MEMORANDUM - dnrec.state.de.us II... · of the crude unit atmospheric heater and ... The cracking takes place when fresh feed ... Agglofilter modules remove particulate matter and

MEMORANDUM The Premcor Refining Group, Inc. SUBJECT: Application for Permits to Construct a Refinery Pollution Control Upgrade Project

(PCUP) at the Premcor Refining Group, Inc.’s Delaware City Refinery – Phase II DATE: February 7, 2005 Page 2

An integral part of these settlement agreements required Motiva and its successors to install pollution control equipment in several major processes to control criteria pollutant and hazardous air pollutant (HAP) emissions. Motiva Enterprises, LLC submitted applications on March 29, 2004 for the construction of the following pollution control equipment and improvement projects collectively termed as the Delaware City Refinery Pollution Control Upgrade Project (PCUP):

• A Belco Technologies prescrubber, a Cansolv amine based regenerative wet gas scrubber and a polishing scrubber to treat sulfur oxide emissions in the exhaust gas from the fluid coking unit’s carbon monoxide boiler.

• A GE Energy and Environmental Research Corporation selective non catalytic reduction system to reduce nitrogen oxide emissions in the exhaust gas from the fluid coking unit’s carbon monoxide boiler.

• A Belco Technologies prescrubber, a Cansolv amine based regenerative wet gas scrubber and a polishing scrubber to treat sulfur oxide emissions in the exhaust gas from the fluid catalytic cracking unit’s carbon monoxide boiler.

• Installation of piping to route the slurry recycle from the fluid catalytic cracking unit to the fluid coking unit’s reactor scrubber.

• Two package boilers (each with a design rated heat input of 216 mmBtu/hour) to supply steam for the regeneration of the amine–based sorbent used in the regenerative wet gas scrubbers.

• Modification of the sulfur recovery area to reroute the ammonia from the pre-combustor of the crude unit atmospheric heater and to allow the processing of the recovered sulfur dioxide from the wet gas scrubbers.

• Concurrent other changes to allow incremental improvements in refinery operations. The basis for the PCUP related constructions and modifications are found in the Second Addendum to Civil Action H-01-0978 between the United States of America (Plaintiff) and the States of Delaware and Louisiana (Plaintiff-Interveners) versus Motiva Enterprises LLC (Defendant), entered on December 24, 2003 in the United States District Court for the Southern District of Texas and in Civil Action No. 18750 NC between Nicholas A. DiPasquale, Secretary of the Department of Natural Resources and Environmental Control, an Agency of the State of Delaware (Plaintiff) versus Motiva Enterprises LLC (Defendant), entered on March 22, 2001. In accordance with paragraph 168a of the Second Addendum, Motiva was required to submit complete applications by no later than March 31, 2004 for all state and federally enforceable permits necessary for the installation of regenerative WGS systems on the DE City FCU and FCCU and other contemporaneous or related projects. The Department had to issue its decision on all state and federally enforceable permits related to the FCU as contained in those permit applications received on or before March 31, 2004, within 8 months of receipt of Motiva’s



application. AQM completed this portion of the PCUP (i.e., Phase I of the PCUP) and issued permits for the following units on November 30, 2004: • Fluid coking unit (FCU) • FCU wet gas scrubber (WGS) • FCU SNCR • Two new package boilers • SRA modifications • Crude unit modifications • Crude unit atmospheric heater modifications The Department has to issue its decision on the remaining state and federally enforceable permits as contained in those permit applications, received on or before March 31, 2004 within 14 months of receipt of Motiva’s application. This memorandum addresses those remaining components as Phase II of the PCUP requiring state and federally enforceable permits by no later than May 31, 2005 (i.e., the 14 month deadline) for the following unit operations: • Fluid catalytic cracking unit (FCCU), the FCCU Carbon monoxide boiler (COB) and

WGS • Coke handling and storage area modifications This memorandum consists of the following 3 parts:

• Part 1: Project description for the FCCU, FCCU COB, WGS and Coke handling and storage area modifications

• Part 2: Regulatory and technical analysis of each affected unit/process whose permits have to meet the May 31, 2005 deadline; and

• Part 3: Draft permits for the affected units, i.e. FCCU, FCCU COB, WGS and Coke handling and storage area modifications

Glossary of Terms: AQM: DNREC’s Air Quality Management BACT: Best Available Control Technology BPD: Barrels per day COB: Carbon monoxide (CO) boiler DCR: Delaware City Refinery DAAQS: Delaware Ambient Air Quality Standards ESP: Electrostatic Precipitator ESPC: Equivalent Sulfur Plant capacity



FCCU: Fluid Catalytic Cracking Unit FCU: Fluid Coking Unit ISCST3: Industrial Source Complex Short Term Model LAER: Lowest Achievable Emission Rate LoTOx: Low Temperature Oxidation LTPD: Long tons per day MACT: Maximum Achievable Control Technology NESHAP: National Emission Standards for Hazardous Air Pollutants NSPS: New Source Performance Standards NSR: New Source Review PCUP: Pollution Control Upgrade Project RACT: Reasonably Available Control Technology RFG: Refinery fuel gas SRA: Sulfur Recovery Area TPD: Tons per day TPY: Tons per year, as determined on a twelve month rolling total WGS: Wet Gas Scrubber



Part 1: Project Description



This part of the memorandum describes the project as it relates to existing operations and the proposed changes subject to review under Phase II review. The PCUP Phase II will impact the following unit operations:

• Fluid catalytic cracking unit (FCCU)1 • FCCU COB3 • Coke handling and storage operations

Fluid Catalytic Cracking Unit (Unit 23): The fluid catalytic cracking unit (FCCU) converts refinery intermediates and purchased feedstock into high octane gasoline and other intermediate products by a catalytic cracking reaction. Overhead gaseous products are fed to the refinery gas plant (Unit 24) for further treatment. The intermediate fractions that are not sent to the gas plant are sent to other refinery units for further processing and handling. The FCCU is comprised of the following units/processes: ! FCCU riser reactor, stripper, regenerator and start up heaters (23-H-1 A and B) each with

a rated design heat input of 82 mmBtu/hour. ! FCCU carbon monoxide boiler (COB, Unit 23-H-3) with a rated design heat input of

679.6 mmBtu/hour Process Description: An integral part of the refining operation involves a chemical decomposition process whereby gas oil (heavy distillate) carbon chains are broken down in the presence of heat and a catalyst into lighter fractions, thus increasing the gasoline yield from the crude stock. This decomposition is carried out in the FCCU. The cracking takes place when fresh feed comes in contact with hot catalyst in a riser reactor. The catalyst and cracked products flow to a reactor where they are disengaged, the products flowing to a fractionator for further processing. The spent catalyst rises through two internal risers to a regenerator where the coke that is deposited on the catalyst is burnt under controlled combustion conditions. The regenerated catalyst is withdrawn and fed to the riser to repeat the process. Regenerator combustion products contain regulated criteria pollutants and therefore must be controlled before being released to the atmosphere. Pollutants in this off-gas include particulates (from catalyst fines), carbon monoxide , sulfur dioxide, nitrogen

1 The permit for the FCCU and the FCCU COB will be processed separately to meet the 14 month deadline in accordance with the provisions of paragraph 168a of the Second Addendum to the CD.



oxides, VOCs that cause odors and hazardous air pollutants. Control measures include internal cyclones for particulate removal and a carbon monoxide boiler to combust the carbon monoxide to carbon dioxide. FCCU feedstocks are complex mixtures of hydrocarbons of various types and sizes ranging from small molecules with four carbon atoms to very large complex molecules of perhaps 60 carbon atoms. The crackability of a particular feedstock is a function of the relative proportions of paraffinic, naphthenic and aromatic species in the feed. The generally accepted theory behind catalytic cracking reactions is based on the formation of carbenium ions. Once formed, the carbenium ion is most likely to crack by $ scission. In other words, the C-C bond in the $ position relative to the positively charged C tends to be cleaved. The olefin may now form a new corbonium ion with the catalyst, thus propagating the reaction. Planned Changes to the FCCU: The planned changes that affect the FCCU include the following:

• Installation of an amine-based regenerative wet gas scrubber (WGS) system and upgrades to the COB to accommodate the WGS installation. This is the primary change planned at the FCCU for compliance with the injunctive relief measure prescribed in the CD. The WGS system will include a water pre-scrubber, an amine-based regenerative scrubber and a caustic polisher along with ancillary equipment needed to support the scrubber (i.e., the regeneration system, air coolers, gas fired package boilers etc). The WGS will replace the existing COB exhaust stack with an integrated scrubber stack system that is designed to remove particulate matter, sulfur dioxide and sulfur trioxide from the FCCU flue gas.

There are 2 main elements in the WGS – the scrubber and regeneration systems. The scrubber element consists of a Belco prescrubber followed by a Cansolv absorber section. The purpose of the Belco prescrubber is to saturate the flue gas with water and to remove particulate matter and sulfur trioxide before the gas enters the absorber section. The main SO2 absorption section of the Belco/Cansolv WGS uses an amine-based scrubber solution in a packed bed absorber tower to remove SO2 from the exhaust stream. The main absorption loop is followed by a polishing scrubber, which is a final packed stage that is separate from the amine-based absorption step and will be used to ensure that the CD driven levels of control are achieved (i.e., 25 ppmvd @ 0 % O2 on a 365 day rolling average basis and 50 ppmvd @ 0 % O2 on a 7 day rolling average basis). The Belco prescrubber section consists of a quench section followed by Agglofilter modules and Cyclolab droplet separators. A low pH is maintained in the prescrubber



section to maximize the SO2 absorption in the absorber section. The quench and Agglofilter modules remove particulate matter and SO3 while the Cyclolabs remove any large entrained droplets carried over from the prescrubber. Blowdown from the prescrubber flows to a purge treatment unit where it is neutralized with caustic and clarified prior to being routed to the refinery’s effluent treatment plant. A packed tower serves as the absorber where the gas is contacted with an amine which absorbs the SO2. Cleaned gas will exit the absorber through a stack mounted on the absorber tower. Rich amine is filtered and heated through an effluent-influent heat exchanger before being fed to the regenerator tower. In the regenerator, the rich amine is steam stripped yielding a high purity SO2 stream that will be routed to the refinery’s SRA. The regenerated lean amine is pumped back to the absorber. Because heat stable salts (HSS) are formed over time, a small slip stream of lean amine is routed to an electro-dialysis unit to extract the HSS from the lean amine. Process Chemistry2: In an aqueous solution, SO2 undergoes reversible hydration and ionization according to the following equations: SO2 + H2O ↔ H+ + HSO3

- ………………..Eq. 1 HSO3

- ↔ H+ + SO3= ………………………Eq. 2

Equations 1 and 2 are half completed at pH values of 1.81 and 6.91 respectively at 18 deg. C. To increase the quantity of SO2 dissolved in the water an amine buffer is added driving the above equilibria to the right by reacting with the hydrogen ions to from ammonium salts: R1R2R3N + H+ ↔ R1R2R3NH+ …………..Eq. 3

2 AQM’s research has found the most concise description of the process chemistry in a technical paper titled “Refinery and Natural Gas Applications of Cansolv System Technology” authored by Paul J. Parisi and John C. Bourdon, presented at “Sulphur 97”, November 18, 1997 in Vienna. This excerpt is a section from the above referenced paper.



In order for the process to be regenerable, the buffering agent should operate in a pH region sufficiently low so as to present a desirable value of SO2 vapor pressure over the solution at the regeneration temperature. Steam stripping the vapor phase SO2 will reverse Equations 1 ~ 3 thus regenerating the amine absorbent. The Cansolv system process is based on a class of diamine absorbents that optimally balance the ability to absorb and regenerate SO2. One of the amine functionalities of the absorbent is strongly basic that it is not thermally regenerable under the Cansolv system process conditions. So, once reacted into a salt by SO2 or any stronger acid, this strongly basic amine functionality will remain permanently as a salt. R1R2N-R3N-NR4R5 + HX ↔ R1R2NH+-R3-NR4R5 + X - …………..Eq. 4 The monoprotonated amine on the right hand side of Eq. 4 is the in-process lean amine which is used to scrub the SO2. This is a HSS. The second amine functionality (the “sorbing nitrogen”) is less basic and it buffers in the desired range for regenerability of SO2, which in practice is about pH 4 for the rich amine and pH 6 for the lean. This buffering range provides the proper balance of absorption and regenerability as shown in Eq 5 below: R1R2NH+-R3-NR4R5 + SO2 + H2O ↔ R1R2NH+-R3-NH+R4R5 + HSO3 - … .Eq. 5 In Eq 5 the anion X – is not shown as it does not participate in the SO2 reaction with the sorbing nitrogen. The nature of X – can affect the functioning of the process: if it is SO3

2- it can contribute to SO2 scrubbing. However, if X – is the anion of a strong acid and is allowed to accumulate more than 1 equivalent per mole of amine, it will neutralize the sorbing nitrogen and thereby decrease the scrubbing capacity of the solvent. Thus the level of the HSS is kept below 1 equivalent per mole of solvent by purging a slip stream to an electro-dialysis purification unit which replaces the non-regenerable sulfate anions by sulfite or bisulfite ions.

• Installation of piping to route the slurry recycle from the FCCU riser to the FCU reactor scrubber. Physically, this modification involves installing a jumper downstream of the reslurry control valve to downstream of the decant oil cooler. From an operational standpoint, this change is expected to reduce the FCCU stack opacity and reduce FCCU HAP emissions.

• Piping and configuration changes that will increase the FCCU feed preheat.



• Upgrade of the COB burner management system. Coke Handling and Storage System (Unit 22): The purpose of the coke handling and storage system is to transfer fluidized petroleum coke from the Fluid Coker Unit (the FCU or coker) where it is produced. Coke exits the coker and is transferred via a pneumatic chute into the coke storage silo. From here the coke is either loaded into railcars or onto a conveyor belt system. Once on the conveyor system, coke is either transported to the power plant where it is used as fuel in the gasifiers or transferred to the coke storage area for subsequent sale and transport offsite. The handling and storage of coke has always caused excess particulate fugitive emissions. The refinery had been ordered to implement several control measures including adding oil to increase the moisture content and covering the conveyor belts used to transport the material. These measures have had little success. Since 2001, the refinery has sampled the quantity of total suspended particulates (TSP) in the ambient air. To date, twenty (23) violations of the State’s 24-hour secondary ambient air quality standard have been monitored. The goal of this project is to control fugitive coke emissions from this process and eliminate the air quality violations. Planned Changes to the Handling and Storage System: As a result of the Pollution Control Upgrade Project (PCUP), the amount of fluidized petroleum coke that will be produced by the coker will increase from 1880 tpd to 2500 tpd. In order to control the expected emissions associated with this increase, Premcor has proposed modifying the system and employing new fugitive coke control measures. The planned changes include:

• Currently, the coke is sprayed with oil to control dust as it is loaded onto the first conveyor belt. Premcor has proposed to modify the process by ceasing to add oil. In order to control fugitive emissions from the dry coke the Company will install three (3) baghouses at the three drop points on the conveyor system as it is moved to the storage area.

• At the end of the conveyor system, the coke drops through a chute to the ground at the

outdoor storage area. It is then moved into the coke storage pile with bulldozer. Premcor has proposed modifying this part of the process by building a warehouse where the coke will be stored. The Company will also load the trucks that transport the coke offsite



inside the building. The stack-out and wetting operations will also take place inside the warehouse.

• The proposed coke storage warehouse will be constructed on the site of the present

outdoor storage piles. An earthen berm, which is 25 feet high and surrounds the coke pile on three sided, is used to help contain the storage pile and protect it from wind erosion. In order for the warehouse to be built, the berm will be removed.

• A truck wash will be constructed so that after being loaded, coke and coke dust can be

removed from the outsides of the trucks so that it wouldn’t be tracked throughout the plant and onto outside roadways.

• Scrappers will be installed at the discharge pulley of each conveyor. This is intended to

mitigate coke emissions from the conveyor discharge chute due to “carry-back” of fines on the conveyor belt.



Part 2: Regulatory and Technical Analysis

MEMORANDUMThe PremSUBJECT: Application f

DATE:Page 13

cor Refining Group, Inc.

or Permits to Construct a Refinery Pollution Control Upgrade Project (PCUP) at the Premcor Refining Group, Inc.’s Delaware City Refinery – Phase II

February 7, 2005

This section evaluates the regulatory and technical considerations for the emission units affected by Phase II of the PCUP – the FCCU and the coke handling and storage facility. In each case the applicable regulation or requirement is stated followed by descriptions of the compliance methodology, monitoring/testing, recordkeeping and reporting requirements. Table 1 provides a summary of the net emissions changes expected after completion of both phases of the PCUP. This netting transaction supplements the netting transaction of PCUP Phase I by appending the emissions changes attributable to PCUP Phase II.

cor Refining Group, Inc.

for Permits to Construct a Refinery Pollution Control Upgrade Project (PCUP) at the Premcor Refining Group, Inc.’s Delaware City Refinery

May 19, 2004

Table 1: PCUP Phase 1 Expected Emissions Changes Baseline Emissions Potential to Emit Projected ∆

Pollutant CU FCU FCCU3 SRA4 Total CU FCU FCCU SRA PB Total Rest ofRefinery

5

Net Change6

SO2 23.4 19215 10503 156.5 29898 48.1 173.8 361 672.1 42.3 1297 1 - 28601NOx 73.8 689.8 451 30.7 1245.3 92.3 TBD TBD 51.9 13.2 TBD 1.2 TBDTSP 18.6 276.6 1,131 6.1 1432.3 42 206.3 TBD 22.3 12.1 TBD TBD TBD

H2SO4 0.4 268.8 458 5.9 733.1 0.7 252.2 309 11.6 0.8 574.3 0 -158.8PM10 19 545.4 848.3 6.1 1418.8 42.9 458.5 TBD 10.8 12.9 TBD TBD TBDCO 0 1296.3 1432 7.7 2736 75.1 TBD 3,768 26 133.9 TBD 0.9 TBD

VOC 0.9 103.5 114.5 0.7 219.6 6.4 TBD TBD 1.3 8.8 TBD 1.0 TBDPb 7.3 E-4 6.9 E-2 NA 1.1 E-4 7.0 E-2 9.1 E-4 9.0 E-2 1.4 E-01 1.8 E-4 8 E-4 2.3 E-1 0 1.6 E-1

TRS/RSC 0 5.8 E-3 NA 7 7 0 7.8 E-3 1.1E-02 12.7 0 12.7 0 5.7 H2S 0 5.8 E-3 NA 7 7 0 7.8 E-3 1.1E-02 12.7 0 12.7 0 5.7

3 AQM has not approved Premcor’s use of October 1, 2002 through December 31, 2003 as being a representative baseline period for the FCCU. Ideally, AQM would have preferred to use the 24 month period immediately preceding October 1, 2002 as the representative period. AQM has asked Premcor to submit emissions data for this 24 month period. That data has not been received so far. The Department presently has access to the inventory data for the calendar years 2001 and 2002 which have been used in this case as an approximation of the representative 24 month desired data. 4 For the SRA, Premcor has used the “current allowable” emissions as the baseline emissions. On the other hand, AQM has used the 12 month period of July 2003 through June 2004 to compare “current actual” emissions to the “future potential” emissions. 5 Emissions changes shown under “Rest of Refinery” include the emissions changes associated with unmodified process heaters and unmodified process units. Emissions from these unmodified sources are expected to change because of the projected units’ utilization changes. Emissions changes associated with the unmodified process heaters were based on the Delaware City Refinery Linear Program model’s predicted changes in the average daily heat input to the various combustion units. The estimated change in average heat input was multiplied by the current emissions factor for each unit’s heaters and converted to a TPY basis. Emissions changes from unmodified process operations were calculated as the difference between the emissions from these units when the crude throughput is 185 MBPD (i.e, the post PCUP operating scenario) and the baseline crude throughput of 165MBPD. TSP and PM10 emissions changes are yet to be determined because the application does not include an AQM 4 application form for the modifications to the coke handling and storage facility. 6 The net change in emissions are the estimated changes after completion of both phases of the PCUP.

MEMORANDUMThe PremSUBJECT: Application

DATE:Page 14



Fluid Catalytic Cracking Unit (Unit 23): Regulation No. 2 - Permits: Section 2.1 (c) of this Regulation states: Except as exempted in Section 2.2, no person shall initiate construction, install, alter or initiate operation of any equipment or facility or air contaminant control device which will emit or prevent the emission of an air contaminant prior to receiving approval of his application from the Department or, if eligible, prior to submitting to the Department a completed registration form for equipment, a facility or an air contaminant control device that is not subject to Section 2.1(a) or 2.1(b), the person shall submit to the Department an application for a permit pursuant to Section 11of this regulation. Section 11.6 of this Regulation states: No permit shall be issued by the Department unless the applicant shows to the satisfaction of the Department that the equipment, facility, or air contaminant control device is designed to operate or is operating without causing a violation of the State Implementation Plan, or any rule or regulation of the Department, and without interfering with the attainment or maintenance of National and State ambient air quality standards, and without endangering the health, safety, and welfare of the people of the State of Delaware. The Department may, from time to time, issue or accept criteria for the guidance of applicants indicating the technical specifications which it deems will comply with the performance standards referenced herein. Furthermore, Section 11.8 states: The following emission rates and/or standards for each air contaminant emitted from any equipment, facility or air contaminant control device shall be specified in each permit issued pursuant to this regulation: a. The rate and/or standard established and/or relied upon in the State Implementation Plan

(SIP) to include the State of Delaware “Regulations Governing the Control of Air Pollution” and regulations promulgated pursuant to Section 111 and Section 112 of the Clean Air Act (CAA); and

b. The rate that was shown under Section 11.6 as not interfering with the attainment and maintenance of any National and State ambient air quality standard, and not endangering the health, safety, and welfare of the people of the State of Delaware; or

c. The rate requested by the applicant. In no case shall this rate be greater than the potential to emit of the equipment, facility, or air contaminant control device; and in no case shall this rate be less stringent than the rate specified in Section 11.8(a) and (b) of this regulation.

This regulation is applicable to the FCCU since changes involve alteration of equipment that emits air contaminants.



The FCCU was constructed in 1956 as a conventional Kellog designed Orthoflow Model B converter unit and a design throughput capacity of 58 MBPD of mixed gas oil derived from heavy sour crudes. The FCCU has been revamped twice; in 1973 its design was modified from a bed cracker to a dual folded riser. The FCCU was revamped again in 1995 when the dual internal riser design was replaced by a single external riser with rough cut cyclones to serve as the catalyst disengaging device. The application indicates the FCCU’s present feed rate is 82 MBPD of refinery intermediate products. This regulation is applicable to the FCCU and the FCCU COB since changes involve alteration of equipment that emits air contaminants. In evaluating and determining the expected FCCU emissions changes associated with the PCUP, Motiva (Premcor’s predecessor) compared baseline emissions during the period commencing October 1, 2002 through December 31, 2003 to the future PTE. Emissions data from this 15 month baseline period was then normalized to appear as annual baseline emissions levels used in the application. In contrast, the emissions changes associated with all other PCUP affected emissions units were made by comparing the 2001 – 2002 baseline time period to the future PTEs. The application states that the data and calculations used to develop an alternative baseline for the FCCU are found in the spreadsheet titled “FCCU Baseline”. The application also states that the alternative 15 month baseline period has been used rather than the 24 month period normally used in determining past actual emissions under PSD because the alternative 15 month period contains the only data available that is representative of normal source operation. Premcor has stated that the basis for the alternative 15 month baseline period being more representative of normal source operation is because this (15 month) period reflects operation of the FCCU under conditions that require compliance with another CD requirement, i.e., the requirement to comply with the 40 CFR Part 60, subpart J carbon monoxide limit of 500 ppmvd. AQM does not agree with the use of the time period beginning October 2002 through December 2003 as being representative of normal operations and consequently finds the baseline calculations provided in the application to be erroneous. AQM’s review of the annual inventory data shows that Motiva reported NOx emissions of 490 tons in 2001 and 411.7 tons in 2002 resulting in an average of 450.8 tons during the baseline period. On the other hand, Motiva’s PCUP application using the October 2002 through December 2003 baseline shows the average NOx emissions to be 1,109 tons. The difference between the two baselines shows an increase of annual NOx emissions of 659 tons allegedly caused by the operational changes made to comply with the NSPS limitation for CO. It is AQM’s assessment that the operational changes in 2002 should have been subject to review under NSR and the appropriate permitting thereof. AQM also disagrees with the PTEs shown in the application for the following reasons:



1. As explained above, an unpermitted increase in NOx emissions of approximately 650 TPY occurred around the same time when operational changes to the unit were made to comply with the NSPS CO limitation.

2. In addition to the unpermitted 650 TPY increase Premcor’s application seeks an additional 24.9 TPY increase thus requesting a PTE of 1,133.9 TPY. AQM not only finds the unpermitted 650 TPY increase to be unacceptable and subject to potential enforcement action, it also finds the requested increase of 24.9 TPY to be totally incompatible with a fundamental goal of the CD by which reductions were to be obtained.

3. The PTEs for H2SO4, and consequently PM107, emissions appear grossly overestimated.

4. The PTE for VOC emissions appear to be based on the FIRE emission factor of 2.8 lb/mmSCF. However, AQM’s review of the FIRE database showed this emission factor to be unrated or unratable (U) which means the data is of questionable or of low quality.

Table 2 shows a comparison of the emissions changes associated with the FCCU. This table is organized as follows: Column 1 lists the pollutants whose emissions changes are being evaluated8; Columns 2 and 3 show the baseline emissions during 2001 and 2002; Column 4 shows the average baseline emissions during 2001 – 2002 and is labeled as “AQM Baseline”; Column 5 labeled as “Premcor Baseline” shows the emissions used in Premcor’s alternative baseline period; Column 6 shows AQM’s calculated PTEs, Column 7 shows the PTEs used in Premcor’s application; Column 8 shows the net emissions changes between the AQM calculated PTEs and the Case 1 baseline levels; and finally, Column 9 shows the net emissions change between Premcor’s PTEs and the Case 2 baseline emissions levels.

7 H2SO4 emissions are a subset of PM10 emissions. The application recognizes this fact and adds the H2SO4 emissions to the TSP emissions to compute the PTE for PM10 emissions. 8 In addition to the pollutants listed here, there are insignificant changes in the emissions of lead, total reduced sulfur compounds and hydrogen sulfide.



Table 2: FCCU Emissions Comparison9 (TPY)

Net ∆ Pollutant 2001 2002 AQM Baseline

Premcor Baseline

AQM PTE10

Premcor PTE AQM Premcor

SO2 9586 11,420 10,503 13,995 361 361 -10,142 -13,634 NOx 490 412 451 1,109 451 1,133.9 0 24.9 TSP 1,245 1,017 1,131 973 TBD 311 TBD -662

H2SO4 41811 498 458 880 309 670.8 -149 -209 PM10 934 763 848.5 1,852.7 TBD 981.8 TBD -870.9 CO 1,369 1,495 1,432 1,673.8 3,768 3,768 2,094 2093.9

VOC 109 120 114.5 112 TBD 116.5 TBD Emissions of criteria pollutants from the FCCU vary with the unit throughput, coke burn rate and sulfur and nitrogen content in the feed, and other process operating parameters. Thus, as the throughput and the coke burn rate increase, the emissions of all pollutants will increase proportionally. The FCCU has an existing permit limit of 18,100 TPY SO2 emissions12. This limit was established as an alternative to limitations on the coke burn rate to 56,000 lb/hour and the average feed sulfur content to 2.12 percent, both on a twelve month rolling average basis.13 It would appear that once construction of the WGS is complete, the SO2 limitation will no longer serve this purpose restrictions will be rendered moot. However, because the FCCU’s requested throughput increase will result in increases in emissions of other pollutants, such as NOx, and the application has not addressed these emission increases, AQM believes these restrictions remain valid and relevant to ensuring emissions rates of these other pollutants do not increase by

9 The Case 1 baselines in this emissions comparison table reflect the calendar year average emissions as reported in the emissions inventory for the years 2001 and 2002. Ideally, AQM would have preferred a 24 month period prior to October 1, 2002, i.e. the date of commencement of operation of the FCCU for compliance with the CO NSPS limit. AQM has asked Premcor in a document titled AQM Discussion Items for the First Technical Meeting on PCUP – Phase II to submit the emissions data for all pollutants for the 24 month period preceding the operational changes made to comply with the NSPS CO requirement. However, to date this information has not been submitted by Premcor. 10 The PTEs shown in this column do not entirely agree with those shown in Premcor’s application. Specifically, AQM does not agree with the following PTEs: NOx 1,133.9 TPY, H2SO4 670.8 TPY, PM10 981.8 and VOC 116.5 TPY. 11 Case 1 baseline H2SO4 emissions are based on 2.85 % conversion of the SO2 to SO3 from stack testing conducted in 2001. 12 Permit: APC-90/0264-OPERATION (Amendment 2)(NSPS) dated July 29, 2004. 13 Permit: APC-90/0264-CONSTRUCTION (Amendment 1)(NSPS) dated June 5, 1997.



significant amounts above their baseline levels. Therefore, from the standpoint of practical enforceability, AQM recommends reinstating the coke burn rate of 56,000 lb/hour. The following technical analysis discusses the emissions changes associated with each pollutant based on the assumptions and design specifications given below: Sulfur dioxide (SO2) Gas flow to scrubber, Qi, w: 69,000 lb.mole/hour (wet) Moisture content, % H2O: 10.95 mole % Oxygen content, % O2, w: 3.0 mole % Inlet SO2 concentration SO2,i,d,0% O2; 1,231 ppmvd Outlet SO2 concentration SO2,o ,d,0% O2; 25 ppmvd The FCU WGS flow data is represented by the following equation:

( ) ( )Q QH O O

H OOH O

i d O i ww i w

, @ ,, , ,% % /

% /%% /

0%2 2

2

2

22 1

1001

1001 100

7921 1 100

=

⎡

⎣

⎢⎢⎢

⎤

⎦

⎥⎥⎥

− −−

+−

⎛⎝⎜

⎞⎠⎟

⎧⎨⎪

⎩⎪

⎫⎬⎪

⎭⎪

⎡

⎣⎢⎢

⎤

⎦⎥⎥

Q QSO

o d O i d Oi

, , , ,,

0% 2 0% 22

6125

10= −

−⎛⎝⎜

⎞⎠⎟

⎡

⎣⎢

⎤

⎦⎥

Based on the above assumptions: Moisture corrected inlet gas flow, Qi, d: [69,000 lb.mole/hour][1- %H2O/100] = 61,454 lb.mole/hour Then, the O2 and moisture corrected inlet gas flow, Qi, d, 0 % O2 is given by the expression: [Qi,,w][1- %H2O/100] [1-{(%O2, i, w /100)/(1- %H2O/100) + 79/21(%O2, i, w /100)/(1- %H2O/100)] = [69,000 lb.mole/hour][1- 0.1095][(1-{(0.0337 + 79/21(0.0337)}] lb-mole/hour = 51,540 lb.mole/hour The WGS is expected to lower the exit concentration of SO2 from 1,231 ppmvd to 25 ppmvd. When this adjustment is applied, the equation for the outlet gas flow under dry, oxygen free and clean condition will be given by the expression: Qo, d, 0 % O2, clean = [ Qi, d, 0 % O2][1- {(1,231-25)/106}] = 51,540 [1- 1,206/106] = 51,478 lb.mole/hour



The contaminant emission rate is: [25 parts SO2 /106 parts of flue gas][ Qo, d, 0 % O2, clean] = [51,478 lb.mole flue gas/hour][ 25 parts SO2 /106 parts of flue gas] = 1.286 lb.mole SO2/hour = 361 TPY Nitrogen oxides (NOx): AQM has not calculated the FCCU NOx PTE and is not establishing an emission limit at this time. However, it is claimed that operational changes made to the FCCU and its COB in order to reduce FCCU CO emissions (to below the NSPS limit of 500 ppmvd) was most likely the cause of the greater than 650 TPY increase in NOx emissions. AQM agrees that the increased FCCU NOx emissions appear to coincide with the changes made to comply with the CO limit. However, AQM is not convinced that the operational changes did not go beyond that purpose. AQM has requested this information, but Premcor has not provided it. AQM has issued Notices of Violation to both Motiva and Premcor, and intends addressing the increase in FCCU NOx emissions in an enforcement context. Simply stated, the FCCU appears to have undergone modifications without permits which resulted in a significant emissions increase of NOx, thus triggering provisions for Regulation No. 25 for installation of Lowest Achievable Emission Rate technology and requirement of offsets. AQM intends establishing a NOx emission limit as part of the enforcement action for this NSR violation. AQM believes the NSR violations, if upheld, would require the installation of technology, such as Low Temperature Oxidation (LoTOx process) where ozone is injected into the FCCU flue gas stream to oxidize insoluble NOx to more soluble oxidized dinitrogen pentoxide, which can then be removed with relative ease in a downstream scrubber used for particulate and SO2 removal. Since the LoTOx process operates optimally below 300 deg F., it does not require heat input to maintain operational efficiency and can enable maximum heat recovery from hot flue gases. Ozone is produced in response to the amount of NOx present in the flue gas and the final NOx emission required. The low operating temperature allows stable and consistent control regardless of variations in flow, load or NOx content. Ozone converts the NO and NO2 to N2O5 which is highly soluble in water making this technology particularly compatible with FCCU WGS applications for PM and SO2 control. The dilute acidic scrubber liquor can be easily neutralized by caustic before being discharged through the WWTP. Some of the salient features14 of the LoTOx technology that are noteworthy include:

14 Confuorto, Nicholas, Sexton, Jeffrey, Barasso, Michael and Suchak, Naresh; LoTOxTM Technology Demonstration at Marathon Ashland Petroleum LLC’s Refinery in Texas City, Texas; Technical Paper Presented at the NPRA Annual Meeting, March 23-25, 2004, San Antonio, TX.



• A successful pilot test was conducted at the Marathon refinery in Texas City. This test

demonstrated NOx emissions can be reduced to below 10 ppmv using LoTOx in conjunction with the EDV WGS

• NOx reduction occurs without conversion of SO2 to SO3 • FCCU catalyst and sulfur compounds do not adversely affect ozone consumption or NOx

removal efficiency • NOx removal efficiency can be instantaneously varied by adjusting the NOx outlet set

point which in turn adjusts the ozone injection rate • High NOx removal efficiencies are obtainable during transient conditions such as start up

and upset conditions • The technology can be retrofitted to existing FCCU WGSs as is the case with

Marathon’s Texas City refinery where it is presently being installed Based on the above discussion, AQM finds the LoTOx control technology as an appropriate and technically viable control technology. Total suspended Particulate (TSP) Matter: Baseline emissions are 973 TPY as shown in the application for the alternate baseline period of October 1, 2002 through December 31, 2003. The average baseline emissions for 2001 and 2002 from AQM’s inventory records is 1,131 TPY.15

The PTE is based on the vendor (Belco) pre-scrubber guarantee of 1 lb of TSP (filterable)/1,000 lb of coke burn. AQM will incorporate this limit in the permit. Sulfuric Acid (H2SO4): Baseline emissions are shown as 880 TPY based on Premcor’s assumption of 4.11 % oxidation factor and a 95 % confidence interval of the test data. AQM questioned the validity of the assumptions made because Premcor had used an oxidation factor of 0.91 % during the PCUP Phase 1 review. Premcor submitted a response on January 5, 2005 which includes a table showing 6 test runs (the first 3 dated 8/18/1994 and the last 3 with unknown dates). AQM’s review of its records show that the last 3 runs may have been performed on 8/1/2001 because the test results in the Table are identical to the results in AQM’s files. In fact, AQM used the 8/1/2002 test results to calculate an oxidation factor of 2.85 % as summarized below in Table 3: 15 TSP2001 = 1,245 tons; TSP2002 = 1,017 tons



Table 3: 8/1/2002 Stack Test Data H2SO4 (lb/hour) SO2 (lb/hour)

Run 1 63.6 2720 Run 2 54.1 2738 Run 3 67.2 2728

Average 61.6333 2728.667 Std. Dev. 6.7678 9.0185

T0.975 4.303 4.303 CC16 16.81 22.40

Average + CC 78.447 2751.072 H2SO4:SO2 0.0285

AQM proposes to use an oxidation factor of 2.85 percent to establish both the baseline H2SO4 emissions during the 24 month period immediately preceding October 1, 2002 and for the PTE. Premcor has been asked to submit the baseline emissions data for the 24 month period immediately preceding October 1, 2002. However, this information has not been submitted at this point. Therefore, AQM will incorporate the annual TPY H2SO4 emissions limit according to the following equation: Baseline H2SO4 emissions = [0.0285 lb H2SO4/ lb SO2] [Average baseline SO2 emission rate17] With respect to the PTE calculations for H2SO4 emissions, AQM research again indicates Premcor’s approach has grossly overestimated the actual quantity likely to be emitted. In a recent study conducted by the ExxonMobil Research and Engineering Company18, FCCU H2SO4 emissions showed no correlation with SO2 with the regression equation being y = -0.0019 x + 5.2796 and a value of R2 = 0.0002. The study further shows that approximately 90.3 % of the H2SO4 histogram data set evaluated (including 47 additional test results from FCCUs equipped with WGSs) showed less than 10 ppm H2SO4 and 90.8 % of the cube root normal distribution of the data set showed less than 10 ppm H2SO4. These results when expressed as a probability of emissions performance indicate only 1 reading in 10 will exceed 10 ppm, 1 in 20 will exceed 12 ppm and 1 in 100 will exceed 16 ppm. AQM’s stack test data with its average H2SO4 16 CC = t0.975 [σ]/√n 17 Because the FCCU has been equipped with a CEMS since 2000, the CEMS data for the 24 monthe period preceding October 1, 2002 will be acceptable. 18 W.X. Nie, G. Shea, T.P. Yarnich and J.H. Siegell of the ExxonMobil Research and Engineering Co.; Analysis Estimates Sulfuric Acid Emissions From FCCU Wet Gas Scrubbers; Oil and Gas Journal; [pages 62 – 64]; January 17, 2005.



concentration of 11 ppmv with a 95 % confidence interval would provide a reasonable basis to estimate the PTE. Thus AQM has based the PTE on an exit concentration of 14 ppmv. PTE H2SO4 = [14 parts H2SO4 /106 parts of flue gas][ Qo, d, 0 % O2, clean] = [51,478 lb.mole flue gas/hour][ 14 parts H2SO4/106 parts of flue gas] = 0.72 lb.mole H2SO4/hour = 309 TPY Particulate Matter with an Aerodynamic Diameter Less than 10 Microns (PM10): Baseline PM10 emissions are shown in the application as being 1,852.7 TPY. The average baseline emissions for 2001 and 2002 from AQM’s inventory records is 848.3 TPY19

However, the baseline PM10 emissions include 879.8 TPY H2SO4 based on Premcor’s assumption of 4.11 % oxidation factor and a 95 % confidence interval of the test data. As discussed above, AQM does not find the use of the proposed oxidation factor of 4.11 percent to be supported by the stack test data at AQM’s disposal. Therefore, AQM will incorporate a condition that annual PM10 emissions to the sum of the TSP and H2SO4 emissions. Carbon Monoxide (CO): Premcor’s alternative baseline CO emissions of 1,674 TPY were obtained from the CEMS installed to comply with the NSPS provisions of the CD. AQM’s research of its inventory database yielded an average CO emissions of 1,432 TPY but this figure was based on the FIRE 6.24 emission factor for petroleum refinery fuel gas combustion (35 lb CO/mmSCF process gas). AQM is cognizant of the fact that the CEMS data used in the alternate baseline represents quality assured real time data that is representative of FCCU CO emissions during the period October 1, 2002 through December 31, 2003. Ideally, AQM would have preferred the 24 month period prior to October 1, 2002, i.e. the date of commencement of operation of the FCCU for compliance with the CO NSPS limit, but the emissions data from this period would have to be obtained from an estimation technique using the FIRE 6.24 emissions factor (35 lb CO/mmSCF process gas) similar to that used for the 2001-2002 baseline emissions reported in the emissions inventory. Furthermore, because operational changes were CD driven to ensure compliance with the NSPS CO limitation, the actual CO emissions prior to making the changes would have been higher than those estimated using the FIRE 6.24 emissions factor, thus artificially inflating the baseline. Therefore, AQM finds Premcor’s lower alternate baseline emissions of 1,674 TPY to be acceptable for evaluating the net CO emissions change. Baseline = 1,674 TPY

19 PM10, 2001 = 934 tons; PM10, 2002 = 762.6 tons



Premcor has requested a PTE of 3,768 TPY based on the FCCU having an uncorrected CO exit concentration limit that is less than 500 ppmvd. Thus PTE CO = [Qi, d][500/106][1 lb.mole/385.5 DSCF][28 lb CO/lb.mole CO][385.5 DSCF/lb.mole][ 4.38 Ton-hour/lb.year] = [61,445 lb.mole/hour][500/106][1 lb.mole/385.5 DSCF][28 lb CO/lb.mole CO][385.5 DSCF/lb.mole][ 4.38 Ton-hour/lb.year] = 3,768 TPY Net increase in CO emissions = 2,094 TPY Volatile Organic Compounds (VOCs): Baseline VOC emissions were calculated using the FIRE 6.23 emission factor for petroleum refinery fuel gas combustion (2.8 lb VOC/mmSCF process gas) Baseline = 112.3 TPY Premcor has requested a PTE of 116.5 TPY based on the FCCU COB combusting a mix of 0.5 mmSCF/hour of RFG and 9 mmSCF/hour of process off gas. The Company has used the same emission factor for both fuels, i.e 2.8 lb VOC/mmSCF. Premcor requested VOC PTE = [2.8 lb VOC/mmSCF][9.5 mmSCF/hour][ 4.38 Ton- hour/lb.year] = 116 TPY. AQM reviewed the FIRE emission factor for VOC emissions from CO boilers and confirmed Premcor’s use of 2.8 lb/mmSCF. However, AQM’s review also found this emission factor to be unrated or unratable (U) which means that the data is of questionable or of low quality. Furthermore, FIRE unequivocally states the emission factor was transferred from natural gas burning boilers assuming process similarity. During the Phase I technical discussions with the Company, AQM raised the question of whether such a factor was adequately representative of COB operations. Premcor did not provide a meaningful response or an alternative to the FIRE emission factor. However, Premcor has since conducted a stack test on the FCU. Although, the final results of this testing has not been submitted at this point, AQM representatives during the test runs observed very low levels of VOCs, perhaps an order of magnitude lower than the PTE sought by Premcor. Therefore, by adopting a similar approach in evaluating FCCU VOC emissions, AQM believes it is necessary to stack test the FCCU to determine the FCCU’s PTE. Therefore, the final permit limit will be an alternative emission limit derived from a stack test conducted under representative operating conditions. AQM believes it is following good engineering and air pollution control practices when it prescribes a meaningful and practically enforceable emission limitation that is supported by a test conducted under representative operating conditions.



Lead (Pb), Hydrogen Sulfide (H2S) and Reduced Sulfur Compounds (RSC): Baseline Pb, H2S and RSC emissions for the 24 month period preceding October 1, 2002 have not been submitted by Premcor up to this point. Premcor developed estimated coke burn rates by multiplying the maximum coke burn rate (71,000 lb/hour) by the ratio of the actual charge rate to the maximum unit charge rate. Premcor then used the estimated coke burn rates to calculate Pb, RSC and H2S emissions by computing the product of the estimated coke burn rates with the respective emission factor for each of these pollutants20. AQM believes the difference in emissions of these pollutants between the two coke burn rates of 71,000 lb/hour vis a vis 56,000 lb/hour will be very small and will not alter the netting transaction and therefore will not trigger any new regulatory requirements Baseline Pb emissions = 1.1 E-01 TPY PTE Pb emissions = 1.4 E-01 TPY Net emissions change = 3.0 E-02 TPY Baseline RSC/H2S emissions = 8.9 E-03 TPY PTE RSC/H2S emissions = 1.1 E-02 TPY Net emissions change = 2.5 E-03 TPY Compliance Methodology: The draft permit will include the following conditions21: FCCU emissions shall not exceed:

• SO2: 25 ppmvd @ 0 % O2 on a 365 day rolling average, 50 ppmvd @ 0 % O2 on a 7 day rolling average and 361 TPY.

• NOx: Reserved • TSP emissions shall not exceed 1 lb TSP/1000 lb coke burn • H2SO4: 40 % reduction across the Belco pre-scrubber and 309 TPY • PM10 emissions shall not exceed the sum of TSP and H2SO4 emissions on an annual

basis • CO: 500 ppmv and 3,768 TPY

20 The emissions factors for these pollutants were obtained the Exxon R&E Guidance. The emission factor for Pb is 4.37 E-04 lb Pb/klb coke burned; the emission factor for RSC/H2S is emissions factor 3.68 E-05 lb H2S/RSC /klb coke 21 TPY is defined as “tons per rolling twelve months”



• VOC as CH4: To be determined. • Pb: 4.37 E-04 lb/Klb coke burn and 1.4 E-01 TPY • H2S/RSC: 3.68 E-05 lb/Klb coke burn and 1.1 E-02 TPY • HAPs (as Ni)22: To be determined. • The FCCU coke burn rate shall not exceed 56,000 lb/hour on a rolling twelve month

basis. Monitoring/Testing:

• SO2: Compliance shall be based on CEMS. The CEMS shall be installed and certified by satisfying the requirements of Performance Specifications No. 2 in Appendix “B” of 40 CFR part 60. The QA/QC procedures for the CEMS shall be established in accordance with the procedures in Appendix “F” of 40 CFR part 60.

• NOx: As explained above, AQM is deferring the incorporation of a new emission limitation at this time, based on a lack of data relating to NOx emissions caused by what appears to be an unpermitted modification to the cat cracker CO boiler which will likely result in an enforcement action. However, AQM will continue to require Premcor to continuously monitor FCCU NOx emissions using CEMS. The CEMS shall conform to the applicable Performance Specifications in Appendix “A” of 40 CFR Part 75. The QA/QC procedures for the CEMS shall be established in accordance with the procedures in Appendix “B” of 40 CFR Part 75.

• TSP (filterable): Compliance shall be based on an initial Reference Method 5 testing in Appendix “A” of 40 CFR Part 60, and annually thereafter.

• H2SO4: Compliance shall be based on an initial Reference Method 8 testing in Appendix “A” of 40 CFR Part 60, and annually thereafter.

• PM10: Compliance shall be based on an initial Reference Method 5/202 testing in Appendix “M” of 40 CFR Part 51, and annually thereafter

• CO: Compliance shall continue to be based on CEMS. The CEMS shall conform to the applicable requirements of Performance Specifications No. 4 in Appendix “B” of 40 CFR part 60. The QA/QC procedures for the CEMS shall be established in accordance with the procedures in Appendix “F” of 40 CFR Part 60.

• VOC as CH4: Compliance shall be based on an initial Reference Method 25A testing in Appendix “A” of 40 CFR Part 60, and annually thereafter.

• Pb: Compliance shall be based on an initial Reference Method 12 testing in Appendix “A” of 40 CFR Part 60. Future compliance shall be based on the stack test based

22 See discussion under applicability of 40 CFR 63, subpart UUU.



emission factor in terms of lb/1,000 lb coke burn rate. The Department reserves the right to require more frequent testing if warranted.

• H2S/RSC: Compliance shall be based on an initial Reference Method 15 testing in Appendix “A” of 40 CFR Part 60. Future compliance shall be based on the stack test based emission factor in terms of lb/1,000 lb coke burn rate. The Department reserves the right to require more frequent testing if warranted.

• All monitor certifications shall be conducted within 60 days of the unit attaining maximum production but not later than 180 days after unit start up. A “Source Sampling Guidelines and Preliminary Survey Form” must be submitted and found acceptable to the Department at least thirty (30) days prior to the performance testing. Results of the Performance Specification testing shall be submitted to the Department, in triplicate, within 90 days after completion of the testing.

Recordkeeping: The Company shall maintain records of CEMS data, stack test results, the daily COB fuel usage, rolling 365 day coke burn rate and FCCU feed throughput. Reporting: Quarterly reports for the preceding quarter shall be submitted to the Department by January 31, April 30, July 31 and October 31 of each calendar year with the following information:

• A summary of all excess emissions for the quarter; • CEMS report to include system calibration and audit results, the actual daily data capture

for the period and details of out of control periods • Exceedances of the rolling 365 day limits of FCCU throughput and coke burn rates.

Regulation No. 3 – Ambient Air Quality Standards: Section 2.1 of this regulation states: No person shall cause the Air Quality Standards specified in this Regulation to be exceeded.

• Premcor has conducted a modeling analysis for normal operation using the EPA approved Industrial Source Complex Short Term (ISCST3 version 02035) model and 5 years of representative meteorological data to show no adverse impacts to ambient air quality standards during normal FCCU operation. Under normal operation with the flue gas routed through the WGS, the majority of the SO2 will be captured. However, during periods of upset operations that require the WGS to be bypassed the pollutant will be emitted uncontrolled to the atmosphere. Premcor’s modeling exercise has not evaluated



the impacts of these uncontrolled emissions on ambient air quality. AQM ran the Industrial Source Complex Short Term (ISCST3 version 02035) model with the 5 year meteorological data supplied by Premcor for uncontrolled emissions using the following stack parameters:

Diameter: 9 feet Height: 230 feet Exit temp. 850 deg F Exit velocity 68.7 feet/sec Flow: 859,668 acfm Table 4 summarizes the AQM’s modeled results.

Table 4: Model Results of Uncontrolled FCU SO2 Emissions Averaging

Time Standard NAAQS (µg/m3) Maximum

value23

(µg/m3)

Maximum fenceline

concentration (µg/m3)

3 hour Secondary 1300 1,409 677 24 hour Primary 365 443 432 Annual Primary 80 31 17

As can be seen from the above table, the model predicts a likelihood of exceeding the 24 hour standard of 365 µg/m3 during periods when the control device is bypassed. During the PCUP Phase I review, AQM had observed a similar model prediction that led to technical discussions with Premcor regarding actions to be taken during periods when the control device has to be bypassed. However, such technical discussions have not occurred during the Phase II review. Therefore, AQM will incorporate a permit condition requiring Premcor to submit alternative operating scenarios for AQM’s approval that address start up and shut down conditions. This shall be submitted at least six (6) months prior to start up of the FCCU WGS.

Regulation No. 5 – Particulate Emissions from Industrial Process Operations: Section 5.2 of this regulation states: No person shall cause or allow particulate emissions from fluid coking operations into the atmosphere in excess of the quantities as indicated in Table 3.

23 The maximum modeled concentrations were located within the property lines of the facility east of Route 9 near the marketing terminal.



Compliance Methodology: As explained above, all particulate emissions (including H2SO4) are considered to be PM10. The PTE for PM10 will be determined as the sum of the PTE for TSP and H2SO4 emissions. Although, the final PM10 PTE is yet to be determined, AQM in exercising its good engineering judgement believes, the installation of the WGS will result in the final PM10 emission limit that will be well below the emission rate prescribed by Table 3 of Regulation 5. Therefore, compliance with this regulation is based on compliance with the emission limit in the permit (see discussion under Regulation No. 2). Monitoring/Testing, Recordkeeping and Reporting: None in addition to those discussed under regulation No. 2. Regulation No. 8 – Sulfur Dioxide Emissions From Fuel Burning Equipment: Section 2.1 states: Except as provided in Section 2.2, no person shall offer for sale, sell, deliver, or purchase any fuel having a sulfur content greater than one (1.0) percent by weight when such fuel is intended for use in any fuel burning equipment in New Castle County. No person shall use any fuel having a sulfur content greater than one (1.0) percent by weight in any fuel burning equipment in New Castle County. This regulation applies only to the FCU COB which meets the definition of fuel burning equipment. Compliance Methodology: The FCCU COB combusts desulfurized RFG and process offgas containing CO and sulfur species that result in 1,235 ppmvd SO2 loading to the WGS. The flue gases are discharged to meet the CD mandated emission level of 25 ppmvd at 0 % O2 on a 365 day rolling average basis and 50 ppmvd at 0 % O2 on a 7 day rolling average basis. Both these limits are several orders of magnitude more stringent than the corresponding regulatory limit of 1 % S in fuel. Therefore, no additional discussion is presented here. Regulation No. 8 – Emissions of Sulfur Dioxide From Industrial Operations: Section 1.1 of this regulation states:



The emission of sulfur dioxide from process operations shall be controlled to a limit that shall meet the ambient air quality requirements. Compliance Methodology: The Company has modeled emissions from the refinery using the ISC3 ST (version 02035) to demonstrate no adverse impacts to ambient air quality. However, the Company has not modeled impacts during atypical operations which frequently result in outages of the COB which in turn causes bypassing the WGS. [See discussion under Regulation No. 3] Regulation No. 11 – Carbon Monoxide Emissions From Industrial Process Operations – NC County Section 2.1 of this regulation states: In New Castle County, no person shall cause or allow the emission of carbon monoxide from any catalytic regeneration of a petroleum cracking system, petroleum fluid coker, or other petroleum process into the atmosphere, unless the carbon monoxide is burned at 1300 F for 0.3 seconds o or greater in a direct-flame afterburner or boiler, or is controlled by an equivalent technique. Compliance Methodology: CO is combusted in the furnace of the FCCU COB. The design operating temperature in the FCU COB at the boiler nose is 1750 deg. F. With a firebox volume of 25,920 ft3, the average residence time of the flue gas is 3.52 sec.24 Monitoring/Testing: The Company shall monitor the FCU COB firebox temperature continuously. Recordkeeping: The company shall maintain records of the firebox temperature for 5 years. Reporting:

24 Tr = 25,920 Ft3/[69,000 lb.mole/hour][385.5 DSCF/lb.mole][2.77 E-04 hour/sec] = 3.52 sec.



The Company shall submit quarterly reports of all periods when the firebox temperature fell below 1300 deg F. Each report shall be submitted within 30 days of the end of the quarter. Regulation No. 14 – Visible Emissions: Section 2.1 of this regulation states: No person shall cause or allow the emission of visible air contaminants and/or smoke from a stationary or mobile source, the shade or appearance of which is greater than twenty (20%) percent opacity for an aggregate of more than three (3) minutes in any one (1) hour or more than fifteen (15) minutes in any twenty-four (24) hour period. Compliance Methodology: As observed from the material balance the WGS stack will have a significant moisture content (about 19 mole %) that will make the use of a continuous opacity monitoring system impractical. Therefore, evaluations of visible emissions will have to be made at a point where the condensed water is no longer visible in accordance with 40 CFR Part 60, Appendix “A”, Reference Method 9, Section 2.3.1. The Company shall maintain a log of daily qualitative observations using Reference Method 22 in Appendix “A” of 40 CFR part 60. In the event visible emissions are observed by reference Method 22, the Company shall conduct a visible emissions evaluation using the procedures described in Regulation 20, section 1.5 (c). At all other times, compliance shall be based on the proper operation of the emissions unit and record keeping. Monitoring/Testing: Conduct visual observations at fifteen-second intervals for a period of not less than one hour except that the observations may be discontinued whenever a violation of the standard is recorded. The additional procedures, qualification and testing used for visually determining the opacity shall be those specified in Section 2 and 3 (except for Section 2.5 and the second sentence of Section 2.4) of reference Method 9 set forth in Appendix A, 40 CFR Part 60, revised July 1, 1982. Recordkeeping: Observation records shall be maintained on site. Records of all maintenance performed on applicable units shall be maintained.



Reporting: The Company shall submit quarterly reports indicating the duration and magnitude of all periods of excess opacity. Regulation No. 17 – Source Monitoring, Recordkeeping and Reporting:

Section 2.1 of this regulation states: Upon written request of the Department, an owner or operator of an air contaminant source shall, at his expense, install, maintain, and use emission monitoring devices, keep records, and make periodic reports to the Department on the nature and amount of emissions from such source. The Department shall make such data available to the public as reported and as correlated with any applicable emission standards or limitations. Section 2.2 of this regulation states: Upon written request of the Department, an owner or operator of an air contaminant source shall, at his expense, sample the emissions of, or fuel used by, that source, maintain records and submit reports to the Department on the results of such sampling. The Department may make such data available to the public as reported and as correlated with any applicable emission standards or limitations. Section 2.2 of this regulation states: The Department may conduct tests of emissions from or fuel used by any air contaminant source. Upon written request of the Department, the owner or operator of the air contaminant source shall provide necessary holes in stacks or ducts, and such other safe and proper sampling and testing facilities, exclusive of instruments and sampling devices, if any are necessary, for proper determinations of the emission of air contaminants. The Department shall have access to and use of monitoring, record-keeping and reporting required by Federal Regulations relating to emissions of air contaminants. The Department may make such data available to the public as reported or received and as correlated with any applicable emissions standards or limitations. Compliance Methodology: Compliance shall be based on the monitoring/testing requirements Monitoring/Testing, Recordkeeping and Reporting: None in addition to those included under the applicability of Regulation No. 2.



Regulation No. 19 – Control of Odorous Air Contaminants This regulation is applicable to all affected emissions units in the refinery. Therefore, its applicability is covered under the heading “Facility Wide Applicable Requirements” in this memorandum. Regulation No. 20 & 40 CFR Part 60, Subpart J – New Source Performance Standards Section 104 (a) of 40 CFR part 60, Subpart “J” is an applicable requirement for the FCCU COB. This requirement restricts the H2S content in the RFG combusted to less than 162 ppmvd. The H2S concentration in the RFG is already monitored by CEMS and thus creates no new requirement. Therefore, no additional discussion is presented here. Regulation No. 24 – Control of Volatile Organic Compound Emissions Section 28 applies to any vacuum-producing system, wastewater separator, and process unit turnaround at petroleum refinery sources. No exemptions are allowable based on size or throughput of a facility. This section is applicable to the crude unit because it contains vacuum producing systems. It is also applicable during unit turnarounds. Standards. Vacuum-producing systems: No person shall permit the emission of any uncondensed volatile organic compound (VOC) from the condensers, hot wells, or accumulators of any vacuum producing system at a petroleum refinery. Process unit turnarounds: The Company shall provide for the following during process unit turnaround: i. Depressurization venting of the process unit or vessel to a vapor recovery system, flare, or firebox. ii. No emission of VOC from a process unit or vessel until its internal pressure is 136 kiloPascals (kPa) (19.7 pounds per square inch atmospheric [psia]) or less. Compliance Methodology for Vacuum Producing Systems: The standard shall be achieved by either of the following:

i. Piping the uncondensed vapors to a firebox or incinerator. ii. Compressing the vapors and adding them to the refinery fuel gas.



Compliance Methodology for Process Unit Turnarounds: The Company shall provide for the following during process unit turnarounds:

i. Depressurization venting of the process unit or vessel to a vapor recovery system, flare, or firebox.

ii. No emission of VOC from a process unit or vessel until its internal pressure is 136 kiloPascals (kPa) (19.7 pounds per square inch atmospheric [psia]) or less.

Recordkeeping and Reporting : The Company shall keep records of the following items:

i. Date of every process unit or vessel turnaround; and ii. The internal pressure of the process unit or vessel immediately prior to venting to the atmosphere.

Section 29: Leaks from Petroleum Refinery Equipment applies to all equipment in volatile organic compound (VOC) service in any process unit at a petroleum refinery, regardless of size or throughput. Compliance Methodology: The Company shall comply with the following standards in sub-sections c through h in Section 29 of Regulation no. 24. c. Standards: General. The Company shall ensure that: 1. Any open-ended line or valve is sealed with a second valve, blind flange, cap, or plug

except during operations requiring process fluid flow through the open-ended line or valve.

2. When a second valve is used, each open-ended line or valve equipped with a second valve is operated in such a manner that the valve on the process fluid end is closed before the second valve is closed.

3. When a double block-and-bleed system is used, the bleed valve or line is open only during operations that require venting of the line between the block valves and is closed at all other times.

d. Standards: Equipment inspection program. The Company shall conduct the equipment

inspection program described in paragraphs (d)(1) through (d)(3) of this Section using the test methods specified in Appendix "F" of this regulation.



1. The Company shall conduct quarterly monitoring of each: i. Compressor. ii. Pump in light liquid service. iii. Valve in light liquid service, except as provided in paragraphs (e) and (f) of this

Section. iv. Valve in gas/vapor service, except as provided in paragraphs (e) and (f) of this

Section. v. Pressure relief valve in gas/vapor service, except as provided in paragraphs (e) and (f)

of this Section. 2. The Company shall conduct a weekly visual inspection of each pump in light liquid

service. 3. The Company shall monitor each pressure relief valve after each overpressure relief to

ensure that the valve has properly reseated and is not leaking. 4. When an instrument reading of 10,000 parts per million (ppm) or greater is measured, it

shall be determined that a leak has been detected. 5. If there are indications of liquid dripping from the equipment, it shall be determined that

a leak has been detected. 6. When a leak is detected, the Company shall affix a weatherproof, readily visible tag in a

bright color bearing the equipment identification number and the date on which the leak was detected. This tag shall remain in place until the leaking equipment is repaired. The requirements of this paragraph apply to any leak detected by the equipment inspection program and to any leak from any equipment that is detected on the basis of sight, sound, or smell.

e. Standards: Alternative standards for valves: Skip period leak detection and repair. 1. The Company shall comply initially with the requirements for valves in gas/vapor service

and valves in light liquid service, as described in paragraph (d) of this Section. 2. After two consecutive quarterly leak detection periods with the percent of valves leaking

equal to or less than 2.0, an owner or operator may begin to skip one of the quarterly leak detection periods for the valves in gas/vapor and light liquid service.

3. After five consecutive quarterly leak detection periods with the percent of valves leaking equal to or less than 2.0, an owner or operator may begin to skip 3 of the quarterly leak detection periods for the valves in gas/vapor and light liquid.

4. If the percent of valves leaking is greater than 2.0, the owner or operator shall comply with the requirements as described in paragraph (d) of this Section but can again elect to use the requirements in paragraph (e) of this Section.



5. The percent of valves leaking shall be determined by dividing the sum of valves found

leaking during current monitoring and valves for which repair has been delayed by the total number of valves subject to the requirements of this Section.

6. The Company shall keep a record of the percent of valves found leaking during each leak detection period.

f. Standards: Alternative standards for unsafe-to-monitor valves and difficult-to-monitor

valves. 1. Any valve that is designated, as described in paragraph (j)(5)(i) of this Section, as an

unsafe-to monitor valve is exempt from the requirements of paragraph (d) if: i. The Company of the valve demonstrates that the valve is unsafe to monitor

because monitoring personnel would be exposed to an immediate danger as consequence of complying with paragraph (d).

ii. The Company shall adhere to a written plan that requires monitoring of the valve as frequently as practicable during safe-to-monitor times.

2. Any valve that is designated, as described in paragraph (j)(5)(i), as a difficult-to monitor valve is exempt from the requirements of paragraph (d) if:

i. The Company demonstrates that the valve cannot be monitored without elevating the monitoring personnel more than 2 meters (m) (6.6 feet [ft]) above a support surface.

ii. The Company follows a written plan that requires monitoring of the valve at least once per calendar year.

g. Standards: Equipment repair program. The Company shall: 1. Make a first attempt at repair for any leak not later than 5 calendar days after the leak is

detected. 2. Repair any leak as soon as practicable, but not later than 15 calendar days after it is

detected except as provided in paragraph (h) of this Section. h. Standards: Delay of repair. 1. Delay of repair of equipment for which a leak has been detected is allowed if the repair is technically infeasible without a process unit shutdown. Repair of such equipment shall

occur before the end of the next process unit shutdown. 2. Delay of repair of equipment is allowed for equipment that is isolated from the process

and that does not remain in VOC service. 3. Delay of repair beyond a process unit shutdown is allowed for a valve, if valve assembly replacement is necessary during the process unit shutdown, valve assembly supplies have

been depleted, and valve assembly supplies had been sufficiently stocked before the supplies were depleted. Delay of repair beyond the next process unit shutdown is not



allowed unless the next process unit shutdown occurs sooner than 6 months after the first process unit shutdown.

Monitoring/Testing: The Company shall comply with the following testing procedures: 1. In conducting the tests required to comply with paragraph (d) of this Section, the Company shall use the test methods specified in Appendix "F" of this regulation 2. The Company shall test each piece of equipment as required under paragraph (d) of this

Section unless it is demonstrated that a process unit is not in VOC service, i.e., that the VOC content would never be reasonably expected to exceed 10 percent by weight. For purposes of this demonstration, the following methods and procedures shall be used:

i. Procedures that conform to the general methods in ASTM E260, E168, and E169 shall be used to determine the percent VOC content in the process fluid that is contained in or contacts a piece of equipment.

ii. Where the test methods in paragraph (i)(2)(i) also measure exempt compounds, these compounds may be excluded from the total quantity of organic compounds in

determining the VOC content of the process fluid. iii. Engineering judgment may be used to estimate the VOC content, if a piece of equipment

had not been shown previously to be in VOC service. If the Department disagrees with the judgment, paragraphs (i)(2)(i) and (i)(2)(ii) of this Section shall be used to resolve the

disagreement. 3. The Company shall demonstrate that a piece of equipment is in light liquid service by

showing one of the following: i. All of the following conditions apply: A. The vapor pressure of one or more of the components is greater than 0.3 kPa at 20°C (0.044 in. Hg at 68°F); standard reference texts or ASTM D2879 shall be used to determine the vapor pressures. B. The total concentration of the pure components having a vapor pressure greater than

0.3 kPa at 20°C (0.044 in. Hg at 68°F) is equal to or greater than 20 percent by weight.

C. The fluid is a liquid at operating conditions. ii. The percent VOC evaporated is greater than 10 percent at 150°C (302°F) as determined

by ASTM D86. 4. Samples used in conjunction with paragraphs (i)(2) and (i)(3) of this Section shall be representative of the process fluid that is contained in or contacts the equipment. Recordkeeping:



The Company shall be subject to the following recordkeeping requirements: 1. Except as noted, these records shall be maintained in a readily accessible location for a

minimum of 5 years and shall be made available to the Department immediately upon verbal or written request.

2. If there is more than one affected facility subject to the provisions of this Section the Company may comply with the recordkeeping requirements for these facilities in one recordkeeping system if the system identifies each record by each facility.

3. When each leak is detected as specified in paragraph (d) of this Section, the following information shall be recorded in a log and shall be kept for 5 years in a readily accessible location:

i. The instrument and operator identification numbers and the equipment identification number.

ii. The date the leak was detected and the dates of each attempt to repair the leak. iii. The repair methods employed in each attempt to repair the leak. iv. The notation "Above 10,000" if the maximum instrument reading measured by the

methods specified in Appendix "F" of this regulation after each repair attempt is equal to or greater than 10,000 ppm.

v. The notation "Repair delayed" and the reason for the delay if a leak is not repaired within 15 calendar days after discovery of the leak.

vi. The signature of the owner or operator (or designate) whose decision it was that repair could not be effected without a process shutdown.

vii. The expected date of successful repair of the leak if a leak is not repaired within 15 calendar days.

viii.The dates of process unit shutdowns that occur while the equipment is unrepaired. ix. The date of successful repair of the leak. 4. A list of identification numbers of equipment in vacuum service shall be recorded in a log

that is kept in a readily accessible location. 5. The following information pertaining to all valves subject to the requirements of paragraph

(f) of this Section shall be recorded in a log that is kept for 5 years in a readily accessible location:

i. A list of identification numbers for valves that are designated as unsafe to monitor, an explanation for each valve stating why the valve is unsafe to monitor, and the plan for monitoring each valve. ii. A list of identification numbers for valves that are designated as difficult to monitor, an explanation for each valve stating why the valve is difficult to monitor, and the schedule

for monitoring each valve. 6. The following information for valves complying with paragraph (e) of this Section shall be recorded in a log that is kept for 5 years in a readily accessible location: i. A schedule of monitoring.



ii. The percent of valves found leaking during each monitoring period as noted in paragraph (e)(6) of this Section. 7. Information and data used to demonstrate that a piece of equipment is not in VOC service

shall be recorded in a log that is kept for 5 years in a readily accessible location for use in determining exemptions as provided in paragraph (a) of this Section.

Reporting. The Company shall comply with the requirements in Section 5 of this regulation. 40 CFR 60, Subpart GGG

The provisions of this subpart apply to affected facilities in petroleum refineries. A compressor is an affected facility. In addition, the group of all the equipment (defined in §60.591) within a process unit is an affected facility.

Standards: The Company is subject to the provisions of this subpart shall comply with the requirements of §§60.482–1 to 60.482–10 as soon as practicable, but no later than 180 days after initial startup.

Compliance Methodology:

Compliance with this regulation is identical to the requirements under the refinery MACT in 40 CFR 63. Therefore, no additional discussion is presented here.

40 CFR 60, Subpart VV The provisions of this subpart apply to affected facilities in the synthetic organic chemicals manufacturing industry.


The Company shall demonstrate compliance with the requirements of §§60.482–1 through 60.482–10 or §60.480(e) for all equipment within 180 days of initial startup. The Company may request a determination of equivalence of a means of emission limitation to the requirements of §§60.482–2, 60.482–3, 60.482–5, 60.482–6, 60.482–7, 60.482–8, and 60.482–10 as provided in §60.484. If the Administrator makes a determination that a means of emission limitation is at least equivalent to the requirements of §§60.482–2, 60.482–3, 60.482–5, 60.482–6, 60.482–7,



60.482–8, or 60.482–10, an owner or operator shall comply with the requirements of that determination. 25

Monitoring/Testing:

The following monitoring/testing requirements are applicable:

(a) In conducting the performance tests required in §60.8, the Company shall use as reference methods and procedures the test methods in appendix A of this part or other methods and procedures as specified in this section, except as provided in §60.8(b).

(b) The Company shall determine compliance with the standards in §§60.482, 60.483, and 60.484 as follows:

(1) Method 21 shall be used to determine the presence of leaking sources. The instrument shall be calibrated before use each day of its use by the procedures specified in Method 21. The following calibration gases shall be used:

(i) Zero air (less than 10 ppm of hydrocarbon in air); and

(ii) A mixture of methane or n-hexane and air at a concentration of about, but less than, 10,000 ppm methane or n-hexane.

(c) The Company shall determine compliance with the no detectable emission standards in §§60.482–2(e), 60.482–3(i), 60.482–4, 60.482–7(f), and 60.482–10(e) as follows:

(1) The requirements of paragraph (b) shall apply.

(2) Method 21 shall be used to determine the background level. All potential leak interfaces shall be traversed as close to the interface as possible. The arithmetic difference between the maximum concentration indicated by the instrument and the background level is compared with 500 ppm for determining compliance.

(d) The Company shall test each piece of equipment unless he demonstrates that a process unit is not in VOC service, i.e., that the VOC content would never be reasonably expected to exceed

25 Equipment that is in vacuum service is excluded from the requirements of §§60.482–2 to 60.482–10 if it is identified as required in §60.486(e)(5).



10 percent by weight. For purposes of this demonstration, the following methods and procedures shall be used:

(1) Procedures that conform to the general methods in ASTM E260–73, 91, or 96, E168–67, 77, or 92, E169–63, 77, or 93 (incorporated by reference—see §60.17) shall be used to determine the percent VOC content in the process fluid that is contained in or contacts a piece of equipment.

(2) Organic compounds that are considered by the Administrator to have negligible photochemical reactivity may be excluded from the total quantity of organic compounds in determining the VOC content of the process fluid.

(3) Engineering judgment may be used to estimate the VOC content, if a piece of equipment had not been shown previously to be in service. If the Administrator disagrees with the judgment, paragraphs (d) (1) and (2) of this section shall be used to resolve the disagreement.

(e) The Company shall demonstrate that an equipment is in light liquid service by showing that all the following conditions apply:

(1) The vapor pressure of one or more of the components is greater than 0.3 kPa at 20 °C (1.2 in. H2O at 68 °F). Standard reference texts or ASTM D2879–83, 96, or 97 (incorporated by reference—see §60.17) shall be used to determine the vapor pressures.

(2) The total concentration of the pure components having a vapor pressure greater than 0.3 kPa at 20 °C (1.2 in. H2O at 68 °F) is equal to or greater than 20 percent by weight.

(3) The fluid is a liquid at operating conditions.

(f) Samples used in conjunction with paragraphs (d) and (e) of this section shall be representative of the process fluid that is contained in or contacts the equipment or the gas being combusted in the flare.

Recordkeeping requirements:

The following recordkeeping requirements are applicable:



(a) The Company shall comply with the recordkeeping requirements of section 60.486. Because

there are more than one affected facility subject to the provisions of this subpart the Company may comply with the recordkeeping requirements for these facilities in one recordkeeping system if the system identifies each record by each facility.

(b) When each leak is detected as specified in §§60.482–2, 60.482–3, 60.482–7, 60.482–8, and 60.483–2, the following requirements apply:

(1) A weatherproof and readily visible identification, marked with the equipment identification number, shall be attached to the leaking equipment.

(2) The identification on a valve may be removed after it has been monitored for 2 successive months as specified in §60.482–7(c) and no leak has been detected during those 2 months.

(3) The identification on equipment except on a valve, may be removed after it has been repaired.

(c) When each leak is detected as specified in §§60.482–2, 60.482–3, 60.482–7, 60.482–8, and 60.483–2, the following information shall be recorded in a log and shall be kept for 2 years in a readily accessible location:

(1) The instrument and operator identification numbers and the equipment identification number.

(2) The date the leak was detected and the dates of each attempt to repair the leak.

(3) Repair methods applied in each attempt to repair the leak.

(4) “Above 10,000” if the maximum instrument reading measured by the methods specified in §60.485(a) after each repair attempt is equal to or greater than 10,000 ppm.

(5) “Repair delayed” and the reason for the delay if a leak is not repaired within 15 calendar days after discovery of the leak.

(6) The signature of the owner or operator (or designate) whose decision it was that repair could not be effected without a process shutdown.

(7) The expected date of successful repair of the leak if a leak is not repaired within 15 days.



(8) Dates of process unit shutdowns that occur while the equipment is unrepaired.

(9) The date of successful repair of the leak.

(d) The following information pertaining to the design requirements for closed vent systems and control devices described in §60.482–10 shall be recorded and kept in a readily accessible location:

(1) Detailed schematics, design specifications, and piping and instrumentation diagrams.

(2) The dates and descriptions of any changes in the design specifications.

(3) A description of the parameter or parameters monitored, as required in §60.482–10(e), to ensure that control devices are operated and maintained in conformance with their design and an explanation of why that parameter (or parameters) was selected for the monitoring.

(4) Periods when the closed vent systems and control devices required in §§60.482–2, 60.482–3, 60.482–4, and 60.482–5 are not operated as designed, including periods when a flare pilot light does not have a flame.

(5) Dates of startups and shutdowns of the closed vent systems and control devices required in §§60.482–2, 60.482–3, 60.482–4, and 60.482–5.

(e) The following information pertaining to all equipment subject to the requirements in §§60.482–1 to 60.482–10 shall be recorded in a log that is kept in a readily accessible location:

(1) A list of identification numbers for equipment subject to the requirements of this subpart.

(2) (i) A list of identification numbers for equipment that are designated for no detectable emissions under the provisions of §§60.482–2(e), 60.482–3(i) and 60.482–7(f).

(ii) The designation of equipment as subject to the requirements of §60.482–2(e), §60.482–3(i), or §60.482–7(f) shall be signed by the Company.

(3) A list of equipment identification numbers for pressure relief devices required to comply with §60.482–4.



(4) (i) The dates of each compliance test as required in §§60.482–2(e), 60.482–3(i), 60.482–

4, and 60.482–7(f).

(ii) The background level measured during each compliance test.

(iii) The maximum instrument reading measured at the equipment during each compliance test.

(5) A list of identification numbers for equipment in vacuum service.

(f) The following information pertaining to all valves subject to the requirements of §60.482–7(g) and (h) and to all pumps subject to the requirements of §60.482–2(g) shall be recorded in a log that is kept in a readily accessible location:

(1) A list of identification numbers for valves and pumps that are designated as unsafe-to- monitor, an explanation for each valve or pump stating why the valve or pump is unsafe- to-monitor, and the plan for monitoring each valve or pump.

(2) A list of identification numbers for valves that are designated as difficult-to-monitor, an explanation for each valve stating why the valve is difficult-to-monitor, and the schedule for monitoring each valve.

(g) The following information shall be recorded for valves complying with §60.483–2:

(1) A schedule of monitoring.

(2) The percent of valves found leaking during each monitoring period.

(h) The following information shall be recorded in a log that is kept in a readily accessible location:

(1) Design criterion required in §§60.482–2(d)(5) and 60.482–3(e)(2) and explanation of the design criterion; and

(2) Any changes to this criterion and the reasons for the changes.

(i) The following information shall be recorded in a log that is kept in a readily accessible location for use in determining exemptions as provided in §60.480(d):



(1) An analysis demonstrating the design capacity of the affected facility,

(2) A statement listing the feed or raw materials and products from the affected facilities and an analysis demonstrating whether these chemicals are heavy liquids or beverage alcohol, and

(3) An analysis demonstrating that equipment is not in VOC service.

(j) Information and data used to demonstrate that a piece of equipment is not in VOC service shall be recorded in a log that is kept in a readily accessible location.

(k) The provisions of §60.7 (b) and (d) do not apply to affected facilities subject to this subpart.

Reporting requirements:

The following reporting requirements are applicable:

(a) The Company shall submit semiannual reports to the Department beginning six months after the initial startup date.

(b) The initial semiannual report shall include the following information:

(1) Process unit identification.

(2) Number of valves subject to the requirements of §60.482–7, excluding those valves designated for no detectable emissions under the provisions of §60.482–7(f).

(3) Number of pumps subject to the requirements of §60.482–2, excluding those pumps designated for no detectable emissions under the provisions of §60.482–2(e) and those pumps complying with §60.482–2(f).

(4) Number of compressors subject to the requirements of §60.482–3, excluding those compressors designated for no detectable emissions under the provisions of §60.482–3(i) and those compressors complying with §60.482–3(h).

(c) All semiannual reports to the Department shall include the following information, summarized from the information in §60.486:

(1) Process unit identification.



(2) For each month during the semiannual reporting period,

(i) Number of valves for which leaks were detected as described in §60.482(7)(b) or §60.483–2,

(ii) Number of valves for which leaks were not repaired as required in §60.482– 7(d)(1),

(iii)Number of pumps for which leaks were detected as described in §60.482–2(b) and (d)(6)(i),

(iv) Number of pumps for which leaks were not repaired as required in §60.482– 2(c)(1) and (d)(6)(ii),

(v) Number of compressors for which leaks were detected as described in §60.482– 3(f),

(vi) Number of compressors for which leaks were not repaired as required in §60.482– 3(g)(1), and

(vii)The facts that explain each delay of repair and, where appropriate, why a process unit shutdown was technically infeasible.

(3) Dates of process unit shutdowns which occurred within the semiannual reporting period.

(4) Revisions to items reported according to paragraph (b) if changes have occurred since the initial report or subsequent revisions to the initial report.

(d) If the Company elects to comply with the provisions of §§60.483–1 or 60.483–2 it shall notify the Department of the alternative standard selected 90 days before implementing either of the provisions.

(e) The Company shall report the results of all performance tests in accordance with §60.8 of the General Provisions. The provisions of §60.8(d) do not apply to affected facilities subject to the provisions of this subpart except that an owner or operator must notify the Department of the schedule for the initial performance tests at least 30 days before the initial performance tests.



(f) The requirements of paragraphs (a) through (c) of this section remain in force until and

unless EPA, in delegating enforcement authority to a State under section 111(c) of the Act, approves reporting requirements or an alternative means of compliance surveillance adopted by such State. In that event, affected sources within the State will be relieved of the obligation to comply with the requirements of paragraphs (a) through (c) of this section, provided that they comply with the requirements established by the State.

40 CFR 63, Subpart CC – NESHAP for Petroleum Refineries

This subpart applies to petroleum refining process units that are located at a plant site that is a major source as defined in section 112(a) of the Clean Air Act; and emit or have equipment containing or contacting one or more of the hazardous air pollutants listed in table 1 of this subpart. The refinery is a major source and the crude unit emits HAPs listed in table 1 of this subpart. Therefore, this is an applicable requirement.

Standards: The Company is subject to the provisions of this subpart shall comply with the requirements of §§60.482–1 to 60.482–10 as soon as practicable, but no later than 180 days after initial startup.


Compliance with this regulation is identical to the requirements under 40 CFR 60, subpart VV discussed above. Therefore, no additional discussion is presented Consent Decree LDAR Requirements: Paragraphs 114 and 115 of the federal CD establish an internal leak definition of 2000 ppm for pumps and 500 ppm for valves in light liquid and gaseous service. These leak thresholds are more restrictive than the standards in 40 CFR 60, subpart VV. Therefore, these are applicable requirements.


Because the leak detection threshold for pumps and valves in light liquid and gaseous service are more stringent than the levels in 40 CFR 60, subpart VV (2,000 ppm (CD )for pumps versus 10,000 ppm (NSPS) and 500 ppm (CD) for valves versus 10,000 ppm (NSPS)), the compliance methodology of the CD is applicable. This requires pumps to be monitored monthly and for repairs to be carried out according to the procedures in Section “F”, paragraphs 116 through 118 of the CD



Monitoring/Testing: The company shall comply with the monitoring requirements in paragraphs 119 through 121 of the CD Recordkeeping and Reporting: The Company shall comply with the recordkeeping and reporting requirements of paragraph 133 of the CD. Regulation No. 39 – Nitrogen Oxides Budget Trading Program: Section 2 of Regulation 39 states: a. Each NOX Budget unit shall hold in its compliance account and/or its overdraft account, as of the NOX allowance transfer deadline of each control period, a quantity of NOX allowances available for deduction that is equal to or greater than the total NOX emissions from that NOX Budget unit for that control period. b. Each NOX Budget unit shall be subject to the requirements of Section 2(a) of this regulation starting on the later of May 1, 2003 or the date the unit commences operation.

The FCCU COB has a rated heat input of 679 mmBtu/hour. Therefore, it is an affected unit under the NOx Budget Trading Program.

Compliance Determination Methodology for Regulation No. 3926: Compliance is based on operating a certified CEMS and the recordkeeping and reporting requirements. Monitoring & Testing: 1. The Company and, to the extent applicable, the NOx authorized account representative of

each NOx Budget source and each NOX Budget unit at the source shall comply with the

26 The requirements of Regulation 39 are unaffected by the PCUP. The discussion on the applicability of this regulation is included for completeness. Since all the Regulation 39 requirements are already included in the NOx Budget Permit, these requirements are not being repeated in this permitting action.



monitoring requirements of Regulation No. 39 § 8.

1. The emissions measurements recorded and reported in accordance with Regulation No. 39 § 8 shall be used to determine compliance by the unit with the NOx Budget emissions limitation under paragraph (c).

Record Keeping and Reporting:

Unless otherwise provided, the owners and operators of the NOx Budget source and each NOx

Budget unit at the source shall keep on site at the source each of the following documents for a period of 5 years from the date the document is created. This period may be extended for cause, at any time prior to the end of 5 years, in writing by the permitting authority or the Administrator.

a. The account certificate of representation under Regulation No. 39 § 6 and all documents

that demonstrate the truth of the statements in the account certificate of representation; provided that the certificate and documents shall be retained on site at the source beyond such 5 year period until such documents are superseded because of the submission of a new account certificate of representation under Regulation No. 39 § 6 changing the NOx authorized account representative.

b. All emissions monitoring information, in accordance with Regulation No. 39 § 8; provided that to the extent that Regulation No. 39 § 5 provides for a 3 year period for record keeping, the 3 year period shall apply.

c. Copies of all reports, compliance certifications, and other submissions and all records made or required under the NOx Budget Trading Program.

d. Copies of all documents used to complete a NOx Budget permit application and any other submission under the NOx Budget Trading Program or to demonstrate compliance with the requirements of the NOx Budget Trading Program.

e. Records demonstrating that any unit exempted under Regulation No. 39 § 3(b) of this regulation is retired. The owner(s) or operator(s) of that unit bears the burden of proof that the unit is retired.

f. The NOx authorized account representative of a NOx Budget source and each NOx Budget unit at the source shall submit the reports and compliance certifications required under the NOx Budget Trading Program, including those under Regulation No. 39 § 7, 8, and 11.

g. Each document submitted to the Department and the Administrator pursuant to this permit shall be signed and certified by the Authorized Account Representative and shall contain the following language: “I am authorized to make this submission on behalf of the owners and operators of the NOx Budget sources or NOx Budget units for which the submission is made. I certify



under penalty of law that I have personally examined, and am familiar with, the statements and information submitted in this document and all its attachments. Based on my inquiry of those individuals with primary responsibility for obtaining the information, I certify that the statements and information are to the best of my knowledge and belief true, accurate, and complete. I am aware that there are significant penalties for submitting false statements and information or omitting required statements and information, including the possibility of fine or imprisonment.”

40 CFR part 63 Subpart UUU –- NESHAP From Petroleum Refineries – Catalytic Cracking Units, Catalytic Reforming Units and Sulfur Plants This regulation is applicable to the FCCU. Premcor’s compliance strategy with respect to the FCCU’s particulate HAP emissions is to route a slurry recycle stream containing these HAPS to the FCU for control at that unit which is presently equipped with an electrostatic precipitator. Therefore, AQM included an appropriate emission limitation in the FCU WGS permit during the phase I review of the PCUP27. AQM is reaffirming as a general condition in this permit that Premcor will have to comply with all the applicable requirements of this regulation. Coke Storage and Handling System (Unit 22): Regulation No. 2 - Permits: Section 2.1 (c) of this Regulation states: Except as exempted in Section 2.2, no person shall initiate construction, install, alter or initiate operation of any equipment or facility or air contaminant control device which will emit or prevent the emission of an air contaminant prior to receiving approval of his application from the Department or, if eligible, prior to submitting to the Department a completed registration form for equipment, a facility or an air contaminant control device that is not subject to Section 2.1(a) or 2.1(b), the person shall submit to the Department an application for a permit pursuant to Section 11of this regulation. Section 11.6 of this Regulation states: No permit shall be issued by the Department unless the applicant shows to the satisfaction of the Department that the equipment, facility, or air contaminant control device is designed to operate or is operating without causing a violation of the State Implementation Plan, or any rule or regulation of the Department, and without interfering with the attainment or maintenance of

27 Condition 2.1.11 of Permit: APC-81/0829-CONSTRUCTION (Amendment 5) dated November 30, 2004.



National and State ambient air quality standards, and without endangering the health, safety, and welfare of the people of the State of Delaware. The Department may, from time to time, issue or accept criteria for the guidance of applicants indicating the technical specifications which it deems will comply with the performance standards referenced herein. Furthermore, Section 11.8 states: The following emission rates and/or standards for each air contaminant emitted from any equipment, facility or air contaminant control device shall be specified in each permit issued pursuant to this regulation: a. The rate and/or standard established and/or relied upon in the State Implementation Plan

(SIP) to include the State of Delaware “Regulations Governing the Control of Air Pollution” and regulations promulgated pursuant to Section 111 and Section 112 of the Clean Air Act (CAA); and

b. The rate that was shown under Section 11.6 as not interfering with the attainment and maintenance of any National and State ambient air quality standard, and not endangering the health, safety, and welfare of the people of the State of Delaware; or

c. The rate requested by the applicant. In no case shall this rate be greater than the potential to emit of the equipment, facility, or air contaminant control device; and in no case shall this rate be less stringent than the rate specified in Section 11.8(a) and (b) of this regulation.

This regulation is applicable to the coke storage and handling system because it involves the modification of existing equipment and the construction and construction of new equipment that will emit more than ten (10) pounds per day of Total Suspended Particulate matter (TSP) and particulate matter with a mean aerodynamic diameter of less than 10 micrometers (PM10). As a result of the Pollution Control Upgrade Project (PCUP), the amount of petroleum coke that will be produced by the coker will increase from 1880 tpd to 2500 tpd. Coke exits the coker and is transferred via a pneumatic chute into the coke storage silo. From here about 285 tons per day (3 railcars per day at 95 tons per car) are loaded into railcars for shipment. The rest will be transferred out of the silo via the zigzag blender to either the storage warehouse for eventual shipment offsite or to the power plant where the coke is used as a fuel in the two gasifiers less frequently (during periods when shut down, reclaim coke from storage will be transferred to the gasifiers, if they are to be operated). The next page shows a flow chart of the coke handling system. The unlabeled transfer belt from Tower #4 to the coke storage warehouse is the C1-belt. The diagram does not show the D-belt coming from Tower #4 used for emergency stacking (discussed later in this memo).





Transfer Tower #2 Tower #2 is the transfer point where coke is dropped from A-belt onto B-belt28. Premcor has proposed enclosing this transfer point within the tower and installing a baghouse to control fugitive emissions expected at transfer points. This transfer point has an emission factor 0.003 lb/ton based upon Table 11.19.2-229 from AP-42 (August 2004 edition). Premcor will operate at a rate of 2,500 tons per day (tpd) and estimates the baghouse control efficiency to be 70%. The TSP emission rate is determined as follows:

( )2500 0 003 10 0 7 3652000

0 41tpdlbton

dayyr

tonlbs

tonyr

× × − × × =. . . . TSP

The PM10 fraction from coke is 74% of the total coke (by weight). The PM10 emission rate is determined as follows:

0 41 0 74 0 30. . .tonyr

TSPtonyr

× = PM10

Transfer Tower #3 The coke being moved (2500 tpd) on B-belt enters Tower #3 and is dropped on either C-belt where it moves on towards the storage warehouse or is dropped onto F-belt and moved to the power plant (estimated to be 1200 tons per day). Similar to Tower #2, Premcor will enclose the transfer point and install a baghouse with an estimated collection efficiency of 70%. Using the same calculations for Tower #2, TSP and PM10 emissions are calculate to be 0.41 ton and 0.30 ton, respectively. Transfer Tower #4 Tower #4 is the transfer point where coke is dropped from C-belt onto D-belt. Even though Premcor estimates 1200 tons of coke per day will be conveyed to the power plant, the conveyor system to the storage warehouse is sized to handle 2500 tpd in the event the gasifiers are offline for prolonged periods of time. In the application, Premcor calculated the emissions from this point based on a throughput of 2200 tpd when in fact 1300 tpd will normally be conveyed into

28 There is no Tower #1. 29 From Chapter 11.19.2 “Crushed Stone Processing and Pulverized Mineral Processing” for a conveyor transfer point



the storage warehouse. Emissions have been estimated using the potential of 2500 tpd. Using the same calculations for Tower #2 and #3, TSP and PM10 emissions are anticipated to be 0.41 ton and 0.30 ton, respectively. AQM disagrees with Premcor’s estimate of 70% control efficiency for the transfer tower baghouses. Typically, baghouses are designed to control particulate matter greater than 3 µm in diameter at efficiencies greater than 99%. At this rate, TSP and PM10 emissions can be reduced to 27 pounds per year and 20 pounds per year, respectively. The company has stated the low collection efficiency is a result of the low inlet grain loading rates. AQM recommends conducting stack tests to ascertain the actual performance of these control devices. The permit requires stack testing all five baghouses and the warehouse ventilation filters during operation. AQM reserves the right to establish appropriate emission limits after the test results have been reviewed. Rail Car Loading Rail car loading operations occur adjacent to the existing coke storage silo. Dry coke is offloaded directly through a loading arm into the rail cars for transport offsite. The loading chute is equipped with a retractable, enclosed vehicle loading spout. Coke dust generated during the loading operation and displaced by the coke filling the car is collected by a vacuum drawn on the annular space around the loading spout and conveyed to an existing dust collector. To control fugitive emissions from the loading operation, AQM is requiring the following operating practices be followed:

1. During rail car loading, all car hatches shall remain closed except for the hatch through which the car is being loaded; and

2. The coke loading chute shall be operated with a dust recovery mechanism that completely covers the hatch’s annular space.

With each rail car able to hold 95 tons of coke, Premcor estimates loading 20 cars each week for a total of 98,800 tons of coke each year. The Company has based coke emissions upon an emission rate of 0.61 lb of TSP emitted per ton of coke loaded from AP-42 Table 11.17-430 (February 1998 edition). It is expected that a good portion of the entrained particulates from the loading operation will settle out in the closed car. Fugitive coke within the closed railcar will be pulled through the loading arm vacuum into the baghouse.

30 From Chapter 11.17 “Lime Manufacturing” for loading into an enclosed vehicle



Premcor has conservatively estimated that all entrained particulates will be directed to the baghouse. The Company estimates its control efficiency at 76.5% but gives no basis for this estimate. The controlled TSP and PM10 emission rates are determined as follows:

( )1040 95 0 61 10 0 7652000

71carsyr

tonscar

lbton

tonlbs

tonsyr

TSP× × × − × =. . . .

71 0 74 52 10. . .tonsyr

TSPtonsyr

PM× =

The permit requires stack testing the baghouse emissions during operation. AQM reserves the right to establish appropriate emission limits after the test results have been reviewed. Coker Silo Baghouse The initial permit issued for the coke handling system was Permit: APC-82/1209-OPERATION dated January 5, 1983 which included this baghouse. This application did not address this device but AQM will incorporate it into this permit and cancel the old permit after the new operation permit is issued. Using the same emission factor that was used for the railcar loading and a control efficiency of 99%, AQM estimates the TSP and PM10 emissions to be:

( )2500 0 61 10 0 99 3652000

2 78tpdlbton

dayyr

tonslbs

tonyr

TSP× × − × × =. . . .

2 78 0 74 2 06 10. . .tonsyr

TSPtonsyr

PM× =

As stated above, AQM is requiring conducting stack tests to ascertain the actual performance of the 5 baghouses. AQM reserves the right to establish appropriate emission limits after the test results have been reviewed. Conveyor Belts To improve reliability of the system, Premcor will be installing new conveyor belts that can accommodate the high temperature coke. The new belts do not allow the use of oil to wet the coke and control fugitive emissions. Therefore, instead of adding oil at the zigzag blender, water



will be added to the coke in the new storage warehouse. The conveyors are covered on the top and two sides. The company will be installing primary and secondary scrappers to scrape the conveyors clean of coke. This will reduce the amount coke carried back on the underside where it is likely to fall to the ground. Premcor estimates the covers and scrappers will reduce the potential for emissions by 79%. The coke rate to the power plant has been calculated using the belt’s potential of 2500 tpd instead of the expected rate of 1200 tpd. The Company has used the emission factor of 0.003 lb/ton from AP-42 but does not give the exact citation. AQM has requested the basis for this emission factor but it has not been supplied by Premcor. Therefore, AQM can not verify if this emission factor is appropriate. The application failed to estimate the emissions from the G-belt section of the conveyor system. Using the same emission factor Premcor used, the controlled TSP and PM10 emission rates are determined as follows:

( )2500 0 003 10 0 79 3652000

0 29tpdlbton

dayyr

tonlbs

tonyr

TSP× × − × × =. . . .

0 29 0 74 0 21 10. . .tonyr

TSPtonyr

PM× =

Premcor estimates TSP and PM10 emissions for the entire conveyor system are 1.74 tons and 1.29 tons, respectively. Other Drop Points Reclaim: Reclaim is the process of reclaiming coke from the storage area and transferring it back to the gasifiers. The most likely reason to do so would be in the event the coker was not operational. Reclaim coke is removed from the storage area by a front end loader and dumped into a below-grade hopper, on to E-belt, then F-belt and G-belt into the power plant’s storage silos. While reclaim operations are not a daily event, the Company assumed an amount of 245 tons per day would accurately represent the amount used annually. Potential emissions from E-belt have not been addressed because the belt runs counter-current to C-belt. When E-belt is moving 245 tpd to the power plant, the coker is not operational and thus delivering 2500 tpd coke to the storage warehouse.



However, there will be emissions when the front end loader is dumping the coke into the hopper. Premcor has estimated these emissions using a control efficiency of 70%. Since the loading is outside with no obvious dust control measures in place, AQM does not agree with this. AQM has used the same emission factor that Premcor uses to estimate the emissions during truck loading operations but does not account for any control efficiency.

245 0 0957 3652000

4 28tpdlbton

daysyr

tonlbs

tonyr

TSP× × × =. .

4 28 0 74 317 10. . .tonsyr

TSPtonsyr

PM× =

F-belt to G-belt: The coke rate to the power plant has been calculated using its potential of 2500 tpd instead of the expected rate of 1200 tpd. The Company has used the emission factor of 0.0049 lb/ton derived from equation #1 from Chapter 13.2.4 in AP-42 (January 1995 edition). This transfer point is not fully enclosed with a baghouse controlling emissions. Premcor has not explained their reason for using a control efficiency of 70% at this point. Premcor has also not explained why the emission factor for this conveyor belt drop point is different from that used for the transfer towers. AQM has estimated the emissions with no control efficiency as a worst case scenario:

2500 0 0049 3652000

2 24tpdlbton

dayyr

tonlbs

tonsyr

TSP× × × =. .

2 24 0 74 165 10. . .tonyr

TSPtonyr

PM× =

G-belt to Storage Silos: Coke is dumped from the G-belt into one of 4 coke storage silos at the power plant. The coke enters through a closed chute. The Company has used the emission factor of 0.0049 lb/ton derived from equation #1 from Chapter 13.2.4 in AP-42 (January 1995 edition). This transfer point is not fully enclosed with a baghouse controlling emissions. Premcor has also not explained their reason for using a control efficiency of 70% at this point. Premcor has also not stated why the emission factor for this conveyor belt drop point is different from that used for the transfer towers. AQM has estimated the emissions with no control efficiency as a worst case scenario:



2500 0 0049 3652000

2 24tpdlbton

dayyr

tonlbs

tonsyr

TSP× × × =. .

2 24 0 74 165 10. . .tonsyr

TSPtonsyr

PM× =

Emergency Stacking: The existing system transfers the coke from C-belt into Tower #4 and onto D-belt where it then is emptied down a chute to ground level and moved into the coke storage area by bulldozer. The Company will leave this section of the process in place in the event an emergency stacking area is needed. This process has been a source of great fugitive emissions in the past. With the existing system, the refinery estimates a control efficiency of 70% because the coke is oiled. The proposed system will be processing dry coke (0.33% moisture). Thus, expected emission will be much greater. The Company has not submitted the measures that will be taken to control emissions nor have the probable emissions been estimated. Coke Storage Warehouse Activities To reduce the potential for fugitive coke emissions, Premcor has proposed building a storage warehouse approximately 125 feet wide, 400 feet long, and 55 feet high. The building is designed to hold a maximum amount of 21,900 tons of coke, but will store approximately 10,000 tons under normal conditions. Premcor estimates the storage warehouse will control fugitive coke emissions at a rate of 70%. The coke will be delivered into the warehouse on D-belt and emptied into one of two pugmills. The pugmills will mix water into the coke to reduce its temperature and to reduce its potential for fugitive emissions. The wet coke exits the bottom of the pugmill and is dropped onto a conveyor belt system. Emissions from dropping the coke into the pugmill are expected to be similar to those above for coke on G-belt to the storage silos; 0.41 tpy TSP and 0.30 tpy PM10. Identical emissions are estimated for the coke leaving the pugmill. The coke is then transferred inside the building over one, two or three conveyor belts depending where the Company wants to pile the material. Again, the expected emissions for a conveyor drop point at 0.41 tpy TSP and 0.30 tpy PM10. If conveyed over three belts, emissions are estimated to be 0.82 tpy TSP and 0.60 tpy PM10.



The Company has estimated the conveyor belt emissions to be 0.41 tpy TSP and 0.30 tpy PM10. The Company has used the emission factor of 0.003 lb/ton from AP-42 but does not give the exact citation. Therefore, AQM can not verify if this rate is appropriate. The Company has estimated emissions when the conveyor belt drops coke onto the coke pile. They are based on equation #1 from AP-42 chapter 13.2.4 and use the building’s estimated control efficiency of 70%. Emissions are estimated to be 0.056 tpy TSP and 0.041 tpy PM10. The majority of the emissions come from loading the coke into dump trucks for transport off-site. Based on an emission factor of 0.0957 lb of TSP per ton loaded from AP-42 Table 11.9-131, these emissions are estimated at 13.0 tpy TSP and 9.7 tpy PM10. On December 4, 2002, Secretary Hughes issued Notice of Conciliation and Secretary’s Order No. 2002-A-0063 to the Delaware City Refinery “in an endeavor to obtain compliance” with the ongoing TSP ambient air quality violations. The Order required the refinery to “submit a plan to totally enclose the coke storage and handling operation.” With a control efficiency of only 70%, AQM does not believe Premcor’s proposal for the storage warehouse meets the Order’s goals. In order to comply with the Secretary’s Order, AQM believes the warehouse must be maintained with a negative pressure and the warehouse’s air handling system must have an adequate pollution control device such as a baghouse. The Company shall submit the proposed control measures for AQM’s approval within thirty (30) days after issuance of the construction permit. Berm Removal To build the storage warehouse, the 25-foot high berm that surrounds the storage pile on 3 sides will have to be removed. The warehouse will be constructed over this area. The Company has not estimated emissions expected from the removal of the berm and other land clearing and grading operations. Permit requirements for controlling of fugitive emissions is detailed under Regulation No. 6 below. The permit requires Premcor to submit a detailed plan for the berm removal and warehouse construction to AQM at least 30 days before the project begins. The plan must include measures that will be taken for controlling emissions and estimated emissions. Emissions Evaluation The Company was issued a construction permit on May 3, 2002 with emission limits of 46 tpy TSP and 32 tpy PM10. Premcor’s application states past actual emissions to 40.26 tpy TSP and 31 From Chapter 11.9 for “Equations for Uncontrolled Open Dust Sources at Western Surface Coal Mines” for truck loading. The coke moisture content is 8%.



29.04 tpy PM10 (the time period was not given). The operation’s potential to emit will decrease with the addition of the three baghouses and the coke storage warehouse. Future potential emissions are estimated to be 25.44 tpy TSP and 18.82 tpy PM10. Premcor estimates the coke handling and storage system will see net reductions of over 14 tons per year of TSP and 10 tons per year of PM10. However, their calculations did not account for the removal of the berm, the existing coker silo baghouse, the warehouse ventilation system, or emergency stacking outside the warehouse. Regulation No. 3 – Ambient Air Quality Standards: This regulation is applicable to all affected emissions units in the refinery. Therefore, its applicability is covered under the heading “Facility Wide Applicable Requirements” in this memorandum. Regulation No. 5 – Particulate Emissions from Industrial Sources Section 2.1 of this regulation states: “No person shall cause or allow particulate emissions into the atmosphere form any source not provided for in subsequent sections of this Regulation in excess of 0.2 grains per standard cubic foot.” All five baghouses and the filters for the warehouse ventilation system shall not exceed this limit. Compliance Methodology: The Company shall install, maintain and use baghouses or filters at Transfer Towers 2, 3, and 4, the rail car loading operation, the coke storage silo, and the warehouse ventilation exhaust. Compliance will be achieved through the testing and monitoring requirements. Monitoring & Testing Requirements: The Company shall demonstrate compliance with this regulation through the following requirements:

(a) Within ninety (90) days after achieving the maximum production rate at which the facility will be operated, but not later than 180 days after initial startup of such facility, the owner or operator shall conduct performance test(s) and furnish the Department with a written report of the results of such performance test(s) in accordance with the following general provisions:



1. One (1) original and two (2) copies of the test protocol shall be submitted a minimum of forty-five (45) days in advance of the tentative test. The tests shall be conducted in accordance with the State of Delaware and Federal requirements.

2. The test protocol shall be approved by the Department prior to initiating any testing. Upon approval of the test protocol, the Company shall schedule the compliance demonstration with the Air Surveillance Branch. The Department must observe the test for the results to be considered for acceptance.

3. The final results of the testing shall be submitted to the Department within sixty (60) days of the test completion.

4. The final report of the results shall be submitted in a format approved by the Air Surveillance Branch, and signed by a corporate official, or his designee, whose signature shall constitute his own, and employer’s certification of compliance, clearly indicating each applicable term and condition of the permit, and whether the test(s) fulfilled the permit condition. The results must demonstrate to the Department’s satisfaction that the emission unit is operating in compliance with the applicable regulations and conditions of this permit; if the final report of the test results shows non-compliance the owner or operator shall propose corrective action(s). Failure to demonstrate compliance through the test may result in enforcement action.

(b) A continuous monitoring device, such as one that monitors the differential pressure

across the control device, shall be installed and maintained to ensure proper operation at all times.

Recordkeeping and Reporting Requirements: The following recordkeeping and reporting requirements are applicable: Plant personnel shall record the differential pressure once every shift while the conveyor system is operating.

Regulation No. 6 – Particulate Emissions from Construction & Materials Handling Section 2.1 of this regulation states: “No person shall cause or allow the demolition of existing structures, buildings, or parts of buildings, in New Castle County or in incorporated areas of Kent and Sussex Counties unless methods are employed to control dust emissions.”



Section 2.2 states: “Such methods may include the application of water or the use of other techniques approved by the Department.” Section 3.1 of this regulation states: “No person shall cause or allow land clearing, land grading (including grading for roads), excavation, or the use of non-paved roads on private property unless methods, as indicated in Section 2.2, are employed to control dust emissions, when the Department determines that such activities could emit dust in quantities sufficient to cause air pollution.” Compliance Methodology: As discussed above, the permit requires Premcor to submit a detailed plan for removing the berm, constructing the storage warehouse, and storing and handling the coke during this period and methods that will be employed to control fugitive emissions. Reporting Requirements: To ensure there is a more timely response to possible exceedances of the secondary Delaware TSP 24-hour AAQS (Regulation No. 3), AQM is requiring the reporting frequency of monitoring results be increased from monthly to weekly until the construction is completed. Section 4 states: “No person shall cause or allow visible particulate emissions of any material being transported by a motor vehicle.” Section 6 states: “No person shall cause or allow stockpiling or other storage of material or transport to or from a storage facility in such a manner as may cause a condition of air pollution.” Compliance Methodology: The Department is concerned about coke emissions from the warehouse as front end loaders load coke into the trucks and as those trucks drive out of the warehouse. The permit requires dust control measures such as control measures to minimize emissions from the openings in the storage warehouse and trucks shall only be loaded inside the warehouse. The permit also requires the coke moisture to be maintained at 8%, trucks must be washed to remove coke and coke dust, and all surfaces where coke routinely accumulates and all road and



vehicle movement areas must be paved. The Company shall use as street sweeper or other approved method to clean the roadways and paved areas. The Company shall propose a cleaning frequency for AQM’s approval within thirty (30) days after the permit is issued. To minimize coke dust during railcar loading operations, all hatches on the railcar shall remain closed except for the hatch that is being loaded. Also, the coke loading chute shall be operated with a dust recovery system that vents to the baghouse. The loading chute must completely cover the hatch’s annular space. Premcor shall propose measures to control coke dust when loading the reclaim hopper within thirty days of issuance of this permit. Regulation No. 14 – Visible Emissions: Section 2.1 of this regulation states: No person shall cause or allow the emission of visible air contaminants and/or smoke from a stationary or mobile source, the shade or appearance of which is greater than twenty (20%) percent opacity for an aggregate of more than three (3) minutes in any one (1) hour or more than fifteen (15) minutes in any twenty-four (24) hour period. Compliance Methodology: The Company shall conduct qualitative observations of the handling system each shift when the system is in operation using Reference Method 22 in Appendix “A” of 40 CFR part 60. In the event visible emissions are observed by reference Method 22, the Company shall conduct a visible emissions evaluation using the procedures described in Regulation 20, section 1.5 (c). At all other times, compliance shall be based on the proper operation of the emissions unit and record keeping. The Company shall conduct a daily quantitative observation of each baghouse in accordance to the Monitoring/Testing requirement below. Monitoring/Testing: Conduct visual observations at fifteen-second intervals for a period of not less than one hour except that the observations may be discontinued whenever a violation of the standard is recorded. The additional procedures, qualification and testing used for visually determining the opacity shall be those specified in Section 2 and 3 (except for Section 2.5 and the second



sentence of Section 2.4) of reference Method 9 set forth in Appendix A, 40 CFR Part 60, revised July 1, 1982. Recordkeeping: Observation records showing the presence or absence of emissions shall be maintained on site. Records of all maintenance performed on applicable units shall be maintained. Reporting: The Company shall submit quarterly reports indicating the duration and magnitude of all periods of excess opacity. Regulation No. 17 – Source Monitoring, Recordkeeping and Reporting: Section 2.1 of this regulation states: Upon written request of the Department, an owner or operator of an air contaminant source shall, at his expense, install, maintain, and use emission monitoring devices, keep records, and make periodic reports to the Department on the nature and amount of emissions from such source. The Department shall make such data available to the public as reported and as correlated with any applicable emission standards or limitations. Section 2.2 of this regulation states: Upon written request of the Department, an owner or operator of an air contaminant source shall, at his expense, sample the emissions of, or fuel used by, that source, maintain records and submit reports to the Department on the results of such sampling. The Department may make such data available to the public as reported and as correlated with any applicable emission standards or limitations. Section 2.2 of this regulation states: The Department may conduct tests of emissions from or fuel used by any air contaminant source. Upon written request of the Department, the owner or operator of the air contaminant source shall provide necessary holes in stacks or ducts, and such other safe and proper sampling and testing facilities, exclusive of instruments and sampling devices, if any are necessary, for proper determinations of the emission of air contaminants. The Department shall have access to and use of monitoring, record-keeping and reporting required by Federal Regulations relating to emissions of air contaminants. The Department may make such data available to the public as reported or received and as correlated with any applicable emissions standards or limitations.



Compliance Methodology: The Company shall continue daily TSP monitoring as required by Permit: APC-82/1209-CONSTRUCTION (Amendment 3) dated May 3, 2002. Monitoring/Testing, Recordkeeping and Reporting: The Company shall submit the monthly TSP monitoring results within the first 10 days of each month. However, during the period of the berm being removed and the warehouse is being constructed, AQM is requiring increasing the reporting frequency. Weekly monitoring results shall be submitted within 7 days so that there can be a more timely response to any exceedance of the secondary Delaware TSP 24-hour AAQS (Regulation No. 3). Regulation No. 19– Control of Odorous Air Contaminants This regulation is applicable to all affected emissions units in the refinery. Therefore, its applicability is covered under the heading “Facility Wide Applicable Requirements” in this memorandum. Regulation No. 30– Title V State Operating Permit Program This regulation is applicable to all affected emissions units in the refinery where construction activity will take place. Therefore, its applicability is covered under the heading “Facility Wide Applicable Requirements” in this memorandum. Facility Wide Requirements: Regulation No. 3: Section 2.1 of this regulation states: No person shall cause the Air Quality Standards specified in this Regulation to be exceeded. Compliance Method:32

32 See also discussion under applicability of Regulation 3 to atypical operations of the FCU and for the ongoing violations of the Delaware Secondary 24 hour Ambient air Quality Standard for Total Suspended Particulate Matter.



Premcor has conducted a modeling analysis for normal operation using the EPA approved Industrial Source Complex Short Term (ISCST3 version 02035) model and 5 years of representative meteorological data to show no adverse impacts to ambient air quality standards. Regulation No. 14: The Company shall not cause or allow the emission of visible air contaminants and/or smoke from any emission unit, the shade or appearance of which is greater than twenty (20) percent opacity for an aggregate of more than three (3) minutes in any one (1) hour or more than fifteen (15) minutes in any twenty-four (24) hour period. Compliance Method: Compliance with the emission standard of this condition shall be demonstrated in accordance with Subsection 1.5(c) of Regulation No. 20 and the recordkeeping requirements of this condition Monitoring/Testing: In accordance with Regulation No. 20 Section 1.5, conduct visual observations at fifteen second intervals for a period of not less than one hour except that the observations may be discontinued whenever a violation of the standard is recorded. The additional procedures, qualification and testing to be used for visually determining the opacity shall be those specified in Section 2 and 3 (except for Section 2.5 and the second sentence of Section 2.4) of reference Method 9 set forth in Appendix A, 40 CFR Part 60 revised July 1, 1982.

• The Company shall conduct weekly qualitative plant-wide stack observations to determine the presence of any visible emissions.

• If visible emissions are observed, the Company shall take corrective actions and/or determine compliance by conducting a visible observation in accordance with the above paragraph.

If no visible emissions are observed or are within permitted limits, no further action is required. Record Keeping: Observation records shall be maintained on site. Reporting: The Company shall report all exceedances of the standard quarterly



Regulation No. 19: The Company shall not cause or allow the emission of an odorous air contaminant such as to cause a condition of air pollution. Compliance Method: Compliance with the emission standard of this condition shall be demonstrated in accordance with the monitoring/testing and record keeping requirements of this condition. Monitoring/Testing: Includes but is not limited to scentometer tests, air quality monitoring, and affidavits from affected citizens and investigators. Recordkeeping: Records of all monitoring/testing shall be maintained on site. Reporting: The Company shall provide reasonable notice to the Department of the possibility of odors being encountered offsite during process upsets. Regulation No. 24: Handling, Storage and Disposal of VOCs Work Practice Standards A. The Company shall not cause, allow, or permit the disposal of more than eleven (11) pounds of a Volatile Organic Compound (VOC), or of any materials containing more than eleven (11) pounds of any VOCs, in any one (1) day, in a manner that would permit the evaporation of VOC into the ambient air. This includes but is not limited to the disposal of VOC from any VOC control devices. This provision does not apply to:

• Any VOC or material containing VOC emitted from a regulated entity that is subject to a VOC standard under Regulation No. 24.

• Any VOC or material containing VOCs used during process maintenance turnarounds for cleaning purposes, provided that the provisions of paragraph (B), (C), and (D) of this condition are followed.

• Waste paint (sludge) handling systems, water treatment systems, and other similar operations at coating facilities using complying coatings.

B. No owner or operator of a facility subject to this regulation shall use open containers for the storage or disposal of cloth or paper impregnated with VOCs that are used for surface reparation, cleanup, or coating removal. Containers for the storage or disposal of cloth or



paper impregnated with VOCs shall be kept closed, except when adding or removing material. C. No owner or operator of a facility subject to this regulation shall store in open containers spent or fresh VOC to be used for surface preparation, cleanup or coating removal. Containers for the storage of spent or fresh VOCs shall be kept closed, except when adding or removing material. D. No owner or operator shall use VOC for the cleanup of spray equipment unless equipment is used to collect the cleaning compounds and to minimize their evaporation to the atmosphere. Compliance Method: Compliance shall be demonstrated by adherence with the VOC handling work practices by providing appropriate training and posting of instructions, and record keeping for storage, use and disposal of VOCs. Monitoring/Testing: The Company shall monitor employee training records on an annual basis and update records as needed. Recordkeeping: The Company shall keep a record of postings, and employee training related to these work practice standards and handling, storage, and disposal of VOCs. Reporting/Certification: The Company shall indicate the compliance status with this regulation in the form of a certification statement in each quarterly report. PEF:CRR:TSD:BAS:klb F:\EngAndCompliance\CRR\05007CRR.doc Pc: Dover File

Documents

MEMORANDUM - dnrec.state.de.us II... · of the crude unit atmospheric heater and ... The cracking takes place when fresh feed ... Agglofilter modules remove particulate matter and