High-Throughput, Off-Chip Microdroplet Generator Enabled by a

Subscriber access provided by UNIV OF TEXAS DALLAS

Article

A High-throughput Off-chip Microdroplet Generator Enabled by a Spinning Conical Frustum Shi-Yang Tang, Kun Wang, Kai Fan, Zilong Feng, Yuxin Zhang, Qianbin Zhao, Guolin Yun, Dan Yuan, Lianmei Jiang, Ming Li, and Weihua Li Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.9b00093 • Publication Date (Web): 12 Feb 2019 Downloaded from http://pubs.acs.org on February 13, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

A High-throughput Off-chip Microdroplet Generator Enabled by a Spinning Conical Frustum Shi-Yang Tang*,†, Kun Wang‡,∥, Kai Fan‡,∥, Zilong Feng‡,∥, Yuxin Zhang†, Qianbin Zhao†, Guolin Yun†, Dan Yuan†, Lianmei Jiang§, Ming Li⊥, and Weihua Li*,†

∥These

authors contributed equally to this work

†School

of Mechanical, Materials, Mechatronic and Biomedical Engineering, University of Wollongong,

Wollongong, NSW 2522, Australia E-mail: [email protected]; [email protected]

‡Department

of Precision Machinery and Precision Instrumentation, University of Science and

Technology of China, Hefei, 230026, China

§ARC

Centre of Excellence for Nanoscale BioPhotonics (CNBP), Department of Physics and Astronomy,

Macquarie University, Sydney, NSW 2109, Australia

⊥School

of Engineering, Macquarie University, Sydney, NSW 2109, Australia

1 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

ABSTRACT Despite that droplet-based microfluidics has been broadly used as a versatile tool in biology, chemistry and nanotechnology, its rather complicated microfabrication process and the requirement of specialized hardware and operating skills hinder researchers to fully unleash the potential of this powerful platform. Here, we develop an integrated microdroplet generator enabled by a spinning conical frustum for the versatile production of near-monodisperse microdroplets in a high-throughput and off-chip manner. The construction and operation of this generator are simple and straightforward without the need of microfabrication, and we demonstrate that the generator is able to passively and actively control the size of the produced microdroplets. In addition to water microdroplets, this generator can produce microdroplets of liquid metal that would be difficult to produce in conventional microfluidic platforms as liquid metal has a large surface tension. Moreover, we demonstrate that this generator can produce solid hydrogel microparticles and fibers using integrated ultraviolet (UV) light. In the end, we further explore the ability of this generator for forming double emulsions by co-flowing two immiscible liquids. Given the remarkable abilities demonstrated by this platform and the tremendous potential of microdroplets, this user-friendly method may revolutionize the future of droplet-based chemical synthesis and biological analysis.

KEYWORDS: Microdroplet; emulsion; microfluidics; off-chip; liquid metal;


Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Droplet-based microfluidics has been extensively developed as a versatile tool for widespread applications in biology, chemistry and nanotechnology.1-6 In conventional microfluidic droplet generators, various geometries adopting cross-flow, co-flow, flow-focusing, step emulsification, and microchannel emulsification schemes have been demonstrated for massive production of microdroplets.3 Precise and effective control of droplet generation can use either passive or active method, where the passive approaches generate droplets without external actuation, and the active approaches incorporate additional electrical, magnetic, acoustic, pneumatic, optical, thermal, or piezoelectric forces into microfluidic systems.1,3 Although microfluidic approaches are relatively straightforward for microdroplet production, rather complicated chip microfabrication process and the requirement for specialized microfluidic facilities and skills could hinder inexperienced researchers to employ such a powerful method. In recent years, numerous off-chip approaches have been explored for producing microdroplets. For instance, Chen et al. developed a spinning micro-pipette liquid emulsion generator.7 This generator spins a glass capillary (inner diameter at the tip is 10 µm) that is attached to an eccentric wheel within oil using a servo motor. Uniform microdroplets can be produced by pumping solution to the capillary while spinning it in oil. Although this method is relatively simple and low-cost, fabricating glass capillaries with such a small inner diameter could be difficult, and the droplet production rate is relatively low since the dispensing rate of water phase is controlled to be less than 2 µL/min. The same group later developed a centrifuge-based droplet generator,8 in which the aqueous phase was forced to pass through a microchannel array using centrifugal force and then pinched off at the nozzles to form microdroplets in air; the formed droplets subsequently flew into the receiving oil to form emulsions. This method allows for the rapid production of microdroplets with uniform size, however, the manufacture of the micro-channel array is rather complicated as it involves multiple microfabrication and surface treatment steps. Similarly, 3 ACS Paragon Plus Environment


Page 4 of 23

microdroplets can also be produced using the membrane emulsification method.9,10 Nonetheless, the variation in the size of microdroplets formed using this method is relatively large. Additionally, other offchip methods such as acoustic droplet ejection,11 electrohydrodynamic atomization,12 particle-templated emulsification,13 compressing water-containing porous elastomeric sponges in oil phase14, and ejecting liquid jet into air15,16 have been demonstrated for producing microdroplets. However, these methods either require complicated operating facilities or have very limited production rate with compromised uniformity of microdroplet size. To overcome the limitations of existing platforms, in this work we developed an integrated off-chip platform enabled by a spinning conical frustum for the versatile and high-throughput production of uniformly dispersed microdroplets. We characterized the capabilities of this droplet generator for controlling the size of the produced droplets using both passive (i.e. operating parameters of the platform) and active (i.e. electric field) methods. In addition to water microdroplets, we further investigated the capability of the platform for producing microdroplets of liquids with high surface tension such as eutectic gallium indium (EGaIn) liquid metal. Moreover, we studied the production of solid hydrogel microparticles and fibers using this platform by integrating ultraviolet (UV) light. Finally, we examined the ability of this generator for forming EGaIn-in-hydrogel-in-oil double emulsions by co-flowing two immiscible liquids.


Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL SECTION Chemicals, Instruments, and Fabrication: EGaIn liquid metal, poly(ethylene glycol diacrylate) (PEGDA, Mn = 700), and 2-Hydroxy-2-methylpropiophenone were purchased from Sigma Aldrich, Australia. pulse width modulation (PWM) motor speed controller (12 V, 8 A), DC motor (12 V, 2.1 kg/cm at 70 RPM), Digital Tachometer, and DC-DC high voltage boost converter were purchased from Jaycar Electronics, Australia. The 34G blunt needle, the 21G blunt needle tubing, ball bearings, and the tungsten steel shaft were purchased online. Poly(methyl methacrylate) (PMMA) frames, gear sets, and needle holder were fabricated using a CO2 laser engraving system (Versa Laser System, Model VLS3.50, Universal Laser System Ltd.). Stainless frustums were fabricated using a computerized lathe machine (CNC Lathe, HAAS Automation Ltd., USA). A UV LED (365 nm ± 5 nm, Model-LS4, Lichtzen Ltd., South Korea) was used to solidify the PEGDA droplets. Syringe pumps (Legato 100, KD Scientific, Holliston, MA, USA) were used to inject the solutions into the needles. Numerical simulations were conducted using the COMSOL 5.2 software package (Burlington, MA, USA). A slow motion camera (FPS1000, The Slow Motion Camera Company Ltd., UK) was used to capture the high-speed videos.



Page 6 of 23

RESULTS AND DISCUSSION Figure 1A illustrates the microdroplet generator. We used a direct current (DC) motor to drive the tungsten steel shaft (diameter and length of 5 and 100 mm, respectively) via a 2:1 gear set, while a stainless steel conical frustum (see Figure S1 of the Supporting Information S1 for details) was fixed at the end of the shaft. The speed of the motor was controlled using a PWM motor controller. A pair of ball bearings was fixed on the shaft to reduce friction and guarantee that the rotation is exactly perpendicular to the shaft axis. We used a luer to luer syringe needle adapter to attach a 34G blunt needle (inner and outer diameters of 60 and 250 µm, respectively; length of 50 mm), and the needle adapter was fixed on the needle holder. The guide rail attached to the needle holder can adjust the distance between the 34G needle and the frustum. Each of the parts was assembled on the PMMA frame produced using a laser cutter. The inset of Figure 1A shows the assembled platform. After construction, the conical frustum and the needle were submerged in to a lab beaker filled with liquid for microdroplet production (Figure 1B). A side-view scheme illustrating the process of microdroplet production is shown in Figure 1C. Upon rotating the frustum at a rotational speed of Ω (RPM), we expect that the induced flow of oil phase on the surface of the frustum can break the water phase applied through the needle into microdroplets. The Reynolds number Re of the oil phase can be defined as: Re = πρcpΩrDO/15ηcp, where ρcp and ηcp are the density and viscosity of the continuous oil phase, respectively; r is the distance from the needle tip to the axis of rotation of the frustum; and DO is the outer diameter of the needle. Assuming a no-slip condition exists at the solid-liquid boundary, we estimated that Re at the tip of the needle is less than 5 even at the maximum possible Ω of 250 RPM. Therefore, the flow of oil phase is laminar at the surface of the frustum and it is possible to produce monodispersed microdroplets using this method similar to the case of utilizing a microfluidic chip with a T-junction. We examined this by applying deionized (DI) water (mixed with blue dye, flow rate Q of 50 6 ACS Paragon Plus Environment

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


µL/min) to the needle and rotating the frustum within the low-viscosity mineral oil (30 mPa∙s) which contains 2 wt% Span-80 and 0.5 wt% Brij L4 surfactants. Figure 1D illustrates the front view (y-z plane) of the experimental results (also see Movie S1); we can see the formation of large droplets (diameter of ~1 mm) before rotating the frustum due to the expansion of the water phase when exiting the needle. When rotating the frustum at the speed of 50 RPM, the platform produced much smaller droplets (diameter of ~300 µm) of similar size (Figure 1D). Interestingly, the produced droplets travelled around the frustum and gradually spiral downward without colliding with each other; the droplets eventually left the frustum and can be collected at the bottom of the beaker (see Movie S1). Such a phenomenon reveals the beauty of using conical frustum. We further studied the spiral downward travelling phenomenon of the microdroplets using numerical simulation. Figure 1E shows the cross-sectional view of the pressure distribution around a rotating frustum, in which we can see the formation of a low-pressure region at the surface of the frustum. Figure 1F shows the distribution of velocity (perpendicular to the x-z plane) for the surrounding liquid at the cross section when we set Ω at 50 RPM. The contour of shear rate around the frustum is given in Figure S2 of the Supporting Information S2. Our simulation shows the induction of a flow travelling downwards along the surface of the frustum, and the formation of a counterclockwise-circulated vortex in the x-z plane. Figure 1G plots the velocity profiles along the line AA' (shown in Figure 1E) that is in parallel with the surface of the frustum; we repeated the plot with different distances (0 to 2 mm) away from the frustum surface. We believe that the formation of such a vortex is due to the increase in flow velocity along the surface of the frustum, and this vortex could gradually push the produced droplets downward to avoid collision. The simulation predicts the existence of an upper flow at the bottom edge of the frustum (Figure 1F). Pleasantly, our experimental results show a good agreement with this prediction, as shown in Figure S3 of the Supporting Information S3. The pressure profiles along the line



Page 8 of 23

AA' and the simulation results for frustums with different θ (see Figure 1C) are presented in Figure S4 of the Supporting Information S4. The frustum with the θ of 70° was used in our experiments to facilitate the adjustment of the gap g between the needle tip and the surface of the frustum (the value of g cannot be adjusted to 0 by the guide rail in our platform when using frustums with the θ smaller than 70°).


Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Experimental setup of the user-friendly platform for high-throughput and off-chip microdroplet generation. (A) Exploded schematic representation of the microdroplet generator, the lower inset shows the assembled system. (B) Actual image of the assembled system. (C) Zoom-in schematic representation of the side view of the droplet production part. (D) Snapshots showing the formation of water microdroplets using the platform, the lower insets are the enlarged images of the needle tip and formed microdroplets. Contours of (E) the pressure distribution, and (F) the velocity profile around a rotating conical frustum obtained from the numerical simulation. (G) Plots of flow velocity profiles along the line AA' in E, the flow velocity profiles along five lines with different distances ranging from 0 to 2 mm away from the surface of the frustum are shown



Page 10 of 23

After understanding the working mechanism, we next characterized the performance of the platform for microdroplet production. Figure 2A shows snapshots taken from a high-speed camera recording the production of a water microdroplet (Ω = 150 RPM and Q = 50 µL/min, also see Movie S2). We can see that under these operating conditions, this generator produces droplets in a dripping mode. Microdroplet breakup occurs when the drag force exerted on the emerging droplet overcomes the interfacial tension that resisting deformation of the droplet.17 Neglecting the effects of the dispersed phase inertia and assuming there is no gap between the needle tip and the surface of the frustum (g = 0), we can estimate the diameter of the produced microdroplets Ddroplet as (see Supporting Information S5 for the detailed deduction of this equation):

𝐷𝑑𝑟𝑜𝑝𝑙𝑒𝑡 =

𝐷𝑛𝑒𝑒𝑑𝑙𝑒2 3𝐶𝑎(1 ― 𝛽)

(1)

where Dneedle is the inner diameter of the needle, β is the ratio between the droplet velocity vdroplet and the velocity of the continuous oil phase vcp (i.e. β = vdroplet/vcp) when droplet breakup occurs, and Ca is the capillary number, which can be defined as:

𝐶𝑎 =

𝜂𝑐𝑝𝑣𝑐𝑝 𝛾

=

𝜋𝜂𝑐𝑝𝛺𝑟 30𝛾

(2)

where r is the distance from the needle tip to the axis of rotation of the frustum (see Figure S5A of the Supporting Information S5), and γ is the interfacial tension between water and oil. Equation 1 indicates that the droplet diameter resulting from unconfined breakup in dripping mode is a function of Ca. Keeping the flow rate Q constant at 50 µL/min, we observed that increasing Ω efficiently reduced the size and polydispersity of the produced microdroplets, as shown in Figure 2B. The decrease in droplet size is due to the increase of Ca (from ~0.26 at 50 RPM to ~1.28 at 250 RPM). We found that Equation 1 can effectively predict Ddroplet at different Ω (see Figure S5B of the Supporting Information S5). The smallest 10 ACS Paragon Plus Environment

Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


diameter of the droplet was ~150 µm at 250 RPM, and we did not observe droplet production in jetting mode at any Ω (see Figure 2B insets). Although it is possible to increase the rotational speed of the frustum by using a DC motor with a higher rotational speed, or by increasing the ratio of the gear set, it should be noted that high rotational speed of the frustum could induce strong vortices that may generate strong shear to further break the produced water microdroplets. The presence of surfactants stabilized the microdroplets. Figure 3C shows the optical images of the collected microdroplets produced using different Ω. Figure 3D illustrates that at a constant Ω (100 RPM in this case), higher Q can result in the production of droplets with a larger size. However, Equation 1 can no longer be used for predicting the droplet diameter since the fluid inertia of the dispersed phase becomes non-negligible at high flow rates. The Weber number (We = ρdpvdp2Dneedle/γ, where ρdp and vdp are the density and velocity of the dispersed water phase, respectively) of the water phase increased from ~0.65 at 50 µL/min to ~2.6 when the flow rate was doubled. We observed that further increasing Q eventually made the platform operate in an unstable jetting mode, inducing Rayleigh-Plateau instability and producing polydispersed droplets, as shown in Figure S6 of the Supporting Information S6. In addition, increasing the gap g between the needle tip and the surface of the frustum can lead to the production of larger droplets, as shown in Figure 2E (Ω = 100 RPM and Q = 50 µL/min). The simulation results given in Figure 1G show the decrease in flow velocity when increasing g. This in turn reduces Ca (from ~0.55 when g = 0 to ~0.30 when g = 5 mm), leading to the generation of larger droplets. Furthermore, we examined the mass production of water microdroplets in a beaker by operating the platform for 30 min (Ω = 50 RPM and Q = 50 µL/min), and Figure 2F illustrates that a relatively large number of droplets (105 to 106) can be produced within 30 min due to the high throughput of this generator. Apart from the passive control techniques illustrated above, we demonstrated that the platform also allows for active control of the droplet production using electric field. Figure 2G shows the side-view 11 ACS Paragon Plus Environment


Page 12 of 23

scheme of the experimental setup. We adjusted the gap g to 2 mm using the guide rail; the needle and the frustum were connected to the positive and negative outputs of a high-voltage DC-DC boost converter, respectively. The simulated electric field distribution for the area between the needle and the frustum is presented in Figure 2H, from which we can see that the tip of the needle has a localized high-intensity electric field. Figure 2I presents the results of the active control over the droplet size using electric field (Ω = 100 RPM and Q = 50 µL/min). The diameter of the produced droplets drastically decreased from ~550 to ~200 µm when increasing the voltage from 0 to 1200 V. This is because when an electric field is applied, the water-oil interface is charged and behaves as a capacitor.18 The accumulated charges can reduce the water-oil interfacial tension and therefore, increase Ca for reducing the droplet size. Interestingly, we observed the formation of a Taylor cone when a large voltage of 1800 V was applied (Figure 2I). At this voltage, the increased charges on the water-oil interface can result in a higher attraction force on water downstream, and the tip of the Taylor cone was stretched to a filament to break into polydispersed droplets due to the Rayleigh instability.18


Page 13 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Production of water microdroplets. (A) Snapshots taken from a high-speed camera showing the formation of a microdroplet on the surface of the spinning frustum. (B) Diameter vs Ω plot for the produced microdroplets, the inset images show the droplet production at the Ω of 50 and 200 RPM, respectively. (C) Optical images of the produced water microdroplets, the inset is a magnified image of a few droplets with the diameter of ~150 µm. Diameter vs (D) flow rate, and (E) gap g plots for the produced microdroplets. (F) Image showing the mass production of water microdroplets within a lab beaker, the boxed image shows the enlarged image of the generated microdroplets. (G) Schematic representation of droplet size control using electric field. (H) Color contour of the electric field for the boxed area in G. (I) Images showing droplet generation without electric field vs the control of the droplet size using three different high voltages of 600, 1200 and 1800 V. Scale bars are 500 µm.



Page 14 of 23

In addition to water droplets, we further investigated the capability of this generator for producing microdroplets of liquid that are conventionally relatively difficult to produce in microfluidic platforms. This includes liquids with high surface tension, such as EGaIn liquid metal (wt75% gallium, wt25% indium). Microdroplets of liquid metal have been explored for developing various applications in microelectromechanical systems (MEMS),19 electrochemical sensors,20 constructing 3D structures,21,22 and fabricating composites.23 In comparison to water, EGaIn has ∼5.3 times higher density and doubled viscosity, but remarkably ∼7.7 times higher surface tension.24,25 In flow-focusing microfluidic devices, due to the high surface tension of liquid metal, liquid with high viscosity need to be used as the continuous stream to generate sufficient shear for pinching off liquid metal into discrete droplets.26-28 However, liquid with high viscosity induces very high pressure drop within the microfluidic channels, causing instability for droplet production and can potentially damage the microfluidic system.28 This droplet generator avoids the use of microfluidic channels and instead, we filled the lab beaker with corn syrup-water mixture (w/w ratio of 2:1, viscosity of ~110 mPa∙s) and only pump EGaIn liquid metal to the needle (flow rate QLM of 50 µL/min). We successfully produced EGaIn microdroplets with a relatively uniform size by rotating the frustum within the corn syrup-water mixture (see Movie S3), and Figure 3A shows the diameter of the obtained microdroplets with respect to Ω. Similar to the case of water, increasing Ω reduces the size of the droplets. The insets of Figure 3A illustrate the production of EGaIn microdroplets. A native oxide layer can form on the surface of EGaIn when expose to oxygen, and this layer can prevent the produced microdroplets from merging, allowing for mass production of the liquid metal microdroplets. Figure 3B shows the optical images of the collected EGaIn droplets produced using different Ω.


Page 15 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Production of liquid metal microdroplets. (A) Diameter vs Ω plot for the produced EGaIn microdroplets, the inset images show the droplet production at the Ω of 50 and 250 RPM, respectively. (B) Optical images of the produced liquid metal microdroplets using four different Ω of 50, 100, 200 and 250 RPM. The inset shows the EGaIn microdroplets collected in a glass vial. Scale bars are 500 µm.



Page 16 of 23

Droplet microfluidic platforms have been widely adopted for microparticle fabrication.1 In general, the production process involves three main steps of 1) producing monodisperse emulsified microdroplets in a microfluidic device; 2) shaping the microdroplets in microchannels; and 3) solidifying the droplets to form microparticles by physical, chemical or photochemical methods.1,29 Following similar process, we further demonstrated the ability of this generator for producing spherical hydrogel microparticles in the lowviscosity mineral oil. Figure 4A shows the scheme of the experimental setup, in which we proposed to use a beam of UV light (power density of ~2,000 mW/cm2) to solidify the microdroplets immediately after production. We prepared the PEGDA solution by mixing 1 mL of PEGDA with 9 mL DI water, and then added 100 µL of 2-Hydroxy-2-methylpropiophenone into the solution. The prepared PEGDA solution was fully mixed using a vortex generator for 5 min before pumping into the platform. We adjusted the gap g to 1 mm and set the flow rate of the PEGDA solution to 20 µL/min. We observed the successful production of PEG hydrogel microparticles using the platform and Figure 4B illustrates that the diameter of the produced particles is reduced to ~100 µm when increasing Ω to 250 RPM. At the same Ω, the size of the PEG hydrogel microparticles was smaller than that of the water microdroplets. This is probably due to the reduction in surface tension of water after adding PEGDA, and this in turn increased the Ca of the system to yield smaller particles. Figure 4C shows the optical images of the collected microparticles produced using different Ω; we didn’t see the coalescence of PEG hydrogel particles after solidification. Interestingly, we discovered that increasing the content of PEGDA to 40% (v/v) can further reduce the surface tension to form a jet instead of droplets after exiting the needle. The jet can solidify after UV exposure to form hydrogel fiber before breaking into droplets, as shown in Figure 4D (Ω = 100 RPM). The diameter of the fiber can be adjusted by varying the flow rate of the PEGDA solution (Figure 4D).


Page 17 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Production of PEG hydrogel microparticles. (A) Schematic representation of the production of the PEG hydrogel microparticles. (B) Diameter vs Ω plot for the produced PEG hydrogel microparticles, the inset images show the PEG microdroplet production at the Ω of 50 RPM. (C) Optical images of the produced PEG hydrogel microparticles using four different Ω of 50, 100, 200 and 250 RPM. (D) Optical images of the produced PEG hydrogel fibers using four different flow rates of 250, 200, 100 and 50 µL/min. Scale bars are 500 µm.

Finally, we demonstrated the capability of this generator for producing core-shell microparticles by forming double emulsions using the co-flow method. Figure 5A depicts the scheme of the experimental setup; we inserted a 34G needle into a 21G blunt needle tubing (inner and outer diameters of 500 and 800 µm, respectively; length of 20 mm), and applied PEGDA solution (10% v/v) and EGaIn liquid metal to the 21G tubing and the 34G needle, respectively, to form the co-flow. Figure 5B shows the snapshots taken from a high-speed camera recording the production of an EGaIn-core/PEG-shell droplet (Ω = 50 RPM, QPEG = 50 µL/min, QLM = 20 µL/min, also see Movie S4). We can see that an EGaIn droplet is 17 ACS Paragon Plus Environment


Page 18 of 23

produced from the inner needle and subsequently encapsulated by the PEG solution. The PEG shell solidified after UV exposure and formed core-shell structured particles, as shown in Figure 5C. Furthermore, particles with multicores can be produced by carefully tuning the flow rate of liquid metal QLM, and Figure 5D shows the number of EGaIn cores increase from 1 to 5 by reducing QLM from 20 to 5 µL/min. Using this similar method, the production of hollow PEG microparticles can be achieved by coflowing PEGDA solution and water, as shown in Figure S7 of the Supporting Information S7.

Figure 5. Production of core-shell microparticles. (A) Schematic representation of the needles adapting co-flow setup for producing core-shell structured microparticles. (B) Snapshots taken from a high-speed camera showing the formation of a single EGaIn-core/hydrogel-shell microparticle. (C) Optical images of a cluster of the EGaIn-core/hydrogel-shell microparticles (QPEG = 50 µL/min and QLM = 20 µL/min). D) Optical images of the multi-EGaIn-core/hydrogel-shell microparticles. Scale bars are 500 µm.


Page 19 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CONCLUSIONS In summary, instead of using conventional microfluidic channels, we developed an innovative userfriendly microdroplet generator for the versatile and mass production of near-monodisperse microdroplets of various liquids in both high-throughput and off-chip manners. The operation of the platform is simple and straightforward as it only requires three major steps: 1) placing the frustum in the suspending medium; 2) pumping liquids into the needle; and 3) activating the motor for droplet production. We also showed that the platform is capable of controlling the size of the produced droplets using both passive and active methods. We expect that using needles or glass capillaries with a smaller inner diameter can further reduce the size of the produced microdroplets. We demonstrated that in addition to water microdroplets, this generator can also produce microdroplets of EGaIn liquid metal that would be difficult to produce in conventional microfluidic platforms. Moreover, this platform can produce solid hydrogel microparticles and fibers by solidifying PEGDA solution using integrated UV light. Finally, we demonstrated the ability of the platform for forming double emulsions by co-flowing two immiscible liquids. Given the distinctive abilities of this droplet generator and the tremendous applications based on microdroplets, this technique possesses the potential to revolutionize the future of microdroplet-based chemical, optical and biological applications.



Page 20 of 23

ASSOCIATED CONTENT This Supporting Information is available free of charge via the Internet at http://pubs.acs.org. Detailed design of the conical frustum; contour of the shear rate; upper flow at the bottom edge of the conical frustum; Simulation results of the conical frustums with different θ; a simple model for predicting the droplet size; operating the droplet generator in a jetting mode; production of hollow PEG particles.

AUTHOR INFORMATION Corresponding Author *S.-Y. Tang: [email protected] *W. Li: [email protected]

ACKNOWLEDGMENTS Dr. Shi-Yang Tang is the recipient of the Vice-Chancellor’s Postdoctoral Research Fellowship funded by the University of Wollongong. Prof. Weihua Li acknowledges the support of an ARC Discovery Grant (DP180100055).

REFERENCES (1)

Shang, L.; Cheng, Y.; Zhao, Y. Chem. Rev. 2017, 117, (12), 7964-8040.

(2)

Mashaghi, S.; Abbaspourrad, A.; Weitz, D. A.; van Oijen, A. M. Trends Anal. Chem. 2016, 82,

118-125. (3)

Zhu, P.; Wang, L. Lab Chip 2017, 17, (1), 34-75. 20 ACS Paragon Plus Environment

Page 21 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(4)

Kaminski, T. S.; Scheler, O.; Garstecki, P. Lab Chip 2016, 16, (12), 2168-2187.

(5)

Li, M.; van Zee, M.; Riche, C. T.; Tofig, B.; Gallaher, S. D.; Merchant, S. S.; Damoiseaux, R.;

Goda, K.; Di Carlo, D. Small 2018, 14, (44), 1803315. (6)

Cole, R. H.; Tang, S.-Y.; Siltanen, C. A.; Shahi, P.; Zhang, J. Q.; Poust, S.; Gartner, Z. J.; Abate,

A. R. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, (33), 8728-8733. (7)

Chen, Z.; Fu, Y.; Zhang, F.; Liu, L.; Zhang, N.; Zhou, D.; Yang, J.; Pang, Y.; Huang, Y. Lab Chip

2016, 16, (23), 4512-4516. (8)

Chen, Z.; Liao, P.; Zhang, F.; Jiang, M.; Zhu, Y.; Huang, Y. Lab Chip 2017, 17, (2), 235-240.

(9)

Dragosavac, M. M.; Vladisavljevic, G. T.; Holdich, R. G.; Stillwell, M. T. Langmuir 2011, 28, (1),

134-143. (10)

Vladisavljević, G.; Kobayashi, I.; Nakajima, M. Microfluid. Nanofluidics. 2012, 13, (1), 151-178.

(11)

Hadimioglu, B.; Stearns, R.; Ellson, R. J. Lab. Autom. 2016, 21, (1), 4-18.

(12)

Agostinho, L.; Yurteri, C.; Fuchs, E.; Marijnissen, J. Appl. Phys. Lett. 2012, 100, (24), 244105.

(13)

Hatori, M. N.; Kim, S. C.; Abate, A. R. Anal. Chem. 2018, 90, (16), 9813-9820.

(14)

Thurgood, P.; Baratchi, S.; Szydzik, C.; Zhu, J. Y.; Nahavandi, S.; Mitchell, A.; Khoshmanesh, K.

Sensor. Actuat. B: Chem. 2018, 274, 645-653. (15)

Visser, C. W.; Kamperman, T.; Karbaat, L. P.; Lohse, D.; Karperien, M. Sci. Adv. 2018, 4, (1),

eaao1175. (16)

Frommhold, P. E.; Lippert, A.; Holsteyns, F. L.; Mettin, R. Exp. fluids 2014, 55, (4), 1716.

(17)

Husny, J.; Cooper-White, J. J. J. Non-newton. fluid 2006, 137, (1-3), 121-136.

(18)

Kim, H.; Luo, D.; Link, D.; Weitz, D. A.; Marquez, M.; Cheng, Z. Appl. Phys. Lett. 2007, 91, (13),

133106.



(19)

Page 22 of 23

Tang, S.-Y.; Sivan, V.; Petersen, P.; Zhang, W.; Morrison, P. D.; Kalantar‐zadeh, K.; Mitchell, A.;

Khoshmanesh, K. Adv. Funct. Mater. 2014, 24, (37), 5851-5858. (20)

Tang, S.-Y.; Ayan, B.; Nama, N.; Bian, Y.; Lata, J. P.; Guo, X.; Huang, T. J. Small 2016, 12, (28),

3861-3869. (21)

Tang, S.-Y.; Zhu, J.; Sivan, V.; Gol, B.; Soffe, R.; Zhang, W.; Mitchell, A.; Khoshmanesh, K. Adv.

Funct. Mater. 2015, 25, (28), 4445-4452. (22)

Ladd, C.; So, J. H.; Muth, J.; Dickey, M. D. Adv. Mater. 2013, 25, (36), 5081-5085.

(23)

Markvicka, E. J.; Bartlett, M. D.; Huang, X.; Majidi, C. Nat. Mater. 2018, 17, 618-624.

(24)

Khoshmanesh, K.; Tang, S.-Y.; Zhu, J. Y.; Schaefer, S.; Mitchell, A.; Kalantar-Zadeh, K.; Dickey,

M. D. Lab Chip 2017, 17, (6), 974-993. (25)

Dickey, M. D. ACS Appl. Mater. Interfaces 2014, 6, (21), 18369-18379.

(26)

Hutter, T.; Bauer, W. A. C.; Elliott, S. R.; Huck, W. T. Adv. Funct. Mater. 2012, 22, (12), 2624-

2631. (27)

Thelen, J.; Dickey, M. D.; Ward, T. Lab Chip 2012, 12, (20), 3961-3967.

(28)

Tang, S.-Y.; Joshipura, I. D.; Lin, Y.; Kalantar‐Zadeh, K.; Mitchell, A.; Khoshmanesh, K.; Dickey,

M. D. Adv. Mater. 2016, 28, (4), 604-609. (29)

Dendukuri, D.; Doyle, P. S. Adv. Mater. 2009, 21, (41), 4071-4086.


Page 23 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For Table of Contents Only


High-Throughput, Off-Chip Microdroplet Generator Enabled by a

Recommend Documents