During sunny and sweltering summer days, much relief comes from sweat. As beads of moisture evaporate on the skin, they pull heat away almost invisibly. It feels like a nuisance, but this is one of the most efficient cooling systems in the animal kingdom.
But not all animals have the ability to sweat. The African elephant is one example, enduring intense heat out on the African savanna with dark gray, inches-thick skin. How does it stay cool?
"Wrinkles," says Dorit Aviv , an associate professor of architecture in the Weitzman School of Design . "Elephants have a network of cracks in their skin that trap water. When they spray themselves, that water stays put and evaporates slowly, cooling them over time."
The concept of cooling without sweat was first pitched to Aviv by her longtime collaborator Shu Yang , a materials scientist at Penn Engineering . The idea stuck with Aviv, whose research focuses on making the built environment more thermodynamically efficient in a warming world, because buildings don't sweat.
In a paper published in Advanced Materials , Aviv, Yang, and Kun-Hao Yu of Syracuse University turned that insight into a new kind of cement-based tile that captures, holds, and slowly evaporates water, cooling surfaces without fans, compressors, or complex moving parts.
Under infrared heating and periodic watering, the temperature beneath these tiles remained a stable 89.6° Fahrenheit. By comparison, cracked commercial stucco rose to 107.6°, while non-cracked stucco climbed to 125.6°.
"In the United States, we spend roughly 90% of our time indoors, and buildings consume about 40% of primary energy—nearly half of that for heating, cooling, and ventilation," says Yang. "A passive, bio-inspired cooling façade like this could lower surface temperatures by 10-20° Fahrenheit compared to traditional stucco."
Unlike mechanical air conditioning, which consumes fossil fuels and contributes to outdoor heat buildup by dumping indoor warmth outside, this is a truly passive system that cools the building's "skin" itself, she explains.
Trapping water to allow it to evaporate and using that phase change to pull heat away is simple in theory. But the researchers found this particularly challenging, because most conventional building materials are terrible at it since water droplets bead up, bounce off, or run straight down the surface before they can do any useful cooling.
The team needed to devise a way to guide each droplet to maximize its cooling potential, which prompted turning to cracks.
"In conventional construction, cracks signal weakness, the beginning of a material's decline," Yang says. "But by engineering where and how they form, we made networks of tiny channels that act like capillaries capable of pulling water across the surface and holding it in place."
Their process begins with a mix of ordinary Portland cement and a porous material called diatomaceous earth (DE), made from fossilized algae. Cast into thin tiles and allowed to partially hydrate, the material is then dried under carefully controlled conditions. As it shrinks, stress builds and releases along pre-designed patterns, forming a predictable lattice of cracks instead of random fractures.
"We realized that storing water and moving water are actually two different problems," explains Yu, who was a postdoctoral researcher in the Yang lab while conducting these experiments. "The microscopic pores inside the material act like tiny reservoirs that quickly soak up each droplet, while the engineered crack network acts like a system of canals that redistributes that water across the surface. It's really the combination of those two features that lets the tile hold onto water instead of losing it."
When water hits the surface, it's quickly absorbed—in milliseconds—rather than bouncing away. It gets wicked into the crack network and is redistributed across the honeycomb-like tile—even against gravity on sloped surfaces. That extended contact keeps the surface wet for longer, allowing evaporation to do its work. The hexagonal geometry forces the water to zigzag laterally rather than drain straight down, which provides sustained cooling for up to 20 hours.
The future of this technology lies in its deceptive simplicity. The team has demonstrated that the DE-cement mixture can be applied to large-area panels using hopper guns, suggesting a path toward on-site fabrication that is both scalable and cost-effective.
"Extreme heat is already the deadliest climate-related hazard, and we've spent decades trying to seal buildings off from the environment," Aviv says. "If we can design materials that work with natural processes instead of against them, we can start to rethink how cities stay livable."
Looking ahead, the researchers envision intelligent control systems wherein watering cycles are automated based on real-time weather data, ensuring a building is "charged" with water only when the forecast demands it.
"Now that we understand how water moves through these patterned tiles, we can start treating water as something we manage very precisely instead of simply spraying more of it," says Yu. "By combining these materials with weather forecasts and automated water delivery systems, we could supply just enough water, exactly where and when it's needed, to maximize cooling while minimizing water use."
Yang recently founded Minerava , a Penn Center for Innovation Venture that seeks to bring carbon-capturing, passive-cooling concrete to modern infrastructure, in partnership with Aviv and others.
Shu Yang is a Joseph Bordogna Professor of Engineering and Applied Science in the Department of Materials Science & Engineering in the School of Engineering and Applied Science.
Dorit Aviv is an associate professor of architecture and the director of the Thermal Architecture Lab at Weitzman School of Design .
Kum-Hao Yu is former postdoctoral researcher in the Yang lab who is now an assistant professor at Syracuse University.
Other authors include and Yunchan Lee of Penn Engineering, Ji Yoon Bae of the Weitzman School, and Qingya Huang of Syracuse University.
This work was supported by the U.S. Department of Energy's Advanced Research Projects Agency–Energy under the HESTIA program (Grant No. DE-AR-0002625) and the National Science Foundation (Grant No. DMR-2309043).
