Inverted perovskite solar cells check nearly every box for next-generation solar power: low manufacturing costs, compatibility with large-scale production, and efficiency numbers in the lab that rival conventional silicon. Yet something buried deep inside the device — invisible to the naked eye and stubbornly difficult to control — has kept real-world performance from matching that promise.
The problem lives at the bottom of the cell, where the light-absorbing perovskite layer meets the layer beneath it. At that hidden junction, tiny structural flaws accumulate quietly, eroding both efficiency and long-term stability before a panel ever sees the sun.
The hidden flaw inside inverted perovskite solar cells
Inverted perovskite solar cells flip the conventional layer order. The hole transport layer sits at the bottom, the perovskite absorber in the middle, and the electron transport layer on top. This reversed arrangement pairs well with solution-based manufacturing methods, making large-panel production more practical and less expensive.
The trouble starts at the very bottom of that stack. Where the perovskite layer meets the hole transport layer — the buried interface — conditions are difficult to monitor and even harder to control during fabrication. The surface beneath is typically hydrophobic, so liquid precursors do not spread evenly across it.
That uneven spread has real consequences. Microscopic voids form, grain boundaries develop irregular grooves, and electronic defects accumulate at the junction. Every flaw opens a pathway for energy to escape before it becomes electricity, and each one accelerates wear under heat and light. Solving this buried-interface problem is the prerequisite for moving perovskite solar technology from a lab curiosity into a manufacturable product.
Crystal seeds that guide growth from the bottom up
Researchers at the Qingdao Institute of Bioenergy and Bioprocess Technology (QIBEBT), part of the Chinese Academy of Sciences, approached the problem from an unexpected angle. Rather than attempting to repair the buried interface after the perovskite film forms, they set out to shape it before crystallization even begins.
Their method is called crystal-solvate (CSV) pre-seeding. Before depositing the perovskite layer, the team places specially engineered low-dimensional halide nanocrystals — with the chemical formula PDPbI4·DMSO — directly onto the substrate surface. These rod-shaped CSV nanocrystals act as structural guides for everything that grows on top of them.
Two things happen at once. The nanocrystals improve how well the perovskite precursor liquid spreads across the typically water-repelling self-assembled monolayer (SAM) surface, and they serve as numerous nucleation centers — giving growing crystals many starting points, producing faster and more directed crystallization with far fewer voids.
A slow-release solvent trick that stabilizes the film
The CSV nanocrystals carry something additional within their structure: dimethyl sulfoxide (DMSO) molecules locked inside the crystal lattice. That embedded solvent turns out to be central to the technique’s effectiveness.
During thermal annealing — the heating step that finalizes the perovskite film — those DMSO molecules are released gradually rather than all at once. This creates what the researchers describe as a “lattice-confined solvent annealing” environment concentrated right at the buried interface, encouraging grain rearrangement and producing a denser, more uniform film than conventional annealing achieves.
“We have developed an integrated approach that simultaneously addresses crystallization regulation and interface stabilization,” said Dr. Xiuhong Sun, co-first author of the study. “This strategy delivers good performance even at buried interfaces, which are notoriously challenging to precisely control.”
From lab cell to large module — with minimal efficiency loss
The real test of any solar fabrication advance is whether it survives the jump from a small lab cell to something closer to a commercial panel. Many techniques that look promising at small scale fall apart as the area grows. This one held up.
The QIBEBT team combined CSV pre-seeding with slot-die coating — an industrial-scale deposition process — to build a perovskite solar mini-module with an aperture area of 49.91 cm². That module achieved a power conversion efficiency of 23.15%, and the efficiency loss from small lab cells to the larger module was less than 3%, outperforming many previously reported studies.
“This technology overcomes the longstanding scaling bottleneck caused by size effects through the combination of induced crystallization and buried interface restoration,” said Prof. Shuping Pang. The use of slot-die coating signals that the technique was designed with real manufacturing conditions in mind — not just optimized for a controlled lab environment.
A platform beyond solar cells
The researchers frame CSV pre-seeding as more than a targeted fix for a single problem. By tuning the organic cations and solvent molecules within the crystal-solvate structure, a broad library of CSV materials could be designed for different applications. The underlying principle — using pre-deposited crystalline seeds to control buried-interface formation in soft-lattice semiconductors — could extend to LEDs, photodetectors, and other optoelectronic devices facing similar interface challenges.
The findings were published in Nature Synthesis on February 27, 2026. Extended outdoor testing and trials across different device architectures are the likely next steps. If the stability gains hold at scale and over time, this technique could quietly become a standard step in perovskite manufacturing.
Carlos is an engineer with strong expertise in technical and industrial topics. He previously worked at international companies such as Siemens and speaks Spanish, German, English, and Italian.
