A New Coal Plant in the U.S.? Once Unthinkable, Now a Strong Maybe

 

 

June 16, 2026 - A $350 million Department of Energy (DOE) coal-revival program has put $18.5 million toward the TerraSpark Energy Campus, a 1.6-GW greenfield project in West Virginia pairing Babcock & Wilcox (B&W) supercritical boilers with Mantel Capture’s molten borate carbon capture. In responses to POWER, developer TerraSpark laid out a 2030 startup target, a 95% to 98% capture design, and a four-pathway plan for the captured CO2.

 

New coal-fired generation has been effectively off the table in the U.S. for more than a decade. The TerraSpark Energy Campus is testing whether that’s still true. The DOE’s selection of the Grant County, West Virginia, project for up to $18.5 million in development funding is a notable signal that surging electricity demand—driven by data centers, manufacturing, and electrification—is forcing a fresh look at baseload coal.

The award will support front-end engineering and design (FEED), permitting work, and early technical studies for the greenfield facility, which would be sited near the existing Mt. Storm energy complex. Combined with $21.5 million in non-federal cost share from the developer, the scoping and design phase carries a total value of roughly $40 million. Project developer TerraSpark—legally TerraPurus Inc., doing business as TerraSpark Inc.—announced the award on June 4.

“Electricity demand in the United States is growing rapidly, and we need infrastructure that can keep up,” said Bill Tolpegin, partner and founder of TerraSpark. “This project is about delivering reliable power when it’s needed, creating skilled, good-paying jobs, and helping advance carbon management technologies that strengthen America’s energy future.”

Four Awards, One Federal Coal Program

 

The TerraSpark award is one of four selections, totaling $350 million, that the DOE announced June 4 under Topic Area 1 of its “Restoring Reliability: Coal Recommissioning and Modernization” Broad Agency Announcement (DE-FOA-0003605). The $525 million program, launched in September 2025, implements President Trump’s executive orders “Reinvigorating America’s Beautiful Clean Coal Industry” and “Strengthening the Reliability and Security of the United States Electric Grid,” and follows the January 2025 declaration of a national energy emergency. The DOE’s 2025 Grid Reliability Assessment, cited in the solicitation, found that coal retirements are outpacing additions of other reliable generation resources—a problem the agency calls “especially acute in regions with constrained transmission and sustained load growth.”

Notably, all awards under the program are structured as milestone-based “Other Transaction Agreements,” tying federal payments to independently verified technical progress. The DOE also extended the Topic Area 1 application deadline specifically to allow applicants to propose the use of federal financing tools such as DOE loan guarantees—a potentially significant lever for projects of this scale as they move beyond the study phase.

TerraSpark’s $18.5 million is the smallest of the four Topic Area 1 awards, reflecting its early, scoping-and-design stage. The other selectees underscore the breadth of the DOE’s coal push. The other Topic Area 1 awards were:

- AES Puerto Rico received $164.5 million toward an $820 million retrofit and life extension of its 510-MW Guayama plant, including a FEED study for post-combustion carbon capture.

- AES Warrior Run received $78 million to recommission its 205-MW Maryland plant, which ceased operations in 2024.

- Terra Energy Center Corp. received $89 million to assess commissioning a new 1,250-MW coal-fired facility with integrated carbon capture in Alaska—meaning TerraSpark is not the only greenfield coal plant in the federal pipeline.

- A separate $175 million tranche, announced in February under Topic Area 2, went to six utilities to modernize and extend existing coal plants serving rural communities.

A Consortium Bet on New Coal

The TerraSpark project assembles a notable roster of partners. Babcock & Wilcox (B&W) announced on June 8 that its anticipated scope would include the design and supply of four 400-MW supercritical boilers and a full suite of advanced emissions control equipment, with Babcock & Wilcox Construction Co. LLC expected to provide construction services. Sargent & Lundy will support project engineering and design, while Advanced Resources International will handle carbon dioxide management, transportation, storage, and utilization planning.

Mantel Capture, a Cambridge, Massachusetts–based startup backed by Shell, Eni, and bp, is the project’s carbon capture technology partner. Its molten borate system is designed to capture up to 98% of CO2 emissions while cutting the energy penalty of capture by more than 97% compared with current technologies—a combination the company says enables capture at less than half the cost per tonne of today’s state-of-the-art systems.

“In an era of energy abundance, the world needs all forms of low-carbon and reliable energy working efficiently together,” said Cameron Halliday, Mantel’s co-founder and CEO. “Projects like TerraSpark’s Energy Campus are an important step forward, demonstrating how next-generation carbon capture can help ensure the industries powering our world remain strong, competitive, and part of a lower-carbon future.”

For B&W, the project is a return to familiar territory. “This project aligns with B&W’s continued commitment to serving the U.S. coal fleet, which is a reliable, secure, and essential component of the nation’s energy mix,” said Jimmy Morgan, B&W executive vice president and Chief Commercial Officer, who added that the company’s boiler and emissions control technologies are designed to “extend the life, improve performance, and enhance the environmental compliance of existing and new coal-fired assets.”

Supercritical Pedigree, Measured Expectations

 

If built, TerraSpark would be among the first new U.S. coal-fired plants in more than a decade—and B&W is leaning on a deep resume to make the case it can deliver. Brandy Johnson, B&W’s Chief Technology Officer, told POWER the company “designed and built the world’s first commercial supercritical boiler in 1957 and has since supplied more than 165 supercritical boilers in North America, Asia, and other regions with high demand for baseload power and advanced coal technologies.” The company also designed, manufactured, and constructed the U.S.’s first and only ultrasupercritical coal-fired boiler (Figure 1), along with its advanced air quality control system.

Still, Johnson was careful to frame the project’s current stage. “This project is currently a FEED study for the overall project which is a minor portion of our overall workload,” she said, noting that B&W has seen continued growth in its aftermarket and services business alongside “growing interest in extending the life of these assets and improving their performance while we continue advancing next-generation, lower-emissions clean coal technologies.”

Asked whether the project signals a broader revival in domestic coal infrastructure, Johnson pointed to a market in motion. “There continues to be a meaningful shift in domestic coal infrastructure work and strong demand for new baseload generation and behind-the-meter AI [artificial intelligence] data center projects, and we are well-positioned to support these projects,” she said. The company is “in ongoing discussions with developers, utilities, and hyperscalers for new baseload generation and behind-the-meter projects using natural gas, biomass, and coal.”

To skeptics who argue that financing, permitting, and the broader energy transition make completion unlikely, Johnson offered a market-based rebuttal: “There continues to be a growing need for power to meet increasing electricity demand, enable infrastructure growth, and strengthen national energy security. B&W is in various stages of discussion around several potential projects, including new natural gas plants and new coal plants.”

Johnson deferred questions on the project’s development timeline, financing structure, permitting strategy, and emissions targets to TerraSpark, which leads those workstreams. She also declined to speculate on the risk of future policy shifts: “It’s inappropriate for us to speculate on future policy changes or potential outcomes, but we look forward to helping all of our customers meet their power generation and environmental controls objectives.” TerraSpark answered most of those questions directly, as detailed below.

The Developer Fills in the Blanks

 

Responding to written questions from POWER, Tolpegin offered the most detailed public picture yet of the project’s schedule, structure, and risk strategy—while declining to release specific cost figures or name investors and customers until the FEED work is complete.

Concerning the schedule, Tolpegin said the FEED study is underway and that commercial operation of the first 400-MW unit is targeted for 2030, with the remaining three units following on a phased basis. A final investment decision (FID) will follow completion of FEED and the parallel permitting, offtake, and financing workstreams. “The project is being developed on a stage-gated basis,” he said, “with each phase designed to retire technical, cost, and execution risk before the next gate.”

On the money, Tolpegin was more guarded. TerraSpark, he said, is a privately capitalized company funding its non-federal cost share through a combination of private equity commitments and strategic partner contributions, but—“consistent with standard practice for private companies at this stage”—it is not disclosing individual investor identities or capital-structure details. Total project capital cost and levelized cost of electricity (LCOE) are still being refined through FEED and were not released. Tolpegin said the project is being engineered to be “cost-competitive with other firm, dispatchable, low-carbon generation options in the region,” and that TerraSpark is weighing a full range of financing structures—including DOE loan programs, private project finance, and strategic partnerships—to be matched to the cost and risk profile that emerges from FEED.

The plant is not envisioned as a merchant generator. Tolpegin described the campus as “an integrated industrial platform” and said TerraSpark is in active discussions with industrial offtakers, including parties with significant load growth in the region, and anticipates “a contracted structure for the substantial majority of the plant’s output.” He declined to name counterparties. The economics, he stressed, hinge on combining reliable low-carbon power with captured CO2 and downstream industrial uses on a single site.

Perhaps the most detailed answer concerned the fate of the captured carbon. Rather than relying on a single outlet, Tolpegin said the project is pursuing a deliberately diversified CO2 management strategy across four “proven and emerging pathways.” He noted Advanced Resources International is “leading the technical characterization, permitting strategy, and sequencing across all four pathways as part of the FEED scope.” Dedicated geologic sequestration into Class VI injection wells is the “anchor” that will handle the majority of captured CO2. Tolpegin highlighted a regulatory advantage: West Virginia holds Class VI primacy, meaning injection-well permits are administered by the state Department of Environmental Protection rather than the U.S. Environmental Protection Agency—which he called “a significant schedule and execution advantage.” The project also plans to sequester CO2 in unmineable Appalachian coal seams, pursue enhanced oil recovery where economics align, and—in later phases—route some captured carbon into on-site manufacturing as industrial feedstock.

Carbon capture, Tolpegin said, is integrated into the plant from day one rather than bolted on later, with Mantel’s molten borate system connected directly to the boiler flue-gas output. The plant is designed for 95% to 98% CO2 capture, with a targeted emissions intensity of 20 kilograms (kg) to 50 kg CO2/MWh equivalent—a fraction of the roughly 800 kg/MWh to 900 kg/MWh typical of an uncontrolled coal unit.

As for the company itself, Tolpegin confirmed that TerraSpark is the operating brand of TerraPurus Inc., a privately held Delaware corporation “founded to develop integrated coal-based industrial energy campuses combining modern generation, integrated carbon capture, and downstream industrial use of CO2 and coal byproducts.” He said the leadership team brings “combined decades of experience in delivering large-scale energy and industrial projects,” supported by a project team that includes B&W, Sargent & Lundy, Mantel, and Advanced Resources International, but Tolpegin did not provide a list of prior projects the company or its leaders have completed.

The Kemper Question

 

The shadow over any new U.S. coal-with-capture project is Kemper—the Mississippi lignite plant whose carbon capture was abandoned in 2017 after costs ballooned past $7.5 billion. Asked directly what makes TerraSpark different, Tolpegin did not dodge the comparison. “We’ve designed the project specifically against the reasons prior efforts struggled,” he said. “We take the Kemper precedent seriously. The project is designed the way it is precisely because of it.”

His argument rests on four distinctions. On generation, the power island is a supercritical pulverized-coal plant—a mature, commercially proven technology—rather than Kemper’s first-of-a-kind lignite gasification, which never ran reliably at commercial scale. On architecture, the 1.6-GW campus is split into four 400-MW units brought online in phases, replacing Kemper’s single-unit, single-commissioning-event model with a profile that lets problems surface on a smaller increment. On the capture technology, Tolpegin argued that the novelty in Mantel’s system “is in the capture chemistry, not in the plant architecture,” which is built around standard absorber and regenerator columns, pumps, and heat exchangers long used in the chemical and refining industries. He noted this is “the opposite of Kemper, where the novelty was in the entire upstream process.” And on risk, he contrasted Kemper’s reliance on a single regulated utility’s balance sheet with TerraSpark’s “privately financed, partner-integrated” structure and stage-gated, FEED-first development.

Whether those distinctions hold up through FEED, permitting, and a multibillion-dollar capital raise is the test the project now faces. But the answers mark a notable shift: where B&W deferred and the press releases stayed high-level, TerraSpark has put a schedule, a capture target, a carbon-storage plan, and a risk thesis on the record. Much of that thesis rests on the capture technology itself—and on a partner taking a leap of its own.

Mantel’s Big Scale-Up

 

For Mantel, the project would represent a dramatic increase in scale. The company’s technology is still early in its commercial journey: its first demonstration project, at Kruger Inc.’s Wayagamack pulp and paper mill in Quebec, Canada, is sized to capture 2,000 tonnes of CO2 per year. In December 2025, Mantel announced a FEED study for its first commercial-scale deployment—a roughly 60,000-tonne-per-year capture system integrated into steam-assisted gravity drainage (SAGD) operations at a Canadian oil and gas facility, supported by Alberta Innovates. This May, the company signed a memorandum of understanding naming engineering firm Wood as its preferred partner for fired-equipment integration across commercial deployments.

Unlike conventional amine-based systems, Mantel’s technology captures CO2 directly from high-temperature processes without cooling and reheating flue gas—the largest cost driver in traditional post-combustion capture. Operating as a liquid at high temperature, the molten borate creates what the company describes as “a self-sustaining thermal loop” that recovers process heat as clean steam and yields a 99.9% pure CO2 stream ready for sequestration or utilization.

“Coal will be an important part of global electricity generation for decades to come,” Halliday said in B&W’s announcement. “Wherever coal operates, it should operate with the best emission control technologies available.”

Politics, Jobs, and Open Questions

 

The project has enthusiastic political backing. Rep. Riley Moore (R-W.Va.), whose district would host the facility, thanked President Trump for “being the first president in my lifetime to prioritize America’s coal industry,” and said the project “sends a message that America’s future is coal-fired.” Moore’s office said the campus would bring nearly 1,000 high-quality jobs to the district; TerraSpark estimates roughly 500 permanent jobs at full buildout, plus hundreds of construction jobs.

Beyond the power block, the developers envision a multi-industry campus that includes a 10-acre West Virginia Coal Innovation and Training Center, managed by West Virginia University, focused on rare earth extraction, advanced materials, and waste-to-value initiatives.

Key numbers still sit behind the FEED process. TerraSpark has not released a total capital cost or LCOE estimate, named its equity investors, or identified the industrial customers it expects to take most of the plant’s output—all of which Tolpegin said would come into focus as the project approaches FID. The $18.5 million DOE award funds development work, not construction, and the much larger capital raise lies further down the road, though the DOE’s openness to loan guarantees under this program offers one potential path.

For now, the consortium will advance engineering, permitting, and development in coordination with the DOE, state and local leaders, and community stakeholders. Whether TerraSpark becomes the proving ground for a new generation of low-emission coal—or another ambitious project that stalls before steel goes in the ground—will be one of the more closely watched storylines in U.S. generation development.