Magnetic actuation and deformation of a soft shuttle

Abstract

Here, we describe the magnetic actuation of soft shuttles for open-top microfluidic applications. The system is comprised of two immiscible liquids, including glycerol as the soft shuttle and a suspension of iron powder in sucrose solution as the magnetic drop. Permanent magnets assembled on 3D printed motorized actuators were used for the actuation of the magnetic drop, enabling the glycerol shuttle to be propelled along customized linear, circular, and sinusoidal paths. The dynamics of the hybrid shuttle–magnetic drop system was governed by the magnetic force, the friction at the interface of the shuttle and the substrate, and the surface tension at the interface of the shuttle and the magnetic drop. Increasing the magnetic force leads to the localized deformation of the shuttle and eventually the full extraction of the magnetic drop. The versatility of the system was demonstrated through the propelling of the shuttle across a rough surface patterned with microfabricated barriers as well as taking advantage of the optical properties of the shuttle for the magnification and translation of microscale characters patterned on a planar surface. The integration of the system with current electrowetting actuation mechanisms enables the highly controlled motion of the magnetic drop on the surface of a moving shuttle. The simplicity, versatility, and controllability of the system provide opportunities for various fluid manipulation, sample preparation, and analysis for a range of chemical, biochemical, and biological applications.

INTRODUCTION

Open-top microfluidic systems offer several advantages compared to their widely used channel-based counterparts. These include simple fabrication procedures that do not rely on microfabrication facilities; simple operational procedures due to the elimination of pumps, tubes, interfaces, leakage, bubble formation, and clogging; and direct sample manipulation due to the elimination of channel walls.^1–3 Notably, open-top microfluidic systems benefit from droplet-based assays. The droplets can be considered soft, small chambers capable of storage, handling, mixing, and reactions of liquid samples, which can be easily placed onto or removed from the substrate.^4–8

Electrowetting on dielectric (EWOD) and magnetic mechanisms have been widely used for the manipulation of droplets in open-top microfluidic platforms. EWOD platforms, utilizing an array of electrodes, enable changing the surface tension and, in turn, the contact angle of droplets by applying an electric field.^9,10 This enables the movement of the droplet from one electrode to another. The integration of computer-controlled, custom-built electronics facilitates transport, mixing, splitting, and dispensing of droplets,^11,12 which benefit multi-step chemical, biochemical, and biological assays.^13–16 Despite its versatility, the motion of droplets is dictated by the configuration of electrodes.^17,18 The droplets should be larger than the size of the electrodes to ensure they can be moved. High voltage power sources are also required for energizing electrodes.

Digital magnetic platforms, on the other hand, take advantage of magnetic fields produced by permanent or electromagnets to actuate droplets. The magnetic force exerted on droplets is proportional to the magnetic susceptibility of the droplet.¹⁹ This suggests that liquids with a high magnetic susceptibility such as ferrofluids can be easily driven using magnetic platforms,²⁰ whereas water-based solutions cannot be actuated due to their low magnetic susceptibility. Several strategies have been implemented to address this limitation. This includes loading of droplets with small (nano to micro) magnetic beads.^21–23 Large magnetic beads (∼1 mm in diameter) can also be immersed in the droplet.²⁴ The droplets can also be coated with a thin layer of magnetic particles, which are referred to as magnetic liquid marbles.²⁵ The droplets are actuated in either dry (air) or wet (oil) liquids.²³

Biocompatible magnetic beads commonly used in immunoassays^26–29 facilitate driving, merging, and splitting of droplets and importantly serve as functional solid substrates,²¹ which enable the extraction, mixing, concentration, and detection of target bio-samples within the droplet.^30–33 This puts the digital magnetic platforms in a prime position for conducting multi-step biological assays, including isolation of target cells,³⁴ detection of antibodies,³⁵ extraction and lysis of target cells followed by extraction, amplification, and purification of nucleic acids,^23,32,36 and targeted drug delivery,³⁷ to name a few.

Digital magnetic platforms provide more flexibility for droplet manipulation compared to their EWOD counterparts. The magnetic platforms do not necessarily rely on prefabricated electrodes, enabling simple and inexpensive open-top microfluidic systems. The motion of droplets is not governed by the configuration or size of the electrodes and can be determined by the user.

In this work, we demonstrate the magnetic manipulation of glycerol droplets (shuttle) loaded with a small droplet containing iron powder in a sucrose solution (magnetic drop). The magnetic drop floats on the surface of the shuttle without being diffused or sunk and, therefore, can be magnetically actuated easily and later extracted from the shuttle without being dissipated. The incorporation of 3D printed gears facilitates the automated motion of the shuttle along customized linear, circular, and sinusoidal paths. We explore the utility of this system for the localized deformation of the shuttle surface. We investigate the propelling of the shuttle on rough surfaces incorporating an array of microfabricated barriers. We also explore the utility of the system for the magnification and translation of microscale characters patterned on the substrate. We also examine the integration of this system with an electrowetting on dielectric platform to further control the motion of the magnetic drop on a moving shuttle.

MATERIAL AND METHODS

Magnetic drop

A supersaturated sucrose solution was prepared by adding 350 g of sucrose to 100 ml of water at 80 °C using a magnetic stirrer. The sucrose solution was then stained by adding 2 ml of red food dye for better visualization. The supersaturated sucrose solution had a significantly small solubility in pure glycerol, as investigated in detail in our previous work.³⁸ This enabled us to actuate the glycerol shuttle with the magnetic drop over extended periods (>2 h) with minimized mixing between the two liquids. Diluting of glycerol with water, which is commonly used for reducing the viscosity of glycerol,^39,40 increased the solubility of the supersaturated sucrose solution with the aqueous glycerol.^38,41 Magnetic drops were produced by adding iron powder to stained supersaturated sucrose solution at three concentrations of 1.38 g/ml (low), 1.41 g/ml (medium), and 1.43 g/ml (high).

Fluidics

A 28 μl droplet of glycerol (0.0356 ± 0.0012 g; n = 10; Sano Consumer Healthcare) with a density of 1260 kg/m³, dynamic viscosity of 1.4 Pa s (99.5% purity), and surface tension of 63 mN/m was used as the shuttle. Glycerol was chosen as the shuttle as due to its stability, non-volatility, accessibility, low cost, and biocompatibility.⁴² Importantly, glycerol has a significantly small solubility in supersaturated sucrose solution,³⁸ which was essential for conducting long-term experiments. These specifications advantaged glycerol over commonly used water-based solutions. The shuttle manipulation experiments were conducted by placing the shuttle on a parafilm substrate pre-coated with silicone oil (Elbesin Silicone Oil B 5, silikon-pro s). This oil has a density of 920 kg/m³, dynamic viscosity of 4.6 Pa s, and surface tension of 19.2 mN/m and was essential to reduce the friction between the glycerol shuttle and the substrate.

Modular magnetic platform

A linear actuator was fabricated to facilitate the axial movement of the magnetic drop or the shuttle at desired axial velocities. This actuator comprised of 3D printed rack and pinion gears, as detailed in Fig. S1 in the supplementary material. The pinion was driven by an NEMA 17 stepper motor, which was activated by an A4988 motor driver chip and controlled by an Arduino UNO microcontroller. A permanent magnet disk (N50, 1.41–1.45 T, D × L = 5 mm × 10 mm) was attached to the rack. A similar actuator was used for the vertical stretching of the shuttle. A circular actuator was fabricated to facilitate the circular motion of single or double magnetic drops over the surface of the glycerol shuttle. This actuator composed of a pair of 3D printed gears, as detailed in Fig. S1 in the supplementary material. The small gear was driven by an NEMA 17 stepper motor and was used to rotate the large gear, which accommodated permanent magnets (N35, 1.17–1.21 T, D × L = 4 mm × 2 mm) at its center. A similar actuator was used for the circular motion of the shuttle.

Magnetic flux calculation

The magnetic flux intensity of the magnets in different configurations was obtained by performing finite element simulations in COMSOL Multiphysics 5.5. The dimensions and properties of the magnets were set according to the specifications of the magnets. The magnetic field generated by permanent magnets was calculated by solving Gauss's magnetic law and the Maxwell–Ampère law in the static mode.⁴³ The results of simulations for each case are provided in Fig. S2 and Table S1 in the supplementary material.

Microfabrication masters

Microfabricated masters with characters (“R,” “M,” “I,” and “T”) were fabricated by spinning 100 μm thick films of SU-8 3050 photoresist (MicroChem) onto 4 in. silicon wafers. The photoresist was then patterned directly with a 375 nm laser with an exposure energy of 600 mJ/cm² using a maskless aligner (MLA150, Heidelberg Instruments, Germany). The non-crosslinked photoresist was then washed away using an SU-8 developer, revealing the patterned structures. A Teflon sheet of 4 μm was then placed onto the microfabricated master to facilitate the visualization of the microfabricated characters immersed inside the glycerol shuttle.

PDMS substrate with barriers

The PDMS film with barriers was fabricated using a mold consisting of an array of rectangular grooves (500 μm × 500 μm × 5 mm) separated by 5 mm. The mold was patterned directly in SU-8 3050 photoresist (MicroChem) using a 375 nm laser with an energy density of 600 mJ/cm² using a maskless aligner (MLA150, Heidelberg Instruments, Germany). A PDMS base and a curing agent (Sylgard 184 Silicone Elastomer Kit, Dow Chemical Co., USA) were mixed at a standard weight ratio of 10:1. PDMS was then poured onto the mold and spun at 100 rpm for 30 s before curing at 85 °C for 5 min. The resulting PDMS film was then peeled off the mold revealing the array of rectangular barriers, as detailed in Fig. S3 in the supplementary material. The PDMS film was cut to 22 × 60 mm pieces and then bonded to a glass slide. Later, the surface was pre-coated with silicone oil to modulate the friction between the barriers and the shuttle droplet during droplet manipulation experiments for comparison purposes.

Image collection and data processing

The motion of the iron drop on the surface of the glycerol shuttle was captured using an SLR camera (Nikon D750, 24 MP) with a Nikon AF-S 50 mm f/1.4 G lens. Shuttle stretching and splitting was recorded at 1000 frames per second (fps) using a high-speed camera (Phantom v1610) with an adjustable lens (Nikon AF-P 18–55 mm f/3.5–5.6 G). The high-speed imaging provided fundamental images during the droplet stretching driven by the magnetic force. Shuttle stretching was analyzed using ImageJ. The motion of the iron drop on the surface of the glycerol shuttle and the motion of the shuttle on the substrate were analyzed using machine vision techniques. Object tracking algorithms such as the Kernelized Correlation Filters (KCFs)³⁸ were used for tracking the object and calculating its velocity.

Contact angle measurements

The contact angle of droplets was measured using a contact angle measurement plug-in ImageJ (https://imagej.nih.gov/ij/plugins/contact-angle.html).⁴⁴ This plug-in estimated the contact angle of a droplet by approximating the droplet as a spherical or elliptical object. The profile of the glycerol droplet was influenced by gravitational forces (due to its large size) as well as the presence of the oil film. As such, the droplet profile was approximated by manually placing five points along the droplet edge.

RESULTS

Movement of a magnetic drop on the surface of a glycerol shuttle

First, we investigated the ability to manipulate a magnetic drop on the surface of a static glycerol shuttle with magnetic actuation (Fig. 1). To do so, we placed a 28 μl glycerol shuttle on a 127 μm thick parafilm sheet. A 0.3 μl magnetic drop with a medium iron concentration of 1.4091 g/ml was loaded on the surface of the shuttle. Our experiments indicated that an N50 permanent magnet positioned 5 mm off the top surface of the shuttle produces sufficient magnetic force to move the floating magnetic drop on the surface of the shuttle (see Video S1 in the supplementary material). The iron powder remained stable in the magnetic drop without being extracted from the drop during magnetic actuation. The magnetic drop floating on the surface of the glycerol shuttle remained stable even after a few hours due to the immiscibility of the sucrose solution and glycerol.³⁹ The glycerol shuttle remained stationary due to the static friction at the shuttle–substrate interface. The automated and highly controllable motion of the magnetic drop was achieved when attaching the permanent magnet to the rack of the linear actuator. Snapshot images show the steady motion of the magnetic drop on the surface of the shuttle [Fig. 1(a) and Video S1 in the supplementary material].

FIG. 1.

Automated, highly controllable, and stable movement of 0.3 μl magnetic drops floating on the surface of 28 μl glycerol shuttles through the using of 3D printed actuators. (a) Linear motion of a magnetic drop with an average velocity of 1.2 mm/s. (b) Circular motion of a magnetic drop with an average tangential velocity of 1.3 mm/s. (c) Circular motion of two magnetic drops with an average tangential velocity of 0.53 mm/s.

The motion of the magnetic drop on the surface of the shuttle can be described by the following equation, relating the magnetic force exerted on the magnetic drop to the shear stress induced by the motion of the magnetic drop in the surrounding viscous liquid:¹⁹

$$\underset{magneticforce}{\underbrace{{\displaystyle \frac{m_{drop}}{{\rho }_{drop}}\chi {\displaystyle \frac{B}{{\mu }_o}\mathrm{\nabla }B}}}}-\underset{shuttle-dropfriction}{\underbrace{0.85\pi U_{drop}{D}_{drop}{\mu }_{shuttle}}}=m_{drop}{\displaystyle \frac{d{U}_{drop}}{dt},}$$ (1)

where $m_{drop}$, ${\rho }_{drop}$, $U_{drop}$, and $D_{drop}$ are the mass, density, velocity, and equivalent diameter of the magnetic drop, respectively, B is the magnetic field produced by the magnet, μ_o is the permeability of vacuum, χ is the magnetic susceptibility of the magnetic drop, μ_shuttle is the viscosity of glycerol, and t is the time. The friction coefficient for a thin circular disk when positioned parallel to a laminar flow is obtained as $13.6/Re$, in which Re is the Reynolds number of the disk in the flow, as detailed in Ref. 45. The friction coefficient at the interface of the shuttle and magnetic drop was estimated by assuming the magnetic drop as a thin disk, which is immersed in glycerol. The friction at the top surface of the magnetic drop in contact with air was ignored due to the negligible viscosity of air. Under this condition, the friction coefficient at the interface of the shuttle and magnetic drop reduced to half and was estimated as $6.8/Re$, in which $Re={\rho }_{shuttle}{U}_{drop}{D}_{drop}/{\mu }_{shuttle}$.

We further investigated the versatility of this mechanism for rotating single and double magnetic drops on the surface of the shuttle. The circular motion of the single magnetic drops was achieved using the rotating actuator coupled with an N50 permanent magnet [Fig. 1(b) and Video S1 in the supplementary material]. The circular motion of double magnetic drops was achieved using the same rotating actuator coupled with an N35 permanent magnet [Fig. 1(c) and Video S1 in the supplementary material]. In the latter, the magnetic strength was reduced to avoid the actuation of the magnetic drop by the opposite magnet.

Movement of the glycerol shuttle

Next, we investigated the utility of the magnetic drop for the movement of the shuttle. This was achieved by increasing the volume of the magnetic drop to 6.3 μl and lubricating the surface with a 100 μm thick layer of silicone oil to reduce the friction. The vertical distance between the permanent magnet and the top surface of the shuttle was set to 8 mm. The combination of these two parameters facilitated the magnetic sliding of the shuttle on the surface [Fig. 2(a) and Video S2 in the supplementary material]. The sliding motion of the shuttle is governed by the magnetic force imposed on the magnetic drop and the viscous induced friction induced at the interface of the shuttle and the lubricant silicon oil, as described below,

$$\underset{magneticforce}{\underbrace{{\displaystyle \frac{m_{drop}}{{\rho }_{drop}}\chi {\displaystyle \frac{B}{{\mu }_O}\mathrm{\nabla }B}}}}-\underset{viscousfriction}{\underbrace{{\displaystyle \frac{k_{shuttle}{\mu }_{oil}{U}_{shuttle}}{h_{oil}}{A}_{shuttle\text{-}base}}}}=(m_{shuttle}+m_{drop}){\displaystyle \frac{d{U}_{shuttle}}{dt},}$$ (2)

in which, $k_{shuttle}$, $U_{shuttle}$, and $m_{shuttle}$ are the shear coefficient, sliding velocity, and mass of the shuttle, respectively, $A_{shuttle\text{-}base}$ is the area of the shuttle base in contact with the lubricant, and ${\mu }_{oil}$ and $h_{oil}$ are the viscosity and thickness of the lubricant oil, respectively.

FIG. 2.

Magnetic sliding of the glycerol. (a) Linear (sliding) motion. (b) Characterization of shuttle sliding velocity at various volumes of the magnetic drop and iron powder concentrations, with the insets showing the representative images of the no-motion, steady-motion, and magnetic drop extraction regimes. (c) Circular motion. (d) Sinusoidal motion.

To get an insight into the sliding motion of the shuttle, we characterized the sliding velocity of the shuttle at various volumes of the magnetic drop loaded onto the shuttle as well as at low (1.38 g/ml), medium (1.41 g/ml), and high (1.43 g/ml) concentrations of iron powder within the magnetic drop while setting the volume of the shuttle to 28 μl [Fig. 2(b)].

Analysis of the shuttle sliding velocity curves indicated the existence of three regimes, which are referred to as no-motion, steady-motion, and iron drop extraction (see Video S3 in the supplementary material). The no-motion regime corresponded to the condition that the magnetic force was less than the viscous friction and therefore could not drag the shuttle $(F_{magnet}<F_{friction})$. This regime occurred at low volumes of iron drop. The steady-motion regime corresponded to the condition that the magnetic force was more than the viscous friction and could drag the shuttle $(F_{magnet}>F_{friction})$. This regime occurred at medium volumes of iron drop. A closer look at the velocity curves reveals the gradual decrease of their slope, which is attributed to the gradual reduction of the $m_{drop}/(m_{shuttle}+m_{drop})$ term in the equation. The magnetic drop extraction regime corresponded to the condition that the magnetic force was more than the maximum capillary force at the magnetic drop and the shuttle before the magnetic drop would be extracted from the shuttle $(F_{magnet}>F_{capillary})$. The maximum capillary force can be estimated as $F_{capillary}=\pi D_{drop}{\gamma }_{shuttle\text{-}drop}$, in which $d_{drop}$ is the equivalent diameter of the magnetic drop, obtained as $D_{drop}={(6m_{drop}/\pi {\rho }_{drop})}^{1/3}$ and ${\gamma }_{shuttle\text{-}drop}$ is the interfacial tension between the shuttle and the magnetic drop. This regime occurred at high volumes of iron drop.

Our experiments suggest that the shifting between the no-motion, steady-motion, and magnetic drop extraction regimes depends on the concentration of the iron powder in the magnetic drop and the volume of the magnetic drop, as presented in Fig. S4 in the supplementary material. At high concentrations of iron powder (1.43 g/ml), the droplet becomes very sensitive to the volume of the magnetic drop such that small changes in the volume of the drop might lead to the transition between the above three regimes.

Our further analysis indicated that during the steady-motion mode, increasing the velocity of the glycerol shuttle is associated with increasing the contact angle of the shuttle. This trend was observed for velocities up to 4.5 mm/s, after which the contact angle does not change significantly, as presented in Fig. S5 in the supplementary material.

We further demonstrated the versatility of this mechanism for the movement of the shuttle along circular and sinusoidal paths. The circular motion of the shuttle was achieved by attaching an A4988 stepper motor to the rack of the linear actuator [Fig. 2(c) and Video S2 in the supplementary material]. The sinusoidal motion of the shuttle by incorporating an additional linear actuator to generate independent motions along the x and y axes [Fig. 2(d) and Video S2 in the supplementary material].

Deformation of the glycerol shuttle

We further investigated the magnetic deformation of static shuttles (Fig. 3). This was achieved by injecting a 28 μl glycerol shuttle onto a parafilm substrate followed by inserting a 9 μl magnetic drop with an iron concentration of 1.4339 g/ml onto the shuttle. Two stacked N50 permanent magnets were positioned 9 mm from the substrate surface (5 mm from the peak of the shuttle). The horizontal movement of the magnet led to the movement of the magnetic drop along the top surface of the shuttle [Fig. 3(a) and Video S4 in the supplementary material]. Due to the increased magnetic force (caused by increasing the magnetic drop volume and the iron concentration within the magnetic drop) resulted in the slight lifting of the magnetic drop, which in turn led to the localized deformation of the glycerol shuttle (Fig. 3(a) and Video S4 in the supplementary material]. We further explored vertical deformation; the magnetic drop aligned itself with the magnet, similar to the conditions presented in Fig. 1(a). Further increase of the magnetic force (caused by reducing the distance between the shuttle and the magnet volume: 7 mm) led to the extraction of the magnetic drop from the surface of the shuttle. In this case, the magnetic force was more than the surface tension between the iron drop and the surrounding glycerol ($F_{magnet}>F_{capillary}$). The process of the magnetic drop extraction occurred very quickly (<75 ms) and therefore was captured using a high-speed camera at 1000 frames per second [Fig. 3(b) and Video S4 in the supplementary material]. The extraction of magnetic drop was accelerated by addition of water to the shuttle, which in turn reduced the surface tension coefficient at the interface of the iron drop-glycerol, as presented in Fig. S6 in the supplementary material.

FIG. 3.

Magnetic deformation of the glycerol shuttle. (a) Horizontal stretching of a shuttle placed between physical barriers: schematic diagram, time-lapse images of the glycerol stretched by horizontal magnetic actuation, and measurement of the stretched shuttle. (b) Vertical stretching of glycerol shuttle by magnetic actuation: schematic of the experimental setup, time-lapse images showing the stretching, and measurement output from the height deformation of the shuttle.

Transmission of characters patterned on a planar surface

We further explored the utility of the magnetically actuated shuttle and its internal light refraction properties to transmit text from planar surfaces (Fig. 4). The characters “R,” “M,” “I,” and “T” were laser printed on a 4-in. silicon wafer. A 4 μm Teflon sheet was placed on the silicon wafer [Fig. 4(a)]. Silicon oil was injected to fill the gap between the silicon wafer and the Teflon sheet. Importantly, the silicon oil generated air bubbles around each letter, providing optical contrast required for the visualization of the characters. The camera was placed with respect to the silicon wafer and, in turn, the characters, which made it almost impossible to visualize the characters. Nevertheless, the internal refraction and the reflection of light inside the glycerol droplet turned the shuttle into a drop lens, causing a 90° translation along with 3× magnification of the characters [Fig. 4(b)]. Time-lapse images show the transmission of the “R M I T” message using the magnetically actuated shuttle [Fig. 4(c) and Video S5 in the supplementary material].

FIG. 4.

Transmission of microscale text messages of a magnetically actuated shuttle. (a) Schematics of the setup, (b) schematic diagram showing the 90° translation and magnification of the imprinted characters by the glycerol shuttle, and (c) time-lapse images showing the transmission of “RMIT” message using the glycerol shuttle.

Driving of a glycerol shuttle over microfabricated barriers

Next, we explored the magnetic actuation of the droplet over uniform surfaces (Fig. 5). To do so, we used a PDMS-based substrate with rectangular-shaped groves patterned onto its surface. The silicone oil lubricant was applied to the surface to reduce the friction. A 28 μl glycerol shuttle was placed between the barriers. A 9.5 μl magnetic drop was loaded onto the shuttle. Stacked N50 permanent magnets attached to the linear actuator were used to slide the shuttle over the barriers [Fig. 5(a)]. The shuttle exhibited a snail-like deformation (see Video S6 in the supplementary material), being stretched up to 160% of its length to pass the barriers and contract to its original size after passing the barrier. This behavior was repeated in the consequent barriers [Fig. 5(b)]. The dynamic variations in the length of the shuttle while passing over the barriers is illustrated in Fig. 5(c). Our extended experiments indicated that increasing the volume of the magnetic droplet (10.2 μl) accelerates the snail-like motion of the shuttle (see Video S7 in the supplementary material). This characteristic is presented in Fig. 5(d), which compares the outline of two glycerol shuttles injected with varying volumes of magnetic drop.

FIG. 5.

Magnetic transportation of a glycerol shuttle over a PDMS substrate with patterned rectangular-shaped barriers, which is coated with silicon oil. (a) Schematics of the setup. (b) Snapshot images showing the snail-like motion of the shuttle over the barriers involving the cyclic stretch and contraction of the shuttle. (c) Dynamic variations in the length of the glycerol shuttle when passing over consequent barriers. (d) Comparing the outline of two glycerol shuttle injected with varying volumes of magnetic drop, showing the accelerated motion of the shuttle with a larger volume of magnetic drop.

Combination of magnetic and electrowetting mechanisms

We also explored the integration of the magnetically driven glycerol shuttle with the electrowetting on dielectric (EWOD) mechanism (Fig. 6). In this setup, the shuttle could be driven by EWOD while the magnetic drop could be independently driven by a permanent magnet (see Video S8 in the supplementary material). In the absence of magnetic actuation, the magnetic drop surfed on the shuttle surface due to surface tension driven shuttle deformation as well as shuttle internal flows, as comprehensively described in our recent work³⁸ [Fig. 6(a)]. In the presence of magnetic actuation, the magnetic drop could be fixed at the same location with respect to the shuttle [Fig. 6(b)] or even undergo customized linear or circular paths on the surface of the shuttle [Fig. 6(c)]. This experiment further highlighted the versatility of our actuation mechanism system to be integrated with well-established drop manipulation mechanisms to obtain more complex and highly controllable motions.

FIG. 6.

The actuation of the magnetic drop under the combination of EWOD and magnetic effects. (a) Surfing of the magnetic drop on a moving shuttle under EWOD actuation. (b) Anchoring the magnetic drop to a moving shuttle under EWOD and magnetic actuation. (c) Customized surfing of the magnetic drop by magnetic actuation on a shuttle which is moved and anchored by EWOD.

CONCLUSIONS

In summary, we demonstrated the magnetic actuation of a glycerol shuttle using a suspension of iron powder in sucrose solution applied onto the surface of the shuttle. Incorporation of a 3D printed rack and pinion actuators enabled the automated manipulation of the glycerol shuttle along customized linear, circular, and sinusoidal paths. Increasing the magnitude of the magnetic force led to the localized deformation of the shuttle surface and the full extraction of the magnetic drop from the glycerol shuttle. Our experiments indicated the ability to propel the shuttle over a rough substrate patterned with an array of microfabricated rectangular-shaped barriers, during which the shuttle experimented a snail-like motion, involving cyclic expansion and contraction. The unique optical properties of the shuttle allowed for the magnification and 90° translation of microscale characters patterned on the substrate, transforming the shuttle into a moving lens. The system can be easily integrated with current electrowetting on dielectric setups. This feature was used for controlling the motion of the magnetic drop over the surface of a moving shuttle. The system is simple, inexpensive, versatile, and controllable. These features can benefit the development of the next generation of soft actuators for handling, manipulation, preparation, and analysis of liquid samples for various chemical, biochemical, and biological applications.^46–48 The controlled motion and positioning of glycerol shuttles can also benefit the fabrication of complex, 3D microfluidic structures in which the glycerol serves as a fugitive ink.^49,50

SUPPLEMENTARY MATERIAL

See the supplementary material for details of 3D printed gear actuators (Fig. S1), calculation of magnetic field intensity (Fig. S2), details of microfabricated rectangular barriers (Fig. S3), further insights into the transition between various shuttle regimes (Fig. S4), contact angle of the shuttle (Fig. S5), magnetic deformation of the glycerol shuttle at different glycerol–water mixtures (Fig. S6), and videos showing the movement of magnetic drop on the surface of glycerol shuttle (Video S1), motion of glycerol shuttle along linear, circular, and sinusoidal paths (Video S2), investigating various regimes of glycerol shuttle motion (Video S3), deformation of glycerol shuttle (Video S4), transmission of characters patterned on a planar surface (Video S5), motion of glycerol shuttle over microfabricated barriers (Video S6), comparison of glycerol shuttle motion injected with different volumes of magnetic drop over microfabricated barriers (Video S7), and combination of magnetic and electrowetting mechanisms (Video S8).

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.