Perovskite solar cells have carried the industry’s heaviest expectations for more than a decade, and the same flaw has stalled them every time: the material degrades under light, heat and moisture far faster than conventional silicon. A research team co-led by the University of California San Diego, working alongside Princeton University and Sungkyunkwan University in South Korea, has now published a fabrication technique in Science that attacks efficiency and durability simultaneously, pushing device performance to 26% while retaining 95% of initial efficiency after eight weeks of continuous simulated sunlight.
- Dual gain: The method improves both power conversion and long-term stability in the same process step.
- The mechanism: Zinc bromide and methylammonium chloride form a temporary liquid phase around growing crystals, then leave behind defect-neutralising zinc.
- The numbers: Devices reached 24% to 26% efficiency and held most of their output across eight weeks of testing.
- The catch: The results come from 1 cm² laboratory cells, leaving the leap to full-size commercial modules unresolved.
A temporary liquid that grows better crystals
The innovation sits inside the crystallisation step. The team added zinc bromide and methylammonium chloride to the solution used to produce a halide perovskite. When heated, the additives formed a low-melting eutectic mixture, creating a temporary liquid layer around the boundaries between growing crystals. That fluid window allowed the grains to rearrange into a more orderly lattice, cutting down the structural defects that bleed energy and accelerate decay.
As heating continued, the methylammonium chloride evaporated, leaving zinc concentrated along the crystal boundaries, where it neutralised the electronic traps that would otherwise interfere with charge movement. Advanced X-ray imaging let the researchers watch the transformation at extremely fine scales: boundary regions turned disordered and liquid-like during heating, then returned to an ordered state as the surplus salt boiled off and crystal growth stopped. The treatment yielded larger crystals, fewer structural defects and a more uniform chemical landscape.
We introduce a new way of tuning the growth of halide perovskites by developing an additive mixture that actively evolves during crystallization. What’s even cooler is that the end product removes performance-limiting defects at the edges of the crystalline grains.
That assessment comes from study co-first author Connor Dolan, a chemical and nano engineering PhD alumnus from the lab of David Fenning, professor of chemical and nano engineering at UC San Diego, who led the work. The team also tested the approach across several perovskite compositions and reported that it worked with the chemistries examined, while noting that the same crystallisation control improved the performance of light-emitting diodes built from the material.
Perovskites made with the eutectic-forming additives had the longest-lasting electron-hole pairs ever reported for any direct-bandgap semiconductor, which speaks to the promise of this approach.
What 26% means next to silicon
Context matters here. Monocrystalline silicon cells that dominate commercial production typically deliver 20% to 23% efficiency in standard modules, while the best premium panels from manufacturers such as SunPower or LG reach 22% to 23%. A perovskite cell holding steady at 26% in industrial production, with the stability demonstrated in this study, would place the material in a clear technical lead rather than a laboratory curiosity.
The durability figure is arguably the more consequential result. Perovskite has long been dismissed as a material that performs beautifully on day one and fades within months. Eight weeks of continuous illumination with 95% retention is not a commercial warranty, but it is a meaningful step past the failure mode that has kept the technology off rooftops. Add the fabrication economics, in which perovskite films can be deposited from solution much like printing, and the case strengthens against silicon’s high-temperature, high-vacuum manufacturing.
From a one-centimetre cell to a rooftop
Scaling remains the open engineering problem. The temporary scaffolding process works on laboratory cells measuring 1 cm², and reproducing that uniformity across commercial modules of roughly 1.6 m² is where promising results usually stall. Encouragingly, the technique demands no exotic equipment or vacuum conditions, unlike several other perovskite enhancement routes, which could shorten the usual path from laboratory to factory.
The companies already building production lines point to the likely timeline. Oxford PV in the United Kingdom, Saule Technologies in Poland and Tandem PV in the United States are all working on silicon-perovskite tandem modules that stack the two materials to combine their strengths. The realistic horizon for competitively priced commercial perovskite modules sits somewhere between 2028 and 2030.
None of this makes silicon obsolete next year, and perovskite has burned expectations before. What makes this result notable is that it targets the two variables that actually decide whether a solar material reaches the market, rather than optimising one at the expense of the other. If the crystallisation control survives scale-up, the economics of distributed generation tilt further toward a material that has spent years being described as almost ready. It may finally be getting closer to genuinely being so.