Researchers at Nanjing University and the Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, have reported a two-terminal perovskite–organic tandem solar cell with a power conversion efficiency of 27.35%, and a certified efficiency of 26.88%. The work is described in Joule under the title “Achieving 27.35% efficiency in perovskite-organic tandem solar cells by improving near-infrared absorption with a low-band-gap acceptor”, and was reported by Perovskite-Info on 7 September and by pv magazine on 11 September.

The certified figure is the one that counts. pv magazine sets it against the 26.4% that the Solar Energy Research Institute of Singapore (SERIS) reported in June 2025 for the same tandem configuration, and calls the new result record-breaking. Neither report names the laboratory that certified the 26.88% figure, and we have not read the full paper, so we cannot yet say whether the result will appear in a recognised efficiency table.

The bottleneck: an underfed bottom cell

In a perovskite–organic tandem, a wide-band-gap perovskite top cell takes the visible part of the spectrum and a low-band-gap organic subcell beneath it collects the near-infrared. The two are connected in series, so the stack can carry no more current than its weaker subcell. According to Perovskite-Info’s summary of the paper, the limit has been the organic subcell’s relatively low photocurrent, which arises largely because its external quantum efficiency spectrum overlaps with the perovskite layer’s near the perovskite absorption edge. In effect, the two layers compete for the same photons.

The Nanjing group’s answer is a new non-fullerene acceptor, Zh-F, built around an electron-rich heteroheptacene core with partially fluorinated side chains. It has an optical band gap of 1.23 eV and absorbs beyond 1,000 nm. Rather than replace the established PM6:BTP-eC9 blend, the team added Zh-F as a third component, making a ternary PM6:BTP-eC9:Zh-F organic subcell, and paired it with a 1.82 eV perovskite top cell.

What the numbers say

These figures are as reported by pv magazine and Perovskite-Info:

  • Tandem: 27.35% efficiency, with an open-circuit voltage of 2.16 V, a short-circuit current density of 15.35 mA/cm² and a fill factor of 82.39%. A reference tandem reached 25.69%.
  • Organic subcell: 19.84% efficiency, 0.853 V open-circuit voltage, 29.00 mA/cm² and a voltage loss of 0.510 V.
  • Perovskite subcell: 19.45% efficiency, 1.32 V open-circuit voltage, 17.63 mA/cm² and an 83.65% fill factor.
  • Stability: an unencapsulated device kept 80% of its initial performance after 744 hours of continuous 1-sun illumination, according to pv magazine.

The gap between the reference device and the Zh-F tandem, 1.66 percentage points on the uncertified figures, is the clearest measure of what the new acceptor adds. We could not find the active area of the record device in either report. Area matters: a small laboratory cell says little on its own about how the approach scales to modules.

The two layers were competing for the same photons. Zh-F moves the organic cell’s absorption further into the infrared.

Why it matters for organic electronics

Perovskite–silicon tandems dominate discussion of tandem photovoltaics, but an organic bottom cell offers something silicon does not: both layers can be deposited from solution, which in principle opens the door to flexible substrates and printed production. That is why the result matters beyond the record. Our August OPV roundup looked at indoor energy harvesting as organic photovoltaics’ nearest market. Tandem work like this is the other track, and it keeps organic absorbers competitive on outdoor efficiency.

What remains unproven is just as clear. 744 hours of unencapsulated operation under continuous illumination is an encouraging laboratory result, but it is a single stress condition. It falls far short of the lifetimes commercial photovoltaics are sold against, and the stability of perovskite top cells remains the open question for every perovskite tandem.

What we’re watching

  • Whether the 26.88% result, and the laboratory that certified it, appear in the next published solar cell efficiency tables.
  • Stability data for encapsulated Zh-F devices under damp heat and thermal cycling, not only continuous illumination.
  • Any demonstration of the ternary blend on flexible substrates or in larger-area devices, which would test whether the gain survives outside a record-scale cell.