The prototype comprises a cascade of as many as six evaporation–extraction chambers. A commercial photovoltaic panel covers the first chamber, where part of the incident sunlight becomes electricity and part becomes waste heat. A capillary wick attached behind the panel transports seawater through an ion-selective material. Water evaporates from the wick, vapour condenses on hydrophobic aluminium foil below, and the heat released by condensation passes through the foil to support evaporation in the next chamber. This sequential recovery of latent heat allows later stages to operate without receiving the full solar flux directly. The uranium-specific wick, designated SAU, contains polydopamine and polyethyleneimine co-coated biochar in a sodium alginate hydrogel supported by carbon fibre. The lithium-specific SALi wick instead contains hydrogen manganese oxide in the same type of conductive composite.

The researchers used TOPSIS, the Technique for Order Preference by Similarity to Ideal Solution, to compare preselected adsorbents on adsorption capacity, equilibrium time and cycling adsorption ratio at a pH near that of typical seawater. The method was applied as a transparent screening and comparison step before experimental validation, not as a substitute for physical reasoning or laboratory testing. It led to the selection of the polydopamine–polyethyleneimine-coated biochar for uranium and hydrogen manganese oxide for lithium. After encapsulation in the alginate composites, the adsorbents lost only about 5–9% of their adsorption capacity after normalization for adsorbent content, while their characteristic micropores remained accessible to dissolved ions. The composites were thermally stable below 200 °C, making them suitable for the reported photovoltaic-wick conditions. Incorporating the adsorbent-loaded alginate composites into carbon fibre also produced thin, flexible structures that could be laterally repeated in larger systems while retaining separate capillary pathways.

Under one simulated sun, a single chamber equipped with the uranium wick produced water at 1.26 kg m−2 h−1 and adsorbed 1.44 g m−2 of uranium, approximately 3.3 times the amount adsorbed in darkness. Its evaporation flux was 1.77 kg m−2 h−1, compared with 0.33 kg m−2 h−1 for the pure-water system. A lithium-equipped chamber adsorbed 2.2 g m−2 of lithium at one sun, 3.1 times its dark result. As irradiance rose from 0.5 to 1.5 suns, photovoltaic output increased from 63.2 to 186.1 mW and water production rose from 0.62 to 1.84 kg m−2 h−1. Adding five stages increased the water-production rate of the six-stage device to 4.01 kg m−2 h−1, largely through latent-heat recovery, while leaving photovoltaic output comparatively little changed. Six-stage uranium and lithium extraction were respectively 3.5 and 3.3 times those of a single stage. Compared with dark operation, the full device increased uranium adsorption by 10.6 times and lithium adsorption by 9.2 times under one sun.

The choice of six stages was based on diminishing thermal returns rather than an arbitrary design target. Calculated phase-change heat-recovery efficiency fell from 82.3% in the second chamber to 47.1% in the sixth as conductive and other irreversible heat losses accumulated. In the first chamber, evaporation accounted for 74.0% of the energy balance, while radiation, convection and conduction represented 6.3%, 6.2% and 1.1%, respectively. By the sixth stage, the capillary wick had cooled to 27.2 °C and the condenser to 26.1 °C, leaving a temperature difference of only about 1 °C across the air gap. Further chambers would therefore provide little additional heat flux. The measured evaporation flux could sometimes appear greater than the directly incident solar energy because incoming water was preheated inside the semi-enclosed assembly, adding sensible heat to vapor generation. These observations characterize a six-chamber prototype rather than a field-scale desalination plant, and no long-term commercial deployment was tested.

Solar enhancement arose from interacting temperature, flow and concentration fields rather than a single effect. Heating increased the diffusion rate of uranyl ions in the SAU composite by 85% when the simulated interface rose from 20 °C to 80 °C, and molecular-dynamics simulations supported faster ion transport. Because uranium and lithium adsorption was endothermic in the tested systems, higher temperature also helped adsorption overcome an activation-energy barrier. Evaporation produced flow normal to the wick, while tangential capillary flow continually delivered bulk solution to adsorption sites, disrupting the low-concentration boundary layer that develops around a stationary adsorbent. Those flows generated a concentration gradient that increased the driving force for ion uptake. The research team’s quantitative decomposition of these contributions was explicitly semi-quantitative: the temperature, flow and concentration fields interact nonlinearly, including through Marangoni convection, so the estimates indicate comparative trends rather than strict physical decoupling. At the higher concentrations used in the principal experiments, temperature was the largest contributor as solar flux increased from 0.5 to 1.0 sun, while concentration made the dominant contribution to uranium enhancement and temperature plus flow made the larger combined contribution to lithium.

Durability was assessed with real seawater spiked with the target ions over 20 consecutive cycles lasting about 40 days. The SAU device maintained water production at an average rate of 3.86 kg m−2 h−1, and the total organic carbon concentration of the collected condensate stayed below 2.0 mg l−1. The outlet salinity of the wick remained below saturation, with no visible salt crystallization or bioscaling detected on its surface. A macroscale unidirectional flow pattern transported ions toward the outlet and limited local accumulation, although the researchers cautioned that long-term antiscaling performance still depends on continued operation. Alkaline or acid washing regenerated the ion-saturated materials and maintained stable adsorption capacity across the cycles. Reported lithium-salt recovery yield and purity were 87.2% and 81.7%, respectively; uranium-compound recovery yield and purity were 84.8% and 77.5%. The wick’s surface morphology, structure, hydrophilicity and mechanical properties changed only slightly after cycling.

The team also compared designs for recovering both ions. In a series arrangement, six SAU units were followed by six SALi units and extracted 35.4 mg of uranium and 64.6 mg of lithium. In a parallel six-stage design, the order of the functional chambers changed the balance between the two targets. Placing three SAU chambers first extracted 45.8 mg of uranium and 41.8 mg of lithium, whereas placing SALi first yielded 67.9 mg of lithium and 32.3 mg of uranium. The hybrid parallel architecture therefore traded some ion-specific capacity for greater integration, uniform light distribution and a shared water-delivery system. A parallel unit with 10-cm SAU and SALi sections kept the outlet below saturation, and capillarities continued transporting seawater at night, although adsorption was weaker without sunlight. Tests with alternative adsorbents also showed that the platform’s thermal and flow effects could enhance uranium capture and could be adapted to caesium and boron; at one sun, a six-stage device adsorbed approximately 0.54 g m−2 of caesium and 1.30 g m−2 of boron.

Outdoor testing provided the clearest demonstration of the combined outputs. The system produced a peak power output of 154.7 W and a daily water collection of 19.82 kg m−2. Outdoor water-production rate did not decline monotonically with wind speed, which ranged from 0 to 2.44 m s−1; the highest rates, 2.55–2.59 kg m−2 h−1, occurred at moderate winds of about 2.0–2.5 m s−1 when solar radiation was also strongest. The result reflects the combined influence of sunlight, ambient temperature and convective airflow rather than a direct causal effect of wind on evaporation. X-ray photoelectron spectroscopy and X-ray diffraction supported the extraction of uranium and lithium, and the collected distilled water contained sodium, potassium, calcium and magnesium below World Health Organization drinking-water thresholds. Selectivity coefficients were approximately 1.07 × 10^5 for magnesium and 204 for vanadium in the relevant tests.

A preliminary technology and economic analysis estimated electricity production at US$0.0589 kW−1 h−1 and clean-water production at approximately US$0.00084 kg−1. It projected extraction costs of US$8.49 kg−1 for lithium and US$136 kg−1 for uranium. These are system-level estimates that allocate costs and revenue across the co-produced electricity, water and recovered materials, rather than stand-alone production prices for each output. Annual net income was calculated to stabilize above US$20 m−2 as the assumed capillary-wick lifetime increased, but the 20-year period used for comparison is a design target and benchmark, not a demonstrated service life. A preliminary cradle-to-grave assessment estimated total global-warming potential at 15.4 kg of carbon-dioxide equivalent per cubic metre of seawater treated. Most allocated product impacts were below industry benchmarks, while uranium compounds had a higher allocated footprint because seawater contains only about 3 μg l−1 uranium. Environmental and economic benefits increased with wick longevity, but biochemical degradation, physical ageing and mechanical deterioration remain constraints that require longer field-scale validation.

Subject of Research: Experimental solar-powered co-production of water, electricity, lithium and uranium from seawater

Article Title: Solar-powered resource separator for clean water production, electricity generation and critical resource recovery

Article References: Yu, Z., Ge, C., Mao, Z., Li, Y., Tang, S., Li, K., Duan, F., & Lu, J. (2026). Solar-powered resource separator for clean water production, electricity generation and critical resource recovery. Nature Water, 4(10), 1307-1317. https://doi.org/10.1038/s44221-026-00700-2

Image Credits: AI Generated

DOI: 10.1038/s44221-026-00700-2

Keywords: solar desalination, seawater resource recovery, lithium extraction, uranium adsorption, photovoltaics, latent heat recovery, adsorption materials, seawater treatment, clean water production, critical minerals, Solar-powered, resource