Modified perovskite cells retain 84% efficiency after extreme thermal cycling

Perovskite Solar Cells Are Finally Ready for Orbit. The Infrastructure Is Being Built Right Now.

The thermal cycling problem that has kept perovskite solar cells out of space applications has a new solution from a lab in Munich. But the more consequential development is happening in South Korea, where Flexell Space and Kongsberg NanoAvionics announced Wednesday they are building solar arrays with perovskite and copper indium gallium selenide cells for K-LEO — South Korea's sovereign national security constellation. Perovskite space solar is no longer a lab result. It is a supply chain decision being made right now.

The Aydin Group at Ludwig Maximilian University published results in Nature Communications this month showing that a dual molecular reinforcement strategy — incorporating α-lipoic acid into the perovskite layer to form a polymer network at grain boundaries, and reinforcing the electrode interface with a dimethylsulfonium-lipoic acid molecule called DMSLA — can significantly reduce the thermal cycling damage that has historically plagued perovskite cells in orbit. After 16 cycles between -80°C and +80°C, the modified cells retained 84% of their initial efficiency, compared to a steeper drop in control cells. Optimized cells hit 26% efficiency, about 3% higher than the control group.

The physics is straightforward: LEO satellites experience extreme temperature swings every 90 minutes as they move in and out of sunlight, ranging roughly from -150°C in Earth's shadow to +120°C in direct sun. Silicon and gallium arsenide handle this through decades of engineering maturity. Perovskite — lightweight, cheap, and highly efficient in the lab — has struggled with mechanical stability under that stress. The grain boundaries and electrode interfaces are where it breaks down. The Aydin team's contribution is a targeted reinforcement of exactly those weak points.

But the commercial side is moving faster than the academic literature suggests. Flexell and NanoAvionics are not waiting for perfect lab results. Their collaboration for K-LEO — announced March 18, with NanoAvionics designing and manufacturing the arrays, Flexell handling quality assurance — explicitly targets integration of next-generation cells into operational satellite platforms. Hanwha Systems is building the satellites. The K-LEO constellation is a Korean national security program, which means it has budget and urgency.

The broader commercial landscape is worth noting. Swift Solar — backed by 42.4 million dollars in funding and with DoD interest in space-based energy applications — acquired Meyer Burger's domestic manufacturing assets this month to accelerate tandem silicon-perovskite cell production in the US. Meyer Burger's bankruptcy earlier this year created an opening; Swift picked up the IP and is using it to pursue gigawatt-scale domestic manufacturing. Meanwhile, samples of perovskite cells spent 10 months on the International Space Station in 2020-2021 as part of exposure testing, with results published in 2026 showing partial but instructive degradation.

The satellite solar array market has historically been dominated by triple-junction gallium arsenide cells — expensive, heavy, and highly efficient. Perovskite offers a meaningful cost and weight reduction at competitive efficiency levels, if the durability problem is solved. The Aydin results suggest it is being solved. The Flexell/NanoAvionics K-LEO deal suggests the industry has already concluded it is solved enough.

The question for investors and builders is timeline. Lab results to flight hardware is a multi-year path with qualification testing, radiation hardness assurance, and legacy system redesign. The K-LEO arrays are in development now. Swift Solar's US manufacturing scale-up is in progress. The next two to three years will show whether perovskite cells can survive the full stress profile of operational LEO — not just 16 thermal cycles in a thermal vacuum chamber, but years of radiation damage, atomic oxygen exposure, and micrometeoroid impacts. The Aydin paper is a data point in that ongoing experiment, not a conclusion.