Manipulation Schemes and Applications of Liquid Marbles for Micro Total T Analysis Systems ⁎ Jing Jin, Nam-Trung Nguyen

Microelectronic Engineering 197 (2018) 87–95

Contents lists available at ScienceDirect

Microelectronic Engineering

journal homepage: www.elsevier.com/locate/mee

Research paper Manipulation schemes and applications of liquid marbles for micro total T analysis systems ⁎ Jing Jin, Nam-Trung Nguyen

Queensland Micro- and Nanotechnology Centre, Griﬃth University, Nathan, QLD 4111, Australia

ARTICLE INFO ABSTRACT

Keywords: Micro total analysis systems (microTAS) provide the opportunity to create complete analytical microsystems by MicroTAS integrating various functional modules, such as sample preparation, separation and detection, into a single chip- Digital microfluidics sized microfabricated device. Microfluidics is the enabling technology for implementing the concept of Liquid marbles microTAS. Liquid marble (LM), as a promising separate digital microfluidic platform, has the great potential to Manipulation enhance the broad applications of microTAS. LMs are small liquid droplets encapsulated by multilayered hy- Applications drophobic particles and have attracted a great interest from the microfluidics research community due to their Microreactors non-wetting property. A LM maintains its integrity and exhibits low friction on various carrier surfaces, enabling the LM to be actuated by external electric, magnetic, gravitational and acoustic fields or other manipulation schemes. LMs can thus serve effectively for the storage and transportation of small liquid volumes. In addition, they have been widely used for the quick detection of water pollution or gas emission and, most importantly, micromixing and microreactions for chemical and biomedical purposes. This paper reviews the recent developments in the manipulation techniques and emerging applications of LMs. The review aims to facilitate better understanding of their use as a unique digital microfluidic platform to promote further advancement of microTAS. The paper begins with different manipulation schemes of LMs according to the nature of actuation energy. Next, it summarizes the diverse applications of LMs for various chemical and biological assays. Finally, this paper concludes with future perspectives regarding the research on LMs in microTAS technologies.

1. Introduction classified as continuous-flow microfluidics and digital (droplet-based) microfluidics (DMF) [17]. In particular, DMF is an emerging liquid- Starting from a humble gas analysis system and microelec- handling technology dealing with the manipulation of discrete droplets, tromechanical systems integrated on silicon chips for chemical analysis, rather than continuous flows of liquid [18]. In DMF, droplets on the the field of micro total analysis systems (microTAS), also called “lab on order of microliter serve as miniaturized reaction chambers. This pro- a chip (LOC)”, has developed rapidly since the establishment of cess has numerous advantages such as minimum reagent requirement, microTAS concept, first proposed by Manz et al. [1]. Currently, mi- fast response rates, low cross-contamination and, more importantly, the croTAS has received much attention from a broad spectrum of scientific capability of performing parallel tests. These advantages make DMF a and engineering disciplines and has been successfully used for diverse perfect candidate for practical LOC, microTAS and point-of-care diag- applications in analytical chemistry [2–9] and biology [10–16]. In nostic devices for clinical use [19–23]. DMF can be further categorized, general, the most important advantage of microTAS is the possibility of based on the droplet type, as droplet-based DMF and liquid-marble- creating complete analytical microsystems by integrating various based DMF [17, 24]. functional modules, such as sample preparation, separation and de- As an individual digital microfluidic platform, liquid marble (LM) is tection, into a microfabricated device and often reducing the entire a rapidly growing research area in microTAS. A LM is a relatively small laboratory room down to a single chip, Fig. 1. liquid droplet encapsulated by a porous protective coating that consists The enabling technology for implementing the concept of microTAS of multilayered micro- or nanometer-sized powders. Most coating is microfluidics, which allows for manipulating samples and reagents powders are hydrophobic, which allows a LM to be manipulated like a within small liquid volumes [7]. According to the way how these small soft solid [25–31]. The porous protective coating physically isolates the amounts of liquid could be handled and manipulated, microfluidics is liquid core from its surroundings but allows for the transport of gas or

⁎ Corresponding author. E-mail address: nam-trung.nguyen@griﬃth.edu.au (N.-T. Nguyen). https://doi.org/10.1016/j.mee.2018.06.003 Received 30 December 2017; Received in revised form 22 May 2018; Accepted 12 June 2018 Available online 13 June 2018 0167-9317/ © 2018 Elsevier B.V. All rights reserved. J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

and practically zero cross-contamination. When two LMs are placed into contact, they do not coalesce naturally, even though they are pressed against each other. The pronounced elastic property of LMs enables them to sustain a completely reversible deformation of up to 30% [44–46]. However, LMs can be forced to merge with each other via impact [47–49], which demonstrates the possibility of using LMs as reactors and mixers at the microscale by coalescing two or more LMs containing different reagents or ingredients. As mentioned above, a LM maintains its integrity and exhibits low friction on various carrier surfaces. This characteristic enables LMs to be actuated in different external fields, such as electric and magnetic fields [50–53], or even to be transported by light-driven motion [54]. In terms of practical applications of LMs, apart from offering controlled liquid manipulation and transportation [27, 55–57], LMs also exhibit a great potential for their use as miniature laboratories in microTAS, where small-scale laboratory operations such as microreactions and detections for analytical chemistry and biotechnology, can be per- formed [35, 58–60]. Based on the concept of microTAS, this paper reviews recent developments in the manipulation techniques and emerging applications of LMs and aims to facilitate better understanding of their use as a novel Fig. 1. Schematic of key functional modules that can simultaneously be in- digital microfluidic platform to promote further advancement of corporated into microTAS. microTAS. The review begins with different manipulation schemes of LMs according to the nature of actuation energy. A discussion follows on the diverse applications of LMs in microTAS for chemical and biological assays. Finally, this paper concludes with perspectives regarding future research on LMs in microTAS technologies.

2. Manipulation schemes of liquid marbles

The manipulation of LMs enables LM-based digital microfluidic platforms to be implemented in different modules of microTAS such as sample handling, reaction and mixing at the microscale and sample detection. There have been some review papers on the fundamental physics, properties and applications of LMs [27–30] after the first re- port on the concept of LM by Aussillous and Quere [26]. However, there Fig. 2. Schematic of the structure of a LM resting on a substrate. are few reviews on the manipulation of LMs except for the work of Ooi and Nguyen [25], which systematically classified the manipulation vapour across the coating layer [24]. Due to the non-stick property schemes of LMs according to the nature of actuation energy. This re- enabled by the air cushions between coating particles and the carrier view paper further summarizes the manipulation schemes of LMs pro- substrate, Fig. 2, LMs exhibit ultra-low friction on various carrier sur- posed in recent years based on this classification method, Fig. 3. faces. This feature allows LMs to move easily across solid and even liquid surfaces without any liquid leakage. Therefore, LMs can be re- 2.1. Electric schemes garded as a promising alternative to superhydrophobic surfaces in the transportation of small amounts of liquid. Compared to bare droplets on Electric-field actuation as a means to manipulate a LM is attractive a superhydrophobic surface or oil-immersed systems in DMF, LMs have because there is no need for physical contact between electrodes and several significant advantages: (i) reduced droplet evaporation rate as liquids inside LMs. Besides, accurate control can be easily achieved by – compared to bare droplets [32 34]; (ii) lower possibility of cross-con- varying the applied voltage. However, this scheme also has some tamination and more accessibility of the liquid inside LMs; (iii) no need drawbacks such as the requirement for high voltage and unavoidable for complicated fabrication of micro-electrodes as compared to elec- charging of coating particles [25]. trowetting-on-dielectric (EWOD); and (iv) low cost and great con- venience for rapid production [35]. 2.1.1. Electrostatic force A LM is formed by simply rolling a liquid droplet over a bed of A number of researchers have demonstrated that an electric field hydrophobic powder, which may produce a random aggregate of par- could deform, move or change the wettability of a LM [27, 28, 51, 52, ticles, or by other methods, such as the electrostatic formation [36] and 57, 61, 62]. The electric energy applied to a stationery LM in an electric dropwise condensation [37]. In the conventional rolling method, the field causes an increase in surface energy that deforms the marble or an hydrophobic powders assemble on the droplet surface, effectively extra increase in kinetic energy that moves the marble. Aussillous and creating a protective coating. LMs thus can rest stably and roll on a solid Quere [27] initially reported the presence of electrostatic effect by substrate or even float and slide on a liquid surface for some time before using a charged Teflon stick to levitate a LM against gravity. They dissipating via evaporation [38–41]. A large variety of coating mate- observed a successive bounce of a lycopodium coated water marble rials and liquid cores have been selected to create LMs successfully and between a metal plate and the charged stick. When the charged stick the combination of powders and liquids will ultimately determine the was displaced and kept at a distance to the marble, a motion of the properties of LMs and corresponding manipulation schemes. The sta- marble was induced. bility and other performance indexes of different LM compositions have Bormashenko's group conducted a series of experiments by applying been studied comprehensively by McEleney et al. [42] and Zang et al. a uniform electric field to LMs [51, 52, 61, 62]. The applied electric [43]. For DMF, samples in LMs can be transported with minimal energy field generated an upward electrostatic force that opposed gravity and

88 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

Fig. 3. An overview of different manipulation schemes of LMs. Each scheme consists of one or several approaches with corresponding cap- abilities shown in the right column. The different outline colours of manipulation approaches mean different levels of manipulation ability on controlling the motion of LMs. Red outline indicates precise control, blue outline indicates proper control and green outline indicates poor control. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

deformed the marbles to a prolate spheroid so that the contact area with the solid surface was reduced, Fig. 4(a). The group established a threshold value of the uniform electric field (15 cgse) for the deformation of LMs. The group further introduced composite marbles consisting of different polar liquids [52] and Janus marbles coated with dielectric/semiconductor particles [61, 62] and reported their actuation with the electric field [52, 61, 62]. Two prepared marbles were brought into contact and pressed to coalesce, resulting in the formation of composite marble or Janus marble. When the composite or Janus marble was put in a uniform electric field, one part of the composite or Janus marble with smaller dielectric constant would climb onto the other part of the marble. Thus, the whole marble was strongly polarized and the dipole moments of two parts of the marble were oriented along the direction of the electric field, Fig. 4(b). Liu et al. [63] focused on the coalescence of water marbles driven by a direct current (DC) electric field. The authors found that two contacting marbles could be forced to coalesce, if they were charged with a sufficiently high voltage, Fig. 5(a). The threshold voltage of the electrocoalescence sensitively depends on the stabilizing particles as well as the surface tension of the liquid core. In addition, it was found that multiple marbles in a chain could be fi Fig. 4. Schematic of the electric manipulation of LMs in a uniform electric eld. driven to coalesce by a sufficiently high threshold voltage that in- Note that the coating particles are not illustrated. (a) Shape deformation of a creased linearly with the number of the marbles, Fig. 5(b). The elec- LM after applying an electrostatic force. (b) The climbing phenomenon of a trocoalescence of LMs could be useful in microreactors for chemical and composite marble consisting of different polar liquids or a Janus LM coated with dielectric/semiconductor particles (with two parts shown in different biomedical reactions involving multiple reagents or ingredients [24]. colours) after applying an electrostatic force. The dipole moments of two parts The wettability of a liquid droplet can be changed in an applied within the composite/Janus LM align themselves in the direction of the electric electric field, which is known as electrowetting. Consequently, EWOD field. (For interpretation of the references to colour in this figure legend, the can modify the contact angles at either sides of the droplet to different reader is referred to the web version of this article.) values and create a driving force to actuate the droplet motion. As mentioned above, LMs serving as an alternative option for bare droplets on a superhydrophobic surface also have an ideal non-wetting property on smooth solid substrates with the contact angle close to 180°. McHale, Newton and other co-workers [28, 57, 64] focused their research on the

89 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

Fig. 6. Schematic of the dielectrophoretic picking of a LM coated with poly- tetraﬂuoroethylene (PTFE) powders in a non-uniform electric ﬁeld.

magnets are available off-the-shelf and can be easily incorporated into various miniaturized devices. Moreover, most importantly, they do not require external power supply. However, magnetic force can only be utilized for LM actuation when a LM has a magnetic coating or consists of a magnetic solution. A magnetic field can change the shape and wettability of a LM, which is known as magnetowetting [70, 71]. In these studies, a sessile ferrofluid marble coated with PTFE powders deformed under the influence of a permanent magnet. Dorvee et al. [47] first reported that two LMs encapsulated by magnetic amphiphilic particles could be manipulated to coalesce using a magnet to perform a reaction. The authors argued that the mixing process of chemical reagents inside LMs resulting from marble coalescence could be aided by increasing the rate at which the two LMs collided. Aussillous and Quere [27] further demonstrated that water Fig. 5. (a) Schematic of the coalescence of two charging LMs resting on a marble coated with a powder mixture of iron and lycopodium or con- polydimethylsiloxane (PDMS) substrate. V, κ, η, γ are the volume, electrical taining a suspension of iron particles could be transported with a conductivity, viscosity and surface tension of the liquid core respectively. The magnet. A LM with magnetic components can be accurately positioned voltage U is applied through the electrodes inserted into the LMs. The current I and moved across various solid and liquid surfaces under the effect of a is monitored through an ammeter during the coalescence process. (b) A chain of magnetic field [35, 50, 53, 55, 56, 59, 72, 73]. Zhao et al. [50] and LMs charged by a voltage and its analogous electric circuit. Bormashenko et al. [56] found that the threshold magnetic force needed to induce marble movement is equal to the frictional force be- electrowetting of LMs and demonstrated the possibility of using a finger tween the marble and its carrier surface. In a magnetic field, the electrode structure to control the motion of LMs coated with lycopo- magnetic surface particles, such as hydrophobized iron oxide, of LMs dium powders [57]. They also used a similar EWOD configuration with rearrange and thus magnetic LMs can be controlled to open and close applied frequency from 1 to 250 Hz to excite the LMs into both an up- reversibly, as shown in Fig. 7, for microreaction, liquid exchange, op- and-down motion and a resonant oscillation involving multiple nodes tical probing [50, 55, 59] and recently, “on-line” quantitative electro- [64]. It was found that the electrowetting of LMs was completely re- chemical/optical detection [35]. versible. Marbles could return to their initial contact angles for both alternating current (AC) and DC electrowetting and without the re- 2.3. Mechanical schemes quirement of exceeding a threshold voltage. 2.3.1. Gravitational force 2.1.2. Dielectrophoretic force Rolling a LM down an inclined plane under the effect of gravity is Apart from the electrostatic force, the dielectrophoretic force pro- one of the simplest methods to initiate its movement. During this rolling duced in a non-uniform applied electric field can also be adopted to process, the gravitational potential energy of the LM converts to its actuate LMs. Inspired by the dielectrophoretic manipulation of cells in kinetic energy if the friction between the LM and the inclined plane is biotechnology [65–67] and droplets in DMF [65, 68, 69], the principle neglected. Aussillous and Quere [74] studied the shape evolution of of dielectrophoresis has been applied for manipulating LMs by Nguyen's LMs rolling down a plane with a specific slope. The team observed group [49]. LMs were not only deformed but also picked by the die- shapes including disks, wheels and peanuts and interpreted the pre- lectrophoretic force in a non-uniform electric field, Fig. 6. They linked sence of different shapes based on a balance between centrifugal and the picking force to the electric field strength and used this approach in capillary forces. Subsequently, Aussillous and Quere [27] investigated the vertical collision of two LMs, which avoids the rotation and posi- the viscous motion of LMs on a slightly inclined plane and summarized tioning problems in the collision process. It is essential for better un- their behaviours and regimes. Yang et al. [75] proved the possibility of derstanding the coalescence process of two LMs. This dielectrophoretic generating a small amount of electricity by rolling an ionic LM down an method provides reliable automatic handling for a wide range of ex- inclined layer of interdigitated electrodes. Furthermore, a LM-based isting non-ferromagnetic LMs without the need to modify their coatings accelerometer prototype was successfully created by Zeng and Zhao or liquid cores. [76] by utilizing the gravitational force. Planchette et al. [48] utilized gravity to control the speed of droplets or LMs that merged with much 2.2. Magnetic scheme larger LMs resting on the bottom substrate.

Magnetic force can work effectively at a relatively long range 2.3.2. Pressure gradient without the specific requirement of physical contact. Small permanent Bormashenko et al. [77] first introduced LM-based micropumps by

90 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

Fig. 7. Schematic of magnetic manipulation of a LM coated with Fe3O4 nanoparticles. From left to right, the Fe3O4 nanoparticles concentrate downwards when a small magnet is brought into vicinity of the marble. This process is reversible as the magnetic nanoparticles can reassemble themselves after the magnet is removed.

[83] and Hashmi et al. [84] reported that the spherical LM could be reversibly flattened into the shape of a flying saucer at low tempera- tures due to the Marangoni convection within the marble. For a sessile LM on a cold substrate, ice usually starts to form at the bottom of the marble. Hashmi et al. [84] inferred that the liquid core impeded ice formation at the bottom edge of the marble due to the warmer coating particles. This resulted in a thermal and surface tension gradient in the liquid core, which drove the Marangoni convection. Zang et al. [83] concluded that the hydrophobicity of the coating material determined the shape of the frozen marble since the Gibbs free energy barrier was influenced by the wettability of the coating material. Utilizing the thermally reversible phase separation properties of poly(N-iso- Fig. 8. Schematic of the LM-based micropump. A PE capillary tube is used to propylacrylamide) (PNIPAm), Yusa et al. destroyed a LM coated with connect two water marbles coated with polyvinylidene fluoride (PVDF) and PNIPAm by lowering the temperature to its lower critical solution point lycopodium powders respectively. Water flows from the PVDF-coated marble [85]. (coloured with potassium permanganate) to the lycopodium-coated marble. (For interpretation of the references to colour in this figure legend, the reader is 2.4.2. pH change referred to the web version of this article.) Fujii and co-workers [86–90] developed various LMs coated with specific particles that are responsive to pH changes. These coating connecting two LMs encapsulated by different hydrophobic powders materials are neutrally charged and remain hydrophobic in an alkaline with a polyethylene (PE) capillary tube, Fig. 8. The difference in surface or acidic environment. However, they become unstable and hydrophilic tension of the two marbles resulted in a difference in Laplace pressure, when the pH value of the carrier liquid changes considerably, which which allowed the liquid core to be transported from one marble to results in the rupture of the LMs. another until the pressures were equalized. The authors also proposed that varying the radii of two connecting marbles could be used to 2.4.3. Irradiation produce different Laplace pressures. This method utilizing pressure The wettability of LMs coated with photoresponsive particles can be gradient does not require external energy supply, which is easy to im- changed by optical irradiation rather than by any physical contact. plement. This micropump could be used for precise delivery of small Ultraviolet (UV) radiation induces surface structure changes of the amounts of liquids, the design of microreactors and other microfluidic coating material, resulting in a significant increase in the wettability of applications. Nevertheless, a way to sustain the pressure difference is LMs to the rupture point. Tan et al. [91] reported that water LMs sta- needed before this concept could be incorporated into a microfluidic bilized by photoresponsive TiO2 particles were stable when shielded device for microTAS. from UV radiation. However, they quickly collapsed after being irradiated with 302 nm UV light. The authors also observed similar UV- induced collapse of oil- and organic-solvent-based LMs with oleophobic

2.4. Other schemes TiO2 nanoparticles. Zhang et al. [73] prepared LMs encapsulated by novel core/shell-structured stimuli-responsive magnetic particles (de- 2.4.1. Temperature change signated RMPs) whose hydrophilicity increased when protonated. The porous shell of a LM prevents the liquid core from coming into These RMPs were in turn coated with silica loaded with photoacid contact with solid and liquid supporting surfaces but allows gas ex- generator (PAG). When irradiated with UV, the PAG protonated the change to occur. Consequently, a LM evaporates, shrinks, deforms and RMPs and caused the coating particles to become hydrophilic, leading eventually collapses over time. This phenomenon and the evaporation to the rupture of the marbles. Nakai et al. [79] also created UV re- rate of LMs have been studied in detail with LMs coated with various sponsive LMs with a coating of spiropyran powders. The method of UV materials [32, 34, 39, 78]. The lifetime of a LM can be controlled by irradiation provides the remote control capability to LM manipulation, adjusting temperature, humidity and the marble volume. In high hu- but may not be suitable for manipulating LMs containing sensitive midity, a LM may last for a few days [34, 79]. The long lifetime of a LM biological samples. Kavokine et al. [54] demonstrated the light-driven is needed for biomedical applications where cell cultures usually re- transport of ﬂoating LMs. LMs were deposited on a water solution quire extended periods of storage time [25]. The importance of containing photosensitive surfactants in the experiment. It was found prolonging marble lifetime was demonstrated by Shen's research group, that irradiation of the solution generated photoreversible Marangoni who grew various types of cells inside LMs [80–82]. ﬂows that transported LMs toward UV light and away from blue light In addition, a LM can also be deformed by freezing. Both Zang et al. when the liquid substrate was thick enough (Marangoni regime). Below

91 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95 a critical thickness, LMs moved in the opposite direction to that of the digital microfluidic platforms can be applied in different modules of surface flow at a speed increasing with decreasing liquid thickness microTAS, such as sample handling, reaction and mixing at the mi- (anti-Marangoni), which was driven by the free surface deformation. croscale and sample detection. A LM maintains its integrity and exhibits Paven et al. [92] recently described that the motion of LMs on a water low friction on various carrier surfaces, enabling the LM to be actuated surface with a simple near-infrared laser or sunlight. They argued that by external electric, magnetic, gravitational and acoustic fields. In ad- using light rather than pH or temperature as an external stimulus al- dition, LMs can be manipulated via pressure gradient, temperature lowed for precise control of the position, area, timing, direction and change, pH change, irradiation, surfactant concentration and self-pro- velocity of the driven motion. This approach makes the idea of trans- pulsion. Among these manipulation schemes, electric, magnetic and porting and releasing the materials encapsulated within the LMs at a acoustic actuations are three key techniques for effectively manip- specific position and time possible. The authors further reported that ulating a LM without physical contact. However, electric manipulation the LMs could function as light-driven towing engines to move small schemes require high working voltage and the magnetic manipulation objects. scheme requires the LM to have a magnetic coating or to consist of a magnetic solution. Moreover, acoustic radiation pressure can only be 2.4.4. Surfactant concentration simply used to levitate LMs and open the shell of a single non-ferro- Bormashenko and Musin [93] demonstrated that changing the sur- magnetic marble. The remaining challenge for researchers in this field face tension of a carrier liquid by adding some organic or inorganic is how to incorporate the concepts of mechanical manipulation tech- solutions altered the integrity of a LM floating on the surface. The au- niques, such as gravitational force and pressure gradient, into a mi- thors concluded that surface tension of the carrier liquid should be crofluidic device. Besides, the manipulation approaches relying on the higher than that of the LM to maintain its integrity. The argument is environmental parameters, such as temperature, pH and irradiation, based on the fact that it is energetically more favourable for the coating may not be suitable for manipulating LMs containing sensitive biolo- particles to adhere to the fluid with lower surface tension. The collapse gical samples. Furthermore, the surfactant-based manipulation techni- of a LM enables the water pollution to be detected with naked eyes. ques would be limited in applications involving live cells as the surfactants are harmful to the cell culture. 2.4.5. Acoustic levitation Zang et al. [94] used an acoustic levitator to manipulate LMs and 3. Practical applications of liquid marbles in micro total analysis thus achieved reversible opening and closing of the particle coating of systems LMs through acoustic levitation. By increasing the sound intensity, the fl shape of the levitated LM changed from a quasi-sphere to a attened Due to their unique properties, LMs as an emerging digital micro- ellipsoid. Meanwhile, the particles migrated from the polar area of the fluidic platform have attracted a great attention from researchers in the shell to the equatorial area. It resulted in the formation of a cavity on past few years. To date, based on various manipulation techniques the marble surface that could be healed with decreasing sound in- mentioned above, the practical uses of LMs range from, but are not tensity. The driving force for the particle motion was attributed to the limited to, the storage and displacement of a small volume of liquid acoustic radiation pressure exerted on the part of the particle surface without any leakage, detection of water pollution, gas sensing, gas-li- protruding into air. The revisable opening and closing of coating par- quid reactions and microreactors for chemical and biological purposes. ticles through acoustic levitation opens up a new possibility for non- ferromagnetic LMs to work as microreactors. 3.1. Microreservoirs 2.4.6. Self-propulsion ff The autonomous locomotion (self-propulsion) of a LM is of great LMs have been e ectively used as microreservoirs for liquid storage, ff interest for its use as the microreactor in DMF, which is attributed to the o ering controlled manipulation and transportation of small quantities – associated chaotic mixing inside the marble. Bormashenko et al. [95] of liquid [17, 27 30, 50, 54, 55, 57]. Zhao et al. [35] pointed out that a reported the self-propulsion of LMs filled with aqueous alcohol solu- magnetic LM possesses a unique feature to encapsulate samples/re- tions and deposited on a water surface. This phenomenon was related to agents after a chemical reaction or analysis by hardening the particulate the Marangoni solutocapillary flow caused by the condensation of al- shell. When the shell of magnetic particles contains low melting point cohol that evaporated from the LMs on a water surface. The Marangoni materials, such as wax, the LM shows a great encapsulation ability to flow in turn enhanced the evaporation of alcohol from the marbles. The form a rigid cover after heating, which facilitates the storage of the non- authors found that adding alcohol to the water supporting surface volatile ingredients to a certain extent. This characteristic can be ap- suppressed the self-propulsion motion. The velocity of the mass centre plied in some biochemical assays that are required to preserve the of the marbles increased with increasing alcohol concentration within samples/reagents after analysis. For instance, for follow-up assessment, the marbles, and was independent of the volume of LMs. The propulsion biobanking and archiving. However, it is noteworthy that chemicals of LMs was finally halted by viscous drag. Interestingly, the self-pro- that are sensitive to heat or salt concentration (e.g., enzymes) are not pulsion motion was not observed when LMs were placed on a glycerol suitable for the encapsulation and storage described above. Besides, the surface. Furthermore, Ooi et al. [41] utilized LMs containing aqueous rigid shell cannot completely prevent the liquid inside the marble from ethanol solutions to generate the Marangoni solutocapillary effect. They evaporating due to the presence of small voids in the particle shell that studied qualitatively the operating conditions of these LMs in the au- are permeable to air. tonomous motion. They also derived the scaling laws relating the dynamic parameters of the motion to the physical properties of the se- 3.2. Micropumps parate system such as the effective surface tension of the marble, the viscosity and the density of the supporting liquid, the coefficient of Bormashenko et al. [77] presented a micropump based on a pair of diffusion of the ethanol vapour, the geometrical parameters of the LMs coated with various powders and connected with a capillary PE marble, the speed, the trajectory and the lifetime of the autonomous tube, Fig. 8. The working principle is based on the difference of the motion. Laplace pressures between the marbles coated with different powders, which allows liquid to be transported from one marble to another until 2.5. Summary the pressures are equalized. They also developed a similar micropump based on a pair of marbles coated with the same powder but with The various manipulation schemes of LMs ensure that LM-based various initial radii.

92 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

3.3. Microsensors

LMs have been innovatively used as microsensors for the quick detection of water pollution and gas emission in chemistry analysis. Bormashenko and Musin [93] proposed the idea of using floating LMs to detect water surface pollution. In their experiment, water marbles coated with hydrophobic PVDF particles floated stably on the water surface, but were destroyed when the water surfaces were con- taminated with silicon oil and kerosene. They also investigated the floating of marbles on ethyl alcohol/water mixtures and various organic and non-organic liquids. As a result, LMs could effectively be used for the detection of water pollution with contaminants that decrease the surface tension, such as oils and petroleum, with the naked eye rather than water sampling. The porous hydrophobic particle layer on a LM keeps the marble separate from the surrounding surfaces, but allows the transportation of gas or vapour. Therefore, this characteristic enables LMs to sense gas or emit gas [60, 87, 96, 97]. Tian et al. [96] proved LMs loaded with different indicators could simultaneously sense different gases or vapours via different mechanisms. When a LM was formed with an ap- propriate indicator solution and placed onto the surface of a gas- or vapour-emitting solid or liquid, the indicator solution was intimately close to the gas- or vapour-emitting source while having no direct contact with the source. Conversely, LMs loaded with water dissolvable gases could be used as a gas emitter. The experimental results implied that different detection mechanisms (e.g. colorimetry and fluorescence) could be employed to carry out simultaneous detections of different vapours or gases independently. They further presented an application Fig. 9. Schematic of the LM-based microbioreactor for blood-type identifica- of gas sensing using LMs in the printing industry and explored the tion. (a) A blood droplet (10 μL) is placed on a hydrophobic precipitated cal- potential to use LM for quantitative gas sensing by paper-based col- cium carbonate powder bed to form the corresponding blood marble. (b) A orimetric measurements. They also mentioned porous LM shell offers droplet of antibody solution (10 μL) is then injected inside the blood marble to possibilities for gas-liquid reactions in another study [60]. In addition, complete the preparation of the microbioreactor. (c) When the corresponding LMs with particulate coating particles have also been used to sense antigens are not present on the surface of red blood cells, no separation is changes in temperature [84, 85], pH [88, 90, 98] and irradiation [79, visible. (d) When the corresponding antigens are present, red blood-cell ag- 92, 99] of the surrounding environment through different regimes. glutination reaction will take place, which will result in the separation of marble colour into two distinct light (top) and dark (bottom) parts. (For interpretation of the references to colour in this figure legend, the reader is re- 3.4. Microreactors ferred to the web version of this article.) As one of the most important microfluidic components, the microreactor plays an essential role in practical applications of LMs involving inside from coming into contact with external surfaces, but allows gases LM coalescence [24] for various chemical [100–102] and biological to transport freely across the shell. Recently, Vadivelu et al. [108, 109] assays [58, 59, 80–82, 103–109]. Arbatan et al. [58] selected human have demonstrated the use of LMs as a three-dimensional cell culture blood grouping (ABO and Rh) as the biological system to demonstrate platform for various cells, such as olfactory ensheathing cells, which the use of LM as a microbioreactor in practical diagnosis involving mainly relied on LM coalescence In addition, Lin's research team [50, human blood, Fig. 9. The authors summarized several significant ad- 59] investigated the optical detection of the magnetic LMs for possible vantages of a LM-based microbioreactor as follows. First, LM requires application as a new discrete microfluidic system, which may form a relatively small amounts of samples and reagents. Second, LM reduces new platform technology to use the magnetic LM as a “smart” micro- biohazards, since the powder-encapsulated biological sample does not reactor for chemical analysis or exploring new chemical reactions, come into contact with the supporting substrate. Third, the bioreactors especially when precious, explosive or highly toxic reagents are used can be controlled by either coalescing marbles containing different and the products need to be stored without separation. Li et al. [107] reagents or by injecting different reagents into one marble. Fourth, demonstrated the transport of electric charge and energy between LMs, marbles are cost effective. The authors further exploited the possibility showing the potential of LMs to function as micro electrochemical re- that LMs work as efficient miniature bioreactors inoculated with Hep actors. G2 cells to culture cell spheroids in vitro [80]. The cell aggregation could be clearly observed after 24 h, while the formation of numerous 3.5. Other applications three dimensional cell aggregates was observed by the end of 10 days. Three major factors, which make LMs suitable for cell culture, are: (i) a In addition to the numerous studies on the function of LMs as mi- porous shell allow O2 and CO2 exchange between cell culture medium croreactors, Lin et al. [35] demonstrated the “on-line” detection of li- and surrounding environment; (ii) hydrophobic powder particles that quid components in magnetic LMs by using electrochemical and optical form the shell and provide a non-adhesive surface, which encourages approaches. These magnetic LMs have a unique magnetic opening the cells to suspend in the medium; and (iii) a confined liquid core feature. By partially opening the particle shell, electrochemical mea- volume that promotes effective contact between cells, hence leading to surement was carried out with a miniaturized three-electrode probe. better aggregation. Inspired by the previous work, Tian et al. [82] de- They also demonstrated that how to conduct the quantitative mea- monstrated the possibility of using LMs to build respirable micro-bio- surement of dopamine with this technique. In contrast, fully opened logical reactors to cultivate microorganisms, which is attributed to the magnetic LMs make it feasible to detect the optical absorbance of the porous and superhydrophobic shell of LMs that prevents the liquid core liquid in a transmission mode. With this optical method, a biological

93 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95 glucose assay was demonstrated. Considering all the mentioned fea- References tures of magnetic LMs, they have numerous applications in clinical diagnostics, biotechnology, chemical synthesis, and analytical chem- [1] A. Manz, N. Graber, H.á. Widmer, Miniaturized total chemical analysis systems: a istry. novel concept for chemical sensing, Sensors Actuators B Chem. 1 (1–6) (1990) 244–248. Further to the applications mentioned above, LMs can be also used [2] D.R. Reyes, et al., Micro total analysis systems. 1. Introduction, theory, and as miniature electric generators [75], microaccelerometers [76], mi- technology, Anal. Chem. 74 (12) (2002) 2623–2636. crocentrifuges and miniature localized viscometers [110], as well as [3] P.-A. Auroux, et al., Micro total analysis systems. 2. Analytical standard operations and applications, Anal. Chem. 74 (12) (2002) 2637–2652. being used in the fabrication of graphene/Ag nanocomposites [111] [4] T. Vilkner, D. Janasek, A. Manz, Micro total analysis systems. Recent develop- and colour pigments [112]. Another feasible application of LMs is in the ments, Anal. Chem. 76 (12) (2004) 3373–3386. pharmaceutical field, where they have been exploited for drug dis- [5] P.S. Dittrich, K. Tachikawa, A. Manz, Micro total analysis systems. Latest ad- – covery utilising hollow granules formed by LMs' evaporation vancements and trends, Anal. Chem. 78 (12) (2006) 3887 3908. [6] D. Janasek, J. Franzke, A. Manz, Scaling and the design of miniaturized chemical- [113–115]. Recently, there have been several systematic review papers analysis systems, Nature 442 (7101) (2006) 374–380. on the practical applications of LMs in chemical, biological and phar- [7] J. West, et al., Micro total analysis systems: latest achievements, Anal. Chem. 80 – maceutical assays [31, 116, 117]. (12) (2008) 4403 4419. [8] A. Arora, et al., Latest developments in micro total analysis systems, Anal. Chem. 82 (12) (2010) 4830–4847. fl 4. Conclusions and future perspectives [9] P. Lisowski, P.K. Zarzycki, Micro uidic paper-based analytical devices (muPADs) and micro total analysis systems (muTAS): development, applications and future trends, Chromatographia 76 (2013) 1201–1214. MicroTAS provide the possibility to create complete analytical mi- [10] S. Lee, S. Lee, Micro total analysis system (μ-TAS) in biotechnology, Appl. – crosystems by integrating diverse functional modules, such as sample Microbiol. Biotechnol. 64 (3) (2004) 289 299. [11] X. Xu, et al., Integration of electrochemistry in micro-total analysis systems for preparation, separation and detection, into a single chip-sized micro- biochemical assays: recent developments, Talanta 80 (1) (2009) 8–18. fabricated device. The central issue of microTAS is microfluidics. As an [12] C. Cakal, et al., Development of a micro-total analysis system (mu-TAS) for the essential component of DMF, LM provides more possibilities to determination of catecholamines, Anal. Bioanal. Chem. 398 (5) (2010) 1909–1917. microTAS applications. The various manipulation schemes of LMs en- [13] J.T. Connelly, et al., Micro-total analysis system for virus detection: microfluidic sure that LM-based digital microfluidic platforms can be applied in pre-concentration coupled to liposome-based detection, Anal. Bioanal. Chem. 402 different modules of microTAS, such as sample handling, reaction and (1) (2012) 315–323. [14] M.L. Kovarik, et al., Micro total analysis systems for cell biology and biochemical mixing at the microscale and sample detection. A LM maintains its in- assays, Anal. Chem. 84 (2) (2012) 516–540. tegrity and exhibits low friction on various carrier surfaces, enabling [15] M.L. Kovarik, et al., Micro total analysis systems: fundamental advances and ap- the LM to be actuated by external electric, magnetic, gravitational and plications in the laboratory, clinic, and field, Anal. Chem. 85 (2) (2013) 451–472. fi [16] C.T. Culbertson, et al., Micro total analysis systems: fundamental advances and acoustic elds. In addition, LMs can be manipulated via pressure gra- biological applications, Anal. Chem. 86 (1) (2014) 95–118. dient, temperature change, pH change, irradiation, surfactant con- [17] N.-T. Nguyen, et al., Recent advances and future perspectives on microfluidic li- centration and self-propulsion. In terms of practical applications of LMs quid handling, Micromachines 8 (6) (2017) 186. fl in microTAS, besides offering controlled liquid manipulation and [18] K. Choi, et al., Digital micro uidics, Annu Rev Anal Chem (Palo Alto, Calif) 5 (2012) 413–440. transportation, LMs have been extensively applied as micropumps, [19] S. Mashaghi, et al., Droplet microfluidics: a tool for biology, chemistry and na- microsensors and, most importantly, microreactors for various chemical notechnology, TrAC Trends Anal. Chem. 82 (2016) 118–125. fl and biological assays. Considering all the features, LMs are envisioned [20] H. Song, D.L. Chen, R.F. Ismagilov, Reactions in droplets in micro uidic channels, Angew. Chem. Int. Ed. Eng. 45 (44) (2006) 7336–7356. as a growing research area in microTAS. LM-based microTAS could be [21] S.Y. Teh, et al., Droplet microfluidics, Lab Chip 8 (2) (2008) 198–220. potentially used in chemical and biomedical applications, as the com- [22] B.H. Weigl, R.L. Bardell, C.R. Cabrera, Lab-on-a-chip for drug development, Adv. – mercial market of microfluidics was valued at 8.28 billion U.S. dollars Drug Deliv. Rev. 55 (3) (2003) 349 377. [23] X. Tao, K. Chakrabarty, Parallel scan-like test and multiple-defect diagnosis for (USD) in 2017 and is expected to grow at a compound annual growth digital microfluidic biochips, IEEE Trans. Biomed. Circuits Syst. 1 (2) (2007) rate of 22.6% during the forecast period (2018–2023), to reach to 27.91 148–158. billion USD by 2023 [118]. [24] J. Jin, et al., Coalescence processes of droplets and liquid marbles, Micromachines 8 (11) (2017) 336. Although there is a variety of studies on the fundamental physics, [25] C.H. Ooi, N.-T. Nguyen, Manipulation of liquid marbles, Microfluid. Nanofluid. 19 properties, manipulation and applications of LMs, many other research (3) (2015) 483–495. aspects of LMs are still not well understood. For instance, how to pro- [26] P. Aussillous, D. Quere, Liquid marbles, Nature 411 (6840) (2001) 924–927. [27] P. Aussillous, D. Quere, Properties of liquid marbles, Proc. R. Soc. London, Ser. A duce LMs reliably in a continuous manner is an open question to the Math. Phys. Eng. Sci. 462 (2067) (2006) 973–999. researchers in the field of DMF. The conventional rolling method of LM [28] G. McHale, et al., Electrowetting of nonwetting liquids and liquid marbles, fabrication is simple and inexpensive. However, it may produce a Langmuir 23 (2) (2007) 918–924. random aggregate of coating particles, which has a significant effect on [29] E. Bormashenko, Liquid marbles: properties and applications, Curr. Opin. Colloid Interface Sci. 16 (4) (2011) 266–271. the stability and manipulation of LMs. Another focus of future research [30] G. McHale, M.I. Newton, Liquid marbles: principles and applications, Soft Matter 7 is how to use LMs as miniature chambers to realize more functions in a (12) (2011) 5473. coherent process based on the concept of microTAS. Further, much less [31] G. McHale, M.I. Newton, Liquid marbles: topical context within soft matter and recent progress, Soft Matter 11 (13) (2015) 2530–2546. is known about the dynamics of LM coalescence, which is crucial to the [32] M. Dandan, H.Y. Erbil, Evaporation rate of graphite liquid marbles: comparison wide use of LM-based microreactors and micromixers in fields of with water droplets, Langmuir 25 (14) (2009) 8362–8367. chemistry and biotechnology analysis. [33] C.H. Ooi, et al., Evaporation of ethanol-water binary mixture sessile liquid marbles, Langmuir 32 (24) (2016) 6097–6104. [34] A. Tosun, H.Y. Erbil, Evaporation rate of PTFE liquid marbles, Appl. Surf. Sci. 256 – Conflicts of interest (5) (2009) 1278 1283. [35] Y. Zhao, et al., Magnetic liquid marbles: toward “lab in a droplet”, Adv. Funct. Mater. 25 (3) (2015) 437–444. The authors declare no con flict of interest. [36] P.M. Ireland, et al., Electrostatic formation of liquid marbles - influence of drop and particle size, Powder Technol. 303 (2016) 55–58. [37] P.S. Bhosale, M.V. Panchagnula, Sweating liquid micro-marbles: dropwise con- Acknowledgments densation on hydrophobic nanoparticulate materials, Langmuir 28 (42) (2012) 14860–14866. fl ffi [38] E. Bormashenko, et al., On the mechanism of oating and sliding of liquid marbles, We acknowledge Gri th University for the International ChemPhysChem 10 (4) (2009) 654–656. Postgraduate Research Scholarship to Jing Jin and the Australian [39] U. Cengiz, H.Y. Erbil, The lifetime of floating liquid marbles: the influence of ff Research Council for grant support DP170100277 to Nam-Trung particle size and e ective surface tension, Soft Matter 9 (37) (2013) 8980. [40] C.H. Ooi, et al., Floating mechanism of a small liquid marble, Sci. Rep. 6 (2016) Nguyen.

94 J. Jin, N.-T. Nguyen Microelectronic Engineering 197 (2018) 87–95

21777. bioreactor, Adv. Healthc. Mater. 4 (1) (2015) 77–86. [41] C.H. Ooi, et al., A floating self-propelling liquid marble containing aqueous [82] J. Tian, et al., Respirable liquid marble for the cultivation of microorganisms, ethanol solutions, RSC Adv. 5 (122) (2015) 101006–101012. Colloids Surf. B: Biointerfaces 106 (2013) 187–190. [42] P. McEleney, et al., Liquid marble formation using hydrophobic powders, Chem. [83] D. Zang, et al., Tunable shape transformation of freezing liquid water marbles, Soft Eng. J. 147 (2–3) (2009) 373–382. Matter 10 (9) (2014) 1309–1314. [43] D. Zang, et al., Effect of particle hydrophobicity on the properties of liquid water [84] A. Hashmi, A. Strauss, J. Xu, Freezing of a liquid marble, Langmuir 28 (28) (2012) marbles, Soft Matter 9 (20) (2013) 5067–5073. 10324–10328. [44] E. Bormashenko, et al., Elastic properties of liquid marbles, Colloid Polym. Sci. [85] S.-i. Yusa, et al., Thermo-responsive liquid marbles, Polym. J. 46 (3) (2013) 293 (8) (2015) 2157–2164. 145–148. [45] S. Asare-Asher, J.N. Connor, R. Sedev, Elasticity of liquid marbles, J. Colloid [86] D. Dupin, S.P. Armes, S. Fujii, Stimulus-responsive liquid marbles, J. Am. Chem. Interface Sci. 449 (2015) 341–346. Soc. 131 (15) (2009) 5386–5387. [46] Z. Liu, et al., Mechanical compression to characterize the robustness of liquid [87] S. Fujii, et al., pH-responsive hairy particles synthesized by dispersion poly- marbles, Langmuir 31 (41) (2015) 11236–11242. merization with a macroinitiator as an inistab and their use as a gas-sensitive [47] J.R. Dorvee, et al., Manipulation of liquid droplets using amphiphilic, magnetic liquid marble stabilizer, Macromolecules 45 (6) (2012) 2863–2873. one-dimensional photonic crystal chaperones, Nat. Mater. 3 (12) (2004) 896–899. [88] S. Fujii, et al., pH-responsive liquid marbles stabilized with poly(2-vinylpyridine) [48] C. Planchette, et al., Coalescence of armored interface under impact, Phys. Fluids particles, Soft Matter 6 (3) (2010) 635–640. 25 (4) (2013) 042104. [89] S. Fujii, et al., Liquid marbles prepared from pH-responsive sterically stabilized [49] J. Jin, et al., Liquid marble coalescence via vertical collision, Soft Matter 14 (20) latex particles, Langmuir 27 (13) (2011) 8067–8074. (2018) 4160–4168. [90] M. Inoue, et al., pH-responsive disruption of ‘liquid marbles’ prepared from water [50] Y. Zhao, et al., Magnetic liquid marbles, their manipulation and application in and poly(6-(acrylamido) hexanoic acid)-grafted silica particles, Polym. J. 43 (9) optical probing, Microfluid. Nanofluid. 13 (4) (2012) 555–564. (2011) 778–784. [51] E. Bormashenko, et al., Electrically deformable liquid marbles, J. Adhes. Sci. [91] T.T. Tan, et al., Photoresponsive liquid marbles and dry water, Langmuir 30 (12) Technol. 25 (12) (2012) 1371–1377. (2014) 3448–3454. [52] E. Bormashenko, et al., Composite non-stick droplets and their actuation with [92] M. Paven, et al., Light-driven delivery and release of materials using liquid mar- electric field, Appl. Phys. Lett. 100 (15) (2012) 151601. bles, Adv. Funct. Mater. 26 (19) (2016) 3199–3206. [53] M.K. Khaw, et al., Digital microfluidics with a magnetically actuated floating li- [93] E. Bormashenko, A. Musin, Revealing of water surface pollution with liquid quid marble, Lab Chip 16 (12) (2016) 2211–2218. marbles, Appl. Surf. Sci. 255 (12) (2009) 6429–6431. [54] N. Kavokine, et al., Light-driven transport of a liquid marble with and against [94] D. Zang, et al., Switchable opening and closing of a liquid marble via ultrasonic surface flows, Angew. Chem. Int. Ed. Eng. 55 (37) (2016) 11183–11187. levitation, Langmuir 31 (42) (2015) 11502–11507. [55] Y. Zhao, et al., Magnetic liquid marbles: manipulation of liquid droplets using [95] E. Bormashenko, et al., Self-propulsion of liquid marbles: Leidenfrost-like levita-

highly hydrophobic Fe3O4 nanoparticles, Adv. Mater. 22 (6) (2010) 707–710. tion driven by Marangoni flow, J. Phys. Chem. C 119 (18) (2015) 9910–9915. [56] E. Bormashenko, et al., New investigations on ferrofluidics: ferrofluidic marbles [96] J. Tian, et al., Liquid marble for gas sensing, Chem. Commun. (Camb.) 46 (26) and magnetic-field-driven drops on superhydrophobic surfaces, Langmuir 24 (21) (2010) 4734–4736. (2008) 12119–12122. [97] Z. Xu, et al., Multi-responsive Janus liquid marbles: the effect of temperature and [57] M.I. Newton, et al., Electrowetting of liquid marbles, J. Phys. D. Appl. Phys. 40 (1) acidic/basic vapors, Part. Part. Syst. Charact. 31 (8) (2014) 839–842. (2007) 20–24. [98] S. Ohno, et al., pH-responsive liquid marbles prepared using fluorinated fatty acid, [58] T. Arbatan, et al., Liquid marbles as micro-bioreactors for rapid blood typing, Adv. Chem. Lett. 45 (5) (2016) 547–549. Healthc. Mater. 1 (1) (2012) 80–83. [99] K. Nakai, et al., Near-infrared-responsive liquid marbles stabilized with carbon [59] Y. Xue, et al., Magnetic liquid marbles: a “precise” miniature reactor, Adv. Mater. nanotubes, Chem. Lett. 42 (7) (2013) 719–721. 22 (43) (2010) 4814–4818. [100] Y.E. Miao, et al., Catalytic liquid marbles: Ag nanowire-based miniature reactors [60] J. Tian, et al., Porous liquid marble shell offers possibilities for gas detection and for highly efficient degradation of methylene blue, Chem. Commun. (Camb.) 50 gas reactions, Chem. Eng. J. 165 (1) (2010) 347–353. (44) (2014) 5923–5926. [61] E. Bormashenko, et al., Shaped composite liquid marbles, J. Colloid Interface Sci. [101] E. Sato, et al., Liquid marbles as a micro-reactor for efficient radical alternating 417 (2014) 206–209. copolymerization of diene monomer and oxygen, Chem. Commun. (Camb.) 51 [62] E. Bormashenko, et al., Janus droplets: liquid marbles coated with dielectric/ (97) (2015) 17241–17244. semiconductor particles, Langmuir 27 (1) (2011) 7–10. [102] Y. Sheng, et al., Silica-based liquid marbles as microreactors for the silver mirror [63] Z. Liu, et al., Coalescence of electrically charged liquid marbles, Soft Matter 13 (1) reaction, Angew. Chem. Int. Ed. Eng. 54 (24) (2015) 7012–7017. (2017) 119–124. [103] E. Bormashenko, R. Balter, D. Aurbach, Use of liquid marbles as micro-reactors, [64] G. McHale, et al., Levitation-free vibrated droplets: resonant oscillations of liquid Int. J. Chem. React. Eng. 9 (1) (2011). marbles, Langmuir 25 (1) (2008) 529–533. [104] F. Sarvi, et al., A novel technique for the formation of embryoid bodies inside [65] T.P. Hunt, D. Issadore, R.M. Westervelt, Integrated circuit/microfluidic chip to liquid marbles, RSC Adv. 3 (34) (2013) 14501. programmably trap and move cells and droplets with dielectrophoresis, Lab Chip 8 [105] M.C. Serrano, et al., Mammalian cell cryopreservation by using liquid marbles, (1) (2008) 81–87. ACS Appl. Mater. Interfaces 7 (6) (2015) 3854–3860. [66] T.P. Hunt, R.M. Westervelt, Dielectrophoresis tweezers for single cell manipula- [106] D. Wang, et al., Liquid marbles based on magnetic upconversion nanoparticles as tion, Biomed. Microdevices 8 (3) (2006) 227–230. magnetically and optically responsive miniature reactors for photocatalysis and [67] X.-B. Wang, et al., Dielectrophoretic manipulation of cells with spiral electrodes, photodynamic therapy, Angew. Chem. Int. Ed. Eng. 55 (36) (2016) 10795–10799. Biophys. J. 72 (4) (1997) 1887–1899. [107] M. Li, et al., Charge transport between liquid marbles, Chem. Eng. Sci. 97 (2013) [68] S.K. Fan, T.H. Hsieh, D.Y. Lin, General digital microfluidic platform manipulating 337–343. dielectric and conductive droplets by dielectrophoresis and electrowetting, Lab [108] R.K. Vadivelu, et al., Liquid marbles as bioreactors for the study of three-dimen- Chip 9 (9) (2009) 1236–1242. sional cell interactions, Biomed. Microdevices 19 (2) (2017) 31. [69] P. Singh, N. Aubry, Transport and deformation of droplets in a microdevice using [109] R.K. Vadivelu, et al., Generation of three-dimensional multiple spheroid model of dielectrophoresis, Electrophoresis 28 (4) (2007) 644–657. olfactory ensheathing cells using floating liquid marbles, Sci. Rep. 5 (2015) 15083. [70] N.-T. Nguyen, Micro-magnetofluidics: interactions between magnetism and fluid [110] X. Han, et al., Spinning liquid marble and its dual applications as microcentrifuge flow on the microscale, Microfluid. Nanofluid. 12 (1–4) (2011) 1–16. and miniature localized viscometer, ACS Appl. Mater. Interfaces 8 (36) (2016) [71] N.T. Nguyen, Deformation of ferrofluid marbles in the presence of a permanent 23941–23946. magnet, Langmuir 29 (45) (2013) 13982–13989. [111] Y. Chu, Z. Wang, Q. Pan, Constructing robust liquid marbles for miniaturized [72] C.H. Ooi, et al., Measuring the coefficient of friction of a small floating liquid synthesis of graphene/Ag nanocomposite, ACS Appl. Mater. Interfaces 6 (11) marble, Sci. Rep. 6 (2016) 38346. (2014) 8378–8386. [73] L. Zhang, D. Cha, P. Wang, Remotely controllable liquid marbles, Adv. Mater. 24 [112] H. Gu, et al., Non-iridescent structural color pigments from liquid marbles, J. (35) (2012) 4756–4760. Mater. Chem. C 3 (26) (2015) 6607–6612. [74] P. Aussillous, D. Quere, Shapes of rolling liquid drops, J. Fluid Mech. 512 (2004). [113] N. Eshtiaghi, K.P. Hapgood, A quantitative framework for the formation of liquid [75] Z. Yang, E. Halvorsen, T. Dong, Power generation from conductive droplet sliding marbles and hollow granules from hydrophobic powders, Powder Technol. 223 on electret film, Appl. Phys. Lett. 100 (21) (2012) 213905. (2012) 65–76. [76] H. Zeng, Y. Zhao, Dynamic behavior of a liquid marble based accelerometer, Appl. [114] N. Eshtiaghi, J.J.S. Liu, K.P. Hapgood, Formation of hollow granules from liquid Phys. Lett. 96 (11) (2010) 114104. marbles: small scale experiments, Powder Technol. 197 (3) (2010) 184–195. [77] E. Bormashenko, R. Balter, D. Aurbach, Micropump based on liquid marbles, Appl. [115] K.P. Hapgood, B. Khanmohammadi, Granulation of hydrophobic powders, Powder Phys. Lett. 97 (9) (2010) 091908. Technol. 189 (2) (2009) 253–262. [78] M.D. Doganci, et al., Liquid marbles stabilized by graphite particles from aqueous [116] E. Bormashenko, Liquid marbles, elastic nonstick droplets: from minireactors to surfactant solutions, Colloids Surf. A Physicochem. Eng. Asp. 384 (1–3) (2011) self-propulsion, Langmuir 33 (3) (2017) 663–669. 417–426. [117] N.M. Oliveira, R.L. Reis, J.F. Mano, The potential of liquid marbles for biomedical [79] K. Nakai, et al., Ultraviolet-light-responsive Liquid Marbles, Chem. Lett. 42 (6) applications: a critical review, Adv. Healthc. Mater. 6 (19) (2017) 1700192. (2013) 586–588. [118] https://www.marketsandmarkets.com/Market-Reports/microfluidics-market- [80] T. Arbatan, et al., Tumor inside a pearl drop, Adv. Healthc. Mater. 1 (4) (2012) 1305.html?gclid=Cj0KCQjw3InYBRCLARIsAG6bfMT- 467–469. aL1aH0aYbk3wXtTQ9sYKVoOuech-8VdrssBHV6HJGY60O6BojEMaArbYEALw_ [81] F. Sarvi, et al., Cardiogenesis of embryonic stem cells with liquid marble micro- wcB.