, Chem. Commun., 2015, 51, 7432.
6. H. Hasinovic and S. E. Friberg, Langmuir, 2011, 27, 6584.
7. T. Nisisako, T. Torii and T. Higuchi, Chem.–Eng. J., 2004, 101, 23.
8. L. B. Zhao, L. Pan, K. Zhang, S. S. Guo, W. Liu, Y. Wang, Y. Chen, X. Z. Zhao and H. L. W. Chan, Lab Chip, 2009, 9, 2981.
9. S. H. Kim, S. J. Jeon, W. C. Jeong, H. S. Park and S. M. Yang, Adv. Mater., 2008, 4129.
10. R. K. Shah, J. W. Kim and D. A. Weitz, Adv. Mater., 2009, 21, 1949.
11. R. F. Shepherd, J. C. Conrad, S. K. Rhodes, D. R. Link, M. Marquez, D. A. Weitz and J. A. Lewis, Langmuir, 2006, 22, 8618.
12. S. Lone, S. H. Kim, S. W. Nam, S. Park, J. Joo and I. W. Cheong, Chem. Commun., 2011, 47, 2634.
13. S. H. Kim, A. Abbaspourrad and D. A. Weitz, J. Am. Chem. Soc., 2011, 133, 5516.
14. Z. Nie, W. Li, M. Seo, S. Xu and E. Kumacheva, J. Am. Chem. Soc., 2006, 128, 9408.
15. K. Maeda, H. Onoe, M. Takinoue and S. Takeuchi, Adv. Mater., 2012, 24, 1340.
16. F. Malloggi, N. Pannacci, R. l. Attia, F. Monti, P. Mary, H. Willaime, P. Tabeling, B. Cabane and P. Poncet, Langmuir, 2010, 26, 2369.
17. S. L. Anna and H. C. Mayer, Phys. Fluids, 2006, 18, 121512.
18. M. Seo and N. Matsuura, Small, 2012, 8, 2704.
19. Q. Zhang, X. Liu, D. Liu and H. Gai, Lab Chip, 2014, 14, 1395.
20. H. Kim, D. Luo, D. Link, D. A. Weitz, M. Marquez and Z. Cheng, Appl. Phys. Lett., 2007, 91, 133106.
21. C. H. Yeh, M. H. Lee and Y. C. Lin, Microfluid. Nanofluid., 2011, 12, 475.
22. T. Nisisako, T. Torii, T. Takahashi and Y. Takizawa, Adv. Mater., 2006, 18, 1152.
23. S. H. Tan, B. Semin and J. C. Baret, Lab Chip, 2014, 14, 1099.
24. T. Thorsen, R. W. Roberts, F. H. Arnold and S. R. Quake, Phys. Rev. Lett., 2001, 86, 4163.
25. G. I. Taylor, Proc. R. Soc. London, Ser. A, 1934, 146, 501.
26. L. J. W. S. Rayleigh, Proc. R. Soc. London, 1883, 14, 170.
27. I. Hayati, A. I. Bailey and T. F. Tadros, Nature, 1986, 319, 41.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed description of microfluidic chip fabrication process, data of diameters and coefficients of variation of various droplets, post-processing and characterization parameters of Janus particles, movies of electrospray processes of pure water droplets and Janus droplets. See DOI: 10.1039/c5ra24531a

---