Hydrophilic hard-magnetic soft robots: A new approach for precise droplet manipulation Open Access

Droplet manipulations such as droplet generation,^1–3 transport,^4,5 splitting,^6,7 and merger⁸ are widely used in biochemical analysis,⁹ material synthesis,¹⁰ and biomedicine.^11–13 Various methods for droplet manipulation with the aid of external stimuli have been developed, including the use of electric fields,^14,15 magnetic fields,^16,17 acoustic waves,¹⁸ light,^19,20 and mechanical vibrations.²¹ Among these, droplet manipulation using magnetic fields is advantageous in terms of contactless control, fast and reversible response, flexible adjustment of the manipulation mode, and good biocompatibility.²² In particular, magnetic fields can be utilized to manipulate discrete microdroplets on an open surface, which allows greater flexibility and lower device costs.^23,24

Depending on the position of the magnetic material with respect to the droplets, the methods of magnetic droplet manipulation can be classified into two categories. The first category involves the addition of magnetic materials^25,26 such as magnetic particles or beads to the droplets. These magnetic materials, driven by the magnetic force, then exert an adhesive force on the droplets to realize droplet manipulation. This method is simple and inexpensive, and the surface of the magnetic particles can be chemically modified.²⁷ However, the driving force is weak because the dispersed magnetic particles do not all converge to the contact line in front of the droplet.²³ The droplet velocity and volume manipulated by these magnetic techniques are also restricted, owing to the low magnetic force exerted on magnetic particles and beads. In addition, the manipulated droplets are prone to become contaminated owing to residues and corrosion of the magnetic particles. Another approach is based on magnetically responsive soft materials composed of magnetic particles and a flexible substrate. These magnetic soft materials manipulate droplets by tuning the wettability of the substrate to the droplets through magnetically controlled deformation.^28,29 Such magnetic soft materials can be magnetically programmed by structural design and magnetization, leading to reconfigurable morphology and wettability. Moreover, they can be made chemically stable by encapsulating the magnetic particles in a chemically inert soft matrix. However, the structural design and preparation of such materials are complicated, and the droplet manipulation is less flexible.

Magnetic soft robots based on magnetically responsive soft materials have attracted widespread attention because of their low stiffness, resembling that of biological tissues, good biocompatibility, and excellent magnetic responsiveness.^30–32 Magnetic soft robots have been applied to targeted drug delivery³³ and minimally invasive surgery.^34,35 In particular, the magnetic programmability of hard-magnetic soft robots (HMSRs) allows them to achieve complex magnetically controlled deformations and precise targeting motions.^36,37 However, since the soft matrices used for such hard-magnetic soft robots are usually polymers with low surface energy, their surface hydrophobicity restricts the range of their applications in droplet capture and manipulation.

Here, we present a hydrophilic HMSR for precise droplet manipulation. This HMSR with a porous structure was constructed by incorporating NdFeB particles and sacrificial sugar particles into Ecoflex elastomers. By performing oxygen plasma treatment on the HMSR, this magnetic robot with a high specific surface area became hydrophilic. Leveraging robust magnetism and hydrophilicity, the HMSR is able to achieve precise droplet capture and transport in a 20 mT magnetic field. In particular, it exhibits three droplet manipulation modes as the magnet velocity increases: droplet transport, droplet splitting, and robot–magnet detachment. Theoretical analysis and experimental results reveal that the critical speed of the HMSR required for droplet transport and splitting is inversely proportional to the droplet volume. Notably, it is demonstrated that the HMSR can transport a 50 μl droplet at a maximum velocity of 200 mm/s and a maximum droplet volume of 900 μl at a maximum velocity of 10 mm/s. Additionally, benefiting from its chemical stability, the HMSR is able to realize precise reaction manipulation of acid, phenolphthalein, and alkaline droplets. Owing to its commendable biocompatibility and droplet adhesion to microparticles, the HMSR also realizes efficient removal of microparticles. This HMSR with high-performance droplet manipulation, good chemical stability and biocompatibility shows great potential for applications in material synthesis, biochemical analysis, and clinical medicine.

HMSRs with dimensions of ϕ5 × 6 mm² were fabricated using a molding method similar to that described in our previous paper.³⁸ Specifically, NdFeB particles with a size of 5 μm (MQFP-B, Magnequench, China), sugar particles with a size of 800–1100 μm (Angie’s Yeast Co., Ltd., China), and Ecoflex 00-30 (Smooth-on Inc., USA) were mixed in a mass ratio of 0.3:0.2:1. The magnetic mixture was poured into a mold and placed on a baking sheet at 85 °C (EH20B, Beijing LabTech Instruments Co., Ltd., China) for 2 h to cure. The cured magnetic mixture could be peeled off from the mold. The cured mixture was then immersed in water and manually squeezed to allow the water to penetrate into the substrate, so that the sugar particles were completely dissolved by water. The porous HMSR thus obtained was then placed in a plasma cleaner (PC-AG-5L, Shuang Microelectronics Technology Co., Ltd., China) and treated with a power of 80 W and an oxygen flow rate of 50 sccm for 7 min. Finally, the HMSR was placed in a strong pulsed magnetic field (1200 V, ME-12150, Magele Technology Co. Ltd., China) for magnetization to rearrange the magnetic moments of its internal magnetic particles.

To make all inner surfaces of the HMSR exhibit hydrophilicity, the plasma treatment time was optimized. In these experiments, the uniformity and infiltration depth of colored ink were examined to characterize the effect of the plasma treatment. To visualize the ink infiltration, transparent Ecoflex foam without magnetic particles was used. Specifically, an ink droplet was added to the surfaces of untreated Ecoflex foam and of Ecoflex foam that had been plasma-treated for 1, 3, 5, and 7 min, respectively. The change in contact angle of deionized water on the HMSR surface before and after 7 min of plasma treatment was tested using a contact angle meter (XG-CAMC, Shanghai Xuanzhun Instrument Co., Ltd., China).

The rolling of the HMSR was tested on the surfaces of polystyrene (PS) sheets, printing paper, and sandpaper, the surface roughnesses of which were about 0.1, 3, and 60 μm, respectively. A permanent magnet (ϕ10 × 2 mm², Suiling Trading Co., Ltd., Quanzhou, China) moving parallel to the substrate was used to drive the HMSR. In this case, the magnetic field strength at the bottom of the HMSR was 20 mT.

To test the droplet capture and transport ability of the HMSR, three droplets (20 μl) containing NaOH solution (0.5M) and phenolphthalein solution were randomly added to the PS substrate. After plasma treatment, the permanent magnet was used to drive the HMSR to move on the PS substrate. The HMSR sequentially moved to the location of the three droplets, and then captured and transported them. The contact angle between the deionized water and the PS sheet was also tested by the contact angle meter.

To characterize the droplet manipulation mode of the HMSR, the behaviors of HMSRs (ϕ5 × 6 mm²) carrying varying volumes of droplets were tested at different speeds of magnet motion. Specifically, 50–1000 μl of deionized water was added to the HMSRs. The magnet was driven at a velocity of 1–800 mm/s. A digital single lens reflex camera (EOS 90D, Canon, Japan) was used to record the movements of HMSR, droplets, and magnet. Red ink was added to the deionized water for visualization.

Three reagents, namely, 100 μl NaOH solution (1 mM), 200 μl phenolphthalein solution, and 300 μl HCl (pH 3.5) were added to the PS substrate. A permanent magnet was used to actuate the HMSR. The HMSR first moved to the NaOH solution droplet. Owing to the high hydrophilicity of the HMSR, it quickly captured the NaOH droplet and subsequently transported it to the phenolphthalein solution droplet, where chemical reaction occurred. At the end of the reaction, the speed of magnet motion was increased to realize sampling of the reaction products. The HMSR then moved to the HCl droplet and transported the HCl droplet in the same way to the location of the NaOH and phenolphthalein droplet to participate in chemical reaction. During the reaction, the HMSR was actuated by the magnet to roll inside the droplet to accelerate the mixing of reactants and speed up the reaction.

Dozens of sugar particles with a size of about 500 μm were randomly placed on the PS substrate. Then, 50 μl of deionized water was added to the HMSR to adhere sugar particles. A permanent magnet was used to actuate the robot to sequentially transport the deionized water to the location of each sugar particle, adhering and transporting the microparticles for removal.

It has been demonstrated³⁸ that hard-magnetic elastomer foam exhibits flexible magnetically controlled deformation and good chemical stability. Moreover, the foam has a large specific surface area and strong adhesion ability to liquids. Therefore, as described in detail in Sec. II A, a porous HMSR for droplet manipulation was prepared based on this hard-magnetic elastomer foam.³⁸ As shown in Fig. 1(a), NdFeB particles and sacrificial template sugar particles were added to Ecoflex liquid. Then, the composition that exhibited the optimal magnetic response according to the literature³⁸ was chosen. Cylindrical magnetic robots (ϕ5 × 6 mm²) were fabricated by pouring the magnetic mixture into the mold and then heated for curing. The cured magnetic robots were peeled off from the molds, and then rinsed with water to remove sugar particles. Thus, a porous HMSR was obtained.

Good wettability between the HMSR and the droplet is essential to enhance adhesion to the droplet for effective droplet manipulation. Plasma treatment is an effective approach to make hydrophobic an Ecoflex surface hydrophilic.^39,40 As described in Sec. II A, oxygen plasma treatment was employed to modify the surface of the HMSR. The effect of plasma treatment time on the penetration depth and uniformity of droplets in the robots was investigated. The results showed that the contact angle of deionized water on the HMSR changed significantly from 115° to 0° after plasma treatment [Fig. S1(a), supplementary material]. Although an HMSR will return to hydrophobicity within 1 h after plasma treatment, immersing it in deionized water will allow the loaded droplets to quickly penetrate into the pores even after tens of days.⁴¹ Furthermore, enhancements in the penetration depth and uniformity of deionized water within the robot were observed with prolonged plasma treatment durations [Fig. S1(b), supplementary material]. The optimal wettability of droplets within the HMSR was achieved with a treatment duration of 7 min.

In contrast to soft-magnetic and superparamagnetic materials, the HMSR possesses high remanent magnetization and exhibits the same magnetic moment orientation as the magnetizing magnetic field after magnetization [Fig. 1(b)]. As shown in Fig. 1(c), a permanent magnet was used to generate an excitation magnetic field in the same direction as the magnetic moment orientation of the robot. When the magnet moved forward, the HMSR was subjected to a strong magnetic force F_m and thus rolled with the magnet. Figure 2(a) shows that HMSR could roll on surfaces with various roughnesses, including smooth PS sheet and coarse sandpaper. Although the magnetic field strength on the substrate surface was only 20 mT, the robot could move at a speed of 70 mm/s. This result demonstrated that the strong magnetic force and the rolling mode allowed the HMSR to overcome a large frictional force and move on surfaces of diverse roughness. Furthermore, benefiting from the precise magnetically actuated motion and the hydrophilicity, the HMSR was demonstrated to achieve targeted droplet capture and transport on the hydrophobic PS surface [the contact angle of which was 92°; see Fig. 2(b) and Video S1, supplementary material].

HMSR utilizes the difference in adhesion forces on a droplet exerted by the hydrophilic magnetic robot and the hydrophobic substrate to manipulate the droplet.^26,42 The HMSR-based droplet manipulation system consists of three parts: an HMSR, a permanent magnet, and a hydrophobic substrate. Driven by the magnetic force, HMSR rolls following the magnet and exerts an adhesion force on the droplet while rolling to the boundary of the droplet. Thus, the droplet is propelled to move forward against the resistance exerted by the substrate. To characterize the droplet manipulation modes of HMSR and obtain useful information for future applications, the effects of the velocity of magnet motion and loaded droplet volume on droplet behavior were tested. As depicted in Figs. 3(a) and 3(b), droplet manipulation by the HMSR exhibited three modes: droplet transport, droplet splitting, and robot–magnet detachment. When the magnet moved at a low speed, the HMSR followed it and transported the droplet in a stable manner. As the magnet velocity increased, the HMSR followed the magnet and broke through the gas–liquid interface of the droplet. In this situation, HMSR split a subdroplet from the parent droplet and subsequently transported it alongside the magnet. With further acceleration of the magnet to about 532 mm/s, the HMSR remained within the droplet and detached from the magnet.

As the magnetic velocity surpasses a certain threshold, the HMSR detaches from the magnet, which means that the droplet is uncontrollable by the magnet. Whether the HMSR can follow the magnet depends on the velocity of the HMSR at the end of the acceleration stage. When the maximum velocity of the HMSR equals or exceeds that of the magnet, the HMSR follows the magnet. Otherwise, the HMSR will become detached from the magnet. As indicated by Eq. (5) and confirmed by the findings shown in Fig. 3(a), the acceleration of the HMSR, a_m, is related to the characteristics of the exciting magnetic field and the HMSR itself, rather than to the droplet volume. Benefitting from the increased F_m and a_m compared with those of magnetic nanoparticles, this critical velocity of the HMSR was improved to 532 mm/s. This indicates that the HMSR could manipulate the droplet at a faster velocity.

These analyses and experiments provide operational guidance for droplet manipulation applications of HMSR. Benefiting from its strong remanent magnetization and adhesion to the droplets, the HMSR, with a volume of 0.12 cm³ and a mass of 0.04 g, could stably transport a maximum droplet volume of 900 μl and reach a maximum droplet velocity of 200 mm/s. Notably, this velocity surpasses the velocity of other magnetic manipulation techniques reported in the literature for transporting droplets of comparable volumes [Fig. 3(d)].^{23,26,42,44,47} Moreover, the actuated magnetic field intensity, only 20 mT, is also lower than that of these reported magnetic manipulation techniques. As indicated by Eq. (8), in addition to droplet volume and magnet velocity, the manipulation behavior of the HMSR is also influenced by the droplet surface tension and density, as well as the composition and surface characteristics of the hydrophobic substrate. These variables play an essential role in determining the droplet manipulation performance of the HMSR and can also be fine-tuned to enhance its efficacy in various applications. In particular, to further improve the droplet volume that the HMSR can transport and the velocity with which it can do so, a superhydrophobic substrate can be used. Such a substrate will decrease the adhesion resistance by reducing the contact angle hysteresis.^48–50 However, ferromagnetic materials with high permeability are not recommended as substrates, because they cause magnetic field shielding of the HMSR, which reduces the driving force.

Precise, programmed, and contactless manipulation of microscale chemical reactions is fundamental for biochemical detection and material synthesis.^51–53 The precise magnetically controlled microdroplet manipulation of HMSRs enables them to manipulate microscale chemical reactions. Moreover, their chemical inertness³⁸ ensures manipulation of corrosive droplets without introducing contamination. As illustrated in Fig. 4, an HMSR was able to achieve sequential reaction manipulation of an alkaline droplet, a phenolphthalein droplet, and an acidic droplet (Video S2, supplementary material). Driven by a noncontact moving permanent magnet, the HMSR first moved to the NaOH droplet (100 μl), captured it, and pressed the TCL to exert an adhesive force on the droplet. Subsequently, the NaOH droplet was transported to the phenolphthalein droplet (200 μl), where it mixed and reacted with the phenolphthalein, resulting in a color change to purple. At the end of the reaction, the speed of motion of the magnet was increased to more than 70 mm/s for droplet splitting according to the test results shown in Fig. 3(a). This step achieved product sampling from the first reaction. The HMSR then transported this reaction product to a specified location, accelerated once more to split the droplet, and thus achieved targeted delivery of the product. Subsequently, the HMSR moved to the HCl droplet (300 μl) and transported it to the phenolphthalein droplet. During the reaction, the HMSR rolled inside the droplet to accelerate the reaction, causing the droplet color to change rapidly from purple to colorless. This experiment verified that an HMSR could precisely manipulate microscale chemical reactions, including reactant capture and transport, reaction control, and product sampling and precise delivery. Such a noncontact HMSR is particularly suitable for remote manipulation of microscale chemical reactions involving toxic materials.

Benefiting from low stiffness and good biocompatibility, tether-free magnetic soft robots offer a promising platform for biomedical applications such as drug delivery and in vivo surgery.³⁰ However, the passive cargo loading mechanisms of existing magnetic soft robots^54,55 hinder them from effectively loading substances in vivo, especially solid materials. Calculi are mineral deposits that are commonly found in the urinary system and typically require pharmaceutical or surgical intervention for removal.⁵⁶ Here, we demonstrate a more efficient approach to removing calculi utilizing HMSR. As shown in Fig. 5, an HMSR was able to accurately maneuver to the location of each microparticle. The HMSR that carried the droplets exerted an adhesive force on the contacted solid microparticles. As it moved, the microparticles were adhered, dissolved, and removed (Video S3, supplementary material). Since the HMSR is chemically inert, in addition to sugar particles that are soluble in water, particles that are soluble in reagents such as acids or bases can also be removed in the same way. Moreover, in the case of particles that are difficult to dissolve, the HMSR can still remove them by exerting an adhesive force on them. Thus, since the HMSR is not highly selective for particles that can be removed, this method can be used for a wide range of particle removal. In particular, the good biocompatibility and chemical inertness of HMSRs allows them to be potentially applied in minimally invasive surgical procedures.

This paper has introduced an HMSR designed for precise droplet manipulation. Hard-magnetic NdFeB particles were incorporated into the HMSR to enhance the magnetic force exerted on it. To augment the adhesion force exerted by the HMSR on droplets, a particle leaching method was used to create a porous structure, and plasma treatment was applied to hydrophilize the HMSR. Thus, this HMSR can precisely target droplets, exerting adhesive forces to effectively manipulate them. The HMSR exhibited three droplet manipulation modes as the magnet velocity increased: droplet transport, droplet splitting, and robot–magnet detachment. It was verified that the critical velocities of HMSR for droplet transport and droplet splitting were inversely proportional to the droplet volume. In particular, the HMSR could transport a 50 μl droplet at a maximum velocity of 200 mm/s in a 20 mT magnetic field. Moreover, the maximum droplet volume that could be transported by the HMSR was 900 μl at 10 mm/s. Leveraging its precise droplet manipulation and chemical inertness, the HMSR successfully manipulated an acid–alkali neutralization reaction and achieved product sampling. In addition, benefiting from the adhesion of droplets to microparticles, the HMSR can actively adsorb liquids owing to its hydrophilicity and can exhibit strong adhesion to microparticles. Thus, the HMSR was able to perform microparticle removal, a functionality valuable for in vivo calculus extraction. This HMSR with its precise droplet manipulation capability has promise for applications in the fields of biochemical detection and biomedicine.

The supplementary material consists of Video S1 illustrating droplet capture and transport, Video S2 illustrating manipulation of chemical reactions of acid and alkaline droplets, Video S3 illustrating microparticle removal, and Fig. S1 showing additional experimental results.

This work was supported by the Science and Technology Program from the State Grid Corporation of China (Grant No. 5700-202155453A-0-0-00): “Development of flexible liquid metal based micro-sensor with anti-electromagnetic interference ability for power engineering applications.”

The authors have no conflicts to disclose.

Xiao Sun: Conceptualization (equal); Formal analysis (equal); Methodology (equal); Visualization (equal); Writing – original draft (equal). Zhenming Li: Funding acquisition (equal); Investigation (equal); Resources (equal); Validation (equal). Chunwei Li: Formal analysis (equal); Methodology (equal). Huimin Zhang: Data curation (equal); Formal analysis (equal); Methodology (equal). Wei Liu: Formal analysis (supporting); Methodology (supporting). Mingyang Liu: Formal analysis (supporting); Methodology (supporting). Lei Li: Conceptualization (equal); Formal analysis (equal). Lin Gui: Conceptualization (equal); Funding acquisition (equal); Methodology (equal); Supervision (equal); Writing – review & editing (equal).

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

Xiao Sun is currently pursuing a Ph.D. degree at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, under the supervision of Professors Lin Gui and Lei Li. She received her Bachelor’s degree in Energy and Power Engineering from Central South University in 2019. Her research interests include hard-magnetic soft materials and their applications.

Lin Gui is a Professor at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. He is also a Professor at the University of Chinese Academy of Sciences. His current research interests include liquid-metal-based microfluidics, microdevices, microsensors, and MEMS, as well as implantable medical instrumentations.

Lei Li is a Professor at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. She received her Bachelor’s degree in Thermal Engineering from Tsinghua University in 2003 and then obtained her Ph.D. degree in Biomedical Engineering from Tsinghua University in 2009. She was a Visiting Scholar at the Massachusetts General Hospital and Harvard Medical School from 2015 to 2016. Her recent work lies at the intersection between micro/nanofabrication, tissue engineering, and flexible electronics.