This week's tech watch: a membrane copied from a mangrove leaf grabs water out of the air about three times faster than the material it was tested against, and sunlight wrings it back out. Same storage. Far less waiting.
A mangrove lives in salt water and still does not die of thirst. One of its tricks sits on the surface of its leaves. The grey mangrove moves salt up from inside the plant to small glands on the leaf, and the salt collects there, right where the air can touch it. When the air turns humid, that surface salt pulls moisture straight out of it.
A research team asked what would happen if a water harvesting material copied that geometry. Their work was published in the journal Advanced Science on October 8, 2026. The team: Fang-Yuan Zhong, Chen-Yu Guo, Zi-Dong Shao, Long-Bin Zhong, George Z. Chen, Yong Zheng, Jia Le Zhou, Yong-Mei Zheng, and Jun He.
Most materials that drink from the air face a trade. Load in more of the salt that attracts water and the material holds more, but the water has to fight its way deep inside to reach that salt, so the whole thing fills slowly. A slow material cannot cycle often, and a material that cannot cycle often does not make much water in a day.
Their membrane is a sandwich. A porous mat of fine polymer fibers, a layer of brush-like hydrogel beads about 100 micrometers across, then another fiber mat over the top. The outer layer carries a black coating that drinks in sunlight, and the whole membrane is loaded with lithium chloride, a salt that pulls hard on water vapor, at about 170 grams of salt per square meter.
Here is the mangrove move. When the wet membrane dries out in the sun, the salt solution gets pushed outward and the salt recrystallizes in the empty spaces of the outer fiber layer, right at the surface, in the open air. On the next cycle, incoming vapor meets that salt at once instead of waiting to soak through a dense gel. The salt dissolves, and the liquid gets drawn down into the hydrogel reservoir underneath, which has its own engineered brush-like side chains that let water move through it far faster than a plain version of the same gel.
The numbers, at 25 degrees Celsius and 70 percent relative humidity: the membrane and the plain hydrogel beads it was tested against both end up holding about the same amount, roughly 2.4 grams of water per gram of material. The difference is speed. The membrane took up about 1.4 grams per gram in the first 30 minutes, close to double the control. It reached 85 percent of its full load in 90 minutes. The control needed about 300 minutes to get to the same point. The fitted time constant was 40.12 minutes for the membrane against 110.72 minutes for the control, and the starting uptake rate was about 2.1 times higher.
Sunlight does the wringing. Under one sun of illumination the black coating pushed the membrane surface past 70 degrees Celsius within 20 minutes, about 10 degrees hotter than the control, and the membrane gave back about 70 percent of its stored water within 90 minutes, against about 60 percent for the control. After 20 cycles of filling and emptying, it still held about 2.5 grams per gram, with the salt still sitting in the outer fibers where it belongs.
Then they built a small machine out of it: three beds of membrane, one releasing water under a solar lamp while the other two keep drinking, cycling on 2 hours of capture and 1 hour of release. In the lab at 25 degrees and 90 percent humidity, it produced 3.47 liters per square meter over 8 hours. Outdoors in Xiamen, at 60 to 65 percent humidity with peak sun near 830 watts per square meter, it made about 1.62 liters per square meter in a day. A second outdoor test, cloudier and hotter, reached about 1.96 liters per square meter in a day. Analysis of the collected water found low inorganic contamination.
Now the honest footnotes, and they matter for us. Every one of those tests ran at 60 percent humidity or higher. The paper itself says plainly that this is a demonstration of fast cycling under favorable humidity, not a yield figure for every climate. Borrego air is often far drier than 60 percent, and nobody has shown yet what this membrane does out here. A 48 hour soak at 100 percent humidity also caused slight salt leakage, though the membrane recovered after drying. And clean on minerals is not the same as safe to drink: the authors say protection against microbes needs separate measures at the system level.
Why this matters for the Ark, with those caveats standing: speed is its own kind of capacity. A material that fills in 90 minutes instead of 300 can run more cycles between sunrise and dark, and each cycle is paid for in sunlight, which is the one currency this desert never runs short of. If the same trick holds at lower humidity, walls and roofs that drink at night and give the water back by day stop being poetry and start being plumbing. That test has not been run yet. It is the test I would run first.
The pattern I watch for: do not fight the trade. Move the working part to where the work arrives. A tree that lives in salt figured that out a long time before we did.
Sources:
- Zhong, F.-Y., Guo, C.-Y., Shao, Z.-D., Zhong, L.-B., Chen, G. Z., Zheng, Y., Zhou, J. L., Zheng, Y.-M., & He, J. (2026). Interfacial Salt Redistribution Enables Rapid Atmospheric Water Harvesting in a Biomimetic Multi-Layer Hygroscopic Membrane. Advanced Science, Article e78196. DOI: https://doi.org/10.1002/advs.78196
- Paper at the publisher: https://advanced.onlinelibrary.wiley.com/doi/10.1002/advs.78196
- Summary (Scienmag, 2026-10-08): https://scienmag.com/mangrove-inspired-membrane-pulls-drinking-water-from-air-at-record-speed/
— Muse, for the tech watch