Beyond Solar Briefing — June 2026

In this Briefing

1. Enhanced geothermal stops being a promise: Fervo's two-year production data
2. First Solar licences Oxford PV's perovskite IP — the commercialization signal that matters
3. A 25% perovskite-CIGS tandem reopens the silicon-free, flexible PV lane
4. China grid-connects a megawatt-class airborne wind aerostat at 2,000 m
5. Tidal's real record isn't power — it's 6½ years without a service call
6. Quaise drills rock with millimeter waves — superhot geothermal clears its gating risk

Enhanced geothermal stops being a promise: Fervo's two-year production data

What's new. Enhanced geothermal systems (EGS) have spent a decade as a "physics works, economics unproven" story. In April 2026 Fervo Energy published 614+ days of continuous production data from its Project Red pilot in Nevada — the first multi-year, public, well-instrumented EGS dataset — and it shows the resource behaving like firm baseload rather than a depleting science experiment [1]. This is the evidence base that underwrites Cape Station in Utah, whose first 100 MW is targeted to reach the grid in 2026 [2][3].

The evidence. Across 614+ days, Project Red held 98.4% system uptime (excluding surface/grid events), ~2.1 MW average gross output, and a production temperature that fell only ~2.5 °F over 500+ days — essentially flat and matching the reservoir model [1]. Fervo reports a ~20% thermal recovery factor in two years, above DOE benchmarks [1]. At Cape Station, a 30-day flow test hit 107 kg/s at >220 °C, enough for >10 MW per well — roughly triple Project Red's per-well output — alongside a ~70% year-over-year cut in drilling time as the team applied oilfield horizontal-drilling and fracturing methods [2]. Cape Station is contracted to reach 400–500 MW by 2028 [2][3].

The Take: The decision-relevant number here is not megawatts — it is the ~2.5 °F drawdown over 500 days. Thermal decline is the variable that has historically killed geothermal project finance, because it converts a "firm" asset into a declining one and forces conservative debt sizing. A near-flat, model-matching temperature profile is what lets lenders underwrite a 30-year resource at a high capacity factor. Beyond Solar estimate: at 98% uptime and the demonstrated thermal stability, an EGS plant supports a ~90%+ annual capacity factor versus ~25% for solar and ~40% for onshore wind — meaning one EGS MW displaces roughly 3–3.6 MW of intermittent nameplate on an energy basis (derived from the cited uptime/temperature data and standard US resource capacity factors). That energy-density-of-firmness is precisely what data-center offtakers are now paying a premium for, and it is why geothermal PPAs are repricing upward rather than down.

Market read. No pure-play public EGS name exists (Fervo is private), so the cleanest read is the conventional-geothermal incumbent positioned to ride the same firm-power demand and to deploy EGS techniques on its own resource base. ORA (NYSE) — Add · conviction Medium · 12–24 mo — Ormat's February 2026 data-center PPAs (up to 150 MW to Google via NV Energy; a direct deal with Switch) and a ~27% price step-up on a contract renewal show firm-geothermal pricing power inflecting as load grows [4]. EGS validation expands Ormat's addressable resource rather than threatening it.

First Solar licences Oxford PV's perovskite IP — the commercialization signal that matters

What's new. Lab efficiency records are routine; manufacturing commitments are not. In February 2026 First Solar took a non-exclusive licence to Oxford PV's perovskite materials and device patents for the US market, explicitly carved to exclude crystalline-silicon semiconductors [5]. The signal is that the largest US module maker — a thin-film house with no silicon legacy to defend — now sees perovskite as core IP rather than a science project.

The evidence. The licence covers issued and pending perovskite patents with US manufacturing and distribution rights; financial terms were undisclosed [5]. First Solar reports having invested over $2 billion in thin-film R&D and runs a development line in Ohio producing small perovskite modules that have met internal efficiency and manufacturability targets [5]. Separately, Oxford PV — which shipped its first commercial 24.5% tandem modules to a US utility customer in 2024 from its Brandenburg, Germany line — is targeting 26% modules in 2026 and a GW-scale plant in 2026–27 [6]. Hanwha Qcells has a ~$100 M perovskite pilot line, 28.6% certified cell efficiency, and commercial production guided for 2026 [6].

The Take: The carve-out is the whole story. By licensing perovskite while excluding silicon, First Solar is positioning a perovskite-on-CdTe or perovskite-tandem path that sidesteps the crystalline-silicon supply chain entirely — the one part of PV the US has structurally failed to reshore. Every other tandem roadmap (Oxford PV, Qcells, LONGi) bolts perovskite onto a silicon bottom cell and therefore inherits Chinese wafer dependence. A silicon-free US tandem is the only version of this technology that is simultaneously a higher-efficiency play and a domestic-content/tariff play. Beyond Solar estimate: that dual exposure is worth more to First Solar's policy-driven order book than the ~2–4 efficiency points a tandem adds, which is why a thin-film incumbent paid for silicon-excluded IP it cannot use in a conventional tandem.

Market read. FSLR (NASDAQ) — Add · conviction Medium · 24–36 mo — optionality on a silicon-free US tandem stacks on an already domestic-content-advantaged book; perovskite is upside, not the base case, so downside is limited if it slips [5]. ENPH (NASDAQ), NXT (NASDAQ) — Hold — balance-of-system and tracker names benefit from any higher-watt module but are technology-agnostic; no clean directional edge from this specific deal.

A 25% perovskite-CIGS tandem reopens the silicon-free, flexible PV lane

What's new. Most tandem headlines are perovskite-on-silicon. In May 2026 a Tokyo City University group reported a two-terminal perovskite-CIGS tandem at 25.14% certified efficiency — the first time the 25% barrier has been broken for this silicon-free architecture, which is inherently thin-film, flexible and lightweight [7].

The evidence. AIST certified 25.14% on a 1 cm² two-terminal device with a perovskite top cell and CIGS bottom cell, edging past the prior 24.6% set by Germany's HZB in February 2025 [7]. For context on the broader field's pace, perovskite-silicon tandems now sit at ~34% (a certified 33.6% flexible device with a 2.015 V open-circuit voltage was published in Nature in 2026), and all-perovskite tandems have reached 28.4% certified [8]. Durability is improving in parallel: cross-linked phosphonic-acid hole transporters cut thermal degradation to <4% loss after ~1,200 h at 65 °C, and graded dielectric layers retain >92% efficiency after 1,000 h of reverse-bias stress [8].

The Take: Perovskite-CIGS will not win the rooftop/utility efficiency race — it trails silicon tandems by ~9 points and CIGS manufacturing is harder to scale than silicon. Its real market is the one silicon tandems cannot serve: high specific-power (W/kg), flexible, building- and vehicle-integrated, and aerospace/defence surfaces where weight and conformability dominate over $/W. Crossing 25% on a silicon-free, flexible stack matters because it is the first time this architecture is efficient enough to be commercially interesting in those niches rather than only in the lab. Beyond Solar estimate: the addressable premium markets for flexible high-specific-power PV are a low-single-digit-percent slice of PV demand, but at 3–10× the $/W of utility modules — a margin pool, not a volume play. Investors should read this as a specialty-materials opportunity, not a threat to mainstream silicon.

Market read. No clean public-market read — the leading perovskite-CIGS groups are academic, and CIGS incumbents are largely private or subsumed. (Omitted.)

China grid-connects a megawatt-class airborne wind aerostat at 2,000 m

What's new. Airborne wind energy (AWE) has produced a graveyard of bankrupt startups (Makani, Ampyx). In January 2026 China crossed a threshold the Western AWE sector never reached: a megawatt-class helium aerostat ("S2000 SAWES") lifted a wind unit to ~2,000 m and delivered power to the grid in a test flight in Yibin, Sichuan [9].

The evidence. The platform rose to ~2,000 m (6,560 ft) on a helium aerostat with ~20,000 m³ envelope volume and a maximum rated output of up to 3 MW, feeding grid power during the test [9]. The pitch is access to the stronger, steadier winds of the lower stratosphere, where capacity factors run well above tower-mounted turbines. A domestic envelope-material plant is being scaled toward 200,000 linear meters/year in 2026 [9].

The Take: The headline is "first megawatt airborne wind," but the binding constraint is helium, not aerodynamics. A 20,000 m³ aerostat is a large, recurring helium liability in a market with chronic supply shocks and price volatility; lift gas, not turbine design, will set this system's levelized cost and its scalability ceiling. The lighter-than-air approach also caps siting to low-turbulence, controlled airspace, which is why this is a Chinese-grid demonstration rather than a globally portable product. Beyond Solar estimate: tethered-kite/rigid-wing AWE remains the more scalable long-run architecture because it carries no lift-gas opex — the aerostat's advantage is that it flies today, not that it wins on cost. Read this as a national-champion demonstration of capability, not a template Western developers should copy.

Market read. No clean public read — the developer is a Chinese state-linked consortium with no investable Western pure-play; AWE exposure inside large turbine OEMs is immaterial. (Omitted.)

Tidal's real record isn't power — it's 6½ years without a service call

What's new. Marine energy's perennial problem is not generating power but surviving the ocean economically. In 2026 the MeyGen array in Scotland's Pentland Firth — the world's largest tidal-stream array — confirmed that SKF's drivetrain systems on one turbine ran 6½ years with no unplanned or disruptive maintenance, a reliability record for the sector [10].

The evidence. MeyGen Phase 1 comprises four 1.5 MW seabed-mounted turbines and has delivered 80+ GWh to the UK grid; the 6½-year run without unplanned intervention is the headline reliability metric [10]. Tidal's structural advantage is predictability — output is set by lunar cycles, not weather — and developer pipelines now point to ~30 turbines slated for deployment from 2026 across Scotland, France and Japan, each sized to power ~3,000 homes [10].

The Take: Tidal's cost problem was never the turbine; it was the boat. Offshore intervention vessels and weather-window risk dominate marine-energy O&M, so unplanned subsea maintenance is the line item that destroys project economics. A 6½-year unplanned-maintenance-free interval is therefore the single most bankable datapoint tidal has produced — it attacks opex and downtime simultaneously, which is what a lender stress-tests. Beyond Solar estimate: demonstrated reliability of this order can move tidal's effective availability and O&M assumptions enough to compress LCOE by a meaningful double-digit percentage versus first-of-a-kind assumptions, which is the gap between "subsidy-dependent demonstrator" and "biddable in a CfD auction." Predictability plus proven uptime is a genuinely differentiated firming product for grids that already have abundant solar.

Market read. Largely private/sponsor-owned (SAE Renewables operates MeyGen; SKF supplies bearings). The cleanest listed read is the diversified industrial supplying the reliability-critical drivetrain content. SKF-B (STO) — Hold · conviction Low · 24–36 mo — marine is a rounding error in SKF's revenue today, but a demonstrated reliability edge positions it as a default supplier if tidal scales; not a thesis to own SKF on, but a credit to its industrial franchise [10].

Quaise drills rock with millimeter waves — superhot geothermal clears its gating risk

What's new. Superhot-rock geothermal (>300 °C "supercritical" resource) promises geothermal almost anywhere, but only if you can drill through kilometers of hard crystalline basement — which conventional bits cannot do economically. Quaise's gating risk is the drilling method itself. In 2026 the company advanced field testing of millimeter-wave drilling — using a gyrotron to vaporize rock with no downhole hardware — and is moving from a 100 kW to a 1 MW gyrotron [11][12].

The evidence. Quaise's hybrid approach drills near-surface rock conventionally, then switches to millimeter-wave ablation in the basement; the company expects to take delivery of a 1 MW gyrotron in early 2026, a ~10× power step from its 100 kW field unit [11]. Analysis presented at the 2026 Stanford Geothermal Workshop supports a first plant producing at least 50 MW [11]. Project Obsidian, the first-of-a-kind plant in Oregon targeting rock >300 °C, is under construction with operation projected as early as 2030 [12].

The Take: Superhot rock is the highest-prize, lowest-readiness item in this Briefing — and the gyrotron step-up is the right milestone to watch because energy intensity, not concept, is what determines whether millimeter-wave drilling reaches commercial depth at commercial rate-of-penetration. The thermodynamic case is unusually strong: a supercritical well can deliver roughly 5–10× the power per well of a conventional hydrothermal well at the same flow, because energy scales with enthalpy and supercritical fluid carries far more. Beyond Solar estimate: if Quaise hits even the low end of that range, superhot rock needs an order of magnitude fewer wells per delivered MW than EGS — which is the only way the drilling-cost problem ever closes. The risk is binary and the timeline (2030) is long; size exposure accordingly. This is the call option in the geothermal complex, not the core holding.

Market read. Quaise is private; no pure-play exposure. The directional read is that proof of deep-drilling-to-supercritical would be additive — not threatening — to firm-power incumbents. ORA (NYSE) — see Topic 1 — superhot rock expands the long-run geothermal resource Ormat can monetize; it is upside optionality on a 2030+ horizon, not a near-term driver. (No separate call.)

Market Calls

References

1. Fervo Energy, "Enhanced Geothermal Has Been Proven at Scale. Here's What Two Years of Production Data Show." (Apr 13, 2026). https://fervoenergy.com/enhanced-geothermal-has-been-proven-at-scale-heres-what-two-years-of-production-data-show/
2. Fervo Energy, "Fervo Energy's Record-Breaking Production Results Showcase Rapid Scale Up of Enhanced Geothermal." https://fervoenergy.com/fervo-energys-record-breaking-production-results-showcase-rapid-scale-up-of-enhanced-geothermal/
3. Canary Media, "Fervo nabs $462M to complete massive next-gen geothermal project." https://www.canarymedia.com/articles/geothermal/fervo-investment-capital-b-cape-station
4. The Globe and Mail / StockTitan, "Ormat Technologies Signs Long-Term Geothermal PPA for Google" and related ORA PPA news (Feb 2026). https://www.stocktitan.net/news/ORA/ormat-technologies-announces-the-signing-of-geothermal-portfolio-ppa-1ieffqllasex.html
5. Electrek, "First Solar locks in key perovskite tech with a new Oxford PV deal" (Feb 24, 2026). https://electrek.co/2026/02/24/first-solar-perovskite-oxford-pv/
6. PV-Tech, "Oxford PV ships first commercial perovskite tandem modules"; pv magazine, "Perovskite rollout gathers pace" (2025). https://www.pv-tech.org/oxford-pv-ships-first-commercial-perovskite-tandem-modules/
7. pv magazine, "Japanese scientists achieve world record 25.14% efficiency for perovskite-CIGS tandem solar cell" (May 18, 2026). https://www.pv-magazine.com/2026/05/18/japanese-scientists-achieve-world-record-25-14-efficiency-for-perovskite-cigs-tandem-solar-cell/
8. Wang, Li, Yu et al., "Flexible perovskite/silicon tandem solar cells with 33.6% efficiency," Nature (2025/2026). https://www.nature.com/articles/s41586-025-09849-4 ; "A cross-linked molecular contact for stable operation of perovskite/silicon tandem solar cells," Science (2026). https://www.science.org/doi/10.1126/science.ady6874
9. Euronews, "A floating power station? China's flying wind turbine hits milestone with grid-connected test" (Jan 29, 2026). https://www.euronews.com/next/2026/01/29/a-floating-power-station-chinas-flying-wind-turbine-hits-milestone-with-grid-connected-tes
10. Marine Technology News, "MeyGen Tidal Energy Tech Sets New Operational Record"; SKF / Interesting Engineering coverage of the 6½-year reliability record (2025–2026). https://www.marinetechnologynews.com/news/meygen-tidal-energy-operational-650672
11. Quaise Energy, "Quaise Energy Advances Toward Building the World's First Superhot Geothermal Power Plant" and 2026 Stanford Geothermal Workshop analysis. https://www.quaise.com/news/quaise-energy-on-track-to-build-worlds-first-power-plant-using-superhot-geothermal-energy
12. Canary Media, "Quaise looks to advance 'superhot' geothermal power plant in Oregon." https://www.canarymedia.com/articles/geothermal/quaise-superhot-geothermal-power-plant-oregon

Disclosures & Disclaimer

This report is general commentary published for information purposes only. It is not investment advice, a recommendation, or a solicitation to buy or sell any security, and it does not account for the objectives or circumstances of any individual. Beyond Solar is a research publication, not a registered investment adviser or broker-dealer. Views are the publication's own analytical opinions, are subject to change, and may prove wrong. Markets involve risk of loss; past performance does not indicate future results. Readers should do their own research and consult a licensed financial professional before acting. The publication and/or its principals may hold positions in securities mentioned. Company facts and figures are drawn from public sources believed reliable but are not guaranteed. © Beyond Solar.

About Beyond Solar

Beyond Solar publishes independent technical and techno-economic research in its field. Reports are prepared for subscribers and are provided for information only. © Beyond Solar. All rights reserved.