Optical-driven miniature robots: driving mechanism, applications and future trends
Xiaowen Wang
*a,
Sen Jia
a,
Yingnan Gao
a,
Changyou Liu
a,
Yaping Wang
a,
Anqin Liu
a and
Wenguang Yang
b
aSchool of Mechanical and Electrical Engineering, Yantai Institute of Technology, Yantai 264005, China. E-mail: wangxiaowen@yitsd.edu.cn; jiasen@yitsd.edu.cn; gaoyingnan@yitsd.edu.cn; liuchangyou@yitsd.edu.cn; wangyaping@yitsd.edu.cn; liuanqin@yitsd.edu.cn
bSchool of Electromechanical and Automotive Engineering, Yantai University, Yantai 264005, China. E-mail: yangwenguang@ytu.edu.cn
First published on 29th July 2025
Abstract
Miniature robots can complete complex tasks at the micro-scale, which have shown great application potential in fields such as biomedicine and environmental monitoring. As a renewable energy source, light is widely used in energy and information transmission. With the maturity of beam modulation and optical microscope technology, optical-driven miniature robots have become a hot topic in the field of miniature robotics due to their programmable nature, high resolution, non-contact nature, high precision, and good biocompatibility. This review introduces the driving mechanism of optical-driven miniature robots, summarizes the progress in their driving control and application capabilities, and looks forward to their future development trends. This review hopes to provide new ideas for the development of optical-driven miniature robots and promote their technological progress and applications. In the future, the cross-disciplinary integration of disciplines will help the research and development of the next generation of miniature robots.
 Xiaowen Wang | Xiaowen Wang received her Bachelor's Degree from the School of Electromechanical and Automotive Engineering and her Master's Degree in Mechanical Engineering from Yantai University in Shandong Province, in 2021 and 2024, respectively. She has been teaching at Yantai Institute of Technology since June 2024. Her current research focuses on microfabrication and microrobots. |
 Sen Jia | Sen Jia received her Bachelor's Degree in Mechanical Design, Manufacturing and Automation from Yantai University and her Master's Degree in Mechanical Design, Manufacturing and Automation, School of Environmental and Material Engineering, China University of Petroleum (East China). She is currently teaching at Yantai Institute of Technology. Her current research focuses on Mechanical Design. |
 Yingnan Gao | Yingnan Gao received his Bachelor's Degree in Mechanical Design, Manufacturing and Automation and his Master's Degree in Mechanical Engineering from Yantai University in Shandong Province in 2020 and 2024, respectively. He has been teaching at Yantai Institute of Technology since July 2024. His current research focuses on bioprinting, and his interests include micromachining, tissue engineering, and nanorobot-based nanomanipulation. |
 Changyou Liu | Changyou Liu received his Bachelor's Degree from the School of Mechanical Design, Manufacturing and Automation and his Master's Degree in Mechanical Engineering from Yantai University, Shandong Province in 2011 and 2013, respectively. He has been teaching at Yantai Institute of Technology since 2023. His current research focuses on machine design. |
 Yaping Wang | Yaping Wang received her Bachelor's Degree in Mineral Processing from the School of Resources and Materials Engineering, Henan Polytechnic University, and her Master's Degree in Materials Science from the School of Environmental and Material Engineering, Yantai University. She is currently teaching at Yantai Institute of Technology. Her current research focuses on Intelligent Materials. |
 Anqin Liu | Anqin Liu received her Bachelor's Degree from the School of Mechanical Design, Manufacturing and Automation and her Master's Degree in Mechanical Engineering from Yantai University in Shandong Province in 2006 and 2009, respectively. She is teaching at Yantai Institute of Technology. Her current research focuses on microrobots. |
 Wenguang Yang | Wenguang Yang received his BS Degree in Measurement & Control Technology and Instrumentation from Yantai University, Shandong, China, and his PhD Degree from the State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China, in 2011 and 2018, respectively. Since January 2018, he has been at Yantai University. His current research interests include bio-printing, microfabrication, tissue engineering and nano-manipulation based on nano-robots. |
1. Introduction
Miniature robots refer to robotic systems with sizes ranging from centimeters to nanometers.1–3 Because of their tiny size, they can enter complex and small spaces such as human blood vessels and cell gaps to perform tasks such as drug delivery, disease diagnosis, and cell manipulation. Thus, they have broad application prospects in the biomedical,4–6 environmental monitoring,7–9 miniature manufacturing7 and other fields. Traditional miniature robot driving methods such as chemical and electric driving have problems such as limited energy supply, easy environment pollution, and difficulty in achieving high-precision control. In contrast, optical-driven action is a new type of driving method, which has significant advantages such as non-contact remote control, high spatial resolution, precise regulation, no pollution to the environment, and good biocompatibility, making optical-driven miniature robots a hot topic in recent years.As a clean and precisely modulated energy source, the wavelength, intensity, polarization state and other parameters of light can be flexibly regulated, which can provide a variety of driving forces for miniature robots.10–15 Through the rational design of the structure and materials of miniature robots, they can efficiently absorb light energy and convert it into mechanical energy or other forms of energy, thus achieving precise motion control. With the continuous progress of materials science, miniature manufacturing technology and optical technology, the performance of optical-driven miniature robots has been significantly improved, gradually developing from simple light-response movement to multi-function integration and intelligent control.
The second section of this article introduces the materials for manufacturing optical-driven miniature robots, and the third section introduces the driving mechanism of optical-driven miniature robots. In the fourth section, we introduce the application of optical-driven miniature robots in the fields of intelligent transportation, environmental protection, and biomedicine. Finally, the fifth section summarizes the challenges faced by optical-driven miniature robots and presents future prospects (Scheme 1). It is hoped that the driving mechanism described in this article will stimulate readers' interest and prompt everyone to conduct in-depth exploration of the manufacturing process, functional expansion and practical application of optical-driven miniature robots from a broader perspective.
 | ||
| Scheme 1 Driving mechanisms and applications of optical-driven miniature robots. | ||
2. Materials
2.1 Photothermal materials
Photothermal materials can absorb light energy and convert it into heat energy, which in turn drives miniature robots. Typical photothermal materials are divided into two types, namely metal nanomaterials and carbon-based materials. In recent years, MXene nanosheets have also received extensive research attention as photothermal materials (Table 1).| Materials | Biocompatibility | Fabrication costing | Driving threshold | Cyclic stability | Advantages | Disadvantages |
|---|---|---|---|---|---|---|
| LCE | Moderate | High | 50–100 mW cm−2 | 103–104 cycles (limited by thermal-oxidative aging) | - Fast light response speed | - Poor biocompatibility |
| - Large deformation and strong structural programmability | - High preparation cost | |||||
| - High mechanical strength | ||||||
| - Better weather resistance than hydrogels, can be used in dry environments for a long time | ||||||
| Hydrogel | High | Moderate | 20–50 mW cm−2 | 102–103 cycles (swelling-shrinkage fatigue) | - High biocompatibility | - Low mechanical strength |
| - High water content, suitable for applications in wet environments | - Slow response speed | |||||
| - Low preparation cost | - Poor salt resistance | |||||
| - Simple photocuring process | ||||||
| - Good flexibility | ||||||
| SMA | Moderate | High | 100–200 mW cm−2 | 104–105 cycles (phase transition fatigue) | - High driving force | - Small deformation |
| - High cycle stability | - Moderate biocompatibility, requires coating treatment | |||||
| - Good fatigue resistance | - High cost | |||||
| - Excellent corrosion resistance | ||||||
| Biomaterials | Extremely high | Moderate | 30–80 mW cm−2 | 101–102 cycles (degradation limit) | - Extremely high biosecurity | - Low light response efficiency |
| - Low immunogenicity | - Weak mechanical properties | |||||
| - Their light response wavelength depends on their modifying group | ||||||
| - Complex preparation process |
Its advantages are high light–heat conversion efficiency and fast response speed. However, its synthesis process is complex and its cost is relatively high.
Its advantage lies in the wide range of material sources, low cost, and good chemical stability. However, when building miniature robots with complex structures, their processing is more difficult.
2.2 Photocatalytic materials
Photocatalytic materials refer to semiconductor catalyst materials (Fig. 1B) required for photochemical reactions to occur. Photocatalytic reactions are based on the unique energy band structure of semiconductor materials. When light with an energy greater than or equal to the bandgap width of a semiconductor material illuminates its surface, electrons in its valence band absorb photon energy and transition to its conduction band, leaving holes in the valence band to form electron–hole pairs. These electrons and holes have a strong redox ability and can migrate to the surface of the material to react with the substances adsorbed on it (Fig. 1A). Electrons can reduce adsorbed oxidizing substances (such as oxygen), while holes can oxidize reducing substances (such as organic matter). The common photocatalytic materials include titanium dioxide (TiO2), zinc oxide (ZnO), and cadmium sulfide (CDs). | ||
| Fig. 1 (A) Photocatalytic reaction of semiconductors under light irradiation. (B) Electrical properties of metal, semiconductor and insulator. | ||
TiO2 is the most widely studied and mature photocatalytic material. It has the advantages of stable chemical properties, high catalytic activity, low cost, and non-toxic nature. However, the main limitation of TiO2 is that it has a wide bandgap (about 3.2 eV) and can only respond to ultraviolet (UV) light. ZnO is similar to TiO2 and has good chemical stability and photocatalytic activity. ZnO can be used to prepare structures with different morphologies (such as nano-rods and nano-flowers) by a variety of methods. However, ZnO exhibits a photoreceptor phenomenon, and its own structure may be damaged under light, resulting in a gradual decline in its photocatalytic performance. This is an urgent problem to be solved in its actual application.
The bandgap width of CDs is relatively narrow (about 2.4 eV), which can respond to visible light, greatly broadening the scope of photocatalytic materials to use the solar spectrum. However, the light stability of CDs is poor, and they are prone to light corrosion during photocatalytic processes, producing toxic cadmium ions, which cause potential harm to the environment and limit their large-scale application. To overcome these problems, methods for the surface modification of CDs or compounding them with other materials are often used, such as the preparation of composite materials of CDs and TIO2 to improve their optical stability and catalytic properties.21
In addition, bismuth-based photocatalytic materials, such as BiVO4 and Bi2WO6, also have good photocatalytic properties in the visible region because of their special crystal structure and energy band arrangement, showing potential application value in the field of environment and energy.22
2.3 Photoinduced deformation material
The phenomenon of photoinduced deformation is based on a specific light response mechanism inside a material. In the case of most photoinduced deformation materials, they possess photoactive groups or light-sensitive substances. When light of a specific wavelength and intensity hits these materials, their photoactive group absorbs photon energy, causing physical and chemical changes within or between molecules, resulting in changes in their internal stress distribution, which are macroscopically manifested as changes in the shape or size of these materials.The photothermal effect refers to the fact that when the photothermal switch is doped in an LCE material, light energy is converted into heat energy after absorbing light. The heat generated is lost to the LCE network, causing the local temperature of the material to increase, resulting in its deformation (Fig. 2B).
 | ||
| Fig. 2 (A) Schematic of the photomechanical effects. (B) Schematic of the photothermal effects. | ||
Photochemical reactions refer to the addition of groups in liquid crystal materials that can undergo photochemical reactions, such as azobenzene and spiropyran. Azobenzene is a typical class of photoresponsive groups, which can undergo reversible cis–trans isomerization reactions under light irradiation. The azo-phenyl groups introduced into the polymer main chain or side chain may be prepared via the azobenzene-based polymer light-induced deformation of the material. Under light, the azobenzene molecule changes from a trans structure to a cis structure, and the shape and volume of the molecule change, causing stress inside the polymer material and causing the material to deform. After the light disappeared, azobenzene reverted from cis to trans, and the material returned to its initial shape (Fig. 2A).
Spirobenzopyran compounds are a class of organic compounds with unique photoinduced deformation properties. Spirobenzopyran compounds undergo reversible photochromic reactions under irradiation of different wavelengths of light. Under UV light, the spiro-ring structure in their molecules will undergo a ring-opening reaction, changing from a colorless spiro-ring form to a colored anthocyanin form with a conjugated structure. This change in molecular structure can cause changes in the dipole moment, shape, and size of the molecule. When a large number of spirobenzopyran molecules undergo this photoisomerization, microscopic stress changes will occur inside the material. The accumulation and synergy of these microscopic stresses result in macroscopic deformation of the material. In the presence of visible light or under the effect of heat, the partial flower crystal form will be closed again and restored to the spiral ring form, and the material will be restored to its original shape, thereby realizing the reversible deformation of the optical drive.
Azobenzene and spirobenzopyran can produce fast and reversible responses to different wavelengths of light, and their light response is reversible. Thus, accurately controlling the wavelength of light and its exposure time can achieve precise regulation of material deformation. This has potential applications in miniature mechanical devices, optical control switches and other fields. However, long-term light exposure may cause material fatigue, which affects the service life. In addition, the colored anthocyanin form of spirobenzopyran compounds is usually thermodynamically unstable and is susceptible to fading or degradation due to heat, light and other factors. Also, certain environmental conditions may cause their photogenic deformation properties to gradually decrease.
The photogenic deformation material of LCEs has a large deformation amplitude and a fast response speed. However, the process for the preparation of these materials is very complex and sensitive to environmental conditions, which limits their large-scale application.
Photoisomerized hydrogels contain photoisomerized groups, such as azobenzene. Under irradiation with different wavelengths of light, the azophthalene group will undergo cis–trans isomerization. This change in molecular structure will cause changes in the internal stress of the hydrogel network, which will cause macroscopic shape deformation. For example, in the azo-phenyl-containing groups of a polymer with another hydrogel monomer copolymer, prepared via light isomerization of the hydrogel under UV irradiation, the azo-phenyl groups transform from the trans to the cis isomer, resulting in hydrogel shrinkage deformation; upon switching to visible light irradiation, azobenzene is restored to the trans isomer, and the hydrogel is gradually restored. Luo et al. used polypyrrole nanoparticles as a photothermal transducer to form a photothermal-responsive composite hydrogel, which can programmable control movement under illumination from an NIR laser.23
By introducing photothermal conversion materials such as gold nanoparticles, carbon nanotubes, and graphene into the hydrogel network, the hydrogel is endowed with photothermal properties. When exposed to a specific wavelength of light, these photothermal materials can efficiently absorb light energy and convert it into heat energy, increasing the local temperature of the hydrogel. Due to the volume phase transition temperature characteristics of hydrogels, the temperature change will cause them to swell or shrink, and deformation of the optical drive will be realized. Guo et al. introduced magnetic nanorods into the prepolarization solution of a hydrogel, causing the hydrogel sheet made via photopolymerization to produce a strain gradient in its thickness. The photothermal properties of the magnetic nanorods are used to generate heat under NIR light to achieve self-rolling.24
The light-driven miniature robot made using this hydrogel has good biocompatibility. However, its mechanical properties are relatively weak, which greatly limits its further application. At present, some researchers have introduced groups or molecules with supramolecular interactions (such as hydrogen bonds) to form dynamic crosslinking points in the hydrogel network, which can improve the mechanical properties of hydrogels. Liang et al. prepared a fully physical cross-linked deformable hydrogel via free radical polymerization at room temperature, and then immersed it in an aqueous Fe3+ solution. The strong physical interaction between the polyacrylic acid (PAA)–Fe3+ metal complex and hydrogen bond endowed this hydrogel with high tensile strength, high toughness and excellent flexibility.
In addition, Liang et al. immersed the polyacrylic acid (PAA) hydrogel structure in an aqueous Fe3+ solution. By selectively coating Fe3+ on its surface, programmable shape deformation of the hydrogel is realized, and the hydrogel strip can be deformed into any shape such as the letters “o”, “d”, “s”, and “w”.25
2.4 Biomaterials
Certain algae (such as chlamydomonas) and bacteria (such as Escherichia coli) exhibit phototaxis and have flagellar structures.27–29 They can swim to a light source through the swing of their flagella, and this phototaxis mechanism provides inspiration for optical-driven miniature robots. Based on the movement principle and control mechanism of chlamydomonas flagella, researchers applied it to the drive system of miniature robots.The structure of viruses is simple and highly specific and precise, and their shell protein can self-assemble to form nanostructures of various shapes.30–32 The self-assembly characteristics of viruses can be used as a framework or carrier for miniature robots. Viruses are modified by genetic engineering technology to express specific functional groups or light-sensitive molecules on their surface, thereby endowing micro-robots with the ability to be driven by light.
3. Driving mechanisms
3.1 Photothermal drive
Photothermal effect driving is based on a material absorbing light energy and converting it into heat energy, which causes the material or its surrounding environment to produce temperature changes, in turn causing a series of physical effects to realize the driving of miniature robots.In the field of optical-driven deformable miniature robots, light assumes the key role of energy fuel, while the body structure of the robot is responsible for performing actual mechanical actions. As important materials, liquid crystals, hydrogels and memory alloys play a key role in the structural design and functional realization of robots.
 | ||
| Fig. 3 (A) (a) Miniature robot made of SMA. (b) SMA exhibits reversible shape changes under irradiation (reproduced from ref. 33 with permission from Wiley Online Library, copyright 2019). (B) Miniature robot with a tapered nanopore array (reproduced from ref. 26 with permission from Wiley Online Library, copyright 2022). | ||
3.1.2.1 Based on deformation drive. Hydrogels are widely used in the field of soft robotics due to their softness, biocompatibility and stimulation responsiveness. Limbless creatures in nature usually live in narrow and difficult-to-enter spaces, which move through the continuous expansion and contraction of their body parts. Inspired by the octopus, Zhang et al. reported a soft swimming device that realizes movement under NIR illumination.34 A light-response driver was made by loading a poly-n-isopropylacrylamide (PNIPAM) hydrogel containing a metal organic framework (MOF) on the surface of a polydimethylsiloxane (PDMS) film (Fig. 4A-a). As an excellent photothermal nanomaterial, MOF can convert light energy into heat energy, resulting in a decrease in the water loss volume of the PNIPAM hydrogel. Due to the porous nanostructure of MOF, the adsorption/desorption of water was accelerated, thereby speeding up the response speed of the actuator (Fig. 4A-b). In addition, by adjusting the pattern of the PDMS film, a hydrogel actuator with a programmable shape was prepared, which can realize different deformation modes under optical drive (Fig. 4A-c). This work provides a reference for programmable intelligent soft miniature robots, thereby illuminating their application prospects in soft robots, biomedical equipment and other fields.34 Another group of studies loaded lightweight polydopamine-modified hollow glass microspheres (PDA-hgmp) into a PNIPAM hydrogel to make an NIR light-driven bionic fish miniature robot (Fig. 4B-a and b). PDA-hgmp can not only be used as an excellent photothermal agent, but also accelerate water adsorption/desorption in the hydrogel, thereby speeding up its response speed (Fig. 4B-c). This bionic fish miniature robot realizes controllable forward and floating/sinking movements under light. This research has potential applications in fields such as micromachines and intelligent systems.35 In addition to the above-mentioned two studies, Xiong et al. enhanced the deformability of a hydrogel by adding lattice metamaterials to make a tetherless optical-driven lattice soft microrobot (LSMR). The in situ rotation speed of LSMR reached 29.38° s−1, which is the fastest peristaltic movement speed among submillimeter-scale miniature robots.19
 | ||
| Fig. 4 (A) (a) Driving schematic of a bionic octopus driver. (b) Materials of the bionic octopus driver. (c) Motion schematic of the bionic octopus diver (reproduced from ref. 34 with permission from ACS Publications, copyright 2022). (B) (a) Driving schematic of a bionic fish driver. (b) Motion schematic of the bionic fish diver. (c) Schematic of its photothermal response. (reproduced from ref. 35 with permission from ACS Publications, copyright 2024). (C) (a) Motion schematic of a bionic caterpillar diver. (b) Fabrication of the driver (reproduced from ref. 36 with permission from MDPI, copyright 2024). (D) (a) Bionic snake microrobot based on CNT/PDMS composite materials. (b) Bionic snake microrobot crawls towards light in the form of wave propagation (reproduced from ref. 37 with permission from Wiley Online Library, copyright 2020). (E) (a) Multi-joint microactuator with multiple deformation modes. (b) Microactuators achieve joint deformation under illumination (reproduced from ref. 38 with permission from Nature Communications, copyright 2023). (F) (a) (i)–(iii) Schematic of human sit-ups. (b) Motion schematic of a droplet-like miniature robot (reproduced from ref. 37 with permission from Wiley Online Library, copyright 2020). | ||
In addition, Jia et al. made a worm-like soft robot based on CNT/PDMS composite materials (Fig. 4C-a). By adjusting the cutting parameters, a miniature robot with a complex micro-pattern can be obtained, which can be deformed into different three-dimensional structures under optical drive (Fig. 4C-b).36 Yang et al. also made a bionic snake microrobot based on CNT/PDMS composite materials (Fig. 4D-a). Under light irradiation, it crawls towards light in the form of wave propagation (Fig. 4D-b).37
In limbed creatures, the traction of joints on muscles triggers the corresponding actions. Miniature robots containing soft joints have been developed and used in soft grippers, artificial muscles and other fields. Xin et al. proposed a micro-joint composed of a PNIPAM hydrogel and metallic silver nanoparticles, developing a multi-joint microactuator with multiple deformation modes (Fig. 4E-a). Independent joint deformation and multi-joint deformation linkage enable the microactuator to imitate various complex humanoid forms, and its deformation time is short (30 ms) and its driving power is low (<10 mW) (Fig. 4E-b). This research found that the dual-joint micro-robotic arm can realize the transportation of multiple micro-cargoes. The micro-actuators introduced in this study have great application potential in micro-transportation, cell manipulation, etc.38 Yang et al. developed a droplet-like miniature robot based on CNT/PDMS composite materials. The robot can be driven by constant white light and perform continuous self-oscillating movement in a wavy manner, imitating human body sit-ups (Fig. 4F-a). In addition, the degree of movement is controlled by controlling the intensity and direction of the incident light (Fig. 4F-b). This research shows great potential in the field of optical-driven bionic soft robots.37
3.1.2.2 Based on bubble drive. Traditional motors are driven by external chemical fuels such as hydrogen peroxide and hydrochloric acid. Therefore, it is very important to develop fuel-self-sufficient motors that can get rid of the reliance on external fuels. At present, catalysts such as biological catalysts (enzymes, glucose oxidase, etc.) and inorganic particles (platinum, MnO2, etc.) have been embedded in hydrogel micro-motors to produce fuel-self-sufficient motors.
Wang et al. prepared a hydrogel motor using a hydrogel as the matrix based on the emulsion template method (Fig. 5A-a). Under light irradiation, graphene oxide (GO) inside the hydrogel absorbs light and converts it into heat. The bubbles produced by the motor under photothermal stimulation can effectively drive the motor (Fig. 5A-b). When a thermal explosion occurs inside the motor, bubbles are ejected violently from one side, and the movement speed reaches 14.78 ± 4.82 mm s−1. This micro soft motor has excellent self-propulsion performance and precise motion control, showing broad practical application prospects.39 Ho et al. fabricated a micromotor with glucose oxidase (GOx)/catalase (Cat) embedded in polyethylene glycol diacrylate (PEGDA) by using microfluidic technology (Fig. 5B-a). GOx and Cat are used to consume glucose as fuel generating oxygen microbubbles. These bubbles not only drive the micro-motors, but also their high surface tension can capture suspended plastic micro-particles in water and remove them from the aquatic environment (Fig. 5B-b). The non-toxic nature of these micro motors makes them suitable for environmental remediation and biological applications, addressing the urgent need for solutions to plastic pollution.40
 | ||
| Fig. 5 (A) (a) Manufacturing process of a hydrogel motor. (b) Motion schematic of the motor (reproduced from ref. 39 with permission from ACS Publications, copyright 2017). (B) (a) Driving principle of a micromotor. (b) Movement trajectory of the micromotor (reproduced from ref. 40 with permission from Elsevier, copyright 2024). (C) (a) Driving principle of a PDA/N micromotor. (b) Enhanced lighting to increase the movement speed of the micromotor (reproduced from ref. 41 with permission from Wiley Online Library, copyright 2025). (D) Driving principle of a micromotor (reproduced from ref. 42 with permission from Wiley Online Library, copyright 2023). | ||
Miniature robots can diagnose, treat and be applied to specific areas with minimal invasiveness, thus showing considerable potential in the biomedical field. However, the complex in vivo environment and extreme driving conditions greatly limit their application in vivo. Therefore, the motion patterns of miniature robots deserve further exploration and research. Li et al. proposed an improved photothermal material, polydopamine (PDA)–NH4HCO3 (PDA/N). PDA was modified by combining it with NH4HCO3 material, enhancing its bubble generation ability (Fig. 5C-a and b). Furthermore, by altering the porous structure of the microrobot, the bubble aggregation pathway is indirectly affected.41 In addition, Chen et al. also fabricated fuel-self-sufficient motors by using the above-mentioned materials, providing more ideas for light-driven microrobots in a high-viscosity internal environment (Fig. 5D). Compared with traditional PDAs, PDA/N microspheres with an obvious spherical multilayer morphology exhibit better photothermal conversion behavior under NIR light. In addition, the doping of NH4HCO3 radicals may be beneficial to directly increase the concentration of free radicals in PDAs, limit the non-thermal radiation transformation process, and improve the photothermal conversion performance of the material.42
3.1.2.3 Driving based on the Marangoni effect. The Marangoni effect refers to the phenomenon of liquid flow due to the gradient of liquid surface tension. In optical-driven miniature robots, photothermal materials are usually used. When irradiated by a specific wavelength of light, the photothermal material absorbs light energy and converts it into heat energy, causing the local temperature of the surrounding liquid to increase. Since the surface tension of the liquid decreases with the increase in temperature, a surface tension gradient is formed on the surface of the liquid. This surface tension gradient creates a driving force that pushes the robot to move on the surface or interface of the liquid. The Marangoni effect is of great research significance in the field of driving microrobots.
Jing et al. prepared a dual-mode-driven Marangoni hydrogel actuator, which combines chemical drive and photothermal drive (Fig. 6A-a). This actuator not only exhibits an excellent ethanol loading capacity, but also continuous and rapid movement. In addition, the actuator can move under IR light irradiation, and by adjusting the lighting position, the actuator can travel in a complex maze (Fig. 6A-b). This hydrogel actuator has great prospects in the field of intelligent transportation and soft robotics.43 In another study, Feng et al. made a drive based on PINPAM/MWCNT composite materials. The drive can be controlled by light and chemical reagents, respectively, and can perform programmable movements (such as linear translation, turning, and rotation) on the surface of water. The flared drive can advance and transport goods under IR lighting (Fig. 6B). This multi-stimulus response Marangoni actuator broadens the application scope of miniature actuators and provides inspiration for bionic robot technology.44 In addition, Chen et al. combined natural bamboo slices with a black amino acid/PNIPAM composite hydrogel to develop an actuator based on a composite hydrogel material (Fig. 6C). The efficient photothermal conversion capability of the black amino acid/PNIPAM composite material allows the actuator to be remotely controlled by light. In addition, bamboo slices with a fiber orientation can be cut at different angles, giving the actuator the ability to undergo programmable deformation. Actuators with various shapes made of composite materials can travel through complex mazes driven by light, and can move in different orientations by controlling the position of the light.45
 | ||
| Fig. 6 (A) (a) Motion schematic of the motor. (b) Movement trajectory of the micromotor (reproduced from ref. 43 with permission from Elsevier, copyright 2025). (B) (a and b) Trumpet-shaped microrobot transports goods (reproduced from ref. 44 with permission from ACS Publications, copyright 2025). (C) Microrobot goes through a maze driven by light (reproduced from ref. 45 with permission from Elsevier, copyright 2022). (D) (a) Motion schematic of a hydrogel actuator. (b) Hydrogel actuator transports goods (reproduced from ref. 46 with permission from Elsevier, copyright 2023). (E) (a) Structural schematic of the microrobot. (b) Movement forms of the microrobot under different lighting conditions (reproduced from ref. 47 with permission from ACS Publications, copyright 2021). | ||
Zheng et al. covalently integrated oxazinyl compounds into a hydrogel network, and the obtained oxazin hydrogel was able to respond to external irradiation with light and heat under visible light irradiation (Fig. 6D-a). Under the effect of a temperature gradient caused by light, the hydrogel actuator moves based on the Marangoni effect at the gas–liquid interface. By controlling the lighting parameters, the actuator realizes the movement of straight, steering, and rotation, realizing the function of cargo transportation (Fig. 6D-b). This research provides a reference for the development of other photostimulation-responsive hydrogels.46 To increase the selection range of light sources, Watanabe et al. mixed two types of photothermal conversion rare earth oxide nanoparticles in a thermosensitive polymer gel to fabricate a multi-wavelength responsive micro-brake (Fig. 6E-a). The micro-brake can realize different forms of movement based on the Marangoni effect in the presence of different wavelengths of light (Fig. 6E-b). This research provides a reference for broadening the movement form of optical-driven microrobots.47
When photothermal materials (such as CNTs, gold nanorods, and GO) are embedded in LCEs as “light control switches”, the materials exhibit unique photothermal response characteristics. These materials can efficiently absorb visible, NIR or IR light, and quickly convert photon energy into heat energy. The photothermal material warms up under light, triggering the LCE to change from an ordered liquid crystal phase to a disordered isotropic phase, thereby achieving programmable reversible deformation. The photothermal driving principle not only gives the material the ability to deform remotely and controllably, but also because of its precise regulation characteristics of optical signals, provides innovative solutions for intelligent bionic devices, flexible actuators and other fields.
Researchers have long been committed to reproducing the functions of natural species in artificial systems, trying to create bionic devices that can act independently and adapt to environmental mutations. Because of their good biosafety, LCE-based soft robots have a flexible and miniaturized body structure, which provides an additional degree of freedom (DOF) to adapt to complex sports scenes and environmental changes, and their unique movement mode driven by photothermal effects gives them the potential to imitate biological movement. Inspired by the morphology and movement morphology characteristics of animals and plants in nature, researchers have developed miniature crawling robots, miniature bouncing robots, and photogenic bionic plant miniature robots. Under light irradiation, the microrobot converts the absorbed light energy into heat energy, causing the polymer network to change from an ordered liquid crystal phase to a disordered isotropic phase, triggering a local shrinkage effect, and then driving the microrobot to complete the bionic movement. This photothermal drive mechanism not only realizes the accurate simulation of biological motion patterns, but also provides innovative solutions for autonomous operations in more complex environments.
Molluscs such as snails, earthworms, and caterpillars rely on body deformation to move. Taking caterpillars as an example, they have a soft cylindrical body and flexible feet. In each crawling cycle, their terminal limbs first leave the ground and stretch forward to complete a new ground-touching positioning. Then, their abdominal feet and limbs are lifted and moved forward in the order from back to front, forming a wavy trajectory transmitted from their tail to their head, pushing the whole way forward through the regular deformation of their body (Fig. 7A-a). This unique movement mode shows the excellent mechanism of organisms to achieve non-rigid support movement through the coordination of flexible body joints. Zeng et al. reported a bionic caterpillar robot made of a single-layer LCE film (Fig. 7A-b). The robot can crawl on different surfaces such as skin and paper under the stimulation of visible light with uniform space and relatively low intensity (Fig. 7A-c).48 With the deepening of research, to enable bionic caterpillar robots to show new actuation behaviors or functions, researchers have developed bionic caterpillar robots based on multi-layer structures. Zhu et al. reported a photothermal response bionic caterpillar robot, where its driving layer is made of LCE and silver nanowire (AgNWs) composite materials, and its passive layer is a polyimide layer. The introduction of the passive layer enhances the strain mismatch and bending moment of the robot, enabling it to realize various actions such as crawling and tumbling (Fig. 7B).49 In another study, to facilitate the control of the bending direction, Xu et al. fabricated a bilayer structure based on a siloxane-based liquid crystal elastomer (Si-LCE) (Fig. 7C-a). Under NIR light, the double-layer Si-LCE structure crawls forward like a caterpillar (Fig. 7C-b).50
 | ||
| Fig. 7 (A) (a) Schematic of the movement of a caterpillar. (b) Bionic caterpillar robot made of a single-layer LCE film. (c) Robot crawling on different surfaces (reproduced from ref. 48 with permission from Wiley Online Library, copyright 2018). (B) Different movement modes of the robot (reproduced from ref. 49 with permission from ACS Publications, copyright 2024). (C) (a) Bilayer structure based on Si-LCE. (b) Motion schematic of an actuator (reproduced from ref. 50 with permission from Elsevier, copyright 2023). (D) (a and b) Motion forms of microrobots with different structures (reproduced from ref. 51 with permission from Wiley Online Library, copyright 2019). (E) (a) Schematic of jumping. (b) Principle of jumping (reproduced from ref. 52 with permission from Wiley Online Library, copyright 2022). (F) (a and b) Restraint fin array structure. (c) Jumping of robot (reproduced from ref. 53 with permission from Wiley Online Library, copyright 2023). | ||
Although the above-mentioned research can enable bionic caterpillar robots to display more complex forms and movement patterns, their multi-layer structure poses the risk of stratification. To overcome this shortcoming, researchers developed a programmable structure. Jiang et al. selectively de-crosslinked the LCE structure by relying on the photolysis of an anthracene dimer, and re-configured the distribution of the driving region (crosslinked) and the non-driving region (de-crosslinked) (Fig. 7D). The bionic caterpillar robot with a planar shape can be reconfigured on demand through selective de-crosslinking to exhibit reversible shape transformation.51 While this method overcomes the risk of stratification, it enhances the motor ability of the bionic caterpillar robot.
Jumping plays an important role in the predation and reproduction of creatures in nature, which is a sudden reaction (Fig. 7E-a). In this process, mechanical instability will quickly transform the structure from one stable form to another, which is achieved by instantly releasing pre-stored elastic energy. The latest research progress in response materials has promoted the development of soft robots that trigger jumping behavior by external stimuli. However, control of the jump time, speed, etc. still faces huge challenges. Guo et al. reported a method that can accurately control the release and emission time of elastic energy. Elastic energy is pre-stored in a light-responsive LCE driver. Light and heat induce the crystal-to-liquid transition of a liquid crystal adhesive latch, thereby releasing the energy and causing ejection (Fig. 7E-b). This method controls the launch speed by flexibly controlling the stored energy and can show the jumping and ejection movements of soft robots and the coordinated movements of multiple soft robots (Fig. 7E-c). The research results provide a practical solution for the rapid motion control of miniature robots.52 In another study, Wang et al. reconfigured a restraint fin array structure to adjust the ejection angle, height, and speed of a soft robot as needed to achieve controllable and programmable ejection movement (Fig. 7F-a and b). In addition, the soft robot has ultra-fast movement speed, powerful ejection and high ejection height (Fig. 7F-c). This research provides a reference for realizing the controllable fast movement and high-power movement of soft robots.53
As natural flexible drivers, muscles exist in the human body in the form of skeletal muscles, smooth muscles, and heart muscles. They can be stimulated by nerve signals to produce diverse deformations and perform complex biological functions. For example, the contraction of the biceps brachii and triceps brachii cooperates to grasp objects, and the contraction movement of the myocardium drives blood pumping. Because LCEs can produce significant reversible deformation under a variety of stimuli, they have become ideal materials for building soft artificial muscles. At present, LCE systems based on the photothermal drive mechanism have been widely used in artificial muscle research and development. They regulate the phase transition process of materials through photothermal effects, simulate the driving characteristics of natural muscles, and provide innovative solutions for flexible robots, biomedical devices and other fields.
Wang et al. constructed a miniature robot made of hard and soft materials that mimic the musculoskeletal system (Fig. 8A-a). Among them, the LCE/CNT composite material is used as a soft material to simulate artificial muscles, whose multi-directional movement can be controlled by optical drive (Fig. 8A-b). The miniature robot has the advantages of ultra-lightness, high flexibility and precise control, and can move along different paths on different surfaces.54 This study preliminarily proved that LCEs are good materials for the manufacture of artificial muscles. However, the mechanical properties of LCEs are poor. Thus, to overcome this shortcoming, Yang et al. prepared an LCE/CNT composite fiber based on electrospinning technology and a two-step crosslinking strategy (Fig. 8B-a). The fiber has a reversible shrinkage rate of nearly 70%, whose tensile strength and response speed are greatly improved. Artificial muscles made of single fibers can achieve a rapid response through photothermal drive (Fig. 8B-b). This research provides a reference for the development of artificial muscles and miniature robots.55 Based on this, to increase the driving force of yarn, He et al. integrated multiple fibers. The results show that one fiber can lift a heavy object of 0.1 g, while four fibers can lift a heavy object of 0.5 g without affecting the response speed. Then, He et al. installed a fiber bundle on a hard arm skeleton to simulate the biceps muscles to achieve rapid elbow movement (Fig. 8C).56
 | ||
| Fig. 8 (A) (a) Bionic musculoskeletal system. (b) Movement trajectory of the bionic musculoskeletal system (reproduced from ref. 54 with permission from Wiley Online Library, copyright 2019). (B) (a) Fabrication of LCE/CNT composite fiber. (b) Artificial muscles made up of one fiber response through photothermal drive (reproduced from ref. 55 with permission from ACS Publications, copyright 2024). (C) Artificial muscles made up of four fibers responding through photothermal drive (reproduced from ref. 56 with permission from Science, copyright 2021). (D) (a) Bionic finger. (b) Fabrication of the bionic finger (reproduced from ref. 57 with permission from Wiley Online Library, copyright 2025). (E) (a) Bionic palm. (b) Under optical drive, the bionic fingers bend (reproduced from ref. 58 with permission from Wiley Online Library, copyright 2018). | ||
To increase the deformation performance and function of LCE artificial muscles, researchers created programmable artificial muscles. Ding et al. integrated the light-responsive o-nitrophenyl moiety and the temperature-responsive hydrogen bond into an LCE (Fig. 8D). The o-nitrobenzyl moiety can cause irreversible reconstruction of the LCE crosslinking network through photochemical reactions, giving the material the characteristics of easy orientation, arrangement and reshaping. The hydrogen bond acts as a “temperature switch” to adjust the mobility of the polymer chain during thermal deformation. The synergistic effect of light-induced irreversible reconstruction and temperature-induced reversible hydrogen bond exchange enables LCE to achieve programmable continuous deformation.57 Wei et al. penetrated an LCE/GO mixture into a silicon opal template, removed the template after UV light polymerization, and prepared an optical drive palm with reversible deformation properties (Fig. 8E-a). The selective deformation behavior of the optical drive is realized in a two-phase LCE structure with alternating macroscopic nematic (N) and isotropic (I) phases, thereby realizing a variety of drives. This bionic palm can make different gestures (Fig. 8E-b). This research represents a significant advancement in narrowing the gap between microrobots and living organisms.58
There are many plants in nature that will deform when stimulated by the outside world. For example, pine cones will contract and relax with changes in external humidity, Mimosa plants will close after sensing external pressure, and sunflowers will grow towards light. Inspired by a variety of natural plants, researchers have created various stimulus-responsive bionic miniature robots. These bionic miniature robots exhibit complex shape deformation after being stimulated by external stimuli, which is needed in artificial intelligence applications. However, in most robots, it is challenging to achieve versatility and configurability at the same time. To meet the needs of multifunctional intelligent microrobots, Zhang et al. designed and manufactured a double-layer soft actuator with shape-programmable PNIPAM-LCE. At the lower critical solution temperature (LCST), the PNIPAM layer transitions between hydrophilic and hydrophobic (Fig. 9A-a). The rapid response of the LCE layer to thermal stimulation enables the double-layer soft actuator to deform quickly. By controlling the lighting time and position, the shape of the double-layer soft actuator is allowed to be programmable. This study proves the ability of double-layer soft actuators to grab objects and imitate bionic aspects such as petal opening and closing (Fig. 9A-b). Also, it provides a reference for the design of multifunctional soft actuators that can operate in different environments.59 Tang et al. proposed a bionic soft actuator based on an azobenzene chromophore and LCE (Fig. 9B-a). The bionic soft actuator can realize dual-stimulus-response and complex bionic deformation, such as blooming flowers and object grabbing (Fig. 9B-b). The bionic soft actuator provides guidance for the design and manufacture of flexible multi-purpose actuators, which have broad prospects in the field of intelligent bionics.60 Wani et al. showed a bionic fly trap made based on light-response LCE, which triggers photoelectric drive through light feedback (Fig. 9C-a). The bionic fly catcher mimics a natural fly catcher, having automatic closure and object recognition capabilities (Fig. 9C-b). This research has opened up the way for soft and autonomous miniature robots.61 Yang et al. reported a tubular structure based on an MXene/LCE composite material, which can be bent toward 360°, and can quickly perceive and track a light source in three-dimensional space, just like the hollow stems of plants in nature (Fig. 9D). This research proposes an adaptive photovoltaic system for solar energy collection, which provides insights into the development of adaptive optoelectronics, bionic microrobots and other fields.62
 | ||
| Fig. 9 (A) (a) Bionic flower. (b) Deformation of the bionic flower (reproduced from ref. 59 with permission from Wiley Online Library, copyright 2025). (B) (a) Deformation of the bionic flower. (b) Gripper grabbing an object (reproduced from ref. 60 with permission from ACS Publications, copyright 2021). (C) (a) Bionic Venus flytrap. (b) Bionic Venus flytrap grabbing an object (reproduced from ref. 61 with permission from Nature Communications, copyright 2017). (D) Deformation of the bionic flower (reproduced from ref. 62 with permission from Wiley Online Library, copyright 2022). | ||
| F = −C∇T(y, t) | (1) |
| C = (9πdpη2ka)/(2ρTkp) | (2) |
Since 2010, researchers have been exploring photodynamic miniature robots based on temperature gradients. These miniature robots have an asymmetric structure, which causes the asymmetric distribution of temperature to circulate the surrounding fluid and move them. In addition, photochemical microrobots driven by photothermal effects often exhibit clustering, and thus they are widely used to simulate biological communities. Sun et al. demonstrated a new type of NIR light-driven micromotor composed of polystyrene microspheres and polydopamine core–shell structure (PS/PDA) (Fig. 10A-a). A single micromotor exhibits a negative phototaxis, while a cluster of micromotors exhibits a positive phototaxis, which is due to the competition between thermal swimming power and hydrodynamic resistance (Fig. 10A-b). In addition, the positive photogenicity can control the shape of the cluster. By adjusting the lighting position, the cluster can show different aggregation forms (Fig. 10A-c). This research provides a new strategy for controlling the collective behavior of micromotors.62
 | ||
| Fig. 10 (A) (a) Photothermal effect of a micromotor. (b) Movement of a single micromotor and clusters of micromotors. (c) Clusters of micromotors with different shapes (reproduced from ref. 62 with permission from Wiley Online Library, copyright 2022). (B) (a) Fabrication of a micromotor. (b) Movement of the micromotor (reproduced from ref. 63 with permission from Elsevier, copyright 2023). (C) Movement of micromotor clusters (reproduced from ref. 64 with permission from ACS Publications, copyright 2024). | ||
Micromotors have attracted much attention because of their intelligent transportation capabilities, especially for targeted drug delivery in dynamic environments such as the blood, tumor interstitial matrix, and blood–brain barrier. Meng et al. covered PDA with quadrangular BaTiO3 (tBT) nanoparticles (NPs), and then loaded ciprofloxacin (CiP) on the PDA cap to prepare a tBT@PDA-Cip micromotor (Fig. 10B-a). When the NIR illuminates PDA, the temperature changes on the tBT NPs, resulting in a photothermal effect. The increase in temperature on one side produces thermal swimming force to drive the movement of the micromotor, enhancing its interaction with bacteria and overcoming the limitations of short acting distance and short life of ROS (Fig. 10B-b). This study improved the antibacterial effect, while avoiding the effects of the extensive use of reactive oxygen species and antibiotics on the human body.63 To perform more complex micro-operations in medical applications, it is very desirable to give micromotors multi-stimulus responsiveness instead of single stimulus responsiveness. Altas et al. made a micromotor with a Janus structure based on Precirol ATO 5 and polycaprolactone (Fig. 10C). Polypyrrole NPs with high photothermal conversion efficiency and magnetic NPs with magnetic response characteristics have been added to these micromotors. The micromotor clusters exhibit a photothermal effect and magnetic response effect. By combining these two response characteristics, the micromotor clusters can overcome the complex liquid environment and the limitations encountered by a single-stimulus drive. This research provides broad prospects for biological sensing and optical detection.64
3.2 Photochemical drive
The photochemical reaction drive depends on the specific chemical reaction of materials under light, resulting in the generation, consumption of the material or a gradient change in the concentration of chemical compounds, etc., thereby providing the driving force for miniature robots. The common photochemical reaction drive mechanisms include self-diffusion drive, self-electrophoresis drive, and bubble drive (Fig. 11). | ||
| Fig. 11 Photochemical drive mechanisms. | ||
As early as 1947, researchers theoretically predicted and established a diffusion motion model. Similar to most chemical reactions, the photochemical reactions of photochemical microrobots involve reactant consumption and product generation. The asymmetry of the reaction causes a concentration gradient to form inside or on the surface of the robot, driving the diffusion of chemical substances, which in turn triggers self-diffusion movement. As photolytic materials, silver halides have been extensively studied. Among them, silver chloride (AgCl) is widely used in the literature.
Zhou et al. explained in detail the driving mechanism of an AgCl micromotor. Firstly, they manufactured an Ag/AgCl micromotor. This study found that AgCl particles undergo a photocatalytic decomposition reaction when irradiated to decompose into Ag, generating H+ and Cl−. Since the diffusion rate of H+ is faster than that of Cl−, an uneven charge distribution is generated around the micromotor, causing the local net electric field to point to the surface of AgCl (eqn (3) and (4)). The micromotor moves toward the side away from AgCl (Fig. 12A-a).
| Ag + H2O2 + 2H+ + 2Cl− → 2AgCl + 2H2O | (3) |
 | (4) |
 | ||
| Fig. 12 (A) (a) Photocatalytic decomposition reaction of AgCl. (b) Movement of a PMMA/AgCl micromotor (reproduced from ref. 65 with permission from ACS Publications, copyright 2018). (B) (a) Movement trajectory of PMMA–Ag micromotor. (b) PMMA–Ag Janus motor moving in an oscillating manner (reproduced from ref. 66 with permission from ACS Publications, copyright 2019). (C) (a) Movement of an Ag/AgCl micromotor. (b) Driving principle of the Ag/AgCl micromotor (reproduced from ref. 67 with permission from Wiley Online Library, copyright 2018). (D) (a) Glucose-powered micromotor exhibiting highly controllable negative phototaxis driven by visible light. (b) By adjusting the content of N-CNT, glucose concentration, or light intensity in the motor, the speed of micromotor can be adjusted (reproduced from ref. 68 with permission from ACS Publications, copyright 2019). | ||
Subsequently, Zhou et al. used polymethyl methacrylate (PMMA)–AgCl Janus micromotors to explore the detailed propulsion mechanism of self-diffusion (Fig. 12A-b). Firstly, the chemical reactions that occur on their surface were studied. When exposed to light with sufficient energy, the micromotor exhibits a negative gravitational axis.65 Based on this, Zhou et al. reported a PMMA–Ag Janus motor that moves in an oscillating manner. Its speed and movement are adjusted by different light intensities and the way light is applied (Fig. 12B).66
Wang et al. reported a blue light-driven Ag/AgCl micromotor showing high motility in pure water, whose speed is 100 times higher than that of a visible light-driven micromotor and 7 times higher than that of a UV light-driven micromotor (Fig. 12C). In addition, Wang et al. also demonstrated the motor potential of micromotors in human saliva, phosphate buffered saline solution, and rhodamine B solution. This research contributes to the application of visible light-driven micromotors in biomedicine and environmental protection.67
To increase the application of micromotors in the biomedical field, it is very important to enhance the biocompatibility of fuels they need. Glucose is an ideal fuel for micromotors. However, the current glucose-powered micromotors are based on enzyme-catalyzed driving mechanisms, and their strict operating conditions and weak driving performance greatly limit their application. Wang et al. reported an efficient glucose-powered micromotor, using cuprous oxide@N doped carbon nanotube (Cu2O@N-CNT) as the raw material, which can be driven by visible light and exhibits a highly controllable negative phototaxis (Fig. 12D-a). The speed of the micromotor is comparable to that of a traditional Janus micromotor powered by toxic H2O2 fuel. By adjusting the content of N-CNT, glucose concentration, or light intensity in the motor, the speed of the micromotor can be adjusted (Fig. 12D-b). The micromotor has excellent driving force in the biological environment and wide application prospects in the field of biomedicine.68
Self-diffusing micromotors are essentially an ion source, and the ion concentration of the surrounding environment will continue to increase during the work process. Because ions have the characteristics of shielding electric fields, when the particle density of these micromotors increases, their movement speed will slow down accordingly, which is easy to form clusters that continue to gather and increase between the particles. Self-diffusion drive has higher efficiency and better control performance. However, its material selection range is narrow, and heavy metal materials are currently used, which leads to poor biocompatibility and limited application in fields such as biomedicine that require high material safety.
The mechanism of self-electrophoresis can be simply summarized as a redox reaction on the surface of a Janus micromotor. Some semiconductor materials such as TiO2 undergo charge separation under specific wavelengths of light, where their valence electrons are excited and transition to the conduction band to form electron–hole pairs. These photogenic carriers have strong redox ability and can react chemically with substances in the surrounding environment to produce high concentrations of H+. To balance the charge, the H+ ions migrate to the metal side. Subsequently, the H+ ions are consumed on the metal side due to reduction. The migration of H+ ions is accompanied by local electrical osmotic flow, causing the micromotor to move in the opposite direction.
Xu et al. used the chemical solution method to prepare a Cu2O miniature robot that can respond to the movement of a 365 nm UV light source, proving its negative phototaxis. The movement direction of the microrobot can be changed with a change in the UV light incident direction.10 However, due to their narrow light absorption spectrum, traditional micromotors can only respond to specific wavelengths of light, limiting their application in fields such as biomedicine and environmental restoration. Jang et al. proposed a multi-wavelength light-responsive micromotor that asymmetrically coated a layer of gold on the surface of TiO2 microspheres. The TiO2 microspheres can absorb light at wavelengths between 300 and 800 nm (Fig. 13A). Whether in H2O2 solution or in pure H2O, the micromotor can respond to UV, blue, cyan, green and red light. By controlling the presence or absence of light, the stop and movement of the micromotor are realized, which are necessary features for the maneuverability of the micromotor.69 Panda et al. fabricated a rod-shaped micromotor by adding photocatalytic materials (TiO2 and Cu2O) at specific positions (Fig. 13B). When exposed to a specific wavelength, the micromotor exhibits different motion behaviors such as linearity and helix. The application of light realizes the control of the motion mode by changing between the UV and visible range. This study provides a reference for micromotors with complex motion patterns.70
 | ||
| Fig. 13 (A) Driving principle of a microsphere (reproduced from ref. 69 with permission from ACS Publications, copyright 2017). (B) Driving principle of a rod-shaped micromotor (reproduced from ref. 70 with permission from Wiley Online Library, copyright 2024). (C) Difference between (a and b) micromotor self-electrophoresis and (c and d) passive electrophoresis (reproduced from ref. 71 with permission from Nature Communications, copyright 2019) (reproduced with permission from ref. 72). (D) Driving principle of a micromotor (reproduced from ref. 72 with permission from Wiley Online Library, copyright 2024). (E) Schematic of a dendritic hematite miniature robot (reproduced from ref. 73 with permission from Wiley Online Library, copyright 2022). | ||
However, most micromotors driven by self-electrophoresis mechanisms have a key defect, that is, they cannot work in high-concentration electrolyte solutions (such as body fluids and blood). Thus, to solve this shortcoming, Zhan et al. proposed the fundamental difference between micromotor self-electrophoresis and passive electrophoresis. This study found that the use of polyelectrolyte coating technology combined with the geometric structure optimization strategy can greatly increase the Dukhin number (Fig. 13C). This study broke the limitations of traditional Helmholtz–Smoluchowski behavior and achieved a favorable dynamic shift. In the case of self-electrophoretic micromotors driven by visible light, this method significantly enhances their ion tolerance. This research has laid an important foundation for the practical application of optical-driven micromotors in complex ion environments.71 In addition, Gao et al. designed a new type of optical-magnetic hybrid micromotor. Under the combined effect of a light field and magnetic field, these micromotors achieve efficient light propulsion and accurate magnetic navigation (Fig. 13D). The optical-magnetic hybrid micromotor can exhibit different motion modes on different surfaces and environments, enabling it to work in complex in vivo environments.72 In another study, Peng et al. proposed a dendritic hematite miniature robot for destroying prostate cancer cells (Fig. 13E). These hematite miniature robots exhibit a negative phototaxis and move on a specified path under the control of an external magnetic field. By controlling the movement of the miniature robot, the non-contact transportation of cells can be realized. This miniature robot contributes to the development of optical-magnetic coupling drive in the field of biomedicine.73
To efficiently realize photogenic bubble drive, there are many considerations in the structural design of micromotors. Their surface usually has a special morphology or modification, such as rough surface texture can provide more bubble nucleation sites and promote bubble generation. Zhang et al. designed an open, hollow TiO2 nanoparticle shell (OMHTNPS) for light-induced bubble propulsion, which is used to effectively degrade organic pollutants (Fig. 14A). OMHTNPS with an asymmetric structure can be obtained by a simple sol–gel method, followed by a simple sintering process to remove carbon microspheres. The shell of the prepared OMHTNPS is mainly composed of 20 nm anatase-TiO2 nanoparticles. When exposed to UV light, the photocatalytic decomposition of H2O2 produces air bubbles, driving OMHNPS to move quickly. By controlling the light intensity and H2O2 concentration, the direction and speed of movement of OMHNPS can be controlled. Experiments have shown that OMHTNPS exhibits the ability to efficiently remove organic pollutants.74
 | ||
| Fig. 14 (A) Schematic of OMHTNPS (reproduced from ref. 74 with permission from ACS Publications, copyright 2022). (B) (a) Diagram of the X-ray device. (b) Driving of a micromotor (reproduced from ref. 75 with permission from ACS Publications, copyright 2019). (C) Micromotor driven by UV and visible light (reproduced from ref. 76 with permission from ACS Publications, copyright 2016). | ||
In addition, materials with light-response functions are combined with support structures to form a composite structure to ensure the high efficiency of light absorption and bubble generation. The shape of the micromotor will also affect the sports performance, and a streamlined structure helps to reduce sports resistance and improve the sports efficiency. Xu et al. developed semi-metal-coated Janus particle propulsion technology based on X-ray drive (Fig. 14B-a). Through a full-field transmission X-ray microscope, the propulsion process of Janus particles in aqueous solution can be observed in real time. This study found that X-ray irradiation caused the decomposition of water on the surface of the particles to intensify, and the rapid growth of air bubbles drove the movement of the particles, whose propulsion speed could be remotely controlled by adjusting the radiation dose (Fig. 14B-b). This technology is expected to be applied to opaque environments, providing a new direction for the combination of light energy micromotors with medical imaging and non-destructive testing.75
In another study, Moo et al. confirmed that spiropyran additives can be used as molecular switches for bubble propulsion micromotor systems. In this study, bimetallic concentric copper/platinum micromotors were selected, using hydrogen peroxide as fuel, which catalyzed decomposition under the action of sodium lauryl sulfate to produce micro-bubbles to promote the micromotor (Fig. 14C). The ratio of spiropyran to partial anthocyanins significantly affects the motor movement and can be adjusted by selectively using UV and visible light. Through UV and visible light irradiation, the speed and movement of the micromotor can be controlled, and the precise operation of a single or cluster of micromotors can be realized.76
With its significant driving force advantages and speed performance higher than photothermal propulsion, bubble propulsion has shown great potential in the field of miniature propulsion. Compared with photothermal propulsion, which relies on a heat exchange gradient between the material and the environment to produce a weak driving force, the speed of bubble propulsion can often reach several times that of photothermal propulsion micromotors. However, the currently commonly used fuels such as hydrogen peroxide not only pose a potential threat to the ecological environment, but also limit their application in sensitive fields such as biomedicine. Due to the randomness of bubble generation and the complexity of the fluid environment, it is difficult to accurately regulate the trajectory of the micromotor, making it difficult to meet the needs of targeted delivery, precision operation and other tasks. Also, problems such as insufficient efficiency and poor stability in the bubble generation process greatly affect the consistency and reliability of the motion performance of the micromotor.
3.3 Optical tweezers drive
Optical tweezers are extremely important and mature technology in the field of miniature object manipulation. In 1986, Ashkin and others of AT&T Bell Labs proposed this groundbreaking technology, and its results were also awarded the Nobel Prize in Physics in 2018. The principle of micro-manipulation by optical tweezers is based on the interaction between light and matter. When a laser light beam is highly focused, a beam with a spatial intensity gradient distribution is formed. In the case of tiny objects, this beam of light will induce them to produce an electric dipole moment; in slightly larger objects, it will cause light refraction. Both these effects will eventually produce a force pointing to the focus of the beam, thereby driving the miniature object to move in the focus direction, realizing the precise grasping and manipulation of the miniature object (Fig. 15A). Villangca et al. developed a new lightweight robot (Fig. 15B-a). This robot uses optical tweezers technology to achieve efficient cargo handling operations. Using silica and polystyrene beads as experimental cargo, the feasibility of this technology was verified (Fig. 15B-b). The material transportation mode created by the robot has great potential and is expected to cause major changes in the fields of targeted drug delivery, which open up a new path for the development of related cutting-edge medical technologies.77 Shishkin et al. further developed a microrobot driven by optical tweezers and demonstrated its ability to manipulate biological objects (Fig. 15C-a). Their research focuses on overcoming the long-term technical problems of the out-of-plane rotation of biological objects in a pure liquid environment and verifies the effectiveness of the technology through live cell experiments. The experimental results show that the new type of miniature robot can accurately manipulate biological objects and realize complex spatial movements (Fig. 15C-b). This innovative achievement provides a more flexible and efficient tool in the field of biomedicine, and it is expected to be widely used in high-precision biological operations such as cell tomography and microsurgery in the future.78 | ||
| Fig. 15 (A) Based on a focused Gaussian beam, particles are subjected to radiation pressure and a gradient force of light intensity. (B) (a) New lightweight robot. (b) Cargo handling operation of the lightweight robot (reproduced from ref. 77 with permission from ACS Publications, copyright 2016). (C) (a and b) Microrobot driven by optical tweezers manipulates biological objects (reproduced from ref. 78 with permission from MDPI, copyright 2020). (D) (a) Electron microscope image of microrobot. (b) Microrobot measures cell elasticity (reproduced from ref. 79 with permission from MDPI, copyright 2020). (E) (a) Electron microscope image of a microrobot. (b) Microrobot measures fiber elasticity (reproduced from ref. 80 with permission from MDPI, copyright 2018). (F) (a) Schematic of a screw and nut. (b) Assembly of the screw and nut (reproduced from ref. 81 with permission from Microsystems & Nanoengineering, copyright 2017). (G) Schematic of a biohybrid micromotor (reproduced from ref. 82 with permission from Nature Communications, copyright 2017). (H) (a) Electron microscope image of a microrobot. (b) Microrobot moves completely freely under the control of optical tweezers (reproduced from ref. 83 with permission from Wiley Online Library, copyright 2017). | ||
Microrobots driven by optical tweezers also have a profound impact in the field of detection. Grexa et al. conducted an innovative study that used optical tweezers to drive microrobots to achieve the accurate measurement of cell elasticity. The shape of the microrobot prepared by a two-photon polymerization method enables the cell to approach the microrobot from any transverse orientation (Fig. 15D-a). In the experiment, endothelial cells grown on a vertical polymer wall were detected and analyzed in the horizontal direction by the microrobot (Fig. 15D-b). The specially shaped microrobot can effectively avoid the risk of light damage to the target cell. The application of optical tweezers technology has successfully expanded the force perception range of cell indentation measurement to the femtonewton level (fN), providing an efficient and feasible alternative to the study of cell mechanical properties and breaking through the technical limitations of traditional vertical indentation experiments.79 Lamperska et al. have developed a dumbbell-shaped miniature robot with a simple structure and diverse functions, which can be accurately controlled by optical tweezers. The dumbbell-shaped miniature robot is equipped with a bead for efficient optical capture at each end, and one end is also equipped with a detection spike (Fig. 15E-a). In the experiment to detect the vibration of suspended cotton fibers in water, the detection spikes of the miniature robot came into contact with the fibers with a diameter of 20 microns (Fig. 15E-b). It is worth noting that in this experiment, only the microbeads directly irradiated by the laser beam will be affected by the refresh rate and addressing rate, while the cotton fibers are not affected by this effect.80
In the development process of future miniature technology, the miniaturization, functionalization and integrated manufacturing of devices have become the core development directions. Köhler et al. proposed assembly technology based on a bottom-up strategy, breaking through the traditional manufacturing model, and relying only on optical tweezers technology to build complex microsystems (Fig. 15F-a). The capture, movement and screwing assembly of screw-nut-like miniature parts are completed through optical tweezers technology (Fig. 15F-b). This assembly technology has been successfully applied to a microfluidic control system to realize the pumping and mixing functions of fluid on a chip. This research has revolutionized the assembly method of micro-components and opened up a new path for the construction of complex hybrid microsystems.81 Vizsnyiczai et al. obtained a fast, reliable and flexible biohybrid micromotor by self-assembling a synthetic structure with a genetically engineered biological propeller (Fig. 15G). Its synthetic components adopt a 3D interconnected structure and are equipped with rotating units, which can accurately capture individual bacteria into the micro-chamber array, so as to give full play to the efficiency of cells in terms of torque output. These bacterial cells have excellent swimming ability, whose swimming speed can be regulated by optical tweezers technology. Through the application of real-time feedback control loops, multiple sets of micromotors can also be directed to rotate synchronously according to preset angular velocities, demonstrating the strong potential of this technology in the field of precision control.82 Avci et al. developed an articulated microrobot by optimizing its contact surface area and used optical tweezers technology to realize its drive (Fig. 15H-a). Experiments proved that the articulated microrobot can move completely freely under the control of optical tweezers without the need for physical connection constraints (Fig. 15H-b). With their three-dimensional spatial control capabilities, articulated microrobots have great application potential in the field of biomedicine. They are expected to be widely used in complex scenarios such as single-cell analysis, embryo injection, polar biopsy, nuclear transplantation, and multi-dimensional imaging of microsurgery.83
Optical tweezers technology has greatly promoted research in physics, chemistry and biology. Its ability has developed from single-target manipulation to multi-target manipulation. However, the limitation of optical tweezers is that the momentum of photons is small. Thus, to effectively manipulate miniature particles, strong light with intensities up to the GW cm−2 range is required, which limits larger applications outside of research. Therefore, many alternative solutions are being developed to achieve similar multi-functional optical tweezers with much lower light intensity.
Optical electric tweezers (OET) and optical tweezers have similar names, but their working principles are very different. OETs mainly manipulate objects through electrodynamic force and electrohydrodynamic force generated by an electric field. Their core lies in the use of structured light to construct a specific electric field distribution pattern on a photoelectric electrode. Compared with traditional optical tweezers technology, the significant advantage of optical tweezers is that they greatly reduce the requirements for light intensity and expand the size range of controllable objects. Due to the low demand for light intensity, the method for the generation of structured light is extremely flexible, which can be achieved either through a fixed mask or with the help of a computer-controlled digital light processing (DLP) projector. This feature greatly improves the efficiency of manipulating a large number of particles or larger objects. For example, Zhang et al. introduced a microrobot that relies on OET to drive. By constructing an irregularly shaped light field, it can drive a submillimeter-scale microrobot to successfully complete a series of complex tasks such as loading, transportation, and delivery. This research has been proven to be used in the fields of cell isolation, RNA sequencing, operation in closed systems, and separation of tissues from heterogeneous mixtures.84 In addition, DLP technology also has unique advantages, being able to spontaneously support separate optical modes of different photon frequencies in the same scanning field. This feature provides strong support for the subsequent application of frequency coding modes to be explored.
3.4 Optical machinery drive
In section 3.1.2, we introduced the driving of liquid crystal soft miniature robots based on the photothermal effect. Here, we will introduce another driving mechanism, namely optical mechanical drive (Table 2). The photomechanical effect refers to the geometric transformation of a material by absorbing light energy after an optical mechanical switch (such as azobenzene) is incorporated into an LCE material. When photomechanical switches such as azobenzene are excited by light, they will have a stress effect on the surrounding LCE molecular network. This stress destroys the original orderly molecular arrangement, which in turn causes the volume of the material to change at the macroscopic scale, and realizes an energy conversion process from the microscopic molecular conformation change to the macroscopic morphological response. Wu et al. reported a double-layer actuator composed of poly(ε-caprolactone) polyurethane urea (PAzo) and polyimide film, and introduced an optical switching group (i.e. azobenzene derivative) in the main chain of PAzo (Fig. 16A-a). By changing the intensity of the incident light, a programmable bending mode is displayed. Wu et al. showed a bionic finger that plays the piano on a smartphone (Fig. 16A-b). This research opened the door to the development of repeatable optical drive actuators.85 In another study, Tahir et al. made a miniature gripper based on anthracene crystals, which can manipulate and carry objects under optical drive (Fig. 16B).86| Driving mechanisms | Advantages | Disadvantages |
|---|---|---|
| Photothermal drive | - The response speed is relatively fast and can be regulated in real time | - Thermal diffusion easily leads to low spatial accuracy, making it difficult to achieve micro-nano level precise control |
| - The driving energy source is simple | - Continuous heating may cause material degradation or thermal damage | |
| - Energy conversion efficiency is relatively low (part of the thermal energy is dissipated) | ||
| Photochemical drive | - The driving threshold is low and the light intensity requirement is mild | - By-products may occur during the reaction process |
| - Precise control at the molecular level can be achieved | - The light penetration depth is relatively shallow | |
| Optical tweezers drive | - Extremely high spatial accuracy | - The driving range is small (only near the laser focus) |
| - Suitable for precise manipulation of biological samples | - The equipment cost is high | |
| Optical machinery drive | - Fast response speed | - The driving force of light pressure is relatively small |
| - No chemical reaction during the actuation process | - High-power lasers are required | |
| - Wide material compatibility | - High precision of the optical system is demanded | |
| Biological drive | - Excellent biocompatibility, which is suitable for in vivo or bio-interface applications | - Poor driving controllability |
| - High energy conversion efficiency | - Weak environmental adaptability (requiring suitable temperature, pH, etc.) |
 | ||
| Fig. 16 (A) (a) Optical switching group in the main chain of PAzo. (b) Bionic finger plays the piano on a smartphone (reproduced from ref. 85 with permission from ACS Publications, copyright 2024). (B) (a) Schematic of a gripper. (b) Gripper grasps an object (reproduced from ref. 86 with permission from ACS Publications, copyright 2024). (C) (a and b) Film spontaneously oscillates under sunlight (reproduced from ref. 87 with permission from Nature Communications, copyright 2016). (D) Self-continuous oscillation behavior caused by sunlight exposure (reproduced from ref. 88 with permission from Wiley Online Library, copyright 2017). (E) (a) Spiral structure of plant tendrils. (b) Three basic self-oscillation modes (reproduced from ref. 89 with permission from Nature Communications, copyright 2021). | ||
In recent years, the continuous drive of soft mechanical microrobots has become a key research direction in the field of scientific research. Among the many research results, the oscillation drive mode of a soft microrobot made based on LCE is particularly outstanding, and the self-oscillating system can realize autonomous and continuous movement under constant and continuous stimulation. This driving method can not only significantly improve the movement speed of the robot, but also has the advantage of easy control. Kumar et al. developed an LCE film doped with visible light in response to fluorinated azobenzene. Experiments have found that the film can spontaneously produce continuous oscillating motion under sunlight (Fig. 16C). The research further revealed that the synergistic irradiation of blue and green light is the key element that triggers the oscillation behavior. This research provides a reference for the development of autonomous and continuous movement of miniature robots, as well as self-cleaning surface technology driven by sunlight, showing great application potential in the field of energy collection and intelligent materials.87 In another study, Gelebart et al. proposed the self-continuous oscillation behavior of oblique liquid crystal networks caused by sunlight exposure (Fig. 16D). The research expands the driving light source from traditional UV light to the field of NIR light, creating conditions for the large-scale use of sunlight into mechanical work.88 In addition, Hu et al. built an optical tunable self-oscillating system, which exhibits a variety of oscillation modes and excellent load capacity. The design of the system is inspired by the spiral structure of plant tendrils, which is prepared using a torsion-free strategy, giving the system a very high DOF of movement (Fig. 16E-a). The results show that the system can accurately and controllably generate three basic self-oscillation modes (i.e. rotation oscillation, tilt oscillation, and up and down oscillation) (Fig. 16E-b). The optical self-oscillating system is expected to play an important role in the field of self-sustaining machinery and equipment.89
When an LCE microrobot driven by the photothermal effect is running in a water environment, the energy dissipation is significant, which will cause a large spatial temperature gradient to form around the microrobot. To produce a photothermal effect sufficient to drive the movement of microrobots, high-intensity light support is often required. In contrast, photomechanical drive technology effectively avoids these limitations. This driving mechanism mainly depends on the interaction between matter and light, and is almost unaffected by environmental factors such as the ion intensity of the surrounding medium.
3.5 Biological drive
The use of biological carriers to manufacture miniature robots has become a very promising technology path. Organisms not only have excellent biocompatibility and degradability, but their tiny size and easy-to-manipulate characteristics provide natural advantages for the research and development of miniature robots. The light-oriented nature of many creatures, that is, their instinct to spontaneously produce movement under the stimulation of light, has brought new inspiration to the control of miniature robots. From the point of view of biological categories, such organisms with photosensitive properties, they can be divided into two types, bacteria and algae. By combining biological characteristics with optical technology, researchers have successfully achieved precise control of the movement of biological objects, laying a solid foundation for the development of new biological hybrid microrobots. | ||
| Fig. 17 (A) Micromotor carries drugs for targeted tumor administration (reproduced from ref. 90 with permission from Wiley Online Library, copyright 2025). (B) Micromotor based on thin-leaf algae performs a variety of tasks in the narrow microenvironment of an organism (reproduced from ref. 91 with permission from Light: Science & Applications, copyright 2024). (C) Micromotor based on Chlamydomonas reinhardtii (reproduced from ref. 92 with permission from Wiley Online Library, copyright 2024). (D) (a) Fabrication of a microrobot based on Chlorella loaded with magnetic particles. (b) Microrobot degrades methylene blue and tetracycline hydrochloride under visible light (reproduced from ref. 93 with permission from Elsevier, copyright 2023). (E) Deep tissue penetration characteristics of NIR light (reproduced from ref. 94 with permission from Elsevier, copyright 2025). (F) (a) Miniature robot system. (b) Miniature robot independently completes a series of complex tasks (reproduced from ref. 95 with permission from Wiley Online Library, copyright 2023). | ||
In the case of flagellate-free microalgae, a variety of driving methods can be coupled to enable them to move autonomously. Chlorella is a typical flagella-free single-celled microalga. Gong et al. proposed a microrobot based on Chlorella loaded with magnetic particles. In this study, Chlorella was used as a biological template, in which magnetic Fe3O4 and BiOX nanoparticles with good photocatalytic properties were deposited sequentially (Fig. 17D-a). This research shows that under visible light, microrobots can degrade methylene blue and tetracycline hydrochloride (Fig. 17D-b). In addition, they have a flexible propulsion ability under the effect of an external magnetic field. This research provides a reference for the construction of a microrobot platform to explain pollutant reduction.93
In a highly competitive living environment, the athletic ability of bacteria is not only the key to their survival, but also closely related to important physiological activities such as biofilm development. The precise regulation of the movement mechanism of bacteria is of great significance for an in-depth understanding of these complex life processes. Although optogenetic tools have been widely used in the control and research of bacterial movement, most of the excitation light of the prior art is limited to the visible spectrum. Xu et al. made miniature robots based on E. coli, which brought a breakthrough. The system is composed of genetically engineered E. coli and orthogonal emission upconverting nanoparticles, which can respond to NIR optical signals. By switching between different wavelengths of NIR light irradiation, the bacteria can be accurately controlled to transition between the two motion states of tumbling and swimming (Fig. 17E). The deep tissue penetration characteristics of NIR light enable this miniature robot to show great application potential in the biomedical field such as targeted drug delivery, and it is expected to open up a new path for the precise treatment of diseases.94
In another study, Pellicciotta et al. used optical drive bacteria as the power source and used 3D printing technology to create a unique miniature robot system (Fig. 17F-a). By regulating the light intensity, the microrobot components can be accurately controlled. To achieve efficient control, an optimized feedback loop is designed. Based on the real-time position and direction information of each microrobot, the central computer calculates and projects a customized light pattern to ensure that each microrobot can receive exclusive control instructions. Based on this, multiple miniature robots can independently complete a series of complex tasks and shuttle between pre-set checkpoints in an orderly manner (Fig. 17F-b). This research shows excellent energy utilization efficiency and scale control potential, which is expected to open a new chapter of application in the fields of biomedicine and environmental monitoring.95
4. Applications
4.1 Environmental restoration
Since the Industrial Revolution, the global industrialization process has accelerated, and a large amount of polluted water has been discharged into natural water bodies such as lakes, rivers, and oceans, posing a serious threat to the ecological environment. Traditional water pollution control methods, such as the placement of detergents in polluted areas and the laying of filter membranes, have certain results, but they have limitations in terms of treatment efficiency and energy consumption. In contrast, miniature motors, especially optical-driven miniature motors, show significant advantages. This type of micro-equipment can remove biological and chemical pollutants with higher efficiency and lower energy consumption, especially in areas that are difficult to reach by conventional methods such as deep pipelines and complex underwater terrain. It has opened up a new technological path for water pollution control and is expected to become key technology to improve the quality of the water environment.In the field of environmental restoration, high-efficiency photocatalysts have become the core technical means of degrading antibiotics and pollutants in water. In recent years, wastewater treatment plants in industrial parks have faced challenges in effectively removing emerging pollutants such as sulfonamide antibiotics, as traditional treatment processes are insufficient to degrade these persistent compounds. To address this issue, Zhang et al. developed a visible light-driven micromotor based on a photocatalyst. This device realizes autonomous propulsion through photocatalytic degradation reaction, which provides an innovative path for water purification (Fig. 18A). This research shows that this micromotor exhibits excellent photocatalytic activity against common pollutants such as sulfonamides, quinolones, and tetracycline antibiotics in water. The high activity of the photocatalyst is mainly due to the modification of carbon quantum dots (CQDs), which significantly enhances the visible light absorption capacity of the material. In addition, the two-dimensional Bi2WO6 nanosheets are inserted into the special structure formed by the flower-like BiOBr, which not only greatly increases the active reaction site, but also strengthens the adsorption of antibiotics molecules. This research shows that the structural engineering design based on CQD modification, which is expected to develop miniature robots with autonomous mobility capabilities, provides new technical solutions for the efficient removal of bacterial pollutants in water, showing broad application prospects in the field of water pollution control.96 Oral et al. proposed an enzyme-immobilized self-propelled zinc oxide (ZnO) microrobot for degrading antibiotics in water bodies (Fig. 18B). Au is introduced in the miniature robot, giving it an “on/off” switching function for light-controlled motion, which realizes flexible control of the motion state. Through physical adsorption technology, the enzyme is stably modified on the surface of the microrobot. Using oxytetracycline (OTC) as a typical pollutant model, Oral et al. explored the pollutant removal performance of the enzyme-immobilized microrobot. The results showed that the successful fixation of the enzyme on the surface of the microrobot significantly enhanced the oxidative ability for the degradation of antibiotics and effectively improved their removal efficiency. This strategy of combining enzyme-immobilized technology with optical-driven microrobots provides a green and efficient new way for the treatment of microbial pollution in water bodies and is expected to play an important role in the field of environmental restoration.97
 | ||
| Fig. 18 (A) Visible light-driven micromotor based on a photocatalyst (reproduced from ref. 96 with permission from ACS Publications, copyright 2022). (B) Enzyme-immobilized self-propelled ZnO microrobot (reproduced from ref. 97 with permission from Wiley Online Library, copyright 2022). (C) Low-energy light-responsive magnetic-assisted cleaning microrobot (reproduced from ref. 98 with permission from ACS Publications, copyright 2024). (D) Optical-driven magnetic-assisted microrobot (reproduced from ref. 99 with permission from Wiley Online Library, copyright 2023). (E) Photoactive BiOI flower-like micromotor (reproduced from ref. 100 with permission from Elsevier, copyright 2022). (F) (a) Fabrication of plasma MXene hydrogel. (b) Movement of micromotor (reproduced from ref. 101 with permission from Wiley Online Library, copyright 2024). | ||
Microplastic pollution has become a serious problem that cannot be ignored in the global ecosystem. Its traces are all over the ocean, soil, and even the atmosphere and other environmental media, posing a continuous threat to marine biodiversity and human health. Although traditional treatment methods, such as chemical flocculation and physical filtration, play a role in the treatment of macroscopic pollutants, these methods have obvious limitations in the face of microplastic particles at the micron or even nano level, making it difficult to achieve their efficient removal. As the problem of microplastics pollution continues to intensify, there is an urgent need to explore more targeted governance strategies to meet the complex challenges posed by microplastic pollution. Wang et al. developed a low-energy light-responsive magnetic-assisted cleaning microrobot. The device is composed of a photocatalytic material, Ag@Bi2WO6, and magnetic nanoparticle, Fe3O4, which provides a solution for the treatment of microplastic pollution in the water environment (Fig. 18C). Driven by the diffusion electrophoresis effect, the Ag@Bi2WO6 polymer material spontaneously assembles to form a low-energy light-responsive clean microrobot, which can continuously adsorb microplastics in water bodies. Studies have shown that the effective effect of microrobots on microplastics is more than 100 microns away, showing strong adsorption properties. Based on this, through the introduction of Fe3O4 nanoparticles, the microrobots gained the ability to manipulate a magnetic field, which can realize clustering operations and efficiently clean up microplastic pollutants in water bodies. With precise individual operation and group collaborative control technology, the microrobot can achieve a 98% pollutant removal rate within 93 seconds. After the cleanup is completed, the microrobot can also be easily recovered under the guidance of a directional magnetic field to avoid secondary pollution. This environmentally friendly and energy-saving microrobot provides a promising new path to meet the challenges of microplastic pollution and promote the innovation of industrial microplastic treatment technology, which is expected to play an important role in the field of environmental restoration.98 In another study, Dekanovsky et al. developed an optical-driven magnetic-assisted microrobot that provides an efficient solution for bisphenol A (BPA) pollution control (Fig. 18D). This microrobot can be embedded with Fe3O4 NPs as a magnetic propulsion engine, giving it the ability to move flexibly in a magnetic field. The grafted bismuth NPs significantly improve the photocatalytic efficiency. The results show that the microrobot exhibits excellent pollutant handling capabilities. Within 10 minutes, 60% of BPA can be removed; after 1 hour of continuous action, the removal rate is almost 100%. Through photocatalytic degradation, BPA is efficiently mineralized into carbon dioxide and water, which fully reflects the significant advantages of this technology in green environmental protection treatment and opens up a new path for the treatment of BPA pollution in water bodies.99
With the deepening of the public awareness of the hazards of excessive sun exposure, the use of sunscreen is becoming more common. However, the extensive use of sunscreen has led to an increase in the concentration of related pollutants in the environment, causing ecological problems that cannot be ignored. In response to this global challenge, Mayorga-Burrezo et al. proposed a sunscreen residue treatment scheme based on photomagnetic response microrobots. They prepared photoactive BiOI flower-like particles, and through surface functionalization treatment, Fe3O4 nanoparticles were modified on the surface of the particles to give the microrobot magnetic responsiveness (Fig. 18E). Experiments have shown that under the irradiation of a pure water system and visible light, the miniature robot exhibits excellent photocatalytic degradation properties. Its outstanding performance is due to the synergistic effect of adsorption and photocatalytic activity. On the one hand, its special structure enhances its adsorption capacity for pollutants. On the other hand, its photoactive materials effectively drive the degradation reaction. Of particular concern is that during the treatment of BP-3, a common ingredient in sunscreen, the magnetic drive enables the miniature robot to dynamically navigate in the water body, which significantly improves its adsorption efficiency for BP-3, thereby greatly enhancing the photocatalytic degradation effect. This research provides new ideas for solving the problem of environmental pollution and shows the huge application potential of photomagnetic response microrobots in the field of environmental restoration.100
Yang et al. used the rapid gelation properties of degradable alginate to successfully synthesize a plasma MXene hydrogel by combining bimetal nano-cubes with Ti3C2Tx MXene (Fig. 18F-a). The hydrogel exhibits excellent pollutant handling ability, which can effectively adsorb more than 60% of dye pollutants in just 2 minutes, whose final removal rate is as high as 90% or more. Yang et al. further integrated a heat-sensitive polymer and superparamagnetic particles into the hydrogel to create a miniature robot with multiple functions and strong light responsiveness. Relying on the dual-drive mechanism of magnetic force and optics, this robot can accurately identify and efficiently remove pollutants in a complex maze-like channel environment (Fig. 18F-b). This miniature robot has broad application prospects and technical advantages in dealing with pollutants in complex real-world environments.101
4.2 Biomedicine
In the field of biomedical research and application, miniature robots, with their tiny size advantages, have shown unique value in scenarios such as drug delivery and cell intervention. Self-propelled miniature robots have the ability to move quickly, can be accurately controlled, and penetrate biological tissues. They can break through the limitations of traditional drug delivery methods and deliver drugs accurately to the target site, opening up a new path for efficient and targeted therapy. However, compared with the open and controllable environment of the laboratory, the blood environment poses severe challenges to the operation of miniature robots. Blood not only has a high viscosity, but it is also in a state of continuous and rapid flow, which makes the autonomous movement of miniature robots difficult. Although traditional chemically driven miniature robots can move at high speed, their chemical raw materials are often toxic, and this limitation makes it impossible to safely apply them in the intravascular environment. In the context of pursuing non-toxic and non-invasive drive, optical drive shows unique advantages. This method can realize remote and precise control of the movement of cells or miniature robots without physical contact. Optical-driven miniature robots have high instantaneous driving energy and can achieve high-speed movement. This characteristic makes them the preferred solution for biomedical miniature robot design. In addition, some miniature robots with photothermal effects can quickly increase the ambient temperature under NIR light and destroy diseased cells and tissues through thermal ablation mechanism, providing non-invasive innovative treatment strategies for disease interventions such as tumor treatment. Fu et al. reported a highly controllable test tube-shaped gold-copper micromotor cluster system. This micromotor can realize autonomous drive under NIR light and spontaneously gather into a group structure (Fig. 19A). The gold section has good photothermal properties and efficiently converts light energy into heat. The high temperature generated by the micromotor cluster center can be directly applied to photothermal therapy to accurately destroy tumor cells. This research not only builds a high-precision and controllable technology platform for the field of tumor photothermal therapy, but also provides new solutions and technical ideas for minimally invasive cancer treatment, which is expected to promote the development of tumor treatment methods in the direction of precision and efficiency.16 In another study, Jiao et al. performed gold sputtering treatment on the surface of mesoporous silica to develop a nanocarrier with a Janus structure (Fig. 19B). Relying on the excellent photothermal properties of the Au surface, the nanocarrier achieves efficient NIR optical drive propulsion, which significantly enhances its ability to interact with cells. At the same time, the porous structure of mesoporous silica greatly improves the drug loading efficiency. The nanocarrier penetrates the cell membrane through thermomechanical effects and realizes controlled drug release, effectively promoting the deep penetration of drugs into tumor tissue. This research not only significantly improved the efficiency of using nanocarriers, but also reduced the exposure of drugs to non-target tissues and reduced potential side effects.102 | ||
| Fig. 19 (A) Micromotors gather into a group structure under NIR light (reproduced from ref. 16 with permission from Elsevier, copyright 2022). (B) Nanocarrier with Janus structure (reproduced from ref. 101 with permission from Elsevier, copyright 2020). (C) Urease-based immobilized TiO2/CD nanotubes (reproduced from ref. 21 with permission from Wiley Online Library, copyright 2022). (D) TiO2@N-Au NW motor driven by visible light (reproduced from ref. 17 with permission from Elsevier, copyright 2022). | ||
Urinary tract infections are a global health problem that plagues millions of people. This bacterial infection is mainly caused by E. coli. Owing to the formation of biofilms on the surface of the bladder or catheter, it is often difficult for conventional treatments to completely remove stubborn biofilms. In response to this problem, Villa et al. developed urease-based immobilized TiO2/CD nanotubes (Fig. 19C). The microrobot uses urea as a biocompatible fuel, swimming autonomously in the urine environment, and responding to visible light. The results showed that after 2 hours of visible light irradiation, the clearance rate of the microrobot to the bacterial biofilm was as high as 90%. The mechanism is that the microrobot robot produces a large number of active free radicals through photocatalytic reaction, which effectively destroys the biofilm structure. This research provides new ideas for the design of high-performance optical-driven microrobots, which have shown great application potential in the treatment of microbial infections, and are expected to provide innovative solutions for solving bacterial biofilm-related diseases.21
In the field of ophthalmological drug treatment, the traditional method of administration based on passive dispersion faces many uncertainties and challenges. Among them, the complex vitreous structure and the presence of biological macromolecules in it seriously limit the effective penetration of drugs and hinder the improvement of therapeutic effects. In response to this problem, Chen et al. developed a TiO2@N-Au nanowire (NW) motor driven by visible light (Fig. 19D). The nanorobot is based on the photoelectric swimming mechanism and can realize autonomous and efficient movement in a vitreous environment. Its nanoscale size is highly matched with the voids of the vitreous body network, and it can penetrate the complex non-uniform non-Newtonian fluid environment of the vitreous body noninvasively to achieve deep drug delivery. This research has opened up a new path for eye treatment and provided a potential technical solution to solve the problem of eye drug delivery.17
4.3 Micro-cargo transportation
In the field of miniature engineering and assembly, just like the logistics system of a real factory, a large number of miniature-scale “goods” urgently need to be accurately loaded and transported efficiently. As the key components of miniature devices, the operating accuracy of these “goods” directly determines the success or failure of micro-manufacturing technology. With the development of miniature technology in the direction of precision and integration, miniature motors that can achieve precise control have become the core elements to break through technological bottlenecks. They not only need to have stable power output and flexible motion control capabilities, but also need to meet the multi-function requirements in complex task scenarios. This is not only an inevitable requirement for the development of micro-manufacturing technology, but also a key driving force to promote miniature engineering to a higher level.Peng et al. used hydrothermal synthesis to prepare Janus hematite/Pt microrobots with cube and walnut shapes. Both microrobots showed motor ability under light irradiation. Due to the asymmetric orientation of the magnetic dipole moment in the crystal, the cube microrobots can self-assemble into ordered micro-chains, while the walnut-shaped microrobots gather randomly. Microchains exhibit different forms of movement under light irradiation, allowing them to capture, pick up, and transport micro-scale objects (such as yeast cells) (Fig. 20A). This photodynamic self-assembled micro-chain has great potential in cargo capture, transportation and delivery.103 Song et al. developed a micromotor with efficient motion and load capacity. The thermal Marangoni convection energy caused by laser focused irradiation can effectively drive the micromotor. By adjusting the laser irradiation position, the micromotor can exhibit different motor behaviors. In the experiment, yeast cells, as cargo, can be effectively captured and transported by micromotors, while maintaining high cell vitality (Fig. 20B). These micromotors provide a way for the development of intelligent micromotors.104
 | ||
| Fig. 20 (A) Microchains exhibit different forms of movement under light irradiation, allowing them to capture, pick up, and transport yeast cells (reproduced from ref. 103 with permission from ACS Publications, copyright 2022). (B) Micromotor captures and transports yeast cells (reproduced from ref. 104 with permission from Elsevier, copyright 2022). (C) (a) Two types of microrobots are driven using a non-focused circularly polarized laser beam. (b) Schematic of a miniature drone (reproduced from ref. 105 with permission from Elsevier, copyright 2022). (D) (a) Soft robot picks up goods. (b) Soft robot transports goods (reproduced from ref. 106 with permission from Wiley Online Library, copyright 2020). (E) (a) Two driving modes. (b) Dual-mode driven microship. (c) Microship realizes intelligent transportation (reproduced from ref. 107 with permission from Wiley Online Library, copyright 2023). | ||
In addition to photodynamic-driven microrobots for intelligent transportation, micro-robots similar to grippers and ships have also been developed for transporting goods. Qin et al. added an optical-driven manipulator to a miniature drone (Fig. 20C). Through optical tweezers technology, the manipulator can accurately capture and transport goods.105 In addition, a miniature soft robot based on light-responsive LCE was created by Pilz da Cunha et al., capable of picking up, transporting, and delivering goods driven by blue light (Fig. 20D). Through the coordinated movement of the “legs”, the miniature robot can move in any direction. Through the picking and releasing mechanism of the “arm” of the miniature robot, the transportation of goods was demonstrated.106 Hou et al. reported a dual-mode driven microship based on the Marangoni effect and a steam jet (Fig. 20E-a and b). Driven by light, the microship not only realizes multi-modal movement such as forward, backward, and rotation at the gas–liquid interface, but also realizes intelligent transportation (Fig. 20E-c).107
5. Conclusions
In the wireless energy supply and control technology of micro-scale devices, light has become one of the most commonly used driving medium due to its advantages of non-contact, precise control and remote operation. Photodynamic microrobots developed based on photosensitive materials and photothermal materials have attracted much attention in recent years and have shown unique application value in many fields such as environmental governance, biomedicine, and miniature transportation.This review systematically combines the core material composition and driving mechanism of photodynamic microrobots, and combines specific research cases to discuss their application and practice in scenarios such as environmental protection, transportation, and biomedical research in depth, thus providing ideas for the innovation and development of this technology in the multi-disciplinary intersection context. It has been found that the material properties of miniature robots significantly affect their applicable environment. Hydrogel-based light-mechanical soft microrobots are more suitable for liquid environments due to their good hydrophilicity and swelling characteristics. LCE mechanical soft microrobots rely on their special molecular orientation and mechanical responsiveness, and can operate stably in both water and gas environments. However, microrobots based on photochemical effects, as well as microrobots that rely on optical tweezers and dielectric electrophoresis (DEP) to drive, are limited by the principle of action and material properties, and can only function in a liquid phase environment. These characteristic differences provide an important basis for accurately matching the application scenarios of microrobots and help promote the development of photodynamic microrobot technology in a more specialized and refined direction.
Although photodynamic microrobots have shown great potential in many fields, their development and practical application still face many technical bottlenecks. In terms of material selection and manufacturing, the micro-nano processing of LCE materials requires high-precision lithography or 3D printing technologies, but their large-scale production is faced with problems such as poor structural consistency and high costs. The photopolymerization process of hydrogels is susceptible to solvent evaporation and uneven cross-linking, leading to discrete mechanical properties of microstructures, which makes it difficult to meet the functional consistency requirements of mass production. In the case of LCEs, soft lithography technology is used to prepare high-precision master molds, and thermal imprinting or solvent evaporation is used to induce the self-assembly of the LCE precursors to form microstructures, solving the efficiency bottleneck of 3D printing. In the case of hydrogels, when UV light is used to initiate cross-linking, supercritical fluids are used as solvent substitutes to avoid the evaporation gradient of traditional solvents.
Furthermore, microrobots that rely on optical micro-manipulation technology usually have a driving force in the order of nano-newtons to piconewtons, making it difficult to drive larger and heavier devices, which greatly limits the expansion of their application scenarios. In addition, these technologies often require complex and expensive equipment, which leads to high research and use costs, hindering the popularization and promotion of technology. To break through these limitations, DEP technology was introduced, with higher driving force and lower light source requirements, providing a new way to control large miniature robots.
The limitations of the ability of light field regulation are also a key problem that needs to be solved urgently. Compared with a magnetic field and sound field, which can realize the precise positioning of objects in three-dimensional space, the light field is currently mainly used for two-dimensional plane control. It has been found that by controlling the light intensity conversion, the spatial movement of the micromotor can be affected to realize its control in different spatial positions. In addition, the poor penetration of light in opaque media (such as blood) seriously restricts its application in the field of biomedicine. At present, the introduction of auxiliary drive methods has become the key to breaking the game. Among them, magnetic field drive has become a very potential solution due to its non-invasive and relatively safe characteristics. The penetration depth of visible/NIR light in biological tissues is typically affected by scattering and absorption. For example, the penetration depth of visible light in soft tissues is less than 1 mm, while that of NIR light can reach 1–2 cm, which limits their applications in deep tissues. By designing core-shell structured nanomaterials, the light absorption efficiency can be enhanced through the surface plasmon resonance effect, reducing the light-driven threshold and the dependence on high-penetration light. However, for specific application scenarios, the integration of innovative technologies is often more advantageous. Furthermore, under high-frequency light modulation, the energy storage and release lag of optical-driven robots leads to a decline in motion trajectory accuracy, failing to meet the real-time control requirements in complex environments. Reducing the heat capacity can mitigate energy storage lag. For example, 3D printing hollow graphene skeletons used to load photothermal materials can enhance the thermal response speed by several times, with minimal temperature fluctuations under high-frequency light modulation.
Although some studies have proposed solutions to the above-mentioned challenges, due to factors such as insufficient technology maturity and complex operation processes, whether these technologies can be successfully applied to clinical medicine still needs further verification. However, with the collaborative research of global scientific research forces, photodynamic microrobotics technology is expected to continue to break through bottlenecks and play a more important role in the fields of biomedicine, environmental governance, and miniature manufacturing in the future, bringing new opportunities and possibilities for interdisciplinary innovation.
Author contributions
X. W. and Y. G. proposed the original idea and planned the configuration; X. W. and A. L. wrote the manuscript; W. Y., S. J. and Y. W. revised the paper for language and quality. All authors have read and agreed to the published version of the manuscript.Conflicts of interest
There are no conflicts to declare.Data availability
No primary research results, software or code have been included and no new data were generated or analysed as part of this review.References
- Y. Chen, et al., Recent Advances in Field-Controlled Micro–Nano Manipulations and Micro–Nano Robots, Adv. Intell. Syst., 2022, 4(3), 2100116 CrossRef
.
- B. Wang, Y. Zhang and L. Zhang, Recent progress on micro- and nano-robots: towards in vivo tracking and localization, Quant. Imaging Med. Surg., 2018, 8(5), 461–479 CrossRef PubMed
- N. Hu, et al., Comprehensive modeling of corkscrew motion in micro-/nano-robots with general helical structures, Nat. Commun., 2024, 15(1), 7399 CrossRef CAS
- B. Chen, et al., Cell-Based Micro/Nano-Robots for Biomedical Applications: A Review, Small, 2024, 20(1), 2304607 CrossRef CAS
- M. Li, et al., An Overview of Recent Progress in Micro/Nanorobots for Biomedical Applications, Adv. Mater. Technol., 2023, 8(11), 2201928 CrossRef CAS
- H. Zhang, et al., Review of the Applications of Micro/Nanorobots in Biomedicine, ACS Appl. Nano Mater., 2024, 7(15), 17151–17192 CrossRef CAS
- J.-y. Hou, et al., Recent advances in micro/nano-robots for environmental pollutant removal: Mechanism, application, and prospect, Chem. Eng. J., 2024, 498, 155135 CrossRef CAS
- M. Urso, M. Ussia and M. Pumera, Smart micro- and nanorobots for water purification, Nat. Rev. Bioeng., 2023, 1(4), 236–251 CrossRef CAS PubMed
- Y. Gan, B. Wang and Y. Lu, Micro/Nanorobotics in Environmental Water Governance: Nanoengineering Strategies for Pollution Control, Small Struct., 2025, 2500058 CrossRef
- L. Xu, et al., Light-Driven Micro/Nanorobot for Biomimetic Optical Communication, IEEE Robot. Autom. Lett., 2024, 9(3), 2287–2294 Search PubMed
- Y. Tao, et al., Four-Dimensional Micro/Nanorobots via Laser Photochemical Synthesis towards the Molecular Scale, Micromachines, 2023, 14(9), 1656 CrossRef
- Y. Gao, et al., Light hybrid micro/nano-robots: From propulsion to functional signals, Nano Res., 2022, 15(6), 5355–5375 CrossRef
- S. Preetam, et al., Empowering tomorrow's medicine: energy-driven micro/nano-robots redefining biomedical applications, Mol. Syst. Des. Eng., 2024, 9(9), 892–911 RSC
- J. Wang, Z. Xiong and J. Tang, The Encoding of Light-Driven Micro/Nanorobots: from Single to Swarming Systems, Adv. Intell. Syst., 2021, 3(4), 2000170 CrossRef
- M. Ebrahimi, et al., Graphene self-powered micro/nano-robots from fascinating research samples to commercial products: Energy supply mechanisms and bio-sensing applications, Carbon, 2024, 230, 119659 CrossRef CAS
- S. Fu, et al., Light driven micromotor swarm for tumor photothermal therapy, Appl. Mater. Today, 2022, 26, 101348 CrossRef
- B. Chen, et al., Visible-light-driven TiO2@N-Au nanorobot penetrating the vitreous, Appl. Mater. Today, 2022, 27, 101455 CrossRef
- Y. Liu, et al., A Light-Driven Carbon Nanocoil Microrobot, Coatings, 2024, 14(8), 926 CrossRef CAS
- W. Xiong, et al., Light-driven Lattice Soft Microrobot with Multimodal Locomotion, 2024.
- L. Xu, et al., An insect larvae inspired MXene-based jumping actuator with controllable motion powered by light, Nano Energy, 2022, 103, 107848 CrossRef CAS
- K. Villa, et al., Enzyme-Photocatalyst Tandem Microrobot Powered by Urea for Escherichia coli Biofilm Eradication, Small, 2022, 18(36), 2106612 CrossRef CAS PubMed
- P. Mayorga-Burrezo, C. C. Mayorga-Martinez and M. Pumera, Light-Driven Micromotors to Dissociate Protein Aggregates That Cause Neurodegenerative Diseases, Adv. Funct. Mater., 2022, 32(1), 2106699 CrossRef CAS
- R. Luo, et al., Gradient Porous Elastic Hydrogels with Shape-Memory Property and Anisotropic Responses for Programmable Locomotion, Adv. Funct. Mater., 2015, 25(47), 7272–7279 CrossRef CAS
- H. Guo, et al., Light-Mediated Shape Transformation of a Self-Rolling Nanocomposite Hydrogel Tube, ACS Appl. Mater. Interfaces, 2020, 12(11), 13521–13528 CrossRef CAS
- Y. Liang, Y. Shen and H. Liang, Solvent-responsive strong hydrogel with programmable deformation and reversible shape memory for load-carrying soft robot, Mater. Today Commun., 2022, 30, 103067 CrossRef CAS
- M.-S. Kim, et al., Surface Nanopatterned Shape Memory Alloy (SMA)-Based Photosensitive Artificial Muscle, Adv. Opt. Mater., 2022, 10(5), 2102024 CrossRef CAS
- M. D. Guiry, How many species of algae are there? A reprise. Four kingdoms, 14 phyla, 63 classes and still growing, J. Phycol., 2024, 60(2), 214–228 CrossRef
- K. Cheng, et al., Progress on control of harmful algae by sustained-release technology of allelochemical: A review, Sci. Total Environ., 2024, 918, 170364 CrossRef CAS PubMed
- S. Liu, et al., Performance, mechanism regulation and resource recycling of bacteria-algae symbiosis system for wastewater treatment: A review, Environ. Pollut., 2024, 362, 125019 CrossRef CAS
- T. DaPalma, et al., A systematic approach to virus–virus interactions, Virus Res., 2010, 149(1), 1–9 CrossRef CAS PubMed
- W. Henle and G. Henle, Seroepidemiology of the Virus, in The Epstein-Barr Virus, ed. M. A. Epstein and B. G. Achong, Springer Berlin Heidelberg, Berlin, Heidelberg, 1979, pp. 61–78 Search PubMed
- P. F. M. Teunis, et al., Norwalk virus: How infectious is it?, J. Med. Virol., 2008, 80(8), 1468–1476 CrossRef PubMed
- M.-S. Kim, H.-T. Lee and S.-H. Ahn, Laser Controlled 65 Micrometer Long Microrobot Made of Ni-Ti Shape Memory Alloy, Adv. Mater. Technol., 2019, 4(12), 1900583 CrossRef CAS
- X. Zhang, et al., Near-Infrared Light-Driven Shape-Programmable Hydrogel Actuators Loaded with Metal–Organic Frameworks, ACS Appl. Mater. Interfaces, 2022, 14(9), 11834–11841 CrossRef CAS
- Z. Zhang, et al., Photothermal-Responsive Lightweight Hydrogel Actuator Loaded with Polydopamine-Modified Hollow Glass Microspheres, ACS Appl. Mater. Interfaces, 2024, 16(18), 23914–23923 CAS
- B. Jia, et al., Light-Responsive Soft Robot Integrating Actuation and Function Based on Laser Cutting, Micromachines, 2024, 15(4), 534 CrossRef PubMed
- L. Yang, et al., An Autonomous Soft Actuator with Light-Driven Self-Sustained Wavelike Oscillation for Phototactic Self-Locomotion and Power Generation, Adv. Funct. Mater., 2020, 30(15), 1908842 CrossRef CAS
- C. Xin, et al., Light-triggered multi-joint microactuator fabricated by two-in-one femtosecond laser writing, Nat. Commun., 2023, 14(1), 4273 CrossRef CAS PubMed
- H. Wang, et al., Emulsion Hydrogel Soft Motor Actuated by Thermal Stimulation, ACS Appl. Mater. Interfaces, 2017, 9(49), 43211–43219 CrossRef CAS
- H. G. V. Ho and P. J. Yoo, Dual-Catalysts-Embedded spontaneously propelling asymmetric micromotors using triple emulsion microfluidic synthesis for highly efficient nano/microplastic removal, Sep. Purif. Technol., 2024, 351, 127952 CrossRef CAS
- M. Li, et al., Manipulation Study of Bubble-Propelled Hydrogel Microbots, Adv. Mater. Technol., 2025, 10(8), 2401265 CrossRef CAS
- Y. Chen, et al., High-Speed NIR-Driven Untethered 3D-Printed Hydrogel Microrobots in High-Viscosity Liquids, Adv. Intell. Syst., 2023, 5(3), 2200311 CrossRef
- L. Jing, et al., Dual-mode driven Marangoni actuator with spontaneity and controllability as a soft robot, Sens. Actuators, B, 2025, 436, 137697 CrossRef CAS
- X. Feng, et al., Programmable Multiresponse Marangoni Actuator Enabled by a PINPAM/MWCNT Composite Material, Langmuir, 2025, 41(7), 4884–4893 CrossRef CAS PubMed
- L. Chen, et al., A bamboo/PNIPAM composite hydrogel assembly for both programmable and remotely-controlled light-responsive biomimetic actuations, Chem. Eng. J., 2022, 446, 137072 CrossRef CAS
- Y. Zheng and C. Li, Photothermally-driven oxazine hydrogel actuators at the water-air interface, Giant, 2023, 16, 100192 CrossRef CAS
- S. Watanabe, et al., Multimotion of Marangoni Propulsion Ships Controlled by Two-Wavelength Near-Infrared Light, Langmuir, 2021, 37(50), 14597–14604 CrossRef CAS PubMed
- H. Zeng, et al., Light-Driven, Caterpillar-Inspired Miniature Inching Robot, Macromol. Rapid Commun., 2018, 39(1), 1700224 CrossRef
- C. Zhu, et al., Light-Driven Liquid Crystal Elastomer Actuators Based on Surface Plasmon Resonance for Soft Robots, ACS Appl. Mater. Interfaces, 2024, 16(50), 69858–69869 CrossRef CAS
- Z. Xu, et al., Programmable liquid crystal elastomers with dynamic covalent siloxane bonds based on a dual self-catalytic strategy, Chem. Eng. J., 2023, 476, 146861 CrossRef CAS
- Z. C. Jiang, et al., Selective decrosslinking in liquid crystal polymer actuators for optical reconfiguration of origami and light-fueled locomotion, Angew. Chem., 2019, 131(16), 5386–5391 CrossRef
- H. Guo, A. Priimagi and H. Zeng, Optically Controlled Latching and Launching in Soft Actuators, Adv. Funct. Mater., 2022, 32(17), 2108919 CrossRef CAS
- J. Wang, et al., Leveraging Bioinspired Structural Constraints for Tunable and Programmable Snapping Dynamics in High-Speed Soft Actuators, Adv. Funct. Mater., 2023, 33(2), 2209798 CrossRef CAS
- Z. Wang, et al., A Light-Powered Ultralight Tensegrity Robot with High Deformability and Load Capacity, Adv. Mater., 2019, 31(7), 1806849 CrossRef
- H. Yang, et al., Fabrication and Photothermal Actuation Performances of Electrospun Carbon Nanotube/Liquid Crystal Elastomer Blend Yarn Actuators, ACS Appl. Mater. Interfaces, 2024, 16(7), 9313–9322 CrossRef CAS
- Q. He, et al., Electrospun liquid crystal elastomer microfiber actuator. Science, Robotics, 2021, 6(57), eabi9704 CrossRef PubMed
- J. Ding, et al., Controlled Deformation Mode and Amplitude of Liquid Crystal Actuators through Orthogonal Light and Heat-Induced Exchanges, Angew. Chem., 2025, e202505172 CAS
- W. Wei, et al., Phototriggered Selective Actuation and Self-Oscillating in Dual-Phase Liquid Crystal Photonic Actuators, Adv. Opt. Mater., 2018, 6(15), 1800131 CrossRef
- Z. Zhang, Z. Li and W. Yuan, Plants Inspired Shape-Programmable and Reconfigurable Actuation Soft Actuators for Adaptive Grasping, Sensing and Recognition, Small, 2025, 2501164 CrossRef CAS
- D. Tang, et al., Bio-Mimetic Actuators of a Photothermal-Responsive Vitrimer Liquid Crystal Elastomer with Robust, Self-Healing, Shape Memory, and Reconfigurable Properties, ACS Appl. Mater. Interfaces, 2022, 14(1), 1929–1939 CrossRef CAS PubMed
- O. M. Wani, H. Zeng and A. Priimagi, A light-driven artificial flytrap, Nat. Commun., 2017, 8(1), 15546 CrossRef CAS PubMed
- M. Yang, et al., Bioinspired Phototropic MXene-Reinforced Soft Tubular Actuators for Omnidirectional Light-Tracking and Adaptive Photovoltaics, Adv. Funct. Mater., 2022, 32(26), 2201884 CrossRef CAS
- J. Meng, et al., Pyroelectric Janus nanomotors for synergistic electrodynamic-photothermal-antibiotic therapies of bacterial infections, Acta Biomater., 2023, 162, 20–31 CrossRef CAS PubMed
- B. Okmen Altas, et al., Multi-Stimuli-Responsive Tadpole-like Polymer/Lipid Janus Microrobots for Advanced Smart Material Applications, ACS Appl. Mater. Interfaces, 2024, 16(12), 15533–15547 CrossRef CAS PubMed
- C. Zhou, et al., Photochemically Powered AgCl Janus Micromotors as a Model System to Understand Ionic Self-Diffusiophoresis, Langmuir, 2018, 34(10), 3289–3295 CrossRef CAS PubMed
- C. Zhou, et al., Photochemically Excited, Pulsating Janus Colloidal Motors of Tunable Dynamics, ACS Nano, 2019, 13(4), 4064–4072 CrossRef CAS PubMed
- X. Wang, et al., High-Motility Visible Light-Driven Ag/AgCl Janus Micromotors, Small, 2018, 14(48), 1803613 CrossRef
- Q. Wang, et al., Glucose-Fueled Micromotors with Highly Efficient Visible-Light Photocatalytic Propulsion, ACS Appl. Mater. Interfaces, 2019, 11(6), 6201–6207 CrossRef CAS PubMed
- B. Jang, et al., Multiwavelength Light-Responsive Au/B-TiO2 Janus Micromotors, ACS Nano, 2017, 11(6), 6146–6154 CrossRef CAS PubMed
- S. K. Panda, et al., Engineering Light-Driven Rod-Shaped Micromotors for Exhibiting Controlled and Tunable Multimode Swimming, Adv. Opt. Mater., 2024, 12(21), 2400590 CrossRef CAS
- X. Zhan, et al., Enhanced ion tolerance of electrokinetic locomotion in polyelectrolyte-coated microswimmer, Nat. Commun., 2019, 10(1), 3921 CrossRef PubMed
- Y. Gao, et al., Template-Guided Silicon Micromotor Assembly for Enhanced Cell Manipulation, Angew. Chem., Int. Ed., 2024, 63(28), e202405895 CrossRef PubMed
- X. Peng, et al., Self-Propelled Magnetic Dendrite-Shaped Microrobots for Photodynamic Prostate Cancer Therapy, Angew. Chem., Int. Ed., 2022, 61(48), e202213505 CrossRef PubMed
- C. Zheng, et al., TiO2-Nanoparticle-Shelled Light-Driven Microcleaner for Fast and Highly Efficient Degradation of Organic Pollutants, ACS Appl. Nano Mater., 2022, 5(11), 16573–16583 CrossRef
- Z. Xu, et al., X-ray-Powered Micromotors, ACS Appl. Mater. Interfaces, 2019, 11(17), 15727–15732 CrossRef PubMed
- J. G. S. Moo, S. Presolski and M. Pumera, Photochromic Spatiotemporal Control of Bubble-Propelled Micromotors by a Spiropyran Molecular Switch, ACS Nano, 2016, 10(3), 3543–3552 CrossRef PubMed
- M. J. Villangca, et al., Light-driven micro-tool equipped with a syringe function, Light: Sci. Appl., 2016, 5(9), e16148 CrossRef PubMed
- I. Shishkin, et al., Auxiliary Optomechanical Tools for 3D Cell Manipulation, Micromachines, 2020, 11(1), 90 CrossRef PubMed
- I. Grexa, et al., Single-Cell Elasticity Measurement with an Optically Actuated Microrobot, Micromachines, 2020, 11(9), 882 CrossRef PubMed
- W. Lamperska, et al., Micro-Dumbbells—A Versatile Tool for Optical Tweezers, Micromachines, 2018, 9(6), 277 CrossRef PubMed
- J. Köhler, et al., Optical screw-wrench for microassembly, Microsyst. Nanoeng., 2017, 3(1), 16083 CrossRef PubMed
- G. Vizsnyiczai, et al., Light controlled 3D micromotors powered by bacteria, Nat. Commun., 2017, 8(1), 15974 CrossRef PubMed
- E. Avci, M. Grammatikopoulou and G. Z. Yang, Laser-printing and 3D optical-control of untethered microrobots, Adv. Opt. Mater., 2017, 5(19), 1700031 CrossRef
- S. Zhang, et al., The optoelectronic microrobot: A versatile toolbox for micromanipulation, Proc. Natl. Acad. Sci. U. S. A., 2019, 116(30), 14823–14828 CrossRef CAS PubMed
- L. Wu, et al., Tunable Light-Responsive Polyurethane-urea Elastomer Driven by Photochemical and Photothermal Coupling Mechanism, ACS Appl. Mater. Interfaces, 2024, 16(15), 19480–19495 CrossRef CAS PubMed
- I. Tahir, et al., Photomechanical Crystals as Light-Activated Organic Soft Microrobots, J. Am. Chem. Soc., 2024, 146(44), 30174–30182 CrossRef CAS PubMed
- K. Kumar, et al., A chaotic self-oscillating sunlight-driven polymer actuator, Nat. Commun., 2016, 7(1), 11975 CrossRef CAS
- A. H. Gelebart, et al., Mastering the Photothermal Effect in Liquid Crystal Networks: A General Approach for Self-Sustained Mechanical Oscillators, Adv. Mater., 2017, 29(18), 1606712 CrossRef PubMed
- Z. Hu, Y. Li and J.-a. Lv, Phototunable self-oscillating system driven by a self-winding fiber actuator, Nat. Commun., 2021, 12(1), 3211 CrossRef CAS PubMed
- Z. Wang, et al., Light-Directed Microalgae Micromotor with Supramolecular Backpacks for Photodynamic Therapy, Adv. Funct. Mater., 2025, 35(5), 2411070 CrossRef CAS
- J. Xiong, et al., Light-controlled soft bio-microrobot, Light: Sci. Appl., 2024, 13(1), 55 CrossRef CAS PubMed
- C.-H. Hsiao, et al., Light-Driven Green-Fabricated Artificial Intelligence-Enabled Micro/Nanorobots for Multimodal Phototherapeutic Management of Bladder Cancer, Adv. Healthcare Mater., 2024, 13(32), 2402864 CrossRef CAS
- D. Gong, et al., BiOX-loaded biohybrid magnetic microrobots for enhanced photocatalysis under visible light, Appl. Mater. Today, 2023, 35, 101915 CrossRef
- W. Xu, et al., Extending visual range of bacteria with upconversion nanoparticles and constructing NIR-responsive bio-microrobots, J. Colloid Interface Sci., 2025, 682, 608–618 CrossRef CAS PubMed
- N. Pellicciotta, et al., Light Controlled Biohybrid Microbots, Adv. Funct. Mater., 2023, 33(39), 2214801 CrossRef CAS
- Y. Zhang, Y. Li and Y. Yuan, Carbon Quantum Dot-Decorated BiOBr/Bi2WO6 Photocatalytic Micromotor for Environmental Remediation and DFT Calculation, ACS Catal., 2022, 12(22), 13897–13909 CrossRef CAS
- C. M. Oral, M. Ussia and M. Pumera, Hybrid Enzymatic/Photocatalytic Degradation of Antibiotics via Morphologically Programmable Light-Driven ZnO Microrobots, Small, 2022, 18(39), 2202600 CrossRef CAS PubMed
- Z. Wang, et al., Low-Energy Photoresponsive Magnetic-Assisted Cleaning Microrobots for Removal of Microplastics in Water Environments, ACS Appl. Mater. Interfaces, 2024, 16(45), 61899–61909 CrossRef CAS PubMed
- L. Dekanovsky, et al., Light-Driven MXene-Based Microrobots: Mineralization of Bisphenol A to CO2 and H2O, Small Methods, 2023, 7(8), 2201547 CrossRef CAS
- P. Mayorga-Burrezo, et al., Hybrid magneto-photocatalytic microrobots for sunscreens pollutants decontamination, Chem. Eng. J., 2022, 446, 137139 CrossRef CAS
- K. Yang, et al., A MXene Hydrogel-Based Versatile Microrobot for Controllable Water Pollution Management, Adv. Sci., 2024, 11(26), 2309257 CrossRef CAS
- X. Jiao, et al., NIR powered Janus nanocarrier for deep tumor penetration, Appl. Mater. Today, 2020, 18, 100504 CrossRef
- X. Peng, et al., Shape-Controlled Self-Assembly of Light-Powered Microrobots into Ordered Microchains for Cells Transport and Water Remediation, ACS Nano, 2022, 16(5), 7615–7625 CrossRef CAS
- L. Song, et al., Light-controlled spiky micromotors for efficient capture and transport of targets, Sens. Actuators, B, 2022, 358, 131523 CrossRef CAS
- J. Qin, et al., Light-driven plasmonic microrobot for nanoparticle manipulation, Nat. Commun., 2025, 16(1), 2570 CrossRef CAS
- M. Pilz da Cunha, et al., A Soft Transporter Robot Fueled by Light. Advanced, Science, 2020, 7(5), 1902842 CAS
- K.-X. Hou, P.-C. Zhao and C.-H. Li, A Light-Driven Actuator Enables Versatile Motion for Smart Transportation and Contactless Delivery, Adv. Opt. Mater., 2023, 11(6), 2202949 CrossRef CAS
| This journal is © The Royal Society of Chemistry 2025 |