A bio-inspired thermo-responsive hydrogel purifier for effective water harvesting in seawater

Yang Xiao† ^a, Yanqiong Bao†^b, Lang Liu†^b, Wanlin Xu^a, Da Li^a, Xiong Zheng*^a, Guangzhao Qin^a and Qing Li^c
aCollege of Mechanical and Vehicle Engineering, Hunan University, Changsha, Hunan 410082, PR China. E-mail: xzheng@hnu.edu.cn
^bKey Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education of China, Chongqing University, Chongqing 400044, PR China
^cSchool of Energy Science and Engineering, Central South University, Changsha 410083, PR China

First published on 6th August 2025

---

---

1 Introduction

Freshwater scarcity has emerged as one of the most critical challenges facing humanity in the 21st century, driven by exponential population growth and rapid industrial expansion.¹ While seawater desalination technologies offer a promising solution to augment terrestrial water resources,² conventional methods such as reverse osmosis and flash distillation remain energy-intensive and environmentally taxing.³ This predicament has catalyzed global efforts to develop sustainable desalination systems powered by renewable energy sources.

Solar-driven interfacial evaporation represents a paradigm shift in desalination technology, combining environmental sustainability with operational efficiency.⁴ Since the pioneering work on solar evaporators in 2014,⁵ significant progress has been achieved through innovations in photothermal materials,⁶ thermal localization strategies,⁷ and anti-fouling architecture.^8,9 Notably, thermo-responsive (TR) hydrogels have recently gained prominence due to their unique phase-transition behavior governed by lower critical solution temperature (LCST) characteristics.¹⁰ These smart materials demonstrate temperature-dependent hydrophilicity switching: maintaining hydrophilic water-absorption states below LCST and transitioning to hydrophobic water-releasing states when heated above LCST through solar irradiation.^11–13 The inherent thermal gradient within the evaporator (surface heating vs. bulk cooling) further induces Janus wettability, creating directional water transport pathways that synergistically enhance evaporation rates and salt rejection.¹⁴ Despite these advancements, conventional interfacial evaporators remain fundamentally limited by their passive water transport mechanisms and single-phase operation.

Recent breakthroughs have introduced a third-generation TR purification system that integrates dual-phase water management through evaporation–guttation coupling. The seminal work by Geng et al.¹⁵ demonstrated this concept using poly(n-isopropyl acrylamide) (PNIPAm) hydrogel-coated melamine foam, achieving unprecedented freshwater yields (4.20 kg m⁻² h⁻¹). Subsequent innovations in multilayer architecture,^16,17 nanocomposite hydrogels,¹⁸ and bioinspired phase-transition designs¹⁹ have progressively elevated production rates to 28.07 kg m⁻² h⁻¹. These systems leverage dynamic swelling/collapse transitions to simultaneously drive surface evaporation and bulk water guttation. Nevertheless, two fundamental challenges remain unresolved: first, inadequate optimization of coupled heat–mass transfer processes constrains maximum yield potential; second, predominant focus on wastewater treatment applications rather than seawater desalination.

Nature's evolutionary solutions provide valuable inspiration for overcoming these technological barriers. The reticulated carpet shark, a small deep-water species, displays a unique anti-predator behavior: inflating its stomach with water to increase apparent size when threatened. It ceases water absorption and returns to normal morphology once the threat subsides. Inspired by the unique mechanisms of the reticulated carpet shark, we have constructed a TR purifier, which integrates a poly(N-isopropylacrylamide) (PNIPAm)/polyvinyl alcohol (PVA) hydrogel matrix with a polyamide (PA) salt-rejection barrier, achieving unprecedented freshwater generation through coordinated absorption–evaporation–guttation cycles. PVA incorporation induces increase in the hydrophilic functional groups of PNIPAm hydrogel and formation of interpenetrating polymer networks, significantly improving swelling kinetics and structural integrity. Vertically aligned microchannels with graphene (GE) dopants create anisotropic thermal conduction pathways for localized surface heating. The nanostructured PA barrier achieves excellent salt rejection while maintaining exceptional interfacial adhesion with the hydrogel matrix. The resultant system achieves record-breaking freshwater production (29.99 kg m⁻² h⁻¹) with sustained salt rejection efficiency and water purification. The working principle of the TR purifier is illustrated in Fig. 1. Our findings not only establish new benchmarks for solar desalination performance but also provide fundamental insights into multiphase transport optimization in smart hydrogel systems.

2 Results and discussion

2.1 Design concept of the purifiers

To optimize the performance of the TR purifier, various hydrogel compositions, GE concentrations, and hydrogel micro-structures are employed to enhance water transport, heat transfer, and light absorption capabilities. Table 1 outlines the components, micro-structures, PA membrane usage, and abbreviations involved.

---

2.2 Characterization

Fig. 2a depicts the comprehensive structural diagram of the purifier, comprising a PA membrane on its surface, a hydrogel inside, and a GE filler. SEM images of two types of internal micro-structures of the hydrogels are presented in Fig. 2b. The hydrogel, produced using normal freeze-drying methods, features porous structures with a diameter of approximately 530 μm that are distributed in a random pattern. In contrast, the directionally freeze-dried hydrogel exhibits a distinctly different internal micro-structure, where the cross-linked structure is frozen into multi-layers with a spacing of roughly 400 μm, offering exceptional channels for directional heat and water transport. GE exhibits a sheet-like structure with a diameter of approximately 30 μm, and there is no overlap between different sheets, indicating a superior dispersion quality (Fig. 2b). Furthermore, the elemental mapping of the hydrogel demonstrates a uniform distribution of C, H, O, and N within the hydrogel (as shown in Fig. 2c), validating the consistent composition of the components within the purifier.

For the TR purifier, exceptional water capture ability is crucial to guarantee its substantial swelling capacity, so the hydrogel's water capture ability is systematically evaluated through water contact angle (WAC) and FTIR spectrum analysis. The WAC of the pure PNIPAm (NV0) stands at 92.9° (Fig. 2d), signifying its relatively poor hydrophilicity. However, the introduction of PVA hydrogel significantly enhances the hydrophilicity, reducing the WACs of PNIPAm to 31.8° and even 0° when blended with 20 and 40 wt% PVA, respectively. Furthermore, at 40 °C (above the LCST), the WAC of the NV20 purifier increases to 75.3°, indicating that the NV20 purifier undergoes a notable hydrophilic-to-hydrophobic transition around the LCST, thereby ensuring efficient guttation functionality.

To explore hydrogel hydrophilicity, we analyzed FTIR spectra of PNIPAm, PVA, and NV20 (Fig. 2e). PNIPAm's spectrum shows amide peaks near 1650 cm⁻¹ and 1540 cm⁻¹, confirming the presence of amide, which forms hydrogen bonds with water, creating a solvation layer that induces an extended coil structure and exhibits hydrophilic behavior. Isopropyl groups (peaks at 2960 and 2930 cm⁻¹) weaken hydrophilicity, resulting in modest hydrophilicity in PNIPAm hydrogels. PVA's spectrum showcases –OH stretching near 3300 cm⁻¹, which promotes hydrogen bond formation and a 3D network structure with water, enhancing PVA's hydrophilicity. NV20's spectrum blends the peaks of PNIPAm's amide and those of PVA's –OH and –CH/–CH₂, which improve NV20's hydrophilicity over pure PNIPAm due to PVA's hydroxyl groups.

Furthermore, the mass ratios of various purifiers in their swollen and collapsed states are depicted in Fig. 2f. An intriguing parabolic relationship emerges between the PVA content and the mass ratio. Specifically, NV0 exhibits a relatively low mass ratio of 1.57 g/g, which can be attributed to the inherent poor hydrophilicity of PNIPAm. In contrast, NV20 achieves the maximum mass ratio of 8.51 g/g. The incorporation of PVA not only bolsters the hydrogel's hydrophilicity but also diminishes its thermo-responsive capabilities. Consequently, further increase in PVA content leads to a reduction in guttation mass, ultimately resulting in a decrease in the mass ratio.

To gain a deeper insight into the swelling/collapse mechanics of the TR purifier, we employ two molecular dynamics (MD) models: one comprising pure PNIPAm and the other incorporating PNIPAm/PVA hydrogels. These models simulate the swelling and collapse conformations of the TR purifiers during phase transitions. Fig. 3g presents the simulated molecular distribution maps. Initially, in both the PNIPAm–water and PNIPAm/PVA–water systems, the molecular chains are meticulously arranged within the water. Upon adjusting the temperature to either 293 or 333 K, the systems attain equilibrium within 12 nanoseconds. The molecular distribution maps distinctly reveal a swollen state at 293 K and a collapsed state at 333 K, respectively.

To quantitatively assess the swelling–collapse transition of the hydrogel, we conducted statistical analyses of the molecular spatial distribution density for both systems, as illustrated in Fig. 2h. Following the temperature shift from 293 to 333 K, the molecular chains in both systems exhibit a tendency to cluster towards the upper and lower boundaries. Consequently, the average distance between the molecular chain and the boundary decreased from 25.9 to 21.3 Å in the PNIPAm–water system, and from 28.4 to 22.6 Å in the PNIPAm/PVA–water system. This finding suggests that the incorporation of PVA attenuates the swelling amplitude within the hydrogel, which agrees well with the above experimental results. Meanwhile, the simulation outcome indicates that the introduction of PVA diminishes the thermal sensitivity of PNIPAm, thereby significantly reducing the guttation rate of the purifier, particularly at high PVA concentrations. Besides, the average number of hydrogen bonds (HBs) in different systems are compared to further reveal the mechanism of the effect of PVA addition on the conformation transition (Fig. 2i). Research has revealed that the capacity of PNIPAm to undergo swelling and collapse in water is profoundly impacted by variations in its HBs both beneath and above LCST.²⁰ In the PNIPAm–water system, the average disparity in the number of HBs between the swollen and collapsed states amounts to 19.8%. However, in the PNIPAm/PVA–water system, this disparity escalates to 35.8%, indicating that the incorporation of PVA enhances the swelling/collapse rate of the TR hydrogel.

To evaluate the long-term durability of the purifier, its mechanical robustness was systematically assessed in the fully swollen state. Initially, the reversible compressive stress–strain curves of NV0 and NV20 purifiers were measured under an 80% compressive strain (Fig. 3a). Both purifiers demonstrated exceptional flexibility, withstanding the 80% strain without failure. However, NV0 exhibited relatively lower mechanical strength, reaching a stress of only 41 kPa at 80% strain. In contrast, the incorporation of 20 wt% PVA significantly enhanced the mechanical properties, increasing the stress to 122 kPa for NV20. Furthermore, as shown in Fig. 3b, NV20 exhibited remarkable resistance to bending, twisting, and stretching, highlighting its structural integrity under various mechanical stresses. These results confirm the outstanding durability of the fabricated purifier. In addition, the long-term mechanical performance of the purifier was analyzed to assess its operational stability. As illustrated in Fig. 3a, the compressive stress–strain curve of the purifier after 50 working cycles nearly overlaps with that of the newly prepared purifier, indicating minimal degradation in mechanical properties over time. Moreover, the physical image of the compression–recovery experiment in Fig. 3c also confirms its excellent durability. These results prove the absence of fatigue fracture or structural damage within the purifier after prolonged operation, underscoring its robust and durable design.

For effective heat transfer from top to bottom in TR purifiers, the thermal conductivities of hydrogels with varying GE contents and micro-structures are measured to identify methods for enhancing directional thermal conductivity. Fig. 3d depicts the thermal conductivities of ordinary and directional hydrogels with a PNIPAm:PVA mass ratio of 8:2 at 25 °C under dry conditions. The thermal conductivity of the ordinary hydrogel is 0.0375 W m⁻² K⁻¹, while the axial and radial thermal conductivities of the directional hydrogel are 0.0602 and 0.0113 W m⁻² K⁻¹, respectively. The addition of GE augments the thermal conductivities in both types of hydrogels. At a GE content of 0.1 wt%, the thermal conductivity of the standard hydrogel rises to 0.1073 W m⁻² K⁻¹. Furthermore, the axial and radial thermal conductivities of the directional hydrogel are 0.1673 W m⁻² K⁻¹ and 0.0265 W m⁻² K⁻¹, respectively. These results indicate that the construction of micro-layers can induce significant anisotropic thermal conductivity in hydrogels, which is pivotal for improving the thermal management quality of TR purifiers.

Apart from enhancing thermal properties, the incorporation of GE significantly elevates the solar absorptance of the hydrogel, thanks to its distinctive electronic structure and profound interaction with light waves.²¹ Fig. 3e illustrates the light absorptance of hydrogels with different GE contents, with a constant thickness of 1.0 mm. The pure hydrogel exhibits relatively high absorptance in the ultraviolet and infrared regions but performs poorly in the primary concentration area of sunlight (300–1100 nm). The addition of GE improves absorptance across the entire spectrum, and a 0.1 wt% GE loading enhances the average absorptance to 95%, proving that the purifier fabricated in this study demonstrates excellent photo-thermal conversion performance. The LCST of the purifier determines its working temperature range, as shown in the differential scanning calorimetry (DSC) curves of NV0, NV20, and NV40 purifiers in Fig. 3f. PNIPAm exhibits an LCST of approximately 34 °C, which aligns closely with the values reported in the literature,¹⁰ and the incorporation of PVA has a negligible impact on this LCST. Consequently, the LCSTs of all purifiers are relatively close to atmospheric temperature, ensuring that the purifier temperature can swiftly reach the LCST under sunlight irradiation.

2.3 Solar water generation

During the process of solar water generation, the initial phase involves the localization of solar energy at the interface through photo-thermal conversion, leading to an elevation in interfacial temperature.²² Consequently, this interfacial temperature serves as an indicator of the purifier's photo-thermal conversion capability. Fig. 4a illustrates the interfacial temperature curves of various purifiers under standard solar irradiance (1 kW m⁻²). For pure hydrogel (90 wt% PNIPAm + 10 wt% PVA), the interfacial temperature ascends gradually, culminating in a final temperature below 30 °C after 25 minutes of exposure, which is even lower than LCST. In contrast, the NV20 purifier experiences a rapid rise in interfacial temperature, reaching LCST within just 3 minutes. After 5 minutes of irradiance, the interfacial temperature attains an equilibrium value and fluctuates around 40 °C. As for the NV20-L purifier, it shares the same equilibrium temperature as the NV20 purifier; however, during the initial stages of illumination, its temperature rise rate is slower than that of the NV20 purifier, attributed to the NV20-L's more efficient heat conduction. To gain a more intuitive understanding of the heat transport within these purifiers, Fig. 4b presents a comparison of the temperature distribution on the sides of the NV20-L and NV20 purifiers at different illumination times. It is evident that the bottom temperature of the NV20-L increases significantly faster than that of the NV20 purifier. This phenomenon is due to the vertical micro-layer structure in the NV20-L, which enhances heat transfer from the surface to the bottom, thereby preventing surface heat accumulation and reducing side heat loss. This outcome proves that the anisotropic structure design is advantageous for improving thermal management in TR purifiers.

Corresponding to the temperature variations, the purifier's conformation undergoes changes due to its thermo-responsive characteristics. Video S1 in the SI showcases a complete working cycle of the NV20-L purifier under lighting conditions. Meanwhile, Fig. 4c displays the physical images of the NV20-L purifier, captured at intervals of 6 minutes. Initially, the evaporator is in a fully swollen state, filled with water. Subsequently, the upper surface rapidly turns white and shrinks, while the remaining part remains swollen due to the delayed heat transfer from the top to the bottom. Gradually, as the light continues to irradiate, the bottom of the purifier begins to shrink once the heat reaches it. Upon completion of the collapse, the purifier adopts a cylindrical shape again, but with a much smaller size compared to its initial state. After a working cycle, abundant freshwater has been produced at the bottom of the container.

To gain a more quantitative understanding of this working process, we measured the diameter of the main body (as defined in the internal diagram) as a function of exposure time, presented in Fig. 4d. The working process can be distinctly divided into four stages. In the first stage, the diameter decreases rapidly due to the swift increase in surface temperature, causing the entire purifier to shrink as the surface collapses. Once the surface contraction is complete, the diameter change becomes minimal, transitioning into the second stage. At this stage, heat is slowly transferred to the bottom, and the bottom does not start to collapse until the temperature exceeds LCST. In the third stage, the diameter begins to decrease rapidly again as heat reaches the bottom. Finally, after the bottom completes its contraction, the purifier attains its minimum dimension and remains so until the end of the working cycle.

After analyzing the working process, we recorded the mass loss of the purifiers through evaporation and guttation. Taking the NV20-L purifier as an example, Fig. 4e illustrates the mass loss as a function of time. Initially, we analyze the mass loss under 1 sun intensity. For evaporation, the mass loss exhibits a nearly linear relationship with illumination time, with a total mass loss of 1.203 kg after 30 minutes of irradiance. However, the mass loss through guttation is an order of magnitude higher than that through evaporation. In the initial stage, the water loss rate is low, even lower than that of evaporation. After 10 minutes, the mass loss rate rapidly increases and maintains a high speed until 20 minutes, gradually decreasing thereafter until reaching equilibrium. Consequently, the TR purifier integrates both evaporation and guttation functions within one system, achieving a significantly higher water yield compared to interface evaporators.

In addition to analyzing the effect of one sun, we also investigated the impact of varying light intensities on the water yield of the TR purifier and adopted two other intensities: 0.5 and 1.5 suns. An increase in solar intensity promotes phase transition, shortening the starting time to 2.5 minutes under 1.5 sun and delaying it to 12 minutes under 0.5 sun. This is attributed to the accelerated temperature rise with higher light intensity. When the purifier gradually completes guttation and the mass tends to stabilize, the total mass loss of the evaporator is consistent under different light intensities, determined by the water absorption capacity of the purifier. Additionally, the mass loss rate from evaporation increases with increasing light intensity.

The calculated water yields considering both evaporation and guttation are presented in Fig. 4f. The calculated evaporation and guttation rates of NV20-L purifier under 1 sun intensity are 2.406 and 27.584 kg m⁻² h⁻¹, respectively, so the total water yield is 29.99 kg·m⁻²·h⁻¹. Light intensity also significantly influences the water yield, with yields of 15.24 kg·m⁻²·h⁻¹ (0.5 sun) and 33.75 kg·m⁻²·h⁻¹ (1.5 sun), respectively. Meanwhile, the 10-day continuous operation shows that the water yield of the purifier fluctuates around the average value and there is no significant decline in water yield with operational time, which indicates that the fabricated TR purifier can be used in long-term freshwater production.

2.4 Literature comparison

In recent years, TR hydrogels have garnered significant attention from researchers in the realm of solar-driven water purification. In this context, we have conducted a thorough analysis of the solar water collection technologies containing TR hydrogels reported in the literature, including interfacial evaporators and TR purifiers, and the comparison of freshwater yields is presented in Table 2.

---

Interfacial evaporators primarily rely on evaporation to generate freshwater. The roles of TR hydrogels in these systems include enhancing water transport and constructing the Janus structure, as previously mentioned. While TR hydrogels can certainly improve the performance of interfacial evaporators, they are still constrained by the fundamental limitations of interfacial evaporation. Consequently, the water yields of evaporators incorporating TR hydrogels reported in the literature typically range between 1.5 and 4.5 kg m⁻² h⁻¹.^11–14

Apart from interfacial evaporators, TR purifiers possess the dual capability of producing freshwater through evaporation and guttation, thereby offering substantial advantages compared to interfacial evaporators. In 2019, Geng et al.¹⁵ anchored PNIPAm hydrogel onto a melamine foam framework, achieving an evaporation rate of 4.2 kg m⁻² h⁻¹. Xu et al.¹⁶ and Wang et al.¹⁷ directly integrated PNIPAm with other hydrogels, which resulted in enhanced guttation rates and freshwater yields of 7.18 kg m⁻² h⁻¹ and 9.80 kg m⁻² h⁻¹, respectively. Cui et al.¹⁸ employed the microgel polymerization method to synthesize P(NIPAm-co-DMAPS) gels, exhibiting exceptional swelling capability and mechanical robustness. These gels demonstrated a water-collecting rate of up to 9.32 kg m⁻² h⁻¹ under one sun irradiation. Mei et al.¹⁹ developed a spherical sandwich hydrogel structure that utilizes reversible phase transitions for water purification, achieving a freshwater yield of 28.07 kg m⁻² h⁻¹ under natural sunlight conditions. In comparison to these pioneering studies, our TR purifier exhibits superior performance, achieving an even higher freshwater yield of 29.99 kg m⁻² h⁻¹. This remarkable outcome underscores the effectiveness of the optimization strategies employed in our TR purifier, significantly enhancing its capacity to produce freshwater.

2.5 Desalination and wastewater purification

In practical applications, a purifier's ability to desalinate and purify wastewater is paramount. Therefore, we have thoroughly evaluated the salt ion separation and wastewater purification capabilities of our purifier.

The desalination process of our purifier primarily hinges on the PA membrane. Fig. 5a showcases the physical appearance of the prepared PA membrane attached to the substrate, which boasts excellent flexibility and can endure substantial bending. The SEM image in Fig. 5b demonstrates that the freshly made PA membrane presents a highly intact and dense layered structure with a thickness of 150 nm, which can effectively block salt ions from penetrating. Fig. 5c presents the physical images of the purifier covered with a PA membrane, in both its swollen and collapsed states. Upon completing a working cycle, the PA membrane adheres tightly to the purifier and remains intact, indicating a robust binding force with the hydrogel surface. In addition, it is obvious that the contraction amplitude of the purifier in seawater after phase transition is greater than in pure water due to the salting-out effect.^23,24

During water absorption in seawater, the membrane flow resistance of the PA membrane is overcome by the osmotic pressure generated by the hydrogel. To evaluate the water absorption ability of the purifier, we systematically characterized the osmotic pressures of both brine solutions and the hydrogel (Fig. 5d). Our engineered NV20-L purifier demonstrates exceptional performance with an osmotic pressure of 12.1 atm, significantly exceeding that of concentrated saline solutions. Hydraulic performance evaluation (Fig. 5e) reveals that under a 9 atm pressure differential in 3.0 wt% NaCl solution, the membrane achieves an impressive water flux of 60.1 kg m⁻² h⁻¹ while maintaining effective salt rejection, ensuring rapid water transport in our evaporation system. Comprehensive salt rejection analysis was conducted using artificial brine solutions following the protocol in ref. 25. As shown in Fig. 5f, the NV20-LM purifier exhibits outstanding desalination efficiency with ion-specific retention rates of 76.3% (NaCl), 81.9% (KCl), 89.5% (MgCl₂), and 91.2% (CaCl₂) at 0.3 wt% total salinity. Notably, retention performance shows moderate concentration dependence, decreasing to 73.4%, 79.2%, 87.7%, and 88.3% respectively when salinity increases to 3.0 wt%. This concentration-dependent behavior aligns with typical PA membrane characteristics where increased ionic strength reduces Donnan exclusion efficiency.²⁶ Long-term stability tests demonstrate remarkable durability, with less than 2.0% average reduction in salt rejection after 50 operational cycles. Meanwhile, the SEM image of the PA membrane after 50 operational cycles also shows good integrity (Fig. 5b). This robust performance persists despite recurring mechanical stress from repeated hydration–dehydration transitions. Collectively, these findings validate our engineered membrane system as achieving an optimal balance between high-throughput desalination capacity and long-term operational reliability in seawater.

In addition to desalination, the water purification capabilities of the TR purifier were rigorously tested using water sourced from the Xiangjiang River in Changsha, China. A compact cuboid purifier, measuring 12 × 10 × 1 cm³, was deployed and floated on the river's surface (Fig. 5g). The ambient river temperature was recorded at approximately 25 °C, conducive for the purifier to efficiently absorb water. Upon exposure to sunlight, the purifier's surface temperature escalated to 40 °C, resulting in the production of freshwater (Fig. 5h). Microscopic images comparing the river water and the purified water are depicted in the lower right corner of Fig. 5g and h. The lake water revealed the presence of numerous microorganisms and pollutants, whereas the treated water was devoid of any microscopic particles. To further assess the treatment quality, resistance measurements were conducted (Fig. 5i). The resistance values measured for river water, treated water, and domestic water amounted to 0.17 MΩ, 0.84 MΩ, and 0.70 MΩ, respectively. These results underscore the exceptional purification performance of our purifier. The comprehensive evaluation of desalination and outdoor wastewater treatment demonstrates that our TR purifier exhibits outstanding performance under natural light conditions, hinting at its promising practical applications in sustainable freshwater production.

3 Conclusions

This paper introduces a mangrove-inspired TR purifier that is innovatively constructed using PNIPAm and PVA composite hydrogels, GE filler, and PA membrane. This unique purifier demonstrates remarkable performance not only in terms of freshwater production but also in its efficiency for salt removal. By cleverly incorporating anisotropic micro-layers and GE filler into its design, the directional thermal conductivity of the purifier has been substantially improved. This enhancement facilitates rapid and efficient heat transfer throughout the entire system, ensuring optimal performance under varying conditions. Furthermore, the hydrogels used in this purifier have undergone meticulous component adjustment to finely tune their water absorption and release properties. This precision allows the hydrogels to maintain optimal performance in terms of water retention and release, further enhancing the overall efficiency of the purifier. The high osmotic pressure exhibited by these hydrogels serves as a powerful driving force for the PA membrane, enabling it to effectively separate water molecules and salt ions from seawater. As a result of these innovations, the purifier achieves an ultra-high freshwater output of 29.99 kg·h⁻¹·m⁻² under one sun. This performance surpasses previously reported solar-powered water purification systems, demonstrating the superior efficiency and practicality of the proposed design. Additionally, the fabricated purifier demonstrates exceptional salt rejection and durability, ensuring high-quality freshwater production. Moreover, the purifier possesses outstanding mechanical strength, capable of enduring harsh environmental conditions without compromising its performance. The findings of this study not only underscore the substantial potential of TR hydrogels for practical use in solar desalination applications but also pave the way for future innovations in this field.

4 Materials and methods

4.1 Materials

The monomer NIPAm, potassium persulfate (KPS, purity > 99%), N,N′-methylenebisacrylamide (BIS, electrophoretic-grade), tetramethylethylenediamine (TEMED, electrophoretic grade), GE and its dispersant SDS, and n-hexane (purity = 99.5%) were supplied by Aladdin Co. Ltd. PVA-1799, m-phenylenediamine (MPD, purity = 99%), and 1,3,5-phenyltricarboxylic acid chloride (TMC, purity = 99%) were purchased from Macklin Co. Ltd. High-quality deionized (DI) water was generated using a UPTC-10L deionized water machine.

4.2 Preparation of the TR purifier

The entire process of evaporator fabrication can be divided into two distinct stages: the fabrication of hydrogel and PA membrane adhesion. The fabrication processes are elaborated by delving into these two stages individually. For illustrative purposes, we employ an NV20-ML purifier with a layered structure and PA membrane as an exemplar.

4.3 Characterization

Purifier component morphologies are inspected using a HITACHI SU8220 SEM. Elemental composition is analyzed by EDS. Evaporator stress–strain curves are assessed using a Hongjin LL-05 machine. Water contact angles are measured with a Biolin Theta Flex meter. FTIR analysis is performed using a Thermo Nicolet iSS spectrometer. Thermal conductivity is evaluated with a Xiangyi DRL-III. Hydrogel DSC profiles are measured using a TA Q2000 DSC instrument. The optical properties of samples are characterized using a Youke T2600 UV-vis spectrophotometer. PA membrane water flux under varying pressures is assessed with a TQFM100-P10 from Shanghai Tongqin.

4.4 Freshwater yield evaluation

A photo-thermal conversion system evaluates the freshwater output of the TR purifier. Before the experiment, the purifier is submerged in the solution to reach full swelling. It's then placed under a solar simulator (Aulight CEL-SE300) with intensity measured and adjusted using a solar irradiance meter (Aulight CEL-FZ-A). The lab environment is maintained at 25 °C and 50% humidity. Mass and temperature variations are documented using a digital balance (Medele Toledo ME104) and a thermometer (Hikvision H21PRO). Water yield (v) is calculated using eqn (1):²⁷

	(1)

where S is the interface area of the purifier, m is the mass lost through both evaporation and guttation, and t is the irradiation duration.

4.5 MD simulation details

We crafted 30 Å PNIPAm and PNIPAm + PVA polymer chains using Materials Studio, then grafted them onto GE walls. Our model had two GE walls 180 Å apart, with polymers evenly spaced 2 Å from the walls. A polymer chain was anchored near the wall for binding. The simulation was analyzed with LAMMPS and visualized with OVITO. A 5 × 5 × 20 nm³ 3D system was established under periodic boundary conditions to explore polymer dynamics. It included 10 PNIPAm, 6 PVA, and another 10 PNIPAm chains, with water molecules filling the space. CVFF was used for polymers, and a rigid model for water. Interactions were described by Lennard–Jones 12–6 potential with arithmetic combination rules for heteroatoms.²⁸ PPPM handled distant electrostatic interactions with RMS accuracy of 10–4.²⁹ Only polymers and water were allowed to move, with wall chains assumed rigid. Water molecules were constrained using SHAKE.³⁰ After preparation, 1000 water molecules were placed in the system. Systems were simulated at 293 K and 333 K. NVT ensemble was used to attain equilibrium within 7 ns, followed by 12 ns data collection. Density distribution along the z-axis at varying temperatures analyzed the component distribution in the hydration system.

4.6 Osmosis pressure measurement

The osmotic pressure of the hydrogel was determined through systematic evaluation of its swelling behavior in NaCl solutions with varying concentrations.³¹ This method is based on the principle that when the hydrogel reaches swelling equilibrium in a specific NaCl solution – indicated by the cessation of both water absorption and volumetric expansion – the external solution's osmotic pressure equilibrates with the hydrogel's intrinsic osmotic pressure. The critical osmotic pressure corresponding to this equilibrium state was quantified using the Van't Hoff equation:³²

π = iCRT	(2)

where π represents the osmotic pressure (Pa), i denotes the van't Hoff factor (i = 2 for NaCl due to its complete dissociation), C indicates the molar concentration of NaCl (mol m⁻³), R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T refers to the absolute temperature (K) during measurement. This equilibrium method provides a reliable thermodynamic approach for characterizing the hydrogel's osmotic properties through its interaction with ionic solutions.

Author contributions

Yang Xiao: writing – original draft, software, investigation, formal analysis, data curation, conceptualization. Yanqiong Bao: data curation. Lang Liu: data curation. Wanlin Xu: software. Da Li: data curation. Xiong Zheng: writing – review & editing, writing – original draft, supervision, project administration, funding acquisition, conceptualization. Guangzhao Qin: writing – review & editing. Qing Li: writing – review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.

Video S1 working process of TR purifier under irradiance. See DOI: https://doi.org/10.1039/d5ta04916d.

Acknowledgements

We acknowledge the support of the National Natural Science Foundation of China (No. 52006059).

References

1. L. Wu, Z. Dong, Z. Cai, T. Ganapathy, N. X. Fang, C. Li, C. Yu, Y. Zhang and Y. Song, Nat. Commun., 2020, 11(1), 521 CrossRef CAS PubMed .
2. B. Liu, Y. Yang, Q. Zhao, Y. Liu, Y. Shen, Y. Ma and W. Que, Energy Environ. Sci., 2025, 18, 6366–6437 RSC .
3. Y. Yang, H. Feng, W. Que, Y. Qiu, Y. Li, L. Guo and Q. Li, Adv. Funct. Mater., 2023, 33(6), 2210972 CrossRef CAS .
4. M. Sheng, Y. Yang, X. Bin, S. Zhao, C. Pan, F. Nawaz and W. Que, Nano Energy, 2021, 89, 106468 CrossRef CAS .
5. Y. Xiao, B. Liu, D. Li, X. Zheng, J. Li and G. Qin, Chem. Eng. J., 2024, 497, 155038 CrossRef CAS .
6. H. Ghasemi, G. Ni, A. M. Marconnet, J. Loomis, S. Yerci, N. Miljkovic and G. Chen, Nat. Commun., 2014, 5(1), 1–7 Search PubMed .
7. Z. Y. Xu, L. N. Zhang, L. Zhao, B. J. Li, B. Bhati, C. X. Wang, K. L. Wilke, Y. Song, O. Labban, J. H. Lienhard, R. Z. Wang and E. N. Wang, Energy Environ. Sci., 2020, 13(3), 830–839 RSC .
8. Y. Yang, H. Zhao, Z. Yin, J. Zhao, X. Yin, N. Li, D. Yin, Y. Li, B. Lei, Y. Du and W. Que, Mater. Horiz., 2018, 5(6), 1143–1150 RSC .
9. P. Tao, G. Ni, C. Song, W. Shang, J. Wu, J. Zhu, G. Chen and T. Deng, Nat. Energy, 2018, 3(12), 1031–1041 CrossRef .
10. J. Liu, L. Jiang, S. He, J. Zhang and W. Shao, Chem. Eng. J., 2022, 433, 133496 CrossRef CAS .
11. S. Cao, J. Jiang, Q. Tian, C. Guo, X. Wang, K. Dai and Q. Xu, Green Energy Environ., 2022, 7, 1006–1013 CrossRef CAS .
12. H. W. Lim, S. H. Park and S. J. Lee, Desalination, 2023, 550, 116368 CrossRef CAS .
13. Y. Liu and X. Zheng, Sol. Energy, 2024, 273, 112507 CrossRef CAS .
14. H. Li, G. Tong, A. Chu, J. Chen, H. Yang, J. Fang, Z. Yang, H. Liu and L. Dong, Nano Energy, 2024, 124, 109475 CrossRef CAS .
15. H. Geng, Q. Xu, M. Wu, H. Ma, P. Zhang, T. Gao, L. Qu, T. Ma and C. Li, Nat. Commun., 2019, 10(1), 1512 CrossRef PubMed .
16. X. Xu, S. Ozden, N. Bizmark, C. B. Arnold, S. S. Datta and R. D. Priestley, Adv. Mater., 2021, 33, 2007833 CrossRef CAS PubMed .
17. M. Wang, Z. Sun, J. Zhou, Y. Chen, X. Mu, Z. Zhang, X. Wang, J. Liu, J. Gao, W. Liu and L. Miao, Sol. RRL, 2023, 7, 2300041 CrossRef CAS .
18. L. Cui, F. Wu, W. Wang, C. Hu, J. Zheng, Z. Zhu and B. Liu, ACS Appl. Polym. Mater., 2024, 6, 3253–3262 CrossRef CAS .
19. J. Mei, Y. Jin, H. Chen, R. Zhou, Z. Mao and J. Wang, Chem. Eng. J., 2024, 498, 155505 CrossRef CAS .
20. L. Tavagnacco, E. Zaccarelli and E. Chiessi, Phys. Chem. Chem. Phys., 2018, 20, 9997–10010 RSC .
21. V. N. Kotov, B. Uchoa, V. M. Pereira, F. Guinea and A. H. Castro Neto, Rev. Mod. Phys., 2012, 84(3), 1067–1125 CrossRef CAS .
22. H. Shan, P. Poredoš, Z. Ye, H. Qu, Y. Zhang, M. Zhou, R. Wang and S. C. Tan, Adv. Mater., 2023, 35(35), 2302038 CrossRef CAS PubMed .
23. H. Shan, Z. Ye, J. Yu, R. Wang and Z. Xu, Device, 2023, 1(4), 100065 CrossRef .
24. P. Poredoš, J. Gao, H. Shan, J. Yu, Z. Shao, Z. Xu and R. Wang, Nat. Commun., 2024, 15(1), 7890 CrossRef PubMed .
25. N. He, Y. F. Yang, H. N. Wang, F. Li, B. Jiang, D. W. Tang and L. Li, Adv. Mater., 2023, 35, 2300189 CrossRef CAS PubMed .
26. J. Schaep, B. Van der Bruggen, C. Vandecasteele and D. Wilms, Sep. Purif. Technol., 1998, 14(1–3), 155–162 CrossRef CAS .
27. F. Zhao, X. Zhou, Y. Shi, X. Qian, M. Alexander, X. Zhao, S. Mendez, R. Yang, L. Qu and G. Yu, Nat. Nanotechnol., 2018, 13(6), 489–495 CrossRef CAS PubMed .
28. M. Alaghemandi and E. Spohr, RSC Adv., 2013, 3(11), 3638–3647 RSC .
29. R. W. Hockney and S. P. Goel, J. Comput. Phys., 1974, 14(2), 148–158 CrossRef .
30. M. Alaghemandi, F. Leroy, F. Müller-Plathe and M. C. Böhm, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81(12), 125410 CrossRef .
31. B. Wang, H. Zhang, N. He, H. Wang, B. Jiang, D. Tang and L. Li, Sep. Purif. Technol., 2023, 314, 123490 CrossRef CAS .
32. P. Atkins and J. de Paula, Physical Chemistry: Thermodynamics, Structure, and Change, 12th edn, Oxford University Press, 2022 Search PubMed .

---

Footnote

† These authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2025