Why is interfacial recombination the #1 efficiency bottleneck?
The biggest single factor limiting efficiency in perovskite-silicon tandems is unwanted electron-hole recombination at the interface between the perovskite top cell and the electron transport layer (ETL). This recombination wastes energy that should become electrical current. A 2024 study in Nature [3] showed that a clever bilayer passivation strategy — a nanoscale lithium fluoride layer followed by a diammonium diiodide molecule — suppressed this non-radiative recombination while maintaining excellent charge extraction. The result was a certified 33.89% efficiency, the first tandem to exceed the single-junction Shockley-Queisser limit of 33.7%. The key evidence gap is that this bilayer approach is still complex and not yet proven on textured silicon at scale; other studies [7] confirm that non-radiative recombination at perovskite interfaces remains a major limitation even in high-performing devices.
A separate 2023 study [4] tackled a related interface problem: the self-assembled monolayer (SAM) used to connect the perovskite top cell to the silicon bottom cell often has uneven coverage on textured silicon, causing energy losses. They solved this by using an ultrathin (5 nm) amorphous indium zinc oxide (IZO) layer, which provided more uniform SAM anchoring sites, boosting certified efficiency to 32.5%. This shows that interface engineering is not a single problem — it involves multiple interfaces (perovskite/ETL, perovskite/SAM, SAM/TCO) — and each one has its own evidence gap regarding how to make it work reliably on textured, industrially relevant silicon.
Why can't we just coat textured silicon with perovskite and call it done?
Textured silicon wafers — the standard in industry because they trap light better — are notoriously difficult to coat uniformly with perovskite. The pyramid-shaped texture causes uneven film growth, leading to defects and lower efficiency. A 2023 study [2] developed an 'anion-engineered additive' strategy to control perovskite crystallization on fully textured silicon, achieving 28.6% efficiency on 1 cm² cells (certified 27.9%) and 25.1% on a 16 cm² mini-module. However, the same study notes that the crystal quality of perovskite on textured silicon is still inferior to that on flat substrates, and the additive strategy is not yet proven at production scale. A 2022 review [6] explicitly calls out the challenge of upscaling from lab-scale (typically <1 cm²) to industry-relevant areas (>100 cm²) as a key evidence gap, with no study here demonstrating a >25% efficient tandem on a full-size (M2 or larger) wafer.
A 2022 Nature Nanotechnology study [1] took a different approach: instead of using standard random pyramid texture, they designed gentle sinusoidal nanotextures that allowed high-quality perovskite film growth while still reducing reflection losses. This improved fabrication yield from 50% to 95% and achieved a certified 29.80% efficiency. The evidence gap here is that these custom nanotextures are not yet compatible with existing industrial silicon texturing lines, so the manufacturing cost and throughput are unknown. The 2024 Science review [5] emphasizes that transitioning from lab to fab requires not just high efficiency but also scalable deposition methods (e.g., slot-die coating, evaporation) that work on textured silicon — and the evidence for these methods at high efficiency is still thin.
How long do these tandems actually last?
Stability is the most daunting evidence gap for commercialization. Silicon solar panels routinely last 25-30 years; perovskite-silicon tandems are not even close. The best stability data in these papers comes from [2], which showed encapsulated tandems retaining over 80% of initial performance after 2000 hours of continuous operation under full 1-sun illumination in ambient conditions. That's about 83 days — impressive for a perovskite device, but orders of magnitude short of the 25-year target. A 2023 review [8] identifies ion migration within the perovskite as a fundamental instability mechanism and proposes eliminating it as a 'cornerstone strategy,' but no study here demonstrates a tandem that passes the industry-standard damp-heat or thermal-cycling tests (IEC 61215).
The 2024 Science review [5] and the 2022 review [6] both stress that stability testing needs to move beyond lab conditions (constant 1-sun, 25°C) to real-world field performance, including temperature cycles, humidity, and UV exposure. The evidence gap is stark: we have no published data from any of these studies showing a tandem operating stably for more than a few thousand hours, let alone years. Until that gap is closed, the technology cannot be considered commercially viable, regardless of efficiency records.
About These Sources
This answer is built on 8 peer-reviewed studies — published from 2022 to 2024, 2 from 2024 or later, 6 in Q1 journals, collectively cited 1,749 times — selected as the most relevant from 8 studies that passed quality screening, drawn from 45 papers retrieved from a database of over 500 million.
Sources used in this answer
Nano-optical designs for high-efficiency monolithic perovskite–silicon tandem solar cells
Designed gentle sinusoidal nanotextures on silicon that improved perovskite film quality and fabrication yield from 50% to 95%, achieving a certified 29.80% efficiency — but these custom textures are not yet compatible with standard industrial silicon texturing.
Efficient Perovskite/Silicon Tandem Solar Cells on Industrially Compatible Textured Silicon
Developed an anion-engineered additive strategy to control perovskite crystallization on industrially textured silicon, achieving 28.6% efficiency (certified 27.9%) on 1 cm² and 25.1% on 16 cm²; encapsulated tandems retained >80% of initial performance after 2000 hours of continuous operation.
Perovskite/silicon tandem solar cells with bilayer interface passivation
Introduced a bilayer interface passivation (lithium fluoride + diammonium diiodide) that suppressed non-radiative recombination, enabling a certified 33.89% efficiency — the first tandem to exceed the single-junction Shockley-Queisser limit of 33.7%.
Enhanced optoelectronic coupling for perovskite/silicon tandem solar cells
Used ultrathin (5 nm) amorphous indium zinc oxide as the interconnecting transparent conductive oxide to achieve uniform self-assembled monolayer coverage on textured silicon, resulting in a certified 32.5% efficiency and an 80% reduction in indium consumption.
Pathways toward commercial perovskite/silicon tandem photovoltaics
A comprehensive review identifying that transitioning from lab to fab requires scalable input materials, manufacturing processes, and a focus on stability, reliability, cell-to-module integration, and accurate field-performance prediction — areas where evidence is currently lacking.
Monolithic Perovskite‐Silicon Tandem Solar Cells: From the Lab to Fab?
A review highlighting key challenges for industrial implementation: upscaling device areas to industry-relevant sizes, achieving >32% efficiency on small cells, and addressing stability as the most critical factor for economic viability.
Maximizing Current Density in Monolithic Perovskite Silicon Tandem Solar Cells
Achieved a current-matched tandem with a certified short-circuit current density of 19.6 mA/cm² and 26.8% efficiency, using optical simulation to guide design; identified non-radiative recombination at perovskite interfaces as a remaining limitation.
Recent Progress in Perovskite Tandem Solar Cells
A review noting that while perovskite/silicon tandems have reached 32.5% certified efficiency, instability and difficulties in large-area realization remain major commercialization challenges; proposes eliminating ion migration as a key strategy for intrinsic stability.