The material at the heart of the study is a covalent organic framework, or COF, built from two molecular building blocks: 1,3,5-tris-(4-aminophenyl)triazine, abbreviated TAPT, and 5,5′-diformyl-2,2′-bipyridine, abbreviated DFBP. These units condense into a rigid, porous, crystalline polymer whose periodic lattice is studded with nitrogen-rich sites that can chelate uranyl ions from solution. COFs have long been attractive for photocatalysis because their ordered conjugated structures absorb light and generate electron-hole pairs efficiently. The problem, the researchers explain, is that in most designs the photogenerated electrons wander randomly through the framework, recombining with holes before they can reach the atomic active sites where uranyl reduction and capture actually happen. Directional induction of photoelectron transport to those atomic sites, they note, has remained a grand challenge in photocatalytic uranium extraction.

The team’s solution was to engineer localized potential wells directly into the framework. These wells are regions of the lattice where the electronic energy landscape dips below that of the surrounding structure, acting like a series of tiny valleys that funnel photogenerated electrons downhill toward the catalytic centers. According to the authors, the construction of these localized potential wells induces multiple electron transport paths toward the atomic active sites, which facilitates the separation of photogenerated electron-hole pairs and enhances photocatalytic uranium extraction from natural seawater. In effect, rather than relying on chance encounters between mobile electrons and uranium-binding sites, the material builds an electrical roadmap that guides charge carriers to their destination.

The performance figures are striking. The optimized TAPT-DFBP COF achieved an average uranium extraction rate of 7.25 milligrams of uranium per gram of material per day, a figure the team reports as higher than those of previously reported active materials. Even more compelling is the demonstration at scale: the researchers fabricated a large-area COF film measuring 150 centimeters by 250 centimeters, an industrial-level dimension that dwarfs most laboratory photocatalyst samples, and deployed it in a flow-through extraction system in real marine environments. That film achieved a photocatalytic uranium extraction capacity of 8.9 milligrams per gram, showing that the laboratory mechanism survives contact with the far messier chemistry of actual seawater, with its competing ions, dissolved organic matter and biofouling organisms.

To understand why the potential wells work, the team deployed a battery of photophysical and computational investigations, including carrier dynamics measurements, characterization of electronic excited states, and density functional theory calculations using the PBE0 functional. These analyses revealed that the potential wells reshape the excited-state landscape of the framework, promoting spatial separation between electrons and holes and opening multiple conduction pathways rather than a single, easily congested route. Electron paramagnetic resonance and related spectroscopic probes tracked how electrons accumulated at active sites and were transferred to adsorbed uranyl species, converting soluble hexavalent uranium into extractable reduced forms deposited on the framework. The bipyridine nitrogen sites serve double duty, both anchoring uranyl ions within reach of the electron pathways and helping define the local energy minima that pull electrons inward.

The broader context makes the advance significant. Uranium is the irreplaceable fuel of nuclear power, and terrestrial reserves, while substantial, are finite and geopolitically concentrated. Seawater represents the ultimate backstop resource, but extracting uranium at parts-per-billion concentrations economically has defeated many approaches, from the amidoxime fiber adsorbents developed by Japanese researchers in the early 2000s to more recent bioinspired membranes, biomimetic nanochannels and uranyl-imprinted nanocages. Cost analyses of braided polymer adsorbent systems have historically suggested that seawater uranium would be far more expensive than mined uranium. Photocatalytic strategies aim to change the economics by using sunlight to actively convert and immobilize uranyl rather than passively waiting for diffusion to a binding site, and the new work shows how to maximize that solar advantage at the level of individual charge carriers.

What distinguishes this study from prior photocatalytic efforts is the explicit engineering of the material’s internal electric field topology. Earlier approaches tuned local charge distribution in multicomponent COFs or built donor-acceptor architectures, often borrowing from the design principles that have advanced artificial photosynthesis of hydrogen peroxide. The localized potential well strategy generalizes that logic: instead of optimizing bulk band structure, it inserts deterministic sinks into the energy landscape that act on every photogenerated electron. Because the wells are built into the covalent connectivity of the lattice rather than grafted onto its surface, they are stable, uniform and compatible with the film-processing chemistry needed for large-area manufacturing, which the 150-by-250-centimeter film demonstrates convincingly.

The practical implications extend beyond uranium. The same principle of directional electron funneling could enhance photocatalytic reactions in which charge carrier recombination is the bottleneck, including hydrogen peroxide photosynthesis, carbon dioxide fixation and hydrogen evolution, all areas where COFs have shown promise. For nuclear energy planners, a sunlight-driven adsorbent that works in real seawater and can be produced in industrial-scale sheets represents a tangible step toward seawater uranium becoming a genuine strategic reserve rather than a laboratory curiosity. The researchers acknowledge that economic deployment will still require durable materials that resist biofouling and can be regenerated repeatedly, and the study’s mechanistic framework provides a rational basis for iterating on those designs.

The study, which received support from the National Natural Science Foundation of China, the National Key R&D Program of China and several other national and provincial programs, was published in Nature Water on 9 September 2026 after peer review by Costas Tsouris, Xiangke Wang and Wenkun Zhu. By showing that the fate of a photogenerated electron can be programmed into the very architecture of a crystalline polymer, the team has turned a fundamental physical chemistry insight into a working technology for one of the most coveted resources in the ocean. If the rates achieved in this demonstration can be maintained over long deployment cycles, the vast uranium wealth dissolved in seawater may finally begin to look less like an untouchable dream and more like an addressable reservoir for the nuclear age.

Subject of Research: Photocatalytic uranium extraction from natural seawater using covalent organic frameworks with localized potential wells for directional electron transport

Article Title: Localized potential wells enabling directional electron transport boost photocatalytic uranium extraction from natural seawater

Article References: Xu, Y., Zhou, Z., Zhao, R., Guo, X., Wang, Y., Liu, Y., Lin, Z., Sun, Z., Yu, P., Luo, M., Wang, J., & Guo, S. (2026). Localized potential wells enabling directional electron transport boost photocatalytic uranium extraction from natural seawater. Nature Water. https://doi.org/10.1038/s44221-026-00685-y

Image Credits: AI Generated

DOI: 10.1038/s44221-026-00685-y

Keywords: uranium extraction, seawater, photocatalysis, covalent organic frameworks, localized potential wells, electron transport, uranium mining, nuclear fuel, Nature Water, charge separation, COF film, marine resources