In 1839, German mineralogist Gustav Rose picked up a strange, lustrous crystal in Russia’s Ural Mountains and named it after the Russian mineralogist Lev Perovski. Nearly two centuries later, that mineral’s synthetic cousins are threatening to upend the solar power industry. A peer-reviewed study published in Nature in early 2025 reports that a flexible perovskite-silicon tandem solar cell has reached 33.6% efficiency, the highest ever recorded for this class of device. For context, the average commercial silicon panel on a rooftop today converts roughly 22% to 23% of sunlight into electricity, according to the European Commission’s Joint Research Centre.
That gap between a lab prototype and a shipping product is where the solar industry’s next multibillion-dollar bet is taking shape. And as of mid-2025, the race to close it has moved from academic curiosity to industrial urgency.
The record and why it matters
The Nature study is notable not just for the headline number but for what the device can do physically. Because the cell is flexible, it can conform to surfaces that rigid silicon wafers cannot: curved building facades, lightweight vehicle roofs, even portable field equipment. The research team provided full device architecture details and independently certified efficiency measurements, making the result the current benchmark tracked by the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL), which maintains the official chart of highest confirmed solar cell efficiencies.
Perovskite-silicon tandems now occupy a growing share of the top entries on that NREL chart. Multiple research groups have pushed laboratory tandem cells past 30%, according to the Joint Research Centre’s 2025 photovoltaics status report. The technology’s trajectory is striking: silicon took decades to climb from early lab records to today’s commercial modules. Perovskite tandems have covered comparable ground in roughly a decade of intensive research.
A separate economic analysis published in Joule by NREL-affiliated researchers quantifies why those efficiency points carry outsized financial weight. The study found that a 2.5 percentage-point absolute gain in tandem module efficiency delivers a cost-per-watt reduction comparable to doubling factory production capacity. In other words, squeezing more power from each cell can substitute for enormous capital spending on new manufacturing lines. The same analysis concluded that tandem modules need to reach at least roughly 25% efficiency to compete on price with conventional silicon panels. At 33.6% in the lab, the technology has cleared that threshold by a wide margin, at least at the cell level.
The commercialization race is already underway
This is no longer a purely academic exercise. Oxford PV, a UK-based company spun out of the University of Oxford, began shipping the first commercial perovskite-silicon tandem solar panels from its factory in Brandenburg, Germany, in late 2024. Those initial modules target the residential market and claim panel-level efficiencies above 24%, a figure that, if sustained in field conditions, would already place them among the highest-performing residential panels available.
They are not alone. South Korea’s Hanwha Qcells has invested heavily in tandem cell research at its innovation centers, and China’s LONGi Green Energy, the world’s largest solar manufacturer by capacity, has published its own perovskite-silicon tandem records. The competitive dynamics are global: European, Asian, and North American players are all positioning for a technology that could redefine panel economics within this decade.
But shipping a first batch of panels and operating a gigawatt-scale factory are very different achievements. No manufacturer has yet published audited production cost data for perovskite-silicon modules at volume. The Joule analysis models cost trajectories, and the Joint Research Centre projects capacity growth, but actual factory-gate pricing remains undisclosed. Until those numbers surface, the economic case for tandems rests on projections rather than industrial proof.
Durability: the make-or-break question
Silicon panels routinely ship with 25-year performance warranties, backed by decades of field data from installations around the world. Perovskite layers are far more fragile. They degrade when exposed to moisture, heat cycling, and prolonged ultraviolet light. Encapsulation techniques have improved significantly, but no peer-reviewed dataset yet documents multi-decade outdoor stability for flexible perovskite-silicon tandems under real-world conditions.
This is the single largest barrier to widespread adoption. A solar panel is a 25- to 30-year financial instrument: developers, banks, and insurers need confidence that the technology will perform over the life of a power purchase agreement. Early accelerated aging tests and short-duration field trials have shown promising results, but they are not substitutes for the kind of long-term reliability data that silicon has accumulated over generations of deployment.
The certification pathway adds another layer of complexity. Perovskite-containing modules must pass the same IEC 61215 and IEC 61730 standards that govern all photovoltaic products, covering electrical safety, mechanical load resistance, and environmental stress testing. Passing those tests is necessary but may not be sufficient: insurers and large-scale buyers may demand additional durability evidence before committing capital.
The lead problem and regulatory unknowns
Most high-efficiency perovskite formulations contain lead, typically in the form of lead halide compounds. The quantities per panel are small, but at the scale of gigawatts of deployment, the cumulative volume raises legitimate environmental and health questions. What happens when a hailstorm cracks a panel? What are the end-of-life recycling requirements?
Researchers are actively developing lead-free alternatives using tin and other metals, but none have matched the efficiency of lead-based formulations. The European Union’s Restriction of Hazardous Substances (RoHS) directive currently grants an exemption for photovoltaic panels, but that exemption is subject to periodic review. No primary statement from the European Commission specifies how perovskite-specific lead content will be treated in future regulatory cycles.
In the United States, the regulatory picture is similarly unresolved. The Department of Energy has funded perovskite research extensively through NREL and other national laboratories, but federal agencies have not yet issued guidance specific to the certification, deployment, or disposal of perovskite-containing modules. For manufacturers planning factory investments today, that ambiguity is a real cost: it widens the range of scenarios they must plan for and slows procurement decisions by risk-averse utilities.
What separates this from previous “next big things”
Solar energy has seen no shortage of promising alternatives to silicon over the years. Cadmium telluride carved out a niche through First Solar’s manufacturing scale. Gallium arsenide cells power satellites but remain too expensive for terrestrial use. Organic photovoltaics have lingered in the lab for decades without breaking through commercially.
Perovskite-silicon tandems are different in one critical respect: they do not ask the industry to abandon silicon. Instead, they layer a perovskite film on top of an existing silicon cell, capturing wavelengths of light that silicon handles poorly. This tandem architecture means manufacturers can, in theory, retrofit or upgrade existing silicon production lines rather than building entirely new factories from scratch. That compatibility with the incumbent technology is a structural advantage that previous challengers lacked.
The speed of progress also sets this technology apart. The first perovskite solar cell was reported in 2009 with an efficiency below 4%. By 2025, tandem devices have surpassed 33%. No other photovoltaic material has climbed that fast. Whether that laboratory momentum can survive contact with factory floors, supply chains, and weather is the open question, but the physics is no longer in doubt.
Where the decisive tests will happen next
For solar developers, utilities, and building owners weighing technology choices over the next several years, the practical signal from the current evidence is clear: perovskite-silicon tandems have crossed the efficiency threshold that credible economic modeling identifies as the minimum for cost competitiveness. The barriers that remain are engineering, manufacturing, and regulatory challenges, not fundamental physics.
That distinction matters enormously. It means the experiments that will determine perovskite’s fate will not take place only in university cleanrooms. They will play out in Oxford PV’s Brandenburg factory, in Hanwha Qcells’ pilot lines, in IEC certification labs, and in the procurement offices of utilities deciding which panels to specify for projects that will operate into the 2050s.
If tandem modules can demonstrate the kind of multi-decade durability that silicon has proven, and if manufacturers can achieve repeatable high-yield production at competitive cost, the 33.6% flexible cell record published in Nature will be remembered as the moment the technology’s ceiling visibly lifted. If they cannot, silicon will continue its reign, absorbing incremental gains from other innovations as it has for half a century. Either way, the next two to three years of factory data, field trials, and regulatory decisions will tell us which path the industry is on. The lab results are in. The industrial verdict is not.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
