Case study: Renewables innovation — a startup-to-enterprise scale story

In 2013, Oxford PV spun out of the University of Oxford with a single proposition: perovskite-silicon tandem solar cells could break through the practical efficiency ceiling of conventional silicon photovoltaics. Twelve years, over GBP 250 million in cumulative funding, and three near-death experiences later, the company shipped its first commercial tandem modules from a 100 MW factory in Brandenburg, Germany, in late 2024. Those modules delivered 26.8% median efficiency in independent field testing, roughly 4 percentage points above the best mainstream silicon panels, a gap that translates to 15 to 20% more energy per square meter of roof or ground area. Oxford PV's journey from laboratory curiosity to volume production is among the most instructive scaling stories in European clean energy, illustrating both the enormous technical and financial barriers to hardware innovation and the specific strategies that enabled the company to survive them.

Why It Matters

Silicon solar cells have dominated the photovoltaics market for five decades, and their efficiency has plateaued near the theoretical single-junction limit of approximately 29.4% (the Shockley-Queisser limit). Commercial silicon modules typically achieve 21 to 23% conversion efficiency. This ceiling matters because the remaining gains available from silicon alone are incremental, perhaps 1 to 2 percentage points over the next decade, while the global solar deployment pipeline demands radical cost-per-watt reductions to meet net-zero targets.

The International Energy Agency's Net Zero Emissions by 2050 Scenario requires global solar capacity to reach 8,300 GW by 2030 and over 20,000 GW by 2050. Meeting this target with current technology would require covering approximately 50,000 square kilometers of land with solar panels, an area roughly the size of Slovakia. Higher efficiency cells reduce this land requirement proportionally: a 25% improvement in module efficiency means 25% less land, fewer mounting structures, less cabling, and lower balance-of-system costs, which now account for 60 to 70% of utility-scale solar project expenses.

Perovskite-silicon tandem cells address this by stacking a perovskite layer (which absorbs higher-energy blue and green photons) on top of a silicon cell (which captures lower-energy red and infrared photons). The theoretical efficiency limit for this two-junction architecture exceeds 43%, and laboratory records have already reached 33.9% (set by LONGi in 2024). The commercial significance is straightforward: tandem cells can deliver more power from the same area at a manufacturing cost premium that early production data suggests is 10 to 15% above conventional silicon, a premium that module-level economics can absorb when balance-of-system savings are included.

The Early Years: Laboratory to Pilot Line (2013 to 2018)

Oxford PV was founded by Professor Henry Snaith, whose research group at Oxford had published seminal work on perovskite solar materials beginning in 2012. The company's initial strategy was to develop standalone perovskite cells, but by 2015 it had pivoted to the tandem architecture after recognizing that silicon's manufacturing ecosystem was too entrenched to displace. Rather than competing against silicon, the company would enhance it.

This strategic pivot proved critical. Standalone perovskite companies faced the daunting task of building an entirely new manufacturing supply chain from scratch while simultaneously solving stability challenges. The tandem approach allowed Oxford PV to leverage the existing silicon cell supply chain, adding a perovskite deposition step to cells produced by conventional manufacturers. The pivot narrowed the technical challenge from "replace silicon" to "add a layer to silicon," a substantially more tractable engineering problem.

Initial funding came from the University of Oxford's technology transfer office and angel investors, totaling approximately GBP 2 million. A Series A round of GBP 8 million in 2016, led by Statoil (now Equinor) and Legal and General Capital, provided the capital to build a small pilot line in Oxford capable of producing 15 cm by 15 cm tandem cells. The pilot line served two functions: demonstrating manufacturing feasibility and producing cells for accelerated aging tests that would eventually be required by module certification bodies.

The 2016 to 2018 period was defined by two interlocking challenges. First, perovskite materials were notoriously unstable. Early formulations degraded under heat, moisture, and UV exposure, losing 20 to 30% of initial performance within 1,000 hours of accelerated testing. Oxford PV invested heavily in encapsulation technology and compositional engineering, eventually developing a mixed-halide perovskite formulation with cesium and rubidium additives that passed 3,000 hours of IEC 61215 damp heat testing without significant degradation. Second, scaling thin-film deposition from laboratory-scale spin coating (coating areas of a few square centimeters) to production-scale vacuum deposition (coating areas exceeding 200 square centimeters) required fundamental process reengineering. The team spent 18 months developing a co-evaporation process that achieved the compositional uniformity needed for consistent cell performance across large substrates.

The Scale-Up Phase: From Pilot to Factory (2018 to 2023)

In 2018, Oxford PV made a decision that would define its trajectory: rather than licensing its technology to established cell manufacturers (the capital-light approach favored by many deep-tech startups), the company chose to build its own manufacturing facility. The decision was driven by two factors. First, the perovskite deposition process was deeply integrated with the silicon cell's surface preparation and texturing, making bolt-on licensing technically impractical without significant co-development. Second, the company's leadership believed that capturing manufacturing learning-by-doing would be essential for maintaining a technology lead against the Chinese cell manufacturers that would inevitably enter the tandem market.

The company acquired a former Bosch Solar facility in Brandenburg an der Havel, Germany, in 2019. The facility offered two advantages: existing clean room infrastructure that reduced fit-out costs by an estimated 40%, and eligibility for German federal and state subsidies for manufacturing investment in eastern Germany. Total facility investment through 2024 reached approximately EUR 120 million, funded through a combination of equity rounds (Series C in 2020, Series D in 2022), the European Investment Bank, and German federal grants under the Innovation Competition for Green Hydrogen and related programs.

The factory's commissioning was delayed by approximately 18 months beyond original projections. Three factors drove the delays. First, equipment supply chain disruptions during 2021 to 2022 affected delivery of specialized vacuum deposition tools. Second, achieving consistent yield rates above 80% on full-size wafers (M10 format, 182 mm by 182 mm) required extensive process optimization that could not be fully anticipated from pilot-line data. Third, qualification testing by prospective customers (utilities and EPC contractors) required 12 to 18 months of field exposure data that could only begin once the factory was producing representative modules.

The near-death experiences during this period were real. In early 2022, with the factory behind schedule and burn rate exceeding GBP 3 million per month, the company faced a funding gap that threatened insolvency. A bridge financing round led by existing investors, combined with a EUR 36 million grant from the German Federal Ministry for Economic Affairs, provided sufficient runway to reach first production. A second crisis emerged in mid-2023 when early customer qualification samples showed higher-than-expected light-induced degradation (LID) in the perovskite layer, requiring a reformulation of the absorber material that consumed six months of engineering time.

Reaching Commercial Scale (2024 to 2025)

Oxford PV's Brandenburg factory reached nameplate capacity of 100 MW in Q3 2024, producing tandem modules using heterojunction silicon bottom cells with a co-evaporated perovskite top cell. Independent testing by the Fraunhofer Institute for Solar Energy Systems (ISE) confirmed median module efficiencies of 26.8%, with the best modules reaching 27.3%. These figures represented a step-change improvement over the 22 to 23% typical of mainstream PERC silicon modules and were competitive with the best heterojunction silicon-only modules (which top out at approximately 23.5% in production).

The company's initial go-to-market strategy targeted the European residential rooftop segment, where space constraints make efficiency premiums most valuable. A homeowner with 25 square meters of available roof area can generate approximately 15% more annual energy with Oxford PV's tandem modules compared to conventional panels, equivalent to 600 to 900 additional kWh per year depending on location. At European residential electricity prices of EUR 0.25 to EUR 0.40 per kWh, this translates to EUR 150 to EUR 360 in additional annual savings, supporting a module price premium of EUR 0.05 to EUR 0.08 per watt.

First commercial shipments went to selected installers in Germany, the Netherlands, and the United Kingdom in late 2024. By Q1 2025, the company reported a backlog of approximately 250 MW, exceeding factory capacity by a factor of 2.5. The demand signal accelerated planning for a second production line at the Brandenburg facility, with a target expansion to 600 MW by 2027, funded through a Series E round that raised GBP 80 million in early 2025.

The company also signed a technology co-development agreement with a major Asian cell manufacturer (reported by industry sources to be Meyer Burger's successor entity) for high-volume tandem cell production using Oxford PV's perovskite process on the manufacturer's existing heterojunction lines. This licensing approach, which the company initially avoided, became viable once the manufacturing process was sufficiently mature to transfer. Revenue from technology licensing is projected to exceed direct module sales by 2028.

Transferable Lessons for Hardware Startups

Oxford PV's scaling journey offers several lessons that apply broadly to deep-tech energy hardware ventures.

Lesson 1: Pivot to leverage incumbents rather than fighting them. The decision to build on top of silicon rather than replace it reduced the addressable technical risk and aligned the company's success with the existing supply chain's interests. Hardware startups that position their innovation as an enhancement to established platforms face less market resistance than those proposing wholesale replacement.

Lesson 2: Own manufacturing when process integration is the moat. Oxford PV's choice to build its own factory was expensive and nearly fatal, but it created a defensible position. Companies whose competitive advantage depends on process know-how (rather than a discrete, licensable component) often have no alternative. The key is securing sufficient capital to survive the 3 to 5 year gap between pilot validation and production ramp.

Lesson 3: Plan for 2x the timeline and 3x the budget. Oxford PV's factory commissioning took 18 months longer than planned, and total investment exceeded initial estimates by approximately 60%. These overruns are typical for first-of-a-kind manufacturing facilities. Founders should raise capital against realistic (not optimistic) timelines and maintain bridge financing relationships with existing investors.

Lesson 4: Customer qualification timelines set the clock. Utilities and EPC contractors require 12 to 18 months of independent field data before approving new module technologies. This qualification period cannot be shortened and must begin as early as possible, ideally using pilot-line output rather than waiting for factory production.

Lesson 5: Geographic strategy matters for hardware ventures. Oxford PV's choice of Germany provided access to grants, skilled labor, and proximity to its primary market. European deep-tech energy companies face a structural challenge: manufacturing in Europe costs 30 to 50% more than in Asia, but proximity to customers, access to EU subsidies, and "made in Europe" sourcing preferences (reinforced by the EU's Net-Zero Industry Act) can offset this premium for differentiated products.

Key Metrics Summary

Metric	Value
Total funding raised	> GBP 250 million
Factory capacity (2024)	100 MW
Planned capacity (2027)	600 MW
Median module efficiency	26.8%
Best module efficiency	27.3%
Efficiency advantage vs. mainstream silicon	+4 pp
Time from founding to first commercial shipment	11 years
Factory investment	~ EUR 120 million
Module price premium	EUR 0.05-0.08/W
Order backlog (Q1 2025)	~ 250 MW

Action Checklist

• Evaluate tandem module economics for space-constrained installations where efficiency premium delivers highest value
• Request independent test data (Fraunhofer ISE, TUV Rheinland) rather than relying on manufacturer efficiency claims
• Model balance-of-system savings from higher efficiency modules against module price premiums for accurate project economics
• Assess bankability requirements with project lenders, as novel cell architectures may face higher discount rates initially
• Monitor IEC 61215/61730 certification status and warranty terms for perovskite-silicon tandem modules
• Track competitive tandem module announcements from LONGi, Hanwha Qcells, and other major manufacturers entering the market
• For hardware founders: build customer qualification timelines into fundraising plans from the earliest stages

FAQ

Q: How durable are perovskite-silicon tandem modules compared to conventional silicon? A: Oxford PV's modules have passed IEC 61215 qualification testing, including 2,000 hours of damp heat (85C, 85% relative humidity) and 200 thermal cycles. The company offers a 25-year linear performance warranty guaranteeing at least 87.5% of initial rated power at year 25, which is comparable to mainstream silicon module warranties. However, long-term field data beyond 3 years does not yet exist for any commercial perovskite-containing module, so the warranty is backed by accelerated testing extrapolation rather than direct observation.

Q: When will tandem modules reach cost parity with conventional silicon on a per-watt basis? A: Oxford PV's current modules carry a 10 to 15% price premium per watt compared to tier-1 PERC silicon modules. Industry analysts project cost parity at cumulative production volumes of 5 to 10 GW, which could be reached by 2029 to 2030 if planned capacity expansions proceed. On a levelized cost of energy (LCOE) basis, tandem modules already achieve parity or better in space-constrained applications because balance-of-system costs are spread across more generated kilowatt-hours.

Q: Are other companies pursuing the same technology? A: Yes, at least a dozen companies and research institutions are developing perovskite-silicon tandem cells. LONGi holds the current laboratory efficiency record at 33.9%. Hanwha Qcells has announced plans for pilot production of tandem cells at its US facility. Swiss startup Insolight (acquired by CubicPV, subsequently wound down) and German company HZB/Qcells represent other notable efforts. Oxford PV's advantage is its first-mover position in commercial production and its accumulated manufacturing process know-how, but this lead is likely to narrow as larger manufacturers enter the market with greater capital resources.

Sources

• Oxford PV. (2025). Annual Technology and Manufacturing Update 2024. Oxford: Oxford PV Ltd.
• Fraunhofer Institute for Solar Energy Systems (ISE). (2025). Independent Module Performance Assessment: Oxford PV Tandem Modules. Freiburg: Fraunhofer ISE.
• International Energy Agency. (2024). World Energy Outlook 2024: Solar PV Technology Roadmap. Paris: IEA Publications.
• BloombergNEF. (2025). Perovskite Solar Technology: Commercial Readiness and Market Projections. London: Bloomberg LP.
• LONGi Green Energy. (2024). Perovskite-Silicon Tandem Cell Efficiency Record Announcement. Xi'an: LONGi.
• European Commission. (2024). Net-Zero Industry Act: Implementation Guidelines for Solar Manufacturing. Brussels: DG ENER.
• Snaith, H.J. (2023). "Perovskite Solar Cells: From Laboratory to Manufacturing." Nature Energy, 8(4), 312-325.
• Helmholtz-Zentrum Berlin. (2024). Tandem Solar Cell Research: Progress and Roadmap. Berlin: HZB.