Synergistic MOF-based composite enabling significant solar-to-water generation enhancement in climate-resilient AWH

Abstract

Solar-driven atmospheric water harvesting (SAWH) holds significant promise for decentralized water supply. However, its widespread application is hindered by two critical limitations: underutilization of high-humidity adsorption windows during nighttime and insufficient desorption during daytime due to the high desorption temperature requirement of conventional sorbents. To overcome these challenges, this study proposes a composite sorbent strategy by synergistically combining the low enthalpy of vaporization of LiCl with the robust adsorption capacity and stability of a metal‒organic framework (MOF, specifically Ni₂Cl₂(BTDD), H₂BTDD = bis(1H−1,2,3-triazolo[4,5-b],[4′,5′-i])dibenzo[1,4]dioxin). This design leverages the complementary properties to achieve lower desorption temperatures (e.g., < 60 ^oC in device level) compared to typical MOF-based systems (usually >90 ^oC in device level), thereby significantly reducing the energy consumption for desorption. Concurrently, the composite exhibits extended adsorption duration within the high-humidity window. Field validation across diverse climatic regions demonstrates the composite’s exceptional wide-range environmental stability and performance. The resulting SAWH device achieves a solar-to-water generation improvement up to 91% in a continental field test. This work presents a generalizable and effective pathway for enhancing SAWH performance through synergistic material engineering, enabling efficient water production and thermal control under varying environmental conditions.

A composite sorbent combining LiCl with MOF material Ni₂Cl₂(BTDD) enables high water uptakes while low desorption temperature, achieving solar-driven water yield over 1 L m^‒2 day^‒1, which has been confirmed in field tests even across diverse regions.

Introduction

More than four billion people worldwide are experiencing a freshwater shortage crisis¹, but the atmosphere contains an abundance of water vapor equivalent to six times the total amount of surface rivers and lakes². Thus, how to convert water vapor in the atmospheric environment into drinkable liquid water has become a research direction of great social significance^3,4. Currently, various studies have demonstrated the feasibility of this research direction. According to different technical routes, these explorations can be roughly divided into three different typical technical routes^5,6: the dew-based atmospheric water harvesting (AWH), which reduces water vapor to below the dew point temperature; the fog capture method, which can be used when the water vapor is close to saturation^7,8; and sorption-based AWH^9,10. Among them, the sorption-based AWH has received widespread attention due to its ability to better adapt to different humidity environments and its ability to be powered by low-grade energy sources represented by solar energy and waste heat^11–13. This technology utilizes the adsorption process of adsorbents to capture atmospheric water vapor, which is then desorbed by the adsorbent driven by heat sources such as solar energy^14,15. The desorbed water vapor can be condensed in the condensation chamber at a temperature higher than the ambient temperature, thereby achieving the conversion of water vapor to liquid water^16,17.

To address freshwater scarcity in both arid and humid regions³, it is necessary to develop highly efficient solar-driven AWH systems with broad climatic adaptability. Many studies have focused on the development of adsorbents, such as metal‒organic frameworks (MOFs)^18,19, hydrogels^20–22, silica gel-based adsorbents^23,24, etc. Among various adsorbents, MOFs have the advantages of large specific surface areas and fast water sorption kinetics, which have received wide attention^11,25. Nevertheless, their water harvesting capacity is often limited by the pore volume. To the best of our knowledge, traditional inorganic hygroscopic salts, represented by LiCl and CaCl₂, exhibit exceptionally high water uptake capacity due to their deliquescence property, which enables a solid-to-liquid phase transition under high-humidity conditions^22,26,27. However, they frequently suffer from issues like leakage and poor cycling stability caused by agglomeration^13,22,28. Therefore, constructing a composite that synergizes the high capacity of inorganic salts and the rapid kinetics of MOFs presents a promising strategy for developing high-performance AWH systems.

As far as we know, composites between hygroscopic salts and MOFs (e.g., LiCl@MIL-101²⁹, LiCl@UiO-66³⁰, CaCl₂@MOF-808³¹) can increase adsorption capacity, and the low enthalpy of vaporization of the confined LiCl solution helps to lower the activation temperature. However, their water uptakes were no more than 0.77 g g^‒1 at 30% relative humidity (RH) and no more than 2.56 g g^‒1 at 80% RH (Supplementary Table 1), which might be ascribed to the lack of strong binding sites for Li⁺ and Cl^‒ simultaneously in their original MOF structures. While we noticed that the MOF Ni₂Cl₂(BTDD) (H₂BTDD  = bis(1H−1,2,3-triazolo[4,5-b],[4′,5′-i])dibenzo[1,4]dioxin) not only exhibits outstanding performance of water capture (Supplementary Fig. 1), but also the completely exposed Ni²⁺ and Cl^‒ sites on the pore channels should be suitable binding sites for stabilizing Cl^‒ and Li⁺, respectively (Fig. 1a, b and Supplementary Fig. 2), which would be helpful to increase the adsorption capacity (details see the section of “Advantages of Ni₂Cl₂(BTDD) as the matrix material” in Supplementary Information).

Fig. 1. Performance characterization of MOF-based composite adsorbent.

Schematic diagrams of a Ni₂Cl₂(BTDD) and b hypothetical LiCl@Ni₂Cl₂(BTDD) structures (Purple, green, gray, blue, and red nodes represent the Ni, Cl, C, N, and O atoms, respectively. Orange represents LiCl inside the pores. H atoms are omitted for clarify). c N₂ (77 K) sorption isotherms of Ni₂Cl₂(BTDD) and LiCl@Ni₂Cl₂(BTDD)_30 (Inset: the pore-size distribution). d Water adsorption isotherms of LiCl@Ni₂Cl₂(BTDD)_30 at 30 °C showing multi-step sorption processes during water capture phase. Step 1: Water uptake rapidly increases at P/P₀ < 0.11 as a result of the chemisorption of active LiCl with increasing humidity; Step 2: Water uptake sharply increases, caused by the deliquescence of LiCl·H₂O at P/P₀ = 0.11; Step 3: Water uptake slowly increases due to the absorption of LiCl solution at P/P₀ > 0.11. e Adsorption/desorption kinetic profiles of the material under adsorption conditions (30 °C, 80% RH) and desorption conditions (80 °C, 3.6% RH). f Multi-cycle stability test of the material under adsorption conditions (30 °C, 80% RH) and desorption conditions (80 °C, 3.6% RH).

On the device level, numerous studies have validated the AWH system based on MOF adsorbent materials. For example, in arid environments such as deserts, the desorption process of adsorbents can be achieved by using concentrated light^29,32, solar vacuum tubes^33,34, electric heating^19,35, and other forms to increase the desorption temperature. At the same time, fans, cooling systems and other methods can be used to lower the condensation temperature and promote the condensation process of water vapor^35,36. Among various system forms, the AWH system driven by photothermal energy has attracted extensive attention because it does not require a power system or energy storage equipment and can better meet the requirements for off-grid deployment^9,32. However, there are two main challenges in the practical application of this system: the matching problem between the low desorption temperature conditions in actual use and the high-humidity adsorption conditions at night³⁷. For this type of system, to reduce or avoid the intervention of the power control system, an operation strategy of adsorption at night and desorption during the day is often adopted^18,33. Nevertheless, the irradiation during daytime desorption fluctuates and is often lower than the common desorption test conditions (1000 W m^‒2). Therefore, in this context, how to sufficiently match the adsorption and desorption conditions under natural environments to achieve water production enhancement has become a critical challenge. Besides, the common adsorption process cannot achieve flexible use of the nighttime adsorption time, that is, continuous adsorption at night through ultra-high adsorption capacity.

Herein, we used Ni₂Cl₂(BTDD) as the matrix and uniformly loaded LiCl with different concentrations on the internal skeleton of the material by impregnation. By virtue of multiple advantages of uniform dispersion of hygroscopic salts in the pore channel, and low desorption temperature due to the low enthalpy of vaporization of the confined LiCl solution, the prepared composite materials exhibit ultra-high water adsorption capacity under low desorption temperature. We conducted comparative tests using the same system architecture as existing research under the same operating conditions¹⁴. Furthermore, the constructed device with MOF-composite achieved a water yield of up to 1 L m^‒2 under 1-sun conditions, and the water production per unit area could increase by 50% compared to the Ni₂Cl₂(BTDD)-based device in outdoor experiments at different locations.

Results

Synthesis and characterizations of LiCl@Ni₂Cl₂(BTDD)

The composite adsorbent LiCl@Ni₂Cl₂(BTDD)_x (x = 10/20/30/40, representing different LiCl solutions) was prepared using a scalable impregnation method in different mass concentrations (wt%) of LiCl solution at room temperature (see “Methods” section for preparation of Ni₂Cl₂(BTDD) and LiCl@Ni₂Cl₂(BTDD)). This approach requires neither pressurization nor ultra-high-temperature operations, with mild reaction conditions enabling effortless large-scale material preparation. Notably, the formulation of 30 wt% LiCl solution, i.e., LiCl@Ni₂Cl₂(BTDD)_30 was chosen for the key structural characteristics and the final device (Supplementary Figs. 3–5), because it enables a synergistic balance of the high water adsorption capacity and fast adsorption kinetics (details see the section “MOF-composite with varying LiCl solution concentrations” in the Supplementary Information).

Powder X-ray diffraction (PXRD) and scanning electron microscopy (SEM) analyses collectively indicated the successful loading of LiCl within the internal pore channels of the MOF material via the impregnation method (Supplementary Figs. 6–7). N₂ adsorption isotherms of Ni₂Cl₂(BTDD) at 77 K (Fig. 1c) demonstrated that the pore volume (1.18 cm³g^–1) and Brunauer‒Emmett‒Teller (BET) surface area of 1839  m² g^–1 (Supplementary Fig. 8), close to the literature reference³⁸. Furthermore, a sharp pore size distribution centered at 2.20 nm for Ni₂Cl₂(BTDD) (Fig. 1a, c) aligns well with the value reported in the literature³⁸. As a contrast, LiCl@Ni₂Cl₂(BTDD)_30 showed a significantly lower pore volume (0.35 cm³g^–1) and BET surface area (504 m² g^–1, Supplementary Fig. 8), which may be due to the occupation of LiCl in the pores. Furthermore, the composite material shows a broader distribution centered at approximately 1.66 nm (Fig. 1c). This significant reduction in pore diameter by about 0.54 nm indicated the formation of a monolayer of LiCl on the internal pore surface (Fig. 1b and Supplementary Fig. 2). In addition, inductively coupled plasma-optical emission spectrometer (ICP-OES) indicated that the LiCl to Ni₂Cl₂(BTDD) molar ratio is approximately 14:1 (Supplementary Table 2), and the distribution of LiCl in MOF pores was confirmed by time-of-flight secondary ion mass spectrometry (TOF-SIMS) for LiCl@Ni₂Cl₂(BTDD)_30 (Supplementary Fig. 9).

Furthermore, the water sorption isotherms of LiCl@Ni₂Cl₂(BTDD)_30 composite sorbent show a distinctive three-step adsorption behavior (Fig. 1d), including salt chemisorption, deliquescence, and solution absorption at different water vapor pressures³⁹. Notably, the material exhibited exceptionally high adsorption capacities under high-humidity conditions, with high water uptake reaching 3.46 g g^‒1 at 30 °C and 80% RH, and 1.72 g g^‒1 at 30 °C and 50% RH. Such high water uptake should be attributed to the deliquescence of LiCl^29–31. To further investigate the desorption performance of the material, we systematically evaluated the desorption behavior at different temperatures (Fig. 1e). The results showed that the developed sorbent can achieve 94.7% desorption of the water uptake at temperatures as low as 60 °C (Supplementary Fig. 10). Subsequently, multi-cycle performance and stability tests showed that the adsorbent maintained a water vapor uptake consistently exceeding 3 g g^‒1 over 10 cycles (as shown in Fig. 1f).

Construction of a modular photothermal-driven AWH system

Considering the application scenarios of solar-driven AWH, this adsorbent material exhibits significant advantages, particularly in arid regions, relative humidity at nighttime can exceed 50% even in dry environments, posing higher requirements for adsorbents to fully utilize long humidity windows of 8‒10 h duration daily (Supplementary Fig. 11, detailed climatic data analysis, please refer to the section: “Climatic data of typical arid and semi-arid regions” in Supplementary Information). While advanced adsorbents are crucial for AWH performance, their potential can only be fully realized through coordinated device-level optimization. Solar-driven AWH system works by alternating between nighttime vapor adsorption and daytime desorption, leveraging the natural day-night cycle. The daytime operation of AWH, which integrates sorbent desorption and vapor condensation, presents a fundamental thermal dilemma: desorption is favored at high temperatures, whereas condensation requires a cold surface (Fig. 2a, for a detailed thermodynamic AWH analysis, please see Supplementary Fig. 12 in the section on Device design and thermal management strategies in Supplementary Information. Driven by this critical insight into the process’s thermal demands, we engineered a highly compact and modular AWH system specifically to decouple these different temperature requirements. This system features a modular design, where the entire panel is composed of 4 cm × 4 cm adsorbent modules (Fig. 2b), and the detailed design drawings are provided in Supplementary Figs. 13 and 14. Such modularity enables flexible adaptation to diverse practical needs and facilitates system scalability for expanded applications. The device mainly consists of two core components: an integrated solar-thermal driven adsorption bed and a thermally enhanced condenser with dual-layer insulated mass transfer plates. Dual-layer insulated mass transfer plates (Fig. 2c) achieve thermal demand decoupling between condensation and desorption through internal thermal partitioning, which can effectively isolate thermal radiation and weaken convection.

Fig. 2. Photothermal-driven AWH device and laboratory testing.

a Working principle of the proposed AWH device. b 3D model of the AWH device. Components 1–5: 1. Double-layer insulation; 2. Photothermal panel; 3. Modular adsorption bed; 4. Dual-layer mass transfer plate; 5. Integrated condensation chamber. c Internal structure and thermal radiation attenuation mechanism. d Variation of water yield during a 7-h experiment. e Temporal variation of temperature distribution characteristics. 1–4: 1. Photothermal plate of MOF-based device. 2. Photothermal plate of the MOF composite-based device. 3. Condensation chamber of MOF composite-based device. 4. Fins of the MOF composite-based device. f Desorbed water vapor and water yield in cyclic testing (cycles 4–12). Adsorption condition: 30 °C and 80% RH. Desorption condition: 30 °C and 35% RH, with irradiation intensity of 1000 W m^‒2.

We conducted laboratory performance tests in the climate chamber (water yield data as shown in Fig. 2d). Experimental results showed that under the system’s operating conditions, the increased water vapor desorption induced a sweating cooling effect. This phenomenon manifested as a temperature drop of over 10 °C (83.6 vs. 97.9 °C in Fig. 2e) on the photothermal panel through the cooling output generated by the adsorbent desorption process. These findings indicate that compared to Ni₂Cl₂(BTDD)-based device, MOF-composite possesses lower desorption temperatures and greater working capacity. Within 7 h, the device produced more than 1 L of liquid water per square meter in total (Fig. 2d), and the thermal efficiency increased by 25% compared with that before the salt composite. Based on this strategy, the proposed device achieves an ultrahigh volumetric water harvesting capacity of 30.3 L m^‒3, ranking among the highest levels reported to date for existing diurnal solar energy-driven AWH devices^{9,10,15,32,40–43} (for a detailed device performance comparison, please see Supplementary Table 2).

In addition, the working conditions adopted in the experiment were the light intensity equivalent to noon sunlight commonly used in existing literature. However, considering the variation of daytime irradiation, such continuous high-intensity irradiation conditions do not last for 7 h. Therefore, the adsorbent must adapt to a lower desorption temperature to meet the actual needs. According to our analysis, the composite strategy we proposed can achieve high-performance desorption at a lower desorption temperature through simple salt hygroscopic factor compounding. Tests under varying irradiation and humidity conditions demonstrate that the device equipped with the developed MOF-composite sorbent achieves strong performance across a wide range of climatic conditions (Supplementary Fig. 15 and Supplementary Table 3 shown in Supplementary Information, section on “Performance of the proposed AWH system under different operating conditions”). Notably, it produced a high water yield exceeding 500 g m^‒2 even under a low irradiation intensity of just 400 W m^‒2. The system developed based on this strategy thus has greater practical application value. We conducted multi-cycle tests at the device level to verify the stability of the combination for the material and system (Fig. 2f). The results of 12-cycle tests at the device level showed that the system could operate stably under the test conditions. During the laboratory cycling tests, the proposed device consistently achieved a water production yield of approximately 42 g, while maintaining a water vapor recovery rate above 90% (see the section of “Multiple cycling performance tests” in Supplementary Information), indicating that the system is capable of stable operation.

Comparative field testing across different climatic regions

To better evaluate the performance of the constructed system in different real-world application scenarios, we conducted field experiments under various climatic conditions in Shanghai (31.0271° N, 121.4460° E), Jinan (36.6599° N, 117.2334° E), and Kunming (24.8644° N, 102.8586° E). During each experiment, all test conditions (including pure powder and composite adsorbent systems) were kept identical, and no auxiliary equipment was used, as shown in Fig. 3a. All tests were performed under consistent protocols; both the MOF-composite and Ni₂Cl₂(BTDD) systems were evaluated in parallel without any auxiliary equipment.

Fig. 3. Outdoor experiments.

a Diagram of outdoor experimental setup. b Temperature data of key components in Shanghai experiments (October 3, 2022). 1–4: 1. Photothermal plate of MOF-based device. 2. Photothermal plate of the MOF composite-based device. 3. Condensation chamber of MOF composite-based device. 4. Fins of the MOF composite-based device. c Climatic data from Shanghai outdoor experiments (October 1–3, 2022). d Nighttime adsorption data and water collected on October 1 from Shanghai experiments (Inset: Water collected in the test).

Outdoor experiments demonstrated that the MOF-composite system consistently outperformed the pure MOF device under varied climatic conditions (Table 1). In Shanghai and Jinan, representing moderate-to-high humidity environments, the composite system collected nearly 50% more water, with the most enhancement observed in Jinan (over 90% increase in water yield). Even under the low-temperature winter conditions of Kunming, the MOF-composite system still achieved effective water production, confirming its broad applicability.

Table. 1.

Performance in field tests

Location	Season	Climate regions	Water desorbed (MOF-composite based system) (g)	Water yield (MOF-composite based system) (g)	Collecting ratio (%)	Water yield (MOF-based system) (g)
Shanghai	Autumn	Humid Subtropical	34.09	32.61	96	22.29
Jinan	Summer	Continental	29.89	27.8	93	14.45
Kunming	Winter	Subtropical Highland	23.84	18.67	94	16.57

A detailed analysis of site-specific performance revealed several key advantages of the proposed MOF-composite material. In Shanghai, the experiment lasted three days (temperature data of the tests on October 3 is shown in Fig. 3b), with no electrical auxiliary equipment used throughout the process. The final experimental results demonstrate that the proposed device with MOF-composite exhibits superior adaptability to daily cycling operations. Under the conditions of a daily maximum solar irradiance of 0.8 sun and an ambient temperature of 33 °C, the maximum driving heat source temperature provided by the photothermal plate is approximately 85 °C, and the duration during which the temperature exceeds 80 °C throughout the day is about 5 h (Fig. 3b). Experimental observations showed that liquid water could appear in the water collection bottle around 9:30 a.m., 1 h earlier than the Ni₂Cl₂(BTDD)-based device, which started to produce water around 10:30 a.m. This indicates that the device improved by the composite material strategy can achieve desorption under lower-RH environmental conditions. Meanwhile, we observed the same phenomenon as in laboratory experiments: the composite adsorbent exhibits higher water vapor desorption capacity, thereby providing greater cooling capacity for photothermal components. Compared to the MOF-based device, the solar thermal panel with MOF-composite demonstrated a temperature reduction effect of approximately 10 °C during testing.

It also extended the effective adsorption period, fully utilizing the high nighttime humidity and achieving continuous moisture capture (Fig. 3d). Different from the Ni₂Cl₂(BTDD)-based device, which reaches adsorption saturation around 11 p.m. at lower water uptake, the proposed device with MOF-composite fully utilizes the high-humidity environment at night, thereby achieving the ability of potential continuous adsorption at night. Compared to the Ni₂Cl₂(BTDD)-based device, the device with MOF-composite achieves ~46% higher water vapor adsorption and desorption capacity (detailed data and analysis can be found in Supplementary Information, section of “Outdoor experimental details” in Shanghai) while realizing precise mass matching between adsorption and desorption phases, thereby enabling stable water production from 30.2 to 32.6 g day^‒1. Furthermore, outdoor testing results confirmed the operational stability of the proposed device with the MOF-composite. These comparisons also demonstrate that utilizing the composite adsorbent’s characteristics of low desorption temperature and high adsorption capacity enables adaptation to variations in operating conditions through both desorption and adsorption pathways, thereby effectively enhancing system efficiency. To verify the reliability of field data, we further replicated Shanghai’s stable noon conditions during the second day in the climatic chamber, which confirmed consistent temperature trends and performance metrics between indoor and outdoor settings (Supplementary Fig. 17).

The experimental results have been further validated in different climatic regions, as exemplified by Jinan and Kunming (Supplementary Fig. 18). In Jinan, despite lower ambient and photothermal temperatures (~10 °C below that in Shanghai as shown in Supplementary Fig. 18a), the composite device nearly doubled the water yield of the pure MOF system, highlighting its superior performance under suboptimal solar irradiation (Supplementary Fig. 18b). In Kunming during winter. The results demonstrated that the adsorption material achieved high-performance liquid water production at a driving temperature as low as 60 °C (Supplementary Fig. 18c, d), corresponding to a saturated vapor pressure only one-third of that at 85 °C (19.923 vs. 57.875 kPa). In previous studies, the required heat source temperature often exceeded 90 °C due to impaired heat and mass transfer during actual system formation^13,15,18,35. Thus, this composite strategy can better enhance the applicability of AWH under conditions of low irradiation and heat source temperatures.

Collectively, the field tests validate that the MOF-composite significantly enhances solar-to-water generation across diverse climates. The solar-driven AWH system operates by utilizing low-grade solar thermal energy to desorb at a lower temperature, thereby achieving the transformation of water vapor into liquid water. Consequently, the improvement in water collection efficiency reflects an overall enhancement in energy efficiency. Specifically, the efficiency improvement reached 91% in Jinan under a continental climate, while an average enhancement of approximately 50% was achieved in medium-to-high humidity regions such as Shanghai and Jinan. These results collectively underscore the significant potential of synergizing MOF-derived composite sorbents with advanced thermal management in AWH devices.

Water quality analysis

The analysis of water quality indicates that, within the allowable precision range of testing instruments (details of testing instruments are introduced in Supplementary Information: Details of water ion concentration analysis), the ion concentration of the proposed water sample is identical to that of distilled water. No lithium ions or nickel ions were detected, confirming its suitability for direct consumption. This result demonstrates that the collected water originates entirely from the condensation of water vapor generated during the desorption process on the condenser walls (as shown in Table 2). Furthermore, it confirms the absence of salt leakage throughout the process.

Table. 2.

Water quality test indicators for harvested water

	Li	Ni	Al		Ca	Fe
Harvested water	Undetected	Undetected		Undetected	Undetected	Undetected
WHO standard	NA	0.02 mg·L‒1		0.2 mg·L‒1	NA	0.3 mg·L‒1
Detection limit	0.0008 mg·L‒1	0.0005 mg·L‒1		0.003 mg·L‒1	0.002 mg·L‒1	0.001 mg·L‒1

Discussion

To improve the efficiency of the solar-driven AWH system, we propose a strategy of introducing hygroscopic salts to construct composite adsorbents, achieving enhanced dynamic adaptability and desorption performance for MOF materials under varying humidity conditions. The composite sorbent based on the selected Ni₂Cl₂(BTDD) offers multiple advantages. First, it significantly reduces the required desorption temperature at the system level. Moreover, the encapsulation of LiCl within its porous structure promotes uniform dispersion of the salt, thereby improving heat and mass transfer efficiency. Furthermore, this structure effectively prevents the aggregation, leakage, or migration of LiCl under high-humidity conditions. The developed composite achieves near-complete desorption at 60 °C and delivers higher water uptake with increasing humidity, offering ideal properties for broad environmental compatibility. The field results in different locations show that the proposed system with MOF-composite can achieve high-performance water vapor production at the system level with a driving temperature of only 60 °C and boost the water harvesting efficiency by up to 91% under the operating conditions in Jinan. These experimental data indicate that by using the proposed environment-adaptive strategy, the composite sorbent can better adapt to the benefits brought by increased humidity and longer moisture absorption time in the future.

Methods

Synthesis of MOF

The synthesis of Ni₂Cl₂(BTDD) was according to the method in the literature¹⁹.

Synthesis of LiCl@Ni₂Cl₂(BTDD)

We initially synthesized the MOF material Ni₂Cl₂(BTDD) via a reflux synthesis method based on previously reported preparation protocols, followed by drying the obtained product. The dried MOF powder was then immersed in a 10/20/30/40 wt% LiCl (Anhydrous LiCl with AR ≥ 99.0% was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.) solution for a 24-h impregnation treatment. Magnetic stirring was applied during the impregnation process to ensure sufficient penetration of the LiCl solution into the MOF pores. The resulting solution was then centrifuged at 1000 rpm (10,619 × g) to achieve preliminary solid-liquid separation, with the material precipitating at the bottom of the centrifuge tubes. Centrifugation was performed for 10 min to ensure complete phase separation. The separated material was subsequently dried at 80 °C to obtain the final composite adsorbent LiCl@Ni₂Cl₂(BTDD). For ease of description, we named the MOF composite material immersed in a 10/20/30/40 wt% LiCl solution as LiCl@Ni₂Cl₂(BTDD)_10/20/30/40, respectively.

Design of AWH prototype

We constructed a branch biomimetic AWH device using inexpensive and reliable 304 stainless steel as the raw material. We have detailed the device’s design drawings and usage details in Supplementary Information: Design schematics of the AWH device and the insulated mass transfer plate.

Experimental details

We used an electronic recorder (TP1000 paperless recorder from SHENZHEN TOPRIE ELECTRONICS CO., LTD.) to record the temperature data during the process. In SI, we have also described some key experimental details and photos taken during the tests, and other experimental information can be provided upon request.

Author contributions

R.W., J.P.Z., and D.D.Z. supervised and led this research. Z.S., X.F., and H.W. designed the experiments and processed the data. S.D., X.F., W.Y.S., Z.S.W., B.J., and H.L. performed experiments. P.P. revised the article. D.D.Z. and R.W. revised the article and optimized the experimental process. All authors discussed the experiments and results.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data supporting the findings of this study are available within the paper and its Supplementary Information files. Source data are available from the corresponding authors upon request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-68946-8.