RESPONSE TO COMMENTS AECI - Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 On January 28, 2010 the Director ofthe Arkansas Department ofEnvironmental Quality gave notice ofa draft permitting decision for the above referenced·facility. During the comment period, the facility submitted written comments, data, views, or arguments on the draft permitting decision. The Department's response to these issues is as follows: Comment #1 The first paragraph ofthe introduction on Page 5 should listthe two fuel oil storage tanks as being 1.533 million gallons each, and the demineralized water storage tank should be listed as 1.35 million gallons. New calculations and emissiontables have been included that reflect the change intank size and corresponding change in VOC emissions. Response to Comment #1 Agree. In addition to the requested change, Specific Congition#35, the Emission Summary Table, and Fee Calculation spreadsheet have been revised to reflect the decrease in permitted VOC emissions. Comment #2 The table on page 7 correctly identifies the fuel oil PMIO BACT determination as being filterable. However, filterable is not noted with the corresponding limit of0.009 IblMMBtu inthe table on page 8 orthe table on page26. Forthe sake ofclarity, and to avoid any confusion going forward, AECI requests that the notation (Filterable) be included with the fuel oil PMIO BACT determination inthe tables on pages 8and 26. Response to Comment#2 Agree. "Filterable" was inadvertently omitted. The draft permit has been revised. Comment #3 The second to last sentence inthe second paragraph ofthe Preliminary Analysis section on page 9 should not reference CO as a pollutantthat requires no further consideration. CO was not required to undergo Class II Area Ambient Air Impact Analysis. Response to Comment #3 Agree. PSD review was not triggered for CO, and ithas been removed from the referenced sentence. Facility: AECI - Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 Comment#4 The Process Description at the bottom ofpage 11 should have the following deletions in orderto be accurate. The word "TPS" should be removed from the first sentence. The "four cell wastewatercooling tower (SN-28 through SN-3l)" shouldbe removed from the second sentence. Response to Comment#4 Agree. The requested revisions have been made. Comment#5 The description on Page 19 for the Fuel Gas Water Bath Heater (SN-33) should list it as 12 MMBtulhr, not 10 MMBtu/hr. Response to Comment#5 Agree. Thetypographical error has been corrected. Comment#6 Specific Condition#17 on page 29 states that the "Testing shall be performed within 180 days of issuance ofPermitNo. 1903~AOP-R7..." While this can be accommodated for the natural gas fuel type, construction for fuel oil combustion is not scheduled to be completed until the end of 2010. Assuming, for example, the final Permit is issued April 1st,this would require the completion ofthe specified testing by October 1st, by which time fuel oil combustion will not be possible. To avoid the necessity for submitting testing extensionrequests, AECI would request the language ofthis specific condition be modified to accommodate fuel oil testing within the allowed schedule for newly modified equipment. AECI proposes the following change be made to the language ofthe condition. REPLACE: Testing shall be performed within 180 day ofissuance ofPermitNo. 1903-AOP-R7 in accordance with Plant Wide Condition#3 and EPA Reference Method 18 as found in 40 CFR Part 60, Appendix A. WITH: For natural gas combustion, testing shall be performed within 180 day ofissuance ofPermitNo. 1903-AOP-R7 in accordance with Plant Wide Condition#3 and EPA Reference Method 18 as found in 40 CFRPart 60, Appendix A. For fuel oil fuel combustion, testing shall be performed 2 of11 Facility: AECI - Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 within sixty (60) days ofachieving the maximum productionrate, butno later than 180 days after initial startup ofthe permitted source on fuel oil in accordance with Plant Wide Condition #3 and EPA Reference Method 18 as found in 40 CFRPart 60, Appendix A. Response to Comment #6 Agree. The requested change has beenmade. Comment#7 AECI requests that additional language be appended to Specific Condition #19(c) on page 30 in orderto clarify howoperation during periods ofstartup, shutdown, and malfunction (SSM) are addressed with respect to complying with the applicable emission limits. NSPS Subpart 40 CFR 60.8(c) allows for an SSM exemptionwhere it states the following (underlined portion): "Operations during periods ofstartup, shutdown, and malfunction shall notconstitute representative conditions for the purpose ofaperformance test nor shall emissions in excess of the level ofthe applicable emission limit during periods ofstartup, shutdown, and malfunction be considered a violation ofthe applicable emission limit unless otherwise specified in the applicable standard." The Subpart KKKK rule is silent on SSM, therefore, those periods are exempt from the standard per Subpart A. AECI proposes to append the following sentence to the current specific condition. Emissions during periods ofstartup, shutdown, and malfunction shall not be considered toward the calculation ofexcess emissions for this standard. Response to Comment #7 Agree. The Air Division agrees to the requested change but also modifies the request to clarify that the exemption only applies to the standards listed in Specific Condition #19(a). Specific Condition#19(c) has been revised by appending the following: Emissions duringperiods ofstartup, shutdown, andmalfunction shall not be consideredtoward the calculation ofexcess emissionsfor the standards listed in Specific Condition #19(a). Comment #8 Specific Condition #19(e) on page 31 references NSPS subpartKKKK in stating the initial compliance testing for NOx and S02 requires testing at four points inthe normal operating range ofthe turbines. This requirement seems to be a carry-over from theNSPS subpart GG requirement, which is no longer applicable, and is currently included in existing PermitNo. 1903-AOP-R6. According to SubpartKKKK section 4400, the testing must be done at any load conditionwithin plus orminus 25 percent of100 percent ofpeak load. Furthermore, according to Subpart KKKK section 4405, facilities withNOx-diluent CEMS may combine the initial 3 of11 Facility: AECI - Dell PowerPlant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 performance test with a standard RATA test. The testing is specified to be conducted at a single load level, within plus orminus 25 percent of100percent ofpeak load. Also, the existing condition specifies that the initial test shall be conducted within 180 days after start-up. This time period has already elapsed for the natural gas fuel type, whose start-up was initiated in 2007. Therefore, AECI proposes the following replacementlanguage for the existing conditionto address both ofthese issues. REPLACE: Specific Condition #19(e) WITH: The permittee shall conduct an initial compliance test for NOx and S02 within 180 days after start-up for fuel oil, and within 180 days ofissuance ofPermitNo. 1903-AOP-R7 for natural gas. The testing shall be conducted for each fuel, at a single load level, within plus orminus 25 percent of100percent ofpeak load. Responseto Comment#8 There are two parts to this comment, and each part will be addressed separately. The first part mentions that Subpart KKKK specifies testing to be conducted at a single load level, within plus orminus25 percent of100percent ofpeak load and that since the combustionturbines are equippedwith CEMS the initial performance test may be combined with a standard RATA test pursuant to §60.4405. The Air Division agrees that Subpart GG and testing at four points in the normal operating range is no longer applicable. The draftpermit has been revised. The second part states that the time period has already elapsed for natural gas testing. The Air Division does not agree with that statement. AECI will be modifying the turbines and their control devices. Therefore, AECI must perform a test for natural gas and a test for fuel oil after the modifications have been completed. The initial test for NOx may be combined with a standard RATA testas stated above. Specific Condition #19(c) has been revised as follows: Thepermittee shall conductan initialcompliance testfor NOx andS02 within 180days after start-upfor eachfuel type. The testingshallbe conductedfor eachfuel, at a single loadlevel, withinplus orminus 25percent of100percent ofpeakload. [Regulation No. 19§19.304, 40 CFR §60.8, and40 CFR §60.4400 and§60.4415] 4 of11 Facility: AECI - Dell Power Plant PennitNo.: 1903-AOP-R7 AFIN: 47-00448 Comment#9 The Cooling Systems source descriptionon page 37 should have "TPS Dell" replacedwith "Dell" in paragraphs four and five. Response to Comment #9 Agree. The draftpennit has been revised. Comment #10 The Fuel Oil Storage Tanks description on page 44 should listthe capacity as 1,533,000 gallons instead of1,750,000 gallons. Response to Comment #10 Agree. The draft pennithas been revised. The US EPA Region 6 comments were sent via email on March 11,2010. Below are the comments from the EPA and the Department's responses to those comments: EPA Comment #1 Page 4-2, Section 4.2, NOx emissions calculation: Please explain the proposed use of "1,215,466 dscfm" as Peak Flue Gas FlowRate. "Peak Flue Gas FlowRate" in Appendix A ofboth versions ofthe Applications (August 2009 and December 2009) was indicated as 924,843 scfm. (See page A-3) Please explainthe discrepancy. Why has the Flue Gas FlowRate inthe calculation increased 131 % from that of August 2009 Application? With the flue gas flow rate increase, did the heat input capacity remain unchanged? Response to EPA Comment #1 The peak flue gas rate of 1,214,466 dscfm is the sumtotal ofacombustionturbine (924,843 dscfm) and a duct burner (290,623 dscfm) peak flue gas rate. The original application (August 2009) proposedNo.2 fuel oil combustion without ductfiring. AECI - Dell later submitted a revised application (December 2009) which proposedto include duct firing. In both applications the heat input capacity ofthe turbines while combusting No.2 fuel oil remainedthe same at 2,112 MMBtu/hr. 5 of11 Facility: AECI- Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 EPA Comment#2 Page 23, Specific Condition #2, it states, "Compliance with this condition will be demonstrated by meeting the requirements set forth in Specific Conditions #3 through #18. Hourly emission rates are based on a worse-case fuel use scenario. [RegulationNo. 18 §18.80l and A.C.A. §8-4 203 as referenced by §8-4-304 and §8-4-311]" Please explain" the worse-case fuel use scenario" and howthe scenario and rates were derived? Also, afterchecking with the listed citation below from RegulationNo.18 §18.801, we did not find a correlation between this Specific Condition and referenced citations in the permit. Please explain. "No person shall cause or permit the emission ofair contaminants, including odors orwater vaporandincluding an aircontaminant whose emission is not otherwise prohibited by this Code, ifthe emission ofthe air contaminant constitutes airpollution." (Regulation 18 §18.801) Responseto EPA Comment#2 ForNo.2 fuel, oil the worst case fuel use is the maximum 15,190 gallons per hour with aheat content of139 MMBtu/gal. This correspondsto a maximum heat inputto the combustion turbines of2,112 MMBtulhr. The emission rates are based on AP-42 emission factors for fuel oil combustion (turbines) and natural gas combustion (duct burners) and the maximum heat input. For natural gas, the worst case fuel use is the maximum 1.86 million standard cubic feet perhour ofnatural gas with a heat content of1,020 MMBtu/MMscf. This corresponds to a maximum heat inputof1,898 MMBtulhrheat input. The emission rates are based on AP-42 emission factors for natural gas combustion and the maximum heat input. Specific Condition #2 contains a table listing the permitted non-criteria pollutants. Those pollutantsare also air contaminants. The citation is the Department's regulatory authority to limitthe emissions ofthose pollutants. EPA Comment #3 Page 25, Specific Condition #3, please clarify whether startup, shutdown, and malfunction (SSM) emissions were included inthe determination for the PMIO BACT emission limitation. BACT emission limits orconditions must be met on a continual basis at all levels ofoperation and be enforceable as apractical matter (contain appropriate averaging times, compliance verification procedures and recordkeeping requirements). Further, the emissions limits inthe permit must be sufficient to demonstrate protection ofshort term ambient standards. 6 of11 Facility: AECI- Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 We are also concernedthatthe statedperformance/compliancetesting frequency does not ensure ongoing compliance with the BACT limits for PMIO in Specific Condition#3. Has ADEQ considered the need to require more frequent PMIO performance testing to help ensure ongoing compliance with the permit limits. Also, we would encourage ADEQ to work with the applicant to correlate its opacity monitoring with any performancetesting for PM. ADEQ should ensure that the periodic testing and monitoring requirement allow the source to demonstrate ongoing compliance, including during SSM conditions (See In the Matter ofCitgo Refining and Chemicals, OrderResponding to Petitions' Request, pp.6-7, Adm'r May 26, 2009) http://www.epa.gov/region07/airltitle5/petitiondb/petitions/citgo_corpuschristi_west_response20 07.pdf Finally, DEQ should consider requiring the installation ofPM CEMS for each Combustion TurbinelDuct Burnerstackifsuch considerationhas not already been made. Response to EPA Comment#3 The BACT limits inthe permit must be met at all times including periods ofstartup and shutdown. The Air Division requires periodic testing where CEMS are not installed, and the Air Divisionhas determined for this facility that the testing frequency is sufficientto protect short term emission limits. The Air Divisionhas considered the optionrequiring PM CEMS but does not see the need to install PM CEMS whenthere is no compelling regulatory requirement to do so and thatperiodic testing accomplishes the same determination ofcontinuous compliance without additional cost to the facility. EPA Comment #4 Page 25, Specific Condition #3, it states, "Visible emissions may not exceedthe limits specified in the following table ofthis permit as measured by EPA Reference Method 9. Compliance with this opacity limit shall be demonstrated by the use ofnatural gas as a fuel and compliance with Specific Condition # 5during combustion offuel oil. Source Opacity Limit Regulatory Citation SN-Oland SN-02 (Natural Gas) 5% Regulation 18 §18.501 SN-01 and SN-02 (No.2 Fuel Oil) 10% Regulation 19 §19.901 In the Specific Condition #5, it states, "The permittee will conduct daily observations while burning fuel oil by apersontrained in EPA Reference Method 9 and keep a record ofthese observations..." 7 of11 Facility: AECI- Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 Neither requirement addresses opacitymonitoring during SSM period when violations are most likelyto occur. Regulation 18 requires"...limitsthe emissions offederally regulated air pollutants on a continuous basis." (Regulation No. 19, Definition of"Emission limitation" and "emission standard"). We recommend that ADEQ consider requiring the installation ofa continuous opacity monitoring system (COMS) to measure and record the opacity ofemissions from the two turbine generators. Response to EPA Comment #4 The Air Division has considered the option ofrequiring COMS butthere is nojustificationto install COMS when there is no compelling regulatory requirement to do so and that daily observation while burning No.2 fuel oil accomplishes the same determination ofcontinuous compliance. The condition will remain unchanged. EPA Comment#5 Page 38, Specific Conditions #39: it states, "The total dissolved solids concentration for SN-04 through SN-15 shall not exceed 8,000 parts permillion in the water. " ADEQ permitted total dissolved solids (TDS) concentration is 8000 ppm but tested actual TDS emission is in the range between 670 -730 ppm. (see Page 1-2, PSD Permit Application) We question why ADEQ used emission factors from AP-42 and AWMA Abstract No. 216, SessionNo. AM-lb, Orlando, 2001 when a concentration based limit can be set utilizing recent test results. ADEQ should correct TDS concentration limit based on recent TDS test results in the Appendix F ofthe PSD Application. AP-42 orother emission factors are helpful for estimating emission levels, they are generally not appropriate for determining compliance with an applicable requirement unless the factor has eitherbeen developed directly from the emission unit in question or substitutes for aproven mass-balance relationship. (See EPA Periodic Monitoring Guidance) http://www.epa.gov/region07/air/title5/t5memos/pmguide.pdf Response to EPA Comment #5 The Air Division does not require apermittee to take a loweremission limitwhen sampling indicates the concentrations to be lowerthan the limitsinthe permit. While the TDS 8of11 Facility: AECI- Dell Power Plant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 concentrations have been tested, emissions from the cooling towers have not. The use ofAP-42 for PM emissions appears to be appropriate due to the lackofemissions testing. AWMA documentwas used because AP-42 is known to overestimate PM emissions. lO EPA Comment#6 Page 38, Specific Conditions #40: it states, "The total dissolved solids concentration for SN-16 through SN-22 and SN-24 through SN-27 shall not exceed 1,500 parts per million in the water. " ADEQ should evaluate the TDS concentration limit based on recent TDS test results from the Appendix F ofthe PSD Application. (See comments made in Item 7) Response to EPA Comment #6 See Response to EPA Comment #5. EPAComment #7 Page 38, Specific Condition #41: it states, "The permittee shall monitor monthlythe total dissolved solids concentrationto demonstrate compliance with Specific Condition # 39 and weekly to demonstrate compliance with Specific Condition# 40. The use ofa hand held meter shall be considered as an acceptable compliance method providedthat the permittee obtains prior approval from the ADEQ Stack Test Inspector Supervisor for the use ofhand held meter and its calibration and maintenance protocol. Otherwise, the permittee shall demonstrate compliance by submitting samples to third party laboratories that use EPA accepted test methods for measuring the conductivity ofthe sample or TDS concentration." Upon submitting such request to ADEQ Stack Test Inspector Supervisor for the use ofhand held meter, AECI should also notify EPA ofthe request for the change intest method. Response to EPA Comment #7 Specific Condition #41 has been revised to state that the permittee shall provide a copy ofthe request to use an alternate compliance method. EPA Comment#8 It does not appear that ADEQ addressed PM2.s emissions inthis permit action. It is unclear whether ADEQ isrelying onthe PM surrogateto satisfythe PM2.Srequirements. Ifso, it is lO necessary to provide a demonstration to supportthe use ofPM as a surrogate for PM2.S. See, lO e.g., 75 Fed. Reg. 6827, 6831-32 (February 11,2010). ADEQ should evaluate whether to require the applicant to submit a revised application or demonstration addressing PM2.S 9 of11 Facility: AECI - Dell PowerPlant PermitNo.: 1903-AOP-R7 AFIN: 47-00448 emissions. See, In re Louisville Gas andElectric, PetitionNo. IV-2008-3 (OrderonPetition). Additional information from the applicant should eitheraddress PM2.5 emissions directly or show howcompliance withthe PSD requirements for PMIO will serve as an adequate surrogate for meeting the PSD requirements for PM2.5 in this specific permit, after considering and identifying any remaining technical difficulties with conducting an analysis ofPM2.5 directly. The permit record must reflect a demonstration to supportthe use ofPMIO as a surrogate for PM2.5. EPA headquarters is developing recommendations for carrying out a PM2.5analysis using modeling and ambient monitoring data. In the meantime, we have worked with other permitting authorities and permit applicants to establish an appropriate PM2.5modeling protocol. Ifthe applicant chooses to address PM2.5 impacts directly, please contactus to develop amethodology that will ensure that an appropriate analysis is performed. Also, ADEQshould specifywhether it is using its NOx and S02 monitoring as amonitoring surrogate for PM25. Although, we did note that the source is not required to monitor for S02 emissions through CEMS. Response to EPA Comment #8 The permit does not address PM2.5directly. PM2.5 New Source Regulations are not yet implemented by the state ofArkansas. As an "approved" state, the state has 3 years from the date ofthe rule to incorporate the provisions into its regulations. In the interim and in accordance with EPA policy, PMIO remains as the surrogate pollutant. See October24, 1997 memo from John Seitz, Director OAQPS, "Interim Implementation ofNew Source Review Requirments for PM2.5" and later affirmed in "Implementationofthe New Source Review (NSR) Program for Particulate Matter Less Than 2.5 Micrometers (PM2.5)", Federal RegisterIVaI. 73, No. 96I Friday, May 16, 2008. ADEQ acknowledges that the EPA proposed a rule on February 11. 2010 to repeal the EPA 1997 policy for the use ofPM10as a surrogate for PM25. However, EPA has not yet made the rule final which would establish an effective date. EPA Comment #9 On January 22,2010, EPA signed a newNational AmbientAir Quality Standard (NAAQS) for nitrogen dioxide (N02)' The new standard is a I-hour standard set at the level of100partsper billion (Ppb). The effective date ofthe newNAAQS is April 12, 2010. The Department should evaluate its regulations to see ifit is required to implementthe newNAAQS as ofApril 12, 2010, should the permitnot be issued priorto that date. Should the Department determine that the existing regulations provide for protection ofanyNAAQS in effect at the time ofpermit issuance (whichwould include the I-hour N02 NAAQS beginning on April 12), then any permit issued on orafter the effective date must contain a demonstration that the emissions increase from the proposed project does not cause or contribute to a violation ofthe new I-hour N02 NAAQS. Since Arkansas regulations incorporate EPA's regulations at 52.2I(k) by reference, EPA believes that Arkansas regulations should be interpretedto incorporate the new I-hour NAAQS upon the effective date ofthat standard. See, 52 Fed. Reg. 24672, 24682 n. 9 (July 1, 1987). 10 of11