US power plants could shed 530 million tonnes of CO2 a year by retrofitting all eligible natural gas combined-cycle (NGCC) and coal facilities with amine-based carbon capture, according to a University of California, Berkeley-led study published June 15 in Nature Sustainability.

That figure — equivalent to more than one-third of the US power sector's current annual emissions — hinges on the federal government continuing to assign a high monetary value to carbon. At US$190 per tonne, the social benefits of carbon capture and storage (CCS) outweigh the costs regardless of solvent choice. At US$6 per tonne, the costs prevail.

What does the model actually measure?

Wilson H. McNeil of UC Berkeley's Energy & Biosciences Institute led the team, with co-authors Robert A. Harley, Chelsea V. Preble and Corinne D. Scown. They ran four components through 2050:

• Which of the 479 coal-fired and 1,971 NGCC generators meet age and size criteria for retrofits
• How the regional grid mix shifts under National Laboratory of the Rockies standard scenarios
• Plant-level emissions changes after CCS
• The monetized human health and greenhouse gas impacts

The hardest constraint is plant age. The average US coal plant is 45 years old; the average NGCC plant is 22.

• 73% of coal CO2 emissions come from plants that do not meet even the loosest CCS criteria (<30 years, >100 MW)
• Only 3% of NGCC CO2 emissions come from plants that fail those criteria
• 97% of NGCC emissions are addressable versus 27% for coal

The team assumed retirements at 52 years for coal and 40 for NGCC, per Lawrence Berkeley National Laboratory data.

Where are the health wins — and losses?

Coal retrofits deliver the largest cobenefits, because plants must install flue gas desulfurization (FGD) to drop SO2 below 1 ppmv before amine solvents can function. The trade-off is ammonia.

First-generation monoethanolamine (MEA) solvents degrade oxidatively, releasing NH3 that combines with sulfates and nitrates to form secondary PM2.5. The study quantifies the imbalance:

• Coal plants with MEA: 25–33% decrease in premature mortality from air pollution
• NGCC plants with MEA: 12% to 4× increase in premature mortality
• Coal plants with CESAR1, a second-generation blend of 2-amino-2-methyl-1-propanol and piperazine: up to 67% decrease

NGCC plants already emit little SO2, so CCS gives them almost no air-quality upside, while even small NH3 increases register as proportionally larger health burdens, especially in the Northeast.

Why does solvent choice swing the math?

CESAR1 carries an NH3 emission factor of roughly 0.1 mg/kWh — more than an order of magnitude below the 82 mg/kWh used for MEA, which the authors characterize as an upper-bound based on 2010 pilot data. The CESAR1 figure derives from limited National Energy Technology Laboratory operations at a residue fluid catalytic cracking refinery.

A 2025 study in Industrial & Engineering Chemistry Research reports that an acid-wash configuration now piloted at a cement plant can cut NH3 below 10 parts per billion. The McNeil team did not model those costs.

How does the social cost of carbon flip the result?

The study ran three carbon values the US federal government has used over the past decade:

• US$190/tonne (EPA 2024): benefits exceed costs for any solvent
• US$51/tonne (interim 2021): benefits still exceed costs
• US$6/tonne: with MEA, costs exceed benefits in every study year

EPA's 2024 regulatory impact analysis raises the value to US$310/tonne by 2050. McNeil and Scown write that assigning a value at the low end of federal practice "can underestimate the societal payoff of CCS."

What decides what happens next?

Postcombustion amine scrubbing has run in refineries and chemical plants for decades, but no US power plant has adopted it at scale. The 11% parasitic load on NGCC and 16% on coal required to regenerate the solvent narrows the pool of plants where capital pencils out. Under a mid-case renewable scenario, coal emissions fall 79% by 2050 through retirements alone, while NGCC emissions fall only 7% — leaving gas as the durable CCS target.

The next milestones to track: any revision to the social cost of carbon in EPA rulemakings, and the pace at which NGCC developers pilot CESAR1 with acid-wash controls. The 1,971-unit NGCC fleet remains the most likely deployment corridor, with 530 Mt of CO2 per year on the table if economics and emissions controls both line up.