Two-Step Freezing Polymerization Method for Efficient Synthesis of High-Performance Stimuli-Responsive Hydrogels.

The widespread use of stimuli-responsive hydrogels is closely related to their synthesis efficiency. However, the widely used thermal-responsive poly(N-isopropylacrylamide) (PNIPAM) hydrogels usually require a time-consuming synthesis process to produce (more than 12 h) and exhibit a relatively slow response speed in the field of cryo-polymerization. In this study, a sequence of thawing polymerization after freezing polymerization by a two-step method of free radical polymerization for the efficient synthesis of PNIPAM hydrogels (merely 2 h) with an excellent comprehensive performance is demonstrated. Results show that the overall performance of the as-synthesized PNIPAM hydrogels is at the top level among reported works despite the significantly reduced preparation time. Moreover, after incorporating multi-walled carbon nanotubes (MWNTs), the PNIPAM hydrogels exhibit a rapid near-infrared (NIR) light-response and programmable shape-morphing capability. It is believed that such a viable and time-saving synthetic method for producing PNIPAM hydrogels of high performance will lay a solid foundation for drug delivery and smart actuators.

Introduction

Stimuli-responsive hydrogels are a kind of macromolecule polymers with three-dimensional network structures. By receiving external environmental stimuli, such as pH,[1] light,[2] temperature,[3] ionic strength,[4] and electric/magnetic field,[5] the hydrogels undertake reconfiguration of a molecular structure, which can dramatically change their volume or physicochemical properties. Due to their unique characteristics, the stimuli-responsive hydrogels show great potential in various applications, including smart sensors/actuators,[6] biomedicine,[7] artificial muscles,[8] and soft robots,[9] which have drawn considerable interests from both academic and industrial communities. Among stimuli-responsive hydrogels, poly(N-isopropylacrylamide) (PNIPAM), as a typical thermal stimuli-responsive hydrogel, has drawn extensive attention due to its biocompatibility, non-toxicity, and high water retention.[10] Most importantly, PNIPAM hydrogels have a lower critical solution temperature (LCST) in water of ∼32 °C close to the physiological temperature.[11] Upon heating above LCST, PNIPAM chains in a solution possess an inverse solubility and change from hydrophilic to hydrophobic, thus undergoing a phase transition from coil to globule to cause a large volume change.[12] However, the volume change of conventional PNIPAM hydrogels usually exhibits a slow stimuli response due to the formation of a “dense skin layer”.[13] The so-called “dense skin layer” significantly prevents the out-diffusion of water molecules from the gel matrix, prolonging the equilibrium swelling and deswelling time of hydrogel, ranging from a few hours to even a few days.[14,15] Such slow responsiveness severely restricts the practical application of PNIPAM hydrogels and becomes one of the most serious challenges to be solved.

By far, various methods have been developed to improve the stimuli-responsive characteristics of bulky PNIPAM hydrogels, which can be mainly classified into four categories, including semi-interpenetrating networks (SIPN),[16,17] chemical grafting,[18,19] porogens,[20,21] and nanofillers.[14,22] The SIPN method introduces linear hydrophilic long-chain polymers (e.g., sodium alginate (SA),[23] poly(vinyl alcohol) (PVA),[24] and poly(vinyl-pyrrolidone) (PVP)[25]) into PNIPAM networks to increase the water release channel, thereby achieving rapid water diffusivity. The elasticity and swelling ratio of the PNIPAM hydrogel improve using the SIPN method, but the thermal response time is still limited to hundreds of seconds.[17] The grafting method used to modify the surface chemical composition and functional groups of original materials is also utilized to improve the thermal-responsive characteristics of PNIPAM hydrogels. Studies have shown that PNIPAM hydrogel can improve its deswelling response by grafting hydrophobic side chains (within 20 min).[26] Fast thermal and pH-responsive hydrogels can be achieved by grafting poly(N,N-dimethylaminoethyl methacrylate) with PNIPAM chains (within 60 min).[27] In addition, the PNIPAM hydrogels grafted with α-cyclodextrin[18] and poly(vinyl alcohol)formaldehyde (PVF)[19] exhibit good biocompatibility, flexibility, and rapid thermal response (90–200 s). However, this preparation method is relatively complicated and time consuming, and the as-prepared hydrogels tend to lack elasticity and fatigue resistance. To eliminate the blocking effect induced by the dense skin layer, porous microstructures were adopted to enrich the water transport channels and thereby improve the thermal responsiveness of hydrogels using porogens, such as cryogenic solvents[20,28] and emulsifiers.[21] Among the porogen methods, freezing polymerization is one of the most effective strategies for preparing porous PNIPAM hydrogels with a fast response due to its simplicity, convenience, and additive-free procedure. For example, poly[NIPAM-co-(di-n-propylacrylamide)] P(NIPAM-co-DPAM) hydrogels prepared by freezing polymerization can achieve rapid swelling and deswelling.[29] However, these porous structures can also reduce the elasticity of hydrogels, causing collapse of the porous structure as the volume shrinks, resulting in the blockage of the water transport channels with a relatively prolonged swelling/deswelling time in tens of minutes. Moreover, the mass swelling ratio of these porous hydrogels is relatively small (usually in the range 10–20). In order to further improve the thermal responsiveness and elasticity of the hydrogel, various nanofillers, such as silica pellet (SiO2),[30−32] tungsten sulfide (WS2) nanosheets,[33] polyaniline (PANI),[32,34] and titania (TiO2) nanoparticles,[35] are added into PNIPAM hydrogels to form hybrid hydrogels due to their high mechanical properties and good dispersibility. Although the above methods combining freezing polymerization and the functional nanofillers can successfully reduce the response time of PNIPAM hydrogels from dozens of minutes to dozens of seconds,[32−34] their production procedures are usually time consuming (e.g., generally requiring at least 12 h or even 72 h for the synthesis of PNIPAM hydrogels) with potential toxicity issues associated with the residual solvent/additives.[36] Therefore, there is still lack of a rapid and cost-effective method to synthesize additive-free PNIPAM hydrogels with an excellent comprehensive performance, including fast response, large swelling ratio, low toxicity, and desired mechanical properties.

In this study, we propose a simple two-step method for the efficient synthesis of additive-free PNIPAM hydrogels. The synthesis of PNIPAM hydrogels is carried out by freezing polymerization (−20 °C) at first and then thawing polymerization (4 °C) (denoted as the FT method), which is contrary to the traditional synthetic methods[14,30−32,35] following the sequence of polymerization (e.g., at 15 °C) first and then freezing step (denoted as the PF method). The proposed FT method can form a 3D ice crystal sacrifice skeleton within 1 h at −20 °C, thereby rapidly forming interconnect porous microstructures in PNIPAM hydrogels during the thawing polymerization step at 4 °C. As a result, PNIPAM hydrogels with a high comprehensive performance, that is, fast response, large swelling ratio (e.g., up to 41), and excellent fatigue resistance (more than 100 cycles at 80% strain), can be achieved within 2 h, which significantly improves the production efficiency. Moreover, doping biocompatible multi-walled carbon nanotubes (MWNTs) offers the PNIPAM hydrogels exceptional near-infrared (NIR) light responsiveness for remote controllability. This PNIPAM-MWNT nanocomposite hydrogel exhibits a fast response and controllable shape-morphing capability via NIR laser irradiation. The proposed two-step synthesis method is cost-effective and time saving for the production of high-performance PNIPAM hydrogels, which is promising for the applications of smart actuators, biomedicine, and drug delivery.

Results and Discussion

Synthesis Strategy

We synthesize PNIPAM hydrogels using the rapid two-step method as shown in Figure . The method adopts the two-step sequence of freezing polymerization first at −20 °C, which is performed in a frozen state and then thawing polymerization at 4 °C. Based on the crystallization theory of the solvent[37] and the exothermic reaction of the redox system, we choose −20 and 4 °C as the experimental temperatures. In the first step of freezing polymerization, the polymerization rate of the hydrogel is extremely low due to the low-temperature effect. Results show that the gel fraction of 1 h freezing polymerization (step 1) is zero (Table S1, Supporting Information). This phenomenon indicates that the formation of ice crystals is not hindered by gelation during the first hour of freezing polymerization, which creates conditions for the porosity of the hydrogel. After 1 h of thawing polymerization (step 2), the gel fraction increases to 66.5% (Table S1), indicating the necessity of thawing polymerization in our two-step process. Notably, the gel fraction increases to 57.8% (Table S1) in 2 h of freezing polymerization (step 1), indicating that gelation starts in the second hour of freezing polymerization. It can be inferred that only the chain extension reaction occurred during the first hour of freezing polymerization without obvious gelation. The relationship between the polymerization time in the different step and the gel fraction is shown in Table S1. The conventional freezing polymerization approach is usually pre-polymerization at first and then freezing polymerization (PF method). This method has a small amount of gelation before the formation of ice crystals, so cross-linking barriers need to be broken during the formation of ice crystals. And the subsequent low-temperature polymerization takes more time to form micro-porous hydrogels with a loose pore wall. Therefore, the whole synthesis process usually requires a relatively long production time over 12 h, which limits massive applications. In contrast, in the first step of our FT method, the aqueous solution of a mixed precursor (containing the monomer N-isopropylacrylamide (NIPAM); the oxidative involved in the redox initiator, ammonium persulfate (APS); the reductive component, N,N,N′,N′-tetramethyl-ethylenediamine (TEMED); and the cross-linker, N,N′-methylenebisacrylamide (Bis)) is frozen at −20 °C to induce the formation of ice crystals (Figure a,b). In this step, the monomers in the unfrozen liquid microphase are subjected to a slow chain extension reaction due to the low activity of the initiator at −20 °C. The free long chains do not retard the formation of ice crystals, thereby ensuring the porosity and pore interconnectivity of the hydrogel network within a short period (merely 1 h) (Figure c,d). In the second step, the rapid thawing polymerization occurs at 4 °C due to the increased activity of the initiator to form a complete three-dimensional cross-linked network (Figure d,e). During this process, ice crystals melt to form pores. Therefore, the entire synthesis period of our proposed two-step method is merely 2 h, which is significantly shorter than the conventional methods.[14,16,17,20,21,23−25,28,31−35,38,39] PNIPAM hydrogels using the FT method (denoted as FT PNIPAM) and the PF method (denoted as PF PNIPAM) were both prepared for the performance comparison. The as-prepared hydrogels were stored in water before use.

Figure 1

Schematic illustration of the synthesis process of the FT PNIPAM hydrogel by a two-step method. (a,b) PNIPAM chains are formed by chemical cross-linking. (c–e) Two-step method to synthesize FT PNIPAM hydrogels. In the first step, (c,d) the ice crystals (blue irregular shapes) and little FT PNIPAM chains are formed by freezing at −20 °C. Pale dots: various components in the mixed solution. Then, in the second step, (d,e) the ice crystals melt and the remaining monomers thawing polymerize at 4 °C.

Thermal Responsiveness of FT PNIPAM Hydrogels

The thermal responsiveness of the as-synthesized FT PNIPAM hydrogels with different polymerization times is shown in Figure . PNIPAM hydrogels would undergo a volumetric shrinkage (Figure a inset) when they are heated above LCST (∼32 °C) due to the coil-globule transition of PNIPAM chains. Our results show that, with a low concentration of chemical cross-linker Bis (0.2 mg/mL), the hydrogels exhibit a consistent change in the swelling ratio (SR, mass ratio), with the SRmax ranging from 37 to 41 (Figure a). Although the polymerization time is greatly shortened, the swelling ratio remains relatively high in our two-step process. In fact, with the fixing of a constant dry weight, the FT PNIPAM hydrogel has a larger swelling ratio than that of the PF method (Figure S1, Supporting Information). For the 2/2 h hydrogel (x/y h refers to freezing polymerization x h and the thawing polymerization y h; see the Experimental Section for details), the thermal-responsive SR reaches up to >41 below LCST, and the swelling ratio variation (ΔSR, ΔSR = SRmax – SRmin) also reaches 38 as the temperature increases from 20 to 45 °C, higher than those of most hydrogels as reported (Figures a,d and S1). More remarkably, the FT PNIPAM hydrogels also exhibit much faster thermal-responsive rates than the PF PNIPAM hydrogels between 20 and 45 °C (Figures b and S2). For example, at 45 °C the volume of the 2/2 h hydrogel (cylindrical shape with a diameter of 15 mm and thickness of 3 mm) could shrink 65 and 91.5% just within 10 and 20 s, respectively. For comparison, the PF PNIPAM requires 40 s to achieve the shrinkage of 71%. And the volume swelling rate of the 2/2 h FT hydrogel is also very fast; it takes 30 s for the complete volume recovery at 20 °C, in contrast to 220 s of volume recovery for the PF PNIPAM (Figure S2 and Video S1). Moreover, the volume swelling-deswelling process of the FT PNIPAM hydrogels is highly reversible and repeatable with the temperature-induced stimulation cycling between 20 °C (volume swelling) and 45 °C (volume deswelling) for 10 cycles without obvious decrease (Figure S3). We use the swelling rate (νswell) and deswelling rate (νdeswell) to represent the thermal response rate of the FT PNIPAM hydrogels. Swelling rate νswell = ΔSR/tswell, where tswell is the full swelling time. Deswelling rate νdeswell = ΔSR/tdeswell, where tdeswell is the full swelling time. To the best of our knowledge, the response rate and the ΔSR value are among the top ones as compared with the related works using the same cross-linker from the literature (Figure c,d).[14,16−25,28,31−35,38,39] Moreover, the time required for synthesizing the high-performance FT PNIPAM hydrogels using our method is the shortest according to the literature (Figure d).[14,16,17,20,21,23−25,28,31−35,38,39]

Figure 2

Thermal responsiveness of FT PNIPAM hydrogels with different polymerization times (fixed Bis concentration: 0.2 mg/mL). (a) Temperature dependence of the equilibrium swelling ratios of the FT PNIPAM hydrogels. The embedded image shows optical images of the PNIPAM hydrogels in swelling states at 20 °C and in deswelling states at 45 °C. Scale bar is 10 mm. (b) Dynamic thermal-responsive swelling-deswelling behaviors of the FT PNIPAM hydrogels at 45 and 20 °C. (c) Speed comparison of thermal responsiveness with PNIPAM-based hydrogels under different conditions. All the following gel-forming polymers are PNIPAM. The porogens: ice,[20] DMSO,[28] and oil-in-water.[21] The SIPN hydrogels: PVA,[24] PVP,[25] skin protein,[16] salecan,[39] SA,[23] starch/SiO2,[31] and hydroxyethyl methacrylate.[17] The hybrid hydrogels: TMOS,[14] TiO2,[35] AuNPs,[22] graphene oxide,[38] PANI/SiO2,[32] WS2,[33] and PANI.[34] The grafted polymers: polyrotaxane (PR)[18] and PVF.[19] (d) Swelling ratio variation and synthesis time with PNIPAM-based hydrogels under different conditions. (c,d) Data points are reproduced with permissions.[14,16−25,28,31−35,38,39]

Mechanical Properties of FT PNIPAM Hydrogels

The additive-free FT PNIPAM hydrogels also exhibit excellent elasticity and fatigue resistance (Bis concentration: 0.2 mg/mL). The compression properties of the fully swollen FT PNIPAM hydrogels (a cylindrical shape with a diameter of 15 mm and thickness of 15 mm) are shown in Figure . The hydrogels have a large water retention capacity, and even under 80% deformation, they can recover to their original shape after the stress is released (Figure a), indicating significantly enhanced mechanical properties compared with PF PNIPAM, which is fragile and easily broken under a strain over 55% (Figure S4). As for the FT PNIPAM hydrogels prepared using different polymerization times, they all have a similar elasticity at 80% strain, but only the corresponding stress is slightly different (Figure b,c). It is indicated that the FT PNIPAM hydrogels using our two-step FT method can be synthesized in 2 h without sacrificing their mechanical performance. Notably, the FT PNIPAM hydrogel can withstand the cyclic compression of 80% strain more than 100 times. Figure d shows that, during the cycle compression test of the 2/2 h hydrogel, the curves of the first compression and the last compression (the 100th) are almost coincident, and the stress value corresponding to 80% strain is stable at around 11 ± 1 kPa (Figure d,e). Notably, the previous work in the literature conducted only 10 cycles of stress tests[33,34] or the non-repetitive tests[17,19] for the mechanical performance characterization of as-synthesized hydrogels. In comparison, we conducted a hundred cyclic stress–strain tests for the FT hydrogels that remained nearly intact without any visible structural damage, indicating that the FT PNIPAM hydrogel has an excellent fatigue resistance and stability. Unfortunately, our FT hydrogels have weak tensile properties due to the gel’s porosity. The poor tensile properties could be improved in future study through a double cross-linking network and the addition of physical cross-linkers.

Figure 3

Mechanical characterization of FT PNIPAM hydrogels with different polymerization times (fixed Bis concentration: 0.2 mg/mL). (a) Optical images showing that 2/2 h hydrogels have a good recoverability and water retention at 80% strain. (b) Stress–strain curves of the hydrogels with different polymerization times. (c) Stress at 80% strain of the hydrogels with different polymerization times. (d) First and last stress–strain curves of 100 cyclic compression tests of the 2/2 h hydrogels. (e) Evolution of stress at 80% strain of the 2/2 h hydrogel with respect to the cycle number.

Morphological Analysis of FT PNIPAM Hydrogels

In order to further explore the polymerization mechanism of our FT method, we perform scanning electron microscopy (SEM) observation on the hydrogels at the end of the first step (Figure S5) and the second step (Figure S6). At the end of the first step, the SEM images show that the gelation degree of the FT PNIPAM hydrogel increases with the increase of the freezing polymerization time. At 2 h of the first step, the pores in the hydrogel are basically formed, and there are many filaments between the pores, which are un-cross-linked PNIPAM long chains. At 9 h of the first step, the filaments between the pores are reduced, and the pore walls piled up. The pore wall structures are basically formed at 12 h of the first step. However, at the end of the second step, the prepared FT PNIPAM hydrogels with different polymerization times all have microscopic porous structures (Figures a and S6) due to the accelerated gelation in the thawing polymerization step (the second step). For example, the pore size distribution of the PNIPAM hydrogel with 2/2 h polymerization time ranges from 30 to 60 μm, and the average pore size is as large as 48 μm (Figure b,c). Notably, the average pore size of FT PNIPAM hydrogels gradually decreases with the increase of polymerization time due to the increase in gelation. In contrast, SEM characterization shows that the PF PNIPAM hydrogels have a smaller pore size and a very low pore interconnection (Figure S7). This is because PNIPAM has partially gelled during the formation of the ice crystals, and the tight network formed by the gelation prevents the formation and distribution of ice crystals[40−42] in the PF method. Moreover, Figure d shows that the hydrogel with 2/2 h polymerization time has the highest area porosity of 68.4% (the area porosity is defined as the ratio of the pore size area to the total area). The average pore size and porosity together determine the water transfer speed in the hydrogel.

Figure 4

Morphological analysis of FT PNIPAM with different polymerization times. (a) SEM image of the freeze-dried 2/2 h PNIPAM hydrogel. (b) Pore size distribution of the 2/2 h PNIPAM hydrogel. (c) Average pore size of the PNIPAM hydrogels with different polymerization times. (d) Porosity of the PNIPAM hydrogels with different polymerization times.

Performance Optimization of FT PNIPAM Hydrogels

The concentration effect of the chemical cross-linker Bis is also investigated to optimize the performance of FT PNIPAM hydrogels. Figure shows the hydrogel properties with respect to Bis concentrations. The PNIPAM hydrogels exhibit a decreased swelling ratio and volume change rate as the Bis concentration increases (Figures a,b and S8). For example, the SRmax of Bis 0.2 (Bis z refer to the Bis concentration (mg/mL) of PNIPAM hydrogels) hydrogel is 41.4 and the ΔSR is 38, while the SRmax and ΔSR of the Bis 2.0 hydrogel are only 21.2 and 16.1, respectively. Besides, the percentages of volume decrease of the Bis 0.2 hydrogel and Bis 2.0 hydrogel are 91.5 and 72%, respectively. Figure S9 shows that the increase of Bis concentration leads to a slight decrease in the pore size and an increase in the thickness of the pore wall of the FT hydrogel. Notably, there are many small pores in the pore wall of these hydrogels. This is because of the tiny ice crystals formed by the water remaining in the pore wall during the freeze-drying process, leaving many small pores on its surface. Interestingly, this change in the pore structure only changes the swelling ratio of the FT hydrogel but has a slight effect on the swelling/deswelling time (Figure a,b). Results indicate that we can adjust the swelling ratio and volume change percentage of the hydrogels by changing the Bis concentration but hardly change the response time. For mechanical properties, the breaking strain decreases monotonously with the increase of Bis concentration (Figure c,d). It is expected that the cross-linking degree of hydrogels becomes higher with the increase of Bis concentration, which leads to thicker pore walls and smaller pores, thus showing increased stress under the same applied strain. For the 50% strain, the stress value of the hydrogel samples gradually increases with the increase of Bis concentration (Figure S10). Without particular specification, 0.2 mg/mL Bis and polymerization time 2/2 h were adopted to produce the optimal FT PNIPAM hydrogel for further experiments.

Figure 5

Thermal-responsive and mechanical properties of FT PNIPAM hydrogels with different Bis concentrations. (a) Temperature dependence of the equilibrium swelling ratios of the FT PNIPAM hydrogels. (b) Dynamic thermal-responsive swelling-deswelling behaviors of the FT PNIPAM hydrogels at 45 and 20 °C. (c) Stress–strain curves of the FT PNIPAM hydrogels. (d) Breaking strain of the FT PNIPAM hydrogels with respect to the Bis concentration.

Shape Morphing of FT PNIPAM-MWNT Nanocomposite Hydrogels

Among thermal-responsive hydrogels, rapid NIR light-responsive hydrogels have received widespread attention from researchers due to their simplicity, remote control, and shape morphing capability.[43,44] These advantages create opportunities for remote manipulation of smart devices, which is promising in many applications. Generally, the high response rate is indispensable in many practical applications, such as NIR light-controlled actuators and soft robots. Therefore, it is of great significance to develop NIR light-responsive hydrogels with fast response.

In this work, an excellent NIR light-responsive hydrogel (Figure ) is also developed by the incorporation of MWNTs with the rapid two-step FT method. When the NIR laser is turned on, the MWNTs absorb the photon energy of NIR irradiation and locally heat the hydrogel above LCST, causing the nanocomposite hydrogel to shrink. And when the laser is turned off, the shrunken volume returns to its original shape quickly due to water absorption (Figure a). In the experiment, the volume of the nanocomposite hydrogels (cuboid shape with a length of 3.5 mm, width of 7 mm, and thickness of 1 mm) shrinks ∼84% within 8 s upon exposure to the NIR radiation (5 W cm–2) and fully recovered within 8 s upon shutoff of the NIR irradiation (Figure b,c). The rapid photo response is also caused by the porosity of the hydrogels (Figure S11). Moreover, the repeatability of the NIR driven-responsive function is investigated by exposing the nanocomposite hydrogels under continuously on and off NIR laser irradiation (power density of 5 W cm–2 and time interval of 8 s). Experimental results show that the NIR light-responsive behaviors of the as-synthesized PNIPAM-MWNT nanocomposite hydrogels are highly repeatable (Figure d and Video S2).

Figure 6

NIR light-responsive characteristics of PNIPAM-MWNT nanocomposite hydrogels. (a) Schematic illustration of volumetric variation of nanocomposite hydrogels controlled by NIR laser exposure and removal. (b) Optical images of volume change of nanocomposite hydrogels under NIR radiation in water. (c) Dynamic swelling-deswelling behavior of nanocomposite hydrogels under NIR radiation in water. V is the volume at time t. V0 is the volume at the equilibrium swollen state. (d) Repeatability of shrinkage–recovery cycles of nanocomposite hydrogels under periodic on-off NIR radiation.

Besides the excellent NIR light responsiveness, the PNIPAM-MWNT nanocomposite hydrogels (denoted as PMNC hydrogels) also possess the shape-morphing capability with controlled NIR irradiation. Figure shows the controlled shape-morphing behavior of a single-layer PMNC hydrogel under the NIR radiation (Figure a,b). Upon placement of a cuboid (length of ∼2 cm, width of ∼5 mm, and height of ∼5 mm) PMNC hydrogel in a Petri dish at room temperature, NIR radiation (5 W cm–2) produces a local volumetric shrinkage along the irradiation direction, as shown in Figure a,b. Moreover, the PMNC hydrogel also exhibits a high responsive rate of NIR light-responsive behavior, which is among the top level according to the literature (Figure c).[33,45−50] The typical shrinkage and recovery time of the PMNC hydrogel are 15 and 8 s, respectively. Furthermore, the deformation direction of the PMNC hydrogel can be precisely tuned by controlling the irradiation position of NIR laser (Figure a,b, Videos S3 and S4). For example, when irradiating the upper surface of the PMNC hydrogel, the hydrogel bends upward. Lastly, we use the hyperelastic gel theory to study large deformations in polymeric gels subjected to an inhomogeneous swelling caused by external mechanical loads. For the kinetics of solvent migration, we make use of the similarity between heat conduction and mass diffusion. Therefore, instead of reformulating an entirely new user-defined element, we can use the inbuilt coupled temperature-displacement analysis in ABAQUS for simulation studies. This analogy significantly reduces the amount of time taken for the finite element formulation of gel-swelling kinetics. Interestingly, the simulation results are in good agreement with the experimental results (Figure a,b), further demonstrating that our model can use the programmed hydrogel structure to guide the design of complex shape-morphing structures.

Figure 7

NIR-guided shape-morphing capability of PNIPAM-MWNT nanocomposite hydrogels. (a,b) Volume shrinkage and direction control behavior under NIR irradiation (5 W cm–1). Cuboid shape with a length of ∼2 cm, width of ∼5 mm, and height of ∼5 mm. The red dotted line represents the irradiation area, and the red solid line arrow represents the irradiation direction. (c) Comparison of actuation speed with PNIPAM-based actuators with different photo-thermal agents. Bending speed = bending angle/(bending time × power density) and unbending speed = unbending angle/(unbending time × power density), where the power density is that of the NIR laser. The photo-thermal agents include carbon materials,[45,50] polymers,[46,49] and transitional-metal dichalcogenides.[33,47,48] (c) is reproduced with permissions.[33,45−50]

Conclusions

In summary, a cost-effective two-step FT method for the efficient synthesis of porous PNIPAM hydrogels with fast response, large swelling ratio, and high fatigue resistance has been successfully developed. The proposed two-step procedure following the freezing polymerization–thawing polymerization sequence ensures the porosity and pore interconnectivity (as opposed to the polymerization-freezing sequence) within a significantly shorter period (merely 2 h instead of a dozen hours or a few days), resulting in an enhanced performance of thermal responsiveness and swelling ratio of the PNIPAM hydrogel. Furthermore, by adding MWNTs, the PNIPAM hydrogel has shown excellent photo-thermal response characteristics, which possesses rapid NIR light responsiveness and programmable shape-morphing capability. The cost-effective and efficient synthesis method for fabricating high-performance hydrogels is expected to pave a way toward various potential applications, including drug delivery, bioengineering, and smart sensing.

Experimental Section

Materials

NIPAM and TEMED were purchased from TCI (Shanghai) Development Co. Ltd. N,N′-Methylenebisacrylamide (Bis), sodium dodecyl benzenesulfonate (SDBS), and MWNTs were purchased from Sigma-Aldrich. APS was obtained from Aladdin (China). Deionized water was used throughout the experiments. All the chemicals were used without further treatment.

Synthesis of FT Hydrogels

FT PNIPAM hydrogels were prepared with NIPAM as the monomer, Bis as the chemical cross-linker, and APS and TEMED as a redox-initiating system by a two-step polymerization. Specifically, fixed NIPAM (0.1 g/mL, 0.5 g), APS (2 mg/mL, 10 mg), and variable Bis (1–10 mg) were dissolved in deionized water (5 mL) to obtain a homogeneous solution in an ice bath. And the mixture was followed by deoxygenation with N2 (99.99%) for 10 min; then, TEMED (10 μL) was added under vigorous stirring at 0 °C for 30 s. Next, the mixed solution was quickly transferred to a refrigerator with a preset temperature. The synthesis was carried out in two steps: the first step is to induce the ice crystal at −20 °C and the second is to further polymerize at 4 °C. A thermostatic refrigerator was used to control the temperature required for the experiment. It was observed that the mixed solution formed a frozen state within 4–5 min at −20 °C. Subsequently, the gels were taken out and immersed in deionized water for 12 h to remove any unreacted reagent and water was replaced every 4 h. After that, the gels were stored in deionized water. In the experiments, the two-step polymerization time was typically set to 1/1, 2/2, 9/9, 12/12, and 24/24 h. The as-prepared PNIPAM hydrogels were designated x/y h, in which “x” represents the freezing polymerization time of the first step (at −24 °C) and “y” indicates the thawing polymerization time of the second step (at 4 °C). In addition, the hydrogel properties were optimized by varying the Bis concentration (0.2, 0.5, 1.0, 1.5, and 2.0 mg/mL). And the hydrogels were marked as Bis z, in which “z” represents the Bis concentration.

Synthesis of PF Hydrogels

PF PNIPAM hydrogels were synthesized from the order of first polymerization and freezing. In brief, NIPAM (0.1 g/mL, 0.5 g), APS (2 mg/mL, 10 mg), and Bis (0.2 mg/mL, 1 mg) were dissolved in deionized water (5 mL) to obtain a homogeneous solution in an ice bath. And the mixture was followed by deoxygenation with N2 (99.99%) for 10 min; then, TEMED (10 μL) was added under vigorous stirring at 0 °C for 30 s. Next, the mixed solution was quickly transferred to a refrigerator with a preset temperature. The mixed solution was first polymerized at 15 °C for 15 min and then frozen at −24 °C for 24 h.

Synthesis of FT PNIPAM-MWNT Nanocomposite Hydrogels

In brief, surfactant SDBS (8 mg/mL, 40 mg), MWNTs (0.5 mg/mL, 2.5 mg), and deionized water (5 mL) were mixed under ultra-sonication for 30 min to form a uniform dispersion. Then, NIPAM (0.1 g/mL, 0.5 g), APS (2 mg/mL, 10 mg), and Bis (0.2 mg/mL, 1 mg) were added to the mixed dispersion and stirred to obtain a homogeneous dispersion in an ice bath. And the mixture was followed by deoxygenation with N2 (99.99%) for 10 min; then, TEMED (10 μL) was added under vigorous stirring at 0 °C for 30 s. Next, the polymerization time of the two-step method was set to 2/2 h to prepare the PNIPAM-MWNT nanocomposite hydrogels.

Determination of Gel Fraction of FT PNIPAM

The gel fraction was measured by the gravimetric method. At the end of the first step, we quickly exposed the sample to air and immersed in a large amount of water to prevent it from continuing to polymerize. Then, we washed the sample three times repeatedly to remove unreacted components. Thereafter, the sample was dried at 90 °C under vacuum until the constant weight was reached. The gel fraction was obtained from the ratio of the gel mass after polymerization to the monomer mass before polymerization. The gel fraction in the second step was measured using the same method.

SEM of Hydrogels

The porous structures of freeze-dried hydrogels were observed by field emission SEM (Nova NanoSEM 450) at an acceleration voltage of 10 kV. The freeze-dried samples were prepared by rapid freezing of swollen hydrogels under liquid nitrogen (−196 °C) for 15 min and drying in a freeze-drier (FreeZone, Labconco) at −48 °C for 48 h. Then, the specimens were coated with a thin gold layer for SEM observation.

Characterization of Thermal-Responsive Properties

The gravimetric method was utilized to monitor the swelling degree of PNIPAM hydrogels in the range from 20 to 42 °C (temperature interval: 3 °C). The specimens were cut into discs with a diameter of 15 mm and a height of 3 mm. The temperature was controlled by a water bath pot. Before each measurement, the hydrogel samples were completely swelled at 20 °C for 1 h. Then, each hydrogel sample was put into the water bath pot. At a certain temperature, the samples were immersed in water for 10 min and then taken out. The surface water of samples was removed by a filter paper to record the sample weights (WT). The swelling ratio (SR) of hydrogel samples was determined according to the following equation: SR = WT/Wd, where Wd and WT were the weight of the dried gel and the weight of swollen gels at different temperatures, respectively. The weights of dry gel were measured with an electronic balance (OHAUS, EX125DZH) after drying each sample under vacuum at 80 °C until the weight become constant.

For the swelling-deswelling behavior, the thermo-responsive equilibrium volume change behaviors of the hydrogel discs were recorded by a digital camera at 20 and 45 °C, respectively. In the volume change behaviors of the hydrogel, we assumed that the hydrogels are homogeneously shrunk and swollen. The dynamic volumetric variation of the hydrogels would be defined as V/V0 = (d2 × h)/(d02 × h0), where V0 represents the initial equilibrium volumes of hydrogels at 20 °C and V represents the volumes of hydrogels at time t in the process of thermal response; d0 and h0 represent the initial diameters and heights of hydrogels at 20 °C, respectively; and d and h represent the diameters and heights at time t in the process of thermal response, respectively.

Characterization of Mechanical Properties

The compressive test of as-prepared hydrogels was measured by a dynamic mechanical analyzer (RSA-G2, TA, USA). The swollen specimens were cut into a cylindrical shape with a diameter of 15 mm and a height of 15 mm and compressed at a strain rate of 10 mm/min. After each test, an appropriate quantity of deionized water was added to the sample to ensure that the sample was in a fully swollen state. The compressive test was operated at room temperature.

Morphology Analysis

The average pore size was calculated by manually measuring the diameter of 100 pores with the imageJ software, and the pore size distribution was statistically obtained from the diameter of 100 pores. The area porosity was calculated by the threshold function of imageJ.

Characterization of NIR Light Responsiveness

The laser head (wavelength of 808 nm, output power of 500 mW, LM8085003D12-AL) was used for the NIR light-responsive tests. In order to measure the volume change of the PNIPAM-MWNT hydrogels responding to the NIR light, the swollen hydrogel samples were cut into cuboids (length of 3.5 mm, width of 7 mm, and thickness of 1 mm) and exposed to NIR light at a power density of 5 W cm–2 in water at room temperature. And to investigate the NIR light-responsive reversibility of the proposed PNIPAM-MWNT hydrogels, the laser was set to turn on and off for 8 s duration. The thermo-responsive equilibrium volume change behaviors of the above samples were recorded by a digital camera. The dynamic volume change ratios were defined as V/V0 = l × w × t/(l0 × w0 × t0), where V and V0 were the volumes of hydrogels at time t and at the beginning (t = 0), respectively; and l, w, t and l0, w0, t0 were the length, width, thickness of the hydrogels at time t and at the beginning of NIR illumination, respectively.

Finite Element Analysis of Hydrogels

The commercial software ABAQUS (SIMULIA) was used to carry out the finite element analysis (FEA), in which the implicit solver was used to calculate the deformations and stress–strain curves. The geometrical nonlinearities were considered in the FEA. Eight-node linear hybrid brick elements with reduced integration were adopted, with refined meshes to ensure computational accuracy. We assume local equilibrium at all material points within the volume and subsequently describe the equilibrium condition by applying the hyperelastic gel theory developed by Hong et al.[51]