Perovskite-silicon tandem solar cells hit 34% in 2026

Efficiency Records and the Race Past 34%

Perovskite-silicon tandem solar cells have surpassed the Shockley-Queisser single-junction theoretical limit of 33.7%, with LONGi Green Energy holding the current NREL-certified world record at 34.85% as of late 2024. This figure represents the culmination of a rapid progression: 33.9% in November 2023, 34.6% in June 2024, and 34.85% by the end of that year — a gain of nearly one full percentage point in twelve months.

The significance of crossing 33.7% cannot be overstated. According to research published by Nature, the tandem architecture captures a broader slice of the solar spectrum by stacking a wide-bandgap perovskite top cell over a silicon bottom cell, enabling energy extraction from photons that a single-junction cell would waste as heat. The theoretical efficiency ceiling for an all-perovskite two-junction stack reaches approximately 47%, meaning there remains substantial headroom above today’s records.

Hanwha Q CELLS has demonstrated 31.6% (certified 30.8%) on industrial Q.ANTUM silicon bottom cells, with 95% efficiency retention after 1,000 hours of maximum power point (MPP) tracking at 25°C — a key stability benchmark for the sector. KAUST’s academic consortium recorded 33.2% on a 1 cm² research cell in April 2024. The gap between small-area cell records and module-level performance remains a defining challenge: Oxford PV’s 26.9% module efficiency on a 60-cell residential format is Fraunhofer CalLab-certified, but it still sits 2–8 percentage points below cell-level benchmarks, a gap that manufacturing scale must close.

Competitive Landscape: Who Leads, Who Follows

The perovskite-silicon tandem competitive field divides into three tiers, each with distinct strategic profiles. LONGi Green Energy and Oxford PV occupy Tier 1 — one holds the efficiency record, the other holds first-mover advantage in commercial production. Aiko Solar, Hanwha Q CELLS, Swift Solar, and the Chinese Tier 1 manufacturers (Trina Solar, JinkoSolar, JA Solar) form Tier 2. Research institutions including Helmholtz-Zentrum Berlin and KAUST anchor Tier 3.

LONGi Green Energy achieved its 34.85% record using a bilayer interface passivation approach — sequential deposition of LiF and EDAI₂ at the perovskite/C₆₀ interface — combined with resistance-increasing nanostructures in carrier transport layers to ensure compatibility with textured silicon. The company’s R&D team of 100+ people dedicated to perovskite is the largest disclosed among Chinese manufacturers, and its progression rate of more than one percentage point of efficiency gain per year is the fastest in the sector. LONGi is also the dual champion in single-junction Si-HJT at 26.81%. Despite this technical leadership, no public mass production timeline has been announced as of March 2026.

Oxford PV occupies a distinct position: it is the first company to operate a commercial-scale perovskite-silicon tandem production line, located in Brandenburg an der Havel, Germany, and has been producing since 2017. Its 60-cell residential module achieved 26.9% efficiency (Fraunhofer CalLab-certified in 2024), and the company is actively searching for a high-volume manufacturing site globally. Oxford PV’s 29.52% cell efficiency (NREL-certified) is lower than LONGi’s record but its module-level performance and IEC reliability testing distinguish it as the most commercially advanced player.

Hanwha Q CELLS demonstrated 31.6% (certified 30.8%) on industrial Q.ANTUM silicon bottom cells using an AlOₓ/PDAI₂ bilayer passivation approach, with 95% efficiency retention after 1,000 hours of MPP tracking — the strongest publicly disclosed stability data in the sector. The company announced a $100 million pilot line investment in 2023. Swift Solar takes a differentiated path entirely, pursuing all-perovskite tandems (not perovskite-silicon) with vapor deposition for flexible, lightweight formats. Backed by exclusive IP licences from MIT, Stanford, and NREL, more than 40 patents, and $60 million in funding (including a $27 million Series A in 2023), Swift Solar targets defence, telecom, space, and vehicle-integrated PV — markets where its partnerships with American Tower Corporation and the US Department of Defense give it a structural advantage.

Among Chinese Tier 1 manufacturers, Aiko Solar leads on patent volume with 12 patents in the dataset — the highest among Chinese manufacturers — focused on interface passivation, carrier transport layer optimisation, and industrial-scale heterojunction integration. Trina Solar, JinkoSolar, and JA Solar each hold 5–10 patents focused on tunnel junction optimisation and module integration. Their strategic posture is that of fast followers: once LONGi or Aiko validate commercial viability, their existing silicon manufacturing scale gives them the capacity to move quickly and drive down cost.

Four Technical Routes Shaping the Perovskite-Silicon Tandem Field

Four distinct technical routes have emerged for perovskite-silicon tandem cells, each with different efficiency ceilings, cost profiles, and manufacturing compatibility. Understanding which route a given player is pursuing is essential for assessing their competitive position and IP exposure.

Route 1: Bilayer Interface Passivation (Dominant Route)

The highest-performing approach uses sequential deposition of an inorganic barrier layer (LiF or AlOₓ) followed by an organic passivator (EDAI₂, PDAI₂, or piperazinium iodide) at the perovskite/C₆₀ interface. This bilayer modulates energy level alignment, reduces trap density, suppresses non-radiative recombination, and blocks ion migration. LONGi’s 34.85% record uses LiF + EDAI₂; Q CELLS’ 31.6% uses AlOₓ deposited by atomic layer deposition (ALD) plus PDAI₂, with island-like structures at grain boundaries enabling nanoscale local contacts. A triple-halide perovskite combined with piperazinium iodide achieved a 32.5% certified efficiency, published in Science in 2023. The performance impact relative to single-layer passivation is +2–3 percentage points of absolute efficiency, with 94–95% efficiency retention after 200–1,000 hours of stability testing.

“Hanwha Q CELLS demonstrated 95% efficiency retention after 1,000 hours of maximum power point tracking — the strongest publicly disclosed stability benchmark in the perovskite-silicon tandem sector.”

Route 2: Simplified Tunnel Junction (Cost-Focused Route)

Aiko Solar, Trina Solar, and Zhuhai Fushan Aixu are pursuing a lower-cost approach that eliminates the separate transparent conductive oxide (TCO) recombination layer, replacing it with heavily-doped amorphous or nanocrystalline silicon (p⁺⁺-a-Si:H / n⁺⁺-a-Si:H) as an integrated tunnel junction. Aiko’s patent CN121751881A introduces single-walled carbon nanotubes (SWCNT) as a tunnelling layer, avoiding the sputtering damage and cost associated with TCO deposition. This route targets the 25–30% efficiency range with lower capital expenditure and faster integration into existing heterojunction (HJT) production lines.

Route 3: TOPCon-Compatible Bottom Cell Architecture

Most high-efficiency demonstrations use silicon heterojunction (Si-HJT) bottom cells, but global silicon manufacturing capacity is shifting toward TOPCon (tunnel oxide passivated contact) technology. TOPCon/perovskite integration is significantly under-developed: only 2–3 patents in the dataset specifically address this combination, including CN117337065A covering a TOPCon/perovskite tandem with tunnel oxide and hydrogenated silicon. A 2023 paper demonstrated poly-SiOₓ carrier-selective passivating contacts enabling 28.1% efficiency in a four-terminal configuration and 23.2% in a two-terminal configuration. The challenge is that TOPCon requires processing temperatures above 700°C for poly-Si deposition, creating a compatibility constraint for perovskite layers.

Route 4: Flexible All-Perovskite Tandems (Niche Route)

Swift Solar’s approach avoids silicon entirely, stacking wide-bandgap and narrow-bandgap perovskite sub-cells deposited by vapor deposition on thin flexible substrates. This enables continuous roll-to-roll manufacturing with deposition times below five minutes and a power-to-weight ratio 30–40% higher than rigid silicon modules. The theoretical efficiency ceiling for an all-perovskite two-junction stack is approximately 47%. Swift Solar projects 28%+ module efficiency and targets markets — defence, space, vehicle-integrated PV, portable power — where weight and flexibility matter more than cost-per-watt.

Patent Surge and Innovation Momentum in Perovskite-Silicon Tandems

Patent activity in the perovskite-silicon tandem sector has accelerated sharply, with 2024 and 2025 together accounting for 68.4% of the 101 patents retrieved in the focused search. The broader landscape is estimated at 300–500 active patent families across China, the US, Europe, and PCT jurisdictions. The average citation count of 28 per patent in the dataset indicates moderate-to-high technical influence, suggesting that filings are substantive rather than defensive padding.

The top three assignees by patent count in the structured dataset are Zhejiang Aiko Solar (12 patents), Zhuhai Fushan Aixu Solar (11 patents), and Shenzhen Black Crystal Optoelectronics (11 patents). This Chinese dominance in patent volume reflects the country’s established silicon manufacturing base and its strategic intent to control the IP stack for the next generation of solar technology. According to WIPO, China has been the leading jurisdiction for solar photovoltaic patent filings for several consecutive years, a pattern that is intensifying in the tandem sub-sector.

European players — notably Helmholtz-Zentrum Berlin and Oxford PV — hold foundational process patents, including nanoimprint lithography for inverted pyramid texturing (Helmholtz-Zentrum Berlin) and multi-junction photovoltaic device architectures (Oxford PV). US innovation is concentrated in next-generation architectures and niche applications, with Swift Solar’s patent portfolio built on exclusive licences from MIT, Stanford, and NREL. The US Department of Energy has supported academic consortia including TEAMUP ($9 million) and the MIT-led ADDEPT programme, reinforcing the US position in upstream research even as manufacturing leadership shifts elsewhere.

White-Space Opportunities and Strategic Entry Angles

Four under-served areas emerge from analysis of the patent landscape and published research, each representing a genuine gap rather than a crowded space where incumbents already hold strong positions.

Large-Area Uniformity and Yield Engineering

Most record efficiencies are achieved on small areas of 1–1.2 cm². Module-level efficiency lags cell-level by 2–8 percentage points due to non-uniform perovskite deposition on textured silicon pyramids, edge recombination, scribing losses, and current mismatch across large-area modules. Multiple patents address “rounded texture,” “inverted pyramids,” and bonding layers to improve perovskite conformality, and edge passivation patents (CN119855369B, CN120614945B) have been filed specifically for module-scale degradation. Entry angles include specialised coating equipment for large-area perovskite deposition (slot-die, blade-coating, spray), laser patterning and low-temperature edge sealing, and in-line quality control using photoluminescence imaging and electroluminescence mapping.

Accelerated Lifetime Testing and Encapsulation

Stability validation is the single largest bottleneck to commercial deployment. The longest publicly reported operational data is approximately 1,000 hours, while field deployment requires 25-year warranties — a gap of 10–20× in validated lifetime. Multiple 2025–2026 patent filings address UV-curing encapsulation, moisture barriers, and ion migration blocking, and academic reviews explicitly identify “lack of standardisation” and “operational stability” as key commercialisation barriers. Suppliers of ultra-low water vapour transmission rate (WVTR) barrier films, automated encapsulation lines compatible with perovskite thermal budgets below 150°C, and predictive degradation modelling tools are all under-served.

TOPCon-Compatible Tandem Stacks

Global silicon manufacturing capacity is shifting from PERC toward TOPCon, yet TOPCon/perovskite tandem integration is represented by only 2–3 patents in the dataset. The fundamental challenge is that TOPCon requires poly-Si deposition above 700°C, which is incompatible with perovskite thermal budgets. Low-temperature TOPCon variants, hybrid TOPCon-HJT architectures, and recombination layer optimisation for the poly-Si/perovskite interface represent an open IP landscape with significant commercial upside as China’s silicon manufacturing base completes its transition to TOPCon.

Four-Terminal (4T) Tandem Retrofit

Most R&D focuses on monolithic two-terminal (2T) tandems, which require precise current matching and a tunnel junction between sub-cells. Four-terminal (4T) tandems are mechanically simpler, avoid current-matching constraints, and enable retrofit of existing silicon module lines. Multiple demonstrations have achieved 28–30% combined efficiency in 4T configurations. Retrofit kits for existing utility-scale silicon plants, semi-transparent perovskite module suppliers for 4T stacking, and optical coupling layers for mechanically-stacked arrays all represent entry points with lower technical risk than competing directly on 2T monolithic efficiency records.

Risks, Milestones, and the Path to Commercial Scale

Five material risks could delay or reshape the perovskite-silicon tandem commercialisation trajectory. Each is grounded in publicly available evidence rather than speculation.

• Stability validation gap: The longest publicly reported operational data is approximately 1,000 hours. Field deployment requires 25-year warranties, demanding either 10–20× longer validation or robust accelerated test correlations that do not yet exist as a standardised protocol.
• Lead regulation: Perovskite films contain lead. Regulatory barriers in the EU and some US states may slow residential adoption, particularly in markets with strict hazardous materials legislation.
• Manufacturing yield: Oxford PV has not publicly disclosed yield rates at its Brandenburg facility. The transition from pilot-scale to GW-scale production may encounter defect modes not visible at lower volumes.
• Silicon roadmap competition: Si-HJT and TOPCon module efficiencies are projected to reach 26–27% by 2025–2026, narrowing the efficiency advantage that justifies the additional cost and complexity of tandem integration.
• China manufacturing overhang: If LONGi, Trina, Jinko, and Aiko simultaneously enter mass production in 2026–2027, rapid cost reduction could commoditise perovskite-silicon tandems before Western players establish manufacturing scale.

According to standards bodies including IEC, there is currently no harmonised accelerated test standard specifically for perovskite-silicon tandem modules — a gap that both regulators and manufacturers are working to address. Oxford PV has tested its modules to existing IEC standards, but the industry acknowledges these were designed for crystalline silicon and may not capture perovskite-specific degradation mechanisms such as ion migration and halide segregation.

The milestone table below summarises the key events to monitor over the next four years, based on publicly stated company plans and technology development trajectories from the source data:

The technology trigger most worth monitoring is whether TOPCon-compatible tandem stacks achieve a certified efficiency above 32% by mid-2027. If that threshold is crossed, it could accelerate Chinese adoption and shift the competitive centre of gravity from HJT-based to TOPCon-based routes — fundamentally reshaping the IP and manufacturing landscape. PatSnap’s innovation intelligence platform tracks certified efficiency announcements, patent filings, and company milestones across the full perovskite solar value chain in real time.

“If Chinese Tier 1 players simultaneously enter mass production in 2026–2027, rapid cost reduction could commoditise perovskite-silicon tandem technology before Western players establish manufacturing scale.”