Skip to content
ContentLora

    Tip: press / anywhere to search.

    Analysis

    Perovskite solar: ready for rooftops, or still a lab record?

    Perovskite-silicon tandem cells reached a certified 35.5% efficiency in July 2026 and first commercial modules shipped in 2024, yet DOE says perovskite PV is not yet manufactured at scale.[1][2][3] The core dispute is durability: panels need decades of life, and a 2026 study found that standard accelerated tests can misjudge how perovskites age outdoors.[4][5]

    Editor reviewedUpdated Materials scienceScience
    Show:

    The efficiency case

    LONGi’s perovskite-silicon tandem record rose from 33.9% in November 2023 to a certified 35.5% in July 2026.[6][1] That is above the roughly 33.7% limit for any single-junction cell; the tandem’s theoretical limit is up to 43%.[7] On larger cells LONGi reported 34.3% at 261 cm² and 32.2% at 274 cm².[8] Oxford PV shipped 24.5% efficient tandem modules to a US utility-scale project in September 2024 and said they can produce up to 20% more than standard silicon panels.[2] It said it planned a gigawatt-scale factory for 2026-27 using the same process as its pilot line.[9] Small research perovskite cells on their own exceed 26%.[10]

    The durability case

    The US Department of Energy lists the main barriers as stability and durability, efficiency at scale, manufacturability and technology validation, and says perovskite PV is not yet manufactured at scale.[3] Perovskite cells ideally need to support decades-long warranties.[4]

    A June 2026 study in Joule, by Helmholtz-Zentrum Berlin and HTW Berlin researchers, compared cells aged 20 months outdoors with lab-aged cells. Standard high-temperature tests (65-85 °C) caused degradation not seen outdoors. Raising light intensity to 2.3 suns reproduced all three outdoor degradation mechanisms and their spatial patterns, but the authors said it is still not precise enough for lifetime prediction.[5]

    The efficiency argument is largely settled: tandems clearly beat single-junction silicon in the lab, and the certified results come from independent test centres.[1][7] The unresolved question is lifetime. If the industry’s standard accelerated tests do not reproduce outdoor ageing, neither optimistic nor pessimistic lifetime claims can be fully trusted yet. That favours the “measure better first” view as the near-term priority.[5] The study’s finding that 2.3-sun illumination preserves outdoor degradation patterns offers a faster screening tool, but its authors stopped short of calling it a lifetime predictor.[5]

    The gap between record cells (35.5%) and the first commercial modules (24.5%) is also large. Part of it reflects normal cell-to-module losses, but it means headline records overstate what buyers can get today.[1][2]

    What to watch

    Over the next 12-18 months, expect further tandem records edging toward 36%, more large-area results and additional pilot-line announcements. The decisive evidence will be multi-year field data and agreed accelerated-test protocols, not efficiency records. If field data over several years show low degradation, the “ready to scale” view gains ground. If new failure modes appear outdoors, the sceptical view strengthens. Confidence: moderate on records continuing, low on durability being settled within that period.[6][5]

    Industry signals

    Keeping samples at 65-85 °C is a very common way to speed up perovskite ageing in the lab. That is the test the June 2026 Joule study found can produce degradation not seen outdoors.[11][5] Mechanical durability is a separate question. A certified 33.6% flexible tandem published in Nature kept 91% of its efficiency after 5,000 bending cycles.[12] Large manufacturers are paying for access to the technology. Trinasolar took an exclusive Chinese licence to Oxford PV’s patents in April 2025, and First Solar a non-exclusive US licence in February 2026; both are company announcements.[13][14]

    Licences show that manufacturers expect perovskites to matter. They do not show that modules will last 25 years or more. Until field data accumulate, efficiency records and licensing deals are signs of intent, not evidence of durability.

    Competing views

    Ready to scale

    Records keep rising, large-area cells exceed 32%, and commercial modules have shipped, so manufacturing scale-up is the main remaining task.[1][8][2]

    Not proven yet

    Stability and durability top DOE's list of barriers, perovskite PV is not made at scale, and decades-long warranties have not been demonstrated.[3][4]

    Measure better first

    The bottleneck is lifetime prediction itself: common hot accelerated tests trigger degradation not seen outdoors, so better test protocols are needed before bankable claims.[5]

    Questions readers ask

    Are perovskite solar panels commercially available?

    In small volumes. Oxford PV shipped the first commercial perovskite-silicon tandem modules to a US utility-scale project in September 2024, but DOE says the technology is not yet manufactured at scale.[2][3]

    What is the main obstacle for perovskite solar?

    Stability and durability. Panels ideally need to support decades-long warranties, and DOE lists durability as the first challenge.[3][4]

    Why is it hard to know how long perovskite cells last?

    A 2026 study found that standard 65-85 °C accelerated tests caused degradation not seen in 20 months outdoors, while raising light intensity to 2.3 suns reproduced outdoor patterns better.[5]

    Sources

    Each numbered claim is a statement we checked against the sources listed with it. Status shows how well established it is.

    1. [1]

      On 14 July 2026 LONGi announced a crystalline silicon-perovskite tandem cell with 35.5% efficiency, certified by the European Solar Test Installation (ESTI), a new world record. confirmedas of 2026-10-10

    2. [2]

      In September 2024 Oxford PV shipped what it called the first commercial perovskite-on-silicon tandem modules, 24.5% efficient, to a US utility-scale project, from its pilot line in Brandenburg an der Havel, Germany; it said the modules can produce up to 20% more than a standard silicon panel. confirmedas of 2026-10-10

    3. [3]

      DOE lists the main perovskite challenges as cell stability and durability, efficiency at scale, manufacturability and technology validation, and says perovskite PV is not yet manufactured at scale. confirmedas of 2026-10-10

    4. [4]

      Perovskite solar cells ideally need to support warranties lasting decades. confirmedas of 2026-10-10

    5. [5]

      A June 2026 Joule study comparing perovskite cells aged 20 months outdoors with lab-aged cells found that standard high-temperature tests (65-85 °C) caused degradation not seen outdoors, while raising light intensity to 2.3 suns reproduced the outdoor degradation patterns. confirmedas of 2026-10-10

    6. [6]

      LONGi's tandem record rose from 33.9% in November 2023 to 34.6% in June 2024, then through 34.85% and 35.2% to 35.5% in 2026. confirmedas of 2026-10-10

    7. [7]

      Silicon-perovskite tandems have a theoretical efficiency limit of up to 43%, above the roughly 33.7% Shockley-Queisser limit for single-junction cells. confirmedas of 2026-10-10

    8. [8]

      LONGi also reported tandem efficiencies of 34.3% on a 261 cm² cell and 32.2% on a 274 cm² cell. confirmedas of 2026-10-10

    9. [9]

      In September 2024 Oxford PV said it planned a gigawatt-scale manufacturing facility by 2026-27, using the same process as its Brandenburg pilot line. reportedas of 2024-09-05· forecast

    10. [10]

      Small perovskite solar cells went from about 3% efficiency in 2009 to over 26%, according to the US Department of Energy. confirmedas of 2026-10-10

    11. [11]

      Keeping samples at elevated temperatures of 65-85 °C is a very common way to accelerate perovskite ageing in the lab. confirmedas of 2026-06-25

    12. [12]

      A Nature paper published online in November 2025 reported a certified 33.6%-efficient flexible perovskite/silicon tandem cell that kept 91% of its initial efficiency after 5,000 bending cycles. confirmedas of 2025-11-10

    13. [13]

      In April 2025 Oxford PV signed an exclusive patent licence letting Trinasolar make and sell perovskite-based PV products in China, with the right to sublicense (company-reported). confirmedas of 2025-04-09

    14. [14]

      In February 2026 First Solar took a non-exclusive licence to Oxford PV's perovskite patents for potential US products, excluding crystalline silicon, and said it runs a perovskite development line in Perrysburg, Ohio (company-reported). confirmedas of 2026-02-24

    Revision history (2)
    1. Page created.
    2. Added standard heat-test practice, flexible-tandem bending stability and new licensing deals as signals of industry commitment.

    Created Oct 10, 2026. Last reviewed by an editor on Oct 10, 2026. Next scheduled review: Jan 10, 2027.

    Cite this page

    "Perovskite solar: ready for rooftops, or still a lab record?." ContentLora, updated Oct 10, 2026. https://contentlora.com/analysis/perovskite-durability-debate

    Spotted an error? Suggest a correction or emailcorrections@contentlora.com.

    Keep exploring

    ☰ Contents