MEMORANDUM

To: Tim Gregori, Southern Montana Electric Generation and Transmission Cooperative, Inc.(SME)

CC: Jeff Chaffee, Bison Engineering, Inc.

From: Paul Skubinna, Air Resources Management Bureau

Date - Time: June 18, 2008 – 09:00

The Department of Environmental Quality (Department) has conducted a preliminary review of the “Addendum to Application for Air Quality and Operating Permits” received June 6, 2008, specifically the particulate matter with an aerodynamic diameter less than 2.5 microns (PM_2.5) Best Available Control Technology (BACT) analysis. Pursuant to Finding of Fact #11 (page 17 lines 16 – 22) of the Final Board of Environmental Review (BER) Order and to ensure that the Department has the information necessary to conduct the required BACT analysis, please consider the following comments and request for additional documentation and information. Note that the following items would constitute application deficiencies if they remain unaddressed upon expiration of the 30-day application review period. In an effort to work with SME to expedite review of the submittal, the Department is providing the attached preliminary review; such that, if information is submitted to the Department by (or before) close of business (COB) June 24, 2008 that renders the application complete, it is the Department’s intent to process your application relative to the time frames established by the June 6, 2008 application receipt date. It is also the Department’s intent to work closely with SME to complete the ordered BACT analysis in a timely, efficient and defendable manner; therefore, if you have any questions, comments or would like to schedule a meeting to discuss this matter, please contact me as soon as possible.

PM_2.5 BACT Submittal Review Summary

Step 1: Identify Alternative Emission Control Techniques.

1. SME needs to list and address the availability of the following control, transferable and innovative technologies, techniques and options (BER Finding of Fact 9a, 9g, 14, 15, 24) and carry them forward through the analysis as appropriate. If the listed controls are found to be unavailable for SME’s circulating fluidized bed (CFB) boiler, SME needs to provide documentation of sources of information used to determine that the technology has not been proven at full scale operation or is not commercially available.

· Bipolar Agglomerator, (Indigo);

· Electrostatic Fabric Filter (General Electric);

· Electrostatic Centrifuge (Pratt and Whitney);

· Membrane Electrostatic Precipitator (ESP), (Environmental Elements Corporation, Southern Environmental, Inc and others);

· Laminar flow ESP (Environmental Elements Corporation)

· Supercritical Boiler Technology, (IV.A.1. and Deseret permit); and

In addition, SME needs to provide a complete list of linked/combined control technology strategies and techniques that will be carried forward for analysis including those above that are found to be available and/with those analyzed in the addendum submittal.(BER FF 9c, 15) (New Source Review Draft Workshop Manual (NSR) IV.A.4. List of Control Options).

2. SME needs to identify which control option(s) or strategy is believed to constitute lowest achievable emission rate (LAER) (BER FF 9f, 18).

3. Dry ESP, Coal Cleaning, and Cyclones and Inertial Separators were eliminated at Step 1 (although Dry ESP was stated to be technically feasible in Step 2). Step 1 is not a technique/technology elimination step unless that technology is demonstrated to be unavailable or not applicable to the emission unit in question. These technologies and combinations of these technologies with other technology options must be carried forward to Step 2 and to other steps, as appropriate. (NSR III.A. last paragraph, second sentence). (BER FF 9d, 9g, 20, 25). See additional comments/requests for information under Steps 2 and 3 regarding these technologies.

Step 2: Eliminate Technically Infeasible Options.

1. Coal Cleaning was eliminated is Step 1. It is not clear whether this was for technical infeasibility or some other reason. SME needs to describe the technical feasibility of coal cleaning in Step 2 (NSR III.B first paragraph second sentence) (BER FF 9c, 9g).

2. SME needs to provide a technical evaluation of the feasibility of control technologies, techniques and control technology combination strategies omitted (bipolar agglomerator, for example) or previously eliminated (cyclones or inertial separators, for example) from Step 1 as articulated above that are determined to be available for the CFB boiler.

Step 3: Rank Remaining Control Technologies by Control Efficiency.

For all information/clarifications requested, if SME does not find technical information (or it is not available) to support control efficiency estimates for the applicable PM_2.5 fraction and/or control technology, SME must provide documentation of sources of information examined to no avail.

1. Wet FGD - Condensable PM_2.5 control efficiency was estimated to be 80% in Step 1 of SME’s submittal based on an Environmental Protection Agency (EPA) technology fact sheet and analysis conducted for the Deseret permit. SME needs to clarify the basis for this estimated control efficiency. For example, is the control efficiency estimate based on an average of the cited work, an arbitrary value based on best professional judgment, an estimate based on vendor research and technology research and development testing; or, some combination of all of the aforementioned.

2. Dry FGD – Condensable PM_2.5 control was estimated to be 79%. This appears to be based on the summation of 75% (assumed to be similar to SO₂ control efficiency) and 15% improvement attributed to dry sorbent injection. SME needs to clarify the mathematical basis for the estimated 79% control efficiency reported for dry FGD. Note the following citation that may contain useful information in determining control efficiency specific to PM_2.5 for dry FGD.

“Advances in control of PM_2.5 and PM_2.5 precursors generated by the combustion of pulverised coal”, International Journal of Environment and Pollution, 2002 - Vol. 17, No.1/2 pp. 143-156.

3. Fabric Filter Baghouse:

· The proposed BACT analysis reported a 30% control efficiency for condensable PM_2.5. SME needs to provide citation or technical analysis of available information/data to support this control efficiency. The following journal citation and others like it may be helpful.

“Interaction between SO₂ from flue gas and sorbent particles in dry FGD processes”, China Particuology, Volume 3, Issues 1-2, April 2005, Page 141, Haiying Qi, Changfu You, Xuchang Xu.

· The proposed BACT analysis estimated 99.5% control efficiency for filterable PM_2.5 based on EPA documented control efficiencies that do not consider size fractionation. SME needs to provide a direct estimate of filterable control efficiency specific to PM_2.5 size fraction (BER FF 9b.). The following journal citations and others like it may be helpful.

“Fine particle and trace element emissions from an anthracite coal-fired power plant equipped with a bag-house in China”, Fuel, Volume 87, Issues 10-11, August 2008, Pages 2050-2057, Honghong Yi, Jiming Hao, Lei Duan, Xiaolong Tang, Ping Ning, Xinghua Li.

“Calibration of an optical particle counter to provide PM_2.5 mass for well-defined particle materials”, Journal of Aerosol Science, Volume 38, Issue 3, March 2007, Pages 325-332, J. Binnig, J. Meyer, G. Kasper.

“Design of baghouses for fines collection in milk powder plants”; Powder Technology, In Press, Corrected Proof, February, 2008, J.R. Gabites, J. Abrahamson, J.A. Winchester.

“Advances in control of PM_2.5 and PM_2.5 precursors generated by the combustion of pulverised coal”, International Journal of Environment and Pollution, 2002 - Vol. 17, No.1/2 pp. 143-156.

· The proposed BACT analysis estimated a 99.9% control efficiency for filterable PM_2.5 using membrane bags, based on EPA documented control efficiencies that do not consider size fractionation. SME needs to provide a direct estimate of filterable control efficiency specific to PM_2.5 size fraction (BER FF9b). For example Process Systems Incorporated advertises a commercially available membrane filter with a PM_2.5 specific control efficiency of 99.9999%. Additionally, the following journal citation and others like it may be helpful in estimating a PM_2.5 specific control efficiency for both filterable and condensable parameters relative to membrane technology.

“The influence of conditioning and regeneration on the separation behavior of rigid surface filters for the separation of particles from gases”, Powder Technology, Volume 120, Issue 3, 22 October 2001, Pages 223-231, Achim Dittler, Heinz F. Umhauer.

4. Cyclones and Inertial Separators – This technology was described in Step 1 of the proposed BACT analysis as having control efficiencies of less than 50%. Pratt and Whitney advertise a commercially available electrostatic centrifugal control technology with a PM_2.5 specific control efficiency of 98%. Provide a PM_2.5 specific control efficiency for this technology.

5. Dry ESP – This technology was eliminated in Step 1; however it was stated to be technically feasible in Step 2. SME needs to identify the PM_2.5 control efficiency for this technology and any available and feasible combination control strategies including this technology in Step 3. (NSR III.C. first sentence).

Step 4: Evaluate Most Effective Controls and Document Results.

1. Control strategies 1, 1a, 2, 2a, 4 and 5, 5a were eliminated based on environmental impacts including disposal of solid and liquid wastes without a description of secondary or collateral effects on environmental resources. Based on the NSR Manual, this is an invalid method for elimination of these control strategies (NSR IV.D.3. 3^rd paragraph 7 and 8^th sentence, 4^th paragraph). SME needs to provide a qualitative and, where possible, quantitative discussion regarding secondary and/or collateral effects on environmental resources due to need to treat and dispose of waste and additional process water requirements.

2. Control strategies 3, 3a and 6 were eliminated based on incremental cost impacts. Cost increments within the analysis were determined using methods that diverge from incremental analysis procedure in NSR. If incremental costs are to be considered in the analysis, SME needs to provide an incremental cost analysis that conforms to methods presented in NSR manual. This includes, at minimum, identification of the dominant control alternative, determination of the least cost envelope, and calculation of incremental costs of the dominant control alternatives relative to each adjacent control alternative along the least cost envelope. (NSR IV.D.2.b. Incremental Cost Effectiveness)

3. Elimination of control options should not be based on energy impacts, incremental costs or environmental impacts individually (NSR IV.D.1, first paragraph; IVD.2.c.; and statements in context at IV. D. 3. second paragraph). SME must clearly document any cumulative and synergistic analysis in the decision making process in describing why the remaining top control strategy was eliminated in Step 4.

HAZARDOUS AIR POLLUTANT EMISSIONS & MACT Applicability

Conceptually, to make the requested emission limitations federally enforceable the Department would propose parameter specific HAP numeric emission limitations on a pound per million British thermal unit (lb/MMBtu) basis. Because the specific HAP limitations (for hydrofluoric and hydrochloric acid, specifically) as proposed are measured on a 1-hour averaging time, the associated heat input limitations would also have to be applied on that same basis to make the effective ton per year limit on HAPs federally and practically enforceable. Therefore, the heat input limitations would have to reflect the 95% (or other, as appropriate) capacity limit on an hourly basis. For example, 95% of the average design heat input rate of 2626.1 MMBtu/hr would equal 2494.8 MMBtu/hr, which could not be exceeded during any hour.