National Park Service

Preliminary Comments on the Greene Energy
Prevention of Significant Deterioration (PSD) Permit Application
March 2005

Background

Wellington Development proposes to construct and operate the Greene Energy (Greene) project consisting of two new 290 (gross) MW Circulating Fluidized Bed (CFB) waste-coal boilers in Greene County in southwestern Pennsylvania. The proposed site is about 185 kilometers northwest of Shenandoah National Park (NP), a Prevention of Significant Deterioration (PSD) Class I area administered by the National Park Service (NPS). SO2 emissions would be controlled to 1,290 lb/hr and 5,649 tpy @ 0.234 lb/mmBtu (24-hour average) by the SO2 removal inherent in CFB, plus a Dry Scrubber; NOx to 551 lb/hr and 2,414 tpy @ 0.10 lb/mmBtu (24-hour average) by Selective Non-Catalytic Reduction (SNCR); and total PM10 to 276 lb/hr and 1,207 tpy (3-hour average) by a fabric filter (baghouse). The project is also subject to PSD for sulfuric acid mist (145 tpy); no additional controls are proposed for this pollutant.

Environmental Impacts of Waste Coal Combustion

Garbage of Bituminous (gob) is a waste product resulting from the removal of unwanted impurities from coal prior to combustion. Large quantities of gob can be found in many coal-mining areas, and the presence of gob piles can result in water pollution as minerals are leeched by rainwater. In some instances, gob piles have caught fire spontaneously, resulting in serious localized air pollution. For these and other reasons, it is desirable to remove gob piles from the environment.

However, the burning of gob also results in environmental impacts. Not only are significant amounts of pollutants released to the atmosphere, but, depending upon the inert content of the gob, the non-combustible portion that is not emitted from the stack remains as a solid waste for re-use or disposal. For example, for every ton of gob burned by the Greene project, over 5,000 pounds of pollutants (mostly carbon dioxide) would be emitted to the atmosphere, including several toxic substances such as arsenic, beryllium, lead, mercury, and radio nuclides. Over 800 pounds of solid waste (not including residual limestone introduced to the boiler) would remain from the combustion and pollution control processes. (See Appendix A.)

While removal of gob can be beneficial to the environment, care must be taken to ensure that combustion of gob does not result in undesirable environmental side-effects.

Best Available Control Technology (BACT) Analysis

Because the NPS Air Resources Division reviews permit applications for projects potentially impacting any park in the National Park System, we are able to gather information on a wide range of projects and emission control technologies. We are pleased to be able to share some of that information with the Pennsylvania Department of Environmental Protection (PA DEP).

BACT definition and process: EPA defines BACT as

an emissions limitation (including a visible emission standard) based on the maximum degree of reduction for each pollutant subject to regulation under the Clean Air Act which would be emitted from any proposed major stationary source or major modification which the Administrator, on a casebycase basis, taking into account energy, environmental, and economic impacts and other costs, determines is achievable for such source or modification through application of production processes or available methods, systems, and techniques, including fuel cleaning or treatment or innovative fuel combustion techniques for control of such pollutant…

It is important to note that, because BACT is an emission limit, that emission limit can be set by the permitting authority without actually specifying the design of the emission source that is to meet that limit. Thus, a permitting authority has the power to set an emission limit that it has judged to represent BACT for a broad source category, and then allow the applicant the freedom to determine how to meet that emission limit. According to the EPA New Source Review Workshop Manual (NSRWM):

Historically, EPA has not considered the BACT requirement as a means to redefine the design of the source when considering available control alternatives. For example, applicants proposing to construct a coalfired electric generator, have not been required by EPA as part of a BACT analysis to consider building a natural gasfired electric turbine although the turbine may be inherently less polluting per unit product (in this case electricity). However, this is an aspect of the PSD permitting process in which states have the discretion to engage in a broader analysis if they so desire. Thus, a gas turbine normally would not be included in the list of control alternatives for a coalfired boiler. However, there may be instances where, in the permit authority's judgment, the consideration of alternative production processes is warranted and appropriate for consideration in the BACT analysis. A production process is defined in terms of its physical and chemical unit operations used to produce the desired product from a specified set of raw materials. In such cases, the permit agency may require the applicant to include the inherently lowerpolluting process in the list of BACT candidates.

So, a permitting authority does have "the discretion to engage in a broader analysis if they so desire."

Clean Coal Technologies A fundamental principle of pollution control is that it is generally desirable to avoid creating the pollution in the first place. According to the EPA NSRWM:

The first step in a "topdown" analysis is to identify, for the emissions unit in question (the term "emissions unit" should be read to mean emissions unit, process or activity), all "available" control options. Available control options are those air pollution control technologies or techniques with a practical potential for application to the emissions unit and the regulated pollutant under evaluation. Air pollution control technologies and techniques include the application of production process or available methods, systems, and techniques, including fuel cleaning or treatment or innovative fuel combustion techniques for control of the affected pollutant. This includes technologies employed outside of the United States. As discussed later, in some circumstances inherently lowerpolluting processes are appropriate for consideration as available control alternatives.

The Greene project is an example of how this can be accomplished using current technology--Circulating Fluidized Bed (CFB). However, we believe that this Clean Coal Technology can yield lower emissions if complemented by more efficient pollution controls.

SO2: PA DEP has proposed CFB with limestone injection, with permit limits of 137.8 lb/hr (each boiler) for 24-hours @ 0.234 lb/mmBtu. (PA DEP should have also included a 3-hr limit—as in the proposed Robinson Power permit—to ensure that the 3-hr National Ambient Air Quality Standards and PSD Increments are protected.)

In general, higher removal efficiencies can be achieved on coals with higher uncontrolled emissions because the “targets” are more concentrated. Conversely, it is usually more difficult to achieve high SO2 removal efficiencies on very clean coals. If all other conditions are held constant, SO2 emissions decrease with a lower concentration of SO2 in the stack gas. So, it is reasonable to conclude that Greene should be able to achieve higher control efficiency than similar CFB boilers burning coal with lower uncontrolled SO2 emissions.

Table 1.a. contains five CFB boilers permitted or proposing to burn coals with lower uncontrolled SO2 emissions but higher 24-hr SO2 removal efficiencies than Greene’s 96.6%. (Appendix B presents the results of application of EPA’s Integrated Air Pollution Control System (IAPCS5a) costing model to Gascoyne. That model estimates that Gascoyne can achieve its 98% overall SO2 removal at a cost of $5,143/ton.) If Greene were to achieve the same 24-hr 97.8% control efficiency proposed for the Gascoyne lignite-burning CFB—24-hr emissions could be reduced by 36%. (Appendix C presents the results of application of IAPCS5a costing model to Greene. That model estimates that Greene can achieve the same 98% overall SO2 removal at a cost of $2,103/ton. This would actually be slightly less than the cost effectiveness estimated by IAPCS5a for Greene to achieve its proposed 97% overall removal efficiency.)

On 30-day and annual bases, Table 1.a. contains six CFB boilers permitted or proposing to burn coals with lower uncontrolled SO2 emissions but higher SO2 removal efficiencies than Greene’s 96.6%. If Greene were to achieve the same 30-day 98.8% control efficiency proposed for the Gascoyne lignite-burning CFB, emissions could be reduced by 66%.

It is also reasonable to conclude that Greene should be able to achieve a lower specific emission rate (lb/mmBtu) than similar CFB boilers burning coal with higher uncontrolled SO2 emissions. A good example is the Spurlock boiler which is proposing to burn waste coals with higher uncontrolled emissions but lower controlled emissions.

H2SO4: The use of CFB and limestone injection avoids the creation of H2SO4 typical of a wet scrubbing system.

NOx: PA DEP has proposed NOx limits of 0.10 lb/mmBtu for a 24-hour average based upon the inherent low-NOx emissions of a CFB boiler plus addition of Selective Non-Catalytic Reduction (SNCR). While this 24-hour limit appears consistent with many other CFBs with SNCR for that averaging period, please note that the Highwood project in MT is proposing a 0.09 lb/mmBtu NOx rate. Table 1.b. also contains four CFB boilers with 30-day NOx limits of 0.07 lb/mmBtu. If Greene were to achieve these lower limits, this would reduce short-term NOx emissions by 10% - , and long-term emissions by 30% or an additional 724 tpy.

However, we would like Greene and PA DEP to further investigate the technical and economic feasibility of SCR. Greene and/or PA DEP should obtain statements from SCR vendors concerning its applicability to this CFB boiler and its effectiveness before reaching these conclusions. Although PA DEP is correct in noting that SCR has not been applied to CFB boilers in the US, it must consider SCR in the context of “technology transfer” as noted by the EPA NSRWM:

The control alternatives should include not only existing controls for the source category in question, but also (through technology transfer) controls applied to similar source categories and gas streams, and innovative control technologies.

In a NOx control technology workshop in Pittsburgh[1], several reports were presented by operators, designers, and vendors of SCR systems that testified to the technical feasibility of applying SCR to coal-fired boilers. If Greene continues its argument of technical unfeasibility, it should file a statement from a reputable SCR vendor supporting that point. When the question of application of SCR to a CFB was raised at the Pittsburgh workshop, one consultant[2] stated that he knew of no reason why it could not be done. (In fact, one presenter[3] in Pittsburgh suggested that addition of limestone, as would be inherent in a CFB, is desirable in counteracting the potential catalyst-poisoning effects of arsenic found in many coals.)

PA DEP appears to have based its assumptions about the energy necessary to operate SCRs on the belief that it would be necessary to reheat gas streams to 700 oF to optimize performance of a conventional SCR system. However, current literature suggests that newer SCR systems may operate at temperatures as low as 250 oF (see attached Table 2). If such lower temperature catalysts could be applied at Greene, reheat costs could be reduced dramatically, thus making application of SCR much more economically feasible. PA DEP should conduct additional analyses to determine the costs associated with re-heating the gas streams to temperatures in the 250 – 600 oF range. Unless Greene can provide better justification for rejecting SCR as technically infeasible, it should proceed to an economic analysis of SCR.

Based upon permit limits for similar CFB boilers, and based upon the ability of SCR to achieve lower emissions than proposed by Greene, we suggest that a 24-hour NOx limit of 0.07 lb/mmBtu or less may be appropriate for BACT in this case. If Greene were to achieve 0.07 lb/mmBtu, this would reduce NOx emissions by 30%.

PM: PA DEP has proposed a baghouse at 0.012 lb total PM10/mmBtu which is the lowest total PM10 limit we have seen for a CFB boiler burning similar waste coalfuel. However, the lowest limit we have seen on total PM10 on any CFB boiler is 0.0088 lb/mmBtu in a permit issued to the Northampton Generating plant in PA. (See Table 1.c.) PA DEP shouldmust show why the Greene facility cannot meet a similar limit.

In summary, we believe that the Greene facility could achieve lower SO2, NOx and PM10 emission limits than currently specified in the draft permit.

Compliance Monitoring

We recommend that a filterable PM10 limit and Continuous Emissions Monitor (CEM) requirements be added. For example, the West Virginia Division of Air Quality (WVDAQ) has included both filterable and condensable PM10 in its permit limit for Greene Longview Power, and proposed that PM emissions be monitored by a CEM within 18 months of boiler start-up or when performance specifications for such monitors are promulgated, whichever comes later.[4] We continue to believe that CEMs are an important tool for monitoring compliance. For that reason, we recommend that a PM CEM be installed upon startup.

State’s/Applicant’s Air Quality Modeling Analysis

Single Source Analysis: The draft permit limits proposed by PA DEP for Greene Energy are internally consistent and consistent with the rates modeled. Following is a comparison of the draft permit and Table 2-1 of the August 30, 2004 application: