Water scarcity is one of the defining challenges of the twenty-first century, driven by population growth, climate change, pollution and the over-extraction of groundwater. Yet one vast reservoir remains largely untapped: the atmosphere, which is estimated to hold around 12,900 cubic kilometres of water in the form of vapour and droplets. Unlike rivers or aquifers, this resource is available to nearly everyone on the planet, even in the driest regions. A new study published in Results in Chemistry describes a carefully engineered material and device architecture designed to pull that water out of the air using nothing more than sunlight, and its findings could reshape how solar-driven water harvesters are built.

The research focuses on sorption-based atmospheric water harvesting, a technology that uses porous materials to capture water vapour even when the relative humidity is low, at or below 40 percent. Traditional approaches struggle in such conditions. Fog capture requires near-saturated misty air, while dew collection demands energy-intensive cooling of dry air to the dew point. Sorption-based systems, by contrast, can operate across a wide range of humidities and can release the captured water using low-grade heat such as sunlight or waste heat. The catch has been the adsorbent itself: conventional materials like zeolites and silica gel need desorption temperatures of roughly 75 to 200 degrees Celsius, which makes them inefficient when the heat source is a modest sun.

Metal-organic frameworks, or MOFs, have emerged as a promising alternative. These crystalline materials, built from metal nodes connected by organic linkers, can be tuned to adsorb large quantities of water and regenerate at temperatures of only about 40 to 80 degrees Celsius. Materials such as MOF-801, MOF-303 and UiO-66 combine high hydrothermal stability, large uptake capacities of up to 0.56 grams of water per gram of adsorbent, fast kinetics and favourable Type IV or V isotherm shapes. But MOFs carry their own weaknesses: thermal conductivities of just 0.1 to 0.4 watts per metre-kelvin, poor absorption of sunlight, and bed designs that throttle the movement of vapour in and out of the crystals.

The research team, led by Zhiyin Duan and Yu Cheng, tackled these limitations with a two-pronged strategy: modifying the material’s composition and optimising the structure of the adsorbent bed. First, they grew MOF-303 in the presence of few-layer graphene, a highly ordered two-dimensional form of carbon whose thermal conductivity can reach 2000 watts per metre-kelvin. Rather than mixing the components after synthesis, which can damage the MOF’s delicate pore structure, they used an in-situ solvothermal method, adding 4, 7, 10 or 15 weight percent of graphene directly into the reaction solution containing aluminium chloride, pyrazoledicarboxylic acid and sodium hydroxide.

The amount of graphene proved critical. At 15 weight percent loading, the composite lost 26.4 percent of its equilibrium water uptake compared with pure MOF-303, because excessive carbon crowds out pore volume and disrupts crystallinity. The sweet spot was 4 weight percent, designated MOF-303-FLG4. This composite preserved nearly all of the original adsorption capacity while boosting the intracrystalline diffusion coefficient of water vapour by 48 to 74 percent across relative humidities from 10 to 30 percent. Microscopy showed the elliptical MOF crystals resting on and between translucent graphene sheets, and nitrogen sorption measurements revealed that the average pore diameter had grown by 90 percent and total pore volume by 21.7 percent, even though the BET surface area dipped only slightly, from 931.87 to 909.84 square metres per gram.

Structural and chemical analyses confirmed the composite’s integrity. Powder X-ray diffraction retained every characteristic peak of MOF-303, with a new reflection at 26.48 degrees confirming the graphene’s presence. Infrared spectroscopy showed that the aluminium hydroxide bridges and carboxylate bands of the framework were unchanged, indicating that the graphene is held by physical interactions rather than chemical bonding. Raman spectra displayed the telltale D, G and 2D bands of layered graphitic carbon. Under simulated sunlight at an intensity of 1000 watts per square metre, delivered by a halogen lamp, the composite’s top surface warmed about 13.5 degrees Celsius more than pristine MOF-303, while its bottom surface ran 7.4 percent hotter, evidence that heat was penetrating deeper into the adsorbent layer.

Durability matters as much as performance, and here the composite also impressed. Over 20 consecutive adsorption-desorption cycles, with adsorption at 30 degrees Celsius and 40 percent relative humidity and desorption at 90 degrees Celsius, the water uptake of MOF-303-FLG4 fell by only 1.7 percent. X-ray diffraction patterns taken after cycling showed no peak shifts, disappearances or new phases, and the graphene reflection survived intact. The surface area declined by about 10.6 percent and total pore volume dropped to 0.458 cubic centimetres per gram, mainly through losses in mesopores and macropores, but substantial accessible porosity remained. Thermogravimetric analysis confirmed that framework decomposition still begins above 400 degrees Celsius, well beyond any operating temperature.

The second half of the study addressed a question often overlooked: how the physical arrangement of the adsorbent shapes performance. The team built packed beds of MOF-303-FLG4 powder with porosities of 0.65, 0.70 and 0.75 and thicknesses from 1 to 4 millimetres, held in containers of PTFE membrane and aluminium frames. Raising the porosity from 0.65 to 0.75 increased adsorption kinetics per unit mass by about 21 percent and desorption kinetics by 5 percent, but porosity 0.70 offered the best balance of capacity and speed. Thickness mattered even more dramatically: cutting the bed from 4 to 1 millimetre accelerated adsorption kinetics by 246 percent and desorption by 179 percent, because thinner layers shorten the diffusion paths that vapour must traverse.

The most striking results came from coated beds, in which the same mass of MOF-303-FLG4 powder was deposited onto copper foam substrates of varying pore density, using silica sol as an inorganic binder and no surfactants or adhesives. Scanning electron microscopy showed the coating following the three-dimensional copper skeleton deep into its pore network, held by mechanical interlocking, and a water droplet vanished from the surface within 253 milliseconds, confirming strong hydrophilicity. Substrates with a pore density of at least 60 pores per inch anchored the powder far more uniformly than the coarser 30 PPI foam, improving heat delivery during desorption. The coated bed reached a thermal conductivity of 0.793 watts per metre-kelvin, more than double the 0.306 of the loose powder, and its bottom-surface temperature under sunlight exceeded that of packed beds by up to 21.28 percent relative to pristine MOF-303.

The payoff is measurable in water. Under adsorption at 25 degrees Celsius and 30 percent relative humidity, the coated copper-foam bed captured 0.254 grams of water per gram of adsorbent within 30 minutes, outperforming reported configurations including MOF-derived nanoporous carbon, a MIL-101/salt monolith embedded in nickel foam, and a packed layer of pure MOF-303, which managed roughly 0.160 to 0.175 grams under comparable conditions. The coated bed’s diffusion coefficient of 1.82 times ten to the minus eighteenth square metres per second nearly doubled that of the packed bed, and its adsorption kinetics improved by 86 percent. The authors propose that graphene sheets act as conductive bridges between MOF crystals and the copper skeleton, dissipating the heat of adsorption during uptake and speeding solar heating during release. Their design recipe, 4 percent graphene, beds 2 to 3 millimetres thick, porosity near 0.7 and copper foam above 60 PPI, offers a practical blueprint for harvesters that could one day turn desert sunlight and thin desert air into a reliable supply of drinking water.

Subject of Research: Solar-driven atmospheric water harvesting using graphene-enhanced MOF-303 adsorbent beds

Article Title: MOF/few layer graphene coated copper foams and packed adsorbent beds for solar-driven atmospheric water harvesting applications

Article References: Duan, Z., Cheng, Y., Huang, K., Kang, J., Fan, W., & Zhao, X. (2026). MOF/few layer graphene coated copper foams and packed adsorbent beds for solar-driven atmospheric water harvesting applications. Results in Chemistry, 31, Article 103934. https://doi.org/10.1016/j.rechem.2026.103934

Image Credits: AI Generated

DOI: 10.1016/j.rechem.2026.103934

Keywords: atmospheric water harvesting, metal-organic frameworks, MOF-303, few-layer graphene, copper foam, solar-driven desorption, adsorption kinetics, thermal conductivity, porous materials, water scarcity, adsorbent bed design, photothermal materials