Skip to main content
NCBI home page
Biomicrofluidics logoLink to Biomicrofluidics
. 2024 May 1;18(3):034102. doi: 10.1063/5.0203482

A bionic mimosa soft robot based on a multi-responsive PNIPAM-PEGDA hydrogel composition

Wenguang Yang 1,a), Xiaowen Wang 1, Xiangyu Teng 1, Zezheng Qiao 1, Haibo Yu 2, Zheng Yuan 1,a)
PMCID: PMC11078265  PMID: 38726372

Abstract

Deformation plays a vital role in the survival of natural organisms. One example is that plants deform themselves to face the sun for sufficient sunlight exposure, which allows them to produce nutrients through photosynthesis. Drawing inspiration from nature, researchers have been exploring the development of 3D deformable materials. However, the traditional approach to manufacturing deformable hydrogels relies on complex technology, which limits their potential applications. In this study, we simulate the stress variations observed in the plant tissue to create a 3D structure from a 2D material. Using UV curing technology, we create a single-layer poly(N-isopropylacrylamide) hydrogel sheet with microchannels that exhibit distinct swelling rates when subjected to stimulation. After a two-step curing process, we produce a poly(N-isopropylacrylamide)–polyethylene glycol diacrylatedouble-layer structure that can be manipulated to change its shape by controlling the light and solvent content. Based on the double-layer structure, we fabricate a dual-response driven bionic mimosa robot that can perform a variety of functions. This soft robot can not only reversibly change its shape but also maintain a specific shape without continuous stimulation. Its capacity for reversible deformation, resulting from internal stress, presents promising application prospects in the biomedical and soft robotics domain. This study delivers an insightful framework for the development of programmable soft materials.

I. INTRODUCTION

Creatures in nature provide inspiration for making hydrogel robots with intelligent behavior (such as motion deformation and environmental awareness). In recent years, researchers have developed soft hydrogel robots with different functions, which can perform various actions such as swimming and walking.1,2 Deformation is an important prerequisite for organisms to survive in nature. Understanding these deformation behaviors can lead to significant technological advances in soft robotics, flexible electronics, and other fields. Mollusc deformable behavior is based on muscle contraction and extension, while plant deformable behavior is mainly due to cellular water uptake and dehydration.3,4 For example, green plants tend to face toward the light, which facilitates more efficient photosynthesis and, in turn, fosters improved plant growth.5–9 Humidity can affect the morphologies of seed pods, rape plants, and impatiens. Under suitable conditions, such as favorable humidity levels, the seeds of these plants are induced to germinate. These deformations are caused by local contraction or expansion due to the asymmetrical distribution of inner stress within anisotropic plant tissues. Taking cues from natural world, researchers are constantly developing smart materials that possess programmable deformability. These materials include multi-stimulus-responsive hydrogel10–14 and liquid crystal elastomers (LCEs).15–22 Hydrogels, in particular, have garnered significant interest from researchers due to their favorable traits such as biocompatibility and deformation reversibility. This has resulted in their growing utilization in the fields of biomedicine23–29 and soft robotics.30–35

To keep the hydrogel in the 3D structure, even in the absence of continuous stimulation, we introduce in this study a double-layer structure that contains an active layer based on poly(N-isopropylacrylamide) (PNIPAM) hydrogel and a passive layer based on polyethylene glycol diacrylate (PEGDA) hydrogel. PNIPAM hydrogel is a typical temperature-responsive hydrogel, and its swelling ratio is affected by temperature changes. As the temperature increases, the swelling ratio decreases. The volume of PNIPAM hydrogel expands sharply at temperatures below its lower critical solution temperature (LCST < 32 °C) and contracts promptly at temperatures above its LCST (LCST > 32 °C).35 The cross-linking density of PNIPAM can be adjusted to generate a non-uniform expansion field with a specific pattern in a single-layer hydrogel sheet. In response to specific stimuli, the 2D sheet deforms into a 3D structure. Additionally, PNIPAM swells in the mixture of ethanol and water at a ratio higher than in pure water or pure ethanol solvent. Hence, it was used as the active layer. In contrast, polyethylene glycol diacrylate (PEGDA) is used as a passive layer because its swelling ratio is little affected by temperature or solvent composition. The double-layer structure would bend toward different angles as the temperature or solvent composition changes. Unlike previous studies shown in Table I, the PNIPAM–PEGDA hydrogel structure was fabricated by using digital micromirror device (DMD) technology instead of techniques such as 3D printing. DMD technology is more efficient and accurate.36–39 Based on the previous research, we have explored the optimal thickness of the active layer to optimize the deformation degree, deformation time, and recovery time. By introducing photomasks with different gray scales, the resulting hydrogel structures were able to deform into different 3D structures. In addition, by studying the performance of the double-layer structure, the driving method and form of the robot were enriched.

TABLE I.

Key findings of present bionic soft robots.

Bionic soft robot Manufacturing technique Response type Response speed Reference
Bionic gripper 3D printing Temperature response Seconds 40
Bionic two-armed manipulator Multiphoton polymerization Ion response Seconds 41
Bionic flower Stereolithography 3D printing pH response; temperature response Minutes 42
Bionic earthworm robot Evaporation technology Air actuation Minutes 43
Bionic caterpillar robot Light curing Light response Seconds 44
Bionic inchworm robot Thermocuring technology Magnetic response Seconds 45
Bionic jellyfish robot 3D printing Temperature response Seconds 39
Open in a new tab

In this study, we investigated the deformation characteristics of differently patterned single-layer PNIPAM sheets, and the self-folding of the material was achieved via its thermal responsiveness. The PNIPAM–PEGDA double-layer structure was subsequently prepared by two-step curing, where PEGDA served as the passive layer for the double-layer structure to maintain a specific shape without continuous stimulation. Mimosa is a kind of stimulus-responsive plant, whose leaves close up immediately when touched and open again after 1–2 min. In this study, we developed a stimulus-responsive bionic Mimosa soft robot with a double-layer structure and investigated its response state and duration under different stimuli. The findings of this study offer valuable insights into ways to precisely control the geometrically oriented programmable deformation of hydrogel structures, thereby contributing to the advancement of deformable tiny robotics.

II. RESULTS AND DISCUSSION

A. Fabrication of light-driven programmable singer-layer structure

Figure 1(a) depicts the photocuring process of the geometrically oriented programmable single-layer hydrogel employed in this study. The prepolymer of PNIPAM hydrogel was prepared by mixing the NIPAM monomer, diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide (TPO), cross-linking agent methylacrylamide (BIS), and organic solvent adequately. In contrast to traditional 3D printing technology, the digital micromirror device (DMD)-based UV curing technology is more accurate and efficient and works well for most photocured resins. Due to the large number of small mirrors arranged on the DMD chip, each small mirror can be flipped independently to control the light cast, which greatly improves the resolution, image quality, and printing speed. In addition, DMD technology has high material adaptability and can be applied to a variety of materials, which provides designers with greater choice space. This light curing device consisted of a UV optical source, a DMD, and a photocuring platform, as illustrated in Fig. 1(b). A fluorinated ethylpropylene (FEP) was pasted onto the glass slide so that the hydrogel could be easily peeled off after curing.

FIG. 1.

FIG. 1.

Open in a new tab

(a) Manufacture of the PNIPAM hydrogel prepolymer. (b) UV curing device. (c) UV curing platform. (d) UV curing process of the PNIPAM hydrogel. (e) Aggregation and dispersion of macromolecular chains of PNIPAM near LCST. Reduction in the hydrogel volume induced by laser illumination. (f) Impregnation of CNTs.

The thickness of the hydrogel sheet depends on the thickness of the polyimide (PI) spacer on both sides of the bottom slide, as depicted in Figs. 1(c) and 1(d). Changing the shape and gray level of the photomask would allow for the creation of a hydrogel structure with the desired shape and anisotropy. Ultimately, the hydrogel structure was obtained by washing off the residual prepolymer on the surface of the hydrogel sheet.

Carbon nanotubes (CNTs) are a good photothermal conversion material, which can efficiently convert light energy into heat energy. To endow the PNIPAM layer with photothermal conversion properties, it was initially placed in hot water above 32 °C, the micropores in the PNIPAM layer lost water, and the volume of it decreased, as shown in Fig. 1(e). Then, the PNIPAM layer was immersed in ice water evenly dispersed with CNTs, allowing it to reabsorb water. As the hydrogel absorbed water, CNTs were aspirated into the micropores of the hydrogel, as shown in Fig. 1(f). Repeat the above process to remove any CNTs that remained on the surface of the PNIPAM layer and had not been aspirated into the micropores. When the hydrogel sheet was placed in cold water, water was sucked into the micropores of the hydrogel, carrying CNTs along with it into the interior of the PNIPAM layer. In contrast, when placed in hot water above 32 °C, the micropores of the PNIPAM layer lost water, but their size decreased, trapping CNTs within them. Thus, the impregnated hydrogel is capable of retaining the CNTs in it even after frequent light drive.

B. Deformation characteristics of singer-layer PNIPAM/CNT structure

Figure 2(a) shows the scanning electron microscope (SEM) images of the PNIPAM hydrogel that has been impregnated with and without CNTs. As depicted in the figure, CNTs were uniformly distributed on the surface of the PNIPAM layer, forming clumps upon drying. Despite the repeated light drive, the surface of the PNIPAM layer remained dark, indicating little shedding of CNTs. When one side of the PNIPAM layer was illuminated by a laser, CNTs on the surface rapidly absorbed light and converted it into heat, causing the volume of the illuminated side to decrease, while the volume of the backlit side remained the same, causing this singer-layer to bend toward the light source. After the laser was switched off, the heat was quickly transferred into the surrounding water, allowing the illuminated side to absorb water and recover from deformation.

FIG. 2.

FIG. 2.

Open in a new tab

(a1) SEM image of hydrogel's internal pores. (a2) SEM image of hydrogel impregnated without CNTs. (a3) SEM image of hydrogel impregnated with CNTs. (a4) Enlarged image of (a3). (b) Relationship between the alcohol content and maximum bending angle of the PNIPAM hydrogel sheet. (c) Swelling ratio of the PNIPAM hydrogel in water at different temperatures. (d1) Relationship between the thickness and bending angle of the PNIPAM hydrogel sheet. (d2) Relationship between the thickness and deformation/recovery time of the PNIPAM hydrogel sheet. (e1) Relationship between light intensity and diameter of the circular hydrogel sheet. (e2) Relationship between light intensity and volume of the circular hydrogel sheet. (f1) Time-dependent temperature of the hydrogel surface under laser illumination at a light intensity of 1.2 W/cm2. (f2) Deformation recoverability of the hydrogel.

We also investigated how the alcohol content of hydrogels affects their deformation capabilities. To accomplish this, we prepared a PNIPAM prepolymer with different contents of alcohol and fabricated rectangular single layer with the same shape, size, and thickness measuring 400 μm. The experimental results demonstrated that the bending angle of the light-driven single hydrogel layer increased proportionally with the alcohol content, but only when it was below 30%. When the alcohol content exceeded 30%, the bending angle of it decreased as the alcohol content increased, and the light response time gradually lengthened [Fig. 2(b)]. Therefore, the alcohol content in this study was set to 30%.

The temperature responsiveness of the PNIPAM hydrogel is shown in Fig. 2(c). We placed a circular hydrogel sheet in pure water and heated the water to observe changes in the hydrogel volume. The experimental results showed that the volume of PNIPAM decreased as the temperature increased. In addition, the thickness of the PNIPAM layer also affected the response speed, denaturation angle, and recovery speed. We produced hydrogel sheets with thicknesses of 200, 300, 400, 500, and 600 μm. In their untreated state, these PNIPAM sheets naturally sagged under gravity. When one side of the hydrogel was illuminated, it bent toward the light source. We found that when the hydrogel was less than 200 μm in thickness, it could not maintain drooping in water due to little buoyancy. When the thickness of PNIPAM sheets ranged from 200 to 400 μm, the bending angle increased with the thickness. When the thickness exceeded 400 μm, the bending angle decreased with the thickness [Fig. 2(d1)]. In addition, as the thickness increased, the response time of the hydrogel decreased, while the recovery time increased gradually. This observation can be explained by the following facts. When the hydrogel was thin (ranging from 200 to 400 μm), little water was expelled from the illuminated side of the hydrogel, as a result of which the deformation was only slight. When the hydrogel was thick (exceeding 400 μm), the illuminated side lost a considerable amount of water. However, due to the backlight side being quite thick, it was difficult for the illuminated side to pull it to deform, so the deformation remained minor. A notable phenomenon was that as the hydrogel thickness increased, so did the amount of water loss on the illuminated side. After the light source was removed, it took a comparatively longer time for the hydrogel to reabsorb water. This, in turn, caused the recovery time to increase with the thickness. As shown in Fig. 2(d2), a PNIPAM layer with a thickness of 400 μm was an appropriate choice for our experimental setup.

In this study, we utilized a near-infrared light source to generate a laser. We analyzed the deformation of the hydrogel induced by laser at different intensities. As the intensity of the laser increased, its temperature rose gradually. As shown in Fig. 2(e1), a circular PNIPAM hydrogel sheet with a diameter of 0.8 mm was illuminated by a laser. As the light intensity increased, we observed a decrease in the volume and response time of the hydrogel. Under illumination, the diameter of the circular PNIPAM sheet decreased with increasing light intensity. When the light intensity was large (greater than or equal to 0.6 W/cm2), the diameter of it decreased to its minimum value, as shown in Fig. 2(e2). When the PNIPAM sheet was immersed in water and subjected to laser illumination with a light intensity of 1.2 W/cm2, the maximum temperature of its surface was approximately 40 °C. After switching off the laser, the heat was quickly diffused to the water, resulting in a rapid decrease in the temperature, as shown in Fig. 2(f1). After repeated light-driven experiments, the volume of the PNIPAM sheet maintained unchanged, as depicted in Fig. 2(f2).

C. Programmed deformation of singer-layer PNIPAM/CNT structure with different patterns

In this study, we used UV curing technology to produce uniform hydrogel sheets with different shapes. In its initial state, the hydrogel sheet remained buoyant on water because no water was sucked into the micropores. As the near-water side started to absorb water, its volume increased, but the volume of the back-water side remained unchanged temporarily, causing the hydrogel to bend toward the back-water side, as shown in Fig. 3(a). With the continued absorption of water, the weight of the hydrogel sheet increased, causing it to completely sink into the water. Consequently, even the back-water side started to absorb water. This, in turn, led to a reduction in the volume difference between the two sides and a gradual recovery of the hydrogel sheet from deformation. The thicker the thickness, the longer it takes for the near-water side to absorb water, so the deformation duration will be longer. In this section, we fabricated hydrogel sheets with a thickness of 200 μm, whose deformation duration was about 1 min.

FIG. 3.

FIG. 3.

Open in a new tab

(a) Deformations and COMSOL simulations of PNIPAM sheets with multifarious micropatterns (scale bar = 3 mm). (b) Deformations and force analysis of hydrogel sheets with 0°, 90°, and oblique microchannels (scale bar = 3 mm). (c) Deformations of PNIPAM sheets with 0°, 30°, 45°, 60°, and 90° microchannels (scale bar = 3 mm). (d) and (e) Deformations of PNIPAM sheets with multifarious micropatterns (scale bar = 3 mm).

The sequence and level of hydrogel curing are determined by the gray scale of the photomask. A lower gray scale value corresponds to great light transmission through the photomask. The photoinitiator absorbs light energy to trigger polymerization. Therefore, the area with a low gray scale value cures first, followed by those with higher gray scale values. Hydrogels cured with a lower gray scale value exhibit a higher Young’s modulus than those cured with a high gray scale value. By controlling the gray level value, we produced PNIPAM sheets containing microchannels at angles of 0°, 30°, 45°, 60°, and 90° to observe their deformations, as shown in Figs. 3(b) and 3(c). For a sheet with 60° microchannels, the internal stress consists of force Fparallel parallel to the microchannel and force Fvertical perpendicular to the microchannel; the resultant force on both ends of the sheet is parallel and opposite in direction. Therefore, the two ends of the 2D sheet curl up and down, respectively, forming a left-handed or right-handed spiral structure. For a sheet with 0° microchannels, the internal stress is also composed of force Fparallel parallel to the microchannel and force Fvertical perpendicular to the microchannel, and the force along the microchannel direction is greater, causing the short edge of the sheet to roll up, forming a short tube structure. Similarly, for a sheet with 90° microchannels, the force along the direction of the microchannel is also greater than the force in the vertical direction, causing the long edge of the sheet to roll up and form a long tube structure. In addition, we also made PNIPAM sheets with multifarious micropatterns, each with a unique deformation, as depicted in Figs. 3(d) and 3(e).

Inspired by the structures of natural plants, we created four bionic plants with directional microchannels: a green-podded bean, a chrysanthemum flower, an Impatiens balsamina, and a mimosa. The bionic green-podded bean is shown in Figs. 4(a1)–4(a3). Green-podded beans utilize elasticity to spread their seeds. When the humidity inside the bean pod reaches a specific threshold, it explodes and automatically ejects seeds in a spiral shape. Based on previous research, we designed and manufactured a PNIPAM strip containing 60° microchannels. Due to the non-uniform expansion, one end of the hydrogel rolled up and the other rolled down, ultimately transforming into a spiral structure. Figures 4(b1)–4(b3) show a bionic chrysanthemum flower, and each petal of it contained several slightly oblique microchannels to enhance its closure. Figures 4(c1)–4(c3) present the bionic Impatiens balsamina. Similar to green-podded beans, Impatiens balsamina rely on elasticity to spread their seeds. The seeds are encased in their spindle-shaped fruit that appears rolled up after spreading. We designed an elliptical PNIPAM sheet with three 0° microchannels, which could be deformed to achieve a 3D roll shape. The bionic mimosa is shown in Figs. 4(d1)–4(d3). To promote the closure of the bionic mimosa, we designed several 0° microchannels on its stem.

FIG. 4.

FIG. 4.

Open in a new tab

(a1) Photomask of the bionic green-podded bean. (a2) Bionic green-podded bean made up of PNIPAM hydrogel. (a3) Deformation of chrysanthemum (scale bar = 3 mm). (b1) Photomask of bionic chrysanthemum flower. (b2) Bionic chrysanthemum flower made up of the PNIPAM hydrogel. (b3) Deformation of bionic chrysanthemum (scale bar = 3 mm). (c1) Photomask of bionic Impatiens balsamina. (c2) Bionic Impatiens balsamina made up of the PNIPAM hydrogel. (c3) Deformation of bionic Impatiens balsamina (scale bar = 3 mm). (d1) Photomask of bionic mimosa. (d2) Bionic mimosa made up of the PNIPAM hydrogel. (d3) Deformation of bionic mimosa (scale bar = 3 mm).

D. Dual-response driven PNIPAM–PEGDA double-layer structure

The programmable single-layer PNIPAM/CNT sheets described in Sec. II C could transform into 3D structures with specific micropatterns due to the non-uniform distribution of internal stress. However, the deformation duration was brief and the sheets would revert to their 2D form in the absence of ongoing stimulation. To keep the desired 3D structure through removing continuous stimulation, we prepared a double-layer hydrogel structure consisting of an active layer and a passive layer by two-step UV curing. The active layer refers to the structure that experiences changes in the volume in response to external stimuli. In this study, we employed a 400 μm-thick PNIPAM/CNT hydrogel sheet as the active layer. The passive layer refers to the structure whose volume remains almost unchanged when subjected to external stimulation. Here, we chose the PEGDA hydrogel to fabricate the passive layer, as seen in Fig. 5(a).

FIG. 5.

FIG. 5.

Open in a new tab

(a) Fabrication process of the PNIPAM/PEGDA double-layer structure. (b1) Relationship between temperature and the swelling ratio of PNIPAM and PEGDA. (b2) Deformation of the double-layer structure under different temperatures (scale bar = 3 mm). (b3) Relationship between the temperature and the bending angle of the double-layer structure. (c1) Response of the PNIPAM volume to the solvent composition. (c2) Response of the PEGDA volume to the solvent composition. (c3) Deformation of the double-layer structure in solutions with different solvent compositions. (d1) Relationship between the solvent composition and the swelling ratio of PNIPAM and PEGDA. (d2) Relationship between the solvent composition and the bending angle of double-layer structure (scale bar = 3 mm).

The PNIPAM hydrogel is a typical temperature-responsive hydrogel. Research findings indicate that its volume shrank rapidly with an increase in the temperature. As shown in Fig. 5(b1), when the temperature increased from 20 to 50 °C, the hydrogel shrank to half of its original size. In contrast, its volume would not change significantly as the temperature decreased. This can explain why the PNIPAM–PEGDA double-layer structure has temperature-responsive characteristics. Specifically, when the temperature was lower than 32 °C, the active layer absorbed water, causing its volume to increase. Meanwhile, since the volume of the passive layer remained nearly constant, the double-layer structure bent, with the outer layer (PNIPAM) and the inner layer (PEGDA). As the temperature increased, the volume of the active layer decreased gradually, causing the curvature of the double-layer structure to decrease gradually. When the temperature was about 37 °C, the curvature of the double-layer structure was almost near 0. As the temperature persisted in increasing, the volume of the active layer continued to decrease, causing the double-layer structure to bend reversely, such that the outer layer consisted of the PEGDA hydrogel and the inner layer the PNIPAM hydrogel, as observed in Fig. 5(b2). The state in which the PNIPAM hydrogel was positioned on the outside was recorded as “positive” and that in which the PEGDA hydrogel was positioned on the outside was recorded as “negative.” Irrespective of whether it was in the positive or negative state, the double-layer structure could achieve a maximum bending angle of 180°, as depicted in Fig. 5(b3).

In addition, the PNIPAM volume is also affected by the solvent composition. As shown in Fig. 5(c1), the swelling ratio of the PNIPAM reached its maximum only in pure water or pure ethanol solvent. However, in a mixed solution comprising both ethanol and water, its volume decreased substantially. In contrast, the volume of the PEGDA hydrogel remained relatively stable, showing no significant change in both pure and mixed solvent solutions, as shown in Figs. 5(c2) and 5(d1). Therefore, the PNIPAM–PEGDA double-layer structure could effectively respond to the solvent composition. In pure ethanol or pure water solvent, the double-layer structure was in a positive bending state, and there was almost no bending in the alcohol mixed solution, as illustrated in Figs. 5(c3) and 5(d2). Notably, stratification did not occur throughout the experimental process, which was due to the covalent bond between the two layers enhanced interfacial adhesion.

E. Dual-response driven bionic mimosa based on double-layer structure

Taking cues from mimosas, we further fabricated a bionic mimosa robot with miscellaneous microchannels. As shown in Figs. 6(a1) and 6(a2), there were three 0° microchannels on each leaf and 90° microchannels on the stem, connecting the leaves for enhanced closure on both sides of the bionic mimosa soft robot. In more detail, for the leaf with 0° microchannels, the force Fparallel parallel to the microchannel is greater than the force perpendicular to the microchannel Fvertical, resulting in closure of leaves. Similarly, the stem with 90° microchannels rolls into a long tube, further promoting the closure of leaves. To maintain the 3D structure of the soft robot in the absence of continuous stimulation, a PEGDA layer was cured onto the PNIPAM layer, as depicted in Fig. 6(b). In pure water at room temperature, the active layer absorbs water and the volume increases, while the volume of the passive layer is basically unchanged because it hardly absorbs water, resulting in the double-layer stem bending toward the passive layer. Then, the double-layer structure was impregnated with CNTs to obtain photothermal response properties. After repeated dyeing, the surface of the active layer was found to be uniformly black, indicating the good adhesiveness of CNTs. Conversely, the passive layer only showed a pale gray color with uneven distribution, indicating a lower CNT content on the surface of the passive layer, thus poorer photothermal conversion capabilities.

FIG. 6.

FIG. 6.

Open in a new tab

(a1) Mimosa-liked photomask. (a2) Single-layer bionic mimosa made up of the PNIPAM hydrogel. (b) Schematic diagram of the double-layer structure. (c) Bionic mimosa unfolded under continuous laser illumination (scale bar = 3 mm). (d) Bionic mimosa unfolded in the alcohol mixed solution (scale bar = 3 mm).

Under the continuous laser scanning of the stem at an intensity of 1.2 W/cm2, CNTs on the surface of the active layer convert light energy to heat energy. As the temperature increases, the water in the active layer loses, and the volume difference between the active layer and passive layer gradually decreases. The long tubular stem opens gradually, allowing the leaves on both sides to open, which lasted about 80 s. After the laser was switched off, the heat quickly dissipated, causing the soft robot to gradually close up as shown in Fig. 6(c) (Movie S1 in the supplementary material). In addition, when the bionic mimosa soft robot was immersed in a 75% alcohol solution for roughly 1 min, it gradually unfolded itself. This can be explained by the fact that the volume of the active layer reduced in the alcohol mixed solution, causing the stem, which was originally curved, to unfold. When placed in 20 °C water, the unfolded bionic mimosa soft robot would close up again in around 40 s, as shown in Fig. 6(d) (Movie S2 in the supplementary material).

III. CONCLUSION

This study combined a programmable PNIPAM hydrogel with a two-step UV curing process to fabricate a PNIPAM–PEGDA double-layer structure. The double-layer structure could realize reversible transformation from 2D to 3D, while maintaining a specific 3D structure in the absence of continuous stimulation. The PNIPAM hydrogel is a typical temperature-responsive hydrogel that can respond to both temperature and solvent composition, while the PEGDA hydrogel is insensitive to both. Therefore, the double-layer structure is doubly responsive. We made a bionic mimosa using the double-layer structure, which remained closed when placed in pure water at normal temperature. With the addition of CNTs, photothermal conversion characteristics were introduced into the bionic mimosa, enabling it to unfold with continuous illumination and recover after the switching off the laser. In addition, reversible unfolding and closing were also realized by changing the solvent composition. The findings of this study serve as a guide for achieving geometrically oriented programmable deformation of hydrogels, which could facilitate the progress of soft robot development.

IV. EXPERIMENTAL SECTION

A. Materials

We obtained the NIPAM monomer (>98%) and BIS (>99%) from Aladdin Biochemical Technology Limited Company (Shanghai, China). The initiator TPO and PEGDA monomer were obtained from Sigma-Aldrich (United States). The CNTs used in this research were multi-walled carbon nanotubes (MWCNTs) obtained from Tanfeng Graphene Technology Limited Company (Jiangsu, China).

B. Synthesis of PNIPAM prepolymer

Put 480 mg of NIPAM, 12 mg of TPO, and 13 mg of BIS into a 5 ml beaker. Then, add 200 μl of ethanol and 700 μl of de-ionized (DI) water into the beaker by using a pipet. Subsequently, put this breaker on a magnetic mixer for about 20 min to obtain PNIPAM prepolymer. The DI water used in this process was subjected to ultrasonic shaking to remove air bubbles.

C. Synthesis of PEGDA prepolymer

In this research, 20% PEGDA prepolymer was used. Put 20 mg of PEGDA monomer and 0.5 mg of hydrogen peroxide catalyst TPO into 5 ml beaker. Then, add 23.85 μl of 30% alcohol solution and 55.6 μl of DI water into this beaker by using a pipet. Subsequently, put this breaker on a magnetic mixer for about 20 min to obtain PEGDA prepolymer.

D. Preparation of PNIPAM/PEGDA double-layer structure

In this research, we obtained a double-layer structure through two-step UV curing technique. First, the PNIPAM layer with a thickness of 400 μm was obtained by curing. After the first curing, wipe off the excess PNIPAM prepolymer with a lens paper. Subsequently, small amount of PEGDA prepolymer was added to the PNIPAM layer. After the second curing, the double-layer structure was obtained.

SUPPLEMENTARY MATERIAL

See the supplementary material for bionic mimosa unfolded under continuous laser illumination (Movie S1) and bionic mimosa unfolded in the alcohol mixed solution (Movie S2).

ACKNOWLEDGMENTS

The authors acknowledge the funding provided by the National Natural Science Foundation of China (NNSFC) (Project No. 62273289), the Youth Innovation Science and Technology Support Program of Shandong Province (Project No. 2022KJ274), and the Graduate Innovation Foundation of Yantai University (GIFYTU).

AUTHOR DECLARATIONS

Conflict of Interest

The authors have no conflicts to disclose.

Author Contributions

Wenguang Yang: Conceptualization (lead); Funding acquisition (lead); Supervision (equal); Writing – review & editing (equal). Xiaowen Wang: Investigation (lead); Methodology (equal); Software (equal); Writing – original draft (lead). Xiangyu Teng: Investigation (equal); Methodology (equal); Software (equal). Zezheng Qiao: Investigation (equal); Methodology (supporting); Software (supporting). Haibo Yu: Supervision (equal); Writing – review & editing (equal). Zheng Yuan: Supervision (equal); Writing – review & editing (equal).

DATA AVAILABILITY

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

REFERENCES

  • 1.Chen Y. et al. , “Bioinspired hydrogel actuator for soft robotics: Opportunity and challenges,” Nano Today 49, 101764 (2023). 10.1016/j.nantod.2023.101764 [DOI] [Google Scholar]
  • 2.Lee Y., Song W., and Sun J.-Y., “Hydrogel soft robotics,” Mater. Today Phys. 15, 100258 (2020). 10.1016/j.mtphys.2020.100258 [DOI] [Google Scholar]
  • 3.Jiao D. et al. , “Programmable morphing hydrogels for soft actuators and robots: From structure designs to active functions,” Acc. Chem. Res. 55(11), 1533–1545 (2022). 10.1021/acs.accounts.2c00046 [DOI] [PubMed] [Google Scholar]
  • 4.Ye Z. et al. , “Liquid-metal soft electronics coupled with multi-legged robots for targeted delivery in the gastrointestinal tract,” Device 2(2), 2(2) (2024). 10.1016/j.device.2023.100181 [DOI] [Google Scholar]
  • 5.Wang Y. and Folta K. M., “Contributions of green light to plant growth and development,” Am. J. Bot. 100(1), 70–78 (2013). 10.3732/ajb.1200354 [DOI] [PubMed] [Google Scholar]
  • 6.Brown R. H., “Growth of the green plant,” in Physiological Basis of Crop Growth and Development (Wiley, 1984), pp. 153–174. 10.2135/1984.physiologicalbasis.c6 [DOI] [Google Scholar]
  • 7.Horton P., Ruban A. V., and Walters R. G., “Regulation of light harvesting in green plants,” Annu. Rev. Plant Physiol. Plant Mol. Biol. 47(1), 655–684 (1996). 10.1146/annurev.arplant.47.1.655 [DOI] [PubMed] [Google Scholar]
  • 8.Wada M. and Kadota A., “Photomorphogenesis in lower green plants,” Annu. Rev. Plant Physiol. Plant Mol. Biol. 40(1), 169–191 (1989). 10.1146/annurev.pp.40.060189.001125 [DOI] [Google Scholar]
  • 9.Hu Y. et al. , “Electrically and sunlight-driven actuator with versatile biomimetic motions based on rolled carbon nanotube bilayer composite,” Adv. Funct. Mater. 27(44), 1704388 (2017). 10.1002/adfm.201704388 [DOI] [Google Scholar]
  • 10.Jain K. et al. , “Tunable LCST behavior of poly(N-isopropylacrylamide/ionic liquid) copolymers,” Polym. Chem. 6(38), 6819–6825 (2015). 10.1039/C5PY00998G [DOI] [Google Scholar]
  • 11.Shin B. et al. , “Hygrobot: A self-locomotive ratcheted actuator powered by environmental humidity,” Sci. Robot. 3(14), eaar2629 (2018). 10.1126/scirobotics.aar2629 [DOI] [PubMed] [Google Scholar]
  • 12.Ma S. et al. , “A light-activated polymer composite enables on-demand photocontrolled motion: Transportation at the liquid/air interface,” Angew. Chem. 131(9), 2681–2685 (2019). 10.1002/ange.201811808 [DOI] [PubMed] [Google Scholar]
  • 13.Li C. et al. , “Fast and programmable locomotion of hydrogel-metal hybrids under light and magnetic fields,” Sci. Robot. 5(49), eabb9822 (2020). 10.1126/scirobotics.abb9822 [DOI] [PubMed] [Google Scholar]
  • 14.Medina-Sánchez M. et al. , “Swimming microrobots soft, reconfigurable, and smart,” Adv. Funct. Mater. 28(25), 1707228 (2018). 10.1002/adfm.201707228 [DOI] [Google Scholar]
  • 15.Shibayev V. P. and Byelyayev S. V., “Prospects for the use of functional liquid crystal polymers and composites: Review,” Polym. Sci. U.S.S.R. 32(12), 2384–2428 (1990). 10.1016/0032-3950(90)90414-2 [DOI] [Google Scholar]
  • 16.Ohm C., Brehmer M., and Zentel R., “Liquid crystalline elastomers as actuators and sensors,” Adv. Mater. 22(31), 3366–3387 (2010). 10.1002/adma.200904059 [DOI] [PubMed] [Google Scholar]
  • 17.Shahsavan H. et al. , “Smart muscle-driven self-cleaning of biomimetic microstructures from liquid crystal elastomers,” Adv. Mater. 27(43), 6828–6833 (2015). 10.1002/adma.201503203 [DOI] [PubMed] [Google Scholar]
  • 18.Zeng H. et al. , “Self-regulating iris based on light-actuated liquid crystal elastomer,” Adv. Mater. 29(30), 1701814 (2017). 10.1002/adma.201701814 [DOI] [PubMed] [Google Scholar]
  • 19.Pilz da Cunha M. et al. , “An untethered magnetic- and light-responsive rotary gripper: Shedding light on photoresponsive liquid crystal actuators,” Adv. Opt. Mater. 7(7), 1801643 (2019). 10.1002/adom.201801643 [DOI] [Google Scholar]
  • 20.Wani O. M. et al. , “An artificial nocturnal flower via humidity-gated photoactuation in liquid crystal networks,” Adv. Mater. 31(2), 1805985 (2019). 10.1002/adma.201805985 [DOI] [PubMed] [Google Scholar]
  • 21.Zeng H. et al. , “Liquid crystal polymer networks and elastomers for light-fueled robotics,” in Photoactive Functional Soft Materials (Wiley, 2019), pp. 197–226. 10.1002/9783527816774.ch6 [DOI] [Google Scholar]
  • 22.Wang M. et al. , “Liquid crystal elastomer actuator with serpentine locomotion,” Chem. Commun. 56(55), 7597–7600 (2020). 10.1039/D0CC02823A [DOI] [PubMed] [Google Scholar]
  • 23.Ashraf S. et al. , “Snapshot of phase transition in thermoresponsive hydrogel PNIPAM: Role in drug delivery and tissue engineering,” Macromol. Res. 24(4), 297–304 (2016). 10.1007/s13233-016-4052-2 [DOI] [Google Scholar]
  • 24.Li H. et al. , “Magnetic actuated pH-responsive hydrogel-based soft micro-robot for targeted drug delivery,” Smart Mater. Struct. 25(2), 027001 (2016). 10.1088/0964-1726/25/2/027001 [DOI] [Google Scholar]
  • 25.Chen L. et al. , “Multi-responsive actuators based on a graphene oxide composite: Intelligent robot and bioinspired applications,” Nanoscale 9(28), 9825–9833 (2017). 10.1039/C7NR01913K [DOI] [PubMed] [Google Scholar]
  • 26.Díez P. et al. , “Biomedical nanomotors: Efficient glucose-mediated insulin release,” Nanoscale 9(38), 14307–14311 (2017). 10.1039/C7NR05535H [DOI] [PubMed] [Google Scholar]
  • 27.Wang J. et al. , “Light-driven micro/nanomotor for promising biomedical tools: Principle, challenge, and prospect,” Acc. Chem. Res. 51(9), 1957–1965 (2018). 10.1021/acs.accounts.8b00254 [DOI] [PubMed] [Google Scholar]
  • 28.Cao Y. et al. , “Self-healing electronic skins for aquatic environments,” Nat. Electron. 2(2), 75–82 (2019). 10.1038/s41928-019-0206-5 [DOI] [Google Scholar]
  • 29.Liang S. et al. , “Microscopic hollow hydrogel springs, necklaces and ladders: A tubular robot as a potential vascular scavenger,” Mater. Horiz. 6(10), 2135–2142 (2019). 10.1039/C9MH00793H [DOI] [Google Scholar]
  • 30.Rus D. and Tolley M. T., “Design, fabrication and control of soft robots,” Nature 521(7553), 467–475 (2015). 10.1038/nature14543 [DOI] [PubMed] [Google Scholar]
  • 31.Katzschmann R. K., Marchese A. D., and Rus D., “Hydraulic autonomous soft robotic fish for 3D swimming,” in Experimental Robotics: The 14th International Symposium on Experimental Robotics, edited by Hsieh M. A., Khatib O., and Kumar V. (Springer International Publishing, Cham, 2016), pp. 405–420. [Google Scholar]
  • 32.Rogóż M. et al. , “Light-driven soft robot mimics caterpillar locomotion in natural scale,” Adv. Opt. Mater. 4(11), 1689–1694 (2016). 10.1002/adom.201600503 [DOI] [Google Scholar]
  • 33.Giorgio-Serchi F. et al. , “Hybrid parameter identification of a multi-modal underwater soft robot,” Bioinspiration Biomimetics 12(2), 025007 (2017). 10.1088/1748-3190/aa5ccc [DOI] [PubMed] [Google Scholar]
  • 34.Pilz da Cunha M. et al. , “A soft transporter robot fueled by light,” Adv. Sci. 7(5), 1902842 (2020). 10.1002/advs.201902842 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Peng X. et al. , “Mechanically strong Janus poly(N-isopropylacrylamide)/graphene oxide hydrogels as thermo-responsive soft robots,” Chin. J. Polym. Sci. 35(10), 1268–1275 (2017). 10.1007/s10118-017-1970-1 [DOI] [Google Scholar]
  • 36.Wang L. et al. , “Multi-responsive PNIPAM–PEGDA hydrogel composite,” Soft Matter 17(46), 10421–10427 (2021). 10.1039/D1SM01178B [DOI] [PubMed] [Google Scholar]
  • 37.Xiao Q. et al. , “PNIPAM hydrogel composite membrane for high-throughput adsorption of biological macromolecules,” Sep. Purif. Technol. 294, 121224 (2022). 10.1016/j.seppur.2022.121224 [DOI] [Google Scholar]
  • 38.Hu S. et al. , “Thermally trainable dual network hydrogels,” Nat. Commun. 14(1), 3717 (2023). 10.1038/s41467-023-39446-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Mourran A. et al. , “Soft microrobots employing nonequilibrium actuation via plasmonic heating,” Adv. Mater. 29, 201604825 (2017). 10.1002/adma.201604825 [DOI] [PubMed] [Google Scholar]
  • 40.Yuk H. et al. , “Hydraulic hydrogel actuators and robots optically and sonically camouflaged in water,” Nat. Commun. 8(1), 14230 (2017). 10.1038/ncomms14230 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Xiong Z. et al. , “Asymmetric microstructure of hydrogel: Two-photon microfabrication and stimuli-responsive behavior,” Soft Matter 7(21), 10353–10359 (2011). 10.1039/c1sm06137b [DOI] [Google Scholar]
  • 42.Odent J. et al. , “Hierarchical chemomechanical encoding of multi-responsive hydrogel actuators via 3D printing,” J. Mater. Chem. A 7(25), 15395–15403 (2019). 10.1039/C9TA03547H [DOI] [Google Scholar]
  • 43.Jiang F. et al. , “Pneumatically actuated self-healing bionic crawling soft robot,” J. Intell. Rob. Syst. 100(2), 445–454 (2020). 10.1007/s10846-020-01187-z [DOI] [Google Scholar]
  • 44.Zeng H. et al. , “Light-driven, caterpillar-inspired miniature inching robot,” Macromol. Rapid Commun. 39(1), 1700224 (2018). 10.1002/marc.201700224 [DOI] [PubMed] [Google Scholar]
  • 45.Shen H. et al. , “A programmable inchworm-inspired soft robot powered by a rotating magnetic field,” J. Bionic Eng. 20(2), 506–514 (2023). 10.1007/s42235-022-00296-9 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

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

Articles from Biomicrofluidics are provided here courtesy of American Institute of Physics

RESOURCES