TerraSpark wins $18.5m DOE award for West Virginia coal-carbon capture campus
DOE's smallest coal-revival grant backs a 1.6-GW greenfield project with unproven molten borate capture, targeting 2030 startup to serve rising data center demand.
The Department of Energy has put $18.5 million toward the TerraSpark Energy Campus, the smallest of four Topic Area 1 grants under its $350 million coal-revival program, backing a 1.6-GW greenfield coal project in Grant County, West Virginia that pairs Babcock & Wilcox supercritical boilers with Mantel Capture's molten borate carbon capture system. Developer TerraSpark, legally TerraPurus Inc., announced the selection on June 4 with a 2030 startup target and a 95% to 98% CO2 capture design.3
US electricity demand is rising on data centers, manufacturing, and electrification, and the DOE is now treating coal with carbon capture as part of the answer. 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 — modest seed funding for a plant that would cost multiples of that to build.3
The $525 million program, launched in September 2025, implements President Trump's executive orders on clean coal and grid reliability. It sits alongside a separate $500 million Defense Production Act package the DOE has committed to support 13 coal-fired power plants and new coal export infrastructure, including up to $425 million for 12 projects strengthening the existing coal fleet.3,2
TerraSpark's $18.5 million is dwarfed by the other Topic Area 1 awards. 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 capture. AES Warrior Run received $78 million. The size gap reflects TerraSpark's early stage: it is still in scoping and design, not engineering.3
The greenfield nature of the project sets it apart from every other award in the program. Retrofitting an existing plant has known costs and existing grid connections; building a new 1.6-GW coal plant with 95% capture from scratch does not. Developers have discussed pairing new coal with capture for years. Few have reached even a funded design phase.3
Mantel Capture's molten borate system has not been demonstrated at commercial scale. A 1.6-GW plant with near-total CO2 capture would be among the largest such demonstrations anywhere if it gets built. Cost overruns on carbon capture projects are routine, and TerraSpark's four-pathway plan for the captured CO2 includes utilization and storage routes that remain unspecified.3
For power market traders and analysts, the award is less about near-term dispatch than about policy direction. New coal in the United States was considered politically closed for roughly a decade. The TerraSpark selection, combined with the Defense Production Act coal funding, marks a concrete shift in federal posture toward baseload generation, even with 2030 as the earliest possible online date.3,2
The DOE is pursuing parallel bets on other advanced generation technologies. Idaho National Laboratory has partnered with NVIDIA on the Prometheus project, using artificial intelligence to cut reactor build times by up to 50% and reduce operating costs by a similar margin, with the DOE committing $293 million through a competitive program. Yet the broader SMR sector has struggled: after supply chain disruptions and technical difficulties across the early 2020s, it remains unclear whether small modular reactor development is advancing at the pace once anticipated.1,4
Canada is moving separately on nuclear. The seven Williams Treaties First Nations are becoming minority owners of the Darlington SMR project through the Indigenous Loan Guarantee program, a political development that may affect project financing and social license rather than near-term capacity.5
The harder test for TerraSpark comes well after this design phase. The $18.5 million in federal money is seed capital; actual construction of a 1.6-GW plant with commercial-scale carbon capture would require hundreds of millions more. The DOE has not committed to further funding beyond scoping, and the economics depend on both future power prices and a CO2 utilization and storage market that remains thin.
The next concrete signal comes from the AES Puerto Rico Guayama FEED study, which carries the largest single award in the program and a clearer existing-asset case. If those engineering numbers come in over budget, coal-with-capture economics face another round of skepticism across the sector. If TerraSpark reaches a final investment decision before 2030, the US baseload generation calculus shifts in a way few would have predicted five years ago.3