One Atom Apart: Crystal Wiring Trick Pushes Perovskite Solar Cells Past 27 Percent

Perovskite solar cells have spent more than a decade rewriting the record books, but the most valuable real estate in these devices has always been the interface, the razor-thin boundary where the light-absorbing perovskite layer meets the contacts that siphon away electric charge. Engineers have long known that tuning this boundary, nudging it to preferentially pass electrons on one side and holes on the other, is the key to squeezing out every last fraction of a percent of efficiency. The trouble is that the standard toolkit for doing so relies on extrinsic chemical doping, adding foreign molecules or ions whose benefits often come bundled with instability, diffusion, and long-term degradation. Now a team led by researchers at Sungkyunkwan University in South Korea, working with collaborators at Shaanxi Normal University and the University of Science and Technology of China, has demonstrated a strikingly different approach: instead of doping the interface, they rewire its crystal structure, and the electronics follow.

The strategy, reported in Nature Materials, hinges on a concept the authors call octahedral connectivity. In lead halide perovskites and their low-dimensional relatives, the fundamental building block is the lead iodide octahedron, a cage of six iodide atoms surrounding a single lead atom. How these octahedra link together, corner-sharing in the classic three-dimensional perovskite lattice, or arranged in distinct sharing patterns in one-dimensional chains, determines how electronic orbitals overlap across the crystal. The researchers showed that by choosing between two organic cations that differ by only a single heteroatom, they could direct the formation of one-dimensional organic lead triiodide phases with fundamentally different octahedral-sharing arrangements, and that this structural difference alone was enough to flip the carrier-selective character of the resulting interface.

The two cations in question are imidazoline-based molecules, close chemical cousins whose subtle difference would seem, at first glance, unlikely to matter much. Yet when each cation is combined with lead iodide to form a one-dimensional crystalline phase, single-crystal X-ray analysis reveals that the resulting octahedral frameworks adopt distinct connectivity patterns. Density functional theory calculations, carried out by the team and validated against the experimental structures, show that these different sharing geometries reshape the coupling between lead and iodide orbitals near the band edges. The consequence is a pronounced shift in the work function and in the absolute positions of the valence and conduction band edges, the very parameters that govern whether an interface extracts holes or electrons. In effect, the crystal architecture acts as an intrinsic, built-in selector for charge carriers, no dopant required.

This structure-property link is what elevates the work from a curiosity to a design principle. Because the electronic modulation arises from connectivity rather than from added chemistry, it should be far more robust against the failure modes that plague doped interfaces, where small molecules can migrate, evaporate, or react over time. The researchers leveraged the two one-dimensional phases as complementary charge-selective contacts, deploying one where hole extraction was needed and the other where electron extraction was needed, on opposite sides of the same perovskite absorber. The resulting device architecture, which the team describes as a dual interfacial carrier-selective design, essentially builds the transport selectivity into the perovskite stack itself, using phases that are chemically native to the lead iodide system.

The performance numbers are the kind that make the photovoltaics community sit up. Single-junction perovskite solar cells built with this structural strategy reached a power conversion efficiency of 27.61 percent in the laboratory, with a certified steady-state efficiency of 27.19 percent, figures that place the devices among the very best perovskite cells ever reported. Just as importantly, the gains were not confined to tiny champion cells. When the team scaled the fabrication to modules with an active area of 655 square centimeters, a size relevant to real-world manufacturing, the modules still delivered 22.26 percent efficiency, demonstrating that the interfacial engineering survives the transition from spin-coated laboratory samples to scalable coating processes, which is where many promising perovskite innovations historically have faltered.

Stability, the perennial Achilles heel of perovskite photovoltaics, also fared well. The devices exhibited excellent operational stability under continuous illumination, a result the authors attribute in part to the absence of extrinsic dopants at the interfaces and to the protective character of the one-dimensional phases themselves. Low-dimensional lead iodide phases have previously been shown to wrap perovskite grains and passivate defects, and the imidazoline-based cations used here appear to confer similar robustness, inhibiting ion migration and shielding the underlying three-dimensional absorber. By combining passivation, defect management, and carrier selectivity in a single structurally driven package, the approach addresses several of the loss mechanisms that normally trade off against one another in perovskite device design.

The scientific heart of the paper lies in the mechanistic analysis that connects atomic arrangement to macroscopic device behavior. Using single-crystal diffraction, the team pinned down the exact octahedral-sharing motifs in each one-dimensional phase. Computational analysis of the crystal orbital Hamilton populations then quantified how the different connectivities alter the energy-resolved chemical bonding between lead and iodide, revealing how orbital coupling near the band edges is strengthened or weakened depending on the sharing pattern. Kelvin probe force microscopy measurements mapped the resulting work function shifts at real device interfaces, while grazing-incidence wide-angle X-ray scattering confirmed that the desired one-dimensional phases form conformally on the perovskite films during fabrication. The convergence of structural, computational, and device-level evidence gives the central claim, that connectivity alone reconfigures carrier selectivity, unusual rigor.

What makes the result especially compelling is its economy. The two cations differ by a single heteroatom, yet that one-atom difference cascades through crystal packing, orbital overlap, band alignment, and ultimately the current-voltage characteristics of a complete solar cell. It is a vivid demonstration of how, in hybrid halide systems, the organic component is not a passive spacer but an active electronic participant whose geometry can be exploited as a design variable. The finding also suggests a broader roadmap: rather than searching for ever more exotic dopants and transport layers, researchers may be able to mine the structural phase space of low-dimensional lead halides for intrinsic contacts with tailored band positions, pairing phases the way one currently pairs dedicated electron and hole transport materials.

The implications extend beyond single-junction records. Certified efficiencies approaching 27 percent bring perovskites squarely into competition with the best commercial silicon cells, and the demonstrated scalability to module dimensions addresses the manufacturing question that investors and panel makers care most about. If the dopant-free, structurally programmed interfaces prove durable over thousands of hours of field operation, the approach could ease one of the main remaining barriers to perovskite commercialization. For now, the study stands as an elegant proof that sometimes the most powerful way to engineer a material’s electronics is not to add something to it, but to rearrange what is already there, atom by atom, octahedron by octahedron, until the crystal itself decides which charges to let through.

Subject of Research: Structurally driven reconfiguration of lead iodide octahedral connectivity to create dopant-free carrier-selective interfaces in high-efficiency perovskite solar cells.

Article Title: Octahedral connectivity reconfigures interfacial carrier-selective properties for efficient perovskite solar cells

Article References: Zhang, Y., Liang, Z., Cho, S. C., Cho, S.-H., Fu, G., Chen, X., Liu, B., Lee, S.-U., Seol, J. H., Liu, S., Zhang, H., Zhao, K., Lee, S. U., Pan, X., & Park, N.-G. (2026). Octahedral connectivity reconfigures interfacial carrier-selective properties for efficient perovskite solar cells. Nature Materials. https://doi.org/10.1038/s41563-026-02722-3

Image Credits: AI Generated

DOI: 10.1038/s41563-026-02722-3

Keywords: perovskite solar cells, octahedral connectivity, carrier-selective interfaces, lead iodide, one-dimensional perovskite, power conversion efficiency, work function tuning, band-edge positions, organic cations, dopant-free contacts, operational stability, solar module scalability