On May 3, 2011, EPA published the highly anticipated proposed Maximum Available Control Technology (MACT) Standard for the utility sector (40 CFR 63 Subpart UUUUU). When considered together with the growing list of proposed and final rules impacting the utility sector (Transport Rule/Cross-State Rule, Coal Combustion Byproducts Rule, Cooling Water Intake Rule, Regional Haze Rule, revised National Ambient Air Quality Standards [NAAQS], Effluent Guidelines Rule, and New Source Performance Standard [NSPS] for GHGs), it’s clear that the environmental landscape for utilities is dynamic and quite complex. With respect to the proposed MACT, EPA has estimated that 1,350 coal- or oil-fired steam electric generating units (EGU) at 525 power plants will be affected at an estimated cost to industry of $10.9 billion. The Utility MACT, in combination with other rules impacting utilities, could result in retirement of some EGUs and full closure of certain power plants. However, as part of the MACT rulemaking, EPA studied power supply and demand issues as they relate to power reliability and believes that power reliability should not be adversely impacted by this rule, assuming capacity retired by these rules can be replaced.
Among the significant history of the Utility MACT rulemaking, most noteworthy are EPA’s March 2005 rules that: 1) removed EGUs from the list of sources requiring MACTs (the “delisting” rule) and, 2) established a national cap and trade program for Hg from EGUs under the Clean Air Mercury Rule (CAMR), inclusive of performance standards for new EGUs. In February 2008, both the delisting rule and CAMR were vacated by the D.C. Circuit Court. The vacatur of the delisting rule meant that EGUs were added back to the list of sources requiring MACTs. Thus, EPA again became required to develop a MACT for EGUs (as was the case in 2000-2005). More specifically, based on a schedule that EPA negotiated with environmental groups and documented in a consent agreement, EPA was required to sign the proposed Utility MACT by March 16, 2011 (as done), and sign the final MACT by November 16, 2011 (which would make the effective date sometime in January 2012).
In the approximately three years since the Court vacated CAMR, EPA has gathered and evaluated data on which to base a MACT for EGUs. The data gathering and evaluation consisted of a three part Information Collection Request (ICR). Statistical analyses were performed on the data collected from the ICR in an effort to establish emission limits that are often referred to as emission floors. EPA’s process to establish the emission floors for EGUs was similar to the process most recently used by EPA to amend the Portland Cement (PC) MACT. When EPA published the proposed amendments to the PC MACT, there were extensive comments on the rigorous methodology EPA used to establish the floors. In short, EPA has proposed the Utility MACT using what some have dubbed the “Franken- MACT” approach – whereby the final limits are the best of best on a pollutant-specific basis and not on a best source achieved basis. The result, in effect, can produce final emission limits that no single unit in existence can meet without additional controls.
The Utility MACT applies to coal- and oil-fired EGUs, with exemptions for EGUs that have the capability to burn coal or oil, but burn these fuels in small quantities compared to other fuels combusted in the unit. A similar rule, the Industrial Boiler MACT, applies to industrial boilers that are not considered EGUs. The definition of EGU in the Utility MACT parallels the definition in the Industrial Boiler MACT, but the Utility MACT clarifies when a boiler qualifies as an industrial boiler or an EGU in situations where multiple boilers serve a common steam generator rated above 25 megawatts electric (MWe).
The Utility MACT defines an EGU as a fossil fuel-fired combustion unit of more than 25 MWe that serves a generator that produces electricity for sale. The definition also addresses the status of cogeneration units as EGUs. The Utility MACT further defines a fossil fuel-fired EGU as an EGU that is capable of combusting more than 73 MWe (250 MMBtu/hr) heat input of fossil fuel. In EPA’s view, 250 MMBtu/hr of heat input to a boiler equates to 25 MW of electric output. Thus, in situations where multiple boilers serve a common generator rated above 25 MWe, each boiler would only be subject to the Utility MACT if the boiler is capable of combusting more than 250 MMBtu/hr heat input. Otherwise, the boiler is regulated under the Industrial Boiler MACT.
Emission Limits (“Floors”)
The proposed rule would require coal-fired units to meet emission limits for mercury, hydrochloric acid (as a surrogate for acid gases), and particulate matter (as a surrogate for non-mercury metallic HAPs). Oil-fired units would also be required to meet emission limits, that address essentially the same pollutants as coal-fired units.
In addition to the proposed emission limits, EPA is proposing alternate equivalent emission standards (for certain subcategories) in three areas: SO2 (as an alternative to HCl), individual non-Hg metals (as a first alternative to PM), and total non-Hg metals (as a second alternative to PM). All sources can opt to comply with the alternative limits for PM, but only sources with scrubbers can comply with limits for SO2 as an alternative for HCl.
In the proposed MACT limits for existing and new EGUs, new EGUs refer to EGUs constructed after the date that the proposed MACT was published (May 3, 2011). All EGUs constructed before this are considered existing. The limits for new units are lower than the limits for existing units;1 the limits for lignite are higher than the limits for other coals; and the limits for pet coke2 (i.e. solid oil) are higher than the limits for other oils. See Table 1 for all of the proposed emission limits.
In establishing emission limits for the Utility MACT, EPA did not subcategorize EGUs to the same extent as the 2005 CAMR boilers subcategorized under the 2011 Industrial Boiler MACT3. In the Boiler MACT, EPA categorized industrial boilers based on the design of the combustion system, (pulverized coal, stoker, and fluidized bed) and included separate limits for each category4. Under the proposed Utility MACT, coal-fired boilers of any design, burning any kind of coal (excluding lignite and gasified coal) are subject to the same emission limits.
The emissions limits summarized in Table 1 are applicable as 30-day rolling averages, and cover all modes of operation, including startup, shutdown, and malfunction. Should a malfunction cause an EGU to exceed a limit, the MACT has specific requirements for making an affirmative defense to civil penalties.
EPA has proposed work practice standards to limit emissions of organic HAPs, such as dioxins and furans. The standards require initial and annual tune-ups of EGUs. Documentation of the tune-ups must be submitted to EPA. The requirements for the tune-ups are detailed in the MACT, and include, among other things, a requirement to test NOX and CO emissions before the tune-up and after any adjustments are made. The inclusion of tune-ups rather than numerical emissions limits for dioxins and furans is a noteworthy contrast to the 2011 Boiler MACT, where specific limits for dioxins and furans were included.
Monitoring and Testing
All affected EGUs must demonstrate initial compliance with emission limits by conducting performance testing and by either conducting fuel analyses or establishing operating limits on parameters EPA has determined are indicative of control device performance, and ultimately indicative of emissions for one or more pollutants. Ongoing compliance is demonstrated by either continuously monitoring the pollutant itself (via Continuous Emissions Monitoring Systems [CEMS]) and/or by continuously monitoring a parameter(s) that is indicative of the pollutant (via Continuous Parametric Monitoring Systems [CPMS]). The approach of using initial performance testing to establish limits for parameters that can be monitored using a CPMS is in not new to the Utility MACT. However, the Utility MACT, as proposed, appears to have some redundancy with respect to the emissions and parametric monitoring requirements, as later discussed.
With coal- and solid oil-fired EGUs, operators must conduct performance tests for total PM every five years and use a CEMS for measuring filterable PM. This is not to suggest that the emission limit for PM includes only filterable PM. The rule indicates that compliance with the PM emission limit must be demonstrated through stack testing for total PM. Further, the rule indicates that sources using a PM CEMS must demonstrate compliance by maintaining the filterable PM concentration at or below the highest 1-hour average filterable PM concentration measured during the most recent performance test. Clearly, this does not allow for any compliance margin on the unit following the performance test.
Coal- and solid oil-fired EGUs are also required to conduct performance tests for Hg every five years and to have a CEMS or sorbent trap monitoring system. Operators have the option to conduct monthly HCl stack testing in lieu of a CEMS or sorbent trap monitoring system. Liquid oil-fired EGUs must conduct monthly or bimonthly performance tests for PM, HCl, and HF establish limits on the pollutant concentrations in the fuel (for HAP metals as well as chlorine and fluorine) and comply with the fuel limits. The fuel limits will be based on the fuel analyses conducted as part of the performance tests.
The requirements of the proposed rule related to fuel analyses, emissions testing, and monitoring require further clarification. The rule allows for PM to be used as a surrogate for non-Hg metal HAPs, but then suggests that all EGUs must conduct fuel analyses. Further, except for liquid oil-fired EGUs, the rule does not appear to allow for fuel analyses alone to serve as the basis for compliance. EPA has established a number of limits on operating parameters that are based on the performance of the EGU at the time of the most recent performance test. For example, a source using a wet scrubber to comply with a limit for HCl is required to maintain the pressure drop and liquid flow rate across the scrubber at or above the lowest 1-hour average flow rate determined during the most recent performance test.
These operating permits apply regardless of whether the source operates an HCl or SO2 CEMS. Thus, an EGU CEMS could demonstrate compliance with the HCl or SO2 limit, at the same time the EGU is out of compliance with a limit on pressure drop or flow rate. It is not clear if EPA intentionally included redundancy with respect to limits on emissions and limits on parameters indicative of emissions. Another example of redundancy is on the limits for PM for coal-fired EGUs. The rule requires PM CEMS, as well as bag leak detection systems for sources that operate baghouses. It is expected that these redundancies will be reversed in the final rule.
As stringent as the proposed rule appears to be, EPA did include a number of provisions intended to increase operational flexibility. Most noteworthy is the exemption for gas-fired units that infrequently burn oil. EPA also allows the averaging of emissions from existing EGUs at a single site. A source could choose to over-control one EGU in exchange for higher emissions at another EGU. Finally, EPA is encouraging sources to take advantage of a Clean Air Act provision allowing a total of four years for bringing existing EGUs into compliance rather than the typical three years required in MACT standards. It is noteworthy that EPA is supporting the idea that an extra year may be needed for compliance due to the backlog that equipment suppliers may face in providing control technologies in the three-year timeframe (with wet scrubbers the most likely candidate for timing challenges).
One area of flexibility that EPA did not include in the proposed rule is the option to set a health based standard for HAPs with established health thresholds. EPA’s position is that health-based standards would potentially be less stringent than the MACT standards, and that they lack detailed information from subject sources and emissions near the subject sources on which to base an analysis. Thus, EPA opted not to consider a health-based option for setting limits.
Impacts to Utility Industry
About half of the electricity in the U.S. is generated from coal-fired EGUs with only about one percent from oil-fired units. Of the coal-fired EGUs expected to be regulated by the Utility MACT, EPA estimates that over 40 percent of the units do not have SO2 or NOX controls. There is general agreement that some controls used to reduce PM, NOX and SO2 can have significant co-benefits for reducing emissions of HAPs. More specifically, oxidized and particle-bound species of Hg are readily captured in PM control devices, with a baghouse being the most effective form of control. SCR devices used for NOX control may further enhance Hg capture by oxidizing elemental mercury across the catalyst. However, the degree of SCR Hg oxidation can be variable, and appears to be coal- and catalyst-specific. The majority of coal fired-EGUs operate with PM control. If the PM control is an older vintage ESP, it will likely be insufficient to reduce Hg to the levels required by the Utility MACT. If the PM control device is a baghouse, then it may be possible to reduce Hg to the required levels without adding Hg-specific controls.
As there are a number of factors that influence Hg emissions from an EGU, it is challenging to develop a control strategy that guarantees that required limits can be met. Activated carbon injection (ACI) has been demonstrated in many applications to effectively reduce Hg; thus, the demand for this material has risen sharply. However, there are also instances where activated carbon has not made a significant impact on Hg emissions. A number of these instances involved elevated SO3 concentrations in the flue gas. SO3 can reduce the effectiveness of ACI at controlling Hg. Other Hg control options that have been demonstrated as effective include injecting halogens (Cl or Br) into the furnace of the EGU or using brominated carbon sorbents. In short, baghouses and/or ACI are expected to be the most common controls installed for the purpose of meeting Hg limits and also for the purpose of meeting limits for total PM or non-Hg metals.
Acid gases control strategy will be more diverse than Hg. Coal-fired EGUs with SO2 scrubbers will likely opt to meet the alternative SO2 limit, as the SO2 limit of 0.2 lb/MMBtu should be easily achievable with a scrubber designed for SO2 control. Coal- fired boilers without SO2 scrubbers, as well as oil-fired boilers, will very likely need to consider additional controls for acid gases. This could include acid-gas specific controls (such as a scrubber designed for HCl), or it could include controls designed for SO2 that will also reduce acid gases, such as wet or dry scrubbers or dry sorbent injection. Dry Sorbent Injection (DSI) is by far the most cost effective means of reducing acid gases and is expected to be the most common control installed for the purpose of meeting acid gas limits.
As mentioned, EPA has estimated that the utility sector will spend over $10 billion to comply with the Utility MACT as proposed. Due to the cost of this rule, as well as other rules also impacting the utility sector, a number of utilities have indicated their plans to retire EGUs. The retirement of EGUs could create a limited supply of power that will serve to drive up the price of electricity. This will have far reaching impacts, not the least of which is the adverse impact it will have for energy-intensive industrial operations.
In contemplating control strategies to meet the Utility MACT, companies should ensure that all aspects of environmental permitting are considered. More specifically, prior to moving forward with a control project, a full review of the impacts of the project should be conducted, including evaluating emissions increases that may be associated with a specific control device. While it may sound counterintuitive to talk about emission increases with respect to a control device, consider an EGU that currently operates unscrubbed on low-sulfur coal. Now, consider that the EGU adds a wet scrubber to reduce both SO2 and acid gases, and at the same time switches to a higher sulfur coal (since the SO2 control from the scrubber will offset the increased SO2 from the higher sulfur coal). This EGU will likely experience an increase in emissions of sulfuric acid, since SO3 converts almost instantly to sulfuric acid in a wet scrubber. Sulfuric acid emissions are regulated as part of the EPA’s New Source Review program. Thus, it is possible that adding a scrubber for the purpose of complying with the Utility MACT could necessitate that the utility obtain a Prevention of Significant Deterioration (PSD) permit to address sulfuric acid.
The proposed Utility MACT is a complex set of emission limits and standards . EPA has estimated the Utility MACT will cost industry nearly $11 billion, making it one of the most expensive rules in the history of the agency. The extraordinary costs are likely to have profound impacts on electricity supply and price. Due to the far-reaching impacts of the rule, many are concerned over the compressed timeline that exists for rule development. EPA received many requests to extend the July 5 deadline for receiving comments on the proposed rule, including several from bipartisan congressional leaders. EPA responded by adding an extra 30 days to the existing comment period, now ending on August 4, 2011.
The extended deadline for submitting comments does not change the court-mandated deadline of November 16, 2011 for having a final rule in place. EPA can request an extension of the November 16 deadline, but the extension would not necessarily be granted. Just recently, EPA failed to obtain an extension of a similar deadline that applied to the Boiler MACT. If the Utility MACT follows a similar path, EPA may be forced to publish a rule without having adequate time to evaluate the numerous comments that have been or will be submitted. For the Boiler MACT, this led EPA to officially reconsider certain aspects of the final rule and to issue a stay on the effective date to fully address stakeholder input.
A final Utility MACT is imminent. Most companies with affected EGUs are in the process of evaluating how to respond. Critical decisions will have to be made for compliance with the Utility MACT, as well as the other new and impending rules that are impacting EGUs. To learn more, register for our upcoming webinar at trinityconsultants.com/webinars.
1 EPA is using “units designed for coal < 8,300 Btu/lb” to differentiate EGUs burning lignite. It is important to note that 8,300 Btu/lb is on a moist, mineral free basis, which means with water but without ash. As received, this equates to around 6,000 Btu/lb.
2 EPA is using “units designed to burn solid oil-derived fuel” to differentiate petroleum coke.
3 On May 16, 2011, EPA issued an administrative stay of the major source MACT until the proceedings for judicial review are completed or EPA completes its reconsideration of the rules, whichever is earlier. On June 24, 2011, EPA announced that it intends to sign a proposed reconsideration rule by October 31, 2001 and to sign a final rule by April 30, 2012.
4 Separate limits apply only to pollutants for which EPA determined emissions are tied to the type of combustor design, as opposed to being tied directly to the fuel type (coal, biomass, oil, etc).