All-water supercapacitor enabled by 1-nm clay channels

Article https://doi.org/10.1038/s41467-026-73924-1 All-water supercapacitor enabled by 1-nm clay channels Received: 1 January 2025 Accepted: 22 May 2026 Vasily Artemov 1,2 , Svetlana Babiy 2, Yunfei Teng 2, Jiaming Ma2, Alexander Ryzhov3, Tzu-Heng Chen 2, Lucie Navratilova2, Victor Boureau Pascal Schouwink2, Mariia Liseanskaia1, Patrick Huber 1,4, Fikile Brushett Lyesse Laloui2, Giulia Tagliabue 2 & Aleksandra Radenovic 2 2 , 5 , Water confined to channels one nanometer thick exhibits electrochemical behavior distinct from bulk water, including enhanced protonic conductivity and large dielectric anisotropy. Here, we exploit these characteristics to design a scalable electrochemical energy storage system-a “blue capacitor”-constructed entirely from naturally abundant materials. By assembling layered clays and conductive graphene, we produce 1-nm-thick channels in which confined water acts as the sole electrolyte. We systematically study different clay types, the electrode composition, and separator thickness using complementary physicochemical and electrochemical techniques. The device operates stably up to 1.6 ± 0.1 V, achieves specific capacitances of 40 F g−1, 97 ± 2% coulombic efficiency, and stable performance over more than 60,000 charge-discharge cycles at a voltage window of 1 V and a scan rate of 10 mA. Structural and dynamic analyses validate the device architecture, water purity, and proton transport in the nanopores. These results demonstrate that nanoconfined water can function as an electrolyte in a macroscopic electrochemical device, providing a platform for exploring sustainable aqueous energy storage systems. 1234567890():,; 1234567890():,; Check for updates Water’s ability to store and transport electric charge underlies a wide range of processes across technology and nature, from electrochemical devices1 to lightning in the atmosphere2 and proton exchange in biological systems3. Clouds, for example, accumulate gigajoules of electricity relying largely on water interfaces4, illustrating that aqueous interfaces can sustain large-scale charge separation (Fig. 1a). Translating this phenomenon into controllable technologies could provide new strategies for sustainable electrochemical energy storage. Yet, despite continued efforts since the pioneering water electrification experiments of Thomson5 and Tesla6, artificial systems exploiting pure water as the active electrolyte for reversible charge storage have remained limited. Current batteries and supercapacitors (Fig. 1b) rely on concentrated electrolytes or metal oxides7,8, which can limit sustainability and scalability. Developing energy-storage concepts based on abundant and environmentally benign materials is therefore an important objective for next-generation electrochemical technologies9. Similar principles may also be relevant for emerging technologies such as biointerfaces and neuromorphic devices10,11. At nanometer scales, water’s molecular structure and dynamics can deviate from its bulk behavior12–15. Under strong nanoconfinement, its dielectric response, proton transport, and interfacial polarization may be significantly modified16–19. These effects have become central topics in nanofluidics and interfacial electrochemistry20–24. In particular, experiments in artificial nanochannels and layered materials have reported fast proton conduction18,19 and dielectric anisotropy16,17, suggesting that nanoconfined water could support efficient charge 1 Hamburg University of Technology, Hamburg, Germany. 2École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland. 3Austrian Institute of Technology, Vienna, Austria. 4Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany. 5Massachusetts Institute of Technology, Cambridge, MA, USA. e-mail: Nature Communications | (2026)17:5014 1 Article https://doi.org/10.1038/s41467-026-73924-1 a b - e- c - e- V Blue capacitor - e- V V Bulk water Interfacial water ~1 nm + + + 1700s 1960s-current This study Electrodes: Nonporous Nano-porous Topologically connected nano-porous Electrolyte: Water Bulk electrolyte Confined water Example: Leyden jar Supercapacitor Blue capacitor Topology type I Topology type II Fig. 1 | Evolution of double-layer capacitors (DLCs). a Leyden jar: an early DLC based on water and nonporous electrodes, storing charge in a nanometer-wide interfacial water layer. b Supercapacitor: a state-of-the-art DLC with high-surface-area electrodes and a separator immersed in a bulk-like concentrated electrolyte or ionic Illite d Kaolinite =O = Si4+ or Al3+ = Al Water 3+ or Mg 10 nm - 2 μm h Water - Alumina plane Silica plane X-ray intensity (a.u.) c MMT MMT 12 9 h → ~ 2 nm Wet clays ~ 1 nm Wet 10−2 0.005 2+ = Na+ or K+ = H+ b 101 0.6 S/m 0.1 Dry Electrical conductivity (S/m) MMT 1 nm a liquid (light blue), enabling high capacitance at the cost of chemical complexity. c Blue capacitor of this study: a new configuration, topologically distinct from the previous two, with a continuous network of strongly confined water as a sole electrolyte (dark blue), continuously crosslinking the electrodes and the separator. MMT Illite Kaolinite 10−5 Bulk water −8 10 Confined water H+ 10−2 Bulk water 6 10−5 3 0 2,5 Dry clays 3,0 3,5 4,0 4,5 Wave vector transfer, q (nm-1) 10−8 −1 10 101 103 105 Frequency (Hz) Fig. 2 | Structure and dielectric properties of clays. a Crystal structure of the three most abundant natural clays. b Enlarged structure of montmorillonite (MMT) clay with an interlayer space accessible to water. c Synchrotron small-angle X-ray scattering (SAXS) data for wet and dry MMT (raw data see in SI Fig. S6), showing interlayer water penetration. d Proton conductivity of clays under wet (top) and dry (bottom) conditions at 25 ± 1 °C (see also SI Figs. S12-15). The inset schematically shows the path of a proton through the confined water in wet clays. Source data are provided as a Source Data file. transport even in the absence of conventional electrolytes (Fig. 1c). However, most studies so far have been limited to nanoscale experimental systems25,26, which, although valuable for fundamental investigations, face challenges related to reproducibility, scalability, and device integration26,27. Natural clays present a unique opportunity to bridge this gap. Composed of layered silica and alumina sheets, they form abundant van-der-Waals heterostructures with interlayer spacings of about 1 nm when hydrated (Fig. 2a). Synchrotron small-angle X-ray scattering shows that hydrated smectite-type montmorillonite expands to accommodate water layers of ~1 nm (Fig. 2b, c), consistent with previous reports28,29. These galleries create extended networks of nanometer-scale aqueous channels, which can support proton transport under hydrated conditions18,19. Indeed, conductivity Nature Communications | (2026)17:5014 2 Article https://doi.org/10.1038/s41467-026-73924-1 measurements (...truncated)