Salt Mixture Helps Grow Cleaner Perovskite Crystals for Solar Cells

A pinch of chemistry borrowed from winter roads could make perovskite crystal growth far more forgiving. Researchers in South Korea and the US have shown that adding two salts to perovskite ink lets the boundaries between crystals melt briefly and regrow with fewer defects. Lab solar cells made this way reached 26.5% efficiency.

The study, published in Science on 8 October, is three weeks old. What is new is a 29 October release from Argonne National Laboratory, whose X-ray scientists helped show how the trick works. It matters because defects at grain boundaries are one of the main reasons perovskite cells lose power and wear out faster than silicon.

How the salt trick improves perovskite crystal growth

Perovskite films are made of many tiny crystals packed together. The seams between them, called grain boundaries, can trap charge and waste energy. The research team added zinc bromide and methylammonium chloride to the solution used to make the film. Together the two salts form a eutectic, a mixture that melts at a lower temperature than either ingredient alone.

“Adding salt lowers the melting point of ice, causing it to melt,” said Yanqi Luo, a beamline scientist at Argonne and a study author. “Our approach behaves the same way.”

According to UC San Diego, the eutectic acts as a temporary liquid around the grain boundaries while the film is heated, giving crystals room to rearrange. The methylammonium chloride then evaporates. The zinc stays concentrated at the boundaries, where it helps neutralise electronic defects, a process known as passivation.

The paper’s abstract says nanoprobe X-ray microscopy “revealed local melting of the eutectic at grain boundaries during annealing that reduced intragranular structural defect density”. Argonne says its Advanced Photon Source and Center for Nanoscale Materials carried out much of that analysis, using snapshots taken at different stages rather than a continuous view. Those snapshots showed zinc atoms drifting towards grain boundaries as crystallisation continued.

“Defects have been the bottleneck for perovskites for almost a decade,” Luo said. “If we can remove the major defects, we can finally start building devices that last longer and perform more reliably.”

The work was a joint effort. The first author, Jason J. Yoo, is at the Korea Research Institute of Chemical Technology (KRICT). The team was co-led from UC San Diego by David Fenning, and the last author, Seong Sik Shin, is at Sungkyunkwan University. Argonne’s role was to help develop and explain the mechanism, not to invent it alone. The paper lists further co-authors at Purdue, Arizona State, Brookhaven, Ajou University, Korea University and MIT.

Perovskite crystal growth results in the lab

The abstract gives the numbers. Solar cells made with the method reached 26.5% power conversion efficiency, with 25.9% independently certified. Minimodules of more than 15 square centimetres reached 23.9%.

For durability, small cells using a polymer hole transporter “retained 85.8% of their initial 24.1% PCE after 1400 hours of continuous operation at 85°C”. The same films, used as light-emitting diodes, reached 26.1% external quantum efficiency.

Connor Dolan, co-first author from Fenning’s lab, said the treated perovskites “had the longest lasting electron-hole pairs ever reported for any direct-bandgap semiconductor”. UC San Diego says the approach worked across the different perovskite chemistries the team tested, and that the perovskite crystal growth gave larger, better-ordered grains with a more even spread of chemical properties.

The paper’s competing-interests statement says the authors have filed a patent application related to the work. No company has announced plans to use the method, and none of the devices described is on sale.

Heat stability matters more than the headline efficiency

Efficiency records grab attention, but they are not what holds perovskites back. The field already has cells in the mid-20s. The harder problem is keeping that output for years on a hot roof, and on that measure this result is more interesting than the 26.5% figure.

The stability test was run at 85C, and the cells held most of their output for 1,400 hours of continuous operation. That is still only about two months, on small cells, with one particular hole-transport layer. Real panels need to last decades. Like recent work on tandem cells that keep 90% at 65C, it is a step towards durability rather than proof of it.

The other test is scale. The method still worked on minimodules of more than 15 square centimetres, and Argonne’s Tao Zhou framed large-area uniformity as “the key requirement for real-world manufacturing”. But that is a long way from the 750 cm² flexible modules reported elsewhere, or the factory output Ricoh is planning for inkjet-printed perovskite cells.

The appeal for manufacturers is that this is a change to the ink and the heating step, not a new production line. Defect control is also a theme in work such as the Oxford-led bismuth sulfide study, which tackled charge loss through the absorber itself. If the salt method can be added cheaply to existing processes, it could become one of several tools for closing the gap between lab cells and panels.

The next step, UC San Diego says, is to test other eutectic chemistries. Argonne adds that its upgraded X-ray source should let researchers watch crystals form continuously rather than in snapshots. Watch for longer stability tests and larger modules before treating this perovskite crystal growth method as a product story.

Sources

• Science: Yoo et al., Volatile eutectics to tailor crystallization for perovskite optoelectronics, 394(6820):228-233 (8 Oct 2026)
• Argonne National Laboratory: Argonne improves perovskite crystal growth for next-generation optoelectronics (29 Oct 2026)
• UC San Diego: A temporary melting liquid could help build better perovskite solar cells and LEDs (8 Oct 2026)

Related stories

• Flexible Perovskite Solar Modules Reach 20.2% at Commercial Size
• Tandem Solar Cells Keep 90% Efficiency After 1,000 Hours at 65C
• Ricoh to Mass-Produce Inkjet-Printed Perovskite Solar Cells in Miyagi