“The climate conversation rarely pauses to ask why the scrubber success is treated as invisible. One reason is rhetorical convenience: once the entire fuel is defined as the problem, partial technological answers become unwelcome.”
In the climate debate, coal is treated as if it still belches the same smoke it did in 1970. This is a major error. Modern coal plants combine a variety of technologies before and after combustion, as well as monitoring systems to verify compliance with today’s environmental laws with criteria pollutants (see Appendix A). As with petroleum, coal today is an environmental product, part of the increasing sustainability of fossil fuels.
Scrubbers, selective catalytic reduction, and the rest of the air-pollution control suite (see Appendix B) are rarely mentioned. That silence is not neutral. It erases one of the clearest engineering successes of the past thirty years and flattens a complicated fuel into a single villain.
Modern coal plants are not the plants of the acid-rain era. Flue-gas desulfurization systems—scrubbers—routinely remove 95 to 99 percent of sulfur dioxide (SO2). Vendors now offer wet limestone systems guaranteed to achieve over 99 percent removal with high availability.
Dry and semi-dry systems have also improved; what was once 80–90 percent is now routinely engineered toward 97–98 percent on suitable coals. Electrostatic precipitators and baghouses capture more than 99 percent of particulate matter (PM); hybrid systems push fine-particle removal still higher.
Selective catalytic reduction systems cut nitrogen oxides (NOx) by 80–90 percent or more when properly operated. Activated-carbon injection and the co-benefits of the other controls have driven mercury reductions well into the high double digits on many units.
These are not laboratory numbers. As of 2022, roughly 82 percent of the remaining U.S. coal fleet carried flue-gas desulfurization, and nearly 58 percent had selective catalytic reduction. Nearly the entire fleet has some form of NOx control.
The result is visible in the emissions record. Under the Acid Rain Program and its successors, power-sector sulfur dioxide emissions fell by roughly 95–96 percent from 1990 levels, while nitrogen oxides fell by about 89–90 percent.
Those cuts occurred while the economy and electricity demand grew for long stretches. Wet sulfate deposition—the marker of acid rain—dropped more than 70 percent. The air got cleaner because engineers installed and operated the hardware, not because coal was abolished.
Examples
Specific plants make the point clearer than fleet averages.
At Ameren’s Duck Creek, Coffeen, and Sioux stations, Hitachi wet limestone forced-oxidation scrubbers that came online between 2009 and 2011 were designed for 99 percent SO2 removal. Performance tests confirmed it: Duck Creek reached 99.83 percent, Coffeen 99.95 percent, and Sioux 99.65 percent, with particulate emissions well below 0.015 lb/MMBtu and strong co-removal of oxidized mercury.
Those units ran on high-sulfur bituminous coal blends and still hit the numbers without organic-acid additives or exotic packing. Prairie State Generating Station in Illinois, a pair of supercritical units burning local bituminous coal, was designed around SCR, dry and wet electrostatic precipitators, and wet FGD.
Early years showed SO2 removal near 98.6–98.8 percent; the plant continues to operate with multi-pollutant controls that also capture fine particulates and mercury. In New Hampshire, Merrimack Station’s single scrubber serving both units cut mercury 97–98 percent and SO2 96–98 percent after it entered service in 2011–2012—well above the 80 percent mercury target set by state law and ahead of schedule.
Southern Company’s Gorgas, Hammond, and Miller plants achieved 99 percent or higher SO2 removal and 100 percent reliability during trial operation with Mitsubishi systems. Babcock & Wilcox installations at Detroit Edison’s Monroe units, Progress Energy’s Asheville Station, and multiple GenOn plants were routinely designed and demonstrated at 97–98 percent SO2 removal while producing saleable gypsum.
Big Investments Pay Off
The investment was large. Cumulative capital spending on NOx, SO2, particulate, and mercury controls at U.S. coal plants from 2000 onward has been estimated in the tens of billions of dollars. Parasitic load is real but limited: a wet scrubber typically consumes 1–2 percent of a plant’s gross output.
That is not trivial, yet it is also not the plant-killing penalty sometimes implied. Operators have learned to optimize liquid-to-gas ratios, slurry chemistry, and fan power so that the energy penalty does not offset the environmental gains.
None of this means coal is carbon-free. It is not. Carbon dioxide (CO2) is a different molecule, and the physics of combustion do not disappear because sulfur is scrubbed. Energy humanism does not require pretending otherwise.
It does require distinguishing criteria pollutants—sulfur dioxide, nitrogen oxides, particulates, lead —from greenhouse gases, and recognizing that the former has been managed at scale.
In contrast, CO2 emission abatement remains the sole focus of policy. Treating every coal plant as an uncontrolled 1970s relic is simply inaccurate. The remaining U.S. fleet, still on the order of 168–180 GW depending on the exact inventory and date, is overwhelmingly equipped with modern controls. Many of the uncontrolled or lightly controlled units have already retired or switched to low-sulfur Powder River Basin coal.
Reliability still matters. Coal plants store months of fuel on site. That feature has repeatedly shown its value during winter storms when natural-gas deliverability tightens, and intermittent sources cannot fill the gap. Capacity accreditation in markets such as PJM reflects the difference: coal units receive far higher firm capacity credit than wind or solar.
Conclusion
With data-center and industrial load growth now material, and with interconnection queues measured in years, retiring dispatchable capacity that already has its in-place environmental controls is not a free environmental lunch. It is a reliability and cost decision whose consequences fall on ratepayers and, in extreme cases, on people who lose power.
The climate conversation rarely pauses to ask why the scrubber success is treated as invisible. One reason is rhetorical convenience: once the entire fuel is defined as the problem, partial technological answers become unwelcome.
Another is that carbon dioxide has no comparable scrubber in terms of maturity and cost. Carbon capture remains expensive, energy-intensive, and sparsely deployed. That reality does not license the claim that air quality has not improved, or that further criteria-pollutant reductions are impossible. It simply means the next engineering problem is different from the last one.
Energy sanity starts with the inventory that already exists. The plants that remain can continue to run with the controls they have, or with incremental upgrades that squeeze out a few more points of removal efficiency or reduce the parasitic load. Some will be needed longer than current retirement schedules assume if load growth continues and gas-plant build rates lag. Others will close when economics or remaining life dictate. Neither outcome requires rewriting the last three decades of scrubber performance out of the record.
The same engineering habit that produced 99 percent sulfur removal can be applied elsewhere—nuclear, high-efficiency gas, advanced coal cycles where they make sense, and whatever dispatchable resources actually keep the lights on. What it cannot tolerate is the selective amnesia that treats coal’s air-pollution controls as if they never happened. The scrubbers worked. The theme of the increasing sustainability of fossil fuel is alive and well.
1. Particulate Matter (PM) Control
2. Sulfur Dioxide (SO₂) Control
3. Nitrogen Oxide (NOₓ) Control
4. Mercury and Toxic Metal Control
5. Hydrogen Chloride (HCl) and Other Gas Control
6. Pre‑combustion Measures
7. Monitoring and Compliance