Energy Weekly Review 2026-05-15

Week In Review

The dominant story of the week was the public-market debut of next-generation geothermal: Fervo Energy’s IPO priced at the top of its range and popped 33% on its first day of trading, pushing its market valuation past $10 billion and validating enhanced geothermal systems as a real category — not a thought experiment — for firm, clean power. That story rhymed with several others. Long-duration storage took a similar institutionalization step when Cerberus Capital Management and Eos Energy announced Frontier Power USA, a purpose-built independent power producer for zinc-bromide batteries with a 2 GWh reservation already on the books. Ford formally launched its grid-scale battery subsidiary and unveiled a containerized 512 Ah LFP product, repurposing capacity originally built for electric vehicles into infrastructure for data centers and utilities. The common thread is firmness: capital is flowing toward technologies that can be dispatched on demand, paired with renewables, or sited next to growing AI load.

Fusion progress kept pace with the storage and geothermal news, in a quieter register. Helion disclosed a downsized “Tiny Merge” testbed intended to compress its iteration cycle ahead of a contractually fixed 2028 Microsoft delivery, and Jefferson Lab opened a spin-polarized fusion experiment that aims to wring more reaction yield out of the same plasma conditions. On the regulatory side, the Nuclear Regulatory Commission’s newly effective Part 53 framework is beginning to be tested against real applications, and NANO Nuclear filed its KRONOS MMR construction permit, one of the first microreactor designs to seek formal licensing under the streamlined pathway.

The bookend stories were a research milestone and a grid milestone. A Chinese group reported an all-perovskite tandem solar cell at 29.80% efficiency using a laser-polishing technique that improves the rear interface — incremental in number but important in method. And in Western Australia’s main grid, batteries supplied a record 37.2% of peak demand on the evening of 9 May, demonstrating the practical capability of large-scale storage in an isolated market. A complementary deal between Alsym Energy and Juniper Energy lined up 500 MWh of sodium-ion deployment in California — a small step toward diversifying chemistry beyond lithium iron phosphate.

Read together, the week reads less like a list of disconnected announcements and more like a transition: the technologies that spent the last decade in research and demonstration are now being capitalized, sited, and tied into actual grids and customers.

Items

Fervo Energy IPO Lands at $10 Billion Valuation as Geothermal Goes Public

Fervo Energy, the next-generation geothermal developer, went public on the New York Stock Exchange on May 13 in an upsized $1.89 billion initial public offering. Shares jumped roughly 33% on their first day of trading, pushing the company’s market valuation past $10 billion and making Fervo the first pure-play enhanced geothermal company to reach public markets at scale.

Enhanced geothermal systems (EGS) differ from conventional geothermal in that they don’t rely on naturally occurring hot water or steam. Instead, operators drill deep wells into hot dry rock, fracture the rock, and circulate water through the fractures to extract heat. The approach borrows directional drilling and hydraulic stimulation techniques developed by the shale industry and applies them to a different objective: producing firm, low-carbon baseload electricity rather than oil and gas. Fervo has reported reducing both drilling time and cost per foot by roughly two-thirds across 14 wells.

The IPO was driven in significant part by demand from AI data-center operators looking for round-the-clock clean power that doesn’t depend on weather. Fervo’s flagship Cape Station project in Utah is on track to begin delivering an initial 100 MW to the grid in October 2026, which would make it the first commercial-scale enhanced geothermal project to do so. The company has also contracted with Google to supply 115 MW from its Corsac Station project in Nevada.

For a sector that has spent decades as a niche of conventional renewables, the IPO is a meaningful signal. Public markets are willing to underwrite firm, low-carbon generation at fossil-scale prices, and enhanced geothermal has graduated from research curiosity to a financeable technology class.

Source: TechCrunch

Eos and Cerberus Launch Frontier Power USA to Scale Long-Duration Storage

Eos Energy Enterprises and Cerberus Capital Management announced on May 13 the formation of Frontier Power USA, a new independent power producer purpose-built to develop, own, and operate long-duration battery storage projects using Eos’ proprietary zinc-bromide Z3 batteries. Cerberus is anchoring the venture with a $100 million equity commitment and is expected to take controlling equity in Frontier Power, while Eos plans to fund its own share via a roughly $150 million rights offering.

Long-duration storage — generally defined as systems that can discharge for six or more hours — has been a chronic gap in the energy transition. Lithium iron phosphate batteries dominate the four-hour-and-under market, but seasonally variable renewables need storage that can shift energy across days, not hours. Zinc-bromide flow batteries use abundant, non-flammable materials and are particularly well-suited to long-duration applications, but commercialization has been slow because project finance for unfamiliar chemistries has been difficult to arrange.

Frontier Power USA is explicitly designed to solve that financing problem. The structure includes a firm 2 GWh capacity reservation from Eos and a project-level insurance framework providing up to $1.5 billion in coverage through Lloyd’s of London syndicates and other highly rated markets. The insurance wrapper is intended to give lenders enough confidence to deploy debt against a chemistry that doesn’t yet have a long track record.

The target customers are commercial and industrial users, AI data centers, and utility-scale projects. If Frontier Power USA can move zinc-bromide from one-off projects to a financeable, repeatable model, it would matter beyond Eos: the same playbook could unlock other emerging long-duration technologies.

Source: Eos Energy Enterprises

All-Perovskite Tandem Solar Cell Hits 29.80% Efficiency Using Laser Polishing

A research team published results on May 12 showing an all-perovskite tandem solar cell with a certified efficiency of 29.80%, achieved through a novel laser-polishing technique applied to the cell’s rear interface. The work pushes all-perovskite tandems — cells made entirely from perovskite materials, without the silicon used in most commercial tandems — closer to the efficiency levels needed for commercial competitiveness.

Perovskite tandems stack two cells with different bandgaps so the upper cell captures high-energy photons and the lower cell captures the lower-energy photons that pass through. All-perovskite versions are attractive because they can be made on flexible substrates and don’t require the energy-intensive crystalline silicon manufacturing process. The challenge has been losses at the interfaces between layers, where defects and roughness reduce performance and stability.

The team’s laser-polishing approach smooths the rear interface of the narrow-bandgap bottom cell, improving its efficiency from 19.64% to 24.07% and substantially improving operational stability under continuous illumination. The full tandem device reached 29.80% efficiency — short of the 34.85% record set by perovskite-silicon tandems but among the highest reported for all-perovskite stacks.

Perovskite-silicon tandems have dominated efficiency records and are expected to reach commercial production first, but all-perovskite stacks remain strategically important because they unlock applications — flexible modules, building-integrated PV, and ultra-lightweight panels — that silicon cannot serve. Progress on interface engineering is the kind of incremental work that will determine whether all-perovskite reaches commercial viability later this decade.

Source: pv magazine

Helion Builds “Tiny Merge” Testbed to Accelerate Its 2028 Microsoft Deadline

Helion Energy disclosed on May 8 that it is building a downsized testbed device called Tiny Merge — less than one-eighth the size of Polaris, its seventh-generation and final prototype — to accelerate research ahead of a contractually fixed 2028 deadline to begin delivering fusion power to Microsoft. The decision reflects an explicit choice to iterate faster on a smaller machine rather than wait for results from the larger one.

Helion’s approach is unusual among fusion startups. Rather than using a tokamak or laser-based inertial confinement, the company uses field-reversed configurations (FRCs) — donut-shaped plasmas without a central solenoid — that are accelerated and merged at high speed. The goal is to compress the merged plasma to fusion conditions, then directly capture the energy released as electricity rather than converting heat through a steam cycle.

Tiny Merge is designed specifically to study FRC formation and merging — the front-end physics that has to work reliably before any downstream compression matters. By making the machine small and inexpensive, Helion can run many more shots, try more configurations, and reach the data that will inform the final design of its Orion power plant. Senior director Michael Hua framed the testbed as enabling iteration with “much less energy and far fewer resource requirements” than Polaris.

The 2028 deadline is real: Helion’s power-purchase agreement with Microsoft (with Constellation as a partner) carries financial penalties if the company fails to deliver. Tiny Merge is a bet that the fastest path to a working power plant runs through more, smaller, cheaper experiments rather than fewer, larger ones — a notably Silicon Valley framing of fusion development.

Source: GeekWire

Western Australia Batteries Supply Record 37.2% of Peak Demand

Battery storage in Western Australia’s South West Interconnected System (SWIS) supplied a record 37.2% of peak evening demand on Saturday, May 9. Energy Policy WA reported the milestone, which capped a weekend during which renewable energy contributed 53% of total supply and an 78% renewable share at one point on the Saturday daytime.

The SWIS is the largest isolated grid in the world to reach this level of battery penetration. It covers roughly 350,000 square kilometres in southwestern Australia and operates independently of the National Electricity Market that serves the country’s eastern states. Because it is isolated, the SWIS cannot import power during shortfalls or export during surpluses — every megawatt of demand has to be met from within the system. That makes it a useful real-world laboratory for what high-renewable, battery-supported grids can do.

The 37.2% figure refers specifically to the contribution of utility-scale batteries during the peak demand window, when the sun has set but household and commercial loads are still high. Batteries that absorbed solar electricity earlier in the day discharged into the evening peak — the textbook duck-curve solution. Coupled with the daytime renewable share, the weekend numbers suggest the SWIS is approaching the kind of operating regime that planners have modelled for the rest of the country.

Penetration milestones above 35% have now been recorded in California, South Australia, and Western Australia — three jurisdictions with very different geography, climate, and grid scale. The pattern suggests that the operational behavior of high-storage grids is more transferable than once feared, and that the limits are increasingly economic and logistical rather than technical.

Source: Energy-Storage.News

Ford Launches Ford Energy with Containerized Grid-Scale Battery Product

Ford formally launched its energy storage subsidiary, Ford Energy, and unveiled its flagship product on May 12: the DC block, a standardized 20-foot containerized battery energy storage system built around 512 Ah lithium iron phosphate prismatic cells. The product comes in two configurations — the FE-250 for two-hour applications and the FE-450 for four-hour applications — and the company plans to begin first customer deliveries in late 2027.

The DC block format is significant because it is engineered around the practical needs of grid integrators rather than the constraints of battery manufacturers. By packaging the battery, balance-of-system electronics, and thermal management into a single containerized unit, Ford is trying to compress the on-site work required to commission a battery project. The 512 Ah cell format itself reflects a broader industry trend toward larger-format prismatic cells, which improve energy density at the pack level by reducing the number of inter-cell connections.

Ford plans to produce 20 GWh of storage annually from its Glendale, Kentucky factory — the same facility originally built for EV battery production under the BlueOval SK joint venture. The repurposing is a practical response to a slower-than-expected EV ramp combined with explosive demand from utility-scale storage and AI data centers. Battery manufacturing capacity that would otherwise have sat idle is being redirected to a market that is currently capacity-constrained.

This is the second major U.S. automotive group to enter the stationary storage market through repurposed manufacturing — a pattern likely to continue as automakers reconcile the gap between their installed battery capacity and slower-than-projected EV demand. For the grid, the net effect is more domestic, U.S.-assembled battery supply at a moment when project developers are scrambling for cells.

Source: pv magazine

Alsym Energy and Juniper Sign 500 MWh Sodium-Ion Storage Partnership

U.S. sodium-ion battery startup Alsym Energy and California renewables developer Juniper Energy announced on May 12 a strategic partnership covering 500 MWh of sodium-ion energy storage deployments. The deal is one of the larger commitments to date for a non-lithium chemistry in a U.S. project pipeline.

Sodium-ion batteries use sodium as the charge carrier rather than lithium. The chemistry has several attractive properties for stationary storage: sodium is abundant and inexpensive, the cells can be built without cobalt or nickel, they operate well at lower temperatures than lithium iron phosphate, and they are inherently non-flammable. The trade-offs are lower energy density — meaning more physical volume per kilowatt-hour — and a less mature supply chain. For grid storage, where space is generally available and safety and cost matter more than weight, those trade-offs favor sodium-ion.

Alsym’s chemistry is positioned for commercial and industrial customers and grid-scale projects where the safety advantage allows installations closer to load — for example, inside or alongside buildings — that would face stricter permitting if they used lithium-based cells. Juniper Energy, as a project developer with a portfolio of California sites, brings the deployment side: identifying projects where the cost, safety, and siting characteristics of sodium-ion can be monetized against specific grid services or customer demand.

The 500 MWh figure is the kind of commitment that moves a chemistry from pilot to early commercial. If even half of that capacity is built and performs as specified, it would substantially improve project-finance terms for the next round of sodium-ion projects in the U.S. — and start to give utilities a credible second source against the global lithium supply chain.

Source: Energy-Storage.News

Nuclear Regulatory Commission Modernizes Processes Under Part 53

The Nuclear Regulatory Commission’s modernized licensing framework — Part 53, which became effective April 29 — is in its first weeks of practical operation, and reporting on May 13 detailed how the agency is restructuring around it. Part 53 is the first new reactor licensing pathway issued by the NRC since 1989, and the first major update to U.S. reactor licensing standards since 1956.

The framework is designed around three principles: risk-informed regulation (focusing requirements on what matters most to safety), performance-based requirements (specifying outcomes rather than prescribing equipment), and technology inclusivity (working for any reactor technology, size, or end use). The previous frameworks — Part 50, dating from the dawn of the commercial industry, and Part 52, which dates from the 1980s — were built around the assumptions and design choices of large light-water reactors, which made them awkward for advanced designs using different coolants, fuels, and operational profiles.

Under Part 53, the NRC expects approval timelines in the 18-month range for designs that meet the framework’s expectations — well below the multi-year reviews that have historically been the norm. Application costs are projected to be roughly half what they would be under the older pathways. The framework is optional: developers can still use Part 50 or Part 52, and a number of advanced reactor companies have already pursued the older routes. But for new designs, Part 53 is now the default path that the agency itself is staffing and tooling around.

Regulatory modernization is not the kind of news that generates headlines on the day it happens, but it is a structural enabler. The economics of advanced reactors depend heavily on review costs and timelines, and a faster, cheaper, more predictable pathway changes which designs can attract capital and which cannot.

Source: Roll Call

Jefferson Lab Opens Spin-Polarized Fusion Experiment

Researchers at the Department of Energy’s Thomas Jefferson National Accelerator Facility in Newport News, Virginia, announced on May 14 the launch of a multi-year experiment to test the feasibility of spin-polarized fusion fuel. The project is funded with $29 million from the DOE and brings together Jefferson Lab, the University of Virginia, and the DIII-D National Fusion Facility in San Diego.

Conventional fusion experiments use deuterium and tritium fuel in which the nuclear spins are randomly oriented. Theory predicts that if the spins of the deuterium and tritium nuclei are aligned, the fusion reaction rate can increase by roughly 50% — meaning a given plasma volume and temperature would produce significantly more power. If this effect can be demonstrated in a tokamak, it would change the economics of fusion power plants by relaxing the requirements on every other parameter.

The experimental challenge is detecting whether the polarization survives the trip through the fusion fuel cycle. Jefferson Lab’s role is to prepare and characterize the polarized fuel and to measure the very small electromagnetic signal — less than a part per million of a single electrical volt — that indicates whether the spin orientation has been preserved. UVA is contributing components; the actual fusion reactions will be measured at DIII-D, where polarized fuel can be injected into an existing tokamak and the resulting yield compared against unpolarized baselines.

Spin-polarized fusion has been theoretically attractive for decades but never tested experimentally in a tokamak. If the effect works as predicted, it would add multiplicative benefit on top of the magnetic confinement and target physics improvements that other fusion programs are pursuing. If it doesn’t survive the plasma environment, knowing that empirically is itself a useful constraint on power-plant design.

Source: WMRA

NANO Nuclear Files Construction Permit Application for KRONOS Microreactor

NANO Nuclear Energy reported on May 14, as part of its second-quarter financial results, that it has submitted a formal Construction Permit Application to the U.S. Nuclear Regulatory Commission for its KRONOS Micro Modular Reactor (MMR). The application is one of the early tests of how the NRC will process advanced reactor applications under its modernized framework.

Microreactors occupy a different niche from gigawatt-scale plants or even the 300-megawatt small modular reactors that have dominated recent nuclear conversation. KRONOS is designed for applications where the host site needs reliable on-site power but cannot economically use a larger reactor — remote communities, industrial sites, data centers, or military bases. At microreactor scale, factory fabrication and standardized designs become possible in a way that large reactors have struggled to achieve.

Construction Permit applications under the existing regulatory framework go through a structured review process before the company can begin building. The submission marks a transition from design and pre-application engagement to formal licensing review — the point at which the actual engineering documentation is scrutinized against the regulatory requirements. The progress of this application will be a useful indicator of how the modernized NRC review timelines hold up in practice.

The microreactor segment as a whole has attracted growing interest from both government and private customers, and 2026 is becoming an inflection year: several designs are entering formal NRC review, the DOE Reactor Pilot Program is pushing demonstration projects toward criticality, and procurement frameworks are emerging for the kinds of customers — particularly federal facilities and data-center operators — that the technology was designed to serve.

Source: GlobeNewswire