A commercially sized tandem solar cell drops indium

TL;DR

• Researchers replaced indium-containing transparent conductive oxides with reactive-plasma-deposited tin oxide (RPD-SnOₓ) in a perovskite–silicon tandem and reported 31.0% certified efficiency at 207.9 cm².
• The same paper reports 94% efficiency retention after 1,000 hours of maximum-power-point tracking at 85°C—a serious stress test, though not a 25-year module qualification.¹
• “Commercially sized” means a mini-module, not a retail-ready panel with proven manufacturing yield, warranty data, and project-finance acceptance.
• The advance matters because indium is a supply-constrained input in transparent electrodes; removing it addresses a real scale constraint while retaining high performance.¹²
• For prospective buyers of solar hardware: change nothing. For PV manufacturers and researchers: the recombination-layer and electrode process is now a credible route worth independently reproducing and cost-modeling.

A 207.9 cm² perovskite–silicon mini-module reached 31.0% certified power-conversion efficiency without indium, according to a paper in Science. That is the interesting number. Not because 31% is a new universal solar record, but because the result holds at a substantially larger area than the 1 cm² cells that dominate tandem-solar headlines.¹

The number that changes the story

Perovskite–silicon tandems work by stacking a higher-bandgap perovskite cell above silicon. The upper layer captures higher-energy photons more efficiently; silicon collects more of the lower-energy light that passes through. The architecture can exceed the practical efficiency ceiling of a single silicon junction.

But tandem devices depend on transparent conductive layers that collect charge and electrically join the two sub-cells. Indium tin oxide is excellent at that job. It is also a poor material around which to build a terawatt-scale industrial story: indium is a by-product metal with a comparatively limited supply base, and conventional sputtering can damage fragile perovskite layers.

The team’s substitution was not simply “use tin instead.” It used reactive plasma deposition to lay down tin oxide at lower particle energies than conventional sputtering. The resulting RPD-SnOₓ served both as the internal recombination layer and as front and rear transparent electrodes. The authors report a 33.2% champion efficiency at 1 cm² for the fully indium-free cell, then 31.0% certified efficiency for the 207.9 cm² mini-module.¹

That 2.2-percentage-point scale gap is not a footnote. It is the real accomplishment. Efficiency normally falls as a device grows because uniform films, electrical resistance, defects, and interconnection become harder to control. Here, the large-area device stayed above 30%.

What happened, precisely

The Science study reports three related results:

1. 33.6% certified efficiency when RPD-SnOₓ was used as the recombination layer.
2. 33.2% champion efficiency for a fully indium-free 1 cm² tandem device using RPD-SnOₓ for both the recombination layer and electrodes.
3. 31.0% certified efficiency for a fully indium-free 207.9 cm² mini-module.¹

The authors attribute part of the performance and stability gain to dense, uniform tin-oxide films that improve anchoring of a self-assembled monolayer. That reduces non-radiative recombination and suppresses halide migration—two failure mechanisms that take output from perovskite devices over time.¹ Specialist reporting independently describes the manufacturing logic and the mini-module scale.²

The actual significance: a materials bottleneck moves

The solar industry has plenty of efficiency records. Most are useful; few alter the manufacturing map.

This one could, because it attacks the supply-chain layer of tandem photovoltaics. The result says high-efficiency tandems do not inherently require indium-based transparent conductors. If the process remains performant at larger areas, with acceptable throughput and yield, it broadens the set of materials and equipment routes available to manufacturers.

That is a different claim from “tin oxide makes tandem panels cheap.” The study does not establish a factory cost per watt, a module bill of materials, deposition throughput, or a levelised-cost-of-energy advantage. It establishes that an indium-free stack can cross 30% at mini-module scale.

The editorial call: the breakthrough is a manufacturing-relevance result—not a consumer-product result. It turns a materials objection into an engineering validation task.

What this is not

It is not a commercial solar panel.

A 207.9 cm² mini-module is approximately 0.021 m² of active area. Full commercial modules are typically well above a square metre and must be packaged, interconnected, transported, installed, warranted, and financed. “Commercially sized” is defensible only in the narrower sense that it is beyond a tiny laboratory cell. It should not be read as “available for purchase” or “ready to replace silicon panels.”

Nor does 1,000 hours at 85°C equal a 25-year field-life claim. It is an important data point: the Science editor’s summary says the mini-modules retained 94% of initial efficiency after that maximum-power-point tracking test, while the paper also reports 65% retention after 105 days of outdoor operation.¹ Those metrics test different conditions; neither settles bankability.

Stakeholder landscape

The quieter connection: mineral constraints become deposition constraints

A clean-energy technology can appear constrained by a mineral, when the more important question is whether manufacturing can deposit a substitute layer with the right conductivity, transparency, uniformity, and interface chemistry.

This paper shifts the problem from “Can the world supply enough indium?” toward “Can RPD-SnOₓ be run cheaply, uniformly, and rapidly enough in a module factory?” That is progress, but it is a different category of risk. Supply risk gives way to capex, process control, yield, encapsulation, and qualification risk.

There is a geopolitical angle as well. The work joins institutions and industrial partners in China, Australia, and the UK. If commercialised, the decisive competitive asset may not be the tin commodity—tin oxide is not scarce in the same way—but the deposition know-how, equipment supply, and manufacturing integration around it.

Durability forecast

• One week: expect a burst of “cheaper, indium-free solar” coverage. The right reading is narrower: a credible mini-module result has been published in a top journal.
• One month: research groups and tandem firms will examine whether the RPD process can be replicated and integrated with textured silicon, fast production lines, and module packaging.
• One year: the story matters if the team or others publish larger modules, third-party qualification results, and manufacturing throughput/yield data. Without those, it remains a strong laboratory-to-pilot milestone.

Recommendations

If you are buying rooftop or commercial solar now

Do not delay a sound solar project waiting for this technology. Ask vendors for the exact module model, IEC certification, product and performance warranty, annual degradation specification, inverter compatibility, and installed cost per watt. This result does not alter today’s buying decision.

If you work on PV manufacturing or materials procurement

Track three numbers before treating RPD-SnOₓ as an industrial option:

1. Deposition throughput and uniformity on full-format substrates;
2. Yield and cost per square metre against incumbent transparent conductive oxide processes; and
3. IEC 61215 / IEC 61730 qualification plus longer-term outdoor degradation after encapsulation.

The paper’s 1,000-hour, 85°C maximum-power-point tracking result is encouraging. It is not a substitute for module qualification or multi-climate field evidence.

If you are a policy-maker or research funder

Do not fund another generic “record efficiency” campaign. Fund the missing bridge: pilot-line deposition, independent reproducibility, full-module packaging, and multi-year outdoor test arrays. That is where this result becomes either an industrial technology or a very good paper.

Uncertainty ledger

• Independent replication: not yet available in the evidence reviewed here.
• Mini-module to full module: unproven; 207.9 cm² is meaningful but far below a standard commercial module’s active area.
• Economics: the study supports a materials-substitution case, not a validated module-cost or LCOE claim.
• Lifetime: stress-test and 105-day outdoor results are promising but insufficient to infer 20–30-year degradation.
• “Indium-free” boundary: the paper describes the tandem cell stack as fully indium-free. A future factory’s broader supply chain, equipment, and packaging would need separate auditing before making a system-level claim.

Bottom Line

A 31%-efficient, 207.9 cm² indium-free tandem mini-module is a real advance because it preserves high performance while removing a material that could constrain scale. It does not mean indium-free tandem panels are ready to buy, finance, or warrant. The next proof is not a higher cell-efficiency number; it is a full-size module made repeatedly, qualified rigorously, and manufactured at a credible cost.

Sources

Tier 1 — Institutional research record. University of Oxford Materials / Bonilla Lab, “Indium-free perovskite/silicon tandem solar cells with tin oxide recombination layer and electrodes,” publication record linking to the Science study, accessed 22 July 2026.