The rapid population growth, worsening pollution, and increasingly severe weather have exacerbated the global water and energy crises. Currently, the shortage of freshwater resources worldwide poses a severe threat to the stability of ecosystems and human health [1–3]. Many people, especially those in remote and rural areas, lack the infrastructure needed to consistently access safe drinking water. Consequently, developing cost-effective wastewater treatment technologies has been a focus of research. Utilizing renewable solar energy through interfacial solar steam evaporation technology has shown great potential for seawater desalination [4–7] and wastewater treatment [8–11]. Although solar photothermal technology offers promising solutions for wastewater issues, traditional evaporators struggle to efficiently treat wastewater. To address this, researchers have combined photothermal evaporators with photocatalysts to achieve both efficient water evaporation and wastewater treatment. For example, Li et al. [12] developed an Ag/MXene membrane that leverages photothermal and photocatalytic synergy to degrade heavy metal ions and organic pollutants during wastewater treatment; Miao et al. [13] designed a metal-organic frameworks (MOFs)-derived C/TiO₂ composite material that simultaneously performs photothermal and photocatalytic functions, enabling solar-driven freshwater production and degradation of organic pollutants in wastewater. As research progresses, photothermal-photocatalytic materials have been demonstrated to degrade volatile organic compounds (VOCs) in distilled water [14]. Additionally, solar-driven photocatalytic evaporators offer the potential for multifunctional applications, achieving dual [15] or even triple benefits [16].

Owing to their three-dimensional network structures with activated water states, excellent mechanical/optical/electronic properties, and multi-scale tunability, hydrogels have gained widespread applications in various fields such as bioelectronics [17–19], biomedical engineering [20], energy conversion [21], strain sensors [22], electrochromics [23], soft robotics [24,25], etc. Particularly, various hydrogel-based evaporators have also been developed for water purification in recent years [6,8,20,21]. For example, Wang et al. [14] developed an MXene/CdS photothermal-photocatalytic hydrogel for efficient solar water evaporation and VOC degradation. Zhao et al. [26] designed a galactomannan/graphene oxide/Fe₃O₄ hydrogel evaporator for solar-driven water evaporation and simultaneous photothermal power generation. Zheng et al. [27] created a magnetically controlled, wind-resistant Janus biomass composite hydrogel with desalination capabilities for the purification of polluted seawater. Lyu et al. [28] developed a simple, low-cost macroporous hydrogel for high-performance atmospheric water harvesting (AWH). Therefore, hydrogels integrating photothermal and photocatalytic effects hold significant potential in this field.

TiO₂, due to its low cost and environmental friendliness, is a potential photocatalyst. However, it can only absorb ultraviolet light, and the rapid recombination of photogenerated carriers significantly reduces its photocatalytic activity [29–31]. To overcome these drawbacks, constructing heterojunctions is an ideal strategy to mitigate the rapid recombination of photogenerated carriers. Since g-C₃N₄ exhibits good visible light absorption, combining TiO₂ with g-C₃N₄ can effectively reduce photogenerated carrier recombination [32–35]. There are relevant studies on this topic, such as the work by Yu et al. [36], who developed an advanced 2D/3D g-C₃N₄/TiO₂@MnO₂ multifunctional separation membrane for light-driven sustainable water purification. Yang et al. [37] designed a hollow octahedral MOF-derived TiO₂ structure combined with ultrathin porous g-C₃N₄, enhancing the degradation efficiency of brewing wastewater.

Given the above analysis, this paper demonstrates that TiO₂/g-C₃N₄ loaded on a hydrogel matrix can achieve efficient solar-driven evaporation and organic pollutant degradation. By incorporating SWCNTs with broad-spectrum absorption and TiO₂/g-C₃N₄ photocatalytic heterojunction, a hydrogel with dual functionalities of photocatalytic degradation and interfacial evaporation is developed. The prepared TiO₂/g-C₃N₄/SWCNT composite hydrogel exhibited excellent light absorption across the full spectrum (approximately 98.9%), efficient photothermal conversion (35.9°C), and a water evaporation rate of 1.43 kg m⁻² h⁻¹. Notably, the hydrogel exhibited a lower evaporation enthalpy than pure water (as low as 1144.7 J g⁻¹). The photocatalytic performance of the hydrogel enabled over 90% degradation of pollutants within 120 min, with almost no decline in photocatalytic performance after five cycles. The primary active species for methylene blue (MB) photodegradation are ⋅O₂⁻ and ⋅OH⁻.

Materials: N,N’-Methylenebis (acrylamide) (MBAA, 99%), ammonium persulfate (APS, ≥98%), acrylamide (AM, 99.0%), melamine (99%), and titanium oxide (TiO₂, 98%) were obtained from Aladdin (Shanghai, China). N,N,N’,N’-Tetramethylethylenediamine (TEMED, 99%) was obtained from J&K Scientific (Beijing, China). Single-walled carbon nanotube (SWCNT) aqueous dispersion (9–10 wt.%) was obtained from XFNano (Nanjing, China). All other reagents were analytical grade and directly employed without further purification.

Preparation of graphitic carbon nitride (g-C₃N₄): g-C₃N₄ was prepared by direct thermal polymerization. Specifically, 5 g of melamine was placed into a crucible, covered with a lid, and heated in a muffle furnace at a rate of 15°C/min to 550°C. The temperature was maintained for 4 h, and then the sample was cooled to room temperature. After cooling, the product was removed and ground into a powder in a mortar for future use.

Preparation of single catalyst hydrogel: First, 0.3 g of TiO₂ nanoparticle/g-C₃N₄ powder was added to 10 mL of deionized water, stirred thoroughly, and then sonicated for 30 min to form a homogeneous aqueous solution. Subsequently, 0.027 g of a diluted aqueous dispersion of single-walled carbon nanotubes (4.5–5 wt.%) was added and stirred well to form dispersion A. Then, 1.1 g of AM monomer was added to dispersion A, stirred until dissolved, and 100 μL of APS (10 wt.%) initiator solution and 0.0125 g of MBAA crosslinker were added to form solution B. The stirred dispersion of solution B was then sonicated for 10 min to form a homogeneous dispersion. Afterward, 50 μL of TEMED catalyst was added while stirring, and after stirring for 1 min, the mixture was transferred to a mold for gelation. After the gelation was completed at room temperature, the hydrogel was demolded and soaked in deionized water for more than 48 h to remove residual substances. It is noteworthy that the TiO₂/SWCNT and g-C₃N₄/SWCNT hydrogels corresponded to TSH and CSH, respectively.

Preparation of composite catalyst hydrogel: First, 0.3 g of TiO₂ nanoparticle and 0.3 g of g-C₃N₄ powder were added to 10 mL of deionized water, stirred thoroughly, and then sonicated for 30 min to form a homogeneous aqueous solution. Subsequently, 0.027 g of a diluted aqueous dispersion of single-walled carbon nanotubes (4.5–5 wt.%) was added and stirred well to form dispersion A. Then, 1.1 g of AM monomer was added to dispersion A, stirred until dissolved, and 100 μL of APS (10 wt.%) initiator solution and 0.0125 g of MBAA crosslinker were added to form solution B. The stirred dispersion of solution B was then sonicated for 10 min to form a homogeneous dispersion. Afterward, 50 μL of TEMED catalyst was added while stirring, and after stirring for 1 min, the mixture was transferred to a mold for gelation. After the gelation was completed at room temperature, the hydrogel was demolded and soaked in deionized water for more than 48 h to remove residual substances. It is noteworthy that in the TiO₂/g-C₃N₄/SWCNT composite hydrogel, the weight ratios of TiO₂ to g-C₃N₄ were 1:1, 1:2, and 2:1, corresponding to TCSH-1, TCSH-2, and TCSH-3, respectively.

Characterizations: The surface morphology and chemical element distribution of TSH, CSH, and TCSH-1 hydrogels were characterized using using a scanning electron microscope (SEM, SU8600, Hitachi, Tokyo, Japan) equipped with an X-ray energy dispersive spectrometer (EDS, ELECT SUPER(C5), EDAX, Commonwealth of Pennsylvania, US). The TiO₂ and g-C₃N₄ powders were analyzed by X-ray photoelectron spectroscopy using an X-ray diffractometer (XRD, Empyrean, MalvernPanalytical, Malvern, UK). The diffuse reflectance properties of the prepared samples were measured using an ultraviolet-visible-near-infrared spectrophotometer (TP720, Tianjin Toupu Instrument Co., Tianjin, China) with an integrating sphere unit and an automatic reflectance measurement unit, and the obtained spectra were corrected for baseline/blank and black. The absorbance (A) was obtained through the Eq. (1): (1) A=1−T−Rwhere R and T are the reflectance and transmittance of the hydrogel, respectively.

Photocatalytic performance measurements: The photocatalytic performance of the hydrogels was evaluated using methylene blue (MB) at a concentration of 10 g L⁻¹ as a model. The light intensity was maintained at one sun (1 kW m⁻²), with a solar simulator as the light source (CEL-HXUV300-T3, Education Au-light Co., Beijing, China). The prepared 3D columnar hydrogels were cut into small pieces and placed in the methylene blue solution before irradiation, kept in the dark for 60 min to achieve adsorption-desorption equilibrium. Subsequently, samples (3 mL) were taken every 15 min and measured using a UV-Vis-NIR spectrophotometer (TP720, Tianjin Toupu Instrument Co., Tianjin, China). Moreover, the degradation efficiency (D_e) was calculated based on the Eq. (2) [38,39]: (2) De=1−C/C0where C0 denotes the initial concentration, and C is the concentration of the solution in different periods. We further evaluate the photocatalytic efficiency based on the pseudo-first-order model according to the Eq. (3) [40]: (3) −ln⁡(C/C0)=ktwhere k is related to the constant photodegradation rate.

Solar steam generation performance measurements: The solar steam generation performance of the hydrogels was tested using a solar simulator with an AM1.5 filter (CEL-HXUV300-T3, Education Au-light Co., Beijing, China). The solar flux was measured by an automatic optical power meter (CEL-NP2000-2A, Education Au-light Co., Beijing, China). The mass change during the evaporation process was monitored in real-time using an electronic analytical balance (Sartorius BAS223, accurate to 1 mg). The surface temperature and infrared thermograms of the hydrogels were monitored in real-time using an infrared thermal imager (HM-TPH21Pr0-3AQF, HIKMICRO, Hangzhou, China). During the tests, the ambient humidity and temperature were ~60% and ~25°C, respectively. Additionally, the evaporation rate (ER) and equivalent evaporation enthalpy (Eequ) were calculated based on the Eqs. (4) and (5) [41]: (4) ER=dm/(S∗dt)where dm denotes to the mass change by using hydrogel as the evaporator for 1 h, S is the evaporation area of hydrogel, and dt is evaporation time. (5) Eequ=E0m0mgwhere E0 and m0 are the evaporation enthalpy and mass change of bulk water; mg is the mass change of hydrogel.

Conductivity and salinity measurements: In the conductivity range of 0.001 to 0.01 wt.‰, the salinity of NaCl solutions exhibits a strong linear correlation with electrical conductivity. A standard curve was established by testing the conductivity of different brines with a conductivity controller (A10CD-AA1, Shanghai Kuosi Electronics Co., Shanghai, China) and used to calculate the salinity of pure water evaporated from simulated seawater by TCSH-1. The conductivity of at least three parallel samples is measured and averaged to accurately determine the salinity.

As shown in Fig. 1a, the hydrogel prepared in the experiment has a three-dimensional columnar structure, designed to increase the evaporation area and thereby enhance the evaporation rate of the hydrogel. TiO₂ nanoparticles and g-C₃N₄ nanosheet dual catalysts are integrated into the hydrogel matrix. The characteristic macroporous structure of the hydrogel not only enables multi-level refraction and diffraction of light within the gel network, thereby enhancing light absorption, but also allows for rapid water absorption and transport (Fig. 1b). In addition, this design specifically addresses the drawback of TiO₂, which can only absorb ultraviolet light, and its reduced photocatalytic efficiency due to rapid photogenerated carrier recombination. Since g-C₃N₄ has well visible light absorption, the heterojunction constructed by the dual catalysts can improve the efficiency of the photocatalyst (Fig. 1c). The incorporation of photocatalytic functionality not only enables efficient sewage purification but also prevents the contamination of bulk water during the evaporation process. Such hydrogels hold promise for multifunctional applications in solar-powered water purification systems.

TiO₂ and g-C₃N₄, as typical photocatalysts, are now widely applied. In this study, g-C₃N₄ powder is synthesized by a direct thermal polymerization method, resulting in a pale yellow color. The microstructure and crystal structure of the two solid catalysts are further explored through Scanning Electron Microscopy (SEM) and X-ray Diffraction (XRD). As shown in Fig. S1, the TiO₂ and g-C₃N₄ solids exhibit different morphological structures at 2 μm. The purchased TiO₂ appears as nanoparticles under microscopic observation, while the g-C₃N₄ obtained by thermal polymerization exhibits a sheet-like structure. Further crystal structure analysis is performed using XRD. As shown in Fig. S2, the diffraction peaks at 2θ = 13.5° and 27.4° correspond to the (100) and (002) crystal planes of the g-C₃N₄ standard sample card (JCPDS 87-1526), respectively. The diffraction peak at 13.5° is attributed to the tris-s-triazine rings, and the sharp diffraction peak at 27.4° is caused by the stacking of aromatic heterocycles. This indicates that g-C₃N₄ is successfully prepared by the direct thermal polymerization method and that the prepared g-C₃N₄ possesses good purity and crystallinity. Subsequently, the samples of TiO₂/g-C₃N₄ loaded on the hydrogel matrix are analyzed using SEM and Energy Dispersive X-ray Spectroscopy (EDS). The results show that the photocatalyst is successfully loaded in the hydrogel, displaying the characteristic porous structure of the hydrogel (Fig. 2).

Excellent light absorption across the full spectrum is essential for achieving efficient photothermal conversion and solar-driven evaporation performance in hydrogels. As shown in Fig. 3a, the prepared single and composite photocatalytic hydrogels exhibit superior light absorption across the entire spectrum (with an absorption rate greater than 98.8%), with negligible light loss. Notably, the light absorption rate of TSH reaches 99.9%, outperforming CSH and TCSH-1. The primary reason for this is that the reflectivity of the TCSH-1 hydrogel across the full spectrum is higher than that of the other two single-catalyst hydrogels (Fig. 3b). Furthermore, we investigate the wetting properties of TSH, CSH, and TCSH-1 through water contact angle measurements. The contact angle of the TCSH-1 hydrogel rapidly decreases to 12.87°, facilitating the dynamic migration of water molecules to the evaporator surface (Fig. S3). As depicted in Fig. 3d,e, the water evaporation performance of TSH, CSH, and TCSH-1 is evaluated by recording the mass change of water under 1 kW m⁻² solar irradiation over 60 min. As the irradiation time increases, the mass change of water in each hydrogel exhibits a linear increase, with the CSH sample showing the highest evaporation rate, approximately 1.63 kg m⁻² h⁻¹. The water evaporation rate of the prepared TCSH-1 shows a decreasing trend with the increase in the number of photocatalysts in the hydrogel components. Furthermore, the different evaporation rates of the hydrogels are verified by testing the equivalent enthalpy of evaporation. Specifically, the evaporation rate is recorded by placing the hydrogels under dark conditions, and the corresponding equivalent enthalpy of evaporation is calculated using Eq. (5). The results indicate that the obtained enthalpy of evaporation correlates well with the hydrogel evaporation rate results, with CSH having the lowest equivalent enthalpy of evaporation, at 1144.7 J g⁻¹ (Fig. 3c). Notably, the equivalent enthalpy of evaporation of the prepared hydrogels is significantly lower than that of pure water (2450 J g⁻¹). The photothermal conversion efficiency of these hydrogels is further demonstrated by monitoring the interface temperature changes of the hydrogels under illumination using infrared imaging technology. Remarkably, the temperature of the hydrogels rapidly rises above 30°C within the first 10 min of light exposure, followed by a slow increase to 35.9°C. The rapid decrease in surface temperature in the absence of light further confirms the high photothermal conversion capability of the hydrogels (Fig. 3f,g).

The long-term stability of materials is essential for ensuring the sustainable and efficient operation of water purification processes. To evaluate the stability of the TCSH-1 hydrogel, it is subjected to a cyclic test in 3.5 wt.% simulated seawater for seven consecutive cycles (Fig. 4a). After repeated swelling-evaporation processes, the evaporation performance of the hydrogel remains consistently at 2.03 kg m⁻² h⁻¹. Furthermore, the evaporation performance of the TCSH-1 hydrogel evaporator is tested under outdoor conditions to assess its practical performance in a real-world environment. As shown in Fig. S4, a fully automatic optical power meter records the solar power density in Nanchang from 9:00 a.m. to 16:00 p.m. on 23rd September. The weather on that day is cloudy to overcast, with a light breeze (wind speed 3–4 level), and the hourly evaporation rates are displayed in Fig. 4b.

Next, the performance of the hydrogel in seawater desalination and wastewater purification is evaluated. Three different simulated seawater samples—World Sea (average salinity 35 wt.‰), Bohai Sea (moderate salinity, 70 wt.‰), and Dead Sea (highest salinity, 200 wt.‰)—are purified. After purification, the salinity of these samples decreases by 3 to 4 orders of magnitude, as shown in Figs. 4c and S5. Additionally, the purified water meets the drinking water standards of the World Health Organization (WHO, 1 wt.‰) and the Environmental Protection Agency (EPA, 0.5 wt.‰). Inductively coupled plasma mass spectrometry (ICP-MS) is used to measure the concentrations of salt ions, including Na⁺, Mg²⁺, K⁺, and Ca²⁺, in the evaporated water. As depicted in Fig. 4d, the concentrations of major salt ions decrease by more than three orders of magnitude after purification, demonstrating the high desalination efficiency of TCSH-1.

The photocatalytic efficiency of the prepared hydrogel is evaluated through the photodegradation test of a typical organic pollutant, methylene blue (MB). First, the hydrogel is placed under dark conditions for 60 min to achieve adsorption-desorption equilibrium, followed by a 120-min photocatalytic degradation test of MB. The results indicate that TCSH-1 exhibits the highest degradation efficiency (Fig. 5a). As shown in Fig. 5b, kinetic analysis reveals that the rate constants (k values) for TSH, CSH, TCSH-1, TCSH-2, and TCSH-3 hydrogels are 0.0104, 0.01293, 0.02039, 0.01664, and 0.01311 min⁻¹, respectively. As shown in Table S1, the TCSH-1 hydrogel evaporator exhibits better photothermal evaporation efficiency [42–44]. More notably, TCSH-1 hydrogel is much better than or comparable to those of reported hydrogel based catalysts [45–47]. Therefore, TSH, CSH, and TCSH-1 are selected for subsequent experiments. As shown in Fig. 5c, the UV-visible absorption spectra indicate that the absorption peak of MB at 664 nm significantly decreases over time, suggesting the gradual degradation of MB. As can be seen from Fig. 5d, the catalytic performance of the photocatalyst slightly decreases after five cycles (10 h). Moreover, the SEM morphology of TCSH-1 remains nearly unchanged after five cycles (Fig. S6), indicating that the synthesized TCSH-1 exhibits excellent stability.

The mechanism of photodegradation MB for TCSH-1 is further studied in Fig. 6. Radicals and holes trapping experiments are adopted by adding p-benzoquinone (BQ), disodium ethylenediaminetetraacetate (EDTA-2Na) and isopropyl alcohol (IPA), which using as ⋅h⁺, ⋅O₂⁻ and ⋅OH⁻, respectively (Fig. 6a). As presented in Fig. 6b, a significant effect is observed after adding BQ and IPA for photodegradation activity, implying ⋅O₂⁻ and ⋅OH⁻ are dominant active species in photodegradation MB. The optical properties of TSH, CSH, and TCSH-1 are investigated. As seen in Fig. 6c, CSH and TCSH-1 display the absorption edge in visible light. In order to obtain Eg of TiO₂ and g-C₃N₄, their UV-vis diffusion reflection spectra and Tacu plots are provided in Fig. 6d, and the bandgap of the semiconductor catalysts are obtained according to the Eq. (6) [48,49]:

(6) (αhV)n=A(hV−Eg)

The Eg of TiO₂ and g-C₃N₄ were 3.1 and 2.74 eV by the straight line to the X axis intercept, respectively.

To better understand the mechanism of this photodegradation MB process, the conduction and valence band positions of TiO₂ and g-C₃N₄ are estimated according to reported literature [50]. The process of charge separation and transfer in TiO₂/g-C₃N₄/SWCNT hydrogel can be illustrated in Scheme S1 according to the above-mentioned results. The electron-hole pairs are generated and the electrons transferred from the VB to the CB when the TiO₂/g-C₃N₄/SWCNT hydrogel absorbed visible light. Because the ·O₂⁻ and ·OH⁻ as the primary active species for MB degradation, a possible Z type heterojunction between TiO₂ and g-C₃N₄ is proposed, the photogenerated electrons of TiO₂ can be migrated to VB of g-C₃N₄, the electrons in CB of g-C₃N₄ is combined the O₂ to produce ·O₂⁻ and the hole in VB of TiO₂ is combined the H₂O to produce ·OH⁻ to degrade MB. Thus, effective separation of the photogenerated electron-hole pairs are achieved.

In summary, TiO₂/g-C₃N₄/SWCNT hydrogel with solar water evaporation and photocatalytic activity is successfully synthesized. The obtained TiO₂/g-C₃N₄/SWCNT hydrogel show well water evaporation channel, photothermal conversion, and charge separation efficiency. Due to these merits, the TiO₂/g-C₃N₄/SWCNT hydrogel demonstrates good water evaporation efficiency and photodegradation performance for MB. Furthermore, the radical trapping experiment is conducted to further understand the mechanism of photodegradation, and the result confirms that the ⋅O₂⁻ and ⋅OH⁻ are the main active species for MB degradation. This study is helpful for designing and developing efficient multi-functional photothermal materials.