On 1 November, it’s lights-out for power inverters from “high-risk” suppliers to European Union-funded grid projects. The EU will no longer pay for Chinese inverters for these projects, although member nations still can if they so choose. As a result, companies across the EU are reviewing their supply chains and preparing to power their fast-growing grids with fewer Chinese components, which the EU has assessed can enable backdoors to remotely control or shutdown grid infrastructure.

The U.S. is enacting its own restrictions for similar reasons. In July, the U.S. Federal Communications Commission added power inverters to its list of foreign technologies that require certification before sale in the country. Then, in August, the White House issued a more expansive executive order banning foreign inverters, transformers, battery storage, generators, and software and firmware. The order also asserts federal authority to review and modify existing installed equipment, but leaves the details to the U.S. Department of Energy (DOE) to explain in a rule in late December.

“The details of the DOE rule are what are going to cause some drama,” says George Gross, emeritus professor of electrical engineering at the University of Illinois in Urbana-Champaign and a former longtime Pacific Gas and Electric Company (PG&E) engineer and manager. “I was surprised that the community did not exactly scream, ‘We are all going to be overwhelmed.’” But that may be because they see enough latitude for purchasing what they need, Gross adds, even as the shape of the supply chain will change.

Power Inverters and Grid Security

Power inverters are a key component for converting DC electricity from sources such as solar farms or long-term battery storage into AC electricity for the grid.

In 2025, industry group Green Power Denmark detected hidden communication channels in some circuit boards bound for installation on power inverters. In a different incident, Célistine Oosting, a security researcher at the Dutch Institute for Vulnerability Disclosure, discovered publicly exposed credentials that allowed firmware updates to her parents’ solar panels. And while there is no public proof of a Trojan horse-type attack on an electrical grid in the U.S. or EU, the vulnerabilities show how enterprising hackers could open a back door.

The risks extend beyond inverters to other components in the long-distance transmission part of the electrical grid, collectively called bulk-power equipment. For example, in 2015 and 2016, hackers used existing weaknesses in Ukrainian electrical equipment to cause blackouts, including by controlling switchboards and circuit breakers.

The U.S. DOE has been wrestling with how to handle grid hacking since before then, and other countries have since developed their own approaches. India’s energy authority has required laboratory malware testing of foreign components since 2020, among other measures. Japan’s equivalent authority has since 2024 required a 30-day advance screening before approving the installation of certain power equipment.

The European purchasing decision, which does not apply to nationally or privately-funded projects, may still force energy companies to replace 4 to 5 gigawatts’ worth of planned bulk-power equipment purchasing from Chinese suppliers from now until 2030, energy consultancy Wood Mackenzie estimated in July. That’s about 14 to 18 percent of the EU’s planned purchases over that period. Inverter manufacturers, renewable energy suppliers, and investment bankers met with EU officials in June to discuss their ability to replace that Chinese manufacturing capacity. The upshot: It is possible, but will be slow and costly.

EU and U.S. Power Supply Chains

EU and U.S. power-supply companies are trying to update their supply chains, even though they know the rules will probably continue to change in the short term. Some are trying to make those rules better suit them.

For example, inverter manufacturer Fronius, in Pettenbach, Austria, warned in June that replacing the Chinese components in its supply chain could take a year. It also advocated for the EU to enact an outright ban on new Chinese bulk-power components rather than just saying it won’t pay for them. That way Fronius and other companies in its position can justify the upfront investment in new, costlier suppliers.

In September, the DOE requested public comment on its rulemaking. Some requested clarification of what components DOE would consider “associated with” bulk-power equipment and recommended keeping the list as narrow as possible to address only elements that allow control of grid components. At least one supply chain verification firm proposed requiring risk-based sampling and physical tests, rather than relying on paper-based statements from importers. The firm also argued that the DOE could define a system in which accredited laboratories that already evaluate other sensitive equipment, such as aerospace components, compete to provide testing services, to add security while keeping down costs for utilities.

Other companies are not waiting around for new rules to make new plans. SMA Solar, in Niestetal, Germany, estimates it stands to gain around 10 percent share of Europe’s utility-scale market, its chief executive said in August, thanks to the fact that it only depends on China for a small percentage of its supply chain (around 2 to 4 percent) and has the capacity to ramp up EU production.

Some U.S. suppliers also stand to benefit. Unico, which is headquartered in Franksville, Wisconsin, designs and assembles a range of energy equipment, including inverters, in the U.S. The company, which also has a presence in the EU, designs its own printed circuit board assemblies, field-programmable gate array-based control loops, and magnetic components (including inductors and transformers) in-house. This allows Unico to retain direct control over firmware development and system architecture. The company does import components from abroad, but already made a point of carefully reviewing its supply chain. “We’re always looking at the sources of supply, traceability, qualification, and commercial scale capacity,” says Unico CEO Michael Canada. “And we’re always looking at the geopolitical risk, and we work to keep those high-risk areas out of our supply chain.”

Canada says it’s hard to tell just how long it will take for U.S. and European companies to replace banned components from their purchasing and existing installations, given the complexity of the supply chain. But the work is evolving, as governments make their concerns more explicit. For example, “cybersecurity is becoming a key aspect in our purchasing decisions,” Canada says.

The most direct way for some of Unico’s competitors to gain trust and access to the market may be to establish a new U.S. manufacturing presence, but that will take time. Right now, grid operators are eager to build as fast as they can to keep up with exploding electricity demand from sectors including AI data centers. For the foreseeable future, Canada says, “There’s a substantial opportunity being an American engineering and manufacturing company.”

This article was originally published by Canary Media.

America’s energy supply crunch is driving a surge of federal and private investment in new electricity generation—from polluting gas turbines to cleaner developments like solar farms and geothermal plants. But gigawatts of power are already available for the taking across the U.S. economy in the form of wasted energy from an unexpected source: motors.

Some 50 million electric motors drive the fans, pumps, and compressors found in virtually every factory and building. Together, they gobble enough electricity to equal nearly a third of total U.S. power demand. Often, the motors use far more energy than they actually need to keep America’s machines humming.

Companies can install devices known as variable frequency drives, or VFDs, to make motors run super efficiently. Yet adoption remains relatively low, in part because of the perceived hassle of retrofitting equipment and because they’re typically not required. On average, only 16 percent of motor systems at industrial facilities and 4 percent of commercial-sector motors use VFDs, according to a 2021 market report by Lawrence Berkeley National Laboratory.

However, there are now more reasons than ever to equip the nation’s motors with power-saving devices, energy experts and device manufacturers say. They’re urging federal agencies, utilities, and deep-pocketed tech giants to spur a wider rollout of VFDs through financial incentives and efficiency standards.

For one, slashing motors’ power use would free up grid capacity for large electricity users like data centers and electric vehicles, they argue. It would help curb costs for factories that are switching from fossil-fueled heating systems to electrified, motor-based solutions like heat pumps—all at a time when industrial power prices are climbing fast. Plus, VFD technology has become far more sophisticated and less expensive to install than earlier iterations.

“This is low-hanging fruit that gives a lot [of energy] back and is reasonably simple,” Nik Sawe, a senior policy analyst at the nonpartisan think tank Energy Innovation, says of the technology.

What are Variable Frequency Drives?

Energy-wise, the main problem with electric motors is that they’re always running at full capacity. Consider an office building, which empties out in the evening. The heating, ventilation, and air-conditioning system can wind down in those hours, with dampers and valves restricting the flow of cold or hot air. But the motors and fans that drive the whole setup are still working in the background, wasting energy.

VFD technology, such as the Danfoss iC7 series, has gained some momentum in recent years among factories and buildings trying to combat rising electricity costs. Danfoss Drives

VFDs are electronic controllers with temperature sensors that continually adjust the speed of motors, automatically ramping up or down to match demand. The technology, which is also called adjustable speed drives, has the potential to curb a motor’s energy use by around 40 to 60 percent, says Neal Elliott, director emeritus of the American Council for an Energy-Efficient Economy.

All those energy reductions add up. Toyota, for example, installed VFDs on over 200 motors inside the HVAC system at its auto manufacturing plant in Georgetown, Kentucky, as part of a US $1.25 million pilot project in 2015. The automaker saved more than $700,000 in annual electricity costs thanks to the VFDs, and it continues to use the technology at other manufacturing facilities across the U.S., the company said by email.

“We’ve known for many, many years that there’s a huge opportunity for energy savings by varying the speed of the motor,” says Elliott, who noted that he literally wrote the book on energy-efficient motor systems in 2002.

Electric Motors in Nearly Every Factory

Certain aspects of the technology have historically held the drives back since they debuted in the late 1960s. Measuring the flow of water, air, refrigerant—whatever is moving through the system that the motors are driving—was for decades an expensive engineering process. Accurately controlling motors to deliver the right amount of juice, at just the right time, faced similar issues.

But recent advances in sensors and machine-learning algorithms have significantly slashed the cost and complexity of using drives, Elliott says. Modern VFDs have also benefited from the increased production of advanced power chips that are used in high-voltage applications such as electric vehicles and solar power inverters.

Another long-standing challenge has been that commercial and industrial building developers often prioritize reducing upfront costs over netting long-term energy savings, and so they don’t always build to the highest efficiency levels, says Zack Valdez, senior director of regulatory and industry affairs at the National Electrical Manufacturers Association, whose members include drive and motor makers.

“If the facility that you’re putting it into doesn’t require it, then oftentimes people will overlook [VFDs], because they’re just trying to build to a minimum set of energy requirements,” he says. And if companies don’t install drives from the start, they’re likely less inclined to take time to retrofit an operating system.

The trade association recently launched a pilot program with manufacturers, utilities, and energy-efficiency groups to collect real-world data about how VFDs reduce motors’ electricity use and lower utility bills. The goal is to demonstrate how the technology can benefit not only individual facilities but also the wider grid—giving utilities a concrete reason to provide rebates for industrial customers.

VFDs Slash Motors’ Electricity Use

Proponents say they hope the results will help the U.S. Department of Energy shape future efficiency measures and incentives that encourage adoption of the drives, including in federally owned buildings. Despite the Trump administration’s open hostility toward home appliance and equipment mandates, agency staff have apparently been receptive to discussing solutions that could meet the soaring power needs of data centers.

“If you added a drive to all the motors that are out there, you could provide [power for] up to five years of data center growth without having to build any new generation,” says Kirk Anderson, director of government relations and public affairs for ABB. The global manufacturing giant makes VFDs domestically near Milwaukee.

Anderson floated the idea of having tech companies help pay for factory owners to install VFDs to free up room on the grid. After all, hyperscalers are already spending many billions of dollars to revive shuttered nuclear plants, support next-generation geothermal projects, deploy fleets of grid-scale batteries, and supercharge the build-out of gas turbines and power plants.

“Instead of ‘bring your own power,’ it should be ‘offset your own power,’” he says, referring to a growing effort by states to get data centers to secure their own power resources and prevent rising costs for regular customers.

Michael Reckamp, the director of enterprise strategy for Danfoss Drives, North America, another VFD manufacturer, said that despite the traditionally low adoption rates, the technology has gained some momentum in recent years among factories and buildings as rising electricity costs and other grid-wide challenges make it harder to justify wasting energy.

Still, “it hasn’t permeated enough to get everyone utilizing it,” he says. “In the current situation, where we’re having so many issues with grids, brownouts, and now data centers wanting to take the load, I think there should be a renewed focus on putting drives out there to really reduce the stress on the grid and improve energy savings.”

One of the singular features of Baotou, an industrial city of 2 million people in Inner Mongolia, is a 10-square-kilometer (4-square-mile) lake that is among the most polluted in the world. The artificial lake, referred to by the name of the dam associated with it, Weikuang Dam (or Baogang Tailings Dam), has for decades collected the effluent from minerals processing for China’s globally dominant iron, steel, and rare-earths industries.

Built in the late 1950s, the lake contains a highly toxic stew of industrial chemicals with substantial amounts of radioactive thorium, the latter resulting primarily from the processing of monazite ores for rare-earths. China has since greatly tightened its environmental regulations and restrictions on mining and minerals processing. For example, largely because of the environmental disaster that is Weikuang Dam, the country doesn’t permit the development or operation of stand-alone monazite mines—that is, mines where nothing is mined except monazite ore.

Now, with industrialized nations in North America, Asia, and Europe spending billions to shore up their rare-earths supply chains, and counting on monazite to help them do it, China’s history and experiences with monazite offer important lessons.

This article is a collaboration between Foreign Policy, the global magazine of politics and ideas, and IEEE Spectrum, the flagship magazine of the IEEE.

Monazite ore is a pillar of China’s rare-earths industry because it is a prime source of some less common but extremely important rare-earth elements, notably dysprosium, terbium, yttrium, and erbium. The country mines and processes vast quantities of monazite ore; it also imports huge amounts of monazite ore for processing.

Most of the domestically mined monazite ore comes from a single mine: the mammoth Bayan Obo complex, 150 kilometers north of Baotou, in Inner Mongolia. Bayan Obo, which is primarily an iron-ore mine, also happens to contain the largest known rare-earth deposit in the world. It’s the source of nearly half of all the rare-earth ore mined on the planet in a typical year. Most of the rare-earth ore mined at Bayan Obo is a type called bastnaesite; it’s the mining of that bastnaesite that makes it legal to mine monazite ore at Bayan Obo as well.

Related: Can a Radioactive Ore Break China’s Grip on Rare Earths?

All told, ore from Bayan Obo was used to produce 170,000 tonnes of rare-earth metal last year, according to David Merriman, research director at the minerals intelligence firm Project Blue. Roughly 30 percent to 40 percent of the rare-earth ore mined at Bayan Obo is monazite. That amount alone would be enough to make China the world’s largest processor of monazite for rare-earths.

China’s 2025 Monazite Imports Topped 70,000 Tonnes

But a more detailed picture of China’s interest in monazite emerges when you consider the country’s imports of the ore. Last year, they amounted to more than 70,000 tonnes, says Thomas Kruemmer, a Singapore-based consultant on minerals and metals. That figure becomes more impressive when you consider the unique challenges of shipping monazite by sea. Because of its radioactivity, concentrated monazite ore is classified as Class 7 dangerous cargo under the International Maritime Dangerous Goods Code. That means, Kruemmer notes, that the ore can be imported through only four ports in China.

In addition, importation of some kinds of monazite into China is supposed to go through a joint venture consisting of the China National Nuclear Corp. and Shenghe Resources, which are at least partly state-owned. But lots of imported monazite doesn’t go through those organizations, notes Merriman. That’s because heavy mineral sands, which may contain roughly 2 percent or 3 percent monazite by weight, are not subject to the same import controls as concentrated monazite ore. So companies import the heavy mineral sands; separate out various profitable concentrates, including the monazite mineral concentrate; and then process the monazite concentrate to recover the rare-earths.

China has lately stepped up its efforts to secure rare-earth ore around the world. One reason, says Kruemmer, is simply to make up for recently lost supplies. In 2025, the U.S. government invested US $400 million in MP Materials, which operates the Mountain Pass mine and processing facility in California. The deal put an end to long-standing contracts under which MP Materials had been selling tens of thousands of tonnes of rare-earth concentrate annually to a trading affiliate of Shenghe Resources.

But other Westerners speculate that China has strategic reasons for stepping up its efforts to acquire heavy mineral sands or other rare-earth assets abroad. By locking down rights to deposits or projects abroad, Chinese organizations are making them unavailable to non-Chinese developers, according to this reasoning.

Merriman points out that your view of these activities largely depends on where you sit. The U.S. government recently made a conditional loan commitment of up to $725 million to U.S. monazite processor Energy Fuels, which has been aggressively securing substantial monazite supplies in Australia, Brazil, and Madagascar. To a Chinese observer, those moves probably look rather strategic.

China, meanwhile, is doing what it has been doing for decades. “China’s playing its game,” says Merriman.

This article appears in the November 2026 print issue as “Radioactive Realpolitik.”

The global campaigns to break China’s dominance of vital rare-earth markets are shaping up into one of the signature geostrategic struggles of the early 21st century. Hundreds of mining companies, government agencies, and industrial facilities scattered across North America, Australasia, and Europe are striving to wrest share from Chinese firms, which control roughly 90 percent of the markets for rare-earth oxides and rare-earth metals.

Rare earths are the ingredients on which vast and crucial manufacturing categories depend: semiconductors, LEDs, lasers, displays, metal alloys, optical lenses, and catalysts for petroleum and chemical processing. Most important, though, are permanent magnets. Essentially all powerful electric motors are based on rare-earth permanent magnets. These motors are the heart of electric vehicles, heating and cooling systems, robots, appliances, and tools. Rare-earth magnets are also in the motors for military systems ranging from tactical drones to ballistic-missile submarines.

This article is a collaboration between Foreign Policy, the global magazine of politics and ideas, and IEEE Spectrum, the flagship magazine of the IEEE.

Succinctly put, any industrial power that wants to remain competitive in vital, high-tech industries must have access to a steady supply of rare-earth elements or products based on them. The traditional supply chain for these elements, originating in China, is increasingly fraught—so much so that some analysts are beginning to use the term “mineral war.” The countries that are now scrambling to piece together alternative supply chains, whether by themselves or in cooperation with each other, face daunting odds. Never mind the sheer momentum of the Chinese rare earths juggernaut. Few industries can match the rare earths business for instability, risk, intrigue, and sheer frustration.

And yet, many rare-earth experts see a way forward, and it depends on a specific kind of ore, called monazite. The ore contains huge amounts of valuable rare earths, including neodymium, praseodymium, and dysprosium.

“The future is monazite,” declares Jack Lifton, an analyst, consultant, and author. Lifton, who has been studying the rare earths industry since 1962, has seen countless trends, companies, and technologies come and go, and he isn’t given to sweeping statements. But on the subject of monazites, he’s bullish. “Right now, in my opinion, monazite is the most important rare-earth ore,” says Lifton, who is also cochairman of the Critical Minerals Institute, an organization that’s promoting the creation of a North American supply chain.

Related: Chasing China’s Monazite Monopoly

That path forward isn’t going to be smooth or straight, Lifton and others are quick to acknowledge. Monazite typically contains radioactive thorium and uranium, which means its handling, storage, and processing are tightly regulated by government agencies. It can be processed only with special techniques and procedures, and the mildly radioactive wastes generated by processing must be stored in isolation indefinitely.

Nevertheless, major investment is now flowing to monazite mines and processing technologies. In the United States, Energy Fuels, which operates the country’s only conventional uranium mill, has developed its own extraction process and is producing mixed oxides of neodymium and praseodymium—and also pilot-scale quantities of dysprosium and terbium oxides, key heavy rare earths that are otherwise produced almost exclusively in China.

Energy Fuels is processing monazite into rare-earth oxides at its White Mesa Mill in southeastern Utah. The body of water at left in the image is a tailings impoundment, where radioactive waste from processing is stored.Energy Fuels

The company has spent more than US $200 million to secure monazite assets in Brazil and Madagascar, and plans to scale its rare-earth production to higher commercial levels, at an estimated cost of $410 million. Energy Fuels’ rare-earth ambitions received a significant boost in June, when the Trump administration announced that it had signed a $725 million conditional loan commitment with the company.

Meanwhile, though, a government-backed project in Saskatchewan, Canada, stands as a cautionary tale for monazite backers. After six years of construction and expenditures of more than CA $100 million (US $72 million), a processing plant there has produced only minuscule quantities of rare earths. The processing plant, built by the Saskatchewan Research Council, was supposed to be producing 40 tonnes of rare earths per month by the end of 2024. That didn’t happen. “Full operation,” with production of 400 tonnes of metal per year was planned for “early 2025.” That didn’t happen, either. The organization is now targeting late 2027—by which time the project will have cost more than CA $216 million, according to the Saskatchewan Research Council’s CEO, Mike Crabtree.

Outside of North America, major monazite projects are also planned by Astron Limited, Hastings Technology Metals, Iluka Resources, Mkango Resources, Sovereign Metals, Steenkampskraal Holdings, and Tronox, among others. Perth, Australia–based Lynas Rare Earths, the largest rare-earths company outside of China, is expanding its monazite mining and processing operations. Most notable is the Lynas mine at Mount Weld, Australia, one of the world’s top monazite resources; the ore is processed at a sprawling facility in Malaysia.

Of course, for countries outside of China, all this surging activity doesn’t guarantee a route to rare earths independence. It’s not even close. So the vital questions now are, What else has to happen to broaden and diversify the rare-earths supply chains? And can all of this be done safely?

What Are the Advantages of Monazite Ore?

Planet Earth has enormous quantities of monazite ore, in two different forms. In Australia, Canada, China, and South Africa, massive deposits of monazite take the form of veins of hard rock that run through mountains and calderas. Much of the new interest in monazite, though, is focused on a second type, called heavy mineral sands. The sands are the result of tens of millions of years of exposure to wind, water, and gravity crumbling mineral-rich rocks into grains, which then accumulated in coastal areas or alongside riverbeds.

A particularly desirable form of monazite ore is found in heavy mineral sands, such as these on a beach in Brazil. Travel Pix/Alamy

Within those grains, the monazite is relatively hard and chemically inert. The monazite might be just a few percent by weight, but it typically contains between 45 and 70 percent rare earths by weight, including, in many cases, all of the necessary rare earths for magnets: neodymium, praseodymium, dysprosium, and terbium.

“The best distribution of rare-earth elements that you can find anywhere is in monazites,” says Constantine Karayannopoulos, former CEO of Neo Performance Materials and now a director of or adviser to several rare-earths firms. “The real money is in those four elements.” Dysprosium and terbium, both examples of heavy rare earths, are particularly sought after and are usually not present in the most commonly mined ore for rare earths, called bastnaesite.

There are vast deposits of monazite-bearing heavy mineral sands in Australia, Brazil, India, Madagascar, Mozambique, and South Africa, with smaller ones in China, Idaho, Indonesia, Malaysia, Sri Lanka, Thailand, the southeastern U.S., and Vietnam.

A geologist points out dark veins of hard-rock monazite ore in the Steenkampskraal mine in South Africa [top]. Ore samples from the mine are collected for analysis [above].Rodger Bosch/AFP/Getty Images

What’s more, collecting the ore simply involves dredging or strip-mining the sands, as opposed to the blasting, drilling, and heavy earth moving required to get at hard-rock ores. Then, too, the sands are already in granular form, so they don’t need to be crushed or ground, like rock ores do. Separating out the heavy minerals from the sands can usually be done at the mining site, using relatively cheap magnetic or gravity-based techniques, leaving a relatively high grade of concentrate for further processing.

And now the bad news about heavy mineral sands. “The problem is they tend to have a lot of thorium,” which is weakly radioactive, says James Kennedy, president of ThREE consulting, a rare-earths advisory firm based in Missouri. “The thorium is the single biggest problem” with using the sands, he says. “If you were mining and processing monazites for rare earths, you would end up with hundreds and eventually thousands of tons of thorium-containing radioactive residues” per processing facility. The only markets for thorium are a couple of tiny ones for welding rods and some optics, he says, so “you have to store it.” And under the rules for nuclear material set by the U.S. Nuclear Regulatory Commission and the International Atomic Energy Agency, any stored radioactive material that doesn’t have a planned use or disposal pathway is considered nuclear waste. “So the [radioactive] liability suddenly far exceeds the value of anything you’ve mined and sold,” Kennedy says.

It wasn’t always this way. Up until the 1950s, monazite was the main ore mined for rare earths. Companies processed heavy mineral sands from Australia, Brazil, and India, and hard rock from South Africa, to produce rare earths for tracer ammunition, steel alloys, cigarette-lighter flints, and, later, color-TV phosphors.

The Mountain Pass mine and processing plants, in California, are the only large-scale rare-earth production facilities in the United States.David McNew/Getty Images

Several developments combined to shift rare-earths production from monazite to bastnaesite. In 1949, a vast deposit of bastnaesite ore was discovered at Mountain Pass in California. By the early 1960s, bastnaesite had eclipsed monazite as the primary rare-earth ore. In the 1990s, with rare-earth production shifting to China, the move toward bastnaesite continued with the exploitation of massive deposits in that country, such as those at Maoniuping, that were either entirely or mostly bastnaesite.

Now, though, any resurgence of rare-earth production outside of China may hinge on a monazite renaissance, and some observers are taking a hard look at the strict regulations that control the handling of thorium and other by-products of monazite processing. Those regulations, Kennedy insists, will throttle the rare-earth ambitions of the world outside China, unless governments take action now.

“You rationalize the regulations around thorium and you will end the rare-earth crisis in the U.S.,” he declares. “You’ll turn rare earths from something that’s always scarce to something that is common and just naturally flows out of other mining.”

How to Scale Rare-Earth Processing Using “Waste”

A prime example of Kennedy’s vision already exists at Energy Fuels, based in Lakewood, Colo. It’s the only company in the U.S. licensed to process, handle, and store compounds of uranium, and among the very few licensed to store radioactive materials of any kind. Those permits give Energy Fuels the distinction of being the only U.S. entity allowed to process monazite ore for its rare earths.

“Supersacks” containing a tonne of neodymium-praseodymium oxides were loaded on pallets at Energy Fuels’ White Mesa Mill.Energy Fuels

The company is moving aggressively to capitalize on that distinction. Energy Fuels’ main business is still uranium—last year it produced nearly 500 tonnes of it. But it has already started selling rare-earth oxides, following a modest $22 million investment in a small rare-earth processing facility at White Mesa Mill in Utah, where it also processes uranium. In September 2025, the company announced that it had sold 1.2 tonnes of neodymium-praseodymium oxide to South Korea–based POSCO International Corp., a leading maker of permanent-magnet assemblies for EV motors. Around the same time, Energy Fuels disclosed that it had raised a further $700 million.

In December 2025, the company announced it had produced small quantities of dysprosium oxide. Tiny amounts of dysprosium are needed for motor magnets to ensure they keep functioning at high temperatures. Today, essentially all of the world’s dysprosium comes from China.

Energy Fuels says it can produce 1,000 tonnes of rare-earth oxide a year at White Mesa Mill. But its rare-earth production there currently shares facilities with the uranium operations, so they cannot both be produced at the same time. Its monazite comes from the chemical company Chemours, which mines heavy mineral sands in Florida and Georgia to produce titanium and zirconium. The monazite ore sold to Energy Fuels is what remains after Chemours strips out the minerals it needs, namely ilmenite and zircon.

The chemical company Chemours mines heavy mineral sands at sites in the U.S. states of Georgia [seen here] and Florida. Monazite extracted from the sands is sold to Lakewood, Colo.-based Energy Fuels, which processes the ore into rare-earth oxides.Chemours

“Effectively, monazites are waste,” Karayannopoulos points out. “So you don’t have to dig another mine” to produce them.

Chemours and Energy Fuels have agreements under which Chemours ships 500 tonnes of monazite annually to White Mesa Mill. Energy Fuels is also negotiating or already has deals that would give it access to enormous reserves of monazite ore in Australia, Brazil, and Madagascar. These will be essential for its ambitions to become a world-class supplier of rare-earth oxides.

Mark Chalmers, former CEO of Energy Fuels, led the uranium company’s entrance into the rare-earths business.Jesse Rieser/The New York Times/Redux

“We’re playing a long game,” says Mark Chalmers, who recently retired as Energy Fuels’ CEO and is now a consultant to the company. “We’re doing the engineering right now on a larger facility that will have the capacity to produce up to 6,000 tonnes per year” of rare-earth oxide, he said in an interview. For comparison, Lynas Rare Earths produced around 12,000 tonnes of rare-earth oxide in its most recent fiscal year.

In announcing the $725 million conditional loan commitment with Energy Fuels, the U.S. Defense Department’s Office of Strategic Capital said the deal would help Energy Fuels scale up its output of rare-earth oxides and also help realize the company’s plans to turn the oxides into metals, a key step preceding the fabrication of magnets.

What Are the Risks in the Rare-Earth Supply Chain?

To see what can go wrong with a monazite project, consider the project in Saskatoon, Canada, where the Saskatchewan Research Council (SRC) has built and demonstrated a pilot plant to produce rare-earth oxides from monazite ore. It’s the only rare-earth extraction and separation plant in the country. In an email, a spokesperson said the plant “is scheduled to be commissioned by year’s end and after a production ramp-up phase, operational within the calendar year 2027.” When in “full operation,” at a time not specified by the spokesperson, the plant will have a capacity somewhere between 400 and 600 tonnes of rare-earth metal per year.

But even that relatively modest output will depend on a reliable supply or “feedstock” of monazite. In an industry small enough that many of the top people are on a first-name basis, SRC’s inability to secure such a supply is an open secret. “Their problem is they haven’t got enough feedstock to run on a continuous basis, so they’re making sample runs,” a well-connected industry observer told IEEE Spectrum. (SRC declined to make an official available for this story.)

In April 2024, SRC announced it had reached agreements with a Vietnamese organization, the Hung Thinh Group, for 3,000 tonnes of rare-earth carbonate per year for five years. It’s unclear how much of that feedstock has made it to the SRC facility. Vietnamese government restrictions on the export of monazite and bastnaesite have been in effect since 2022 and became law this past January.

But the real twist here is that one of the largest, highest-grade monazite deposits in the world is just 800 kilometers from Saskatoon, near Alces Lake in Saskatchewan. It’s likely that the SRC facility was conceived and built with that monazite deposit in mind. After all, as a government-owned entity, SRC says its mission is “to help the people of Saskatchewan strengthen the economy with quality jobs and a secure environment.”

The Alces Lake deposit is owned by Appia Rare Earths & Uranium Corp. In an interview, Appia’s CEO, Tom Drivas, talked up the possibility of a link between SRC and Appia. “SRC would be our logical option because, you know, it’s in the same area, same province. They need the monazite. And if we can supply the monazite, that’s sort of a natural, right?”

It probably won’t happen, though. The logic of rare-earth mining is utterly different from the logic of mining a commodity like, say, copper. Multiple experts contacted by Spectrum suggested that the cost of developing a monazite mine at Alces Lake, including such necessary infrastructure as roadways, electricity, and water, would be in the hundreds of millions of dollars. Such costs, and the mercurial exigencies of the rare-earths market, make it unlikely that the government of Saskatchewan would commit that kind of money, even with help from the Canadian federal government. These factors help explain why Canada, which has some of the most attractive deposits of rare-earth ores on the planet, along with impressive mining companies and resources, has no operating mines for rare earths.

Meanwhile, SRC is pursuing other options. Under a deal announced in May with REalloys, in Euclid, Ohio, REalloys will spend $20.6 million for upgrades to SRC’s Saskatoon facility, while SRC develops technology to produce rare-earth metals from rare-earth oxides. The deal also gives REalloys the right to purchase up to 80 percent of the SRC plant’s output. Where will the feedstock come from? Hard to say, but REalloys also struck a deal with U.S. Critical Materials Corp. for 10 percent of the hypothetical output of a proposed mine at Sheep Creek, in Montana. Sheep Creek reportedly has substantial amounts of hard-rock monazite that’s unusually low in thorium. But it’s far from certain that those deposits will ever be mined. The proposal for developing a mine there, in a breathtakingly scenic area near the headwaters of the Bitterroot River, has attracted fierce opposition from environmental groups, outdoor enthusiasts, and local officials.

Taken together, SRC’s recent moves have left industry veterans nonplussed. “I’m not sure what their business model is,” said one. “They say, ‘we’re a research institute. Our mandate is to help industry.’ The reality is, they’re competing with industry.”

SRC has also drawn scrutiny for some questionable business practices. In 2022, SRC board member Dennis Fitzpatrick was fired after raising concerns about the qualifications of an engineer hired to manage and direct operations at SRC’s Saskatoon plant. The president and CEO of SRC at the time, Mike Crabtree, was responsible for the hire. A subsequent investigation of Fitzpatrick’s claims by the Canadian Broadcasting Corp. detailed a troubling web of relationships and practices surrounding the SRC’s rare-earths initiative. (Fitzpatrick did not respond to multiple emails.) For example, past business associates of Crabtree’s were reportedly appointed to positions of authority without any other candidates being considered and without their connections to Crabtree being made public.

In one case, in 2021, Crabtree disclosed that he was offering a venture capital firm, Westbridge Capital, an opportunity to invest in the rare-earths plant. But according to Fitzpatrick, Crabtree never mentioned that the VC firm was run by an SRC board member who also happened to be one of Crabtree’s business associates—or that Crabtree himself was a trustee of Westbridge and an investor in the company, as well as a paid advisor.

Crabtree remains as SRC’s CEO.

Managing Toxic By-products From Rare Earths

Putting aside SRC’s missteps, the organization does have one advantage: It holds multiple licenses from the Canadian Nuclear Safety Commission to handle, process, and manage radioactive materials. And that’s huge.

Most of the radioactive burden associated with processing monazite ore comes from substances produced during processing. In one of the early steps, pulverized, powdery ore is dried and then leached in strong acids—a process called cracking—to produce a rare-earth concentrate. Waste products of this step include a slurry containing thorium hydroxide and much smaller amounts of uranium hydroxide. The rare-earth concentrate then undergoes an extraction process to isolate specific rare earths, such as neodymium or dysprosium. More thorium and also radium are produced during these steps.

Waste barrels were photographed in 2024 at Energy Fuels’ White Mesa Mill, where ores containing uranium and rare-earths are processed and refined.Jesse Rieser/The New York Times/Redux

The thorium hydroxide slurry, meanwhile, is typically deposited in heavily regulated waste cells, called tailings impoundments or tailings ponds. The thorium hydroxide, either in the form of a slurry or a powder, must be stored in isolation from the environment indefinitely. Energy Fuels stores waste from its longstanding uranium production operations, as well as its new rare-earth operations, in tailings impoundments at its White Mesa Mill facility. The impoundments are just 10 kilometers from the community center of the Ute Mountain Ute Tribe, a group of Indigenous people, which has vigorously opposed operations at White Mesa Mill for more than 20 years.

Virtually all of the thorium is the isotope thorium-232, which is only weakly radioactive. Its radiation is alpha particles, which cannot penetrate human skin. Nevertheless, improper storage can have serious consequences, particularly if the thorium is inhaled or ingested.

“Have a look at the gravitational separation of HMS [heavy mineral sands] in Nigeria,” Thomas Kruemmer, a rare-earth consultant and analyst, wrote in an email. “There is lots of dust in the Nigerian processing sites, which the workers inevitably breathe in. And inside your respiratory system there isn’t any protective skin, so alpha radiation develops its full potential. This is something many people do not consider.”

In 2012, protestors in Malaysia [top] rallied against the Lynas rare earths processing facility [above], then under construction near Kuantan. Left: AFP/Getty Images; Right: Goh Seng Chong/Bloomberg/Getty Images

Moreover, adds Kruemmer, “it is not thorium itself. It is the decay isotopes of thorium that are highly radioactive.” One decay product of thorium is polonium, he points out, “the preferred poison for [killing] Russian dissidents.”

In 1982, a company called Asia Rare Earth, a subsidiary of Mitsubishi Chemical, began extracting yttrium from monazite ore in Perak, Malaysia. Local residents soon began complaining of various illnesses. With the consent of the Malaysian government, the company had built a waste channel to a nearby disposal site, where the waste was stored in open steel barrels and in an unlined waste pond. The waste overflowed from the pond whenever it rained heavily. In 1984, a Japanese professor measured radiation levels at and near the pond and found them to be, in the most extreme cases, hundreds of times higher than internationally accepted maximums. A subsequent investigation found extremely high levels of birth defects and eight cases of leukemia in the small community near the waste dump.

In 2011, at age 69, Lai Kwan prepared to bathe her 29-year-old son, Cheah Kok Leong [top]. Cheah Kok was born with severe mental disabilities that Lai Kwan believed were caused by exposure to radiation that she received while working at a rare earths processing plant in Bukit Merah, Malaysia [bottom, photographed during its disassembly].Rahman Roslan/The New York Times/Redux

A protracted legal fight among local residents and Asia Rare Earth and the Malaysian government culminated in the closure of the plant in 1992. By that time, so much radioactive material had been carelessly dumped that cleanup took more than two decades and cost Mitsubishi an estimated $100 million. The episode still haunts the rare-earths industry in Malaysia, where Lynas now processes monazite concentrate at the largest rare earths processing facility in the world. That plant, in Kuantan in the state of Pahang, has been the subject of organized and active local opposition that began even before its commissioning in 2012. The Malaysian government has sporadically tightened its regulations on the plant, stipulating that, for example, radioactive wastes be shipped out of Malaysia or that the processing stages that create radioactive residues be accomplished outside of Malaysia.

Kennedy notes that if monazite ore is to become a major source of rare earths, much larger quantities of thorium will need to be disposed, probably more than could be economically accommodated in lined and isolated tailings ponds.

One of the very few options for using such large amounts of thorium involves nuclear power. Years ago, countries including the U.S., Canada, Norway, and Germany had thorium-reactor research programs, and India has had a slow-moving program for decades. But today, the only operating thorium power reactor is in China. The Shanghai Institute of Applied Physics of the Chinese Academy of Sciences has been operating, since 2023, a 2-megawatt molten-salt-cooled reactor that is thorium-fueled.

The world’s only operating thorium reactor is a 2-megawatt unit at the Shanghai Institute of Applied Physics of the Chinese Academy of Sciences. In October 2025, technicians used mechanical arms to analyze samples of the reactor’s molten-salt fuel.Jin Liwang/Xinhua/Getty Images

A couple of Western startups are keeping the thorium reactor dream alive. A Franco-Dutch company called Thorizon says it has designed a reactor that would be fueled by a mixture of thorium and spent nuclear fuel. Startup Copenhagen Atomics is working on a 1-MW reactor that’s a prototype for a 100-MW, mass-manufacturable, molten-salt reactor.

Processing of monazite ore could provide fuel for an enormous number of thorium reactors far into the future. Consider a single large facility producing 5,000 tonnes of rare-earth oxide a year. If the monazite contains 7 percent thorium, the plant would also be producing roughly 700 tonnes of thorium a year. According to Rusty Towell, a professor at Abilene Christian University in Texas who directs an advanced-reactor project there, a tonne of thorium would be sufficient to generate about 350 MW (electric) for a year. So the large processing facility alone would be throwing off enough thorium annually to keep 2,450 100-MW reactors operating for a year.

That amount of power—245 gigawatts—is about half of all the generating capacity growth forecast for the United States over the next decade, a period during which analysts expect accelerating electricity growth due to increasing numbers of data centers and electric vehicles.

What Could Strangle Rare-Earth Supply Chains?

Virtually all rare-earths experts foresee a growing role for monazite. But given the problems with thorium, local opposition to some projects, and political instabilities connected to others, some have muted expectations. “The sole and single monazite hope rests with Iluka’s expensive project in Australia, IMHO,” writes Kruemmer. Iluka Resources, based in Perth, is now building a massive rare-earths refining facility at Eneabba, 275 km north of Perth. It will process a million-tonne monazite stockpile that accumulated over decades from the company’s other mining operations. The cost of the project is estimated at AU $1.8 billion (about US $1.3 billion).

In a US $1.3 billion project, Iluka Resources is building a massive facility in Eneabba, Australia, to extract and separate rare earths from a million-tonne stockpile of monazite, visible in the background of this photo.Iluka Resources

For other experts, though, the biggest challenges facing monazite projects isn’t thorium, or radiation, or regulations. It’s the blend of quirks, misconceptions, and self-dealing that dog the entire rare-earths industry.

“The rare-earth mining business is the Afghanistan of mining,” quips Kennedy. It’s a place where people and companies go with grand ideas, only to become ensnared in quagmires.

One fundamental problem is the highly speculative nature of almost every aspect of the rare-earths business, which lends itself to backroom deals. In particular, the industry is plagued by speculators who buy stakes in a project with no intention of staying with the project through completion; they cash out as soon as they can profit from any future favorable news—a technique known as “pump and dump.”

Consider the Trump administration’s decision to take a 10 percent stake in USA Rare Earth in exchange for $277 million and to provide a $1.3 billion secured loan to the company. USA Rare Earth has the rights to a deposit called Round Top near Sierra Blanca, Texas. The company hired the Wall Street firm Cantor Fitzgerald to raise $1 billion in equity financing for the deal. Cantor Fitzgerald, as it happens, is run by the sons of Howard Lutnick, Donald Trump’s Commerce Secretary.

Meanwhile, there is no rare-earths mine at Round Top, and it’s a distinct possibility that there never will be.

“The current administration consistently fails to seek expert advice and is recklessly pouring billions of taxpayer dollars into speculative [rare-earth element] processes and projects, like Round Top, that have little chance of commercial success,” says Terry McNulty, a mining consultant based in Tucson.

Geoff Atkins, vice president of business development of the Canadian rare-earths firm Ucore Rare Metals, had advice for his colleagues at a convention of the Critical Minerals Institute in 2022. “Instead of focusing on your speculative value,” he said, “instead of being able to say we’ve got the biggest project in the world, instead of, ‘we’ve got a [net present value] of $3 billion or $4 billion,’ how about you say, ‘we’ve got a project which is going to get into operation, and this operation is going to provide the greatest value to our shareholders in the long term, not just in the short term when we can pump this up and, you know, the [early] investors can sell out.

“Because unless there is that shift,” he warned, “we’re not going to have any hope of getting the volumes we need to give the certainty to the downstream supply chain.”

In short, the rare-earths industry outside of China needs to grow up. And sooner rather than later.

This article appears in the November 2026 print issue as “Can a Radioactive Ore Break China’s Grip on Rare Earths?”

The latest front in the ongoing global transformer shortage may be a disagreement over steel.

To be precise, that steel is a specialty called grain-oriented electrical steel (GOES). It’s a key ingredient in the majority of the world’s transformers—and in Europe, it’s now a trade flashpoint. Last month, the European Commission, the European Union’s main executive branch, significantly raised the cost of importing GOES from outside the EU.

While EU steelmakers believe the measures are necessary to protect their industry at a turbulent time, many in the energy sector have been far more critical. They believe restricting GOES imports will only make it harder for Europeans to find the steel their transformers need—and risk slowing down Europe’s electrification.

“European producers can currently neither supply the volumes nor the highest-performance grades necessary to meet the demands of the transformer industry,” says Mark Becker-von Bredow, Director of Energy, Electrification and Climate at the German Electrical and Digital Industry Association (ZVEI).

The Basics of Grain-Oriented Electrical Steel

GOES is not a garden-variety steel, which is an alloy of iron combined with carbon to provide strength. An electrical steel like GOES instead blends iron with silicon, which improves the metal’s electrical properties.

There are several electrical steel varieties, but the “grain-oriented” in GOES separates it from the others. Examine any steel under an electron microscope, and you’ll see a mosaic of tiny crystal grains. Unlike in most steels, the grains in GOES are aligned in one direction.

This trait reduces energy loss in situations where a magnetic field flips back and forth along that direction—which is precisely what alternating current does many times per second in a transformer’s core. GOES cores can minimize power loss, emitted heat, and audible noise, all making for a more efficient transformer. That’s why GOES is the dominant material in transformer cores today.

GOES, however, also differs from conventional steel in its economics. “Because it’s a specialty good, it’s more comparable to other specialty goods like, say, computer chips,” says Ryan Alimento, an analyst at the U.S.-based Breakthrough Institute.

Compared to the number of steelmakers around the world, there are far fewer specialized GOES producers. As with chipmakers, their products can vary. A higher-quality GOES loses less energy and helps build a better transformer. Transformer manufacturers may pay close attention to where their GOES comes from.

“The global leaders—the companies that make the world’s best GOES—by and large are in [South] Korea and Japan,” Alimento says. “In the U.S. and in the EU, domestic manufacturers are not capable of making that quality of GOES, because it is a heavily protected trade secret.”

In fact, just two EU steelmakers today (Thyssenkrupp and Stalprodukt) manufacture GOES at all. This is partly because Europe’s electricity demand stayed relatively flat for decades, and demand for grid components like transformers has likewise been modest.

Europe’s Growing Electrification Needs

Today, the story has changed. Renewables, electrification, data centers, and the EU’s own ambitious grid expansion plans mean that transformers in Europe are on the rise. According to the European Commission’s numbers, from 2021 to 2025, the EU’s GOES consumption increased by 32 percent.

Yet EU steelmakers have seen few benefits. Their share of the EU GOES market dropped from 66 percent to 43 percent over that period, even as the volume of GOES imports more than doubled. More than half of those imports now come from China, fueling fears that overcapacity at Chinese steel mills is undermining their EU counterparts. Since December 2025, Thyssenkrupp has slashed GOES production in response to “the massively increased inflow of low-priced imports, especially from Asia.”

It was in this environment that EU steelmakers pushed the European Commission to act. The EU’s earlier steel tariffs didn’t touch GOES, and while the EU had set minimum import prices on electrical steel from certain countries since 2015, steelmakers felt more action was needed.

On 18 September, the commission unveiled a preliminary package of “safeguard measures.” It raised the minimum prices for both imported GOES metal and imported transformer cores. These measures are temporary, scheduled to expire in February 2027, by which time the European Commission must decide if and how to make them permanent.

The GOES issue is not limited to Europe. The U.S. has just one GOES mill whose product is considerably less energy-efficient than its East Asian counterparts. The U.S. raised tariffs on GOES in 2018, but unlike the EU, its tariffs didn’t cover finished GOES products until August 2025. The result was a surge of manufacturers using Asian GOES to make transformer cores in Canada or Mexico, then exporting them to the U.S.

EU GOES Producers and Output Limits

European steelmakers who blame low-cost imports have welcomed the measures, stating that they’re necessary to protect their industry and its jobs.

However, many in the EU’s energy sector feel that focusing on cost misses a larger problem: European manufacturers simply cannot provide enough of the metal that meets transformer efficiency standards.

“These grades are currently not available from European producers in sufficient quantity and quality,” says ZVEI’s Becker-von Bredow. “Until European mills can provide these grades at the required scale, imports from trusted partners such as Japan and South Korea will remain indispensable.”

For this reason, ZVEI and others in the energy sector oppose the safeguards. They believe the measures will only complicate obtaining the needed GOES or encourage producers to manufacture whole transformers outside Europe and import them.

“Based on our initial assessment, the introduction of safeguard measures will result in an increase in the production costs of transformers in Europe,” says Diederik Peereboom, Secretary General of T&D Europe, a group of European grid technology companies.

If the EU wants to strengthen its GOES makers, Peereboom says, then it will need comprehensive policies for upgrading EU GOES production. “Nothing in the [safeguard measures] guarantees—or even asks for—EU GOES producers to expand output, add production lines, or develop the high-permeability and specialty grades that EU mills do not currently manufacture,” he says.

This article was originally published by Canary Media.

The geothermal startup Fervo Energy is now selling power to the grid from its Cape Station project in Utah—marking the first time a utility-scale, next-generation geothermal facility has achieved such a feat worldwide.

The Houston-based firm turned on an initial 33-megawatt unit late last week and has been testing and ramping up output since. On Thursday, Fervo will officially begin commercial operations, supplying carbon-free electricity to Shell Energy, a retail power provider in California.

Cape Station, a $2 billion project that will eventually be 500 MW, is unique among the world’s commercial geothermal plants. While other facilities draw on naturally occurring hot water and steam, Fervo has engineered its own geothermal resource deep underground. The startup uses drilling techniques from the oil and gas industry to create reservoirs in hot, dry rocks that generate steam to produce electricity.

“Traditional geothermal would have never been done in southern Utah,” says Dawn Owens, Fervo’s senior vice president and head of development and commercial markets.

With the first unit online, Fervo says it is working to commission two more 33-MW power blocks by 1 January, 2027. The project’s remaining 400 MW—which is supported by a new agreement with Google—is under construction and slated to come online in 2028.

Fervo’s Next-Generation Geothermal Journey

Fervo already operates a 3.5-MW commercial pilot plant in Nevada, which went online in late 2023. Owens says the pilot, called Project Red, was important for demonstrating the technology itself. But firing up Cape Station proves that Fervo’s approach can achieve utility scale and export power to wider markets.

“This is massive for both Fervo and the geothermal industry at large,” she says.

Fervo, which went public in May, is leading the fast-growing movement to develop clean, always-on geothermal power to meet soaring demand from data centers and balance growing supplies of intermittent solar and wind energy. Next-generation geothermal firms have raised billions of dollars in private capital in recent years to develop and deploy technologies that can harness energy in harsh, deep environments.

Drilling continues at Fervo Energy’s Cape Station geothermal plant in Beaver County, Utah, where the company plans to ramp up production to 500 MW. Fervo Energy

Today, geothermal represents just 0.4 percent of total U.S. electricity generation every year. Expanding geothermal’s share of the mix will require not just building more projects like Cape Station but also tackling high upfront drilling costs and other economic and regulatory barriers—all while working to limit potential impacts on communities and the environment, experts say.

“Fervo’s first power achievement shows investors the size and scale of the opportunity to fuel many more such milestones across developers and sites in the future,” says Mitchell Smith, who serves as grid lead for the Mountain West Geothermal Consortium. The bipartisan initiative is calling for increased public and private financing “to help bring more projects online responsibly and at the speed of demand,” he added.

Enhanced Geothermal Energy Gets Cheaper

Fervo has declined to disclose the terms of its power purchase agreements for in-progress and future development, which together represent about 1,000 MW, or enough to power roughly 1 million U.S. homes every year. Phase 1 of Cape Station is set to deliver power at $7,000 per kilowatt, while Phase 2 should come at $5,500 per kilowatt, the company previously told Canary Media. (By comparison, the average cost to build a natural gas–fired power plant was $2,157 per kilowatt in 2025.)

Along with slashing drilling costs, enhanced geothermal projects could improve their economics by delivering power when it’s most valuable to utilities, according to researchers. While geothermal plants traditionally provide around-the-clock, baseload power, next-generation facilities might operate flexibly during certain periods—like in the evening, when solar output declines—to reduce overall grid costs. Geothermal plants could also operate “behind the meter” to directly power data centers so that the costs of adding new electricity capacity aren’t passed on to utility customers, Owens says.

“We’ve been saying all along that our technology makes geothermal more scalable and affordable in more places,” she says. On Cape Station’s powering-up, she adds, “That’s just something no next-generation geothermal company has ever been able to do before.”

As billions of dollars are poured into building data centers, many projects face a constraint that goes beyond getting the facilities constructed: securing enough electricity to power them. San Francisco–based Rune Energy is flipping the script: Instead of building data centers and waiting for grid power, the startup is putting data centers where there’s excess power the grid can’t absorb.

The company’s modular data centers, called Renewable Energy Linked Intelligent Compute (RELIC), connect directly to the direct current (DC) electricity generated by a solar array. Each RELIC module is a self-contained ecosystem with its own network equipment, cooling system, power electronics, and GPUs, all within a container about 8 feet long, 2 feet wide, and 5 feet high. A module can draw up to 100 kilowatts of power and weighs about 2000 pounds.

“It’s our estimate that there’s 50,000 gigawatt hours of energy wasted by utility-scale solar every year,” says William Layden, Rune’s cofounder and CEO. A solar farm can produce more electricity than the grid can absorb at particular times, especially when generation is high and demand is relatively low. Curtailing that generation means reducing output even though the panels would otherwise be producing electricity.

“We’re already in an energy abundant future,” says Varun Palivela, Rune’s cofounder and CTO. “But the energy abundance is stranded in time and space.” The power constraints facing large data centers, he adds, are “not a real bottleneck, but a resource allocation issue.”

The direct current advantage

In addition to curtailment, electricity is wasted due to equipment outages, conversion losses, and an issue called inverter clipping. Inverters typically convert DC electricity produced by solar panels to alternating current (AC) for delivery to the grid. When a solar array produces more DC power than its inverter can handle, the power gets clipped by the inverter.

Since Rune taps into the DC power before this conversion, they get ahead of some of the potential losses. Constance Crozier, an assistant professor at Georgia Tech’s School of Industrial Systems Engineering who has studied data center energy demand, says Rune’s main advantage won’t come from cutting out this conversion, though. “DC to AC conversion is typically quite efficient—95 to 97 percent—so I don’t expect the efficiency savings to be exceptional,” she says. “I would imagine the main advantage is reduced component cost.”

The company signs power-purchase agreements with solar-farm owners, buying electricity directly from them (and, in the process, giving the solar facility income for power that would otherwise go unused). Utility-scale solar has a standardized voltage around 1500, Palivela says. While Rune’s technology doesn’t require an inverter to convert DC power to AC, it does use DC to DC converters to step down the voltage to a lower range. “You need to get the power in a conditioned state to allow the GPUs to consume it,” Layden says.

A modular approach

Layden emphasizes that Rune is a product company, not a construction company. RELIC modules are manufactured at Rune’s facility in Mountain View, Calif., and are designed for ultrafast and simple deployment once they reach a solar farm. “You literally take two wires, plug it into the DC link, and you’re done,” says Layden, adding that Rune recently fulfilled a request to bring a data center online within a week. “We were able to do that,” Layden says. “We are the fastest time to compute.”

The modular approach also sidesteps some of the supply-chain bottlenecks facing conventional data centers. A traditional facility can require large transformers, switchgear, cooling systems, and other specialized electrical equipment, some of which can take months or even years to procure.

One Relic module can contain anywhere from 128 Nvidia B300 GPUs to over 2,000, and multiple modules can then be interconnected to form a large network of GPUs. “The only difference [between Rune and a conventional data center] is that it’s not in a centralized building connected to transformers and switchgear,” Palivela says. “We sell this to inference customers and those who are doing small training runs or fine-tuning training runs on state-of-the-art models.”

Rune’s approach is part of a broader push by data-center operators to go behind the meter, connecting directly to power generation to avoid relying on the grid. The strategy has gained attention as data centers struggle to secure new grid connections, with companies exploring direct connections to nuclear plants, natural-gas generators, and other sources of power.

When the sun doesn’t shine

The big question regarding Rune’s technology is the intermittency of solar power; what happens when it’s cloudy for several days in a row? “We have batteries integrated at the Relic level to make sure we always have compute 24/7,” says Layden. Sizing the batteries correctly is key, Crozier says, particularly because Rune modules don’t have a backup grid connection. “They need to size the batteries to store enough energy for nighttime and cloudy day operations, and generate enough solar to fill these batteries on ‘good solar’ days,” she says. This also means that some of the available power at any given time is directed toward battery charging rather than directly powering compute.

The company does extensive modeling to figure out the optimal data center and battery sizing, using historical data to understand how much power is likely to be available under different conditions at each site where it operates. Current sites are in California, Texas, and Massachusetts.

“Far more significant flexibility can be obtained by co-locating batteries with a data center, especially if they’re paired with a generation source,” Crozier says. “This way the data center can remove part of the load from the grid without changing its operation.”

Rune is planning to scale its operations to 100 megawatts of compute over the next 12 months.

If there is any place that would be off-limits to AI, the control room of a nuclear power plant certainly sounds like one. The nuclear industry is historically cautious and risk averse—understandable given the possible catastrophic consequences of an accident or a mistake. Even well-trained managers struggle with the operational complexity of a nuclear reactor. It’s not the kind of setting that seems well suited to a powerful but error-prone new technology.

Reality tells a startlingly different story. A variety of companies have begun to offer AI solutions for nuclear power. This year, nearly the entire fleet of 94 U.S. nuclear reactors has been offered the chance to integrate AI into its operations, and most have taken it.

In August, California-based Atomic Canyon launched NIVA, the Nuclear Industry Virtual Assistant, which was developed in cooperation with nuclear-industry groups. The assistant was pilot-tested in nuclear plants run by Constellation Energy and is now available to the United States’ entire fleet of reactors. (Constellation declined to be interviewed for this story.) Meanwhile, the AI-for-nuclear startup Nuclearn says its products have gone to work with more than 65 U.S. partners, a total that includes many traditional plants but also an integration with NuScale Power Corp. that is working toward advanced small nuclear reactors. Microsoft and Nvidia have collaborated on a project to apply AI toward optimizing “the entire life cycle of a nuclear plant, spanning site permitting, design, construction, and continuous operations.”

The reason for the rapid adoption, according to Nuclearn CFO and cofounder Jerrold Vincent, is twofold. First, the technology companies banking their futures on AI are also keen on a revival of nuclear power to provide the electricity for all those data centers. Vincent says AI is also ideal for addressing some of the nuclear industry’s most pressing problems. Many of the things that make reactors so difficult to operate—intricate regulations, along with complex systems for management and maintenance—are exactly the kinds of tasks that AI is good at. The industry is staggering to deal with aging hardware and an aging workforce. If it is going to survive, and have a real shot at a nuclear revival, it will need help from machine intelligence, Vincent says.

“There’s this huge growth in demand from advanced-reactor developers and in new nuclear,” Vincent adds. “But the existing workforce has been spread really thin. So most of the time when we’re coming into a customer, [they're saying,] ‘I’ve had jobs open and I can’t fill these positions. We need to do the things required by regulation. How the heck are we going to do it?’”

AI Streamlines Nuclear Documentation

Right now, the principal output of a nuclear reactor (aside from electricity) is paperwork. Nuclear plants produce piles of documentation to demonstrate regulatory compliance and to record how they respond to any potential problem, from a crack in the sidewalk to a fault in the reactor.

“Every engineering decision has to be evaluated, independently verified, and checked,” Vincent says. Nuclearn’s AI products help plants search through large repositories of data more efficiently to more quickly file paperwork with the federal government or figure out how a particular problem was solved in the past. “If you think of it as ChatGPT for nuclear,” he says, “it’s actually not a bad place to start.”

Rob Austin, the Electric Power Research Institute’s leader for the NIVA project, says the project started a year ago when Constellation chairman Joseph Dominguez issued a challenge to the nonprofit industry groups that act as storehouses of information about regulatory compliance, maintenance and repair, and mechanical history. He asked the groups to find a way to make that industry knowledge available with the convenience of a large language model like ChatGPT, Copilot, or Gemini but keep it secure at the same time.

The task fell to Atomic Canyon, whose CEO Trey Lauderdale previously worked to integrate AI into health care, another conservative industry. Living and working downstream from Diablo Canyon, California’s last operating nuclear plant, he began to see how nuclear energy and AI could form a mutually beneficial relationship: A revival of nuclear power could help feed electricity to energy-hungry data centers, but only if AI could help the industry thrive again.

“The goal was to make decades of operating experience, hundreds of thousands literally of operating experience reports, tens of thousands of technical documents, and a whole lot of information available for people to find.” —Rob Austin, Electric Power Research Institute

A major hurdle is the public’s opinion of nuclear power, which soured decades ago after notorious accidents and worries about waste and only recently has begun to recover. But the resulting halt in construction of new nuclear plants has created another hurdle of its own: a loss of institutional knowledge, Lauderdale says. As the industry stalled in the U.S., fewer people entered careers as nuclear scientists or technicians. The result is a people problem. Nuclear power has a remarkably skilled but quickly shrinking workforce tasked with keeping old reactors online and possibly starting up new ones to meet surging demand for electricity.

“There is absolutely no way we as an industry will be able to have this nuclear resurgence or nuclear renaissance without augmenting the capacity of this knowledge-based workforce,” Lauderdale says.

For him, the answer is an AI that can intelligently search the repository of data that the nuclear workforce has created over the decades. When Atomic Canyon won the NIVA contract in 2025, the first step was to download 53 million pages of publicly available data from the Nuclear Regulatory Commission to train the model on the operations, vernacular, and regulations particular to nuclear power. Lauderdale then struck a partnership with Oak Ridge National Laboratory to build the first nuclear-specific AI models.

“Technically, they’re sentence-embedding models,” Lauderdale says. “Fancy way of saying, we taught AI how to seek the nuclear language. Nuclear is almost like its own vernacular.” The goal is that today’s nuclear workers can ask questions about every aspect of plant operations, from filling out regulatory forms to procuring the right tools and materials, simply by writing a query and letting the AI sort through all of the accumulated institutional knowledge.

Unifying Nuclear Data for AI Searches

When Lauderdale began to work with Diablo Canyon on integrating AI into the nuclear industry, it became apparent that one bottleneck was simply getting decades’ worth of a nuclear plant’s documentation—procedures, best practices, safety recommendations—into an easily searchable format. Information across the industry was scattered and not standardized, which is why three of the biggest nonprofit entities in the industry (the Institute of Nuclear Power Operations, the Electrical Power Research Institute, and the Nuclear Energy Institute) moved in 2025 to create unified datasets, an idea that would birth the NIVA AI assistant developed at Atomic Canyon.

For example, EPRI’s Austin says, nuclear engineers might run into a problem with one of the house-size, thousand-horsepower vertical pumps that circulate vast quantities of water used for cooling purposes. In their search for a solution, the workers have access to plenty of information—too much information, perhaps. They can read through EPRI’s maintenance guides about those components. They can also call upon tens of thousands of incident reports stored with the Institute of Nuclear Power Operations about things that have gone wrong with pumps in the past. AI’s task is to synthesize all those reports to create a guide to the main failures that strike a pump, and, crucially, to show all its work so engineers can find the documentation it used to make its recommendations.

“I want to know all the other nuclear power plants that have that same part,” Lauderdale says. “Have any of them seen a similar issue? How did they fix it? What’s the troubleshooting look like? If you think about nuclear, we have a tremendous amount of information. The challenge is, how do you collectively gather knowledge from that information dataset? And that’s what AI is really, really good at doing.”

“If you think of it as ChatGPT for nuclear, it’s actually not a bad place to start.” —Jerrold Vincent, Nuclearn

Secrecy is a major hurdle to modernizing nuclear operations. AI systems that chiefly help with documentation and data search aren’t subject to the strictest rules, as they would be if they interacted with components that control the power plant. But nuclear data itself is largely proprietary or protected by federal law.

Nuclearn offers only single-location solutions to deal with this, so that a power plant’s data can never leave that site. NIVA can only search databases that already required plant-level security clearances, Austin says, so the same rules hold true for using the AI assistant to hunt through that deluge of data. Technology cannot wipe away these human barriers, but it does provide new incentives to find ways to share data more openly.

“The goal was to make decades of operating experience, hundreds of thousands literally of operating experience reports, tens of thousands of technical documents, and a whole lot of information available for people to find, and do so in a way that’s going to be much more accessible and easy to use, but at the same time preserving data security and that critical information, keeping control over it,” Austin says. That’s exactly what NIVA does.

Nuclearn’s “Equipment AI” system lets engineers pose queries about nuclear-equipment operations and generates answers that draw on decades of industry experience.Nuclearn

AI-Assisted Nuclear-Plant Inspections

So far, both the Nuclearn and NIVA large language models used in the nuclear world report to humans and don’t go anywhere near power plant operations. Still, the inevitable question arises: What if they could?

Vincent and Austin both state that such a feat is far away, but they can envision the pathway there. Consider the litany of welds throughout a nuclear plant, Austin says. All of those welds currently must be inspected for quality and endurance by a human. One pilot project at EPRI is testing the ability to have AI look at the mountains of data generated by such inspections to see whether it can flag areas of concern. A human remains in the loop with that project, Austin says, to review all of the AI’s recommendations, and no power plant is allowing AI to take over its inspections. But the project marks a step toward using AI on direct plant operations rather than only guidance for the human operators.

“I cannot see, in the short or the long term, an environment where AI is making decisions around the operation of the plant, because of the regulatory nature and also because of the risk nature,” Lauderdale says. “You don’t want machines making decisions that could impact the operations. So, I believe we’re going to have humans in the loop for a very, very long time.”

This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore.

Garlic can pack a punch, awakening our tastebuds with a real zing. Who knew that it can also be used to generate shockingly high voltages? At least, the peels of garlic do.

In a recent study, researchers in India used garlic peel and plastic to create a novel self-powered sensor and demonstrated how it could be used as part of a real-time home security-monitoring system. A single layer of garlic peel folded within polyethylene proved to be a durable power source even after six months in storage.

The results, published 20 August in IEEE Sensors Letters, demonstrate how biowaste can be repurposed for simple, low-cost technological solutions.

The sensor is a type of triboelectric nanogenerator, in which two different materials—one positively charged and the other negatively charged—are layered together. When movement or pressure causes the surfaces to repeatedly contact and separate, this produces an alternating current that can be harvested or used directly as a sensing signal.

“It’s a simple, low-cost way to convert everyday mechanical motion—footsteps, vibrations, a door opening—into usable electrical output,” explains Deepak Bharti, an assistant professor in the department of electronics and communication engineering at the Malaviya National Institute of Technology, in Jaipur, India, who conducted the study with Ph.D. candidate Manas Tiwari.

Why use garlic peel as a voltage source?

In past studies, Bharti and Tiwari have explored the use of other biowaste products for powering tech solutions, such as peanut skin and the dry, pulpy residue of sugarcane. In their most recent study, they focused on garlic peel for a number of reasons.

“Garlic peel is an agricultural and household biowaste that’s produced in large volumes but almost never repurposed—it typically just gets thrown away,” Bharti explains. “We were interested in whether a material like this, which is lightweight, biodegradable, and widely available, could perform competitively as a triboelectric layer, instead of relying on synthetic polymers.”

He adds that, structurally, garlic peel has a naturally textured, layered surface that the researchers found favorable for contact electrification.

In their study, they ground up garlic peel, which is positively charged, and layered it over a negatively charged sheet of polyethylene.

A single layer of garlic peel and plastic together produced an open-circuit voltage of about 210 volts and a very low short-circuit current of about 45 microamperes, with a power density of 2.16 watts per square meter. When the researchers added an additional layer of garlic on the other side of the polyethylene sheet, this roughly doubled the output, to about 400 V and 65 µA. The combination of high voltage and low current is characteristic of this class of nanogenerators, which makes them suitable for low-power sensing devices and energy harvesting.

Bharti says he is not only surprised at how the second layer of garlic peel nearly doubled the output, but also by how durable the garlic peel proved to be as a power source—maintaining its performance after six months of storage and 200 mechanical stress cycles.

“Seeing the device perform essentially the same after six months of ambient storage was a strong validation that a ‘waste’ material like this can hold up as a practical sensor material, not just a lab curiosity,” he says.

The researchers tested their sensor as a home-security system, detecting when a door is opened. Manas Tiwari and Deepak Bharti

Demonstrating a home-security system

The researchers then incorporated their novel garlic peel sensor into a home-security system, whereby the sensor is sandwiched between the top of a door and the door frame. When the door is opened, the garlic peel and plastic layers separate, triggering a voltage spike that is relayed to a microcontroller and a Bluetooth module. (While the garlic peel sensor requires no external power source, the microcontroller in this experiment uses a power bank.)

The Bluetooth module wirelessly transmits the signal to a smartphone running on a custom Android app that the researchers built. The app then generates a real-time, voice-based alert notifying the resident that the door has been opened.

Over the course of 50 tests, the sensor system was able to identify when the door was being opened with 92 percent accuracy.

Bharti says in future work he is interested in studying a curious phenomenon that occurred in these experiments. While stacking two layers of garlic and polyethylene doubled the electrical output, stacking more than two layers did not yield significantly greater outputs. Bharti plans to study this phenomenon in more detail, along with other biowaste materials that could prove useful as triboelectric materials.

Dielectric metasurfaces have moved to the forefront of nanophotonics, offering flat, low-loss alternatives to conventional bulk optical elements for controlling the amplitude, phase, and polarization of light. These structures are of growing interest to researchers and engineers working on sensing, energy harvesting, and flat optics, as their performance hinges on precisely engineered optical resonances. Full-wave finite element simulation, combined with semianalytical multipole decomposition in the COMSOL Multiphysics® software, gives us a way to not only predict these resonances but also uncover their underlying physical origins.

In this webinar, Dr. Pavel Terekhov, postdoctoral researcher at National Institute of Standards and Technology, will trace how a single quadrumer meta-atom, originally studied for its magnetic octupole response, evolves into two distinct light manipulation regimes. He will first revisit the foundational single-particle results that motivated this work, then show how arranging quadrumers into a periodic crystalline silicon metasurface produces anomalous absorption enhancement, governed by two independent multipole mechanisms coexisting in the same structure. Building on this, he will then introduce ongoing work on a gallium nitride metasurface, where the complex interplay of four different multipoles is used to sculpt reflection and transmission spectra including the quasi-bound-states-in-the-continuum (q-BIC) manipulation.

Attendees will see how COMSOL Multiphysics® and multipole decomposition connect full-wave simulation and analytical insight, turning abstract resonance behavior into physically interpretable design rules. The broader takeaway is that multipole-based simulation is not just a diagnostic tool but a design strategy: It enables the on-demand tailoring of absorption, reflection, and transmission in dielectric metasurfaces, with direct relevance to sensing, energy harvesting, and future optical device applications.

Key Takeaways:

• Learn about modeling multipole resonances in dielectric metasurfaces using full-wave finite element simulation and semianalytical multipole decomposition in COMSOL Multiphysics®.
• See how multipole-based simulation can be used to understand and control absorption, reflection, and transmission in silicon and gallium nitride metasurfaces.
• Explore how different multipole mechanisms, including quasi-bound states in the continuum (q-BICs), can be engineered to tailor optical responses for sensing, energy harvesting, and flat optics.
• Gain insights into how simulation and multipole decomposition can help researchers and engineers turn complex resonance behavior into practical design strategies for future optical devices.

It’s disheartening how much power gets generated and then promptly lost as it travels through grid networks. This leaking of electricity happens when it vanishes as heat as well as when it is pilfered by thieves and nonpaying customers.

More than half of the countries that track these metrics lost at least 10 percent of their electricity in 2023, according to the World Bank. Losses topped 20 percent for 24 of those nations. Two countries lost more than half of what they generated.

With numbers this high, cutting down on losses makes sense. Electricity demand is rising beyond what many grid operators can supply; reducing waste would help meet some of that demand without having to build new power plants. Plus, when the power comes from fossil fuels, any loss means emitting even more greenhouse gases into the atmosphere. And electric losses hit the bottom lines of power providers, which ultimately pass those costs on to everyone else.

The trouble is, reducing electricity losses is a hard and expensive process that takes a long time. Typically, the less maintained the grid infrastructure, the more electricity that’s lost. And the more fragile the region’s law enforcement and government, the more prevalent the power theft. Natural disasters and war make things worse.

Delhi’s Power Grid Comeback

Fixing a power grid requires a systemic approach across many sectors. There’s no one technology that will solve the problem. At the outset, the obstacles to success may feel insurmountable. Equipment across entire grid networks must be updated. Multiple arms of government must agree to reforms and coordinate to ensure power providers are set up to succeed. Regulations must be written or revised, investments made, cultures changed.

The city of Delhi did all those things. Over the past 25 years, it cut its electricity losses from about 50 to 5 percent. How the city pulled off that impressive feat is the focus of “The Epic Comeback of Delhi’s Power Grid” by Mini Shaji Thomas, an electrical engineer at the university Jamia Millia Islamia who has lived in Delhi since the 1990s. She gives us a view from the inside—as a resident and a power systems expert.

Delhi is a shining example, but some other regions have significantly lowered electricity losses over the last quarter century too. The country of Georgia went from losses of over 16 percent in 2002 to about 8 percent in 2023. In Singapore, losses dropped from 6.6 percent to a nearly nonexistent 0.2 percent over the same time period.

Global Electricity Theft Crisis

But there are many parts of the world where electricity losses remain a problem or have gotten worse. In Jamaica, where power theft is rampant, losses have hovered between 21 and 28 percent for years. Argentina’s losses nearly doubled between 2015 and 2023, going from an all-time low of about 12 percent to an all-time high of nearly 24 percent. The main problem: Transmission and distribution companies lacked the capital to maintain and upgrade their networks, which left equipment operating under stress. A delay in the installation of smart meters has allowed thieves to more easily siphon power and tamper with meters.

Thomas says she hopes her account of Delhi’s grid comeback will serve as a blueprint for others. It’s possible to replicate the sweeping changes Delhi made, she says. But it “requires a concerted effort from all stakeholders, customers, the utility, the government, and their employees.”

It’s 6 a.m. on a cold January morning in 2002 in New Delhi. It’s still dark outside, and I’m in the kitchen preparing breakfast, packing lunches, and getting my two children ready to catch the school bus when, for the third time in a week, the power goes out. No lights, no mixer to finish my daughter’s puttu—her favorite rice dish—no kettle, no toaster. The bathroom is dark, and the kids are upset.

It will probably be hours before the power comes back on, so I grab a flashlight and light the candles that are set up around the house for these occasions. We’re behind schedule now. We pack the food we have, bundle up as the house turns chilly, and head outside, leaving a mess in the kitchen. We make our way to the bus stop in the dark—the streetlights are out, too—only to discover my daughter has missed her ride. Again. I’ll be late for work at Jamia Millia Islamia, a university where I am a professor of electrical engineering and teach power systems and smart grids. I just hope the power is on there.

This was a common scene for my family and all of Delhi in the early 2000s. Power outages happened almost daily and lasted hours. When the power was on, the quality was so poor that it would dim lights, flicker screens, and wreak havoc on appliances. Customer service at the power utilities essentially didn’t exist.

A child walks in July 2007 past a store in New Delhi specializing in reconditioned generators. The fear of power cuts during summer heat spurs demand for these generators so that residents can produce their own power.Nicholas Bradley/AFP/Getty Images

These problems had been getting worse through the 1980s and 1990s. The cause: an aging distribution grid bereft of crucial technologies, and electricity providers with little accountability. The situation became so bad that the city was losing more than half of its power through obsolete equipment and theft. These staggering losses meant that utilities got paid for only a fraction of the electricity they were trying to deliver. And the lack of funds prevented them from investing in better grid infrastructure.

But over the last quarter century, a remarkable effort by the government and the city’s distribution utilities has turned Delhi’s grid into a reliable, modern system. Power losses have shrunk from over 50 percent in 2002 to 5 to 6 percent in 2026—on par with France and Belgium, and better than Greece and Serbia. Delhi’s grid reliability index, a measure of how often electricity can be counted on, stood at around 70 percent in 2002 and has now topped 99.9 percent.

The bustling Main Bazar in the Paharganj neighborhood of Delhi increasingly uses more nighttime electricity, but reductions in electricity loss help counter demand. iStock

With reliable power, businesses across the city have blossomed. The streetlights are bright. The number of electric vehicles, including city buses, is growing daily. Quality of life has improved. Today, my family is comfortable year-round in our home despite Delhi’s scorching summers and cold winters. The chaos of losing power no longer hinders me from getting to work. The city still has problems—pollution, overcrowding, noise—but thankfully, reliable power is no longer among them.

The transformation of Delhi’s grid can serve as a model for other cities that suffer from decrepit power infrastructure. Regions of Albania, Argentina, Bangladesh, Brazil, Estonia, India, Kenya, Pakistan, Sri Lanka, Uganda, and Venezuela are reeling from heavy losses in their distribution grids. Their problems look like Delhi’s 25 years ago. I believe it’s possible to improve electricity in these places by adapting the changes Delhi made. Here’s an inside look at how the city accomplished it.

Delhi’s Power Grid and Energy Mix

The city of Delhi hosts the capital of the Republic of India, and sits along the Yamuna River in the northern part of the country. It’s home to about 23 million people and is one of the most densely populated areas in the world. Delhi’s grid includes thousands of kilometers of power lines, and peak electricity demand reached an all-time high this year of 8,748 megawatts. The city currently buys 76 percent of its power from central generating companies and private players from neighboring states. Energy generation within the city is restricted to natural gas and renewable sources. Nearly 48.5 percent of the city’s power comes from coal, about 26.5 percent from natural gas, and the rest from carbon-free sources, led by hydropower at 15.6 percent.

Tata Power-DDL replaced about 5 kilometers of overhead lines with underground cables, which reduced electricity loss and improved the aesthetics of Delhi’s streets, such as the Janta Flats in the Shalimar Bagh neighborhood.Tata Power-DDL

By the early 2000s, Delhi’s nearly 100-year-old power distribution system was in serious disrepair. Everything was old—lines, transformers, circuit breakers, switches. New grid technologies were needed to keep up with new kinds of electricity loads, but there was little money to upgrade components.

The shabby state of the grid caused many problems, most notably high electricity losses, where electricity vanishes primarily as heat. The cause of the losses was a classic electrical problem: too much current flowing through a network that wasn’t designed to carry it efficiently.

To understand the problem, it helps to understand how modern power grids work. Typically, they include generation, transmission, and distribution. After power is generated, transformers convert the electricity to high voltage levels—typically 132, 220, 400, or 765 kilovolts in India. Transmission lines then carry the power over long distances to receiving substations that are closer to where customers need electricity. Transformers then step down the voltage (to 66, 33, or 11 kV in India) and distribution lines branch out, carrying the power to customers. The whole grid works primarily on alternating current.

Electricity Losses by Country, 2002 vs. 2023

Distribution networks carry both active and reactive power. Active power is the energy used to perform useful work (and is measured in watts). Reactive power is the power that flows back and forth in an electric circuit, building electric and magnetic fields (measured in volt-ampere-reactive, or VAR). Although it doesn’t perform useful work, reactive power is necessary for many devices, such as induction motors, transformers, and computers (typically any circuit or device with inductance or capacitance elements).

When there are a lot of devices consuming reactive power on the same line, the overall current carried by the line—the sum of the active and reactive current—must increase. The more current in the line, the more the line heats up and the more energy that’s wasted as heat.

In addition to current, resistance in the line will increase losses as well. Resistance is when electrons encounter opposition as they move through the conductive material (typically aluminum in a power grid). Longer lines with many branches and connection points will increase resistance. The rule of thumb is that line loss equals the square of the current multiplied by the resistance.

Reactive power creates a second problem: It causes the voltage along the line to drop. And when the voltage falls, many modern electrical devices try to maintain roughly the same level of performance by drawing more current. That higher current produces even greater losses in the line and causes the voltage to fall further.

Meter Technology Impacts Electricity Losses

In a healthy grid, the utility will take compensatory measures to lower the current and maintain the voltage all the way to the ends of the lines. But in Delhi, this wasn’t happening. The result was a vicious cycle. Reactive loads increased the current, the higher current increased energy losses and lowered the voltages, lower voltages forced devices to draw more current and further increased the losses.

In some parts of Delhi, the effect was so severe that residents took matters into their own hands. A colleague of mine who lived in a different part of the city constantly experienced voltage that was too low for her appliances to operate reliably, so she had to install her own voltage stabilizer. At my home, we bought an inverter and battery system to keep a fan and a few lights running during the many outages.

Electricity Loss and Theft in Delhi

The losses in Delhi weren’t caused solely by technical problems. Theft of electricity was rampant, by both the powerful and the powerless (in both senses of the word). Businesses, residential customers, and utility employees with vested interests would siphon electricity from the grid. It was easy to illegally hook into a streetlight or a distribution line running close to one’s house or factory. Utilities didn’t have the resources to identify theft or penalize offenders. Even if they could, the courts were already overburdened, and an electricity regulatory commission that could push for reforms had not yet fully formed.

Updated meters have made billing easier and more accurate. Tata Power-DDL

Making matters worse, the utilities and their employees were rarely held accountable for their actions, and so corruption plagued the system. Junior engineers and line workers, many of them lacking appropriate technical skills, were tasked with handling nearly every issue, including outages, flickering, and bill payment. This was too much authority in the hands of people with too little training.

On top of that, customers didn’t pay their bills. Meters were old, frequently faulty, and easily tampered with. Utility employees would take a meter reading by visiting the customer’s property, noting the reading in a book, entering it in a ledger or on a computer back at the office, and converting it into an electricity bill that would get dropped off at the customer’s property. This process left a lot of room for incorrect billing.

To pay a bill, customers had to stand in long queues at the utility offices, which had limited business hours. Not wanting to take off a half day of work for this, many customers simply didn’t pay. And there was no penalty for not paying—there were no regulations allowing the utilities to cut off a customer’s power. (I paid my bill by having a family member stand in line for me.)

The combined commercial and technical losses left Delhi’s utilities collecting payment for less than half of the electricity they were supplying in the early 2000s.

India’s Electricity Act and Power Reforms

Such problems weren’t unique to Delhi. On average in 2002, state utilities across India experienced electricity losses of nearly 37 percent. My country desperately needed systemic reforms, but authority over electricity was split between the central and state governments so any decision-making was fractured. States managed most of the generation, as well as transmission and distribution, while the central government oversaw generation that supplied multiple states, such as hydropower, fossil fuel plants, and nuclear plants. The central government could push reforms, but the states determined whether those reforms would succeed. Making matters worse, most states put a single organization in charge of generation, transmission, and distribution, giving that entity too much control and reducing transparency and competition.

A team of technicians with BSES Rajdhani Power maintains an insulator string on a large power transformer in 2011. BSES Rajdhani Power

In 2001, India’s central government began writing some historic legislation that became the landmark Electricity Act, 2003. Among the grand reforms aimed at transforming the country’s power industry, it unbundled state oversight of grid networks, creating separate entities for generation, transmission, and distribution. It also opened up the power sector to privatization. It allowed large electricity customers to bypass local distribution companies and purchase electricity from competitors or build their own power plants. It created a central regulatory agency responsible for determining interstate tariffs and promoting market competition in the power sector. And it created mechanisms for prosecuting electricity theft.

Hundreds of capacitor banks have been installed in Delhi to supply reactive power at strategic locations and help stabilize voltage.Tata Power-DDL

In 2002, Delhi was already taking drastic action to fix its grid. The organization overseeing Delhi’s distribution, the Delhi Vidyut Board, was broken up and two private companies—BSES (now part of Reliance Infrastructure), and North Delhi Power Limited (now known as Tata Power-DDL)—took over distribution. They faced a Herculean task. Tata Power-DDL, serving the northern half of Delhi, would have to tackle a combined commercial and technical electricity loss of 53.5 percent. BSES, whose territory was split between two subsidiaries, was facing 51.5 percent losses in South Delhi and 63.1 percent losses in East Delhi.

“The company inherited a deteriorated and overloaded network, massive power theft, weak billing and collection systems, inaccurate consumer records, and an aging, largely untrained workforce,” Dwijadas Basak, CEO of Tata Power-DDL, told me. There were over 100,000 unresolved billing complaints, 20,000 pending connection applications, and frequent supply failures, which had severely eroded consumer trust, he added. Both Tata Power-DDL and BSES devised sweeping reforms and human resource development initiatives. The companies followed their own paths over the years, but ultimately implemented similar changes, with similar results.

Delhi’s Electricity System Overhaul

Fixing Delhi’s grid was a journey that involved all stakeholders, including customers, city authorities, and utility employees at all levels. The utilities revamped their organizational structures, diminishing the power of junior staff and creating separate teams to focus on specific tasks. Long-term employees of the erstwhile Delhi Vidyut Board received training from the up-and-comers at the new companies.

On the technical side, both companies installed digital control systems that let them monitor and operate the grid from a central location. Known as SCADA, or supervisory control and data acquisition, the systems offered a bird’s-eye view of the infrastructure, including the status of equipment, voltage, current, power flow, and switch positions, with updates in seconds. This helped the companies identify areas of high loss and theft and make faster decisions based on accurate information.

The SCADA (supervisory control and data acquisition) system at Balaji Estate in Delhi’s Kalkaji neighborhood serves as the nerve center of BSES Rajdhani Power’s distribution network in South and West Delhi. It enables real-time visibility, remote control of grid operations, fault identification and isolation, and load management. BSES Rajdhani Power

The utilities also replaced aging transformers and circuit breakers and created extensive maintenance plans for equipment. In 2002, 11 percent of the transformers in the region were failing at any given time. That rate is less than 1 percent today, according to Tata Power-DDL. Crucially, the companies installed hundreds of capacitor banks, including some mobile ones, to supply reactive power at strategic locations. This improvement reduced the total current flowing in the distribution lines and helped stabilize the voltage. They also installed voltage regulators at points in the system where voltage tends to drop.

To reduce theft, the companies replaced bare distribution wires with insulated lines—a single cable for three phases—which made it harder to tap into the lines. The cables also reduced outages because they’re better at preventing ground faults, which can occur when, say, a tree branch falls on the line.

Workers received better sensors and tools to do their jobs safely and accurately. For instance, they were given helmet-mounted voltage sensors, which are safer than handheld ones, and thermal scanning tools to detect hidden defects in the insulation of high-voltage equipment that could otherwise have led to catastrophic failures.

To reduce inaccurate billing and meter tampering, the companies replaced the old electromechanical meters with digital ones that are read with handheld devices. In some locations, radio-frequency-based group metering systems were installed by Tata Power-DDL to consolidate multiple customers’ meters into one. The data is then wirelessly transmitted to a central database, eliminating the need for individual meter readings. The companies are now trying smart meters, which give consumers more control over their electricity bills and give utilities remote control of some equipment (with the customer’s consent).

To encourage people to pay their bills, the utilities installed kiosks that are available 24 hours a day, and they created a web-based payment system and mobile app. Incentives for early bill payment and community-engagement programs also helped. Assistance from Delhi’s law enforcement considerably reduced electricity theft.

Tata Power hired women living in the 223 slums it serves in the northern parts of the city to knock on neighbors’ doors and remind them to pay their power bills. These payment collectors [left and center], known as abhas, were photographed while speaking with a customer [right] in the Sanjay Basti area of New Delhi in 2017. Prashanth Vishwanathan/Bloomberg/Getty Images

In areas where theft was particularly rampant and losses were as high as 83 percent, according to Tata Power-DDL, the companies took a different strategy. These pockets of Delhi were predominantly occupied by low-income families. Tata Power-DDL, and later BSES, worked to improve the water supply for these residents and provide educational opportunities, such as instruction in reading and writing in Hindi as well as financial literacy. These efforts focused on the women, who were at home more, and paid them to collect electricity payments from their neighbors. Bill payment rates from these areas are now on par with those of other parts of Delhi.

In recent years, some customers have been installing rooftop solar panels to take advantage of subsidies and incentives. This trend can reduce electricity losses further because the energy generated at the customer end reduces current in the distribution lines. Customers are also installing more LED lights and energy-efficient appliances, reducing the load in the system.

BSES is using AI to help detect theft. The algorithms analyze consumption patterns in pockets where losses are higher than they should be. The company is also using AI to forecast demand, fine-tune operational efficiency, and provide chatbots for customers.

Quality of Life Improves in Delhi

Life in Delhi is better than it was 25 years ago. I’m not worried that the power may go out and force me to reschedule my activities. My uninterrupted Wi-Fi gives me peace of mind, and my heating and cooling systems keep me and my family comfortable. I rarely need to use our old inverter and battery.

The sharp rise of e-rickshaws in Delhi has increased demand on the power grid. Sajjad Hussain/AFP/Getty Images

The number of businesses in Delhi has increased substantially, in part because of the access to quality power. People can confidently buy products that depend on electricity. In fact, the city’s peak electricity demand has tripled since 2002 due to the increase in population, commercial activity, and use of electrical gadgets.

And then there’s the benefits to the planet. One unit of electricity that isn’t frittered away is one less unit that must be generated, not to mention the reductions in carbon emissions.

Still, there’s work to do. Some areas of Delhi continue to have high losses, driven partly by the illegal charging of e-rickshaws. Elsewhere in India, the states of Himachal Pradesh, Madhya Pradesh, Maharashtra, and Telangana still experience losses of about 17 to 23 percent despite the sweeping Electricity Act, 2003. There are many reasons for the ongoing losses: long distribution lines to remote villages, less digitization, and inefficiencies in billing and collection of payments.

These regions, and others around the world, can learn from Delhi’s grid comeback. Recently, power losses have increased substantially in countries such as Argentina, Greece, Jamaica, and Morocco, according to the World Bank, and some of the causes are similar to those that Delhi faced back in 2002.

Meanwhile, Australia, most countries in North America and Europe, and a few countries in Asia and Africa experience low electricity losses as they invest regularly in their distribution infrastructure and the ethical enforcement of rules. In China, for example, losses have gradually been cut in half, from 7.1 to 3.4 percent. In Latvia, losses plummeted from 25 to 5.8 percent.

What’s important is a comprehensive approach. Technologies like smart metering, AI, and analytics certainly help, but equally important is that people in the field are trained and take responsibility for their jobs, and that laws are enforced and payments collected.

“Sustainable loss reduction cannot happen through technology alone,” Abhishek Ranjan, CEO of BSES Rajdhani Power told me. “Technology is an important enabler, but long-term success comes from combining it with disciplined execution, operational accountability, and strong consumer engagement.”

This article was updated on 30 September, 2026 with clarification about the full name Tata Power-DDL.

Geothermal startup Endurance Energy began drilling a well last week into the seafloor in an attempt to harness energy from an underwater volcano. The well, located 1,500 meters below the surface of the Pacific Ocean off the Oregon coast, will give the company access to hot fluid from Earth’s crust, which will spin the turbine of an underwater power generator.

The project takes an elaborate approach to tapping geothermal energy that’s far more accessible elsewhere. Nearly every other geothermal company is operating on dry land while Endurance is adding a host of engineering challenges by doing it in deep water. But the company is betting that the approach will prove valuable for small islands needing an alternative source of energy as they try to wean off expensive, imported diesel.

Subsea geothermal is “a slightly dingbat idea, at least for grid generation,” but it has potential in certain use cases such as generating power for islands, says Roland Horne, director of Stanford’s geothermal program. But he remains skeptical. Companies have long tried and failed to make offshore geothermal work at scale, and questions remain about logistics and performance in the harsh conditions of the deep ocean.

Endurance, based in Seattle, acknowledges the ambitiousness of the plan. “There’s very, very little precedent” for successfully generating power with subsea geothermal technology, says Endurance CEO Andrew Redd. The Oregon project “really serves as a learning pilot for us,” he says.

How to Drill for Offshore Geothermal Energy

Over the past several weeks, Endurance has been using remote sensing to survey the Pacific Ocean’s seismic and volcanically active Juan de Fuca Ridge in search of a few target locations to install its system. From the command center aboard the Ocean Guardian, a Stabbert Maritime vessel, the company’s team identified a few flat areas on the seafloor near the Axial Seamount, the region’s largest and most active volcano, about 500 kilometers from the Oregon coast.

Last week, the company chose a spot and drilled a shallow, 16-meter well to gain access to hydrothermal fluid, or briny water, at around 250 °C. That temperature is within the 250–350 °C “sweet spot” that will allow the team to “still use a lot of pretty normal terrestrial materials,” Redd says. The well’s bore hole, at 5 centimeters in diameter, is considerably narrower than those of land-based geothermal systems, which draw from wells that are about 20–30 cm in diameter.

Endurance Energy’s 100-kilowatt power system, Adélie, sits fully assembled on the dock at the company’s Seattle headquarters. Dan Chosich

Next week the team plans to submerge its 100-kilowatt power system, dubbed Adélie, and place it on the seafloor about 20 meters away from the well. A submarine robot will then unspool flexible piping to connect the well to the generator and turn on the flow of hydrothermal fluid.

The power system’s organic rankine cycle technology uses a closed-loop heat-transfer process to generate electricity by spinning a turbine. This kind of generator has not been hooked up to a well on the seafloor before, so the team selectively chose materials that can withstand the undersea environment, Redd says.

Once fluid from the well is connected to the power system, electricity production will begin. But for this deployment, the company will be focused more on data collection than power. A fraction of Adélie’s electricity capacity will be used to power sensors, and the rest will be released back into the water as heat. Temperature and flow-rate data from Adélie will travel back to shore in real time via an existing fiber-optic cable network operated by the National Science Foundation’s Ocean Observatories Initiative–part of the reason Endurance chose the site.

The Juan de Fuca Ridge “is an extremely well-researched, well-studied, and well-surveyed part of the ocean floor,” Redd says. “So we’re really focusing on how much can we learn from this process while being very good stewards of the vent systems.” Before the deployment, Endurance took baseline temperature and environmental measurements so that it can later evaluate the technology’s impact on the surrounding ecosystem.

Redd says the company has already learned a lot through the system’s design and coordination with existing offshore resources, like the array of research cables lining the ocean floor between the Axial Seamount and the coast of Oregon.

Tonga Turns to Subsea Geothermal

Endurance Energy aims to operate on the Axial Seamount for a year, Redd says. By the end of 2027, the company plans to deploy a 2-megawatt unit off the coast of the Kingdom of Tonga. Like many island nations, Tonga faces high prices for diesel, which has to be shipped from abroad at the whims of geopolitics. The country’s partnership with Endurance is part of a strategy to build the island’s energy independence.

Unlike the pilot near Oregon, the Tonga system will generate power for the island, and won’t be installed quite so close to active volcanoes, where seismic activity is a risk.

Endurance Energy’s power system connects to a seafloor fiber-optic network that runs all the way to shore, allowing the company to monitor and control its system in real time.Canadian Scientific Submersible Facility

Endurance will have to lay submarine cables to transfer the power to shore which will add to the overall cost, the exact details of which Redd says are still being determined. He says Endurance’s system will be competitive with the cost of the diesel upon which many Pacific Island nations like Tonga currently rely.

Endurance will surely face doubts about the viability of subsea geothermal for large-scale generation, especially when there are plenty of terrestrial geothermal resources waiting to be tapped. “It’s not like oil and gas; that we ran out and had to look somewhere else,” Horne says.

But Endurance’s team is steadfast in its belief that its technology will be cost-competitive at scale. And as for building out all of those underwater cables–Endurance sees that as an opportunity. Offshore, there’s more flexibility with the transmission corridor than onshore, where existing transmission lines bring limitations, says Jennifer Kenyon, Endurance’s director of policy and external affairs.

“If offshore wind figured it out…and they’ve managed to be cost-competitive in parts of the world, I think there’s [a] ‘there’ there,” Kenyon says. “We just need to build to get to it.”

This article was updated on 25 and 28 September, 2026. It was corrected to reflect the accurate owner of the Ocean Guardian vessel, the manner in which excess heat is released into the water, and the depth of the well that Endurance drilled.

This white paper gives distribution engineers and utility teams a practical overview of the substation exit, the short and high-consequence section where station capacity divides into individual feeders. It explains how covered, spacer-supported overhead construction can reduce common contact-driven faults while leaving room for future circuits.

What you will learn about:

• Why a fault near the substation exposes more customers than one farther along the feeder, and why the first spans out of the station carry so much reliability weight.
• How covered conductors and spacer-cable systems differ from bare overhead conductors, and why covered conductor is not treated as touch-safe insulation.
• Which engineering factors shape a sound exit design, including conductor rating, protection coordination, grounding, and structural loading.
• How overhead spacer cable compares with conventional bare overhead and underground shielded cable across footprint, reliability, and lifecycle cost.
• How to move a project from concept to commissioning using staged design gates and a clear performance specification.

Typically, if you swallow a battery, it’s a mistake that calls for a trip to the hospital. But that’s not the case with a new ingestible paper battery developed by scientists at MIT. The battery could be used to power medical devices inside the body before safely breaking down.

In tests in live pigs, the paper battery “generated sufficient power to perform meaningful functions inside the gastrointestinal tract, including meter-scale wireless communication and continuous electrical stimulation over several days,” says Giovanni Traverso, a professor of mechanical engineering at MIT.

Smart Pills and Ingestible Electronics

There is a broader trend in healthcare to develop smart pills. These ingestible electronics can analyze and treat the body from within in real time. However, ingestible electronics need a safe and reliable source of power.

Conventional alkaline or lithium-ion batteries can prove deadly if their protective casings leak inside a person. Previous research has sought to develop energy-harvesting techniques for ingestible electronics, but these usually generate low and inconsistent levels of power. Scientists have also designed biodegradable batteries, but these often can only store limited amounts of energy.

Traverso and his colleagues developed a paper-based battery with electrodes made from magnesium and molybdenum trioxide, as detailed in a paper published 21 September in Nature Chemical Engineering. Prior work had explored both materials for use as biodegradable batteries but had used bulky adhesives to bind large particles of these materials into batteries, resulting in thick electrodes that took longer to break down.

The new batteries instead use cellulose nanofibrils (essentially, paper) to bind the electrode materials to the battery. These fibrils are sturdy, can load electrode particles into their porous structures, and can be thin and biodegradable, Traverso says.

Battery electrolytes—which help electric charges shuttle between a battery’s electrodes to discharge electricity—are often toxic. The new paper battery uses a biodegradable ionic liquid gel based on choline chloride and lactic acid as its electrolyte, which yielded more stable performance than several other biodegradable electrolyte approaches, Traverso says.

“Encapsulation was also critical,” Traverso says. “If gastric fluid reached the battery too quickly, its capacity declined prematurely. If the encapsulation persisted for too long, the device would not degrade as intended. Natural wax coatings provided a way to delay fluid penetration and tune the operating lifetime.”

Researchers developed bioresorbable electrodes for an ingestible paper battery. After 90 days under accelerated conditions, the electrode material has almost completely dissolved.Source images: Mehmet Girayhan Say, Ada Erus, et al.

Paper Battery for Gastrointestinal Devices

In tests in live pigs, the batteries generated a peak voltage of 1.84 volts and remained operational in the stomachs of the swine for up to three days. They could power an RFID tag in a pill that, after it was swallowed, maintained stable wireless communication with a receiver 1.5 meters away. Similar smart pills could theoretically help doctors track whether a patient has indeed taken prescribed medication. The batteries could also power a capsule that electrically stimulated the stomach for 20 minutes, an electroceutical technique used to treat gastrointestinal conditions involving impaired digestive function, nausea, or loss of appetite. There was no notable tissue damage from this stimulation. The RFID tag and the circuit board used for electrical stimulation was not biodegradable, but was naturally excreted.

“The battery could potentially power ingestible sensors that measure temperature, pH, pressure, motility, or biochemical signals; systems that transmit physiological information; controlled drug-delivery devices; temporary electrophysiological recording systems; and gastric or intestinal electroceutical devices,” Traverso says. “It may be particularly valuable for devices designed to remain in the stomach for several days, because a bioresorbable power source could reduce or eliminate the need for retrieval.”

The scientists are now prototyping devices that combine these new batteries with biodegradable antennas and RFID systems. The goal is a human clinical trial of a device that can monitor how well patients are taking medicine, “which we aim to start in about two years,” Traverso says.

A fusion startup on a shoestring budget hopes to leapfrog more established companies by reviving a technology that’s been sidelined for half a century.

The startup, Terra Fusion, in Baltimore, is developing a reactor that uses a relatively simple approach known as magnetic mirrors to confine hot plasma–the environment where fusion reactions can occur. The company is testing its strategy on an experimental reactor at the University of Maryland, built with parts from secondhand MRI scanners and repurposed brewing equipment.

The goal is to eventually provide a route to commercial fusion at a fraction of the cost of competing systems, says Carlos Romero-Talamas, CEO at Terra Fusion. To that end, the company is developing a more powerful proof-of-concept device that aims to exceed scientific breakeven, where a reactor generates more power than is used to heat and confine the plasma inside of it.

It’s a key step toward the ultimate goal of generating useful amounts of carbon-free energy. More than 50 fusion companies around the world are working toward that goal, and over the past year have attracted a record US $4.5 billion in new funding, surpassing government spending on the technology for the first time.

Most of that money is going to large organizations working on technologies that appear closest to breakeven. These include giant toroidal tokamaks and pulsed laser systems. But that doesn’t mean it’s too late to try a different approach, says Jason Cassibry, professor of mechanical and aerospace engineering at the University of Alabama in Huntsville and an advisor to several fusion startups.

“Magnetic mirrors have the advantage of being simple,” he says. “If a nation or company wants to be first and have the highest power density, they might go with a tokamak. But if they want something that’s cheaper to build and easier to maintain, they might go with a mirror concept.”

Terra Fusion is now in a race to demonstrate a bargain-basement reactor before its mainstream rivals.

How do magnetic mirrors confine plasma?

Building a nuclear fusion reactor is an attempt to bottle the same reaction that powers the sun. The tremendous gravitational force within stars propels hydrogen nuclei to fuse together into helium, releasing vast quantities of energy in the process. Scaling that down to a reactor on Earth requires either confining an incredibly hot plasma with magnetic fields or compressing and heating targets filled with fuel, in both cases usually deuterium and tritium. Decades of fusion experiments pushing the limits of magnet, laser, and materials technologies have shown just how difficult that is.

Magnetic mirrors are a particular configuration of natural or artificial magnetic fields that can trap energetic particles for long periods. They can be found in the Earth’s Van Allen radiation belts, for example, which can capture electrons and protons from the sun for years, causing hazards to spacecraft and astronauts traveling between 1,000 and 60,000 kilometers in altitude.

During the Cold War, researchers used traditional electromagnet coils at each end of a vacuum chamber to form magnetic mirrors on Earth. The hope was that the mirrors would confine hydrogen nuclei within a plasma long enough for them to be heated to the point of fusion.

But those efforts failed. The plasmas proved unstable, and the mirrors were leaky, allowing energetic particles to escape. When the world’s largest and most expensive magnetic mirror reactor was mothballed at Lawrence Livermore National Laboratory the day after being completed in 1986, magnetic mirrors appeared to be a dead end. Enormous donut-shaped tokamaks seemed to offer a smoother path to fusion breakeven.

But some mirror research trickled on. At the University of Maryland, engineers were intrigued by earlier experiments that suggested that the magnetic mirror strategy could work if the plasma was rotated at supersonic speeds. This could stabilize the plasma and reduce leaks. From 2004 to 2010, Romero-Talamas, who was a postgrad student at the time, and other researchers operated this kind of reactor, called the Maryland Centrifugal Experiment, which gave some impressive results.

“With only a few megawatts of power input, we were getting similar [temperature] results to the Livermore spheromak [a reactor similar in shape to a tokamak] with hundreds of megawatts,” he says.

Nuclear Fusion Startups Seek Breakeven Power

In 2010, the University of Maryland applied for U.S. Department of Energy funding for a larger magnetic mirror device but was unsuccessful. Tokamaks were in vogue, particularly the gargantuan ITER reactor being developed by China, the European Union, India, Japan, Russia, South Korea, and the United States.

It was not until 2020 that Romero-Talamas, now an associate professor, secured a major grant: $5.2 million from ARPA-E (the U.S. federal government’s Advanced Research Projects Agency–Energy). The sum was a tiny fraction of the more than $1.5 billion the U.S. had already spent on ITER (now nearing $3 billion), but it would be enough to build a reactor, dubbed the Centrifugal Mirror Fusion Experiment (CMFX) if Romero-Talamas was frugal.

As with many fusion reactors, CMFX’s priciest components would be its superconducting magnets. Instead of buying them brand-new, Romero-Talamas repurposed them from two Philips medical MRI scanners by stripping out the scanners’ patient beds, scanning coils, and plastic cases. He also modified glass beer bottles for use as insulators and acquired some donated equipment from the Air Force Research Laboratory. The Maryland Space Grant Consortium paid for summer interns to work on the project.

Although the scanner magnets could generate the strong 3 Tesla magnetic fields CMFX required, the cavity for human patients around which they were designed was larger than Romero-Talamas had planned for, forcing a redesign of the reactor.

Inside the CMFX reactor, rotation begins [top left] and the plasma is still relatively cold, indicated by brighter light. The plasma starts heating up [top right] as it supersonically rotates. As the plasma reaches its highest temperatures (about 10,000,000 Kelvin), the camera struggles to capture any light [bottom]. Artur Perevalov

The team designed CMFX with a central electrode running through the vacuum chamber between the mirrors that operates at a voltage similar to power transmission lines. The interaction of the electric and magnetic fields drives a plasma of deuterium ions to rotate around the electrode at well over a million meters per second, creating a centrifugal force that balances the plasma and magnetic pressures. The faster the plasma spins, the tighter and hotter it burns.

Results published last year confirmed that rotating the plasma suppressed the worst instabilities, reduced particle losses, and helped heat the plasma. At its best, CMFX reached about one-tenth of the numbers needed for thermonuclear fusion in each of temperature, density, and plasma confinement time.

“You might say, Wait, you’re still a factor of a thousand away just from breaking even,” says Romero-Talamas. “But this is a huge achievement. The machine performed exactly as we designed it, which is very unusual in plasma physics, where usually everything goes wrong.”

The oversize bore of the MRI magnets even helped in the end, as the larger vacuum chamber made it easier to control impurities in the plasma. In 2024, Romero-Talamas felt confident enough to take a sabbatical from the university and form Terra Fusion.

Is Terra Fusion’s Reactor Scalable?

The young company is part of a recent resurgence in interest in magnetic mirrors, says Cassibry. “The magnetic mirror seems to be on the upswing. New concepts are coming out because the tools to analyze them in detail require very robust computational models, and we just have the tools now,” he says.

There are now mirror startups based in India and Sweden, and another ARPA-E funded university spin-out in the U.S.: Realta Fusion, which emerged from the WHAM reactor at the University of Wisconsin-Madison. WHAM’s larger $10 million federal grant was enough for two custom-made magnets that use high-temperature superconducting (HTS) tape to generate a massive 17 Tesla field.

With the powerful magnetic field doing most of the confinement, WHAM, which stands for Wisconsin HTS Axisymmetric Mirror, only has to spin the plasma just enough to avoid instabilities. That means Realta can avoid putting electrodes in the burning heart of the reactor as Terra Fusion does. “That thing’s going to get chewed up in a power plant,” says Realta’s chief science officer, Derek Sutherland. “We think our approach is a much lower risk path.”

Terra Fusion’s first milestone is to build a proof-of-concept reactor called TFEC-1 (Terra Fusion Energy Corporation 1), using 10 Tesla magnets and a conductor operating at 10 times the voltage of CMFX. TFEC-1’s ambitious 2029 timeline to breakeven is similar to better-funded rivals like Commonwealth Fusion Systems and Helion Energy.

In Terra Fusion’s favor, the centrifugal mirror reactor will be much cheaper to build, says Romero-Talamas. “Our machine is much, much simpler, and a lot more straightforward in the engineering sense,” he says.

For example, Terra Fusion will eschew the latest high-temperature superconductors for magnets that need to be refrigerated close to absolute zero. This is older technology that’s available from a wider range of vendors at a much lower cost. The next step for the company is to build a production prototype.

Cassibry thinks that both Realta and Terra Fusion’s magnetic mirrors show promise. “Most likely, the scaling suggests that confinement and stability will persist all the way up to breakeven,” he says. “But whenever you build something, it’s always very humbling because there’s some piece of physics that you didn’t see coming.”

Ultimately, Terra Fusion hopes to be able to sell small, modular fusion reactors in the mid-2030s, generating up to 100 megawatts to power data centers, cargo vessels, and spaceships. Realta is shooting for a 50-MW reactor in a similar time frame that would be used in industrial processes that require intense heat, like concrete and steel production.

In the meantime, Terra Fusion’s demonstration reactor could be used to produce valuable medical isotopes, or as a test bed to develop neutron-resistant materials for rivals, Romero-Talamas says. The conductor that his team is developing to withstand the reactor’s harsh environment could also translate to more efficient and durable electrodes in industrial electric arc furnaces, potentially opening up another revenue stream.

“If I thought there was a better machine or a better concept to work on, I would be working on that right now,” says Romero-Talamas. “But I truly think that this is the way to get to scalable net energy.”

This story was updated 23 September 2026. Due to an audio transcription error, Carlos Romero-Talamas was misquoted as saying that his fusion reactor was getting the same temperature results as Livermore’s tokamak, when he actually said it was getting the same results as Livermore’s spheromak. The quote has been corrected.

This article is brought to you by CoolIT, an Ecolab Company.

Beyond 250 kW a server rack can no longer be cooled by a hybrid approach of liquid and air. At this density a 70/30 liquid-air split leaves 75 kW of air load. The air cooling system needed to move it brings cost and complexity few operators will accept. The answer is near-total heat capture. Liquid takes effectively all the heat, air falls below 1 percent of the load, allowing the server to run fanless.

CoolIT builds these loops today from modular coldplate blocks proven across six generations of fanless designs. Processor thermal design power (TDP) keeps climbing generation over generation. This rising heat load is now cascading into the memory, networking, storage, and power components that once ran comfortably on air.

The heat escaped the chip

For years the story stayed simple. Cool the processor and let air handle the rest. That balance has shifted. As TDP climbs, heat spreads outward from the processor and cascades into the components around it. Memory, networking, storage, and power now run hot enough to demand liquid of their own. Engineers designing the next generation of AI servers face a board where heat capture rises with every launch.

Beyond 250 kW per rack, air cooling becomes the bottleneck. Near-total liquid heat capture enables fanless AI server designs built for the next generation of computing.

New parts, new rules

Unlike processors, which are cooled as flat rectangular packages, these peripherals come in a wide range of shapes, sizes, and mounting requirements, each with its own thermal limits. Some run cooler than the processor case temperature, others run hotter, which leaves them sensitive to a design tuned only for CPUs and GPUs. Operators need purpose-built solutions here, matched to the part rather than stretched across the board.

CoolIT engineers meet this with a deep toolkit. Conductive plates, vapor chambers, heat pipes, and thermal transfer plates move heat from components closer to the liquid path. Riding coldplates enable pluggable components. Each solution stays true to the component it serves.

CoolIT Customer Showcase: How GWDG Cools HPC & AI Systems with CoolIT’s Direct Liquid Cooling CoolIT

One loop, one server

Cooling the parts is one challenge. Uniting them is the real work. Full heat capture means folding every one of these solutions into a single server loop that distributes coolant effectively and remains easy to install. Connection reliability, coolant routing, and the time it takes to assemble the loop at rack integration determine whether a design thrives in production or stalls on the bench. CoolIT builds these loops from proven modular blocks, so operators gain performance and deployment speed within the same solution.

Density forces the decision

Rack power continues to climb toward 1 MW, and the case for liquid grows stronger at every step. A 70/30 split of liquid to air holds comfortably at lower density. Past roughly 250 kW it stops working. The 30 percent left to air becomes a 75 kW load inside a single rack, and moving that much heat demands a parallel air system whose cost and footprint few operators will accept. Adding density only widens the gap.

As rack power continues to climb toward 1 MW, CoolIT’s modeling places full heat capture as the standard server design for flagship rack-scale products through 2028.

The simpler, more efficient answer is to capture the heat in liquid and drop air to less than 1 percent of the total load. True 100 percent remains almost impossible to reach in the strictest sense, so the honest and achievable target is near-total capture. That distinction matters to engineers who value precision, and the direction stays clear either way. Full heat capture moves from a premium option to a mainstream requirement as density rises, and CoolIT’s modeling places it as the standard server design for flagship rack-scale products through 2028.

CoolIT delivers it

CoolIT scales heat capture all the way to 100 percent using modular coldplate building blocks proven across six generations of fanless server designs. Engineering teams are already working on designs for the maximum density racks coming next. As the cascade spreads and racks grow denser, near-total heat capture becomes the design that keeps AI running.

Talk to CoolIT about building a server loop engineered for total heat capture.

Nowadays, solar cells are typically found on sunny surfaces on land or aboard satellites in space. A new study suggests photovoltaics can also work below the ocean’s surface, generating useful amounts of electricity underwater even at depths of 10 meters.

“With submerged solar cells, submarines or autonomous underwater vehicles could achieve far greater range and mission endurance than they do today,” says Wen-Hua Zhang, dean of Yunnan University’s school of materials and energy in Kunming, China. Many other underwater electronic devices, such as sensors, cameras, lights, and communications systems “could also benefit from longer operating periods and much broader deployment coverage in the ocean, to support a subsea Internet of Things (IoT) in the future,” Zhang says.

More than 70 percent of Earth’s surface is covered by water, encompassing vast areas where solar cells could potentially be placed. However, water strongly absorbs sunlight with long wavelengths—those greater than 630 nanometers, corresponding roughly to red and beyond to infrared, microwave, and radio wavelengths. This limits conventional solar panels, which can absorb sunlight with wavelengths ranging from roughly 400 to 1,100 nm, or from blue to infrared.

To overcome this challenge, Zhang and his colleagues designed solar cells that could make full use of shorter wavelength sunlight. They fabricated perovskite solar cells that cover the spectrum of light that does filter through 5 to 10 meters of water, which consists mostly of blue-green 400- to 600-nm wavelengths.

The scientists found that while their photovoltaic cells attained a light-to-electricity conversion efficiency of 17.08 percent under the conditions usually found on land, they achieved 34.71 percent efficiency when given the spectrum of light one would expect submerged 10 meters underwater. Most standard land-based solar panels achieve an average efficiency between 20 and 25 percent.

The researchers also integrated their cells into underwater mini-robots equipped with rechargeable lithium-ion batteries, deploying them near Weizhou Island in the South China Sea. They encapsulated these cells in glass, synthetic rubber, and epoxy resin to protect them from seawater. The robots could automatically maintain a fixed depth to help the scientists see how well the cells performed at different distances underwater.

At a depth of 2 meters, 115 square centimeters of these encapsulated cells could generate a modest 1,416 milliwatt-hours over two hours—roughly half the capacity of a rechargeable AA battery; at a depth of 10 meters, these cells could still generate 324 mWh of energy over two hours. The performance of these cells at 10 meters exceeded Zhang’s expectations by about three- to six-fold, he says.

How long can submerged solar cells survive?

Compared to conventional silicon solar panels, perovskite solar cells are cheaper, easier to make, and more efficient, making them a strong choice for underwater applications. But they also often rapidly degrade in performance under intense light and high temperatures. In contrast, the low temperatures and weak light typically found underwater could allow for more durable operation.

The researchers tested this possibility with lab experiments where the new cells were subjected to the kind of light seen underwater at 10 meters. Given the minimal level of degradation the researchers saw at standard operating temperatures of 25 °C—the annual average water temperature near Weizhou Island—they estimated the cells were capable of continuously operating for roughly 5.5 years before degrading to 80 percent of their original efficiency. (In contrast, standard perovskite cells begin to deteriorate after just one year of use on land, although researchers worldwide are investigating ways to significantly extend perovskite cell lifetime.)

“The submerged solar cells show application potential as a real solution for underwater energy,” Zhang says. “Massive deployment of submerged solar cells could extend photovoltaics from land and space into the ocean, addressing the urgent demand for durable offshore energy supply. Ultimately, this could support the growth of the blue economy and may open a new chapter for marine energy.”

Submerged photovoltaics face a number of potential obstacles, Zhang notes. For instance, ocean currents could make it difficult to deploy these devices. Seawater corrosion and encrusting from barnacles and algae could also reduce their long-term performance.

In the future, Zhang would like to see just how deep underwater solar cells can operate. He would also like to integrate them with energy storage devices to create all-in-one units to accelerate practical applications. In addition, standardized testing protocols are needed to analyze submerged photovoltaic performance; these could enable comparisons across different labs to advance the development of these devices, he notes.

The scientists detailed their findings on 11 September in the journal Joule.

Construction underway on a remote island nearly 400 kilometers west of India’s mainland could demonstrate, for the first time, the commercial viability of a vast, untapped renewable resource: the ocean’s thermal layers. Here, on the Lakshadweep Archipelago, surface waters are more than 20 °C warmer than the depths below, and with the right equipment that temperature differential can be exploited to generate electricity.

Known as ocean thermal energy conversion, or OTEC, such power plants can deliver renewable power around the clock. The scale of the opportunity is immense thanks to extensive regions of the world’s oceans with sufficiently large temperature differences between surface and deep waters. In just the waters around the United States, for example, OTEC could satisfy the country’s entire power demand, according to a 2021 report by the U.S. Department of Energy’s National Laboratory of the Rockies.

India’s project will tap 1,000-meter-deep seawater to generate 65 kilowatts of electricity and produce 100,000 liters of potable water per day. Developed by India’s National Institute of Ocean Technology (NIOT), it’s a small project, but a big step toward a commercial breakthrough for OTEC. “What we are doing is scalable in numbers for islands and remote communities. This can be a market strategy,” says Purnima Jalihal, who led the project before retiring last year as head of the energy and freshwater division at NIOT.

The project is part of a fresh wave of enthusiasm behind OTEC. Kilowatt-scale projects have been completed recently in China and the Canary Islands, and more are planned for Hawaii, Taiwan, and Japan. While previous efforts to commercialize OTEC have come up short amidst engineering snafus, efficiency shortfalls, and funding constraints, today’s efforts benefit from advances in key devices and engineering capacity.

Still, far bigger projects are needed before investment in OTEC will take off, says Robert Varley, who led defense contractor Lockheed Martin’s OTEC program from 2006 to 2017. Most OTEC proponents agree, saying the field needs a demonstration of at least a few megawatts, which is an order of magnitude larger than the longest-running projects to date. “Once that happens, I think the world changes for OTEC,” Varley says.

How Ocean Thermal Energy Conversion Works

Earth’s oceans absorb much of the sun’s daily irradiation and excess heat trapped in Earth’s atmosphere. But most of this thermal energy stays within 100 meters of the ocean’s surface. The deep seas remain frigid, dominated by meltwater from Earth’s frozen poles.

Since the 1970s, a few dozen projects have sought to harness that heterogeneity. It’s challenging because the temperature gap the technology aims to exploit is relatively narrow. Even in tropical zones, most OTEC plants will have to make do with a 25 °C difference between the surface waters and the cold waters 1,000 meters below. (By contrast, conventional thermal generators function off of a 400 °C temperature spread.)

To harness the small temperature differential, most groups generate power by heating and cooling a working fluid that naturally boils at a low temperature, such as ammonia. Surface water is pumped into the plant, and its heat is transferred to the ammonia, causing it to evaporate. That vapor then drives a turbine to generate power. Finally, to repeat the cycle, the ammonia’s heat is transferred to cold water piped in from the deep, causing the ammonia to condense.

An offshore, closed-cycle OTEC system requires a heat source and a heat sink. The warm water passes through a heat exchanger (evaporator) that is in contact with ammonia or other working fluid. The working fluid evaporates into a vapor, which expands and drives a turbine. Then the vaporized working fluid enters another heat exchanger (condenser) that is in contact with cold water. This condenses the working fluid back into a liquid, which is then pumped back into the evaporator to complete the cycle. Both the cooled warm water and the warmed cold water are discharged into the ocean, passing through the heat exchangers.NOAA

Squeezing power from a meager temperature delta requires a lot of surface and deep water. A 6.4-MW OTEC plant proposed on Taiwan’s east coast by Taipei-based industrial conglomerate TCC Group Holdings would suck up to 9,500 liters of cold seawater per second from 600 meters below the surface. A 10-MW design Lockheed envisioned for installation in China in 2013 would have processed 40,000 liters of seawater per second, but it was never built.

Pumping on that scale favors placing plants on offshore platforms, which shortens the pipes required since they can extend straight down to reach lower depths. But offshore platforms increase costs, and operating in punishing open ocean conditions increases risk. When filled with water, an offshore OTEC platform’s dangling cold water pipe will weigh tens of thousands of metric tons, and any movement from rough seas will place immense stress on its connection to the platform.

To address one of the risks of being located offshore, London-based startup Global OTEC deployed in the Canary Islands the world’s first platform purpose-built for OTEC and hurricane readiness. The platform, floated in April and paid for by government grants, is a 1:7-scale, non-working model of a 2.5-MW demonstration plant.

Sea Solar Power in Jacobus, Pa., is developing a 25-MW OTEC plant. Sea Solar Power

An operating plant will be required to convince investors of the technology’s reliability. Jacobus, Pennsylvania-based Sea Solar Power, established in 1972 to pursue OTEC, is designing a 25-MW plant that it vows would satisfy investors by paying for itself if its electricity serves an island grid reliant on expensive diesel power. Sea Solar president James H. Anderson Jr. says his firm survived to date thanks to revenues from its manufacturing spinoffs, but he says the company is now seeking a US $5–10 million investment to finish the plant’s “pre-feasibility” engineering design. Building the plant would cost at least US $120 million more.

Islands Seek Renewable Energy from OTEC

India’s project in the Lakshadweep Archipelago aims to show that small OTEC plants can be reliable and cost-effective. The key is using simpler technology and producing more than power.

Rather than adding a cycle that evaporates and condenses a separate working fluid like ammonia, NIOT’s plant will evaporate seawater at low temperature. To achieve that, pumps will create a vacuum, which lowers the atmospheric pressure around a liquid, decreasing its boiling point. This enables the Archipelago’s approximately 29 °C surface water to evaporate more easily. That resulting ‘steam’ will then spin a low-pressure turbine before it is condensed using 7 °C water. Then, instead of returning all of its condensed water back to the ocean, NIOT’s plant will divert some of it to generate fresh water.

NIOT’s technology emerged after a failed bid to go big in 2002. The Chennai-based government lab sought to demonstrate the world’s first 1-MW OTEC process, 40 kilometers off India’s southeast coast. But during construction, workers dropped the project’s 1,000-meter-long cold water pipe, losing it to the depths of the ocean. It was a confidence-shaking accident that Global OTEC founder and CEO Dan Grech calls “the biggest mistake in OTEC history.”

After the incident, NIOT pivoted, adapting its technology to address the needs of Lakshadweep. Like many island communities worldwide, the people of this archipelago rely on pricey imported diesel to generate electricity, and they face mounting water stress due to overdrawn aquifers and rising seas. NIOT initially addressed the drinking water crisis with desalination plants.

These plants use vacuums to evaporate hundreds of liters of warm surface water per second and condense the vapor to freshwater using deep sea cold water. The first plant started up in 2005 in Kavaratti, Lakshadweep’s capital city, ending islanders’ consumption of salty water. Doing so slashed rates of hypertension and gastrointestinal illness. “That was a moment of triumph for NIOT because nobody else had thought of doing this,” says Jalihal, the project’s former leader at NIOT.

But Lakshadweep’s desalination plants still rely on the islands’ expensive diesel-powered grids for electricity. So they’ve turned to OTEC for power. NIOT’s OTEC-enhanced desalination project, now under construction on Kavaratti, will use its low-pressure turbine to generate enough power to operate desalination grid-free.

It’s not technology that will scale up, because the low-pressure turbines require much larger blades. But it could provide a small return on the 500 million rupee (US $5.3-million) installation, saving about 20–30 million rupees (US $210,000–$315,000) per year in annual diesel costs, according to one report.

And the design could be replicated at many of the world’s developing islands. NIOT has discussed sharing the technology with the Maldives and Mauritius. That would be another small evolutionary step for OTEC. But, as Lakshadweep’s experience shows, the local economic and health impact could be impressive.

This article appears in the November 2026 print issue as “Islands Tap the Ocean for Energy.”

Dozens of materials today outperform the world’s most common superconductor—niobium-titanium (NbTi). But no competing material can beat niobium-titanium’s capacity to be manufactured at scale and rolled out into usable wire.

Two startups say that AI and robots could help close that gap.

The San Francisco–based Periodic Labs and Cambridge, Mass.–based Quantum Formatics are searching for materials that work at temperatures higher than 10 kelvins (−264 °C), the transition temperature of NbTi, below which it loses all electrical resistance.

Quantum Formatics has several candidate materials for a new generation of superconductors. Company founder Jason Gibson says the company is about a year away from making a prototype wire that can be used in qualification tests to demonstrate superconductor performance.

“We’re trying to get operating temperatures of 10 to 20 K,” says Gibson. “We focus on moderate-temperature superconductors with excellent mechanical properties that lead to good manufacturability. This approach means we can adopt a standard wire-manufacturing process.”

Periodic Labs cofounder Ekin Doğuş Çubuk says his company, established in September 2025, has developed a technique based on X-ray diffraction. XRD, as it’s called, is a method of mapping the atomic structure of a material by bouncing X-ray beams off it.

Periodic Labs’ XRD procedures help the company to study candidate superconductors and automate its superconductor discovery process, says Çubuk.

Periodic Labs cofounder Ekin Doğuş Çubuk says the company has pioneered an AI-powered robot-arm system that can test 1,000 candidate superconductors per day. Richard Morgenstein/Periodic Labs

How Can Robots and AI Help Discover New Superconductors?

Because no human can sift through thousands of XRD patterns a day, the company is developing machine learning algorithms to use XRD and measure the magnetic properties of materials to speed detection of new superconductors.

“We have built robots that can physically use X-ray diffraction machines,” Çubuk says. “Robotic arms load samples, and then the measurement gets done, sent to an LLM. The robot takes a sample out, puts in a new sample, and then the LLM analyzes all the X-ray diffraction patterns to determine if the experiment worked—and if it didn’t work, what to do next.”

Çubuk adds that developing and creating new materials from scratch hasn’t been easy to automate.

“Our biggest bottleneck is being able to make these different materials for the first time,” says Çubuk, who was previously a research scientist at Google DeepMind.

Meanwhile, according to competing teams also using AI to make progress in the field, new superconductor discoveries are hotly anticipated today.

“Magnets could come very fast if we are very lucky and find something very easily,” says Päivi Törmä, the leading physicist of the Super C consortium of European and American universities aiming to make the first room-temperature superconductors by 2033.

New superconductors can boost energy efficiency in data centers and hospital magnetic-resonance-imaging equipment by an order of magnitude, she says.

“I made estimates based on the best data that I could get,” Törmä says. “For sure, it’s 10 times more, but it can easily be 100 or 1,000 times more energy efficient.”

Periodic Labs has been conducting 100 experiments a day, according to Çubuk. The company is also opening a second lab to accelerate testing.

“That’s going to ramp up to 1,000 attempts a day,” Çubuk says. “I think that will be by far the highest-throughput superconductivity research ever done. We want to see by doing 1,000 good attempts a day if we can discover really exciting superconductors.”

Quantum Formatics is using what it calls the System, its proprietary AI-accelerated superconductor-discovery algorithm. The company has published three articles in Nature Computational Materials describing how its automated discovery system works.

Quantum Formatics founder Jason Gibson says AI can assist in the discovery of hardened superconductors for powering strong magnets for nuclear fusion, since existing magnets turn brittle in a reactor’s high-radiation environment.Minks Media

“We realized there was a critical gap between where the research of superconductor discovery was and what actually gets implemented in a device,” says Gibson, the company’s founder. “Dozens of superconductors have been discovered that vastly exceed the superconducting properties of NbTi. However, nearly all have failed to reach large market penetration due to the challenge of manufacturing.”

Gibson points to the importance of ductility—a material’s capacity to be reshaped and, in the case of Quantum Formatics, made into wires flexible enough to be wound into superconducting magnets.

What Technologies Could Benefit From New Superconductor Discoveries?

For comparison, rare-earth barium copper oxides (ReBCO) are a popular superconductor, with transition temperatures in excess of liquid nitrogen’s 77 K boiling point. (A superconductor that’s coolable with inexpensive liquid nitrogen, as opposed to the more costly liquid hydrogen or liquid helium, is an attractive prospect for superconductor researchers.)

The Cambridge, Mass.–based Commonwealth Fusion Systems uses ReBCO tape in two experimental-fusion reactors the company is building in Massachusetts and Virginia.

However, superconducting magnetic tape in a fusion reactor also is exposed to a substantial amount of radiation, which can degrade the tape.

“Radiation is a very challenging thing,” says Dan Brunner, former chief technology officer at Commonwealth. “That’s one of the challenges with the ReBCO class of superconductors. It’s a very fine crystal structure. Because it’s crystalline, it’s brittle, and that makes it hard to engineer into systems.”

Quantum Formatics is working with companies like Commonwealth Fusion and Realta Fusion, based in Madison, Wis. Superconducting magnets account for up to half of the $3 billion cost of a fusion reactor, Gibson says.

“The one challenge we’re really trying to address in terms of ReBCO is its radiation tolerance,” Gibson adds. “It does not perform well under radiation, but it is essentially the only option that produces high enough magnetic fields to allow you to have a compact fusion plant.”

Çubuk says that Periodic Labs is also investigating commercial applications like power transmission for data centers.

“Data centers are getting more power hungry, and they have to use these extremely thick copper wires to bring power into the racks,” Çubuk said. “If somebody could make superconductor tapes work with liquid nitrogen, those tapes could be much thinner and much smaller and could carry much more power without the resistance loss.”

For the foreseeable future, Çubuk says, old-fashioned trial and error will remain an important part of the discovery process.

“From that perspective,” he says, “it’s very interesting to think about how superconductors get discovered—and how they might get discovered in the future.”

In a display case on the lower level of the Faraday Museum at the Royal Institution in London, there’s an unassuming stack of gray metal discs and blotting paper. It’s not at all obvious that this humble object is the starting point of today’s multibillion-dollar global battery industry. The object’s invention in 1799 grew out of a disagreement that Alessandro Volta—the Italian physicist for whom the unit of measurement for electrical potential is named—had with his friend Luigi Galvani over a dead frog.

The Debate Over Animal Electricity

Galvani was a well-respected Italian physician. In the 1770s, he began investigating the use of electricity to stimulate the muscles of dissected frogs. Armed with an electrostatic generator and an early type of capacitor called a Leyden jar, he was able to create a charge, store it, and then zap his animal specimens at will. He was intrigued when the frog legs twitched as if they were still alive. He spent the last three decades of the 18th century studying the phenomenon, and in 1791, he published De viribus electricitatis in motu musculari commentarius (Commentary on the Effect of Electricity on Muscular Motion).

Luigi Galvani spent decades investigating what he believed to be a natural electric force emanating from animals. Universal History Archive/Getty Images

Galvani saw the frog as embodying an “animal electricity,” an innate vital force that activated nerves and muscles, similar to what had been observed in (living) electric eels and torpedo rays. For Galvani, the frog was an electrical machine analogous to a Leyden jar. The brain was the source of the electrical charge; the nerves conducted the electrical fluid; and the muscles stored opposite charges. The illustrations in his 1791 book are fabulous—frog legs spread all over his laboratory table!

Galvani was wrong in thinking that his frogs were electrical machines, but he was right that the muscle contractions were caused by electric signals.SSPL/Getty Images

At first, Volta, chair of physics at the University of Pavia, concurred with his friend. But after beginning his own experiments, he concluded that Galvani was wrong and that the frog generated no electricity at all. He thought of the frog as nothing more than an electroscope, an instrument to indicate the presence of an electrical charge. Volta posited that the source of the charge Galvani observed came from two different metals in contact with the frog. He termed this “metallic electricity.”

Alessandro Volta came to disagree with Galvani’s theory of animal electricity.Apic/Getty Images

To prove his point, Volta created an “artificial electric organ.” He stacked alternating discs of copper and zinc, separated by cardboard, blotting paper, or cloth soaked in brine or acid. When the top and bottom plates were connected, an electric current flowed through the stack. As opposed to a Leyden jar, which is essentially a capacitor that can store an electric charge and release it in a brief powerful discharge, his stack of discs generated its own electricity through a chemical reaction and delivered a sustained low-current output.

Volta didn’t publicly demonstrate or announce his artificial electric organ until after Galvani died in 1798. But when he finally did, in 1799, it immediately began upending science. Just six weeks after Volta wrote to the Royal Society about his invention, the English scientists William Nicholson and Anthony Carlisle used a voltaic pile to run a current through water to separate it into hydrogen and oxygen. They had discovered chemical electrolysis. Humphry Davy later used a large voltaic pile to isolate a number of elements, including potassium, sodium, calcium, strontium, and barium. Early piles petered out after a few hours. Users who stacked up more metal discs to make more powerful piles found the weight of the discs squeezed out the moisture in the paper or cloth.

Invented in 1799, Volta’s “artificial electric organ” (later known as the voltaic pile) was the first battery. Volta presented this one to Michael Faraday in 1814.Royal Institution of Great Britain/Science Source

One of the most enthusiastic users of the voltaic pile was Galvani’s nephew, Giovanni Aldini, who spent much of his career defending his uncle’s ideas. Aldini created spectacles across Europe in which he used voltaic piles to shock the carcasses of livestock and, occasionally, the bodies of recently executed convicts. Vivid descriptions in the popular press, as well as Aldini’s own writings, raised the question of whether electricity could bring the dead back to life. Mary Shelley provided her answer in her 1818 novel, Frankenstein; or, The Modern Prometheus. In an introduction to an 1831 edition, Shelley cites galvanism as one of her inspirations for the monster’s reanimation process.

Beyond Winners and Losers in Scientific Debates

Scientists and historians share a common trait: They like stories with clear winners and losers. The narrative of competition helps drive a narrative of progress that makes it look like humanity is always moving forward. In the case of Galvani and Volta, Volta is usually depicted as the clear winner in the debate over animal versus metallic electricity. The Encyclopedia Britannica goes as far as to write that “with his announcement of the first electric battery in 1800, victory was assured for Volta.”

But both science and history are more nuanced than that. In fact, Galvani and Volta were both partially right and partially wrong. There was no universal force of animal electricity, but Galvani was correct that electrical signals caused muscle contractions, which he discussed in his anonymous 1794 publication Dell’uso e dell’attività dell’arco conduttore nella contrazione dei muscoli (On the Use and Activity of the Conductive Arch in the Contraction of Muscles). Volta was right to push back on Galvani’s animal electricity theory, but he was wrong that electrophysiological effects require two different types of metal, or any metal at all; the circuit in the voltaic pile was closed by the wet paper or cloth.

It seems a little presumptuous for the Encyclopedia Britannica to declare Volta the winner and Galvani the loser. Volta definitely thought his friend was wrong, but he waited until after Galvani’s death to make his views public. It’s closer to the truth to say they were both genuinely curious to understand the nature of electricity. In the process, they unknowingly helped develop different fields of inquiry: electrophysiology for Galvani and electrochemistry and battery science for Volta.

RELATED: Who Really Invented the Rechargeable Lithium-Ion Battery?

Such an outcome is actually quite common in scientific disagreements. For example, Isaac Newton’s dispute with Christiaan Huygens over the nature of light—did light consist of particles, or corpuscles, as Newton termed them, or waves, as Huygens contested—breaks down today into quantum optics and classical optics. Similarly, Louis Pasteur’s and Justus von Liebig’s debate over fermentation (microorganisms versus chemical decomposition) led to two complementary fields: microbiology and biochemistry.

Maybe instead of looking for winners and losers, we would be better off expanding our horizons and considering the multiple paths of inquiry and discovery. Writing in 1816, toward the end of his career, Volta graciously acknowledged Galvani’s pioneering work, saying “it contains one of the most beautiful and surprising discoveries and the germ of many others.” What new revelations are waiting to develop out of today’s scientific debates?

Part of a continuing series looking at historical artifacts that embrace the boundless potential of technology.

An abridged version of this article appears in the September 2026 print issue as “The First Battery.”

References

On 20 March 1800, a year and three months after the death of Luigi Galvani, Alessandro Volta wrote a letter (in French) to Joseph Banks, president of the Royal Society, describing his invention of an artificial electric organ. It was read before the Society on 26 June and published in Philosophical Transactions on the last day of that year as “On the electricity excited by the mere contact of conducting substances of different kinds.”

The Smithsonian Institution Libraries used their rare books in the online exhibit The Body Electric, which has more information on both Galvani and Aldini.

The website of the Whipple Museum in Cambridge, England, has a number of pages devoted to frogs, including a very informative description of the role frogs played in Galvani’s experiments and how those led to Volta’s work.

In a significant policy twist, on 14 August, Spain’s government decreed a three-year extension for a pair of nuclear power plants with combined capacity of about 2 gigawatts. In 2019, the plants at Almaraz, 200 kilometers southwest of Madrid, had been scheduled to shut down in 2027. A phaseout of nuclear power was a major plank of Spain’s coalition government, so the delay has set off a conflict with one of the coalition parties.

The decree blames the extension on “the crisis in the Middle East,” but the decision comes as multiple factors have slowed or reversed several European countries’ policies aimed at phasing out nuclear power. “The Ukraine war, the Iberian blackout, and rising natural gas prices have changed things a lot,” says energy economist Diego Rodríguez Rodríguez of the Complutense University in Madrid, who is also a former board member of the Spanish National Markets and Competition Commission.

“The government could see that the system was going to need a lot of battery storage to catch up to its booming solar photovoltaic capacity, and they have used the Strait of Hormuz thing as cover.”

Europe’s Nuclear Power Policy Shift

Even before Russia’s 2022 invasion of Ukraine, some European Union countries were reconsidering plans to shut down nuclear power plants or even considering how to add nuclear-generating capacity. The invasion raised household electricity prices as EU nations scrambled for alternatives to Russian natural gas. In 2023, Sweden added nuclear to its energy-mix target alongside renewables. In 2025, Belgium voted to repeal its 2003 nuclear phaseout. In 2026, France canceled plans to shut down more than a dozen reactors, called for new reactors, and raised its targets for nuclear’s role in the national energy mix. Poland is planning to build its first nuclear power plants in 2028.

Spain’s nuclear capacity has hovered at around 20 percent of the national electricity mix since a 2019 phaseout agreement. Nuclear power has contributed a relatively price-stable alternative to natural gas and has also provided reliable base load to a grid that incorporates a growing share of time-variable renewable energy sources.

Spain’s long-stated goal has been to use renewables such as wind and solar to replace its most carbon-intensive energy sources; to get there, the government has provided a mixture of incentives. Since 2022, Spain has added more than 20 GW of photovoltaic capacity. As of 2024, wind was contributing about 23 percent, and solar 17 percent of the country’s electricity mix.

All of that added renewable energy has created significant technological challenges. Grid operators must balance reactive power (essentially stored power that is shifted back and forth through the grid) with active power (the power drawn by users) to maintain a steady electricity supply. Outdated grid-management rules exempted solar power plants from the same reactive power control contributions required of legacy plants, potentially exposing the grid to instability.

On 28 April 2025, a series of power oscillations, voltage gaps, and reactive power control failures set off a cascade of disconnections across the Iberian Peninsula. The resulting blackout affected more than 60 million people for about 12 hours. While Spain’s political parties debated how much of the blame belonged to the grid operator or renewables or other factors, the coalition government, with its tiny majority, could not afford another similar hit to the grid. Nuclear power offered a way to increase stability while the country shored up the resilience of its electric grid.

The Almaraz power plant is in Navalmoral de la Mata, in western Spain. Carlos Criado/Europa Press/Getty Images

Spain Buys Time for a Grid Makeover

For now, Spain’s grid lacks sufficient storage and transmission capacity to absorb new PV as quickly as it is being added. Instead, many installers are going behind the meter, building almost one-fifth of the country’s solar PV at industrial sites that can use it themselves, such as at data centers.

But upgrades are coming. In July 2025, Spain’s Ministry for Ecological Transition and Demographic Challenge (MITECO) announced a EUR 750 million (US $874 million) investment in 65 new resilience measures, including synchronous condensers, flexible AC transmission systems, substation improvements, and switching relays. The country is also adding three more interconnections to France to meet European recommendations.

Investment in energy storage is also booming: At the time of the blackout, Spain had only 28 megawatts in battery storage capacity, but that number almost sextupled to 193 MW by April 2026. Spain is funding 2.2 GW in near-term large-scale energy storage projects and has a national target of 22.5 GW (including pumped hydrogen and other nonbattery systems) by the time of the rescheduled shutdown of its Almaraz nuclear plants in 2030.

Slowing down the nuclear shutdowns buys Spain’s embattled electric grid affordable base power with stable voltage and avoids using more price-volatile, expensive natural gas. The Almaraz extension raises the question of which other nuclear plant shutdowns the government might delay, given that several were scheduled to end operations around 2030 as well. “The national consortium for dismantling the plants wouldn’t even be able to dismantle them all at once anyway,” Rodríguez says.

Critics of Spain’s energy-policy pivot have argued that extending nuclear power will unfairly cut into the profits of investors in solar, who will have to curtail their power generation more at peak times. Natalia Fabra, an energy economist at CEMFI, in Madrid, wrote this spring in the Journal of the Spanish Economic Association that “keeping nuclear plants online longer may lower prices and emissions at first, but it can also dampen clean investment incentives and lead to higher prices and emissions later on.”

There is no serious talk of building more nuclear power in Spain, Rodríguez says. He regards the delay more as a realistic move to better amortize the seven existing Spanish nuclear plants while the country invests more in storage and grid resilience. The 14 August decree itself insists that “the requested extension does not compromise the contribution of Spain to the renewables objective.”

There are 17 rare earth elements. While almost all of them have vital industrial uses, you could be forgiven for believing, based on news coverage, that there are only four that matter: neodymium, praseodymium, dysprosium, and terbium.

To be sure, those are the essential ingredients in the powerful permanent-magnet motors that go into electric vehicles, heating and cooling systems, appliances, tools, fighter jets, bombs, and many other systems. But the industrial and national-security importance of rare earth elements goes far beyond magnets. Among the 13 other rare earths are two—erbium and yttirum—whose importance is no less vital, and which are now facing a looming geopolitical deadline this November. Depending on how negotiations play out, the post-deadline framework could erode U.S. national security—or redraw U.S.–China trade relations.

Erbium is used in the amplifiers that boost optical signals in the fiber-optic cables that crisscross the planet. Yttrium is a key component of the thermal barrier coatings that protect turbine blades and other structures from blistering heat in jet engines and in the combustion turbines so in demand to power new data centers.

Yttrium is more broadly used than erbium; besides thermal barrier coatings, it is also used in white LED lights and displays, microwave filters, and solid-state lasers, among other applications. Also, “small quantities of yttrium feed into trillions of multilayer ceramic capacitors that keep our electronics world going, including AI data centers,” notes Thomas Kruemmer, a rare-earths consultant based in Singapore.

Erbium and yttrium, Kruemmer writes in an email, are good examples of how “tiny quantities of rare earths can have an outsized impact by enabling core functionality in their respective applications.” Taken together, all of these rare earths “literally affect almost anything we can switch on and off.”

Chinese dominance of the markets for these two rare earths is essentially 100 percent. Take erbium: “There is no one that I’m aware of, outside of China, that makes erbium oxide on a commercial scale,” says Eric Bender, vice president of strategy and corporate development at Tronox, a mining, processing, and chemicals company headquartered in Stamford, Conn. Erbium oxide is the molecular form that is typically produced by separation facilities and bought by industrial users of the element.

For yttrium, too, industrialized economies are entirely dependent on China. China mines 90 percent of the world’s yttrium, but it processes essentially all of the world’s mined ore into an industrially usable form, such as yttrium oxide.

Yttrium coatings help turbine blades in generators and jet engines survive the heat. Sergii Zhmurchak/iStock

China’s Monopoly and the Yttrium-Erbium Supply Squeeze

Twelve rare earth elements, including erbium and yttrium, are subject to global export restrictions instituted by China in April and October 2025. The restrictions for some of them, including erbium but not yttrium, were suspended for one year in November 2025. Exports of yttrium, and a few other rare earths, have become erratic and sharply reduced. In March 2026, China approved a single, 60-tonne shipment of yttrium oxide to the United States, followed by a 10-tonne shipment in April. But overall, shipments of yttrium to the U.S. are down around 75 percent this year in comparison with 2025.

In November, China is expected to either extend or remove the suspensions for some or all of the 12 rare earths currently under export controls. And yet there is little movement apparent on the part of U.S. and European governments to cope with the possible cutoff of yttrium and erbium supplies. “Very few folks outside of China are talking yet about ‘How do we produce erbium, how do we produce yttrium?’” says Gareth Hatch, managing director of Strategic Minerals Advisory. “We’re lagging behind the geopolitics, the realities, of the day.”

The most important thermal barrier coating is yttria-stabilized zirconia. GE Vernova is one of the world’s largest manufacturers of turbines and a major user of the compound. At an investor day meeting last December, GE Vernova’s CEO, Scott Strazik, said the company had enough yttrium to last into 2026, but did not specify how far into 2026. An emailed request for elaboration, sent to GE Vernova’s media organization, did not elicit a response by press time.

Nevertheless, there are clear signs that the reduced availability of yttrium is having an effect on the industry. Oerlikon Metco is one of the largest of the small group of companies that produce yttria-stabilized zirconia and the equipment that applies it to the blades and other parts of jet-engine and combustion turbines. The erratic availability of yttrium has “thrown us upside down in the last year or less,” says Riston Rocchio-Heller, an engineer and lab manager at the company. Demand for combustion turbines in the United States is now at an all-time high, driven mainly by the need to power new data centers, including huge “hyperscale” facilities for AI training and inference and general cloud computing.

Yttria-stabilized zirconia is a ceramic produced by mixing yttrium oxide with zirconium dioxide. Alternatives to it exist, but some unique advantages of the yttrium-containing version have made it the standard choice, according to Roccio-Heller. Researchers are trying to find better alternatives, he says: “A lot of collaborative work is going on right now,” he reports. But it’s unlikely to come to fruition soon.

The erbium market is dominated by just a couple of applications: the coloring of consumer glass and erbium-doped fiber amplifiers. These amplifiers are spaced along long-haul fiber-optic lines, typically every 80 to 120 kilometers on a terrestrial cable, and every 50 to 80 kilometers for a submarine cable.

The amplifiers consist of a piece of erbium-doped optical fiber, usually between 10 and 30 meters long. The doped fiber is illuminated, or “pumped,” with laser light at 980 or 1,480 nanometers. That laser light excites the erbium atoms into a higher energy state. When a weak optical signal enters the length of fiber, it triggers the excited erbium atoms, which fall to lower energy states and release photons in the exact phase, direction, and wavelength—1,550 nm—of the incoming signal. That wavelength corresponds to the spectral “window” in which silica optical-glass fibers cause the least loss of the signal. That’s why nothing but erbium will do here: It naturally re-radiates at the necessary wavelength.

Engineers are starting to use erbium-doped amplifiers in the free-space optical communication systems that connect low-Earth-orbit (LEO) communications satellites, such as the Starlink satellites, with each other and with ground stations. Groups of these LEO satellites are being launched into orbit on an almost daily basis. Because their signals don’t go through glass, satellites could use a much wider variety of wavelengths. But engineers are using the erbium amplifiers anyway to take advantage of their relatively low cost and the large existing industrial base for the 1550-nm hardware, says Michel Corriveau, a researcher at MPB Communications, in Pointe Claire, Quebec, Canada, a leading provider of erbium-doped fiber amplifiers.

Corriveau says MPB obtains erbium-doped fiber for its amplifiers from a couple of companies, including Lightera (formerly OFS) in Denmark, which is now a business unit of Japan-based Furukawa Electric. An executive at Lightera, Annette Lundby, declined to comment on the company’s ability to secure stocks of erbium.

Project Vault and a geopolitical deadline

Erbium and yttrium, along with 14 other rare earths and a couple of dozen critical minerals, are stockpiled by the U.S. government. This National Defense Stockpile is managed by the Defense Logistics Agency, which does not comment on its specific contents. In February 2026, the Trump administration announced a US $12 billion public-private initiative called Project Vault to stockpile rare earths and other critical materials. The funding includes $10 billion in financing from the Export-Import Bank of the United States—the largest single financing in the bank’s history.

Curtis Stough, a “defense stockpile liaison” at the Defense Logistics Agency, did not respond to a request for an interview. The U.S. Geological Survey, which tracks availability of critical minerals, also declined to make anyone available for an interview. The White House press office, too, did not make anyone available. But spokesman Kush Desai did provide a statement, which read, in part: “The Administration continues to take a multifaceted and nimble approach to reshore critical mineral supply chains back to the United States.”

That approach will surely be tested in the event that China’s 2025 export restrictions take full effect, which could happen on 10 November. But a lot will probably happen before then, says Hatch, making the outcome hard to predict. “This whole thing is ostensibly about dual-use control of materials” that have essential military uses. “But in reality, it’s part of the bigger trade picture.” So Hatch expects a wider discussion, leading up to November, involving leading-edge GPU chips, extreme-ultraviolet lithographic equipment, high-bandwidth memory, design-automation software, and other tech goodies that the United States and its allies are currently denying China.

In a tranquil Boston suburb, on the far edge of a horse farm, where pasture gives way to woods, a crane lowers an enormous electrode into a borehole. The electrode, a half-meter-long cylinder with copper-tipped arms to ensure good contact with the borehole walls, descends—deeper, deeper—through layers of spongy sandstone to the hard, marbled roots of an ancient mountain range hundreds of meters below ground. Here the rock is tight; there are few cracks for water or gases to flow. But that’s about to change.

A stone’s throw away, a second electrode—a twin of the first—has been fixed in another borehole at the same depth. From above ground, a pair of high-voltage generators cabled to the two electrodes fires a series of pulses.

Tsss!…Tsss!…Tsss!…Tsss!…Tsss!….

Each discharge, heard faintly at the surface, is like a miniature, subterranean lightning strike. The rock between the electrodes heats. Pressure builds. Then, suddenly, the rock splits into a spiderweb of fractures.

On a horse farm outside of Boston, a worker sets up the well where Eden’s electrode will be lowered with a winch.Bob O’Connor

Eden GeoPower, the Massachusetts-based startup performing this peculiar field test, calls the technology electrical reservoir stimulation. The company’s tagline: “We break rocks with electricity.”

Eden’s researchers hope their rock-breaking technique will someday aid mineral mining, tap geothermal heat, or create geologic storage areas for carbon. But there’s an even more intriguing use that could create a whole new category of energy production: generating hydrogen underground.

The dream of a hydrogen-powered economy dates back to the 1970s, when petroleum shortages and rising concerns about pollution from fossil fuels sparked visions of cars, ships, planes, and industrial machines running on hydrogen instead of carbon. Hydrogen is often touted as a clean fuel because when it’s burned or consumed in fuel cells, it emits only water and heat. However, it currently takes more energy to make than it yields, and the cheapest and most common way is by reacting steam with methane, a potent greenhouse gas.

How to Break Rocks With Electricity

It’s possible to make zero-carbon hydrogen by splitting water with electrolyzers powered by renewable energy. But in most cases, the process is too expensive to be economical—a reality that burst the hydrogen-hype bubble in the early 2020s. Global demand for hydrogen in 2024 reached approximately 100 million tonnes, containing energy equal to only about 3 percent of the world’s annual energy consumption. Most of it is used as chemical feedstock for petroleum refining and for making fertilizers and plastics.

The frustrations of manufacturing clean hydrogen have convinced many entrepreneurs and scientists to instead seek the element underground. For the past half-decade, dozens of companies around the world have been hunting for buried stores of hydrogen, called natural or geologic hydrogen. But with a commercial-scale operation yet to be proved, Eden and a handful of other startups and research groups are chasing the more audacious scheme of producing geologic hydrogen artificially.

This approach, known as stimulated geologic hydrogen or engineered hydrogen, turns subterranean rock formations into giant hydrogen factories. It typically involves injecting water into iron-rich rock, which oxidizes the iron and releases hydrogen as a by-product. Fracturing the rock, as Eden is doing, creates a network of conduits for the water to reach iron-bearing minerals.

The concept of stimulated hydrogen is so new that few have had a chance to test it. Proponents say that if it works—which is a big “if”—it could provide almost unlimited energy for the indefinite future. There’s one way to find out: Start breaking rocks.

There’s Plenty of Underground Hydrogen

Hydrogen is the simplest and most abundant element in the universe, the stuff of stars and galaxies. Geologists have long known that Earth generates hydrogen gas through natural water-rock reactions, but until recently, the occurrence was regarded as a curiosity. The gas is so light that most experts assumed it all escaped through pores and cracks in Earth’s subsurface and didn’t accumulate in useful quantities.

During a demonstration at Eden’s testing site near Boston, an employee displays a central component of the company’s proprietary electrode. Bob O’Connor

Inklings that they were wrong emerged in the 19th and 20th centuries, when researchers in the former Russian Empire and Soviet Union reported hydrogen seeping from mines and wells. But in the ongoing frenzy for fossil fuels, these observations were largely overlooked or forgotten. Scientists later discovered hydrogen spewing from hydrothermal vents in the seafloor and feeding so-called eternal flames, like those of Türkiye’s Mount Chimaera, where ancient athletes lit torches for the first Olympic games.

Then, in 1987, in the village of Bourakébougou, Mali, people drilling a water well noticed a breeze blowing out of the hole. According to local lore, a worker leaned in for a closer look, a lit cigarette dangling from his mouth. The air instantly ignited, burning a brilliant blue.

The crew capped the well, which stayed sealed for 25 years until, in 2012, a Malian oil and gas prospector confirmed the ground contained a large reservoir of hydrogen. The prospecting company, now called Hydroma, had a small electrical plant constructed to convert the gas into power for the village’s residents. Soon after, startups in Australia, Canada, the United States, and elsewhere began searching for more hydrogen stores. By 2025, large multinational petroleum and mining companies were getting in on the game.

To date, hundreds of exploratory wells have been drilled across the globe. But although researchers have documented widespread hydrogen deposits, none have proved capable of producing the gas at rates and quantities needed for commercialization. “We’ve poked a lot of holes, and nobody has found the gusher—or at least they’re not talking about it,” says Douglas Wicks, a former program director at the United States’ Advanced Research Projects Agency—Energy who now advises companies pursuing geologic hydrogen.

A wellhead guides multiple lines downhole: fluid hose, electric cables, rope, control for a sealing device, and sensor communication. Bob O’Connor

Wicks says that in 2022, while at ARPA-E, he got “dragged into the rabbit hole of geologic hydrogen” by Emily Yedinak, then a Fellow at the agency, who was trying to convince her colleagues to take it seriously. “I was the ultimate doubter,” Wicks says. The astronomical price of electrolyzers had made him skeptical that clean hydrogen was a viable pursuit. Plus, if Earth really did contain vast pools of hydrogen, then surely humanity, which had been digging for natural resources for thousands of years, would have found them by now, he reasoned.

But after talking with geologists—who pointed out that people historically hadn’t found hydrogen because they hadn’t been looking for it—Wicks changed his tune. “I got the epiphany that geologic hydrogen is not just an accumulation; it’s a chemical reaction,” he says. “And if it’s a chemical reaction, then it can be stimulated.”

Finding large accumulations of geologic hydrogen entails stumbling on a Goldilocks set of conditions. You need iron-rich source rocks that have already produced or are producing bountiful hydrogen. You also need porous reservoir rocks that can hold sizable quantities of gas migrating from the source rocks. And you need solid cap rocks above the reservoir that trap the gas underground.

To stimulate hydrogen, however, you don’t need this just-right geology. All you need are iron-rich rocks, and then you can generate the hydrogen yourself.

“These rocks are everywhere,” Wicks says. “If you look at the amount of iron that’s within drilling range of Earth’s crust, you’re talking about quadrillions of tons of hydrogen being accessible. If we’re 1 percent successful just in the United States, we could power the economy for thousands of years.” A back-of-the-envelope calculation convinced him that the cost of stimulated geologic hydrogen could easily compete with hydrogen made from methane. “If we get the technology right,” he concludes, “this could be huge.”

Wicks wasn’t the first person to propose the idea, but he was the first to allocate major funding. In 2024, under his leadership, ARPA-E awarded US $20 million to 16 teams aiming to advance stimulation technologies and research. Winning ideas included fracturing rocks with fluid pressure or mechanical stimuli, exposing them to catalysts to speed hydrogen-generating reactions, and manipulating native microbial communities to enhance production. Eden’s rock-breaking project, the lone electricity-based approach, received $900,000.

Eden GeoPower’s Underground Rock Fracturing

Paris Smalls, Eden’s CEO, founded the company in 2017 as a 23-year-old graduate student at MIT. For his Ph.D. in civil and environmental engineering, he was studying the effects of electricity on rock strength and became interested in enhanced geothermal systems, which require fracturing hot, dry rocks to circulate water through them for extracting heat. This is typically done by hydraulic fracturing, or fracking—a technique borrowed from the oil-and-gas industry that involves injecting high-pressure fluids.

Fracking is controversial because it can cause earthquakes and groundwater contamination, and many regions have banned the practice. From an engineering perspective, it’s also imprecise. The fractures it forms are large and difficult to control. “You can’t get enough fractures where you want because the water ends up just going through the same cracks,” Smalls explains. Electricity, he knew from his Ph.D. work, could create more extensive and finely tuned fracture networks, enabling geothermal systems to produce more heat with less environmental risk.

To determine how permeable its fracture networks are, Eden measures fluid pressure downhole and flow rates at the surface. Bob O’Connor

Smalls immediately grasped that the same rock-breaking strategy could be used for mineral mining, carbon sequestration, and extending the life of oil and gas wells. But he hadn’t considered using it to make hydrogen. So when Wicks invited him to apply for the hydrogen program at ARPA-E, he was confused. “I didn’t get it at all,” Smalls says. “I’m like, ‘I break rocks. How am I going to generate hydrogen?’”

Not long after, Smalls met Alexis Templeton, a geomicrobiologist at the University of Colorado Boulder who had become an expert in geologic hydrogen by studying microbes that consume the gas and the mineralogical transformations that create it. “There was a lot of early interest in whether or not you could engineer the production of hydrogen from rocks,” Templeton recalls. “And the rocks with some of the best potential have all the right chemistry, but they need water. Nobody was excited to do hydraulic fracturing. So everyone was wondering, ‘Well, how are we going to get the water in?’”

Eden’s technology, Templeton understood, could be the answer. She agreed to join the company part-time as its lead geochemist, a position she held from 2023 to 2025. During that time, Eden ran its first pilot experiment, in an oil field in Oman, near where Templeton was already doing her own hydrogen research. The initial setup used DC power to send a steady flow of tens of kilowatts between electrodes in two wells. When Smalls’s team tested it in a petroleum reservoir made of soft, chalky carbonate, the rock fractured readily, increasing oil production by 30 percent.

But when they did the same test in hard rocks, like those needed for hydrogen and geothermal systems, they didn’t fracture much at all. So the team went back to the drawing board and came up with a fix: pulsed power.

Using Pulsed Power for Rock Fracturing

The idea of breaking things using pulsed power—short, concentrated bursts of electrical energy—originated with a mid-20th-century experiment in Soviet-era Russia. As the story goes, a physicist and inventor named Lev Yutkin was out in a thunderstorm when he saw lightning strike a log underwater. Rather than burn, as it would in air, the log exploded, as if blown up by dynamite. Intrigued, Yutkin tried to reproduce the spectacle in his lab. He placed a dinner plate in a water tank, dipped in two wire electrodes, and released a high-voltage pulse. The ensuing spark, he discovered, instantly ionized the water molecules between the electrodes into a plasma channel, which then rapidly expanded, creating a shock wave that shattered the plate.

Yutkin described the phenomenon in his 1955 book Electrohydraulic Effect. He later proposed numerous fanciful uses for it, such as cleaning pipes or breaking up kidney stones, which inspired real tools in use today, including electrohydraulic drills and rock-crushers, and a kidney-stone-busting medical device called a lithotripter. The following decades saw advances in pulsed-power systems and experimental techniques to better understand the complex physical processes involved. By the 2020s, when Smalls’s team began investigating it for subterranean rock fracturing, the technology seemed ripe for use, although that particular application had been little explored outside the laboratory.

“We essentially generate a plasma channel in the rock itself,” says Rafael Villamor-Lora, vice president of R&D at Eden. “This channel then expands very, very rapidly,” fracturing the rock with a shock wave. Bob O’Connor

Eden’s scientists first experimented with pulsed power on thumb-size hard-rock cylinders. Instead of submerging each sample in water, however, they placed a pair of electrodes at opposite ends of the cylinder and delivered pulses directly to the rock. Using this dry-pulse method, drawn from Smalls’s and others’ research, the team found they could form plasma in tiny, moist pockets between mineral grains. “We essentially generate a plasma channel in the rock itself,” explains Rafael Villamor-Lora, Eden’s vice president of research and development. With enough pulses, the fast-swelling channel, as in Yutkin’s investigation, induces a shock wave that fractures the rock.

To bring the technology to the field, Eden needed voltage high enough to break through meters of solid rock. The obvious solution was a Marx generator, which converts low-voltage DC power into high-voltage bursts by slowly charging multiple capacitors in parallel and then rapidly discharging them in series such that their voltages add up to produce one massive pulse. (Marx generators are commonly used in high-energy physics experiments and to simulate lightning strikes on power lines.) Eden custom-built two devices—named Zeus and Thor after the gods of thunder—which together can release a surge of several hundred kilovolts.

This time, the plan worked. In 2025, in an abandoned gold-and-silver mine in Colorado, Eden used Thor to successfully fracture a hard, igneous column, increasing its permeability tenfold.

Ezra Frank, a mechanical engineer at Eden, works on Zeus, Eden’s custom Marx generator. Bob O’Connor

In March this year, the company began setting up the test site on the Massachusetts horse farm to refine its systems and gather more data on how the technology performs in different geologic environments. Its engineers are also designing more powerful generators to discharge stronger and faster pulses. Because Zeus and Thor consume very little power—akin to running a toaster or two—it takes about a minute to store enough energy to fire a maximal pulse. It then takes around 100 pulses to penetrate around 10 meters of hard rock. So fracturing over longer distances or at multiple depths can take hours to days. That means Eden’s biggest cost is labor, not energy.

Smalls says Eden signed an agreement with a geologic hydrogen startup—he declined to say which one—to demonstrate electrical fracturing in a field pilot of stimulated hydrogen, which could begin late next year. Eden will need to prove its technology can help coax the gas from the ground at a profitable rate and cost.

“It’s no question whether we can produce hydrogen,” Villamor-Lora says. “The question is whether we can produce it fast enough to be economical.” In the lab, Eden researchers found they could generate up to four times more hydrogen from rock samples using the pulsed-power technique, compared with the amount found in unfractured samples. But that may not be enough to make stimulated hydrogen commercially viable without some additional technology.

Other Approaches to Stimulated Geologic Hydrogen

One of the biggest challenges in stimulating hydrogen is that there’s no obvious go-to recipe. Beyond the basic ingredients of water and iron, many factors affect how much hydrogen is generated and for how long, and fractures are only one factor. Laboratory studies have shown, for example, that the ideal temperature for maximizing hydrogen production is around 200 to 300 °C. Acidity, rock and water chemistry, and microbial inhabitants are other important considerations.

Making the puzzle more complex, each rock formation is different and may require different stimulation techniques or a combination of them. “There isn’t a single solution that will work everywhere,” says Alexei Tcherniak, CEO of the hydrogen startup GeoKiln. “You have to know the geology you’re operating in.”

Some promising rock formations, he points out, may already be fractured or porous enough to become saturated with water but too cool to make ample hydrogen naturally. To solve this problem, his company, based in Houston, uses a system of underground heaters originally developed for improving flow in heavy oil reservoirs and converting solid organic matter in young shale rock into extractable oil and gas. The heaters, which are commercially available, can be installed in boreholes drilled into hydrogen source rocks, similar to Eden’s electrodes. Tcherniak says that GeoKiln is ready to start field testing as soon as it can raise the capital.

Other researchers are exploring the use of catalysts—metal or chemical salts that speed hydrogen-generating reactions—which, they say, could replace or complement fracturing or heating to increase hydrogen production at less cost. Vema Hydrogen, for instance, is betting on a mixture of boiler-heated water and proprietary catalysts. “What I can say about our catalysts is basically what they are not, which is not toxic, not expensive, and not dangerous,” says Florian Osselin, Vema’s chief science officer. The company, also headquartered in Houston, has begun drilling pilot wells in Canada to test its mysterious brew. By injecting it into semi-permeable rock, Vema expects to achieve commercial production rates without fracturing. “We’ve done field-scale numerical simulations that give us a lot of confidence,” Osselin says.

Another stimulation method, proposed by the Denver-based startup Koloma, aims to expose more rock surface for generating hydrogen by mimicking natural weathering. The technique involves adding carbon dioxide to water and injecting the fluid at specific times to control for factors like acidity and gas concentrations. The carbon dioxide reacts with the water to form an acid that breaks down mineral chains in rock pores, thereby increasing the pores’ surface area, explains Tom Darrah, the company’s CTO, who studied and patented the method as a professor at Ohio State University. “I call it micro-pitting because the texture goes from smooth to rough,” he says. As with fracturing, more surface area means more hydrogen production—if you can get the formula right.

Rita Esuru Okoroafor, an energy resources engineer at Texas A&M University, is studying the effects of various stimulation approaches, including fracturing, catalysts, and carbon-dioxide injection, on hydrogen generation. Her data, based on laboratory tests of rock samples from around the world and numerical models of stimulated geologic hydrogen systems, suggest that none of these approaches alone will sustain hydrogen production at rates needed for long-term commercial development. “We’re still fine-tuning our models, but they’re telling us that we’re going to need a lot of fracturing, we’re going to need catalysts, and then we’re going to need restimulation,” she says.

The process of generating hydrogen, Okoroafor explains, will eventually consume all the readily available iron in exposed rock surfaces, causing production to plummet. By accelerating hydrogen generation, catalysts also accelerate its decline. “When these reactions happen very fast, they also die very fast,” she says. They also leave behind mineral precipitates that can clog existing cracks. In a recent study, she found that hydrochloric acid helps clear the debris, expose fresh rock surfaces, and reopen water pathways to restore production.

It’s too early to know which technologies will win out in the race for geologic hydrogen and if stimulation will even be needed to make it a viable industry. What’s more, production is just the first step toward commercialization. Many questions remain. Once hydrogen is flowing from the ground, how will the gas be purified? How will it be stored and transported? How will the industry be regulated? What are the environmental risks, and how will they be mitigated? What will be the cost?

“With all these wars and gas prices going up, we need to be preparing for the future,” Smalls says. But as is often the case with nascent technology development, life gets in the way. At the horse farm, fracturing started in June after being delayed for months, first by a snowstorm and then minor equipment failures and other logistical snags. “Everything takes longer than you think,” Smalls says. Still, he’s unfazed, ever the optimist. “I like to go after things that other people are afraid to.”

This article appears in the September 2026 print issue as “How To Get Hydrogen From a Stone.”

This article was corrected on 22 September. In the original version, we did not accurately describe a Marx generator. It converts low-voltage DC power into high-voltage bursts by slowly charging multiple capacitors in parallel and then rapidly discharging them in series such that their voltages add up to produce one massive pulse. The article has been updated with this definition.

New York isn’t particularly inundated with new data centers, but it just became the first U.S. state to ban them. In an executive order on 14 July, Governor Kathy Hochul imposed a moratorium of up to a year on data centers that consume 50 megawatts of power or more.

The ban arose from concerns about data centers’ power and water use and effects on residential electricity bills. Such worries have risen sharply across the U.S. over the last year as more developers propose projects and request grid connections. Data-center energy use in the country is expected to exceed 650 billion kilowatt-hours by 2050, doubling to quadrupling today’s consumption.

Virginia, Texas, and California have received an onslaught of data centers, sparking public backlash and sending grid operators scrambling to keep up. But in New York that’s not the case. Electricity margins in the state’s grid are running thin and power bills are rising, but data centers aren’t to blame—for now.

“I don’t think we’ve actually seen those fights in New York,” says Daniel Zarrilli, who most recently served as the chief climate and sustainability officer at Columbia University and was previously New York City’s chief resilience officer and chief climate-policy advisor. “We’re seeing the anticipation of those fights, which is what I imagine led to the moratorium.”

So the governor’s move is preemptive—and it alone won’t fix the state’s energy woes. “Anytime you open the paper, figuratively speaking, these days, there’s all this discussion about data centers,” says Kevin Lanahan, senior vice president of external affairs and corporate communications at New York Independent System Operator (NYISO), which runs the state’s power grid. “But the increase in demand that we have experienced and been tracking really has not been attributable to data centers yet.”

Rather, Lanahan says, New York’s razor-thin electricity margins are mostly a result of the increasing electrification of the state’s building and transportation sectors. That is only expected to continue, and Lanahan says it will drive forecasted grid strain alongside new manufacturing facilities in NYISO’s territory.

“We have some modern manufacturing and some traditional manufacturing in the queue that promises to increase the demand going forward,” Lanahan says. For example, he cites a Micron semiconductor plant under construction near Syracuse that could consume over 1 gigawatt when complete.

Making matters more difficult, NYISO’s aging grid infrastructure isn’t up to the task of meeting new demand. “What we need is new dispatchable generation,” he says. “Hard stop.”

Why New York’s Power Grid Is Vulnerable

New York’s move is an indicator that the U.S.’s burgeoning anti–data center movement has gained traction. Bans on the power-hungry facilities have proliferated over the past year at the county and municipal levels, but this marked the first statewide prohibition. Texas on Monday followed suit in the form of a directive by Governor Greg Abbott ordering the state’s grid operator to pause new data-center connections until it completes an audit of proposed facilities.

The New York ban comes just a month after NYISO announced that New York’s grid wouldn’t have much headroom this summer. Only 417 MW of extra power is available during typical peak demand, NYISO estimates—the lowest summer reserve margin since 2017. Additional pressure on the grid, such as from extreme temperatures, could make margins even thinner and force the use of higher-polluting backup plants.

Since 2019, 2.9 GW of power generation have been added to New York’s grid, but 4.4 GW have come offline, according to a NYISO report published in June. The retired facilities were largely fossil-fuel powered and have been taken out of the system because of factors such as mechanical issues and performance decline.

As a result of the mismatch, summer reserve margins have decreased by 1,810 MW since 2019. “We have what we assess to be the oldest-generation fleet in the country,” Lanahan says. During heat waves, like one that occurred over the July 4 weekend, NYISO must turn on some of those old facilities. “We’re relying routinely…on old steam, peaking units that were built sometimes 70 years ago,” he says.

Meanwhile, the requests for new connections to the grid keep coming. In total, about 14 GW worth of large electricity load await review in NYISO’s grid interconnection queue. Twelve of those gigawatts come from data centers and crypto-mining facilities spread across 40 projects. All but two exceed the moratorium’s 50 MW threshold. NYISO will need to study the requests to see how they’ll impact its system before allowing them to connect. Lanahan says the agency is planning for only about 3 GW to come to fruition within the next 10 years, because projects routinely withdraw from the process or are subject to delays.

“Together we’ve got these two bookends: the increased poor performance of the system because we’re relying on aging infrastructure, and then that future challenge to meet,” Lanahan says.

Solutions such as demand response to manage the increased load aren’t enough, he says. Some facilities could build their own power generation—something NYISO is developing rules around in accordance with a Federal Energy Regulatory Commission directive issued earlier this summer.

IEEE Spectrum reached out to Hochul’s office to determine what the state government is doing to address New York’s shrinking reserves and aging fleet. Senior energy and environment communications advisor Ken Lovett says that the state is prioritizing renewable energy such as wind and solar, plus a nuclear reliability initiative that will add a backbone of 5 GW to the system. “Governor Hochul will never sacrifice the reliability of the grid,” Lovett says.

Large manufacturing facilities, such as a Micron semiconductor plant currently under construction in upstate New York, are a major factor driving future electricity demand in the state.Mike Groll/Office of Governor Kathy Hochul

New York’s Data Center Ban Buys Time

So New York’s ban on data centers is addressing only part of the problem, and a largely future one at that. It will have no effect on NYISO’s interconnection queue, Lanahan says, as the executive order only bans environmental permitting of data centers, not their grid connection. “Projects are still there,” he says. Project developers are “still working with us on the exchange of information that’s requisite to complete what’s called a system impact study. That’s the responsibility we have in that process.”

New York’s moratorium is set to lift when the state’s Department of Public Service completes an assessment of the impact of data centers. Governor Hochul called for using the time, which could last up to a year, to build a regulatory framework that protects the state’s electricity supply, utility bills, and the environment.

Zarrilli, formerly at Columbia University, says that it’s “hard to imagine” the moratorium being extended beyond a year. The governor “continues to support AI and the data centers that drive it,” Lovett says.

Asked about the usefulness of the moratorium, Zarrilli says he’s “hopeful” that the pause can give the state time to plan for the future grid issues posed by data centers. Still, he says there’s the possibility that a year is simply not enough time.

“There’s a scenario for sure where the state takes this time and comes up with a thoughtful path forward for managing the growth and strengthening the grid, and doing it in a way that’s affordable for New Yorkers,” he says. “And then there’s clearly a possible scenario where the time runs out and we haven’t done those things, and then it’s a scramble to figure out what to do next. That latter option would be tragic.”

Today’s U.S. electrical grid, among the largest, most complex systems ever built, is operating at its limit. The combination of rapid industrial growth, more frequent extreme weather, and a record surge in electricity use has pushed the grid to its breaking point, according to the U.S. Department of Energy.

Built decades ago for a more predictable world in which power came mostly from centralized coal or gas plants and electricity use grew at a steady pace, the grid faces unanticipated strain due in part to growing demand from data centers. The jobs of professionals managing the infrastructure have evolved from traditional engineering tasks to complex, fast-moving challenges.

Industry reports show that millions of modern digital sensors, smart meters, and grid monitors are generating nonstop waves of information. The sheer volume of data requires instant, automated computer analysis because human operators cannot process it fast enough.

Pressure on utilities stems from two sources: a spike in electricity demand and a shift in how power is generated.

An example of the operational strain can be seen at the regional level. With the recent deployment of artificial intelligence tools and high-performance computing, data centers require immense amounts of energy to operate. The largest power transmission utility in Texas recently reported a staggering 220 gigawatts of new connection requests, driven largely by a surge in AI and cloud-computing facilities, according to a CNBC report.

Alongside the rise in regional demand, global energy networks are absorbing an unpredictable variety of weather-dependent renewable energy such as wind and solar. The switch creates a volatile operating environment wherein supply and demand are balanced, second by second, to prevent blackouts.

The challenges are compounded by the vulnerability of the grid’s physical and digital framework.

More-frequent severe weather events cause costly disruptions, such as the devastating winter freeze that crippled the Texas grid and record-breaking heat waves that have overloaded transformers.

Simultaneously, the energy networks’ digital architecture faces threats. As utilities replace outdated analog equipment with smart meters and control systems, they are increasingly vulnerable to cyberattacks.

To overcome physical and digital vulnerabilities, grid reliability organizations, such as those conducting North American security simulations like GridEx, emphasize that the grid must become smarter, more agile, and completely automated. Energy researchers are noting that the key to this change lies in integrating AI across every layer of utilities’ operations.

The AI imperative

According to energy industry experts, using AI to manage power systems is no longer a futuristic research project; it has become a baseline operational necessity. Grid analysts emphasize that traditional grid-planning methods are too slow to handle rapid energy dynamics or to balance volatile renewable energy in real time within decentralized power systems such as microgrids.

AI can fill the gap by processing vast amounts of data instantly. Machine learning algorithms can quickly analyze information from thousands of sensors, historical usage patterns, and weather forecasts to predict issues before they happen.

An industrial digitization study conducted by McKinsey & Co. indicated that integrating advanced data and automation across infrastructure networks could reduce system design errors, decrease equipment downtime by up to 50 percent through predictive maintenance, and extend the lifespan of power machinery by up to 40 percent.

From forecasting energy spikes to automatically fixing localized voltage drops, AI acts as the digital backbone of a self-healing grid, experts say. Deploying the complex systems requires a new workforce: power engineers who understand data science, as well as data scientists who understand electricity.

Upgrading the Workforce

To bridge the gap between groundbreaking AI research and practical field deployment, IEEE Educational Activities, in partnership with the IEEE Power & Energy Society, has launched the online Artificial Intelligence for Power and Energy Systems course program.

The program explores core challenges threatening modern utilities. Rather than treating AI as an unverified black box that operates without human supervision, the curriculum focuses on safety, asset preservation, and strict reliability standards.

The curriculum is designed to educate power system engineers, utility managers, and data scientists tasked with modernizing the grid. The program was developed by Fangxing “Fran” Li, professor of electrical engineering and computer science at the University of Tennessee in Knoxville and chair of the IEEE Working Group on Machine Learning for Power Systems.

Five learning modules

The program breaks down the technical transition into five modules that bridge high-level theory with real-world solutions:

AI fundamentals. This module teaches engineers how basic machine learning models apply to power grids. It discusses how specialized neural networks solve complex power-flow calculations and how AI models can safely transition from computer simulations to physical, high-voltage equipment.

Accelerating grid control. Learners are taught to leverage deep reinforcement learning, an AI approach that uses trial and error, to accelerate automated grid adjustments during emergency power events.

Forecasting and data analytics. Using predictive modeling, engineers learn how to predict sudden demand surges, variable wind and solar outputs, and fluctuating wholesale electricity market prices to keep power affordable and available.

Physics-informed and safe AI. To address trust—a barrier to utility AI adoption—this course covers AI models hard-coded to obey the laws of physics. The approach is designed to ensure that automated algorithms never make erratic choices that damage grid equipment.

Generative AI and next-generation tech. Learners can explore the frontier of utility technology, including graph neural networks and large language models. This module highlights how generative AI can process complex, interdisciplinary data to streamline utility planning, emergency responses, and regulatory reporting.

The algorithmic literacy and practical execution tools provided by the course program can help convert systemic risks into grid resilience.

For individual access, visit the IEEE Learning Network. If you are looking for customized organizational options, contact a content specialist to discuss volume pricing.

Despite setbacks in sodium-ion battery development in the U.S., the startup Peak Energy says it can be the one to finally achieve the ambitious application of using the tech to store energy for the grid.

When U.S. start-ups Natron Energy and Bedrock Materials shut down sodium-ion battery operations last year, they joined an ignominious list of more than dozen failed Western battery companies.

CATL, based in China and the world’s largest battery company, then dropped a bombshell in April, announcing it would supply 60 gigawatt-hours of sodium-ion cells to the grid storage provider HyperStrong. The largest sodium-ion battery order in history suggested China was on its way, as with lithium-iron phosphate in previous years, to dominating yet another promising chemistry.

The Colorado-based Peak Energy insists it can succeed where companies like Natron failed. Company executives say its sodium-ion tech can compete directly with low-cost lithium-iron phosphate (LFP) batteries, which currently dominate grid storage. Unlike Natron, which ran out of money and investors’ patience, Peak Energy has a giant in its corner: General Motors. Like Tesla and other automakers, GM is moving aggressively into grid storage to keep massive battery factories humming in the wake of slumping EV demand.

Peak Energy has formed a partnership with the automaker to ultimately deploy sodium-ion batteries at grid scale. In July, the company announced it will build a US $71 million, 17,000 square-meter factory near Sacramento, with capacity to produce 4 GWh of sodium-ion batteries annually, enough to power 4 million homes.

Sodium-Ion vs Lithium-Iron Phosphate Batteries

As with other sodium-based designs, Peak Energy’s cells can’t yet match the energy density of LFP batteries. Company executives from Peak Energy and GM, which has partnered with Peak to codevelop and manufacture the batteries, freely admit they can’t currently compete with current LFP prices on a per-cell basis.

Yet Peak Energy says its passively cooled storage system will still cost operators 20 percent less over its lifetime compared with LFP storage. Cameron Dales, Peak Energy’s cofounder and chief commercial officer, says the company’s GS1.1 system will store energy for 20 years, over roughly 20,000 cycles, and still retain 80 percent of its capacity. For LFP, a basic durability benchmark pegs them at 70 percent capacity after 8,000 cycles.

Peak Energy’s case is helped by a booming market for energy storage, to back up AI data centers and to store excess solar and wind energy. Volatile lithium prices have major players looking for a steady alternative.

Incumbent LFP batteries, Dales says, were initially designed for EVs, where cell costs and energy density are critical for driving range and affordability. But utilities and storage operators are focused on entirely different metrics. They want batteries that last the longest, at the lowest overall cost, to maximize returns and justify massive capital investments.

The company’s technical edge, Dales says, is that its stable cells that can operate safely at temperatures roughly double the typical operating temperatures of LFP, which performs best at or near room temperature.

“You need to keep an LFP cell at 25 °C, give or take, or it will rapidly degrade,” Dales says.

The generous temperature tolerance of Peak Energy’s cells allows a passive cooling system with no pricey, trouble-prone fluid cooling loops or moving parts such as fans or pumps. That makes the modular system well-suited for data centers or grid support in desolate areas, where operators are finding abundant cheap land, high potential for solar energy, and a low risk of natural disasters.

“You don’t have to power a refrigerator in the desert for 20 years to keep the system operating properly,” Dales says.

The GS1.1 system stacks slender prismatic cells into modules roughly the size of a king-size mattress. A 36-module system stores 3.1 megawatt-hours. Large-scale projects might combine dozens or hundreds of units, generating enough juice to power a small city.

Proponents note that sodium is the sixth-most abundant element on earth, roughly 1,000 times as abundant as lithium. And unlike lithium, which must be sourced from far-flung regions fraught with environmental or human-rights issues, the world’s largest, purest deposits of trona are found in the Green River Basin in Wyoming. Trona, which is composed of sodium carbonate, sodium bicarbonate, and water, supplies the United States with 90 percent of its soda ash, which is the basis for battery-grade lithium carbonate salt. However, analysts note that while the raw material may be plentiful, its processing is dominated by China.

“The supply chain tends to get overlooked, but it’s just a massive issue,” says Varnika Agarwal, a battery research analyst at Benchmark Mineral Intelligence.

Agarwal says sodium-ion tech holds promise. But a lot has to go right for sodium ion to carve out a viable niche in the U.S. Benchmarks project that less than 1 percent of newly deployed storage in the United States will be sodium ion this year, less than 4 percent by 2030, and 5 percent globally.

For now, she notes, Peak Energy is buying its commercial cells via contracts with Chinese suppliers, which dominate both processing of its raw materials—however cheap-and-abundant they may be—and cell production. The U.S. is basically just getting started, with Peak Energy’s California factory slated to come online in 2027.

NFPP Cathodes in Sodium-Ion Batteries

Natron was banking on long-shot “Prussian Blue” electrodes, a form of blue pigment that acts as a sponge to soak up and release sodium ions. Peak’s batteries, however, rely on sodium iron pyrophosphate (NFPP) cathodes, which are chemically and structurally similar to lithium-iron phosphate in an LFP battery, known for superior safety and long cycle life.

NFPP is fast becoming an industry standard. CATL has also settled on NFPP for its core chemistry. This helps make Peak Energy’s cells largely “drop in,” able to be manufactured at existing battery plants such as GM’s—a huge advantage for market viability.

Kurt Kelty, Tesla’s former battery guru and a globally recognized battery expert, is now vice president of batteries and sustainability at GM. Kelty says GM is backing Peak Energy for several reasons, including his familiarity with and respect for its own former Tesla execs.

Peak Energy executives [from left] Landon Mossburg, Kurt Kelty, and Cameron Dales show off one of the company’s prismatic cell batteries.Lucas Breuer/Peak Energy

GM has been testing Peak Energy’s cells, in 170 and 190 ampere-hour formats, at its Wallace Battery Cell Innovation Center in suburban Detroit, in the same labs where GM is developing its proprietary lithium-manganese-rich battery chemistry. Kelty says the battery life of competing cells, from a full range of global producers, “falls off a cliff” during high-temperature testing. But Peak Energy’s batteries are withstanding extreme testing at up to 55 °C, with notably little effect on their lifespans.

“The cell is kicking butt over everything,” Kelty says. “We can get 20 years of lifetime without a cooling system, and that’s the key.”

The batteries are showing a round-trip efficiency of 96 percent, a significant 2 to 3 percent better than LFP. (“Round trip” refers to the amount of energy a battery discharges, relative to the amount used to charge it).

Eliminating active cooling, Kelty adds, allows a near-silent system that could be used in public buildings. In addition, it ditches the plumbing that’s a potential trouble spot for leaks and maintenance.

“It also reduces parasitic power losses, because you’re not using power to cool the system,” Kelty says.

Peak and GM executives believe sodium-ion cells themselves will reach price parity with LFP around 2028, based in part on discussions with Chinese cathode suppliers. Sodium ion, they say, is just entering its steep slope of cost reductions, where LFP’s savings have largely been realized over 20 years.

Peak Energy’s Sodium-Ion Battery Projects

In March, Peak Energy announced it will join with RWE Americas for another pilot system near Milwaukee. That would become the first-ever use of sodium-ion backup on the Midcontinent Independent System Operator, the regional grid operator for 15 central states and Canada’s Manitoba province.

Peak Energy also plans to begin supplying up to 4.75 GWh of batteries to Jupiter Power, an independent storage developer, through 2030. The deal, worth up to $500 million, includes an initial 720 MWh of storage, including in Texas, the nation’s largest single announced deployment of the batteries to date.

Ask Dales about winners and losers in battery chemistries, and he’ll tell you it’s the wrong question. In a world of planes, trains, phones, drones, and every imaginable device, it makes no sense that one “super battery” would rule them all.

“Sodium ion is just another sister technology to lithium,” he says. “But instead of going higher energy density, it’s going lower energy density, and you’re paying for that with better stability and safety at a lower cost.”

This story was updated on 23 July, 2026 to correct the timeline for Peak’s California facility. It is scheduled to come online in 2027, not 2028 as originally stated.

In the chill of a London spring night, under overcast skies, iconic Trafalgar Square opens around me. Admiral Nelson rises on his pedestal, the National Gallery rests behind, the church of St Martin-in-the-Fields sits nearby. From the 13th century, the site served as the Royal Mews for hawks and then horses. By 1844, it was a public space at the heart of one of the biggest cities in the world.

Despite the square’s presence through that grand sweep of history, it’s not why I’m here. My interest is far more specific: I want to find out what happens to spaces like this when artificial light, specifically from light-emitting diodes (LEDs), intrudes. My companion tonight is Simon Thorp, a local lighting designer, who crouches in the shadows near Nelson’s spire, light meter in hand. “Two lux,” he reports, “and it’s very comfortable here.” Two lux is 10 to 20 times the illuminance of a full moon. We can see each other clearly, and a nearby sign assures us that closed-circuit television (CCTV) is in operation for safety’s sake.

Trafalgar Square captures the relationship between lighting and darkness that exists in almost every city, suburb, small town, and village around the world. The lighting here is a mishmash of old technologies and new, of shadow and glare, the ornamental gas lamps fronting the National Gallery all but washed out by the LEDs inside modern versions of traditional “brass and glass” fixtures a few meters away—21st-century technology housed in 19th-century designs.

The ugly truth of artificial lighting today is that in parts of London, as in many cities around the world, lighting levels are excessive, with unshielded illumination blasting in all directions. And the irony is that too much light invites danger: It creates shadows, impedes our vision, and gives the illusion, without the reality, of safety. Thorp notes that modern CCTV cameras are “pretty great” even at low-light levels, while harsh light makes it hard for both human eyes and digital sensors to see.

“The more bad light we add, the more bad light we think we need,” says Thorp. “We can’t see because of the light we’ve added. And it makes areas that were perfectly okay seem darker.”

Among the costs of this excess, the most alarming may be its toll on human health (and that of other animals) by disrupting circadian rhythms, impeding the production of melatonin, and contributing to sleep disorders that are tied to every major modern disease. New research shows that increased exposure to blue light from LEDs is having “substantial biological impacts” such as suppression of the sleep hormone melatonin and an increased risk for obesity, certain cancers, and type 2 diabetes.

A panoramic view from the Eiffel Tower looks down the Pont d’Iéna and the Palais de Chaillot [in the center], with the imposing silhouette of the La Défense business district visible on the Parisian skyline.

Luigi Avantaggiato

I have come to London and Paris—which led the way in the expansion of public street lighting in the 19th century—because they embody both the current enormity of the problem as well as a future certain to be lit by trillions of chips: controllable, tunable, and energy-efficient LEDs.

Living with artificial light at night

The standard justification for nighttime illumination is public safety. Lighting experts’ term for the phenomenon is “artificial light at night.” While people won’t often admit it, the desire for light at night seems to stem from a primal fear of the dark. Darkness is where the bad guys hide. And if dark is bad and light is good, then more light can only be better.

This assumption has guided our use of nighttime light for hundreds of years. And yet, high-lumen output doesn’t necessarily correlate to a reduction in crime, research has found. In other words, if we relied on the data as much as we do our primal anxieties and paused those anxieties long enough to learn how light and darkness interact, our nights would almost certainly be lighted differently—especially now that we have the extraordinary technology that is the light-emitting diode.

A 19th-century gas lamp [white square] manufactured by William Sugg & Co. next to a pedestrian path in Trafalgar Square, along with the architectural floodlighting on the neoclassical facade of the National Gallery, showcase the interplay between modern and historic lighting systems.

Luigi Avantaggiato

The first visible red light-emitting diode was invented by Nick Holonyak in 1962, but LED lighting technology took several decades to develop, before exploding in recent years. Just a decade ago, LED streetlamps were rare. By 2019, more than half of U.S. streetlights were LEDs, and that number is predicted to top 90 percent by 2030. Similar uptake has occurred around the world, even in developing countries, where inexpensive Chinese-made LED fixtures are increasingly common.

This rapid global migration to LEDs represents a shift in the fundamental physics of how we illuminate our world. From oil lamps and candles to gas lamps, early examples of artificial light at night relied on a burning wick, an incredibly inefficient way to create light. An incandescent bulb is effectively a heater that happens to produce light as a by-product, so it squanders nearly all of its energy as heat.

By contrast, LEDs use semiconductors to convert electricity into light. Through this process of electroluminescence, LEDs use up to 90 percent less energy than incandescent bulbs do, which has enabled municipalities to realize an immediate energy savings of 50 percent or more. This fact alone has fueled the technology’s worldwide adoption.

But LEDs aren’t just more efficient and less expensive. The use of solid-state technology gives the lights an extraordinary life-span, often measured in decades rather than years. This significantly lowers the maintenance costs, as city workers spend far fewer hours replacing broken or burned-out lights. Even more striking, by manipulating the properties of the semiconductor material, engineers can dictate the precise color and intensity of the output, something that gives LEDs incredible versatility. For a lighting designer like Thorp, LEDs offer countless possibilities.

Digital control for smarter lighting

Wandering from Trafalgar Square along the edge of St. James’s Park, Thorp and I find ourselves near Westminster Bridge, one of nine city bridges that in 2021 were part of the Illuminated River project, meant to make the Thames more beautiful at night. Each bridge now features a new LED lighting scheme that moves and changes color and intensity to create a coordinated work of art. But Thorp is frustrated that the project did nothing to correct the often glary lighting on the riverbanks. “Why don’t you pay the money to correct all of this bad lighting instead of adding new lighting?” he says. LED technology, he points out, has the potential to fix that problem.

The Illuminated River artwork for Blackfriars Bridge uses a color scheme that closely complements the red pillar supports that remain from the original Blackfriars Railway Bridge.

Luigi Avantaggiato

In fact, this may be the most meaningful potential of LEDs: the ability for a community to control when, where, at what levels, and in which colors its lights shine. A public space like Trafalgar Square could be lit more brightly during rush hour, then dimmed as the night progresses, the lights not only connected to one another but to the surrounding streetlights and commercial lights. Anywhere in the world, LED streetlights could be programmed to rise and fall in brightness depending on the time of night or time of year. They could even be turned off during bird migrations, to reduce the number of birds that are disoriented by the lights and ultimately killed in collisions with reflective and illuminated windows.

Up to now, LED public lighting has largely not been part of any comprehensive plan to curtail and control nighttime illumination. Most LED installations have simply replaced older, inefficient “dumb” electric lighting with newer “dumb” LEDs and thus made light pollution worse. The main reason? Because LED lighting is cheaper, we tend to use more of it—a literally shining example of the Jevons paradox. Even as awareness of light pollution grows, we aren’t yet taking advantage of LED technology’s full potential.

The good news is that we could start tonight. At DarkSky International, the world’s foremost organization fighting light pollution, CEO and executive director Ruskin Hartley tells me the organization has five principles for responsible outdoor lighting: It should be useful, targeted, low level, controlled, and warm-colored. “They’re enabled because of the capabilities of LEDs,” Hartley says.

The technology to control LEDs will be part of the solution. In the olden days of the analog era, a streetlight was either on or off. To change a lighting schedule, you had to physically rewire a circuit. Today, the Digital Addressable Lighting Interface (DALI) protocol turns each luminaire into part of a network, with its own digital address and a driver that reports to a central server. With DALI, the lighting is managed through software rather than physical switches.

Unfortunately, most LED streetlights have been deployed without this technology because it costs more. But Paul Drosihn, general manager of the DALI Alliance, says that using such controls makes the LEDs much easier to maintain. Before the new digital protocol, it took an average of nearly three visits to identify, diagnose, and repair a defective luminaire. “Now it’s one,” Drosihn says. “Saving the cost of sending two guys on a cherry picker to replace those [lights] is immeasurable. What I just described to you is pretty much all the utilities need to know.”

The intricate cast-iron understructure of Westminster Bridge glows in vibrant green and teal light as part of the Illuminated River public art project, a color palette selected to echo the green benches of the nearby House of Commons.

Luigi Avantaggiato

What’s more, DALI allows the tuning of the lights’ spectrum so that they become warmer and less disruptive as the night progresses. Digitally connected, full-spectrum luminaires allow cities to transform the nocturnal experience—saving money, increasing health and safety, and creating a warmer and more appealing atmosphere at night.

This digital intelligence is equally transformative for the “bleed lighting” that spills from building interiors, Drosihn says. “You don’t think of night lighting as coming from inside buildings, but it does,” he says. “Particularly in the States, you drive through any major city and all the lights are on, on every floor of every high-rise, even if no one is home.”

Already, the use of digital controls for interior lighting has become commonplace in some European cities, Drosihn says. By integrating DALI with occupancy sensors and building-management systems that monitor HVAC, electrical systems, and security networks, a skyscraper can become a dynamic participant in the urban environment—dropping a floor’s interior lights to zero the moment the last person leaves. As a result, electronic controls combined with LEDs can act like a dimmer switch for a city’s entire skyline.

White light blights the night

With all the possibilities from LED technology, why are our nights too often lit with harsh and clinical light, casting glare and creating shadows, disrupting human and ecological health, erasing the stars from our skies? The answer starts with the color of LEDs. At first glance, an LED streetlight looks like a collection of small white bulbs, but it’s not. To produce a light we perceive as white, most manufacturers coat a blue semiconductor core with a yellow phosphor material that absorbs a portion of that high-energy blue light. The problem is that this “white” light is still heavily blue, which is exactly the color no species has evolved to expect at night.

And because blue light is the second most energetic part of the visible spectrum (violet is the most), it doesn’t just illuminate our streets and invade our homes. Blue light also scatters in the atmosphere more easily than any other color, which helps to create the hazy, illuminated fog known as sky glow over every city of any size.

“Cooler” colored LEDs in the 4,000- to 6,500-kelvin range offer the most lumens at the lowest cost, so most early adopters installed these blue-rich white lights. The good news is that LED technology has continued to advance, and a growing number of communities are choosing warmer-colored streetlights that have less blue. (Phoenix, for example, converted 100,000 streetlamps to 2,700 K LEDs in 2020.) And, of course, light pollution isn’t just a result of LEDs. Older lighting technology also adds to the glare—bright white metal-halide lights, especially—and cities are loath to replace something that isn’t yet broken.

But our main failure isn’t a technical one. It’s that we have yet to revise our thinking about lighting at night. We use LED technology just as we did the old sources of light. As a result, we have largely offset the gains that were promised in terms of reducing energy consumption and carbon emissions by making light pollution worse, and have so far let an incredible opportunity go unrealized.

Can the City of Light do it right?

Across the Channel in Paris, the failure to realize the potential of LEDs feels even more palpable. Unlike London, which suffered heavily from German bombs, the lovely 19th-century Paris that Baron Haussmann created largely escaped destruction in World War II. To nearly 50 million annual tourists, the beautiful uniformity of the architecture is instantly recognizable. But the City of Light’s nocturnal atmosphere is also part of the draw, and extensive attention has been given to relighting its buildings and monuments. When I wander into the Cour Carrée in the Louvre, for example, I’m stunned by rows of amber LEDs that together create a warm glow along the palace facades. When I see the Eiffel Tower, first from a distance walking along the Seine and then up close, I find myself staring as I would at a campfire, the structure’s metalwork amber-lit with more than 336 high-pressure sodium bulbs.

Still, the city’s night lighting is far from perfect. With millions of residents, thousands of stores and restaurants, and 300,000 streetlights, the city overall is among the world’s brightest. Even at the base of the Tower, bright white LED lamps illuminate the African émigrés selling cheap berets and Eiffel Tower trinkets. And the city has been replacing its old sodium streetlights with new LEDs, swapping the warm yellow tones for which the city has long been known for bright white lamps no one wants to look at.

Street vendors display souvenirs at the foot of the Eiffel Tower. Their merchandise is lit by harsh white LED systems, which starkly contrast with the warm golden sodium-vapor light illuminating the tower above.

Luigi Avantaggiato

Nonetheless, the potential is here. In 2019, France introduced a nationwide law to reduce levels of light pollution, setting rules about both public lighting (preventing light from being projected above the horizontal) and private lighting such as stores, which are required to turn off their exterior and shop window lights after 1 a.m. In addition, an increasing number of French communities dim or turn off municipal lights after midnight to save energy and reduce carbon emissions. Although light pollution worldwide continues to increase by nearly 10 percent per year, France has managed to reduce its overall level. Chloé Beaudet, a researcher at Université Paris-Saclay, documented local light-reduction measures and found people generally agreed with the notion of dimming or turning off the lights, mainly for energy savings and ecological concerns.

“What I find is that people living in urban areas, they accept this kind of policy,” she tells me. “They’re like, okay, I don’t really use public space at night as a pedestrian, so what’s the point of having lights on?” For her, a key takeaway is that one lighting level does not fit all areas. “I think there is really a need for policy that is differentiated according to the neighborhood.”

In another positive development, the country has been minimizing artificial light to create ecological corridors designed to protect nocturnal species such as birds, bats, and insects. These corridors are connected and dark, mitigating the disruption to the 30 percent of vertebrates and more than 60 percent of invertebrates that are nocturnal. Even for city dwellers, this trame noire (“dark infrastructure”) helps to raise awareness of why controlling light pollution is important for life on Earth. Nationwide laws to control light pollution, the ability to light different parts of a city differently, dark corridors to protect biodiversity—these are exactly the kind of changes made possible with LEDs.

The iconic I.M. Pei Pyramid glows softly at the center of the Cour Napoléon at the Louvre Museum, its warm LED illumination flowing through the geometric glass-and-metal structure with a symmetrical framing of the surrounding historic pavilions against the night sky.

Luigi Avantaggiato

That’s not all. Almost until 1920, astronomers at the Paris Observatory were still gazing at the Milky Way. That’s impossible nowadays, but it could happen again. Despite the bright white LED streetlights now lining so many Paris streets, networks of LEDs using controls could lower lighting levels enough each night, so that the Milky Way could once again be visible over the French capital. And in the process, Paris could become the City of Light in ways that would set an example for other parts of the world.

New lighting demands new thinking

“I think we should aspire to have cities that see the stars,” Simon Thorp says when I mention this view of Paris. “You just need everything to be coordinated.”

Nearing the end of our London walk, having turned from the river and back up toward the Strand, Thorp brings me down narrow Carting Lane behind the Savoy Hotel, to where a gas fixture tops a thick lamppost, an original from 1870. A small plaque reads, “The last remaining sewer gas destructor lamp in the city of Westminster.” Thorp explains that the thick pole hides a tube that allowed methane from the sewers to get burned off at the mantle. “An early example of renewable energy,” he jokes.

The fire-orange flame is pleasing to the eye. But even here, on a narrow lane with no vehicle traffic, in a touristy area of the city, the flame is overwhelmed by a nearby, unshielded LED security light. Thorp shakes his head. “It’s stunning that someone could put in a light like that and think, ‘Great, nice job.’”

Here is the crux of contemporary artificial lighting at night. We know how to light well, and LEDs give us the ability to do so. But while our technology is 21st century, too often our thinking about light and darkness, safety and security, is stuck in the past. We could be doing so much more with this technology than we are. We could relight our nights in ways that would not only reduce energy and maintenance costs but also bring a slew of benefits, including healthier nights for humans, safer skies for nocturnal creatures, and a restoration of the stars.

In 2026, the tale of these two cities and their artificial light at night is that of a brilliant technology that we’ve engineered but haven’t yet learned to master. In short, we have yet to change the way we think about artificial light at night and to use it more thoughtfully and carefully—as we might, as one hopes we will.

This article appears in the August 2026 print issue as “The Wrong Way To Light a City.”

The stroke of a pen in Utah set off a solar chain reaction across the United States. Last March, the Western state legalized plug-in photovoltaic: solar panels small enough they can plug into a typical household outlet, helping power the house. Dozens of states followed suit shortly thereafter. Balcony solar, as the tech has become known since it could be deployed on the balcony of an apartment building, is now legal in states such as Colorado and Virginia. By July 2026, balcony solar bills had passed one or both houses of the legislatures states including New York, New Jersey, and California.

The U.S. embrace of this technology happened so fast that legalization has raced ahead of regulations needed to make sure it works safely. No panels meet newly created certification standards, though one microinverter announced last week does. The country still must solve a suite of legal and technical problems, said Craig Morris, the CEO of the German plug-in solar trade association Bundesverband Steckersolar. “If you get the law,” he said, “then your work begins.”

Morris knows the growing pains because Germany is where balcony solar came of age. In the 2010s, tinkerers realized they could use newly available microinverters, from large solar arrays, to send AC power from a single, affordable panel to a household plug. Balcony solar took off in popularity there as a clean and affordable energy solution after the 2022 Russian invasion of Ukraine sent natural gas prices soaring. Now, Germans can buy plug-in panels at Aldi or Ikea, and the country has more than 2 gigawatts of balcony solar deployed.

In December 2025, Germany finalized a regulatory standard to ensure these devices could safely send electricity back into a home power outlet and even the grid. Germany’s balcony solar law forbids net metering, in which grid operators pay customers who return electricity to the grid, because it creates so much uncertainty for grid managers. Morris’s organization got German firefighters to agree the technology isn’t dangerous. German law also protects renters whose landlords don’t want them to mount solar panels for aesthetic reasons. Property owners need a legitimate reason to oppose a resident’s request, such as necessary repairs to the balcony.

Coming to America

Ken Boyce watched all this from afar. Boyce is vice president of principal engineering at UL Solutions, the firm that creates testing and certification standards for a wide variety of electrical technologies, which now includes balcony solar. The speed of U.S. adoption of balcony solar caught him by surprise. “That Utah legislation passed and we said, “Oh, how about that?” Soon after, Boyce and a cadre of engineers got to work. By the end of 2025, UL Solutions had developed a basic framework for UL 3700, its attempt at a U.S. testing and certification standard for balcony solar.

The standard addresses numerous concerns about the safety of plug-in PV. The simplest problem is that household wiring simply wasn’t designed to handle power flowing into the outlet from a plug. Unlike plugs on ordinary appliances, a solar panel’s plug can be electrically conductive with the device unplugged, which may surprise unwary customers. Plug-in panels inject power into the home electrical system downstream of circuit breakers and other overcurrent protections, making those safeguards blind to the extra load. Electricity flowing backward through the GFCI (ground fault circuit interrupter), the electrical safety device built into power outlets, could cause GFCIs to fail. “There’s this kind of insidious aspect to it,” Boyce said, “and now you’ve bypassed that protection that’s required to protect consumers from electric-shock hazards.”

All of these factors make a plug-in solar panel much different from household appliances, such as a toaster or refrigerator, says Nadav Enbar, a program manager at the nonprofit Electric Power Research Institute (EPRI). Nevertheless, he says it’s crucial that regulators treat the balcony solar panels like an appliance. Its ability to plug in and just work—without the owner needing agreement from a utility, permits, or an electrical upgrade—is central to the technology’s mass appeal and its ability to get solar into so many more homes.

Making Balcony Solar Safe to Plug and Play

So, UL3700 addresses those dangers. In addition to specific requirements such as the thickness of insulation, it also requires manufacturers to account for other known risks–for example, by using wiring methodologies that lend additional overcurrent protection or by using GFCIs built to withstand bidirectional current. Plug-in solar must also power down in the case of a local blackout. This addresses one of the chief concerns cited by power utilities: that plug-in PV, were it to send power onto the grid during an outage, poses a danger to linemen.

How plug-in PV manufacturers satisfy those rules is up to them. But because UL3700 and balcony solar panels are so new, no panels have achieved the standard so far. Enbar notes that the UL standard is a high bar to clear, and, in fact, some of its requirements may not be solvable at this moment. For example, the bidirectional GFCI disconnect is not yet commercially available.

Balcony solar is momentarily caught in limbo: with one recent exception, the systems now available, and going online in states like Utah, are either not certified or cite some other electrical certification that wasn’t written for plug-in PV, according to EPRI.

The last piece of the puzzle is transparency. David Eisenhauer, a representative of the power utility SoCal Edison, told IEEE Spectrum that while Edison supports its customers gaining access to plug-in solar, “it’s really important that customers notify us before they start to think about or install plug-in or balcony solar. That’s going to help us ensure that they’re installing certified equipment, help us track connections so that we can balance and operate the grid.”

This may be easier said than done: EPRI’s research shows that even in industry leader Germany, unregistered balcony solar panels outnumber registered panels two to one. Some customers simply don’t want to do the extra homework to register a purchase, while others, Morris says, might not want to publicize their plug-in PV or let landlords know about it. Given the likely rise of unregistered “guerilla” solar, plug-in PV just has to work safely, every time, no matter how many people set up a solar panel on their balcony or in their backyard.

This story was updated on 21 July 2026 to reflect the fact that a microinverter announced during the story’s reporting does meet UL 3700. Thanks to eagle-eyed reader Stephen Mayer for pointing this out. It was updated again on 24 July 2026 to correct UL Systems to UL Solutions. We apologize for the error.

Data center developers and hyperscalers are experimenting with 800 volt direct current (800 VDC) inside their largest data centers. It enables them to pack more compute power in much smaller spaces to run the latest AI applications; 800 VDC also reduces waste heat from electric wires, lowers cable weight, and eliminates equipment needed in multiple AC to DC to AC conversions.

The Large Load Working Group (LLWG) of the Electric Reliability Council of Texas (ERCOT) is evaluating how to accommodate several 800 VDC facility designs. Data center power infrastructure provider Dimaag, for example, presented in April a way to protect the grid from AI’s massive load variations, which can switch from almost nothing to hundreds of megawatts and back again several times per second. DiMaag’s system also enables data centers to stay connected to the grid, or “ride through,” during short drops in voltage. This ability is known as Low Voltage Ride Through (LVRT).

“Grid infrastructure wasn’t built for the large load oscillations from AI data centers and large load disconnections due to voltage dips,” says Sadha Kameswaran, vice president of business development at Dimaag.ai.

Dimaag’s solution comprises batteries and control software that work with devices that switch electricity from AC to DC called rectifiers. This system, located between the grid and an AI data center, isolates DC load fluctuations from the grid and delivers LVRT support that can respond to demand shifts in real time at scale. Further, it lowers the number of AC to DC conversions required. Hyperscale data center developers are now evaluating this approach, Kameswaran says.

A power system study conducted by Electric Power Engineers (EPE) on behalf of Dimaag evaluated the technology in a simulated data center environment with regard to load smoothing, LVRT compliance, provision of backup power, and demand response. EPE concluded that this approach was an “effective solution for addressing the power quality and stability challenges inherent in modern data center operations.” The Dimaag module was found to be effective in buffering extreme DC-side power fluctuations and delivering LVRT response.

“Voltage ride through is key to maintaining balance between power generation and load,” said Woody Rickerson, COO of ERCOT during a panel at the Data Center World conference.

Power Electronics Alternatives

Companies are developing other ways of helping large loads integrate into the grid. The National Laboratory of the Rockies (NLR), for example, has tested ON.energy’s AI Uninterruptible Power Supply against ERCOT’s large electronic load (LEL) ride-through requirements. Another alternative from Ramboll acts as a shock absorber for the grid, helping utilities and grid operators to add more renewable energy sources while lowering risk, preventing outages, raising resilience, and lowering development costs.

One of Ramboll’s innovations is to address reactive power, a component of electricity that does not perform useful work but is essential in maintaining electrical and magnetic fields and controlling voltage levels. Ramboll have developed a Static Synchronous Compensator (STATCOM) that adds more controls than traditional STATCOMs that provide damping capabilities to detect and neutralize any voltage oscillations by absorbing energy from the oscillations and converting it into useable power.

Protecting ratepayers and the grid

This work to address data center load variability comes at a time of mounting public and governmental pressure. The Governor of Texas just ordered the Public Utilities Commission (PUC) and ERCOT to safeguard residential and small business ratepayers from added costs due to the expansion of AI data centers and to require that data centers fund any new electric infrastructure their operations require.

That was followed by the PUC approving ERCOT’s proposal on how to connect large electricity users such as AI data centers while protecting the reliability of the grid via provisions for LVRT and load smoothing. As well as incurring the costs for additional substations, transmission, and power electronics equipment for their new facilities, large load customers are encouraged to agree to curtail power usage at times of local transmission constraints. This comes at a time when ERCOT has 438 GW of large load interconnection requests, almost all from AI data centers.

“Texas is experiencing an energy transformation unlike anything we have seen before,” said ERCOT President and CEO Pablo Vegas in a press release. “This new process represents a fundamental shift in how ERCOT manages the significant growth of large load interconnection, providing a structured, transparent path forward that protects reliability for Texans while supporting the state’s continued economic growth.”

AI versus grid pace

Nvidia, AI hyperscale developers, and their supply chain partners are pressing ahead with their plans for 800 VDC data centers. Nvidia envisions that 800 VDC facilities will begin appearing in 2027 to support its forthcoming GPU design. However, these organizations are moving at a pace that is far in advance of the power industry.

ERCOT is known as one of the most progressive grid operators in the nation. Texas has more solar energy, wind power, and battery energy storage systems (BESS) than any other state, but it is far from ready for 800 VDC. The LLWG continues to evaluate 800 VDC.

For now, ERCOT continues to review a variety of 800 VDC, but no decisions are imminent on the preferred approach.

“There are a lot of solutions out there if you are listening for them, as help can come from many areas,” said Rickerson.

This article was updated on 17 July 2026 to add several paragraphs in the opening section inadvertently omitted from publication.