Sulfuric Acid Mist: Regulating Uncertainties

Although sulfuric acid mist has been recognized as an air
pollutant for decades, it only emerged as a significant problem for the utility
industry in the early 2000s.[4]
In 2001, after the General James M. Gavin Power Plant installed a type of nitrogen
oxide controls called selective catalytic reduction devices (SCRs), sulfuric
acid mist emissions unexpectedly spiked from 9000 to 11,000 pounds per day to
allegedly more than 64,000 pounds per day.[5]
In Cheshire, Ohio, a small village of 200 people in the shadow of the Gavin
plant, residents reported asthma-like symptoms and noted corrosion and
discoloration of paint on cars and houses, as blue plumes of sulfuric acid
periodically drifted through the village.[6]
The owner of the plant, American Electric Power (AEP), eventually paid $20
million to buy out most of Cheshire. A decade later, the village remains mostly
empty.[7]

The utility industry responded to the Gavin incident by investing
significant time and money to study the sulfuric acid mist problem.[8]
EPA has also responded by paying closer attention to sulfuric acid mist from
power plants and bringing a handful of sulfuric acid mist enforcement actions.[9]
Current and future enforcement cases involving sulfuric acid mist pose a number
of challenges. In cases brought under the Clean Air Act’s (CAA) Prevention of
Significant Deterioration of Air Quality (PSD) Program, utility companies,
regulators, and courts may struggle to determine what emissions limits and
controls are required for sulfuric acid mist. This struggle is based on
uncertainties about the precise conditions under which sulfuric acid mist
forms, how it can be controlled, how emissions can be monitored, and most
importantly, at what emissions levels it poses a threat to human health and the
environment.

The Gavin incident and subsequent studies have dramatically improved
the utility industry’s understanding of sulfuric acid mist. Sulfuric acid mist
emissions strongly correlate with the sulfur content of coal: the higher the
sulfur content, the higher the sulfuric acid mist emissions.[10]
But the precise circumstances that result in the formation of sulfuric acid
mist have been much more difficult to unravel. Experts believe that vanadium
and other constituents in coal may increase sulfuric acid mist formation.[11]
In addition, boiler design and oxygen levels in the flue gas appear to
influence sulfuric acid mist formation. High temperatures in boilers increase the
formation of sulfuric acid mist with the mist forming at the highest levels in
a temperature band above approximately 800 degrees.[12]
Finally, ambient conditions, including wind and water content in the air, also
influence sulfuric acid mist formation and its impacts.[13]
This means that even if all other factors remain constant, weather conditions
may result in higher or lower ambient concentrations and can increase the risk
of human exposure to sulfuric acid mist.

The uncertainties and complexities associated with sulfuric
acid mist are further compounded by its relationship to other pollutants. Most
troubling, as discovered at the Gavin plant, there is a clear relationship
between the use of SCRs to reduce nitrogen oxide emissions and increases of
sulfuric acid mist.[14]
A study of power plants equipped with SCRs found that 98 percent of the plants
were expected to emit sulfuric acid mist at levels above 5 ppm, a level that
might result in environmental impacts.[15]
But enforcement actions brought against plants that have installed SCRs raise
the troubling specter of potentially penalizing utilities for their efforts to
reduce their environmental impact. In exercising enforcement discretion,
regulators may be forced to balance the need to reduce nitrogen oxides and
their regional impacts with the need to protect communities from the more
localized impacts of sulfuric acid mist.[16]
And as with any pollutant from power plants, industry, government, and the
public must weigh the environmental and health benefits of sulfuric acid mist
control on one side, versus energy supply and demand and the potential
increased cost of electricity on the other. This Article briefly outlines the
scientific and legal complexities facing the utility industry and environmental
regulators in developing sulfuric acid mist control strategies. Next, it
compares the economic and environmental tradeoffs of different control
strategies. Finally, it recommends control strategies that provide the utility
industry operational flexibility while ensuring that human health and the
environment are protected from sulfuric acid mist and recognizes that there may
not be a one-size-fits-all solution to reduce sulfuric acid mist emissions at
power plants.

I. Regulating Sulfuric Acid Mist Under the Clean Air Act

A. Prevention of Significant Deterioration Provisions

One of the most important avenues for regulating sulfuric
acid mist under the CAA is through the New Source Review (NSR) program.[17]
Under NSR, major new and modified stationary sources in areas that are
unclassifiable or that meet the National Ambient Air Quality Standards (NAAQS) are
subject to PSD permitting.[18]
Notwithstanding certain exceptions, NSR and PSD are triggered either by new
construction, or a physical change or change in the method of operation of an
existing facility that results in a significant net increase of emissions of a
pollutant. The threshold for “significant increase” varies by pollutant. For
sulfuric acid mist, the threshold is an increase of 7 tons per year.[19]
Facilities subject to the PSD program must submit a PSD permit to the
permitting authority and implement Best Available Control Technology (BACT) for
each regulated pollutant. Thus, unless an exception applies,[20] any power plant that makes a
physical or operational change to its plant resulting in an increase of
sulfuric acid mist emissions of more than 7 tons a year must obtain a PSD
permit and apply BACT to limit sulfuric acid mist emissions.

B. Best Available Control Technology

BACT is an emissions limit “based on the maximum degree of
reduction of each pollutant subject to regulation under” the CAA “on a
case-by-case basis, taking into account energy, environmental, and economic
impacts and other costs,” that the permitting authority “determines is
achievable.”[21]
EPA recommends, and most permitting agencies apply, a top-down BACT analysis
that ranks all available control technologies for a regulated pollutant in
descending order of effectiveness.[22]
Following this approach, the most stringent control alternative is selected
unless technical considerations, energy, environmental, or economic impacts
lead the permitting authority to conclude that it is not “achievable.”[23]
Because BACT is assessed on a case-by-case basis, it may differ significantly
from one power plant to another power plant. Geography, fuel sources and types,
and plant configuration can impact the technical feasibility of controls and the
cost effectiveness of controls.[24]
Not surprisingly, permitting authorities, the utility industry, and the public,
which can comment on PSD permits,[25]
may disagree over how technical, environmental, economic, and site-specific factors
should influence BACT determinations.

C. Control Technology Available for Sulfuric Acid Mist

One of the most effective control options for sulfuric acid
mist is a wet electrostatic precipitator (WESP), a particulate control device that
removes particles, including sulfuric acid mist, from flue gas by using an
electrostatic charge.[26]
WESPs are extremely efficient at removing sulfuric acid mist, but they can cost
$50 million to $200 million and require a significant amount of energy, up to 0.5
percent of the plant’s gross output, to operate.[27]
Because BACT analyses require economic assessment of control options, including
a calculation of removal costs on a per ton basis, utilities may effectively
argue that a WESP is not required under BACT if the amount of sulfuric acid
mist removed is less than several thousand tons per year. There is no bright-line
rule for per ton removal costs under BACT, but costs above five or six thousand
dollars per ton for controls have been referenced as approaching the upper
limit of the threshold of economically feasible technology required under BACT.[28]

A wet electrostatic precipitator. Image credit to Hitachi Plant Technologies, Ltd.

The current preference of the utility industry for sulfuric
acid mist control appears to be sorbent injection, an option that is much more
economical than WESPs in the short term. Dry sorbent injection uses nozzles to
spray a dry powder, typically magnesium, lime, or trona (a sodium-based
mixture), into the flue gas.[29]
The sorbent binds with the sulfuric acid mist and removes it from the flue gas
stream. But there are limits on the use of sorbent control. Excessive use of
sorbents can clog equipment, so power plant engineers may need to experiment
with different levels of sorbent injection. They may also need to balance
sorbent injection with other control methods, including configuration changes
that increase the amount of time sulfuric acid mist remains in the stack.[30]
The longer the residence time for sulfuric acid mist in the stack, the more
opportunity the sulfuric acid mist has to bind with sorbent.[31] Facilities can also maximize sulfuric
acid mist control if they mill sorbent into smaller particles that increase the
surface area of sorbent and improve its potential to capture sulfuric acid mist.
However, even with these measures, there remains a saturation point beyond
which increasing the amount of sorbent injected will not further reduce the
amount of sulfuric acid emissions. Other plant improvements, including
installation of low catalyst SCRs,[32]
which reduce, but do not eliminate the impacts of SCRs on sulfuric acid mist formation,
or switching or blending fuel with low or medium sulfur content coal,[33] may be used to supplement
sorbent injection.

Baghouses, which are large filters
designed to capture soot and other particulates,[34]
can also reduce sulfuric acid mist emissions. One study indicates that
baghouses can remove up to 90 percent of sulfuric acid mist.[35]
As with WESPs, however, installing baghouses can be a significant capital
expenditure.[36]
Utilities may balk at the expense and argue that the technology is not economically
feasible under a BACT analysis.

Improvements in sorbent control may be increasing
regulators’ and industry’s confidence that sorbent injection, while not
achieving the same reductions in sulfuric acid mist as WESPs, can reduce
sulfuric mist emissions to levels that are sufficient to protect human health
and the environment at a fraction of the cost.[37]
The effectiveness of sorbent injection, however, may depend on proper
calibration and maintenance of sorbent injection rates over time and a consistent
fuel source. If these inputs are not constant, sulfuric acid mist emissions
could spike. To mitigate the possibility of fluctuations in sulfuric acid mist
emissions, operators need to build in some compliance headroom by ensuring that
day-to-day emissions of sulfuric acid mist are marginally lower than levels
that could result in opacity problems or violate permit limits. Once a control
strategy has been adopted, power plants and enforcement authorities need to
monitor the effectiveness of the controls over time and in different operating
scenarios. Some power plants may need to continue to experiment with a variety
of controls to find a solution that provides the best balance of sulfuric acid
mist reduction, control of other pollutants, and power plant performance.

D. Opacity Violations

The appearance of the tell-tale blue plume of sulfuric acid
mist from the stack of a power plant often indicates a different violation of
the CAA. In addition to, or in lieu of, claims brought under the CAA’s PSD provisions,
EPA may bring claims against power plant owners and operators for opacity
violations under the Act’s New Source Performance Standards (NSPS) (Section 111
of the Act) or the applicable State Implementation Plan.[38]
“Opacity means the degree to which emissions reduce the light and obscure the
view of an object in the background.” 40 CFR § 63.2. At 100 percent opacity,
no light is visible through a plume. At zero percent opacity, a plume is completely
transparent. While opacity is not itself a pollutant, it serves as a surrogate
for particulate matter pollution, including sulfuric acid mist pollution, from
power plants.[39]

The NSPS for fossil-fuel-fired steam generators provide that,
for power plants constructed after August 17, 1971, gases emitted from the
facility cannot “exhibit greater than 20 percent opacity except for one
six-minute period per hour of not more than 27 percent opacity.”[40]
For facilities not subject to the NSPS (built prior to August 1971), opacity
limits can vary depending on the applicable State Implementation Plan. In
Kentucky, for example, facilities are not to exceed 40 percent opacity except
for one six-minute period per hour of not more than 60 percent opacity.[41]
In Texas, older facilities cannot exceed 30 percent opacity averaged over a six-minute
period.[42]

Opacity problems associated with sulfuric acid mist can
occur with emissions as low as 3-4 ppm.[43]
The utility industry reports that sulfuric acid mist concentrations of only 10
ppm can result in opacities greater than 40 percent, the upper opacity limit
for many older power plant units.[44]
Other industry guidance indicates that opacity problems can occur at sulfuric
acid mist concentrations above 5-15 ppm.[45]
The impacts of sulfuric acid mist emissions on stack opacity can fluctuate
depending on operating conditions. Opacity monitors can provide accurate opacity
data and send an immediate warning to plant operators if limits are exceeded. But
many power plants have had to install wet scrubbers to address sulfur dioxide
emissions, which makes it difficult or impossible to monitor opacity within
their stacks. As a result, many plants have either removed their opacity
monitors or replaced them with particulate monitors. Continuous sulfuric acid
mist monitors, which would best address the current monitoring problem, are
still under development.[46]

As a result of the unavailability of effective monitoring
devices for sulfuric acid mist, in many instances the only method to determine
opacity at a power plant is through visual observation of the plume. Method 9
readings, which rely on the judgment of a trained inspector, are the most
common method of visual opacity assessments.[47]
Although inspectors are typically experienced and well trained, Method 9 is
subject to judgment, memory, and the human eye.[48] Inspectors record 24 consecutive
observations (typically in six-minute increments) and average the results. An
accurate opacity test requires ideal weather conditions because the plume
cannot be clearly observed on cloudy days. The difficulty and uncertainty
associated with visual opacity readings are a cause for concern for both
regulators and the industry.[49]
On the one hand, regulators have to undertake opacity readings onsite and in
clear weather to document opacity violations. On the other hand, an opacity
reading made by a regulator visually assessing real time emissions from memory
is difficult for a defendant to refute.

If a pattern of opacity violations at a power plant can be
established, EPA or state environmental authorities may argue that significant
controls are required to eliminate the visible sulfuric acid mist plume.
Enforcement authorities may obtain civil penalties for each day a plant exceeds
opacity limits, or obtain significant injunctive relief that, in some cases,
approaches a BACT-like remedy.

II. Human Health and the Environment and Emissions Limits

Perhaps the thorniest issue with regard to regulating
sulfuric acid mist is determining at what levels emissions of sulfuric acid
mist threaten human health and the environment. Studies indicate that sulfuric
acid mist can impact the health of children with asthma at 70 micrograms per
cubic meter, and can impact normal adult lung function at 100 micrograms per
cubic meter.[50]
The impact on health, however, is a function of exposure duration, individual
sensitivity, and exposure to other air contaminants.

West Maple Street at Route 7, Cheshire, Ohio, 2002. Photo credit to Franz Jantzen. Mr. Jantzen photographed various locales in Cheshire, Ohio once a year between 2002 and 2004, when American Electric Power bought up much of the land in the the town in response to complaints by residents of severe sulfuric acid mist pollution. Mr. Jantzen returned one final time in 2009. NPR and Architizer both published blog posts on Jantzen's Cheshire, Ohio Project.

In the early 2000s, toxicologists from the Agency for Toxic
Substances and Disease Registry evaluated the impacts of sulfur dioxide and
sulfuric acid mist on the village of Cheshire, Ohio, near the Gavin power
plant, and concluded that emissions posed a public health hazard to some
residents.[51]
The highest officially recorded levels of sulfuric acid mist in Cheshire were
approximately 120 micrograms per cubic meter of air, but there were unofficial
reports of levels as high as 200 micrograms per cubic meter of air.[52]
It was difficult for investigators to determine the duration of individual
exposure to sulfuric acid mist, but exposures ranged from several minutes to
several hours.[53]
Residents in Cheshire were also exposed to high levels of sulfur dioxide and
metal oxide particulates and investigators indicated that the presence of these
and other co-contaminants might also have had health impacts.[54]

Following the lawsuit brought by Citizens Against Pollution,
AEP agreed to emissions limits of 14 ppm of sulfuric acid mist at the Gavin
plant. By comparison, in the only settled case to date in which EPA has
directly addressed sulfuric acid mist, involving the Hoosier Energy Company,
the parties agreed to limits of approximately 2.5 ppm (.007 lb/mmBTU).[55]
But this limit was only one part of a larger settlement that required a number
of significant improvements to the facility at a total cost of $250 million to $300
million.[56]
Additionally, Hoosier had a lengthy compliance period, almost two years, to
meet the sulfuric acid mist emissions limit, and an additional year before it
was subject to stipulated penalties to meet these limits.

Although there appears to be some consensus in the utility
industry that emissions of sulfuric acid mist above 5 ppm may result in opacity
problems, many power plants may be disinclined to agree to similarly low
limits. This hesitation is in part a function of the difficulty of measuring
sulfuric acid mist.[57]
Although there is an accepted EPA methodology for stack testing for sulfuric
acid mist, the industry has expressed concern that this method does not provide
accurate results.[58]
Sulfuric acid mist emissions also may fluctuate between stack tests. While sorbent
controls may ensure compliance with a 5 ppm limit most of the time, ambient
conditions or variations in fuel could result in higher emissions.[59]
Both EPA and the industry are concerned that no continuous emissions monitoring
device is commercially available for sulfuric acid that has the requisite
sensitivity to detect changes of less than 1 ppm in the stack.

In the absence of recurrent, visible opacity problems, power
plants may remain unaware of the potential significance of sulfuric acid mist emissions
until stack testing can be performed, at a relatively high cost, on a quarterly
or biannual basis. Infrequent stack tests threaten both utility operators and
regulators. Utility operators may worry that one high stack test could be used
as evidence of continuous non-compliance with emissions limits, while
regulators may be concerned that stack tests do not provide adequate monitoring
of sulfuric acid mist emissions and could fail to identify non-compliant
facilities.

III. Uncertainty and Balancing

As EPA seeks more stringent regulation of other air
pollutants, the utility industry and regulators will need to keep close tabs on
sulfuric acid mist emissions. Perhaps the most vexing problem for regulators
and industry alike is the uneasy relationship between sulfuric acid mist
control and control of nitrogen oxides. A significant sulfuric acid mist
problem first emerged, at Gavin and elsewhere, with the adoption of SCRs used
to reduce nitrogen oxide emissions. When SCRs are combined with high-sulfur coal,
as they were at the Gavin plant, sulfuric acid mist emissions can increase
dramatically.

In response to the now decade-old problem with SCRs, the
utility industry developed low-acid conversion catalysts that reduce the
sulfuric acid conversion rate.[60]
But SCRs still involve a trade off with generally higher emissions of sulfuric
acid mist.[61]
Power plants have sought to compensate for the higher sulfuric acid mist
emissions with many of the control technologies described above, but several of
these strategies, including WESPs and baghouses, may not be economical if used
solely to control sulfuric acid mist. Sorbent injection, while typically the
most economically feasible control method for sulfuric acid mist in the short
term, may be insufficient to control very high levels of sulfuric acid mist
emissions.

The LIFAC (limestone injected into the furnace with activation of untreated calcium oxide) sorbent injection desulfurization process. Image credit to the US Department of Energy, Office of Fossil Energy.

Following the recent regulations for hazardous air
pollutants, including mercury and acid gases, power plants may increasingly
turn to sorbent injection.[62]
The need to control mercury and acid gas may have a positive impact on sulfuric
acid mist control because power plants may not only invest in superior sorbent
injection systems, but may also install baghouses that increase the
effectiveness of sorbent controls.[63]
Additional controls that target sulfur dioxide, including scrubbers, should
also serve to reduce sulfuric acid mist emissions.

In the current regulatory landscape, optimizing the
performance of coal-fired power plants and ensuring that they run efficiently
within permit limits is increasingly difficult. Finding the operational “sweet
spot” at power plants may require a significant investment of time and money. In
order to reduce sulfuric acid mist emissions and balance other regulatory
requirements, enforcement authorities and the utility industry must develop
control strategies tailored to individual power plants. Regulators must allow
time for calibration of sulfuric acid mist and other air pollutant control
strategies and stack testing. For their part, plant operators must think ahead
and consider the impacts of implementing control schemes for multiple
pollutants. In an environment of regulatory uncertainty, it may be worthwhile
for the utility industry to consider adopting conservative control strategies
with multi-pollutant control benefits that increase operational flexibility,
reduce the risk of future non-compliance, and anticipate the possibility of
stricter, long-term emissions limits. The high costs of control equipment, complex
maintenance and operation schedules of power plants, and threat of future
enforcement actions leave little room for trial and error.