Entanglement in Smart Hydrogels: Fast Response Time, Anti‐Freezing and Anti‐Drying

The common techniques to improve hydrogel's mechanical properties include increasing crosslinking density and forming crosslinked double‐network hydrogel, which may cause some hydrogels to lose their smart functionalities. Inspired by entanglement‐induced strengthening, a simple approach to introducing hydroxypropyl cellulose (HPC) fibers entangled with different smart hydrogel matrix systems are reported. Different from the conventional methods which hinder the movement of the polymer network, through entanglement with HPC fibers, the composite hydrogel shows both improved Young's modulus and toughness and more importantly improved smart functionalities including response speed, anti‐drying, and anti‐freezing capabilities and cycle stability. This strategy provides a new design rule to fabricate durable and strengthened smart hydrogels which can be used in smart windows, sensors, and soft robots.

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adfm.202211027. and sensors. [13,14] Conventional hydrogels strengthen techniques such as forming crosslinked double-network hydrogel, [15,16] increasing crosslinking degree, [17,18] forming high crystalline polymer chain, [19,20] and salting [21,22] will form rigid covalent bonds and restrict the movement of polymer chains, which could deactivate the smart function in some hydrogels. [22] Meanwhile, there is literature that reported the application of reinforcement agents such as cellulose nanofibers, [23][24][25] carboxymethyl cellulose, [26] bacterial cellulose, [27] and acidochromic regenerated cellulose [28] in hydrogel with the main purpose to improve its mechanical strength. However, as these cellulose-based fibers scatter the light, the reinforced hydrogel cannot be used for applications requiring high optical transparency. Moreover, in real-world applications, the retained smart functionality, anti-freezing, anti-drying, and enhanced durability are also of prime importance for smart hydrogels. Enhancing entanglement in hydrogel [29][30][31] has been recently reported to improve the mechanical property of hydrogel as it helps to transmit tension between the chain and dissipate the elastic energy. The introduction of entanglement does not hinder the movement of polymer network nor embrittle the hydrogel, which could be applied to smart hydrogel as the phase change process of it involves the moving of polymer network.
Inspired by the literature, [29][30][31] we reported a simple approach to enhance the entanglement in smart hydrogel through compositing matrix with hydroxypropyl cellulose (HPC) nanofibers. Compared with conventional smart hydrogel in which the polymer networks interconnect with each other through crosslinking (Figure 1 left), the HPC fibers entangle with the matrix and form a weak hydrogen bond (H-bond) (Figure 1 right). As the result, the mechanical property of smart hydrogel increases with its smart functionality preserved since the entanglement and H-bond do not hinder the movement of polymer network structure. Through compositing HPC with thermal-responsive poly(N-iso-propylacrylamide) (PNIPAm), the new PNIPAm-HPC composite shows ≈10and 22-times improvement in Young's modulus and toughness compared with pure PNIPAm hydrogel. More importantly, compared with the control, the smart functionality of PNIPAm-HPC composite has been improved with better cycle stability, faster response speed, and enhanced anti-drying/anti-freezing abilities.

Introduction
The stimuli-responsive smart hydrogel has the advantages of biocompatibility, [1,2] conductivity, [3,4] and flexibility, [5,6] which is of interest for robots, [7,8] medical devices, [9,10] smart windows, [11,12] We demonstrate the smart window applications using the HPC entangled PNIPAm hydrogel which shows energy-saving ability with high radiative cooling (RC) capability. It saves 16.1% annual heating/cooling energy compared with commercial low-E glass. We further investigated the general applicability of this entanglement enhancement technique. Similar to PNIPAm-HPC system, significant performance improvements in mechanical integrity, cycle stability, response speed, and anti-drying/anti-freezing abilities were observed in PNIPAmpolyvinyl alcohol (PVA)-HPC system. This new strategy of enhancing entanglement provides a simple but universal way to strengthen the hydrogel with improved smart functionalities.

Figure 2a
shows the Fourier transform infrared spectra for pure PNIPAm (control sample) and PNIPAm-HPC composite hydrogel with different HPC content. All four samples showed absorption peaks at 1253, 2972, 2934, and 2882 cm −1 , and these peaks are due to the CN stretching vibration and CH vibration peak of saturated hydrocarbon (methyl, methylene, and methenyl, respectively) in PNIPAm. [32] It is worth mentioning that the characteristic peak at 3307 cm −1 (marked by the black arrows in Figure 2a) is weakened with the HPC content increasing, while in the pure HPC control sample, there is no 3307 cm −1 peak observed. As this peak relates to the scalation vibration of the N-H bond, [33] the weakened absorption peak may be attributed to the formation of intermolecular hydrogen bonds between HPC and PNIPAm. Similarly, the shifting of NH bond stretching peak at 1648 cm −1 for pure PNIPAm hydrogel to 1658 cm −1 in the PNIPAm-13% HPC shows the effect of forming hydrogen bonds to the stretching of CO. Moreover, as there is no new characteristic peak observed in the composite hydrogel, there is no new chemical bond formed between PNIPAm and HPC. The bond that connects two types of hydrogels is hydrogen bond. The dynamic rheological test result of PNIPAm-HPC composite hydrogels and control were shown in Figure 2b. Both samples have a storage modulus (G′) higher than the loss modulus (G″) over the whole frequency range, demonstrating elastic behaviors. Moreover, the G′ and G″ values of PNIPAm-HPC hydrogels are higher than those of control. It is worth mentioning that PNIPAm-HPC hydrogels have lower transition temperatures compared with both PNIPAm and HPC by investigating with Differential Scanning Calorimetry (DSC) (Figure 2c). The transition temperatures for PNIPAm and HPC are 32 °C and 34 °C, respectively, while the transition temperature is reduced to ≈29 °C after the composition. Photo of control and composite hydrogels at different temperatures are shown in Figure 2d. At 20 °C, all the hydrogels show good transparency. While at 40 °C, composite hydrogels become opaque, and the control turns translucent. The Scanning Electron Microscope (SEM) images of freeze-dried PNIPAm-HPC, control samples, and pure HPC samples are shown in Figure 2e. Compared with porous structure in pure PNIPAm control sample and fibrous HPC sample, PNIPAm-HPC sample has a fiber structure. At 20 °C, the fibers are thin and elongated to allow light to pass through. However, at 40 °C, the fibers aggregate, and their diameter increase from ≈200 nm to ≈5 µm. As the result, the aggregated fibers serve as scattering centers of light and the transmittance of hydrogel reduces. [34] Meanwhile, the Atomic Force Microscope image of sheared HPC solution shows that the HPC film has a smooth and continuous surface with the presence of islands and gaps on the surface ( Figure S1, Supporting Information).  The mechanical properties of PNIPAm-HPC composite and pure hydrogel were evaluated with a tensile tester. The control can be stretched from 2 cm to 6 cm before breaking (Figure 2f). In contrast, due to the introduction of entanglement, the PNIPAm-HPC hydrogel can be stretched to 9, 12, and 19 cm with increasing composition with 5%, 9%, and 13% HPC respectively. The stretch-stress curve shows the same trend: with the HPC content increasing, the hydrogel shows improved mechanical strength (Figure 2g). By compositing with HPC, the hydrogel has 10 times higher in Young's modulus (20.0 ± 0.2 kPa vs 2.1 ± 0.2 kPa) and 21.8 times increasing in toughness (114 ± 0.2 kJ m −3 vs 5 ± 0.3 kJ m −3 ) against baseline ( Figure 2h; Table S2, Supporting Information). The significant increase in Young's modulus and toughness illustrates the importance of enhancing entanglement and these results are consistent with the previous reports. [29,30]

Multifunctionalities: Anti-Freezing, Anti-Drying, Smart Functionality, Response Time, and Durability
We explore the newly improved functionalities of the entangled smart hydrogel. DSC was used to investigate the freezing performance of the composite (Figure 3a). It can be observed that with increasing HPC content, the freezing point of hydrogel decreases. The rheological behavior change of hydrogels around the freezing point was thereby investigated (Figure 3b  seen that all four samples experienced abrupt changes in G′ and G″. From 5 °C to −5 °C, the control is elastic as the G′ is higher than G″. When the temperature is below -5 °C, both G′ and G″ increases dramatically. This phenomenon indicates that the control has lost its elasticity and become fragile. In contrast, PNIPAm-13% HPC keeps its elasticity at −13 °C. The  lowered freezing temperature might be due to the hydrogen bond formed between HPC and PNIPAm. By calculating the interaction energy of molecules with density functional theory (DFT) simulation, it was shown that the interaction energy between water and HPC is lower than that of water-water (Table S3, Supporting Information). Therefore, the HPC tends to form H-bond with water, which will disturb the forming of a tetrahedral H-bond network between water molecules. [35] As a result, H-bond introduced by HPC prevents water crystallization at low temperatures. Since in harsh environmental conditions such as in cold winter, good anti-freezing performance is essential, [36] PNIPAm-HPC hydrogel becomes a good candidate for application in variable conditions. It is worth mentioning that the good stretchability at room temperature ( Figure 2f) is preserved even after freeze-drying. Compared with the rigid pure control aerogel, the PNIPAm-HPC aerogel is able to be stretched, folded, and twisted (Movies S1 and S2, Supporting Information).
As a smart hydrogel, the response time is one of the key performance indexes. The response time of PNIPAm-HPC hydrogel was evaluated by recording its transmittance change in the heating and cooling process. Figure 3c is the response time of PNIPAm-HPC hydrogel during heating and cooling by recording the time for the hydrogel to fully change its transmittance. In the heating process from 20 °C to 40 °C, the control sample took ≈37 s while the PNIPAm-13% HPC hydrogel took 25 s to change from transparent to opaque. Meanwhile, the control sample took 15 s to restore its transparency. In contrast, PNIPAm-13% HPC only required ≈8 s. The curves of transmittance@650 nm against heating/cooling time are shown in Figure S2 (Supporting Information). The faster response time of PNIPAm-HPC hydrogel is demonstrated in Figure 3d. After 20 s of heating, PNIPAm-HPC has become opaque while the control is still translucent. Similarly, the composite hydrogel takes ≈10 s to become transparent during cooling and the control takes a longer time.
Moreover, PNIPAm-HPC composite hydrogel shows better anti-drying performance ( Figure 3e) and durability (Figure 3f) than the control. In the dry environment with 30RH% and 20 °C, the weight loss of PNIPAm-13% HPC due to water evaporation was 40% less than the baseline. Similar to the antifreezing property, the hydrogen bond formed between water and HPC prevents water from evaporation and drying. The good antidrying performance of PNIPAm-HPC hydrogel makes it a potential choice for open environment applications such as artificial skin, [37,38] building envelopes, [39] and information processing platforms. [40,41] At the same time, the PNIPAm-HPC hydrogel kept higher toughness than the baseline after 100 heating-cooling cycles as the enhanced entanglement maintains the mechanical integrity of polymer networks during repeated heating-cooling. The thermal degradation behavior of the control and composite hydrogels was analyzed using thermogravimetric (TGA) analysis. The main decomposition peaks of PNIPAm-HPC hydrogels moved to higher temperatures, with the maximum decomposition peak occurring at ≈425 °C, compared with that of the control at ≈400 °C. (Figure S3, Supporting Information).
Hydrogel is a well-studied thermochromic material used in smart windows. [11,42,43] We further demonstrated the thermochromic performance of the entangled hydrogel by comparing the solar modulation ability of pure and composite hydrogels. Figure 3g shows the UV-vis spectra for the control and composite hydrogels. It is worth noting that the two decreases in the transmittance at 1430 and 1930 nm corresponding to the OH stretching in water and the binding of OH stretching to the HOH bending. [44] From the spectra, it can be seen that composite hydrogels have a high luminous transmittance (T lum , 80.7%). Meanwhile, the composite, especially the PNIPAm-13% HPC sample has a better solar modulation ability (ΔT sol ) than the control (64.5% vs 52.1%). The longwave infrared (LWIR) emissivity spectra of PNIPAm-HPC hydrogel and PNIPAm at 20 and 40 °C are shown in Figure 3h and Figure S4 (Supporting Information), PNIPAm-HPC hydrogel shows high LWIR emissivity (ε LWIR 0.95 at 20 °C and 40 °C). As a near unity ε LWIR is preferred to strengthen the radiative cooling effect of material, [43,45] the PNIPAm-HPC hydrogel has good potential for RC application. Therefore, the high ΔT sol and ε LWIR are preferred in tropical weather regions where cooling is dominant. [46] The weather condition in Singapore is chosen to evaluate the monthly energy consumption and both clear glass and commercial low-E glass were used as a benchmark to compare with PNIPAm-HPC (Figure 3i). Among the three samples, PNIPAm-HPC saves 24.6% and 16.1% energy compared with clear glass and energy-saving commercial low-E glass annually. The energy consumption simulation illustrates that PNIPAm-HPC composite hydrogel demonstrated energy saving with high RC capability and solar modulation ability.
Aging test was conducted to evaluate the durability of PNIPAm-HPC hydrogel ( Figure S5, Supporting Information). In the continuous 30 days of aging, the hydrogel preserved its T lum and ΔT sol with negligible performance deterioration. In summary, the enhanced entanglement and weak H-bond in smart hydrogel could not only improve mechanical strength such as Young's modulus and toughness but also maintain smart functionality with faster response speed, enhanced anti-drying/freezing performance, and better durability, than pure hydrogel (Figure 3j).

Applicability in other Smart Hydrogel Systems
In the previous section, HPC is introduced into PNIPAm smart hydrogel to enhance entanglement and introduce weak H-bond, giving improved mechanical properties, wider operating temperature range, enhanced anti-drying and antifreezing ability, cycle stability, and more importantly, the retained smart functionality with a faster response speed. We further deploy the same strategy to understand the general applicability of enhancing entanglement through compositing HPC into PNIPAm-PVA system. Similar to PNIPAm-HPC system, a mechanical property improvement was observed in PNIPAm-PVA-HPC system (Figure 4a). The Young's modulus of PNIPAm-PVA hydrogel composited with 3.5% HPC improved 65% compared with the pure one (1.7 ± 0.  Figure 3c, the 3.5% HPC content sample shows a response time decrease of 10 s (heating) and 48 s (cooling) compared with the control sample. In PNIPAm-HPC system shown in Figure 3c, the response time of PNIPAm-HPC for heating is longer than for cooling. While for PNIPAm-PVA-HPC system shown in Figure 4e, the response time in heating process is shorter than the response time during cooling. This phenomenon might attribute to the addition of PVA. The anti-drying performance of PNIPAm-PVA-HPC hydrogel is demonstrated in Figure 4f. both PNIPAm-PVA-HPC hydrogel and the control were exposed to the 25 °C, 50RH% environment for 48 h. The original diameter for both samples was 1.8 cm. After exposure, the diameter of the control sample decreased to ≈1.5 cm while the composite hydrogel was ≈1.7 cm. The less shrinkage of composite hydrogel indicates its good anti-drying performance. In the durability test, HPC composited samples show less toughness deterioration than the control after 100 heating/cooling cycles (Figure 4g). Lastly, the freezing point of hydrogel decreases with increasing HPC concentration (Figure 4h). In summary, similar to PNIPAm-HPC system concluded in Figure 3j, the performance of PNIPAm-PVA systems such as modulus, response time, anti-drying/freezing ability, and durability were improved through compositing with HPC.

Conclusions
We reported a simple and universal way to improve smart hydrogel's modulus and toughness while retaining its smart functionality via enhancing entanglement and introducing weak H-bonds. In this approach, HPC fibers were composited with different hydrogels (PNIPAm and PNIPAm-PVA) to provide entanglement. The Young's modulus and toughness of both systems have been improved compared with the control. More importantly, the smart functionality has been retained with faster response speed, a wider application temperature range, better anti-drying/freezing ability, and good cyclability. This entangled hydrogel has been demonstrated in thermochromic smart windows giving higher solar regulation ability and a near unity long wave infrared emissivity making it a good choice for tropical weather conditions. We believe this strategy could be extended to other stimulus-responsive hydrogels with better mechanical integrity and improved smart functionalities.
Preparation of PNIPAm-HPC Composite Hydrogel: NIPAm (1.0 g), 0.02 g BIS and variable amounts of HPC (0, 1, 1.8, and 2.6 g for 0%, 5%, 9%, and 13% samples, respectively) were dissolved into 20 ml DI water at room temperature. The mixture was stirred at 1500 rpm for 30 min to obtain a homogeneous solution. Then, 0.02 g APS was added to the solution; followed by stirring at 1500 rpm for 10 min. Subsequently, 6 µm TEMED was added to the solution. The solution was stirred for 3 min at 1500 rpm. Finally, the prepared solution was poured into a mold. The mold and the solution were then put at 4 °C for 2 h to obtain the final composite hydrogel. Preparation of PNIPAm-PVA-HPC Composite Hydrogel: NIPAm (1.0 g), 0.4 g PVA, 0.02 g BIS and a variable amount of HPC (0, 0.3, 0.5, and 0.7 g for 0%, 1.5%, 2.5%, and 3.5% sample, respectively) were dissolved into 20 ml DI water at room temperature. The mixture was stirred at 1500 rpm for 30 min to obtain a homogeneous solution. Then, 0.02 g APS was added to the solution; followed by stirring at 1500 rpm for 10 min. Subsequently, 6 µm TEMED was added to the solution. The solution was stirred for 3 min at 1500 rpm. Finally, the prepared solution was poured into a mold. The mold and the solution were then put at 4 °C for 2 h to obtain the final composite hydrogel.
Characterization: The SEM images of PNIPAm, HPC, and PNIPAm-HPC hydrogel were taken with SEM Supra 55 (Carl Zeiss). To obtain the microstructure of hydrogels below and above Lower critical solution temperature (30 °C), the two samples of PNIPAm-HPC hydrogel were placed at 20 and 40 °C separately for 1 h. Then the two samples were immediately submerged in liquid nitrogen until fully frozen. Due to the extremely low temperature of liquid nitrogen, the microstructures of hydrogel at different temperatures were preserved. [11,43] The frozen hydrogels were placed into a freeze-dryer for freeze-drying process for 72 h. The freeze-dried aerogels were subsequently used to take the SEM images.
The FTIR spectra of hydrogels were collected with an FTIR spectrometer (TENSOR 27, Bruker) in the range of 4000-500 cm −1 . The dynamic rheology test of the hydrogels was performed using a modular smart advanced rotational rheometer (MCR302, Anton Paar) equipped with 25 mm parallel plates. The strain amplitude of the dynamic rheology test was 0.5%.
The nominal stress-stretch curves were measured by the tensile compression testing machine (Instron 4465, Instron) at a strain rate of 0.2 s −1 . The hydrogel specimens used in this test had a dimension of 20 mm in length, 5 mm in width, and 3 mm in thickness. Tensile tests were performed until the fracture of hydrogel samples. The stretch was calculated by dividing the deformed length by the initial length. Young's modulus was calculated from the slope of the stress-stretch curve, while the toughness was obtained by integrating the area under stress-stretch curve.
Anti-Freezing Performance Characterization for Hydrogel: The freezing point of hydrogel was determined with a differential scanning calorimeter (DSC Q10, TA Instruments) with a ramping rate of 5.0 °C min −1 from 10 to −25 °C. The rheology test at low temperatures was conducted in the rheometer (MCR302, Anton Paar) equipped with a temperaturecontrolling attachment. The set of 8 mm parallel plates was used with a strain amplitude of 0.5%. The value of G′ and G″ were recorded from 5 °C to −15 °C.
Response Time Measurement for Hydrogel: The response time of composite hydrogel was estimated by plotting transmittance at 650 nm against heating-cooling time. A UV-vis spectrometer (AvaSpec-ULS2048L StarLine Versatile Fiber-optic Spectrometer, Avantes) with a heating-cooling stage (PE120, Linkam) equipped was employed in this test. Data recording was started when the sample temperature reached 40 °C for heating and 20 °C for cooling. The transmittance was measured with a time interval of 1 s. For the ease of comparing response times for different samples, the transmittance was normalized during the plotting.
Anti-drying Performance and Cycle Stability Measurement: The antidrying performance was measured by the weight loss of hydrogel in the environment of 30RH% and 20 °C. For hydrogels with different HPC content, a 2 g sample was taken for evaluation. The weight loss of the sample was recorded for a continuous 360 min.
The cycle stability of hydrogel was evaluated by measuring its mechanical properties after heating/cooling cycles. Pure PNIPAm and PNIPAm-HPC were used in this test. The Young's modulus and toughness were recorded for the original state, after 10 heating-cooling cycles, after 50 heating-cooling cycles, and after 100 cycles. The tensile compression testing machine (Instron 4465, Instron) was used in this test.
Optical Performance Characterization for PNIPAm-HPC Hydrogel: A UV-vis spectrometer system (AvaSpec-ULS2048L StarLine Versatile Fiber-optic Spectrometer and AvaSpec-NIR256-2.5-HSC-EVO, Avantes) with a heating-cooling stage (PE120, Linkam) equipped was employed to collect the UV-vis-NIR spectra of hydrogel in this manuscript. The integral luminous transmittance (T lum , 380-780 nm), IR transmittance (T IR, 780-2500 nm), and solar transmittance (T sol , 250-2500 nm) were calculated by Equation 1: In this formula T(λ) is spectral transmittance, ψ lum (λ) is the standard luminous efficiency function of photopic vision in the wavelength range of 380-780 nm. [47] On the other hand, ψ sol (λ) is the solar irradiance spectrum for air mass 1.5 (corresponding to the sun standing 37 ° above the horizon with 1.5-atmosphere thickness, which corresponds to a solar zenith angle of 48.2°). [48] LWIR Emissivity Measurement for PNIPAm-HPC Hydrogel: The LWIR emissivity of PNIPAm-HPC was measured with a dual-band emissivity measuring instrument (IR-2, Shanghai Chengbo Photoelectric Technology) with a heating stage equipped. PNIPAm-13%HPC hydrogel was used in this test. LWIR emissivity values of five spots were recorded and the average value was used. Meanwhile, the emissivity curve was plotted according to Kirchhoff's law of thermal radiation: ε(λ) = A(λ) = 1 − T(λ) − R(λ). [46] R(λ) and T(λ) are the spectral LWIR reflectance and the spectral LWIR transmittance that was measured by FTIR spectrometer (Perkin Elmer Frontier) with an integration sphere attached, respectively.
DFT Simulation of Interactive Energy Between Molecules: To study the interaction energy between molecules in PNIPAm-HPC composite hydrogel, 7 molecular models were used in the DFT simulation; namely: 1) water molecule and water molecule (W-W); 2) PNIPAm and water molecules (P-W); 3) HPC molecule and water molecule (H-W); 4) PNIPAm and HPC (P-H); 5) PNIPAm coexisted with HPC and water (P-H-W); 6) HPC molecule and HPC molecule; 7) PNIPAm molecule and PNIPAm molecule. These molecular models were shown in Figure S8 (Supporting Information).
All calculations including geometry optimizations and frequency analysis were performed on Gaussian 09D01 program suites, [49] and the M062X-D3 functional was chosen. [50] Geometry optimizations and frequency analysis were performed at the DFT M062X-D3/6-311+G (d, p) level. Considering the effect of solvent, the IEF-PCM water solvation model was chosen. [51] Interaction energies were calculated based on the enthalpy (Corrected by electron energy) of corresponding species.
Energy Consumption Simulation for PNIPAm-HPC Composite Hydrogel: EnergyPlus was used in the energy consumption simulation. In this simulation, a building model with the dimensions of 8 m in length, 6 m in width, and 2.7 m in height was used. The floor area was 48 m 2 and total external wall surface area of the building was 75.6 m 2 . Four windows with the dimension of 3 m in width and 2 m in height were installed in the four orientations to avoid the impact of orientation. The window covered 31.7% of the total wall surface area. The structure of the model house was shown in Figure S9 (Supporting Information). Hourly weather data for a Typical Meteorological Year was employed as the external boundary conditions. The weather data of Singapore was used in this simulation. [52] In this simulation, the energy usage based on the unit of MJ m −2 was calculated based on the floor area. The properties of samples used in this simulation were listed in Table S5 (Supporting  Information).

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.