air operating permit for celotex corporation

STATEMENT OF BASIS/FACT SHEET for WEST DEPTFORD ENERGY STATION DRAFT PREVENTION OF SIGNIFICANT DETERIORATION PERMIT AND

TITLE V OPERATING PERMIT SIGNIFICANT MODIFICATION Program Interest (PI): 56078 / Permit Activity Number: BOP120001

I. FACILITY INFORMATION The West Deptford Energy Station (WDES) is located at 3 P a r a d i s e R o a d in West Deptford Township, Gloucester County, New Jersey and consists of natural-gas fired combined-cycle turbine power generating facility. The facility is owned by West Deptford Energy Associates Urban Renewal, L.P., and is operated by West Deptford Energy, LLC; both are owned by LS Power. The WDES is classified as a major facility based on its potential to emit 84.5 tons per year of volatile organic compounds (VOC), 240 tons per year of nitrogen oxides (NOx), 592 tons per year of carbon monoxide (CO), 104 tons per year of total suspended particulates (TSP), 150 tons per year of particulate matter with an aerodynamic diameter less than 10 microns (PM10), 145 tons per year of particulate matter with an aerodynamic diameter less than 2.5 microns (PM2.5), and 3.37 million tons per year of greenhouse gases in a carbon dioxide equivalent (CO2e) basis. The facility i s currently n o t a major source of Hazardous Air Pollutants (HAPs) and will remain a non-major source of HAPs after the Phase II expansion. A major HAP emitting facility is designated as major when the allowed emissions exceed 10 tons per year of any individual HAP or 25 tons per year of any combination of individual HAPs that may be emitted simultaneously. This permit allows individual HAP to be emitted at a rate to not exceed: 360 lb per year of acrolein, 13,400 lb per year of formaldehyde, and 7,200 lb per year of toluene. II. AREA ATTAINMENT CLASSIFICATION The Federal Clean Air Act (CAA) sets National Ambient Air Quality Standards (NAAQS) for six common air pollutants. These commonly found air pollutants (also known as "criteria pollutants") are particulate matter, ground-level ozone, carbon monoxide (CO), sulfur dioxide (SO2), nitrogen dioxide (NO2), and lead. The US Environmental Protection Agency (USEPA) also classifies areas as “attainment” or “nonattainment” for each criteria pollutant, based on the magnitude of an area's problem. Nonattainment classifications are used to specify what air pollution reduction measures an area must adopt, and when the area must reach attainment. Currently, the entire State of New Jersey is designated as nonattainment for the 8-hour ozone NAAQS and portions of the State are designated as nonattainment for the daily SO2 NAAQS. This facility is located in an-ozone nonattainment area of the State in which the ambient air concentration exceeds 8-hour ozone NAAQS. III. BACKGROUND AND HISTORY The facility’s current operating permit was approved on December 19, 2013. The original Prevention of Significant Deterioration (PSD) permit was issued on May 6, 2009. This PSD permit modification with a revised acid rain permit application was received on January 19, 2012 to construct phase II expansion. Please refer to the attached explanation sheet for the structure and configuration of conditions of approval, included in the Facility Specific Requirements section of this permit. This modification would allow the following changes to the facility’s approved operating permit and PSD permit:

Allow the construction and operation of WDES – Phase II, a 427 MW (maximum net output with duct firing) combined-cycle power generating facility with required state-of-the-art air pollution control technology proposed to be operated by West Deptford Energy II, LLC. The proposed Phase II expansion will consist of one “F-class” General Electric (GE) or Siemens combustion turbine generator (CTG) that will utilize pipeline natural gas only. A heat recovery steam generator (HRSG) downstream of the combustion turbine will recover heat from the exhaust gases to generate steam. The HRSG will be equipped with a 770 MMBTU/hour (HHV) natural gas-fired duct burner for supplementary firing and the WDES- Phase II expansion will include a steam turbine generator (STG). Supporting ancillary equipment for the Phase II expansion will include a 40 MMBtu/hr (HHV) natural gas fired auxiliary boiler with low NOx burners that will operate on natural gas exclusively for 4,600 hours per year or less, an 8-cell mechanical draft cooling tower, a 1,000 kW emergency diesel generator and a 282 HP emergency diesel fire pump.

1



This modification will also increase the allowable emission limits for the significant source operations as listed in the following table: Allowable Emission Limits

Facility’s Potential Emissions from all Significant Source Operations (tons per year) VOC (total

)

NOx CO SO2 TSP (total

)

Other (total

)

PM10 (total

)

PM2.5 (total

)

Pb HAPs

(total)

CO2e (total)

Current Permit

54.2 169 548 26.8 58.4 173 93 89.7 0.00057 9.8 NA

Proposed Permit

84.5 240 592 41.3 104 233 150. 145 0.00123 10.5 1,220,516

Change (+ / -)

30.3 71 44 14.5 45.6 60 57 55.3 0.00066 0.7 1,220,516

VOC Volatile Organic Compounds PM10 Particulates under 10 microns NOx Nitrogen Oxides PM2.5 Particulates under 2.5 microns CO Carbon Monoxide Pb Lead SO2 Sulfur Dioxide HAPs Hazardous Air Pollutants TSP Total Suspended Particulates CO2 e Carbon Dioxide equivalent Other Any other air contaminant regulated under the Federal Clean Air Act

2



IV. AIR CONTAMINANT EMISSIONS Table 1 lists the proposed maximum allowable emissions of air contaminants from this project in pounds per hour (lbs/hr) and appropriate concentration limits.

TABLE 1 MAXIMUM ALLOWABLE EMISSIONS FOR PHASE II COMBUSTION TURBINE/HRSG UNIT

(Operating Conditions That Produce Worst-Case Emissions Vary By Pollutant) (Base load Operations with Supplemental Duct-firing)

Air Contaminant

units Maximum Allowable Emissions

for Siemens Turbine Maximum Allowable

Emissions for GE Turbine Fuel: Natural Gas Fuel: Natural Gas

Nitrogen Oxides (as NO2) lbs/hr1 PPMVD @ 15% O2

2

23.0 2.0

22.44 2.0

Carbon Monoxide (CO) lbs/hr PPMVD @ 15% O2

10.5 1.5

10.24 1.5

Volatile Organic Compounds (VOCs)3

lbs/hr PPMVD @ 15% O2

4

1.0

3.9 1.0

Sulfur Oxides (SO2) lbs/hr

6.56

6.4

Total Suspended Particulates (TSP) lbs/hr lbs/MMbtu3

15.10 0.0048

15.10 0.0050

Particulate Matter less than 10 microns (PM10)

lbs/hr lbs/MMbtu

21.55 0.0069

21.55 0.0071

Particulate Matter less than 2.5 microns (PM2.5)

lbs/hr lbs/MMbtu

21.55 0.0069

21.55 0.0071

Ammonia (NH3) PPMVD @ 15% O2

2.0

2.0

Formaldehyde lbs/hr

0.62

0.6

Green House Gasses in terms of Carbon Dioxide equivalent (CO2e)

lbs/hr lbs/MW-hr

371,448 947

362,406 955

NOTES: 1. lbs/hr = Pounds per hour emissions per turbine.

2. PPMVD (@ 15% O2) = parts per million by volume on a dry basis (corrected to 15 percent oxygen).

3. lbs/MMbtu = pounds per million British thermal units

3



Table 2 shows the PSD Applicability Determination for this project. Based on this table, this project is subject to PSD for CO, NOx, TSP, PM10, PM2.5 and CO2e.

TABLE 2 - PSD Applicability Determination (based on TPY)

Air Contaminants Applicable to Project

Phase I Current Potential to Emit

Revised Potential to Emit under Phase I **

Proposed Emission Increase under, Phase II *

Phase I and Phase II

PSD Significant Net Emission Increase Threshold Criteria

PSD Applicable

CO 548 385.5 206.5 592 100 Yes

NOx 169 158.8 80.8 239.6 40 Yes

SO2 27 26.8 14.5 41.3 40 Yes

TSP (Total) 58.4 58.4 45.3 104.0 25 Yes

PM10 (Total) 93 93 57.3 150.3 15 Yes

PM2.5 (Total) 89.7 89.7 54.8 144.5 10 Yes

VOC (Total) 54.2 54.2 30.3 84.5 NA NA

Sulfuric Acid Mist 6.3 5.3 3.2 8.5 7 Yes

CO2e 2,151,582* 2,122,880* 1,249,217 3,372,097* 75,000 Yes * CO2e is not directly limited in the operating permit for phase I equipment. The PSD permit for the phase I was approved before the tailoring rule effective date January 2, 2011. The CO2e emissions for Phase I in the table were calculated from the allowable fuel use.

** The phase I emissions were revised in the modification prior to installation of the phase I equipment and these limits were to be applied.

NOTES

1. Maximum potential emissions determined using worst case potential to emit calculations are based on the following operating scenarios for two turbines and associated duct burners:

• 7,650 hours of natural gas-fired combustion turbine operation which includes up to 700 hours of natural gas- fired start-up/shutdown operation, and up to 3,500 hours of natural gas-fired duct burning operation.

• Cooling towers: 8,760 hours per year; • Auxiliary boilers: 4,600 hours per year on natural gas; • Limited operation (100 hours/year) of emergency diesel fire pumps and emergency diesel generators.

Based on Table 2, the potential increase in annual emissions, this modification is also considered a major PSD source since the proposed increases in CO, NOx, TSP, PM10, PM2.5, SO2, Sulfuric acid and CO2e are greater than the PSD significant increase emission thresholds for those pollutants. Thus, the source is determined to be subject to PSD review and Best Achievable Control Technology (BACT) requirements for CO, NOx, TSP, PM10, PM2.5, SO2, Sulfuric acid and CO2e.

4



Table 3 shows the Non-Attainment New Source Review (NNSR) (N.J.A.C. 7:27-18) Applicability Determination for this project. Based on this application this project is subject to N.J.A.C. 7:27-18 for NOx and VOC. This is because the potential emissions of NOx and VOC which are precursors for ozone, are greater than the N.J.A.C 7:27-18 thresholds of 25 tpy. Table 3 also shows that the proposed CO and PM/TSP, and PM10 emissions from WDES – Phase II are greater than the N.J.A.C 7:27-18 thresholds. As per N.J.A.C 7:27-18.2(b)2, the facility is required to conduct an air quality impact analysis to show that the proposed emissions of CO, PM/TSP and PM10 would not result in a violation of an National Ambient Air Quality Standard (NAAQS) and the New Jersey Ambient Air Quality Standard (NJAAQS). N.J.A.C 7:27-18 is further discussed in Section IX.

TABLE 3 - N.J.A.C. 7:27-18 Applicability Determination

Air

Conta-minants

Total Increases in Allowable Emissions

from the new Turbine/HSRG and ancillary equipment

(TPY) IA

Sum of all creditable emission reductions at the facility during the Contemporaneous

Period (TPY) DC

Net Emissions Increase

(TPY) NI1

Significant Net

Emissions Increase

Threshold (TPY)

N.J.A.C 7:27-18

Applicable

CO 206.5 0 206.5 100 Yes2

NOx 80.8 0 80.8 25 Yes

SO2 14.5 0 14.5 40 No TSP (Total) 45.3 0 45.3 25 Yes2

PM10 (Total) 57.3 0 57.3 15 Yes1

PM2.5 (Total) 54.8 0 54.8 NA No3

VOC (Total) 30.3 0 30.3 25 Yes

Note:. 1. The net emission increase from the project was determined using the following formula specified in N.J.A.C. 7:27-18.7: NI = IP + INP + IF + IA - DO – DC, Where: NI = net emission increase at the facility IP = any increase(s) in the allowable emissions of the air contaminant that occurred during the “contemporaneous period” and that were authorized by permits issued by the Department INP = any increase(s) in the allowable emissions of the air contaminant that occurred during the “contemporaneous period” and that came from any equipment for which no permit was in effect at the time of the increase IF = any increase in the fugitive emissions of the air contaminant from the facility during the “contemporaneous period” IA = any proposed increase in allowable emissions of the air contaminant from the new or altered equipment or control apparatus that is the subject of the permit application DO = any increase(s) in the allowable emissions of the air contaminant that occurred during the “contemporaneous period”, if emission offsets were secured for these increases from the facility or from another facility DC = the sum of all creditable emission reductions at the facility during the “contemporaneous period”, not including any creditable emission reductions previously used as emission offsets at the facility or any other facility

5



2. Modeling requirements only. 3. Not a Subchapter 18 pollutant.

V. AIR POLLUTION CONTROL TECHNOLOGIES for BACT/LAER The WDES is required to evaluate Lowest Achievable Emission Rate (LAER) for each applicable pollutant (NOx and VOC) that exceeds the significant net emission increase level of any air contaminant listed in Table 3. LAER is the most stringent emission limitation contained in the implementation plan of any State for a particular source category, or which is achieved in practice by a particular source category, whichever is most stringent. The WDES is required to evaluate Best Available Control Technology (BACT) for each applicable pollutant (CO, TSP/ PM10/PM2.5, NOx, and CO2e) that exceeds the significant net emission increase level of any air contaminant listed in Table 2. BACT is an emission limitation based on the maximum degree of reduction for each regulated pollutant taking into account technical feasibility, energy, economics and other environmental factors.

1. Nitrogen Oxide (NOx) Control Technologies Description of NOx Control Technologies Nitrogen oxides are formed during the combustion of fuel in the turbine and are generally classified as either thermal NOx or fuel-related NOx. Thermal NOx results when atmospheric nitrogen is oxidized at high temperatures to yield nitrogen oxide (NO), nitrogen dioxide (NO2) and other oxides of nitrogen. The rate of formation is proportional to temperature in the combustion chamber. Fuel-related NOx is formed from the oxidation of chemically bound nitrogen in the fuel. Fuel-related NOx is minimal for natural gas combustion; NOx emissions from the combustion of natural gas are primarily from thermal NOx. Distillate oil contains some chemically bound nitrogen, and NOx emissions from oil combustion are comprised of both thermal and fuel-related NOx. Reduction in NOx emissions can be achieved using combustion controls and/or flue gas treatment. Available combustion controls include water or steam injection and low emission combustors that reduce NOx before it is formed. Flue gas treatment comprises of back-end controls, also called add-on controls that remove NOx from the exhaust gas stream once NOx has been formed. Selective Catalytic Reduction System (SCR) represents the current state-of-the-art for back-end controls for gas turbine and is considered the most stringent control technology for combustion turbines operating in both simple- and combined-cycle mode. NOx Controls for Combustion Turbines and Duct Burners Four technologies for controlling NOx emissions from the proposed combustion turbines were evaluated as follows: 1. Selective Catalytic Reduction System (SCR) In SCR process 19% aqueous ammonia (NH3) is injected directly into the exhaust gas and then exhaust gas is passed over a catalyst bed to react with formed NOx, converting NOx and injected ammonia to nitrogen and water. Environmental impacts associated with the use of SCR include storage and use of NH3, additional PM formation (nitrites) and catalyst disposal. The SCR promotes partial oxidation of sulfur compounds to SO2 to SO3 that ultimately form sulfuric acid and ammonia sulfates. 2. Selective Non-Catalytic Reduction (SNCR) SNCR is another method of post combustion control of NOx emissions. SNCR selectively reduces NOx into nitrogen and water vapor by reacting the flue gas with a reagent like urea or ammonia. The SNCR system is dependent upon the reagent injection location and temperature to achieve proper reagent/flue gas mixing for optimum NOx reduction. SNCR systems require a fairly narrow temperature range for reagent injection to achieve a specific NOx removal efficiency. The optimum temperature range for ammonia injection is 1,500° to 1,900°F. The NOx removal efficiency of a SNCR system decreases rapidly at temperatures outside the optimum temperature window. Operation below this temperature

6



window results in excessive ammonia emissions, also referred to as ammonia slip. Operation above the temperature window results in increased NOx emissions. 3. Dry Low-NOx Combustors (DLN) Dry Low-NOx (lean pre-mix) combustors stages a fuel combustion, lowering flame temperatures, thus by reducing the amount of thermal NOx formation without the use of diluents such as steam or water. 4. Lean Burn Combustion Typical gas turbines are designed to operate at a nearly stoichiometric ratio of fuel and in the combustion zone, with additional air introduced downstream. This is the point where the highest combustion temperature and quickest combustion reactions (including NOx formation) occur. Fuel-to-air ratios below stoichiometric are referred to as fuel-lean mixtures (i.e., excess air in the combustion chamber). The rate of NOx production falls off dramatically as the flame temperature decreases. Thus, very lean, dry combustors can be used to control emissions by reducing thermal NOx formation within the combustion chamber. The lean combustors typically are two-staged premixed combustors designed for use with natural gas fuel. The first stage serves to thoroughly mix the fuel and air and to deliver a uniform, lean, unburned fuel-air mixture to the second stage. The project has proposed to install a DLN combustion system with SCR on to achieve an emission limitation of 2.0 PPMDV corrected to 15% O2 on natural gas for all normal operations and thereby, meeting BACT/LAER requirements. NOx Control Technologies for Auxiliary Boiler In addition to SCR, and SNCR the following three control technologies for NOx were evaluated for the auxiliary boiler: 1. SCR and SNCR SCR emission control technology is not considered technically feasible for the proposed auxiliary boiler because the design effectiveness of an SCR is not achieved until the flue gas temperature reaches between 400 and 800°F. The auxiliary boiler will run exclusively on natural gas and with Dry Low NOx Burner (DLNB). During the turbine start-up procedure, pollutant emission concentrations a re elevated. The proposed auxiliary boiler will be required to supply steam in an expedited manner to minimize the duration of the combined cycle unit start-up. For this same reason, SNCR was also not found to be technically feasible for the auxiliary boiler. 2. Dry Low-NOx Burners (DLN) Dry Low NOx Burners reduce NOx through staged combustion. Staging partially delays the combustion process, resulting in a cooler flame, which suppresses thermal NOx formation. NOx emission reductions of 40 to 85 percent (relative to uncontrolled emission levels) have been observed with Low-NOx Burners. 3. Flue Gas Recirculation (FGR) In an FGR system, a portion of the flue gas is recirculated from the stack to the burner. The recirculated gas is mixed with combustion air prior to being fed to the burner. The FGR system reduces NOx emissions because the recirculated gas reduces combustion temperatures, thus suppressing the thermal NOx mechanism. FGR also reduces NOx formation by lowering the oxygen concentration in the primary flame zone. Together, Low-NOx Burners and FGR are capable of reducing NOx emissions by 60 to 90 percent. Technical Review of NOx Controls The proposed auxiliary boiler will be limited to natural gas firing only. One of the purposes of the auxiliary boiler is to supply steam during the start-up of the combined cycle unit.

7



The expansion project has proposed to install DLN Burners along with FGR and the use of natural gas to comply with BACT and LAER for the auxiliary boiler. The proposed NOx emission limit for the auxiliary boiler is 0.011 lbs/MMBtu (equivalent to 0.44 lb/hr or 1.0 TPY). The project has also proposed to take a restriction on the amount of natural gas usage for the boiler equal to 180 MMscf/yr, equivalent to 4,600 hours annually, operating at 100 percent load. The Department has reviewed the proposed NOx emission limitations for auxiliary boiler and found them to be BACT and LAER. These proposed NOx emission limitations will also comply with SOTA limit of 0.035 lbs/MMBtu for this size boiler firing natural gas. NOx Controls for Emergency Engines To comply with BACT and LAER limits, the emergency diesel generator (EG) and fire pump (FP) will operate on Ultra Low Sulfur Diesel (ULSD) exclusively, which is considered a clean fuel with low emissions and will meet the emission limits of NSPS IIII. The EG will comply with NSPS Subpart IIII requirements of “Non-methane hydrocarbon (NMHC) + NOx” <= "4.8 " g/HP-hr, CO <= "2.6 " g/HP-hr, PM <= " 0.15" g/HP-hr, while the FP will comply with NSPS Subpart IIII requirements of “NMHC + NOx” <= "3.0 " g/HP-hr, CO <= "2.6 " g/HP-hr, PM <= "0.15 " g/HP-hr. The proposed NOx emission limits for the EG and FP are 15.9 lbs/hr or 0.79 TPY and 1.87 lbs/hr or 0.093 TPY, respectively. The expansion project has also proposed to restrict the hours of operation for emergency diesel generator and fire water pump of less than or equal to 100 hours per year each equipment.

2. Volatile Organic Compound (VOC) Control Technologies Description of VOC Control Technologies The turbine, duct burner, auxiliary boiler, EG and FP are combustion emissions sources. By incomplete combustion of fuel volatile organic compounds (VOC) is generated.. Since potential emissions of VOC from the turbines and ancillary equipment are above N.J.A.C. Subchapter 18 threshold of 25 TPY, VOC emissions are subject to LAER analysis. Control of VOC emissions is dependent on combustion efficiency of the above equipment. High combustion temperatures and long residence times in the combustion zone minimizes VOC emissions, both of which result in increase of NOx emissions. As NOx emission limits are being driven lower and lower, achieving this balance has become more of a challenge, often resulting in a “trade-off” among pollutants.

Oxidation catalysts along with good combustion practice have been used to reduce both CO emissions and VOC emissions. However, the high temperature necessary to make oxidation catalyst effective for VOC reduction has the undesirable result of causing substantially more conversion of SO2 to SO3 and subsequently to the increase in PM/PM10/PM2.5 emissions due to the formation of sulfates and ammonium salt formation. VOC controls for Combustion Turbine and Duct burner: The most stringent VOC control levels for combustion turbine and duct burner has been achieved with advanced low NOx combustors and/or catalytic oxidation for carbon monoxide (CO) control. The facility proposes the installation of an oxidation catalyst for CO control that also reduces VOC emissions. The proposed VOC emissions limits when burning natural gas are 1.0 PPMDV and 0.7 PPMDV corrected to 15% O2 at 100% load with duct burner and without duct burner, respectively. The Department has determined that this limit meets LAER.

VOC controls for Ancillary Sources No technically feasible post-combustion control methods have been identified to assure the reduction of VOC emissions from auxiliary boilers. VOC emissions are controlled by good combustion practices. As 8



discussed in the section for CO controls below, an oxidation catalyst used to control CO emissions from a boiler, may also reduce some VOC emissions. The proposed VOC emission limit for the auxiliary boiler is 0.0015 lbs/MMBtu (equivalent to 0.06 lb/hr or 0.14 TPY), which meets LAER limitations for boilers of this size firing natural gas. Both emergency engines will comply with the NSPS Standards of Performance for Stationary Compression Ignition Internal Combustion Engines (40 CFR 60 Subpart IIII). The proposed VOC emission limit for the EG is 0.38 lbs/hr or 0.019 TPY). The proposed VOC emission limitation for the FP is 0.69 lbs/hr or 0.035 TPY). The Department has reviewed and found these proposed VOC emission limitations to be LAER.

3. Carbon Monoxide (CO) Control Technologies: Description of CO Control Technologies Carbon Monoxide is usually generated due to the incomplete combustion of fuel. Several factors lead to incomplete combustion including insufficient O2 availability, poor air/fuel mixing, cold wall flame quenching, reduced combustion temperature, decreased combustion residence time and load reduction. CO emissions are minimized by good combustion practices that oxidized all carbon and hydrogen contained within the fuel to form CO2 and H2O. Oxidation Catalyst as Combustion Control After combustion control, the only practical control method to reduce CO emissions from combustion of fuel is the use of an oxidation catalyst. Exhaust gases from the combustion equipment are passed over a catalyst bed where excess air oxidizes the CO to carbon dioxide (CO2). CO reduction efficiencies in the range of 80 to 90 percent can be achieved, although CO reduction may at times be somewhat less than the design value. Modern data acquisition and control systems through a combination of operator and software driven process adjustments optimize a combustion performance resulting in minimizing pollutant emissions, including CO. Technical Review of CO Controls: CO Controls for Combustion Turbines and Duct Burners The Phase II expansion project turbine will be equipped with oxidation catalyst to reduce CO emissions. This project proposes the use of oxidation catalyst as BACT for CO emissions along with process control and good combustion practices. The control system will reduce inlet CO concentrations approximately 50 to 70% depending on steady-state operating load condition. The oxidation catalyst will be located in an optimum temperature region within the HRSG immediately upstream of the SCR ammonia injection grid and downstream of the gas-fired duct burner. The proposed emission limitation when firing natural gas is 0.9 PPMDV and 1.5 PPMVD corrected to 15% O2at 100% load without duct firing and with duct firing, respectively. The Department has reviewed these limits and found them to be BACT. CO Controls for Auxiliary Boiler: Although an oxidation catalyst has been used to reduce CO emissions from boilers, it is not considered technically feasible to use it with the auxiliary boiler since the auxiliary boiler is required to supply steam quickly to the combined cycle units during the startup procedure and the oxidation catalyst requires a high flue gas temperature to achieve effective control. A more effective method of reducing emissions, including CO, is by good combustion controls and restricting operation on an annual basis.

9



The expansion project has proposed CO emission limitations for the auxiliary boiler of 0.036 lbs/MMBtu (equivalent to 1.44 lb/hr or 3.3 TPY), which meets BACT and the Department’s SOTA limitations for boilers < 100 MMBtu/hr. CO Controls for Emergency Engine and Fire Pump: The Phase II expansion project has proposed the CO emission limitation for the emergency diesel generator of 2.6 g/bhp-hr (8.6 lbs/hr or 0.43 TPY) and the CO emission limitation for the emergency diesel fire pump of 2.6 g/bhp-hr (equivalent to 1.62 lbs/hr or 0.081 TPY) and restricted hours of operation of 100 hrs/yr as BACT. The Department has reviewed these emission limitations and found them to be BACT.

4. Particulate Matter /Particulate Matter less than ten microns/Particulate Matter less than 2.5 microns (PM / PM10/ PM2.5 ) Control Technologies

Description of PM / PM10 Control Technologies Sulfur and nitrogen in the fuel are oxidized forming post-combustion particulate aerosols known as PM, PM10, and PM2.5 emissions due to incomplete combustion. Particulate emissions can also form to some extent from trace elements in the fuel. Sulfur in the fuel when combusted turns into Sulfur dioxide (SO2). Some of SO2 is converted to sulfur trioxide (SO3) in presence of a catalyst in SCR. Formation of sulfate aerosols occurs from use of ammonia (NH3) injected to control NOx with SCR technology. The EPA fine particle rule states that SO2, NOx, VOCs, and NH3 can all combine to the formation of PM2.5. All of the particulate matter emitted from the turbines is thus conservatively assumed to be less than 2.5 microns in diameter. Therefore, PM10 and PM2.5 emission rates are assumed to be the same. To assure that PM10, and PM2.5 annual emissions will comply with the permit limit and not exceed PSD limit, initially eight quarterly stack tests will be performed on site. These stack tests will be evaluated and permit limits for these pollutants may be revised accordingly. Technical Review of PM / PM10/ PM2.5 Controls A review of approximately 295 natural gas-fired combined cycle facilities from the USEPA‘s RACT/BACT/LAER/Cleaning House (RBLC) and recently issued air permit searches lists PM/PM10 emission limits ranging from 0.0013 to 0.1400 lb/MM Btu. In many instances, the pollutant listed in the RBLC database is TSP or PM. TSP and PM typically only includes the filterable portion of particulate matter; therefore, many of these limits cannot be compared to the proposed project. Control technologies, good combustion practice and low-sulfur, should be considered the driving factor for proposing BACT. Particulate matter is formed from non-combustible constituents in the fuel or combustion air, or from formation of ammonium sulfates post combustion. The WDES is not aware of any combustion turbine project that has been required to install add on controls for PM, PM10 or PM2.5. Post- combustion controls, such as baghouses, scrubbers and electrostatic precipitators (ESP) are not technically feasible for combustion turbines due to the high pressure drops, the large flue gas volumes and the low concentrations of PM/PM10/PM2.5 present in the exhaust gas. The combustion of clean burning fuels is the most effective means for controlling PM emissions from combustion equipment. WDES is proposing exclusive use of natural gas as the fuel for turbines and duct burners. The emission limit proposed for PM10/PM2.5 is 21.55 lb/hr (0.0071 lb/MM Btu for GE turbine and 0.0069 lb/MM Btu for Siemens turbine) and, for PM/TSP, is 15 .1 l b / h r ( 0.0050 lb/MM Btu for GE turbine and 0.0048 lb/MM Btu for Siemens turbine) when firing natural gas in the combustion turbine with duct burners operating. The emission limits proposed by WDES – Phase II for PM10/ PM2.5 is 10 lb/hr (0.0044 lb/MM Btu for GE turbine and 0.0043 lb/MM Btu for Siemens turbine and for PM/TSP is 6 lb/hr (0.0026 lb/MM Btu for GE turbine and 0.0026 lb/MM Btu for Siemens turbine) when firing natural gas in the combustion turbine without duct burner operating.

10



The use of natural gas (or other low ash content fuels like ULSD) is regarded as BACT for PM, PM10, and PM2.5. The Department reviewed and found the proposed limits for PM, PM10, and PM2.5 are BACT.

5. Sulfur Dioxide (SO2) Control Technologies Summary Emissions of SO2 are formed from oxidation of sulfur in fuel. Natural gas and 15 PPM sulfur ULSD are the stringent degree for SO2 control. Natural gas is the only fuel for the combustion turbine, duct burner, and auxiliary boiler and ULSD is the only fuel for the EG and FP. Description of SO2 Control Technologies There have been no applications of SO2 removal technology to CTs because low sulfur fuels are used. Sulfur removal technologies, such as scrubbers or flue gas desulfurization, are most effective with high inlet SO2 concentration, unlike sulfur dioxide concentration of CT exhaust gas characteristics on either gas or ULSD. Given the low SO2 exhaust stream concentrations, scrubber technology is not considered to be technically feasible for the proposed CTs. Technical Review of SO2 Controls for turbine and duct burner Some of the SO2 produced during the combustion process is converted to various sulfate compounds as the exhaust gases pass through the SCR and oxidation catalysts. The emission rates shown here incorporate the conversion of some of the SO2 to sulfate. Natural gas will contain 0.5 gr S/100 SCF (annual average) and 0.75 gr S/100 SCF (short-term maximum). Short-term emission rates are based on the expected short-term maximum sulfur content; long-term emission rates are based on the expected annual average sulfur content. ULSD will contain a sulfur content of no more than 15 parts per million by weight (PPMW). The proposed BACT SO2 emission rates are: 6.40 lb/hour and 4.78 lb/hour for the GE CT with duct firing and without duct firing, respectively; and 6.56 lb/hour and 4.94 lb/hour for the Siemens CT with duct firing and without duct firing, respectively. The Department reviewed and found the proposed limits for SO2 is BACT. Technical Review of SO2 Controls for Auxiliary Boiler The Project proposes to fire natural gas in the auxiliary boiler to meet BACT for SO2. The maximum proposed SO2 BACT emission limit is 0.0021 lb/MM Btu or 0.084 lb/hr. The proposed SO2 emission limit is below the 0.05 lb/hr N.J.A.C. 7:27-22 Appendix, Table A “Thresholds for Reporting Emissions of Air Contaminants” The Department reviewed and found the proposed limits to be BACT. Technical Review of SO2 Controls for Emergency Engines (EG and FP). WDES proposes to use only ULSD with a sulfur content limit of 0.0015% sulfur or approximately 0.002 lb SO2/MM Btu, which is well below the NSPS limit of 0.06 lb SO2/MM Btu for the emergency diesel generator and the fire pump. The sulfur dioxide emission limitations for emergency diesel generator and the emergency diesel fire pump are below the N.J.A.C. 7:27-22 Reporting Thresholds. The Department reviewed and found the proposed limits for SO2 is BACT

6. Sulfuric Acid (H2SO4) Control Technologies Summary Sulfuric acid forms when the sulfur is oxidized to sulfur trioxide (SO3) in the presence of catalyst and then condenses at lower temperatures in the presence of water. A small fraction of the SO2 formed from the fuel bound sulfur will be oxidized to SO3 during combustion and across the SCR catalyst and oxidation catalyst.

11



WDES proposes natural gas, an inherently low sulfur fuel, as the exclusive fuels for the combustion turbine, duct burner, and auxiliary boiler. The FP and EG will be fueled by ULSD. The maximum proposed for fuel sulfur limit for natural gas is well below the NSPS limit. Description of H2SO4 Control Technologies A conservative estimate of H2SO4 mist production has been made by assuming that all of the SO3 in the exhaust converts to H2SO4. The actual H2SO4emission rates are expected to be lower, as the majority, if not all of the SO3 should react with the NH3 slip from the SCR to form ammonium sulfate/bisulfates at low flue gas temperatures. Technical Review of H2SO4 Controls for turbine and duct burner There have been no applications of H2SO4 removal technology to CTs because low sulfur fuels are used. H2SO4 removal technologies, such as wet scrubbers are most effective with high inlet concentration, unlike CT exhaust gas characteristics on either gas or ULSD. Given the low sulfuric acid exhaust stream concentrations, scrubber technology is not considered to be technically feasible for the proposed CTs. As with SO2, use of mist eliminators to control H2SO4 mist emissions is not technically feasible due to the very low exhaust concentrations. ULSD will contain a sulfur content of no more than 15 PPMW. The proposed BACT H2SO4 emission rates are: 0.96 lb/hour for the GE CT with duct firing and 0.71 lb/hour for the GE CT without duct firing; and 0.98 lb/hour for the Siemens CT with duct firing and 0.74 lb/hour for the Siemens CT without duct firing. The Department reviewed and found the proposed limits to be BACT. Sulfuric Acid Controls for Auxiliary Boiler The Project proposes to fire natural gas in the auxiliary boiler to meet BACT for sulfuric acid. The maximum proposed H2SO4 BACT emission limit is 0.00048 lb/MM Btu or 0.019 lb/hr. The proposed H2SO4 emission limit is below the 0.05 lb/hr N.J.A.C. 7:27-22 Appendix, Table A “Thresholds for Reporting Emissions of Air Contaminants” Sulfuric Acid Controls for Emergency Engines. WDES proposes to use only ULSD with a sulfur content limit of 0.0015% sulfur or approximately 0.002 lb SO2/MM Btu, which is well below the NSPS limit of 0.06 lb SO2/MM Btu for the emergency diesel generator and the emergency diesel fire pump, The Sulfuric Acid mist emission limitations for emergency diesel generator and the emergency diesel fire pump are below the N.J.A.C. 7:27-22 Reporting Thresholds.

7. Greenhouse Gas (GHG) Control Technologies The main sources of GHG emissions for the Phase II project are the combustion turbine, duct burner, and auxiliary boiler. GHG emissions are also generated from the operation of the diesel engines, which are intended for limited operation (emergency and ready-testing). On June 3, 2010, EPA issued a final rule that “tailors” the applicability provisions of PSD for greenhouse gas (GHG) emissions. Under the tailoring rule, a new source (facility) that commence construction after July 1, 2011 is subject to PSD permitting requirements for GHG emissions, if the potential GHG emissions from the new source are greater than 100,000 tons/year, or if the source is otherwise subject to PSD for another pollutant and its GHG potential emissions are equal to or greater than 75,000 TPY. The Phase II is subject to PSD permitting requirements for GHG emissions. For PSD purposes, GHGs are considered a single air pollutant Carbon Dioxide Equivalent (CO2e) defined as the aggregate group of the following: • Carbon dioxide (CO2) • Methane (CH4) • Nitrous oxide (N2O)

• Hydro fluorocarbons (HFCs) • Perfluoro carbons (PFCs) • Sulfur hexafluoride (SF6)

12



a) Description of GHG (CO2e) Control Technologies:

The major constituent of CO2e emissions for combustion sources is CO2, which accounts for over 95% CO2e emissions. Therefore it is necessary to control CO2 emissions. Carbon Capture and Sequestration (CCS): EPA has classified CCS as an add-on pollution control technology that is “available” for large CO2-emitting facilities including fossil fuel fired power plants. Carbon sequestration is a geo-engineering technique used to remove the CO2from an exhaust gas stream and store it permanently in underground reservoirs (typically depleted oil or gas reservoirs) or other geological features. Ideal geological formations for sequestration include depleted oil and gas fields and deep ocean masses. The vicinity surrounding the proposed WDES does not contain saline aquifers, depleted oil or gas reservoirs, or deep coal beds and therefore is not suitable for the injection of CO2 underground. Alternative sequestration techniques include converting CO2 to baking soda or algae based carbon capture. Power plants similar in size to the proposed WDES – Phase II may require a significantly larger site area to accommodate the additional process facilities associated with CO2 capture. Such facilities could include CO2 compressors, scrubbers, oxygen production plants or other carbon capture equipment. Depending upon the capture technology employed, the site area for the capture equipment may approach the size of the site area of the power generating plant itself. In addition, capturing, scrubbing and compressing CO2 requires much energy and would increase the fuel needs of the plant. The long term storage of CO2 is a relatively new concept and has mostly been demonstrated on a pilot-scale. While current technologies could be used to capture CO2 from new fossil fuel fired power plants, they have not been demonstrated at the scale necessary to establish confidence for power plant application. 1 Further, the technology has not been demonstrated as suitable for the proposed site location. Therefore, CCS is not considered to be technically feasible for the project. The Department requested the applicant to analyze the cost effectiveness of adding CCS to the project. The analysis showed that CCS would add an estimated $239 million to the project’s construction costs – a 78% increase. This is an extremely high cost in comparison to the overall project cost. In accordance with recent EPA and EPA Environmental Appeals Board (EAB) precedent and EPA’s GHG Permitting Guidance2, the information provided by the applicant shows that installation of CCS on the Phase 2 expansion project is cost prohibitive. In addition to the high construction and operating costs associated with CCS, the carbon capture equipment requires a substantial amount of energy to operate, thereby reducing the efficiency of the WDES.

b) Thermal Efficiency: The design base load net heat rate at ISO conditions for the GE turbine is approximately 6,722 Btu/kWhr on a higher heating value (HHV) based on new and clean conditions without duct firing at steady state. This heat rate reflects the projects net power production which means that the output is net of plant auxiliary loads consumed by operation of the plant and is on HHV fuel basis. This heat rate is equivalent to a net plant efficiency of 56.3% Lower Heating Value (LHV). The design base load net heat rate at ISO conditions for the Siemens turbine is approximately 6,876 Btu/kWhr on a higher heating value (HHV) based on new and clean conditions without duct firing at steady state. This heat rate reflects the projects net power production which means that the output is net of plant auxiliary loads consumed by

1 See Report of the Interagency Task Force on Carbon Capture and Storage, p.50 (http://www.epa.gov/climatechange/policy/ccs_task_force.html).

2 PSD and Title V Permitting Guidance for Greenhouse Gases - EPA-457/B-11-001, U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Air Quality Policy Division, March 2011.

13

http://www.epa.gov/climatechange/policy/ccs_task_force.html)



operation of the plant and is on HHV fuel basis. This heat rate is equivalent to a net plant efficiency of 55.1% Lower Heating Value (LHV). The heat rate and efficiency is based on the gas turbines at base load without duct firing. The appropriate heat rate limit for the WDES – Phase II turbine was determined by applying the following compliance margins to the base heat rate, consistent with other recent GHG BACT applications:

• A 3.3% design margin reflecting the possibility that the constructed facility will not be able to achieve the design heat rate.

• A 6% performance margin reflecting efficiency losses due to equipment degradation prior to maintenance overhauls.

• A 3% degradation margin reflecting the variability in operation of auxiliary plant equipment due to use over time.

Based on the above criteria, WDES for Phase II proposes as BACT net heat rate for the Project of 7,582 Btu/kWh (HHV) for a GE turbine and 7,756 for a Siemens turbine, corrected to ISO conditions of: • Ambient Dry Bulb Temperature: 59ºF • Ambient Relative Humidity: 60% • Barometric Pressure: 14.69 psia

• Fuel Lower Heating Value: 20,549 Btu/lb LHV • Fuel HHV/LHV Ratio: 1.109 • Gas Turbines at base load with duct firing off

The proposed net heat rate for the Phase II turbine is consistent with other recent GHG BACT determinations for the Calpine Russell City Energy Center in Hayward, California, the Cricket Valley Energy Center in Dover, New York, the Hess, Newark Energy Center in New Jersey, the CPV Woodbridge Energy Center in NJ, the Calpine Channel Energy Center in Texas, the Calpine Deer Park Energy Center in Texas, and the LCRA Ferguson Plant in Texas which contain CO2e limits for each facility’s natural gas fired combined cycle power plant, and efficiency limits ranging from 7,522 to 7,730 Btu/kWh for the natural gas fired combustion turbines (operating at 100% load, ISO conditions and without duct firing). b. Technical Review of Proposed GHG Controls GHG Controls for Combustion Turbines and Duct Burners The Phase II proposes to operate the combustion turbine in combined cycle mode with natural gas as the exclusive fuel and proposes as BACT a heat rate limit of 7,582 Btu/kWh at full load ISO conditions without duct firing (based on net output) for a GE turbine and 7,756 Btu/kWh at full load ISO conditions without duct firing (based on net output) for a Siemens turbine. In addition to the heat rate limit, WDES is also proposing, as BACT, an emission limit of 947 lb CO2/MW-hr (gross), for the Siemens turbine and its associated duct burner and an emission limit of 955 lb CO2/MW-hr (gross), for the GE turbine and its associated duct burner. WDES – Phase II is also proposing an annual limit CO2e of 1,249,217 tons per year for the turbine and duct burner. This is the worst case value for either of the turbine models. Compliance with the proposed CO2e limit would be demonstrated through the use of CO2 CEMS along with fuel usage and emission factors for methane and nitrous oxide. The Department has reviewed the proposed BACT limits for GHGs for the Phase II turbine and found them to be BACT and they are consistent with other recent BACT determinations for similar size turbines. GHG Controls for Auxiliary Boiler To reduce GHG emissions from the auxiliary boiler, WDES is proposing followings:

1 . To use natural gas which has the lowest CO2 emissions compared to other combustion fuels, 2. To limit its operation 4 ,600 hours per year and

14

3. To operate it efficiently. 4. Annual limit CO2e of 10,946 tons per year

The proposed limits for GHGs have been reviewed by the Department and found to be BACT. GHG Controls for Emergency Engines A search of the RBLC for “carbon dioxide” did not yield any results for emergency diesel engines similar to that proposed for the WDES – Phase II Project. The reduction of GHG emissions from the emergency diesel generator and fire pump will be achieved by limiting the hours of operation. The proposed limits for GHGs have been reviewed by the Department and found to be BACT. VI. CASE-BY-CASE DETERMINATIONS This project is subject to BACT pursuant to 40 CFR 52.51 and to Lowest Achievable Emission Rate (LAER) pursuant to N.J.A.C. 7:27-18.3(b) VII. AIR QUALITY ANALYSIS

Bureau of Technical Services (BTS) has determined that emissions of criteria pollutants from WDES Phases I and II will not cause or significantly contribute to violations of the National and New Jersey Ambient Air Quality Standards, as well as the Class 1 and Class 2 PSD increments. In addition, the modeling has predicted no exceedances of NJDEP’s cancer and non-cancerous health guidelines due to its emissions of hazardous and air toxic pollutants. Because the project’s emissions caused impacts that exceeded the significant impact levels (SILs) for 1-hour NO2 and 24-hour PM2.5, a multisource air dispersion modeling analysis was conducted that included numerous other sources in the area as well as representative background air quality. No violations of the 24-hour PM2.5 NAAQS or Class 2 PSD increments were predicted. The modeling did identify violations of the 1-hour NO2 NAAQS. However, the WDES does not significantly contribute to these violations. The modeled violations will be addressed by the Department through better quantification of NOx emissions from the out-of-state sources, removal of those source in the proximity of the background NO2 monitor to avoid double counting, permit revisions to the sources in New Jersey that are predicted to have major contributions to the violations, and exploring the use of more refined techniques in estimating the NO to NO2 conversion rates in the atmosphere. The Air Dispersion Modeling and Risk Assessment are discussed in the attached Summary Memorandum.

VIII. BASIS FOR MONITORING AND RECORDKEEPING REQUIREMENTS The facility’s operating permit includes monitoring, recordkeeping and reporting requirements that are sufficient to demonstrate the facility’s continued compliance with the applicable requirements consistent with the following:

1. Provisions to implement the testing and monitoring requirements of N.J.A.C. 7:27-22.18, the recordkeeping and reporting requirements of N.J.A.C. 7:27-22.19, and all emissions monitoring and analysis procedures or compliance assurance methods required under the applicable requirements, including any procedures and methods promulgated pursuant to 40 CFR 64; and

2. Where the applicable requirement does not require direct periodic monitoring of emissions, the Department requires periodic monitoring of surrogate parameters sufficient to yield reliable data from the relevant time period that are representative of the facility's compliance with the permit. WDES will be required to conduct stack testing within 180 days of the startup of the turbines/duct burners and every five years to demonstrate the ability of the facility to operate within the approved emission limitations. WDES will be required to conduct stack testing of the auxiliary boiler within 180 days of startup to demonstrate the ability of the facility to operate within the approved emission limitations. Continuous Emission Monitors (CEMs) and recorders for NOx and CO will be required for the turbines/duct burners. The scope of the stack testing and CEMS is detailed in the facility specific requirements of the draft compliance plan. In addition, monitoring of fuel consumption is required for auxiliary boiler and monitoring of hours of operation for the emergency generator and fire pump are required for purposes of monitoring compliance

15

with annual emission limits. Monitoring of Total Dissolved Solids (TDS) concentration in the cooling tower circulating water is required for the cooling tower to monitor compliance with the TSP and PM10 limits.

3. In situations where the underlying applicable requirement did not specify any periodic testing or

monitoring, the following factors were considered in the evaluation and determination of the appropriate methodology for compliance demonstration for each emission unit: • Pollutant’s potential impact on public health and environment. • Emission unit and control device (older, less reliable equipment generally require more monitoring to

ensure ongoing compliance). • Compliance history and margin of compliance. • Emissions variability and process stability (emissions units with highly variable process rates or

materials generally require more monitoring to ensure ongoing compliance) • Quantity of emissions (emissions units that will have more impact on the environment generally

require more monitoring to ensure ongoing compliance).

IX. APPLICABLE STATE AND FEDERAL RULES N.J.A.C 7:27-18 – offset requirements The WDES – Phase II project was determined to be subject to N.J.A.C 7:27-18 for emissions of NOx, CO, VOC, TSP, and PM10. The WDES – Phase II p ro j ec t is subject to N.J.A.C 7:27-18 for NOx and VOC as the potential emissions of these two ozone precursors are greater than 25 tons per year (the threshold for severe ozone nonattainment, which applies to the entire state of New Jersey). The potential emissions of NOx, CO, VOC, TSP, and PM10 are also greater than N.J.A.C 7:27-18 thresholds. Gloucester County is non-attainment for ozone (precursors NOx and VOC). Applicable requirements include application of LAER technology and acquisition of emission offsets. Compliance with LAER has been discussed in Sections C and D above. Offset requirements are discussed below. This modification shows an increase of 80.8 tons per year (TPY) of NOx emissions and 30.3 TPY of VOC emissions during the contemporaneous period (last five years). This increase is greater than the Sub 18 significant net emission threshold of 25 TPY. The facility is therefore required to purchase 105 and 39.4 tons of NOx and VOC, respectively, offsets pursuant to N.J.A.C 7:27-18, at an offset ratio of 1:3 tons reduction to 1.0 ton of NOx and VOC emission increase. The applicant has proposed to obtain 105.0 tons of NOx and 39.4 tons of VOC offsets from AFG Industries, Cinnaminson, Burlington County and Griffin Pipe Products, Florence, Burlington County, respectively. The Department has verified that those emission reductions would comply with the provisions at N.J.A.C 7:27-18.5. If this permit is approved, the offsets would need to be secured and transferred to the applicant through a permit significant modification prior to initiation of operation, pursuant to N.J.A.C. 7:27-18.3(f). In accordance with N.J.A.C. 7:27-18.3(c)2, WDES has conducted an analysis of alternative sites within New Jersey and considered alternative sizes, production processes, including pollution prevention measures and environmental control techniques, demonstrating that the benefits of the newly constructed WDES outweigh the environmental and social costs imposed as a result of the location, construction, and operation of the WDES. The Department has found that the benefits of the WDES will significantly outweigh the potential environmental and social costs imposed as a result of construction and operation of the WDES.

16

Other NJ State Requirements: This modification is subject to New Jersey Air Pollution Control Regulations, codified in N.J.A.C. 7:27-1 through 34, as applicable, including N.J.A.C. 7:27-18 as discussed above. A complete text of these regulations is available at: http://www.nj.gov/dep/aqm/rules27.html 40 CFR 52.21 PSD Requirements: DEP has determined that the WDES – Phase II is classified as a major modification to a PSD source and is subject to all applicable requirements of the federal PSD regulations codified at 40 CFR 52.21. PSD applicability is determined on an individual pollutant basis. Based on the potential annual emissions in Table 2, the WDES – Phase II was determined to be subject to PSD requirements for emissions of NOx, CO, PM, PM10, PM2.5 and greenhouse gases (CO2e).

In addition to PSD regulations codified at 40 CFR 52.21, the WDES – Phase II is subject to the following subparts of NSPS codified at 40 CFR 60 and MACT codified at 40 CFR 63. This modification is also subject to Federal regulations listed below. NSPS - Subpart A: New Source Performance Standards General Provisions; NSPS - Subpart Dc: NSPS for industrial steam generating units greater than or equal to 10 MM Btu/hr

but less than 100 MM Btu/hr (auxiliary boiler); NSPS -Subpart IIII: NSPS for stationary Compression Ignition internal combustion engine, NSPS - Subpart KKKK: NSPS for stationary gas turbines. MACT - Subpart ZZZZ Maximum Achievable Control Technology for Stationary Reciprocating Internal

Combustion Engines Acid Rain Program at 40 CFR 72 The Acid Rain Permit is proposed pursuant to the air pollution control permit provisions of Title IV of the federal Clean Air Act, federal rules promulgated at 40 CFR 72, and state regulations promulgated at N.J.A.C. 7:27-22. These rules require facilities operating “affected units” that are subject to the Acid Rain Program to obtain an Acid Rain Permit for those units. Pursuant to Title IV of the Clean Air Act, the United States Environmental Protection Agency (USEPA) has not previously approved sulfur dioxide allowances for the Unit 3 proposed for WDES – Phase II. Each allowance provides authorization to emit up to one ton of sulfur dioxide during a specified calendar year. In accordance with USEPA’s rules, WDES – Phase II may sell or purchase allowances on the open market in order to more accurately reflect current operation. The representatives for WDES are Gordon Holk and Douglas Mulvey. X. Environmental Justice (EJ) Analyses Pursuant to Executive Order (“EO”) 12898 of February 11, 1994 (Federal Actions to Address Environmental Justice (EJ) in Minority Populations and Low-Income Populations), DEP requires PSD Applicants to provide information necessary to determine if the project is subject to any specific federal executive order or State initiatives and policies regarding overburdened communities. The Applicant conducted an Environmental Justice (EJ) analysis to determine whether the construction and operation of the WDES – Phase II would have a significant adverse and disproportionate effect on an “environmental justice community.” The EO requires federal agencies to consider disproportionately high adverse human health or environmental effects of their actions on minority and low income populations. New Jersey has a similar order in Executive Order 131 that requires state agencies, including NJDEP, to allow for public participation in decisions that affect environmental quality and public health. The “USEPA Region 2 Interim Environmental Justice Policy” (USEPA, 2000) (Interim Policy) provides guidance in conducting the EJ Analysis. The Interim Policy defines an “EJ Community” as:

17

http://www.nj.gov/dep/aqm/rules27.html

“A minority and/or low income area suffering a disproportionate and adverse environmental burden as a result of the unfair or unequal development, implementation, or enforcement of environmental laws, regulations or policies.” An “adverse environmental burden” is defined by the Interim Policy as: “When there is an acknowledged health or welfare standard for the burden in question, the burden is adverse only when it exceeds that standard. When there is no standard, the decision is based on additional site-specific analysis” WDES Phase II utilized the Interim Policy guidelines to develop its EJ Analysis. An area within 3 miles of the WDES was evaluated. The EJ analysis did not indicate any environment justice communities within this analysis area. On the basis of the available information, the proposed project does not have a disproportionately high and adverse human health or environmental effects on any environmental justice community XI. Endangered Species Analysis WDES Phase II is located in West Deptford Township. Swamp pink (federally threatened) has historically occurred in West Deptford Township, Gloucester County, New Jersey. Because no suitable habitat is on the WDES – Phase II site, no adverse effects will occur to swamp pink or its habitat from the proposed Project. No adverse effects are anticipated on other federally listed species either. The PSD air quality modeling analyses are discussed in detail above and in the attached Summary Memorandum. WDES Phase II project emissions are in compliance with the NAAQS, NJAAQS, and PSD Class I and Class II increments, and will not have an adverse impact on soils, vegetation, or visibility. XII. FACILITY’S COMPLIANCE STATUS The Responsible Official at the facility has certified that the facility currently meets all applicable requirements of the Federal Clean Air Act and the New Jersey Air Pollution Control Act. Based on this certification, the Department’s evaluation of the information included in the facility’s application, and a review of the facility’s compliance status, the Department has concluded that this air pollution control operating permit should be approved. Prior to the expiration of the Operating Permit’s five-year term, the facility will be required to apply for a renewal, at which time the Department will evaluate the facility and issue a public notice with its findings. XIII. ATTACHMENT Memorandum dated March 31, 2014, of the air dispersion modeling and risk assessment summary from the Bureau of Technical Services. XIV. EXEMPT ACTIVITIES The facility’s operating permit does not include exempt activities such as office and interior maintenance activities, maintenance shop activities, food preparation facilities, cafeterias and dining rooms, etc. A complete list of exempt activities, as allowed by the Operating Permit rule, can be found at N.J.A.C. 7:27-22.1.

18

MEMORANDUM TO: Bashir Bouzid, Bureau of Air Permits FROM: Joel Leon, Bureau of Technical Services

DATE: March 31, 2014

SUBJECT: West Deptford Energy Station, Refined Modeling PSD Impact Analysis West Deptford Township, Gloucester County PI# 56078, BOP 12-0001

The Bureau of Technical Services (BTS) has completed its review of the West Deptford Energy Station Refined Modeling PSD Impact Analysis. This modeling analysis was performed in support of a Prevention of Significant Deterioration (PSD) air permit. The current West Deptford Energy Station consists of a combined cycle 700 MW power plant which is under construction and will soon begin operations (known as Phase 1). The Phase 1 power plant includes two Siemens F5 combustion turbines that will exclusively fire natural gas. Each turbine will exhaust to a heat recovery steam generator equipped with natural gas-fired duct burners. Control devices include dry low-NOx combustors, selective catalytic reduction (SCR) system, and oxidation catalyst. In addition to the combustion turbines, other sources of air emissions at the facility are two wet mechanical draft cooling towers, an auxiliary boiler, an emergency diesel generator, and an emergency diesel fire pump. This modeling analysis was submitted because of the proposed Phase 2 expansion at the West Deptford Energy Station. This modification will add 427 MW and consist of a third gas-fired combustion turbine, steam turbine generator and ancillary combined cycle equipment. The combustion turbine will fire natural gas and either be a Siemens SGT6-5000F(5) or a General Electric 7FA. Phase 1 and 2 in combination will bring the total energy output for the facility to a nominal 1100 MW. The proposed project was subject to PSD review for nitrogen oxides (NOx), carbon monoxide (CO), sulfur dioxide (SO2), sulfuric acid mist, particulate matter less than 10 microns (PM-10) and fine particulate (PM2.5). The project will be subject to nonattainment new source review for

DEPARTMENT OF ENVIRONMENTAL PROTECTION

CHRIS CHRISTIE CLIMATE AND ENVIRONMENTAL MANAGEMENT BOB MARTIN Governor DIVISION OF AIR QUALITY Commissioner

P.O. Box 420 Mailcode 401-02 KIM GUADAGNO TRENTON, NJ 08625-0420 Lt. Governor 609 - 984 - 1484

New Jersey is an Equal Opportunity Employer l Printed on Recycled Paper and Recyclable 19

NOx and volatile organic compounds (VOC). Facility-wide (Phase 1 and 2) emissions of criteria pollutants are as follows: NOx = 240 tons/yr, CO = 592 tons/yr, SO2 = 41 tons/yr, PM-2.5 = 144 tons/yr, PM-10 =150 tons/yr, and H2SO4 = 8.5 tons/yr. Air quality impacts were evaluated for the combined emissions of both Phase 1 and Phase 2.

• The modeling analysis predicted that the air quality impacts from the emission increases of CO, annual NO2, PM10, annual PM2.5 and SO2 will be below their respective significant impact levels (SIL). Consequently, the proposed modification is not projected to cause or contribute to a violation of a PSD Class II increment or the New Jersey or National Ambient Air Quality Standards (NJAAQS/NAAQS) for these pollutants and averaging times. PM2.5 24-hour emissions during normal operations and NO2 1-hour emissions during periods of hot start-up were both predicted to exceed their SILs.

• A multi-source modeling analysis for PM2.5 was conducted which included other sources

in New Jersey, Philadelphia, Delaware and Pennsylvania. This modeling predicted that emissions from all sources included in the multisource modeling with background added will not cause a violation of the PM-2.5 24-hour NAAQS or Class II PSD increment.

• A multi-source modeling analysis for NO2 was conducted that included other sources in New Jersey, Philadelphia, and Pennsylvania. While the multisource NO2 modeling predicted violations of the NAAQS, emissions from the proposed project did not significantly contribute to the modeled violations of the 1-hour NO2 NAAQS. These modeled violations will be addressed by the Department through better quantification of NOx emissions from the out-of-state sources, removal of those source in the proximity of the background NO2 monitor whose impact is being double counted, permit revisions to the sources in New Jersey that are predicted to have major contributions to the violations, and exploring the use of more refined techniques in estimating the NO to NO2 conversion rates in the atmosphere.

• The health risks from air toxic emissions of the proposed facility will be negligible.

• The project’s air quality impact on the Brigantine National Wildlife Refuge Class I area

will not adversely affect the air quality related values (AQRV) or violate a Class I PSD increment.

A summary of the modeling analysis conducted for the proposed project is attached. c: Frank Steitz (DAQ)

Alan Dresser (BTS) Greg John (BTS) Dave Owen (BAP) Piyush Desai (BAP)

20

Air Quality Dispersion Modeling Analysis of the West Deptford Energy Station, March 2014

DOCUMENTS Refined Modeling Report - PSD Impact Analysis and Multisource Modeling Protocol for West Deptford Energy Station (dated September 30, 2013, revised January 8, 2014), prepared by Enviromet, Division of AKRF Incorporated. Refined Modeling Report, PSD Impact Analysis, Multisource Modeling Protocol for West Deptford Energy Station, Supplement 1 (dated November 25, 2013), prepared by Enviromet, Division of AKRF Incorporated. Multisource Dispersion Modeling Report for West Deptford Energy Station (dated January 13, 2014), prepared by Enviromet, Division of AKRF Incorporated. West Deptford Energy Station Phase II Transmittal of Supplement to Multisource Dispersion Modeling Report Supplement 1(dated February 19, 2014), Supplement 2 (dated March 10, 2014, and Supplement 3 (dated March 18, 2014), prepared by Enviromet, Division of AKRF Incorporated. CONCLUSION BTS has determined that emissions of criteria pollutants from West Deptford Energy Station Phases 1 and 2 will not cause or significantly contribute to violations of the National and New Jersey Ambient Air Quality Standards, as well as the Class I and Class II Prevention of Significant Deterioration (PSD) increments. In addition, the modeling has predicted no exceedances of NJDEP’s cancer and non-cancerous health guidelines due to its emissions of hazardous and air toxic pollutants. Because the project’s emissions caused impacts that exceeded the significant impact levels (SILs) for 1-hour NO2 and 24-hour PM-2.5, a multisource air dispersion modeling analysis was conducted that included numerous other sources in the area as well as representative background air quality. No violations of the 24-hour PM-2.5 NAAQS or Class II PSD increments were predicted. The modeling did identify violations of the 1-hour NO2 NAAQS. However, the West Deptford Energy Station does not significantly contribute to these violations. The modeled violations will be addressed by the Department through better quantification of NOx emissions from the out-of-state sources, removal of those source in the proximity of the background NO2 monitor whose impact is being double counted, permit revisions to the sources in New Jersey that are predicted to have major contributions to the violations, and exploring the use of more refined techniques in estimating the NO to NO2 conversion rates in the atmosphere. PROJECT DESCRIPTION West Deptford Energy Station proposes to add a third combustion turbine to its 700 MW combined cycle power plant located in West Deptford, Gloucester County, New Jersey. The

1

initial operating permit issued May 2009 was for Phase I of the project (two Siemens F5 combustion turbines, each exhausting to a heat recovery steam generator equipped with natural gas-fired duct burners. Control devices include dry low-NOx (DLN) combustors, selective catalytic reduction (SCR) system, and oxidation catalyst. In addition to the combustion turbines, other sources of air emissions at the facility are a mechanical draft cooling tower, 40 MMBtu/hr auxiliary boiler, a 1 MW emergency diesel generator, and a 282 hp emergency diesel fire pump. The significant modification to the current permit (Phase 2) consists of one “F-class” General Electric or Siemens combustion turbine, a heat recovery steam generator, cooling tower and auxiliary boiler and ancillary equipment. WEST DEPTFORD ENERGY STATION POTENTIAL-TO-EMIT The proposed project is subject to PSD review for nitrogen oxides (NOx), carbon monoxide (CO), particulate matter less than 10 microns (PM-10) and sulfuric acid mist (H2SO4). The potential to emit (PTE) annual emission limits for the facility are listed in Table 1.

Table 1. West Deptford Energy Station Maximum Potential Emissions and Significance Thresholds

Pollutant Phase I PTE (TPY)

Phase 2 PTE a (TPY)

Facility PTE (TPY)

PSD Significant Emissions Thresholds (TPY)

Carbon Monoxide 385.5 b 206.5 592 100 Nitrogen Oxides (NOx) 158.8 b 80.8 239.6 40/25 c

Particulate Matter (PM-10) 93.0 57.3 150.3 15 Particulate Matter (PM-2.5) 89.7 54.8 144.5 100 Sulfur Dioxide (SO2) 26.8 14.5a 41.3 40 Lead 0.00057 0.00066 0.0012 0.6 VOCs 54.2 30.3 84.5 40/25 c Sulfuric Acid Mist 5.3 3.2 8.5 7 Total HAPs 6.6 3.9 10.5 d 10

a. Reflects the higher of either the Siemens F5 or GE F7A combustion turbines. b. Reflects BOP13-0003 modification to Phase 1, pervious emissions were 548 tons/yr for CO and 169

tons/yr for NOx. c. Represents Subchapter 18 significant emission increase threshold. d. Includes acrolein, formaldehyde, lead, and toluene.

STEADY-STATE TURBINE OPERATION The air quality impact during steady-state operation was evaluated for a variety of turbine operating loads at ambient temperatures ranging from -5 oF to 104 oF. Thirty-six operating scenarios were identified to be representative of the full range of proposed operation. For the load screening, the maximum 24-hour PM2.5/PM10 impacts occur under the scenarios listed below:

Phase I – Combustion turbine only (no duct burner) at 60% load and 104◦F Phase II (Siemens) – Combustion turbine with duct burner at 100% load and 0◦F Phase II (GE) – Combustion turbine with duct burner at 100% load and 100F

2

The maximum 1-hour NO2, CO, and SO2 impacts and maximum 8-hour CO impact occur under the load screening scenarios listed below:

Phase I and II (Siemens) – Combustion turbine with duct burner at 100% load and -5◦F Phase II (GE) – Combustion turbine with duct burner at 100% load and 59◦F

The scenarios predicted to have the highest ambient impacts were used to determine whether the project had a significant impact. OTHER FACILITY EMISSION SOURCES Other emission sources at the facility include a Phase 1 natural gas fired auxiliary boiler, a Phase 1 12-cell mechanical draft wet cooling tower, a Phase 2 8-cell mechanical draft wet cooling tower, and a Phase 2 natural gas auxiliary boiler. There is also an emergency generator and a fire pump both fired by ultra-low sulfur diesel fuel. The Applicant will comply with the emergency generator and fire pump requirements outlined in Division of Air Quality’s July 29, 2011 memorandum. The emergency generator and the fire pump are therefore exempt from the 1-hr NO2 and SO2 modeling requirements and will be only modeled for annul NO2 impacts. CRITERIA POLLUTANT EMISSION RATES AND STACK PARAMETERS Normal Operations Table 2 lists the short-term normal operations emissions modeled to determine significance and demonstrate compliance with ambient air quality standards. Emission rates are listed for each source type at the proposed facility. Table 3 lists the annualized emission rates for the three combustion turbines. The operating case using the Siemens equipment Phase 2 alternative was selected for the annual modeling based upon the results of the load screening. The annualized emission rates for the auxiliary boilers were assumed identical to the short-term emission rates. This is a conservative assumption because the auxiliary boilers will be limited by permit condition to operating no more than 4,600 hours per year. Table 4 lists the stack parameters for the sources modeled.

Table 2. Normal Operation /Worst-Case Scenario Short-term Emissions (lb/hr)

Source CO NOx SO2 PM-10 PM-2.5 Phase I Turbines

(per turbine) 12.03 19.76 5.64 15.0 16.64 a

Phase 2 Turbine 14.0 23.0 6.56 21.55 23.45 a Phase I Aux. Boiler 1.44 1.40 0.084 0.2 0.29 a Phase II Aux Boiler 1.44 0.44 0.084 0.2 0.23 a

Phase I Cooling Tower (per cell – 12 total)

-

-

-

0.102

0.038b

Phase II Cooling Tower (per cell – 8 total)

-

-

-

0.113

0.043b

a. Included are 8 % of SO2 emitted converted to sulfate and 6% of NOx emitted converted to nitrate. b. Per BTS policy, total cooling tower PM-2.5emissions of less than 1 lb/hr are considered negligible.

3

Table 3. West Deptford Energy Station Normal Operation/Annual Emissions (lb/hr)

Emission Source Phase 1 Turbines a

(per turbine) Phase 2 Siemens a

Turbine

NOx (lb/hr) 18.84 20.14

PM10 (lb/hr) 9.95 12.05

SO2 (lb/hr) 3.06 3.31

PM2.5 (lb/hr)b 11.33 13.53 a. Reflects emissions with duct firing at 100 percent load at 59 deg. F ambient temperature. b. PM2.5 emissions include secondary PM2.5

Table 4. Source Location and Stack Parameters For Annual Average Modeling at West Deptford Energy Station

SOURCE ID

UTM-East Coord.

(meters)

UTM-North Coord.

(meters)

BASE ELEV.

(meters)

STACK HEIGHT (meters)

STACK TEMP.

(degrees K)

STACK EXIT VEL.

(m/s)

STACK DIAMETER

(meters) Phase I HRSG 1 a

481077.86 4410211.7 5.0 64.01 361.48 26.45 6.1

Phase I HRSG 2 a

481117.15 4410211.7 5.0 64.01 361.48 26.45 6.1

Phase II HRSG

480980.49 4410211.4 5.8 64.01 356.48 21.73 6.1

Phase I Auxiliary Boiler

481060.91

4410141.7

5.0

38.10

422.04

15.10

0.69

Phase II Auxiliary Boiler

480967.49

4410150.8

5.8

33.53

422.04

15.10

0.69

Phase I Cooling Tower (per cell) b

481043.30

4410275.0

5.0

18.59

5 deg F above

ambient

6.86

10.26

Phase II Cooling Tower (per cell)b

480972.00

4410266.0

5.0

18.59

5 deg F above

ambient

6.86

10.26

a. Combustion turbine stack parameters represent peak load with duct firing at 59 F ambient temperature for all averaging periods with the exception of 24-hour PM10/PM2.5 for which the worst-case operating scenario of 50% load and no duct firing at 59 F ambient temperature was modeled.

b. Parameters are provided per cell. Coordinates are provided for the first cell. Startup and Shutdown The facility emissions and stack parameters during the transient conditions of start-up and shutdown differ from those during normal, steady-state operation. Short-term emissions of PM-2.5, PM-10, and SO2 are highest during steady state operation and are therefore not modeled for startup/shutdown scenarios. During transient conditions, NO2 and CO emissions can be

4

significantly greater than those during steady-state operation. For this reason, CO (1-hour) and NO2 (1-hour) are required to be modeled for the startup/shutdown impact. The length of the startup process depends on the length of time that the unit has been shut down. The worst-case start (cold-start) takes three hours to complete. For warm start and hot start, this time period is shorter. For NO2, startup characteristics significantly affect emissions, even during hot startup. Thus, West Deptford Energy Station will accept a permit restriction that all three combustion turbine generators undergo a staggered start. This means that the Phase II unit will not start in the same hour as the Phase I turbines. For NO2, the staggered start was conservatively modeled as a scenario where there is a 30 minute delay between the start of turbines, and the Phase 2 turbine (GE or Siemens) is operated at worst-case steady state for the entire hour averaging period. Because the 1-hour NO2 NAAQS is a statistical standard, the 1-hour NO2 impact was not assessed for cold start and warm start since they occur infrequently (a maximum of 10 and 50 times a year) and are not expected to significantly affect the 98% of the annual distribution of the daily maximum 1-hr NO2 concentration. The 1-hour NO2 impacts were only accessed during hot startup which will occur up to 350 times per year. For CO, it was conservatively assumed that both Phase 1 turbines were starting up and shutting down simultaneously in the modeling. Because the CO NAAQS is not a statistically based standard all startup (cold, warm, hot) modes were evaluated. Load screening results showed that the impacts from the Siemens combustion turbine generator (CTG) were higher than the General Electric CTG, therefore, the emissions and stack parameters for the Phase 2 Siemens CTG were modeled for all of the transient scenarios. Table 5 lists the parameters for each of four scenarios (cold start, warm start, hot start and shutdown), and for each mode. West Deptford Energy Station modeled a maximum of 25 cold starts, 350 hot starts, 50 warm starts and 350 shutdowns.

Table 5. Parameters and Emission Rates during Turbine Startup (per turbine)

Units

Cold Start Warm Start Hot Start Shutdown

GE

Siemens

GE

Siemens

GE

Siemens

GE

Siemens Duration Mins. 180 120 50 40

No. per year # 25 50 350 350

Exit Temperature oF 166 173 166 173 186 173 166 173

Exit Velocity ft/s 35.9 64.0 48.7 64.0 79.9 64.0 79.9 64.0

NOx Lead Turbines

lbs/hr 167 176 94 113 42 49 32 40

Lag Turbine

lbs/hr 89 117 59 74 21.09 22.32 32 40

CO Lead Turbines

lbs/hr 1,950 a 1,310 a 638 a 260 a

Lag Turbine

lbs/hr a 13.6 a 13.6 a 13.6 a 13.6

a. For CO, Siemens emissions were modeled for both the GE and Siemens scenarios, because the Siemens equipment has higher emissions under all of the scenarios listed.

5

BACKGROUND AIR QUALITY The most recent three years of available monitored background concentrations at representative background monitoring stations are listed in Table 6. With the exceptions of PM2.5, 1-hr NO2 and 1-hr SO2, the selected background concentrations are the highest annual and the highest-second-highest short-term values.

Table 6. Background Air Quality (µg/m3) Pollutant Averaging

Time Monitoring

Station

2010

2011

2012 NO2 Annual Chester, PA 47.4 47.1 42.9

1-hour (a) Chester, PA 94.1 86.5 86.5 SO2 Annual Chester, PA 17.2 7.7 5.2

1-hour (b) Chester, PA 89 34 39.3 3-hour Chester, PA 77.5 31.4 34.8 24-hour Chester, PA 34 22.2 22.8

CO 1-hour Camden 802 687 2,519 8-hour Camden 458 344 1,603

PM-10 24-hour Chester, PA 70 45 39 PM-2.5 Annual Gibbstown 9.3

24-hour Gibbstown 20.8/15.2/22.3/16.6 (c) (a) 1-hour 3-year average 98th percentile value for NO2 is 89 ug/m3

(b) 1-hour 3-year average 99th percentile value for SO2 is 54.1 ug/m3 (c) PM2.5 24-hour 3-year average 98th percentile value for winter, spring, summer, and fall, respectively

For PM2.5, three-year averages of the monitored annual average concentrations and the 98th percentile seasonally averaged 24-hr average concentrations are used as the annual and the 24-hr backgrounds, respectively. Three year averages of the 98% (99% for SO2) of annual distribution of daily maximum 1-hour average concentrations are used as the 1-hour NO2 and 1-hour SO2 background concentrations. It should be noted that the use of the 98% of the 1-hour NO2 calculated on an annual (not seasonal and/or hourly) represents a conservative Tier 1 estimate of the background value. The background values in Table 6 were added to the modeled impacts to demonstrate compliance with the National and the NJ Ambient Air Quality Standards. MODELING METHODOLOGY The air quality modeling analysis of the facility impacts was performed using the U.S. Environmental Protection Agency’s atmospheric dispersion model AERMOD (version 12345). Land use in the immediate vicinity of the site was determined to be urban. A Cartesian Grid of 2,069 receptors was modeled. The grid was centered on the Phase 2 HRSG stack. The receptors were spaced as follows: - 50 m spacing at the fenceline out to approximately 500 m -100 m spacing from 500 m out to a distance of 1,500 m -250 m spacing from 1,500 m out to a distance of 3,000 m -500 m spacing from 3,000 m out to a distance of 5,000 m

6

-1,000 m spacing from 5,000 m out to 10,000 No receptors were placed on the facility property. Receptor elevations were assigned by using the EPA’s AERMAP program (version 11103) to extract elevation data from the USGS 30 meter digital elevation model files. The latest version of BPIP-PRIME (version 04274) was used to assess the building downwash effect. The GEP stack height for the two turbines is calculated at 275 ft. The 210-foot turbine stacks are modeled for building downwash influence on pollutant ground level concentrations. Five years of meteorological data (2008 – 2012), with surface observations taken at Philadelphia International Airport, located 2 kilometers north of the facility, and upper air data from Dulles International Airport, which is approximately 217 kilometers southwest of the facility, were used in the modeling. The meteorological data used for the air quality modeling analysis is considered representative of the West Deptford Energy Station site. In the single-source modeling of the West Deptford Station, EPA’s Tier 2 NO to NO2 conversion factors of 0.8 for the 1-hour impacts and 0.75 for the annual impacts was applied to the predicted NOx values. The Auer Analysis and previous modeling conducted for West Deptford Energy Station has been classified as rural. However, recent AERMOD guidance has addressed instances when the classic Auer Analysis involves a large body of water and the modeling domain just outside the 3 kilometers radius is characterized by strikingly different land use. For West Deptford Energy Station, 20% of the 3 kilometer radius is open water and the modeling domain includes a large portion of the Philadelphia-Camden-Wilmington Metropolitan Statistical Area. Within 20 km of the West Deptford site the population is approximately 1,750,000, or 1,393 people per km2. The AERMOD Implementation Guide states that where the MSA is not clearly identified, urban classification is justified where the population density exceeds 750 per km2. The population and possibility that a heat island effect exists within the modeling domain supports the use of urban dispersion coefficients for this modeling analysis. Justification for the 1-hour NO2 and the 24-hour and Annual PM-2.5 Interim Significant Impact Levels (SILs)

There is currently no EPA proposed or promulgated SILs for the 1-hour NO2, the 24-hour PM-2.5, or the annual PM-2.5 NAAQS. As a result, BTS has developed interim SIL levels for the purposes of demonstrating that an air quality impact is insignificant and would not significantly cause or contribute to a violation.

BTS has been using the 10 ug/m3 value as the NO2 interim SIL in its air permit reviews since April 2010. It was at that time the Northeast States for Coordinated Air Use Management (NESCAUM) adopted this value as an interim SIL. The technical justification for use of this value is contained in the document NESCAUM Recommendations on the Use of an Interim Significant Impact Level (SIL) in Modeling the 1-Hour NO2 NAAQS (dated April 21, 2010). The document is in Attachment 1 of this memo. BTS believes this value provides the proper balance between the BTS resource burden of the review of a multisource analysis and protecting

7

the NAAQS from being significantly impacted by single source. As a result, a 1-hour NO2 interim SIL of 10 ug/m3 has been used in the modeling analysis.

EPA’s promulgated PM-2.5 annual SIL of 0.3 ug/m3 and 24-hour SIL of 1.2 ug/m3 were vacated by the D.C. Circuit Court on January 22, 2013. However, given the background PM-2.5 levels in the vicinity of the West Deptford Energy Station the use of these SILs in this analysis is technically justified.

WEST DEPTFORD ENERGY STATION MODELING RESULTS The air quality impacts from the proposed West Deptford Energy Station combustion turbines were examined along with ancillary equipment which included duct firing within the HRSGs, the auxiliary boilers, and the cooling towers. Startup and shutdown conditions were also evaluated. Unless otherwise noted, the impacts reported represent the highest of either the Siemens or GE Phase 2 combustion turbine. With the exception of 24-hr PM2.5, 1-hr SO2 and 1-hr NO2, all pollutant concentrations used for comparison to the SILs utilize the maximum predicted concentration at any receptor over the entire 5-yr modeling period for each applicable averaging period. For 24-hr PM2.5, each year’s maximum 24-hr impact value at each individual receptor is averaged over the five years. The highest 5-yr average value across all receptors was used to compare with the 24-hr PM2.5 SIL of 1.2 ug/m3. For 1-hr NO2 significant impact analysis, the maximum 1-hr impact for each year was predicted and then averaged over the 5-yr modeling period for each receptor. The highest 5-yr average value across all receptors was used to compare with the interim 1-hr NO2 SIL of 10 ug/m3. The only pollutant and averaging time exceeding a SIL level during normal operations was the predicted 24-hour PM-2.5 impact of 2.0 ug/m3. The location of the maximum 24-hour PM-2.5 impact was 758 meters northeast of the Phase 2 turbine stack. The three combustion turbines were responsible for 90 percent of the impact at that location. The radius of the significant impact area was approximately 1.5 km. Modeling results for short-term NO2 during hot startup predicted an impact of 29.9 ug/m3 for the Siemens equipment and 30.0 ug/m3 for the GE equipment. Both of these exceed the SIL of 10 ug/m3 out to a distance of 5.8 km in all directions. Modeling results for short-term CO emissions showed that the transient operation case that results in the highest impacts was the cold start. One-hour CO impacts for the cold start condition were predicted to be 1,250.6 ug/m3 which is 63% of the 1-hour SIL of 2000 ug/m3. Therefore, CO required no further analysis. In summary, the results of the single source modeling indicate that the facility had modeled concentrations below the significant impact levels (SILs) for CO, PM-10, annual PM-2.5, SO2, and annual NO2 in both Class I and Class II areas and adequately demonstrated compliance with the NAAQS and PSD Class II increments. Results show that 1-hour NO2 and 24-hour PM-2.5 impacts were above the SIL. In Table 7 the facility’s predicted impacts are compared to their respective SILs.

8

Table 8 shows that when those pollutants with insignificant impacts are added to background concentrations, the proposed emissions increase due to the proposed West Deptford Energy Station will not cause or contribute to a violation of a NAAQS or NJAAQS. Table 9 shows emissions from the proposed facility will not cause a violation of the existing Class 2 PSD increments. Even though the source will have a significant impact, no other PM-2.5 sources in the area will consume increment and therefore no multisource modeling is needed for the purposes of determining compliance with the PM-2.5 PSD Class 2 increments. This is the first major stationary source or a major modification subject to PSD regulations that has submitted a complete application in this air quality region after the area was designated attainment for PM-2.5. As a result, the application has set the PM-2.5 Minor Source Baseline Date for this region.

TABLE 7 Comparison of West Deptford Energy Station’s Impacts to SILs

Pollutant

Avg. Time

Max. Predicted Concentration

(ug/m3) (a)

Significance

Level (ug/m3)

SO2

1-hour 2.1 7.8

3-hour 1.9 25

24-hour 0.7 5

Annual 0.03 1

NO2 1-hour 30.0 c (hot startup)

7.0 c (normal operations) 10

Annual 0.35 1

CO 1-hour 1251 (cold startup)

12.7 (normal operations) 2,000

8-hour 9.3 500

PM-2.5 24-hour 2.0 c 1.2

Annual 0.14 0.3

PM-10 24-hour 3.6 5

a. Impacts represent normal turbine operation unless otherwise noted. b. Impacts above SIL are in bold. c. 5-year average of each year’s maximum impact at a given receptor.

9

TABLE 8 Comparison of West Deptford Energy Station’s Impacts to NJAAQS/NAAQS

Pollut-ant

Avg. Time

Max. Pred Conc.

(ug/m3)(a)

Background Air Quality / Monitor

(ug/m3)

Total Impact

(ug/m3)

NAAQS (ug/m3)

SO2

1-hour 1.7 54.1 / Chester, PA 55.8 197

3-hour 1.8 77.5 / Chester, PA 79.3 1,300

24-hour 0.6 34 / Chester, PA 34.6 260(c)/365

Annual 0.03 17.2 /Chester, PA 17.2 60(c)/80

NO2 1-hour 24.2 (hot startup)

5.9 (normal) 89 / Chester, PA b (hot startup) 94.9 (normal) 188

Annual 0.35 47.4 / Chester, PA 47.8 100

CO 1-hour 1130 (cold startup)

12.3 (normal) 2,519 / Camden 3,649 (cold startup) 2,531 (normal) 40,000

8-hour 8.2 1,603 / Camden 1,611 10,000

PM-2.5

24-hour 1.4 Varies seasonally / Gibbstown b 35

Annual 0.14 9.3 / Gibbstown 9.4 15

PM-10 24-hour 3.5 70 / Chester, PA 73.5 150

a. Values represent maximum predicted average calculated as appropriate for NJAAQS or NAAQS. b. Multisource modeling conducted to determine the total impact.

c. Values represents the secondary New Jersey ambient air quality standard (NJAAQS).

Table 9. Comparison of Maximum with PSD Class 2 Increment

Pollutant Averaging

Period Maximum Class II Impact a (µg/m3)

PSD Class II Increment (µg/m3)

SO2 3-hr 1.82 512

SO2 24-hr 0.62 91

SO2 annual 0.03 20

NO2 annual 0.35 25

PM2.5 24-hr 2.19 9

PM2.5 annual 0.14 4

PM10 24-hr 3.6 30

PM10 annual 0.71 17

a. Impacts represent the maximum concentration modeled in any receptor during the five years.

10

MULTISOURCE MODELING METHODOLOGY Inventory The multisource modeling analysis was conducted for the two pollutants and averaging times whose impacts exceeded the SILs, i.e. the 1-hour NO2 during hot startup and 24-hour PM-2.5 concentrations. The multisource modeling was performed to assess the impacts of the West Deptford Energy Station plus other major sources of PM-2.5 and NO2 in the surrounding region. Multisource impacts were modeled using the worst-case operating scenarios for the single source modeling, including the hot-start and shutdown scenarios. A modeling methodology similar to that used in the single-source modeling was used in the multisource modeling. Because a newer version of AERMOD and AERMET (version 13350) was released before the multisource modeling protocol was approved, AERMET version 13350 was used to regenerate the 2008 through 2012 meteorological data and AERMOD version 13350 was used in the multisource modeling. Receptors were only placed within the significant impact area of 1.5 km for the 24-hour PM-2.5 modeling and 5.8 km for the 1-hour NO2 where the single source modeling predicted SIL exceedances. For NO2, inventory information was compiled for point sources at facilities located at up to a distance of 20 – 30 km from the West Deptford Energy Station. Point sources beyond a distance of roughly 10 - 12 km from the WDES were modeled using a screening procedure (AERSCREEN) and were not included if AERSCREEN predicted a non-significant impact at the edge of the SIA. For PM2.5, inventory information was compiled for point sources at facilities using the same data sources and the same screening procedure as was applied for NO2. The sources included in the multisource modeling analysis, their stack parameters, and their emission rates are listed in Appendices E and F of the Refined Modeling Report - PSD Impact Analysis and Multisource Modeling Protocol for West Deptford Energy Station BTS made several changes to the emission inventories listed in Appendices E and F. Based on information in the Navy Surface Warfare Center Title V permit and discussions with the Philadelphia AMS, the PM-2.5 emissions were changed to the following: Gas Turbine LM2500-2A = 12.6 lb/hr, Gas Turbine LM2500-2B = 12.6 lb/hr, DD(X) Test Cell MT30/LM500/RR-4500 = 5.07 lb/hr, and P-104 Test Cell = 7.72 lb/hr. The Building 87 Testing Diesel Generator was removed from the NOx inventory. In addition, the Exxon-Mobil test Engine Center is in the process of renewing their Title V permit. It is anticipated that the test engines NOx permitted emission rate will be changed from 28.3 lb/hr to 6 lb/hr. Both of these emission rate scenarios were run in the NO2 multisource analysis. It should be noted that the 98th percentile of the annual distribution of daily maximum 1-hour values averaged across 2010 -2012 was used as the 1-hour NO2 background value. Use of this value represents a conservative “first tier” background concentration. The following is stated in EPA’s March 1, 2012 memo entitled “Additional Clarification Regarding Application of Appendix W Modeling Guidance for the 1-Hour NO2 National Ambient Air Quality Standard”

11

…use of a “first tier” assumption for a uniform monitored background contribution may represent a level of conservatism that would obviate the need to include any background sources in the modeled inventory if, for example, the number of nearby sources which could contribute to the cumulative impact is relatively few and the available ambient monitor would be expected to reflect their cumulative impacts reasonable well or conservatively in relation to the modeled design value based on project emissions. The locations of the following sources are all within approximately 4 km of the background NO2 monitor in Chester PA: Covanta Delaware Valley, Monroe Energy, and the Chester Exelon Generating Station. In addition, when the wind blows towards the West Deptford site, these facilities are upwind of the monitor. As a result, their emissions are double counted (i.e., the impact of their NOx emissions are being captured by both the monitor and the modeling). These sources were removed from the NO2 multisource modeling analysis to avoid this double counting these emissions. As BTS evaluates the modeled violations of the 1-hour NO2 NAAQS in the future, additional sources may be removed from the multisource modeling. A total of 153 stacks were in the NOx multisource inventory, 163 stacks were in the PM-2.5 multisource inventory. The NO2 multisource inventory included the emission points at 70 facilities; 30 facilities in New Jersey, 14 facilities in Pennsylvania, 25 facilities in Philadelphia, and 1 facility in Delaware. The PM-2.5 multisource inventory included the emission points at 49 facilities; 11 facilities in New Jersey, 15 facilities in Pennsylvania, 22 facilities in Philadelphia, and 1 facility in Delaware. These facilities are listed in Table 10.

Table 10. List of Facilities Included in Multisource Refined Modeling Study

Major Emitting Facility PM2.5 NO2 NEW JERSEY X X NUSTAR Asphalt Refining X X Eagle Point Power Generation LLC X X Paulsboro Refining Company LLC X X Logan Generating Plant X X Pedricktown Cogen Plant X X Johns Manville X Carney’s Point Generating Plant X X ARDAGH Glass X X Vineland Municipal Electric Utility - VMEU X X Vineland Municipal Electric Utility – Down Station X X Gerresheimer Glass Millville X X Gerresheimer Glass Vineland X Solvay Solexis Specialty Polymers USA LLC X ExxonMobil Research & Engineering Co X Johnson Matthey Inc X Colonial Pipeline Co X Mickleton Energy Center X Eagle Point Dock X

12

Wheelabrator Gloucester Company LP X Camden County Energy Recover Assoc LP X Camden Plant Holdings X Calpine Deepwater X Ferro Corp X Mannington Mills X AMCOR ASP Glass America X Durand Glass Mgt. X Cumberland Energy Center X Sherman Ave Energy Center X 87th Air Base Wing Maguire AFB X Aluminum Shapes LLC X Ardagh Glass Packaging X PHILADELPHIA Grays Ferry Cogeneration Partnership Schuylkill Station X X Exelon Generation - Delaware Generating Station X X Exelon Generation - Schuylkill Generating Station X X US Naval Foundry and Propeller Station X X Naval Surface Warfare Center Carderock Division X X PAID Steam Boiler Plant Philadelphia Naval Business Center X X Aker Philadelphia Shipyard, Inc. Philadelphia Naval Business Center X X The Children's Hospital of Philadelphia X Exelon Generation Company Southwark Generating Station X X Plains Products Terminals LLC X X Southwest Water Pollution Control Plant/Biosolids Recycling Center X X Philadelphia Energy Solutions (Former Sunoco Refineries in PHL) X X Inolex Chemical Company X X Veolia Energy Edison / Phila X X SUNOCO Chemical, Frankford Plant X X Veolia Energy / Schuylkill Station X X Exelon Generating Company - Richmond X X Phila. Gas Works Richmond LNG Plant X X Temple University Health Sciences Campus X X Temple University Main Campus X X Bluegrass Container CO. X X Newman & Co. X X U. S. Mint X Kinder Morgan Liquids Terminals X Northeast Water Pollution Control Plant X X PENNSYLVANIA X Kimberly Clark PA LLC/Chester OPR X X Covanta Delaware Valley LP/Delaware Valley Res Rec X Monroe Energy LLC/Trainer X

13

SUNOCO Partners Mkt. & Terminal X X Boeing Defense Space X X Exelon Generation Eddystone Plant X X Liberty Electric Power X X PQ Corp Chester X X Swarthmore College X X Exelon Generating Chester Station X Delaware Col Regional Water Treatment Plant X X Congoleum Corp. X X SUNOCO Marcus Hook Refinery X X FPL Energy Marcus Hook X X FPL Energy MH50 LP Marcus Hook X X Riddle Memorial Hospital X Villanova University X DELAWARE SUNOCO Marcus Hook Refinery X

Use of PVMRM

When modeling 1-hour NO2 impacts in the multisource modeling analysis, EPA’s Tier 3 Plume Volume Molar Ratio Method (PVMRM) option was used. The 2008 through 2012 observed hourly O3 concentrations from NJDEP’s Clarksboro air quality monitor (approximately 4.7 km south of the West Deptford site) were input. The EPA recommended default NO2/NOx stack ratio value of 0.5, and the NO2/NOx ambient equilibrium ratio of 0.9 were used. Though still considered a screening technique, PVMRM is designed to provide more accurate NO to NO2 conversion rates than the Tier 2 method.

MULTISOURCE MODELING RESULTS PM-2.5 The results of the PM-2.5 multisource modeling analysis are summarized in Table 11. Table 11 demonstrates that the proposed West Deptford Energy Station will not cause or significantly contribute to a violation of the NAAQS. The PM-2.5 multisource maximum 24-hour 8th high concentration is predicted to occur approximately 1.45 km northeast of the West Deptford Energy Station. By far, the major contributors to the maximum 24-hour 8th high concentration are the Navy Surface Warfare Center Carderock Division and the Philadelphia Energy Solutions (formerly the Sunoco Refinery). Their combined impact is approximately 11 ug/m3. The results are identical when a Phase 2 Siemens or GE turbine is assumed at the West Deptford Energy Station.

14

TABLE 11 Comparison of Predicted Multisource PM-2.5 Impacts to NAAQS

Pollutant

Avg. Time

Multisource Maximum

Conc. (ug/m3)

Background Air Quality/Monitor

(ug/m3)

Total Impact (ug/m3)

NAAQS (ug/m3)

Max. WDES Impact at Predicted NAAQS Violation (ug/m3)

PM-2.5

24-hr

13.6 (a)

20.8/15.2/22.3/16.6 (b)

(Gibbstown, NJ) 32.3 35

none

a. Estimated based on the average of the seasonal background values. b. PM2.5 24-hour 3-year average 98th percentile value for winter, spring, summer, and fall, respectively Nitrogen Dioxide The results of the NO2 multisource modeling analysis are summarized in Table 12 and 13. Table 12 demonstrates that the proposed West Deptford Energy Station with a Phase 2 Siemens turbine will not cause or significantly contribute to a violation of the NAAQS. When the Exxon-Mobile facility is modeled with a 28.3 lb/hr emission rate, the maximum 1-hour NO2 8th high occurs next to the Exxon-Mobil facility. When the Exxon-Mobile facility is modeled with a 6.0 lb/hr emission rate, the maximum 1-hour NO2 8th high occurs approximately 5.5 km north-northwest of the West Deptford Energy Facility. This location, just north of the Philadelphia International Airport, is in Pennsylvania. The maximum 1-hour NO2 8th highest concentration in New Jersey is predicted to be 250 ug/m3. Table 13 demonstrates that the proposed West Deptford Energy Station with a Phase 2 GE turbine will not cause or significantly contribute to a violation of the NAAQS. The locations of the maximum 1-hour NO2 8th high with the GE turbines occur at the same locations as with the Siemens turbine. The major contributors to the high NO2 concentrations predicted in the West Deptford NO2 significant impact area were the following: Exxon-Mobil (only during the 28.3 lb/hr scenario), NUSTAR (now called Axeon), Philadelphia Naval Surface Warfare Center, Philadelphia Energy Solutions (formerly Sunoco Refinery), Exelon Eddystone Plant, and the Exelon/Veolia/Grays Ferry Cogeneration Schuylkill Station. The first two of these sources are in New Jersey.

15

TABLE 12. Comparison of Predicted Multisource 1-Hour NO2 Impacts to NAAQS with a Phase 2 Siemens Turbine (Hot Start-Up)

Exxon-Mobile Emission Rate

(lb/hr)

Multisource Maximum

Conc.a

(ug/m3)


(ug/m3)


NAAQS (ug/m3)

Max. WDES Impact at Predicted

NAAQS Violation (ug/m3)

28.3 316.3

89

(Chester, PA)

405.3

188

8.6

6.0 202.9

89

(Chester, PA)

291.9

188

9.4

a. EPA’s Tier 3 PVMRM used for NO to NO2 conversion.

TABLE 13. Comparison of Predicted Multisource 1-Hour NO2 Impacts to NAAQS with a Phase 2 GE Turbine (Hot Start-Up)

Exxon-Mobile Emission Rate

(lb/hr)

Multisource Maximum

Conc.a

(ug/m3)


(ug/m3)


NAAQS (ug/m3)

Max. WDES Impact at Predicted

NAAQS Violation (ug/m3)

28.3 316.3

89

(Chester, PA)

405.3

188

8.6

6.0 202.9

89

(Chester, PA)

291.9

188

9.2

a. EPA’s Tier 3 PVMRM used for NO to NO2 conversion. BRIGANTINE CLASS I AREA ANALYSIS The only PSD Class I area within 300 kilometers of the facility is the Brigantine National Wildlife Refuge, which is located approximately 75 kilometers to the southeast of West Deptford Energy Station. The Federal Land Manager (FLM) for Brigantine was contacted by the facility and the FLM determined that an assessment of the facility impacts to air quality related values (AQRVs), i.e., sulfate and nitrate deposition was not required. Emissions from the proposed project were evaluated to assess the impact on the Class I PSD Increments at the Brigantine National Wildlife Refuge Class I area. The Class I impacts were modeled with AERMOD and five years (2008-2012) of hourly surface meteorological data from Philadelphia International Airport with concurrent upper air data from Sterling, VA. AERMOD was used as a Class I screening tool. Class I area screening receptors were developed by placing an arc of receptors at 50 kilometers from the facility site in line with the Brigantine Class I area. The actual Class I receptors and heights for the Brigantine Wilderness Area were used. The Class I modeling assumed rural land use because the plume from West Deptford Energy Station would travel over land classified as rural. Maximum modeled concentrations were then compared to both the PSD

16

Class I SILs and increments as presented in Table 14. Emission of SO2, PM-10, PM-2.5 and NO2 were all predicted to be below their SILs and PSD Class I Increments.

TABLE 14. Comparison of Maximum Predicted Impacts for Criteria Pollutants to Class I Significant Impact Levels and Increments

Pollutant

Averaging Time

Maximum Predicted Class I Impact

(ug/m3)

PSD Class I Significant Impact Level

(ug/m3)

SO2 3-hour 0.123 1.0 24-hour 0.033 0.2 annual 0.0013 0.1

PM2.5 24-hour 0.063 0.07 annual 0.005 0.06

PM10 24-hour 0.063 0.3 annual 0.005 0.2

NO2 annual 0.006 0.1 RISK ASSESSMENT A risk assessment was conducted to assess the possible adverse health effects due to inhalation exposure to the air toxics emitted from the natural gas fired combustion turbines at the West Deptford Energy Station. A total of 6 air toxics were included in this study. The highest 24-hour and annual concentrations predicted by the single source AERMOD modeling analysis with the 2008-2012 meteorological data were used. Modeling results indicate that the long-term cancer risk for the carcinogenic air toxics (formaldehyde and lead) is less than one in a million. In addition, hazard quotients for all pollutants with health effects other than cancer are predicted to be less than 1.0. Table 15 lists the emission rates and the risk assessment performed for the six air toxics. In summary, the health effects due to air toxics emissions from the West Deptford Energy Station are predicted to be negligible.

Table 15. Air Toxic Pollutants Health Risks

Pollutant

Total Turbine Emissions (ton/yr)

70-Year Exposure

Cancer Risk

Long-Term Hazard

Quotient

Short-Term Hazard

Quotient

Acrolein 0.198 -- 0.0012 0.003 Ammonia 200.9 -- 0.0011 0.003

Formaldehyde 7.5 0.03 in a million 0.0021 0.006 Lead 0.00124 0.000003 in a million 0.00006 0.0003

Sulfuric Acid 9.1 -- 0.0014 0.01 Toluene 3.9 -- 0.000002 0.00001

17

Attachment 1

NESCAUM Recommendations on the Use of an Interim Significant Impact Level (SIL) in Modeling the 1-Hour NO2 NAAQS

Background and Importance of SILs On February 9, 2010, the U.S. Environmental Protection Agency (EPA) published a new 1-hour nitrogen dioxide (NO2) National Ambient Air Quality Standard (NAAQS) at a level of 100 ppb (approximately 188 µg/m3). This new standard became effective on April 12, 2010, which means that permits issued under EPA’s prevention of significant deterioration rules (40 CFR 52.21) on or after April 12, 2010, must contain a demonstration that allowable emissions from any new major stationary source or major modification will not cause or contribute to a violation of the new 1-hour NO2 NAAQS (see EPA’s Fact Sheet). EPA has not yet proposed a significant impact level (SIL) for the 1-hour NO2 NAAQS, yet states are expected to begin implementing the standard immediately. It is EPA’s policy to exempt sources from conducting comprehensive, multisource modeling if their estimated maximum ambient impacts for a given pollutant are less than the SIL. Therefore, it is important for NESCAUM states to allow permit applicants to use an interim 1-hour NO2 SIL in the permitting process. EPA-defined SILs currently exist for PM10, CO, SO2 and the annual NO2 NAAQS. The NESCAUM Permit Modeling Committee previously developed and recommended interim SILs for the PM2.5 NAAQS (see NESCAUM Technical Guidance on Significant Impact Levels (SILs) for PM2.5, dated December 8, 2006; http://www.nescaum.org/topics/permit-modeling). In practice, if the modeled ambient impacts from a proposed project are less than the respective SIL, the project:

• is presumed to not cause or significantly contribute to a PSD increment or NAAQS violation, and • is not required to perform multiple source cumulative impact assessments.

Without a 1-hour NO2 SIL, permit applicants would be obligated to perform a cumulative modeling analysis in essentially all instances – an analysis which may unnecessarily consume regulatory agency resources, especially given the large number of NO2 major sources that are being proposed across the region. The use of an interim 1-hour NO2 SIL would also make the NSR process more efficient without a detrimental effect on air quality. Therefore, its use is advantageous to both permit applicants and NESCAUM state agencies.

18

http://edocket.access.gpo.gov/2010/pdf/2010-1990.pdf

http://www.epa.gov/air/nitrogenoxides/pdfs/20100122fs.pdf

http://www.nescaum.org/topics/permit-modeling

The recommendations for the use of 1-hour NO2 SILs, below, were developed by the NESCAUM Permit Modeling Committee to assist permit applicants and states in preparing and reviewing air quality modeling analyses. The technical basis for these recommendations is provided in the Appendix.

19

Summary of NESCAUM Permit Modeling Committee Recommendations on Use of an Interim 1-hour NO2 SIL To facilitate air quality modeling reviews of permit applications and other modeling assessments, the NESCAUM Permit Modeling Committee recommends that the following 1-hour NO2 SIL can be used by state air agencies until such time that EPA formally adopts a 1-hour NO2 SIL: 1-hour NO2 SIL = 10 µg/m3, with a form based on:

• the highest five year average of modeled 1-hour maximum NO2 concentrations predicted each year at a given receptor, if using five years of National Weather Service meteorological data; or

• the highest modeled 1-hour NO2 concentration for one year of site-specific

meteorological data. Conversion of nitr ic oxide (NO) to nitrogen dioxide (NO2) can be approximated with a three tiered screening system similar to the tiered procedures specified in Section 5.2.4 of EPA’s Guideline on Air Quality Models:

• tier-1 assumes 100 percent conversion of NO to NO2, • tier-2 assumes a NO2 to NOx (NO + NO2) ratio of 75 percent, and • tier-3 allows case-by-case use of a site specific ratio derived using techniques such as the

Plume Volume Molar Ratio (PVMRM) or the Ozone Limiting Method (OLM). The interim 1-hour NO2 SIL is recommended for use by permit applicants and the states in the NESCAUM region to determine if a proposed source or modification is required to perform a multiple source cumulative impact assessment. The extent and complexity of any cumulative analysis conducted when the interim SIL is exceeded will be determined by individual states. The interim SIL can also be used at the discretion of individual NESCAUM states to determine if a source is causing or significantly contributing to a violation of the 1-hour NO2 NAAQS.

20

http://www.epa.gov/scram001/guidance/guide/appw_05.pdf

http://www.epa.gov/scram001/guidance/guide/appw_05.pdf

April 21, 2010

Appendix

NESCAUM Recommendations on the Use of an Interim Significant Impact

Level (SIL) in Modeling the 1-Hour NO2 NAAQS

Basis for the Recommended Interim 1-Hour NO2 SIL The NESCAUM Permit Modeling Committee considered three issues when developing a recommendation for the 1-hour NO2 SIL: (1) the value of the SIL, (2) the form of the SIL, and (3) the use of a default, or a tiered system of NO to NO2 conversion rates.

1) Value of Interim 1-Hour NO2 SIL Option 1 – Use the existing annual NO2 SIL (1 µg/m3). Option 2 – Use a value based on the ratio of the annual NO2 SIL to the annual NO2 NAAQS ((1 µg/m3 / 100 µg/m3) x 188 µg/m3 = 1.9 µg/m3). Option 3 – Develop a value based on the only other criteria pollutant with a 1-hour NAAQS, carbon monoxide (CO), using the ratio of the 1-hour CO SIL to the 1-hour CO NAAQS ((2000 µg/m3 / 40,000 µg/m3) x 188 µg/m3 = 10 µg/m3 (rounded up from 9.4 µg/m3)). Option 4 – Use a value based on the EPA’s draft July 23, 1996, NSR Reform proposal recommending 4% of the Class I increment as the Class I SIL, where the 4% value was based on EPA’s definition of de minimis emission rates for NAAQS impact demonstration purposes (see 45 FR 52676, August 8, 1980). (188 µg/m3 x 0.04 = 7.5 µg/m3).

The spatial and temporal variations of short-term 1-hour impacts tend to be much more volatile than longer averaging times such as an annual average. In addition, the new 1-hour NO2 NAAQS will be applied to hot spot type modeling near major roadways, not unlike the 1-hr CO NAAQS. A very low SIL will result in frequent multisource cumulative modeling for NO2, a resource intensive activity that in many cases will have limited usefulness.

Recommendation: Option 3, Interim 1-Hour NO2 SIL = 10 µg/m3

2) Form of 1-Hour NO2 Interim SIL Option 1 – Use the highest modeled 1-hour NO2 concentration using five years of NWS meteorological data or using one year of site-specific meteorological data. This form is similar to many of EPA’s current SILs. Option 2 – Use the highest modeled 1-hour NO2 concentration predicted each year at a receptor, then average over five years if using NWS meteorological data; or use the highest modeled 1-hour NO2 concentration for one year of site-specific data. This option is similar in form to the 24-hour PM2.5 NAAQS, another permit modeled criteria pollutant with a probabilistic, not deterministic, NAAQS (see Stephen Page memo Modeling Procedures for Demonstrating

21

Compliance with the PM2.5 NAAQS, dated March 23, 2010; http://www.epa.gov/scram001/Official%20Signed%20Modeling%20Proc%20for%20Demo%20Compli%20w%20PM2.5.pdf ). Option 3 –Use the highest of the modeled 8th highest daily maximum 1-hour average concentrations in a year predicted over five years. This option reflects the approximate form of the 1-hour NO2 NAAQS (i.e., highest of the 98th percentile of the annual distribution of daily maximum 1-hour average concentrations predicted over five years). Given the similarities in the form of the 1-hour NO2 NAAQS with that of the 24-hour PM2.5 NAAQS, it would seem likely that when a 1-hour NO2 SIL is promulgated by EPA it will reflect the form of option 2. Recommendation: Option 2: The form of the interim 1-hour NO2 SIL would be the highest five year average of the modeled maximum 1-hour NO2 concentrations each year at a receptor, if using five years of NWS meteorological data. For one year of site-specific meteorological data, it would be simply the highest modeled 1-hour NO2 concentration.

3) NOx to NO2 conversion rate Option 1 – Assume 100 percent conversion of exhaust gas NO to NO2. Option 2 – A two tiered screening approach, where tier-1 assumes 100 percent NO to NO2 conversion, and tier-2 assumes a NO2 to NOx (NO + NO2) ratio of 75 percent (this is EPA’s ambient ratio method, ARM, annual national default conversion). Option 3 – A three tiered system, where tier-1 assumes 100 percent NO to NO2 conversion, tier-2 assumes a 75 percent NO2/NOx ratio, and tier-3 allows a case-by-case use of a site specific ratio derived using more refined techniques such as the Ozone Limiting Method (OLM) or the Plume Volume Molar Ratio Method (PVMRM). Both are in AERMOD. Option 3 is similar to the tiered procedures specified for modeling annual NO2 impacts in Section 5.2.4 of the Guideline on Air Quality Models. There is also an EPA Clearinghouse memo endorsing application of the 75 percent NO2/NOx ratio when modeling the annual NO2 SIL (see Daniel J. deRoeck memo: Use of the Ambient Ratio Method for Modeling Significant Ambient Impacts of NO2, dated March 15, 2002;. http://www.epa.gov/region07/air/nsr/nsrmemos/m200203.pdf) Option 3 would give the States the most flexibility. It should be noted that PVMRM is a non-Guideline technique, so if used in PSD permit modeling, EPA Regional Office approval should be obtained. Recommendation: Option 3: A three tiered system would provide maximum flexibility.

22

http://www.epa.gov/scram001/Official%20Signed%20Modeling%20Proc%20for%20Demo%20Compli%20w%20PM2.5.pdf

http://www.epa.gov/scram001/Official%20Signed%20Modeling%20Proc%20for%20Demo%20Compli%20w%20PM2.5.pdf

http://www.epa.gov/region07/air/nsr/nsrmemos/m200203.pdf

FACILITY NAME (FACILITY ID NUMBER) BOP050001

Emission Unit: U1 25 MM BTU/hour Boiler burning Fuel Oil and Natural Gas Operating Scenario: OS Summary OR OSXX Boiler burning Fuel Oil

New Jersey Department of Environmental Protection Facility Specific Requirements

Explanation Sheet for Facility Specific Requirements

Ref.# Applicable Requirement Monitoring Requirement Recordkeeping Requirement Submittal/Action Requirement

1 Conduct a comprehensive stack test at emission point PTXX at least 18 months prior to the expiration of the approved operating permit to demonstrate compliance with the CO, NOx, TSP and VOC emission limits.[N.J.A.C. 7:27-22.16(e)]

Other: Stack emission testing. Stack test shall be conducted for CO, NOx, TSP, and VOC emissions (add language as needed). Based on any 60- minute period. [N.J.A.C. 7:27- 22.16(e)]

Other: Stack test results . [N.J.A.C. 7:27-22.16(e)]

Stack Test - Submit a protocol, conduct stack tests, submit result s: As per the approved schedule. Submit a stack test protocol to the Bureau of Technical Services (BTS) at PO Box 437, Trenton, NJ 08625 at least 30 months prior to the expiration of the approved operating permit. [N.J.A.C. 7:27-22.18(e)] and [N.J.A.C. 7:27- 22.18(h)]

Item Number

Description of applicable requirement

Rule citation (subchapter, section, and paragraph) for the

applicable requirement

Rule citation for the monitoring

requirement

Rule citation for the recordkeeping requirement

Rule citations for the submittal/action

requirement

Air contaminants

Monitoring method to ensure compliance

Records to be kept

Actions to be taken by the

facility

Submittal requirement

Emission unit number (assigned by the

facility) Brief description of emission unit

OS Summary lists all rules and requirements that apply to an emission unit,

regardless of operating scenarios. Emission unit may contain one or more

pieces of equipment and the corresponding operating scenarios

OSXX denotes the operating scenario number and lists the rules and requirements that apply to a particular scenario. An operating scenario represents various ways (or scenarios) a piece of equipment can operate.

Activity Number (assigned by the

Department)

12/14/10 23

air operating permit for celotex corporation

Documents