On May 6, 2009, EPA proposed amendments to the National Emission Standards for Hazardous Air Pollutants from the Portland Cement Manufacturing Industry (PC MACT) in response to numerous petitions and court mandates. The proposed revisions present significant challenges to the portland cement industry by establishing a new lower maximum achievable control technology (MACT) floor that would require additional controls and monitoring not widely implemented on existing cement kilns. Should the methods utilized by EPA to amend the PC MACT become common in other MACT regulations, the challenges facing the cement industry could become ubiquitous among U.S. industrial facilities subject to MACT requirements.
Summary of Proposed Standards
The standards contained in the original PC MACT, the 2006 revisions, and the new 2009 proposed standards are summarized in Table 1. In the proposed rule, EPA established new MACT floors for new and existing sources of particulate matter (PM), total hydrocarbon (THC), mercury, and hydrochloric acid (HCl) emissions.
Procedures for Resetting the MACT Floor
EPA’s methodology for resetting the MACT floor reflects several recent court rulings. In the original PC MACT, limits for mercury, HCl, and existing source THC were not established, as EPA determined that these emissions can be directly influenced by raw materials (or were determined to be of low risk in the case of HCl) and these pollutants were generally notactively controlled across the industry. However,the prior rulemaking approaches were rebuttedin several court decisions that directed EPA in setting MACT floors, including the following:
- Sources with low HAP emissions, due to low levels of HAP in their raw materials, can be considered best performers in establishing the MACT floor [Sierra Club v. EPA, D.C. Cir. 2007]
- Floors for existing sources must reflect the average emission limitation achieved by the best performing 12 percent of existing sources, not levels EPA considers to be achievable by all sources [Sierra Club v. EPA, D.C. Cir. 2007]
- EPA cannot set floors of “no control.” EPA must set floor standards for all HAP, including those not controlled by at-the- stack control devices [Sierra Club v. EPA, D.C. Cir. 2007]
- EPA cannot ignore non-technology factors that reduce HAP emissions [Sierra Club v. EPA, D.C. Cir. 2007]
In response to these court decisions/directives, EPA established a process to adopt a strict floor setting exercise in the proposed PC MACT rulemaking in which only the rate for the pollutant in question is used to determine the MACT floor, independent of emissions of other pollutants from the same kiln (i.e., the MACT floor was set based on pollutant specific basis – regardless of whether any one kiln could collectively meet the floor for all or related pollutants). The cement category has more than 100 sources and the floor for existing sources is the average of the top 12 percent of sources. In a stringent interpretation of this floor setting requirement, EPA used an average of the top 12 percent of sources even when the data population for a particular pollutant was less than 30 kilns. In the past, EPA used the average of the top 5 sources when data from less than 30 sources was available in setting a MACT floor. Under the revised approach used by EPA in this rule, the proposed existing source floor for THC is based on the average of the top 2 kilns and the existing source floor for HCl is based on the average of the top 4 kilns. Although this approach is more stringent, in its assessment of the data used to set the various floors, EPA concluded that the difference between the average of the top 5 sources and an average of the top 12 percent of sources is relatively small.
The proposed HCl MACT floor is an example where EPA fully utilized its regulatory discretion in setting emission standards. In past evaluations, EPA determined that no cement kiln, under worst operating conditions, would emit HCl at levels that would exceed acceptable health-based levels. Furthermore, modeling conducted by EPA shows that if a health-based standard was proposed, its value would be 23 ppmv, much higher than the proposed limit of 2 ppmv. EPA is not relying solely on health-based limits due to the substantial reductions in emissions of other pollutants, including SO2 and other acid gases, PM, and ammonia, that will result from more stringent HCl control.
In general, this revised approach to setting MACT floor would likely present compliance challenges to any industry, but particularly for mercury, THC, and HCl emissions from cement kilns. Throughout the cement industry, emissions of these pollutants are rarely actively controlled and there is limited mitigation potential available through good operating practices. In other words, the emissions of these pollutants are directly related to their concentration in the kiln feed, which is fundamentally tied to the economic viability of a given cement plant. EPA chose not to establish kiln subcategories based on raw material HAP content, arguing specifically that in the case of mercury, there was not a distinct breakpoint for establishing separate subcategories based on the mercury content of lime feed. Therefore, the MACT floor is not established by sources with good operating practices, kiln technology, or controls currently demonstrated throughout the industry, but rather the HAP concentration of the raw materials readily available to their site. As a result, the only option for complying with these limits will be the installation of costly add-on controls under a “one limit fits all” proposal.
One downside of the “one limit fits all” MACT setting approach is that it does not account for case specific costs, including availability of raw materials, like a traditional best available control technology (BACT) analysis. For example, older, less-efficient plants that have low mercury and organics in their raw materials may bear less burden of MACT compliance compared to modern, well-controlled plants with greater compliance costs due to less optimal mercury or organic raw material sources. In the extreme case, EPA identified two cement kilns that utilize high mercury content raw materials and may not be able to meet the existing source mercury limit even if they apply best controls.
For a new 1.3 million tpy kiln, EPA estimates the average cost of compliance with the proposed standards, including installation and operation of required control devices and CEMS devices, will be $17.6 million in capital cost with an annualized cost of $4.62 million. Given the similarities in emissions standards for new and existing sources in the proposed rule, the same controls will likely be required for both. The compliance costs for existing kilns may be even greater than for new kilns, since retrofitting existing equipment is typically more expensive than installing new equipment, and existing sources will not have the option of selecting a site with lower mercury and organic content in the quarry limestone. Limestone is the largest raw material input (~80 %) to a kiln and, in many cases, is the largest contributor to Mercury emissions.
Another court decision addressed in the May 2009 proposal is the December 2008 ruling vacating the Start-up, Shutdown, and Malfunction (SSM) exemption for MACT requirements. Under this ruling, MACT emission standards now apply at all times, including during SSM events. Since the information used to develop the proposed PC MACT amendments was gathered well before the December 2008 SSM ruling, EPA acknowledges that its database has no information on emissions during SSM events. However, EPA relies on the recent SSM ruling to justify applying the MACT floor at all times, including during SSM events, since there is no data indicating that emissions during SSM differ from normal operations. EPA further states, “We believe that startup and shutdown are both somewhat controlled operating modes for cement kilns… so that emissions during these operating modes may not be significantly different from those during normal operation.” EPA requests com- ments on emissions during periods of SSM and, if emissions are different during SSM events, recommendations for defining these periods and how standards may be applied.
Possible Implications for Other Industries
In addition to the immediate concerns this proposed rule presents to the cement industry, the methods utilized in the rulemaking could have far-reaching effects. First, EPA has not yet established a formal process for addressing the SSM vacature. While EPA may initiate procedures to set SSM specific limits in the future, the proposed PC MACT rule demonstrates that the onus will be on industry to provide evidence for separate emissions limits for SSM events and “normal” operation. However, since SSM events are no longer exempt from the established MACT floor standards, any emissions testing data showing that emissions are higher and require a separate MACT floor standard during SSM events would effectively demonstrate periods of non-compliance. Given this delicate situation, facilities should continue to track how EPA addresses the SSM vacature.
Furthermore, EPA’s proposal to control levels for all pollutants from cement kilns could, if applied elsewhere, lead to more new limits for other MACT standards. The proposed limits for HCl from cement kilns, for which no negative health effects were determined, could set a precedent for EPA to require costly HAP controls in an effort to lower emissions of non-HAP pollutants that will also be mitigated by the HAP control device.
EPA’s proposed action on the PC MACT could be duplicated in a re-assessment of the MACT floor for other source types. On January 14, 2009, the NRDC and the Sierra Club petitioned EPA, requesting a comprehensive assessment of its existing regulations under 40 CFR Part 61 and 63 to ensure that each standard fully complies with the Clean Air Act. Similar to past petitions against the PC MACT, this petition contests instances where EPA established a MACT floor of “no control” due to lack of currently demonstrated control measures or emissions below acceptable health based levels and asserts that EPA must establish new standards that reflect the actual emission levels achieved by the relevant best performing sources. For the 29 sources categories specifically named in the petition, the proposed PC MACT amendments raises the possibility of significant change to existing source requirements across U.S. industry.