You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:
On Thea Energy’s $100 million Series B, plus more of the week’s big money moves.

Nuclear is once again a dominant theme this week, with fusion startup Thea Energy landing a $100 million Series B that will help it expand its magnet manufacturing capabilities. While $100 million is nothing to scoff at, it somehow sounds modest alongside some of this year’s other deals, which include a $450 million Series A for Inertia Enterprises and $240 million for Shine Technologies. This week also brought the news that small modular reactor startup Newcleo plans to go public via SPAC later this year, bringing to mind the exuberance of the 2021 SPAC boom, in a deal expected to net a cool $429 million.
Elsewhere, gridtech company Utilidata raised fresh capital after (surprise!) pivoting to the data center market, while a standalone battery storage developer and operator is betting there’s still plenty of money to be made in the increasingly crowded ERCOT market.
Thea Energy officially joined the growing ranks of fusion companies to surpass $100 million in total funding this week, raising a $100 million Series B round led by the U.S. Innovative Technology Fund to scale its magnet manufacturing operations as it targets a demonstration reactor by 2030. Thea is a part of the Department of Energy’s Milestone-Based Fusion Development Program, which seeks to accelerate efforts for commercial fusion power. In January, the DOE certified Thea’s preconceptual pilot plant design, making it the first of the program’s eight awardees — who will split $46 million in federal funding — to see its reactor architecture validated.
Unlike many top-funded fusion startups, which are building donut-shaped tokamak reactors, Thea Energy is betting on a stellarator design. Traditional stellarators resemble a helical tokamak, which require manufacturing and installing dozens of huge, twisted magnets, but Thea’s approach deviates from the norm. Instead, it relies on hundreds of small, planar magnets arranged in the more familiar donut-shaped configuration, which the company’s artificial intelligence software controls individually. That enables Thea to create the same complex magnetic field within a far simpler and more manufacturable shell.
Thea plans to use the new capital to build a second facility in New Jersey to complement its existing lab and to double its headcount as it seeks a site for its demo reactor later this year. The startup is aiming to bring its subsequent commercial pilot online by 2034, on par with the timeline laid out by fusion industry leader Commonwealth Fusion Systems. According to Gaetano Crupi, USIT founder and billionaire investor Thomas Tull “believes the stellarator is the right architecture for commercial fusion, and Thea Energy is the company that makes it commercially viable.” As Crupi put it in a press release, that’s because “Thea Energy’s breakthroughs shift complexity from precision mechanical fabrication to software-defined controls.”
Newcleo is the latest small modular reactor startup seeking a quick pathway to the public markets via a SPAC merger, announcing plans to list on the Nasdaq in the second half of the year after merging with a blank-check firm. The deal values the European fuel and reactor developer at $2.4 billion, and is expected to deliver about $429 million in fresh capital. It comes just months after Newcleo raised $88 million in a growth financing round as the company expands into the U.S. market while continuing to fund projects across Europe.
Newcleo stands out in the crowded SMR field through its fuel and cooling strategy. It plans to run its 200-megawatt reactors on recycled fuel made from nuclear waste products like recovered plutonium and depleted uranium, and cool its reactors with liquid lead rather than water. Because liquid lead has such a high boiling point, lead-cooled reactors can operate at atmospheric pressure, reducing the need for the complex, high-pressure systems used in conventional nuclear plants and potentially improving safety along the way.
The company has already raised over $760 million to date, and CEO Stefano Buono told the Wall Street Journal that the pending SPAC could carry it through 2028 or 2029. Even that won’t be enough, however, for Newcleo to reach its target of opening a fuel factory by 2031 and bringing a commercial reactor online the following year. Not to mention that SPACs — a once rare go-to-market strategy — have a checkered history in the SMR industry. After NuScale went public via SPAC in 2022, its flagship project collapsed, taking its stock down with it and underscoring the risks that pre-revenue companies face when their early failures unfold in the public markets. On the other hand, shares of Sam Altman-backed startup Oklo’s have surged since it went public via SPAC in 2024, reaching a market cap over $11 billion, though it also has yet to build a reactor.
Newcleo’s capital push may also be tied to its strategic partnership with Oklo, as it has preliminary plans to invest up to $2 billion to develop advanced nuclear fuel facilities in the U.S. in partnership with the SMR pioneer. Earlier this week, the DOE selected Oklo — and by extension, Newcleo — to enter “advanced negotiations” to receive surplus weapons-grade plutonium for use in reactor fuel.
What’s that I hear? Another climate tech company has pivoted to the data center market? While Utilidata — an artificial intelligence-powered gridtech company — initially set out to give utilities granular insight into household-level electricity usage and grid data, it’s now raised a $40 million extension round to accelerate its shift into the data center market. As I wrote following last year’s initial $60 million tranche of Series C funding, Utilidata initially set out to get its hardware module inside residential smart meters — which it managed to do at pilot scale — to enable faster fault detection and eventually even automate load management at the household level.
Now, Utilidata is taking this same principle and applying it to the booming data center market, where so many climate tech companies are finding their first customers. The company developed its AI platform in collaboration with Nvidia, installing its modules on server racks to help data centers optimize power allocation across its facility. The company says it measures power consumption a million times per second, such that if usage on one rack is low, it can reroute electricity to parts of the data center that need it. Much like electric grids, data centers also overbuild their capacity to ensure they can handle sudden spikes in demand or hardware failures. Utilidata wants to tap into that headroom by managing power flow in real time.
Utilidata’s first commercial data center deployment is set to go live next month in Montreal in partnership with European AI cloud provider NexGen Cloud, with the startup targeting a 50% increase in the data center’s usable processing power. It also plans to use this latest funding to increase headcount by 25% this year as it builds out operations at its new Ann Arbor headquarters, which opened in February.
In some later-stage funding news, battery energy storage developer, owner, and operator Goshe Energy Storage just secured up to $40 million in strategic financing from S2G investments. As I wrote last week, S2G recently raised a $1 billion fund aimed at helping growth-stage companies commercialize, though this latest commitment actually comes from a different arm of the firm — its Special Opportunities team. This division focuses on non-dilutive financing, in this case providing Goshe with a HoldCo loan backed by the company’s portfolio of energy storage projects. Rather than lending to a specific project, a HoldCo loan gives Goshe flexible capital that can be used to fund its broader growth.
Founded in 2022, Goshe specializes in acquiring late-stage battery storage projects and getting them over the finish line by securing capital and managing the construction process into commercial operations. Thus far, all of its announced projects are in Texas’ ERCOT electricity market. Alongside this financing announcement, Goshe said that its first project — a 100-megawatt battery storage plant in Bexar County, Texas — is now fully operational after securing $288 million in project financing. The company also expects to bring its second project, a 180-megawatt storage facility, online in the following few months, with two additional ERCOT projects slated to begin construction later this year.
This funding is the latest sign that infrastructure investors have grown comfortable backing battery energy storage projects, with a record 24.3 gigawatts of new battery storage capacity projected to come online in the U.S. this year alone. The wholesale ERCOT market, however, is no longer the guaranteed moneymaker that it was just a few years ago. Between January 2024 and January 2026, ERCOT more than tripled its battery storage capacity, driving battery revenues down as the market has become increasingly crowded. In this landscape, there may be a growing number of stranded projects for Goshe to acquire, though it’ll also have to be increasingly selective.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Roads bring people, and people start fires.
The United States has more roads than you could possibly imagine. Eighty-three percent of the land in the Lower 48 lies within one kilometer of a road; if you’re seeking isolation, the furthest you can get away from one is likely only about 25 miles, in a far-flung corner of Yellowstone National Park.
The Trump administration wants to build even more. Earlier this week, the U.S. Department of Agriculture filed to rescind the nearly quarter-century-old Roadless Area Conservation Rule, which protects about 45 million acres of pristine national forest lands from the construction of — and dissection by — new permanent roads. The USDA’s given reason? That new roads will provide the access forest managers and fire practitioners need to better prevent wildfires in the nation’s most remote locations.
Fire ecologists immediately cried foul. Researchers have long understood that roads enable wildfire ignitions by bringing people — who are responsible for starting nearly 85% of fires — into the landscapes. Just this past January, new research found that wildfires ignited four times more often within 50 meters of a road than in an untracked, intact forest. “The notion that you can prevent fires by building roads seems to me precisely backwards when you look at what the science says,” Ben Goldfarb, the author of the road ecology book Crossings, told me.
But this past spring, Americans got a good idea of what wildfires look like when there aren’t roads around. Lightning storms in Northern Ontario ignited fires in an area so remote that officials found it “impossible to get firefighters on the ground” to fight them, per The New York Times, or even to react early with airplane water tankers. The result? More than 1.8 million acres burned in the province so far this year, with the resulting smoke causing the Midwestern U.S. and New England to experience some of its worst air pollution in decades.
“There’s a duality — roads are neither necessarily good nor bad from a fire perspective,” Eric Kennedy, an associate professor of disaster and emergency management at York University, told me. “They bring opportunities for ignition and they bring opportunities for firefighting.” Those opportunities include the aforementioned access for fire personnel, as well as serving as a fuel break so crews can gain a foothold against an approaching conflagration. In a populated area, more roads can also mean more evacuation routes when there is a disaster, preventing potentially deadly traffic jams.
Forest defenders were already suspicious of the administration’s motivations when it comes to wildfire policy. “There’s all of the Trump administration directives to increase logging on public lands, which rescinding the Roadless Rule helps to facilitate,” Goldfarb noted. Environmental groups have pointed to attempted legislation such as the Fix Our Forests Act, which removes obstacles for forest management methods, including timber harvest, as another example of how the administration is allegedly using wildfire as a cover to cut down and sell more trees.
Viewed in the context of recent changes by the administration to weaken the Endangered Species Act — namely, narrowing the definition of “harm” to a species to exclude disturbances to its habitat — rescinding the Roadless Rule can appear to follow a kind of rapacious internal logic that “wildlife doesn’t need habitat, and we can build roads wherever we want to disrupt” the forest, Goldfarb went on.
Fires igniting in remote areas is also not a new problem. Agencies adapt to the fire conditions in their areas, such as Quebec, which has an entire apparatus for fighting fires in tractless wilderness, including shuttling in fire crews via float plane. “You can fight fires via helicopter. You can also build temporary roads under the Roadless Rule,” Goldfarb said. As one Montana-based National Forest manager of 25 years recently wrote for a local newspaper, in his experience, “the Roadless Rule doesn’t pose an insurmountable barrier to good land management; it simply requires baseline analysis and thought before impacting the landscape.”
Those who are cynical about the Trump administration’s motivations also pointed me toward the grandiose scale of the Roadless Rule rescission. Fire managers frequently talk about the need for tailored, local, and precise responses to America’s wildfires, which run the gamut from grass fires to chaparral fires to forest fires in regions that both do and do not have histories of regular burning. Policymakers would more appropriately approach wildfire management fireshed by fireshed, they say, and through proposed management plans. Perhaps most notably, the Roadless Rule protects about half of the nearly 17 million acres of the Tongass National Forest, a temperate rainforest and one of the wettest locations in North America, which “does not experience wildfires like those in other places,” the Alaskan environmental conservation group SalmonState wrote in a statement with other advocates and business groups.
Most cynical, though, is the argument that the Trump administration is proposing rescinding the Roadless Rule at the same time that it has gutted the Forest Service that is supposed to maintain all those roads. The agency already struggles with an overwhelming backlog of maintenance projects, from washed-out bridges to erosion problems that impact the water quality in drought-stressed areas. If the USDA were really interested in using roads to combat wildfires, the line of thinking goes, then it would be investing more in the Forest Service, people told me, not less.
“The wildfire challenge really calls upon us to be able to hold different dimensions and different layers and seemingly contradictory ideas at the same time,” Kennedy said, again emphasizing that one can make the case that roads have benefits in certain contexts and scenarios. But while there may be a valid line of debate about when, where, and how roads can help with wildfire management, using the cudgel of a rescission, it doesn’t appear to be one the administration is interested in having.
The facility will power OpenAI’s 10-gigawatt data center in Pike County, Ohio.
The Trump administration aims to complete its environmental review of what would be the biggest fossil fuel power project in the country in just a few months, Heatmap has learned.
This news follows Monday’s announcement from OpenAI that it intends to lease a new 10-gigawatt data center under development in Pike County, Ohio, financed by a mixture of money from a SoftBank subsidiary and the chip company Nvidia. This AI hyperscale facility — known as the PORTS-Pike project — is expected to draw power from the largest gas power facility ever built in the United States, a 9.2-gigawatt facility sited on federal lands that would be built and owned by the Energy Department.
According to OpenAI, the data center campus will be built and started up in phases, with the first 800 megawatts starting construction this year and operational in 2028. That first phase will rely mostly on existing power infrastructure operated by AEP Ohio. How things progress from there will depend at least in part on the permitting and construction timelines for the new power plant.
Building large infrastructure of any kind on federal land or with significant federal investment typically triggers a review under the National Environmental Policy Act. I’ve been curious to find out what kind of review this particular project was going to get, especially after the administration allowed a NEPA review for a solar project to be repurposed for a data center on federal lands earlier this year.
Turns out some information about the PORTS-Pike permitting process is public. Before OpenAI confirmed its involvement with the site, the Trump administration added the project to the federal FAST-41 permitting dashboard, where it posts regular updates on the timeline for getting federal sign-offs. Per the lone federal notice available about the PORTS-Pike project, it will include “several data center buildings and power plants.” That will require at least two federal greenlights: an Army Corps of Engineers permit and approval from the Fish and Wildlife Service, which is being consulted about potential endangered bats in the project area.
The NEPA permitting work for this historically large data center-plus-fossil fuel power project began on July 10 and will conclude on December 23, the day before Christmas Eve, according to the Trump administration’s estimates. This comes after paperwork to begin the review was submitted to the Army Corps in May, per the federal notice — a total timeline of about seven months.
Those familiar with NEPA and the debate over permitting reform will likely be surprised by the speed of this review. It’s moving fast in part because the project is receiving just an Environmental Assessment, the lesser and smaller type of analysis than the EIS. I do not know why the government decided to take this route because the government’s NEPA review determination is not currently public, but I have asked the Army Corps to explain this move.
I’m not sure exactly how air permitting will fit into this NEPA review, as the Clean Air Act isn’t listed as a review step on the federal dashboard. The Ohio EPA has primary authority over permitting projects like these under the Clean Air Act, and I’ve reached out to them to confirm whether PORTS has submitted a permitting application. The state agency’s permitting database does not have any information on air permitting for the project, though it does include reports from third-party consultants confirming wetlands and protected species warranted reviews from the Army Corps and Fish and Wildlife.
Lastly, these timetables are not sacrosanct. Under the Fiscal Responsibility Act of 2023, agencies are supposed to complete environmental assessments within one year, but nevertheless they regularly fail to meet them. The White House’s Council on Environmental Quality said in a report to Congress last year that from mid-2023 to mid-2025, the Army Corps was the agency that most often missed these statutory NEPA deadlines for environmental assessments.
Still, news of this speedy review for a priority Trump project is sure to excite pro-data center advocates who see expedited construction as an imperative in the global AI arms race. It’s also guaranteed to put a foul taste in the mouths of environmentalists already frustrated by federal revisions to NEPA regulations they say elide analysis of climate impacts.
What’s undebatable in all this is that, as my colleague Robinson Meyer wrote, the PORTS project could ignite a new era of mega-gas plants. This permitting timeline couldn’t be more important for the future of the data center boom — and the nation’s greenhouse gas emissions.
SB Energy, the SoftBank subsidiary behind the data center project, did not provide comment before publication.
A new front opens in the data center wars.
A series of lawsuits filed in federal court asks a big question – are data center moratoria constitutional?
In early August, data center developer DC Blox sued the city of Nashville in federal court to overturn a zoning moratorium stopping them from building a hyperscale facility adjacent to the city zoo. “The Data Center Moratorium, moreover, is a targeted attack against DC BLOX, in violation of federal constitutional protections,” the suit argued, claiming that it defied the corporation’s due process and equal protection rights.
Around the same time, another developer – Wixom Industrial One – filed a federal lawsuit against the city of Wixom, Michigan, to try and “invalidate the city’s illegal police power moratorium” blocking their data center.
These two cases were far from novel or the first of their kind, and they’re now a fresh front in the battle over hyperscale data centers. At least that’s what some who work on these cases say: In April, attorneys with the law firm Vorys published a “client alert” asserting “many moratoria may be vulnerable to statutory, procedural, and constitutional challenges.” The attorneys advised that constitutional arguments against moratoria “may be stronger where a government singles out data centers without a sound factual basis, treats similar land uses differently without a reasonable basis, or adopts a restriction driven more by political pressure than by defensible planning or regulatory objectives.”
Months later, according to court documents, the Vorys attorneys who authored the alert now represent real estate firm Thor Equities in a federal case against the Ohio city of Urbana, arguing the city’s decision to reject their data center project broke “fundamental protections” under the U.S. Constitution. (Vorys and Thor Equities did not respond to requests for comment.)
It’s unclear how many of these kinds of cases have been filed to date. Data on federal court cases is quite opaque. But legal experts and industry attorneys tell me we should expect them to be on the rise as developers seek whatever tools they can find to get projects built.
“Bringing a lawsuit like this is fairly cheap, something they can do at a relatively low cost, and imposes a real cost on local governments to defend themselves,” said Daniel Metzger, director of the Cities Climate Law Initiative at Columbia Law School’s Sabin Center. “The cases out there will be bellwethers. And if successful, there’ll be a lot more of them.”
What developers probably want looks a lot like Hill County, Texas, where an LLC proposing an $80 million data center project was stymied in May by the state’s first countywide moratorium. (It predated Governor Greg Abbott’s temporary freeze of data center development in Texas by three months.) Within a period of only a few weeks, the LLC sued and the county rescinded the pause on approvals. The case was dropped a month later. Local reports state the county had to afterwards pay the corporation $100,000 in legal fees – a drop in the bucket compared to what a drawn-out court battle would have cost the rural county.
Metzger said whether the companies will win these cases is ultimately not the point – their goal is to win a finished data center, not a judicial ruling. By filing expansive litigation in the national court system, a hypothetical developer can exhaust the coffers of a city or county with legal expenses that are chump change compared to would-be billions in private financing for compute infrastructure.
“These lawsuits may deter some local governments from taking steps to oppose data center development, just because of the cost it would impose on them to defend a lawsuit, even if they know they have a strong legal basis for the action they want to take.”
Those I spoke to in private practice about data center developers’ constitutional arguments agreed with Metzger’s assessment that it’s too early to tell whether the companies will win. Generally, they said, a city or county will win this kind of case if it demonstrates a rational basis for its decision-making and courts typically want to defer to governmental autonomy. The onus will be on the developers to prove a moratorium was meritless – that’s the due process challenge – or unfairly targeted their industry in a way other sectors don’t face, which is the basis of the equal protection claim.
“What they’re saying is in essence that these actions the municipality is taking are arbitrary and capricious, which is one of the sort of catch-all standards,” Thomas Allen, a partner at K&L Gates, told me. “They say the laws lack a rational basis. And then they make equal protection claims, saying data centers are being singled out because of political concerns as opposed to actual things relevant to the legislature’s directive. They’re not basing their decisions on the underlying merits of the project but reacting to political pressure.”
“It’s a reliance question and it’s about the treatment of their projects,” added Laura Morton, an attorney with Ashurst Perkins Coie. “It’s always been important to talk about and engage with communities where your infrastructure is planned. Here, I think this is the developers going in, maybe having conversations, and then suddenly they’re getting a reversal after already receiving these approvals and making investments based off of what the conversations and rules were.”
The likelihood of these constitutional challenges reaching higher courts anytime soon is quite low. It’ll be a long time before we see one of these cases reach a verdict, let alone some kind of appeals process come to fruition. Nevertheless, the new legal ambiguity around these local restrictions is an important new facet of the data center wars, including for developers.
“Companies want to act within the law to get [things] done, so whatever tactics they can do to help get the project over the line that are legal and ethical, they may try those,” Allen told me. “And if that includes the pressure of a lawsuit, that’s a judgment they’ll have to make.”