Download Air Pollution Source Test: Emission Rates of Asphalt Pollutants in Wilmington, CA and more Slides Engineering in PDF only on Docsity!
AP42 Section: 11.
Reference Number: 46
Title: Report Of Air Pollution Source Testing For Selected Air Toxics At Industrial Asphalt, Wilmington, California, Engineering-Science, Inc., Itwindale, CA,
August 5,1992.
... ..^.
REPORT OF AIR POLLUTION SOURCE TESTING FOR SELECTED AIR TOXICS AT INDUSTRIAL ASPHALT) WILMINGTON, CALIFORNIA
Conducted On:
April 30 - May 4, 1992
Submitted To: INDUSTRIAL ASPHALT 3200 San Fernando Road
Los Angeles, California 90065
And NATIONAL ASPHALT PAVEMENT ASSOCIATION 5 100 Forbes Boulevard NAPA Building Lanham, Maryland 20706-
Submitted On: August 5, 1992
Submitted By: ENGINEERING-SCIENCE. INC. 6060 N. Invindale Ave, Skte J P.O. Box 2007 Invindale, California 91706
I
'0 SECTIONS RESULTS Introduction Table 1: Table 2: Table 3: Table 4: Table 5: Table 6: Table 7:
Table 8:
Table 9: Table 10: Table 11: Table 12:
Table 13:
Summary of EPA 5 , Particulate Matter Summary of EPA 202, Particulate Matter and Condensibles Summary of CARB 429, PAH (Actual Excess Air, mg/dscm) Summary of CARB 429, PAH (100% Excess Air, mgldscm) Summary of CARB 429, PAH (Actual Excess Air, ppm, vlv) Summary of CARB 429, PAH (100% Excess Air, ppm, v/v) Summary of CARB 429, PAH (Actual Excess Air, lb/hr) Summary of CARB 429, PAH (100% Excess Air, lb/hr) Summary of EPA 001 1, Formaldehyde Summary of EPA 18, Methane, BTEX (Actual Excess Air) Summary of EPA 18, Methane, BTEX (100% Excess Air) Summary of EPA 25A(THC), 6C(S@), 7E(NOx), lO(CO), and (Actual Excess Air) Summary of EPA 25A(THC), 6C(SO2), 7E(NOx), lO(CO), and (100% Excess Air)
LIST OF APPENDICES
APPENDIX A EPA METHOD 5
Particulate Matter
Particulate Matter and condensibles MODIFIED CARB METHOD 429 Polycyclic Aromatic Hydrocarbons & Cresols
APPENDIX B EPA METHOD 202
APPENDIX C
APPENDIX D EPA METHOD 18
Methane & BTEX
APPENDIX E EPA METHOD 0011 Formaldehyde CEM DATA REDUCTION, STRIP CHARTS, AND QUALITY ASSURANCE Total Hydrocarbons Oxygen, Carbon Dioxide Sulfur Dioxide Nitrogen Oxides Carbon Monoxide APPENDIX G PROCESS DATA
APPENDIX F
SECTION 1
INTRODUCTION
a
REPORT OF AIR POLLUTION SOURCE TESTING FOR
SELECTED AIR TOXICS AT INDUSTRIAL ASPHALT,
WILMINGTON, CALIFORNIA
SECTION 1
INTRODUCTION
During the period April 30 through May 4, 1992, Engineering-Science, Inc., (ES), of
Invindale, California, conducted air pollution source testing to quantify the mass emission rates of selected air toxics from the baghouse serving the rotary dryer at the Industrial Asphalt (IA) facility i n Wilmington, California. The facility is a batch type asphalt plant and operates under Authority-to-Construct granted by the South Coast Air Quality Management District
(SCAQMD). Testing was conducted to determine air toxic emissions for use in calculating
emission factors for air toxic emissions inventories as required by recent amendments to Title 111 of the Clean Air Act. The testing program was conducted at the request of the National Asphalt Paving Association (NAPA), and will be used as "pooled source test" data.
A total of three test runs for each pollutant was conducted. Testing was performed to determine mass emission rates of particulate matter (PM), condensible particulate matter (Condensible PM), polycyclic aromatic hydrocarbons (PAH), cresols, formaldehyde (HCHO), and volatile organic compounds (VOC) including methane, benzene, toluenes, ethyl benzene, and xylenes (BTEX). In addition, continuous emissions monitoring (CEM) was conducted to determine concentrations of total hydrocarbons as methane, sulfur dioxide ( S 0 2 ) , nitrogen oxides (NO2), carbon monoxide (CO), oxygen (02), and carbon dioxide (CO2).
The testing program was coordinated by Messrs. Dwight Beavers and Terry Prentice of IA. The ES testing team was comprised of Messrs. Rollie Rosario, Mike Edwards, Dwight Wieman, Cory Holt, Cesario Mangaoang, and Rob Pearman. Mr. Tom Brumagin of NAPA assisted in the coordination and planning of the testing program. Also present as observers were Messrs. Don England and Steve Falzarano of ES. IA control room personnel monitored and documented plant operating conditions and throughput.
SECTION 2
PROCESS DESCRIPTION
EMISSION SOURCE INFORMATION
e
a
e
SECTION 2
PROCESS DESCRIPTION AND
EMISSION SOURCE INFORMATION
2.0 Process Description
Plant operations at Industrial Asphalt, Wilmington, California involve the drying, sizing, and combining of aggregate with hot asphalt to produce asphalt concrete for use in paving. The batch plant is supplied from aggregate storage bins maintained on-site and heated asphalt delivered by tank transport. The primary source of emissions at the plant are from the rotary aggregate dryer which is vented, via a cyclone and baghouse, to the stack. The final asphalt product is temporarily stored, and then transferred to dump trucks.
2.1 Emission Source Description
The baghouse stack discharge is approximately 25' above grade, with an internal diameter of 33 inches. The sampling ports are located about 45" (> 1 diameter upstream) from the exit to the atmosphere and about 20' (5 diameters downstream) from the nearest flow disturbance. There were only two sampling ports available, and were oriented 90" to each other in the same cross-sectional plane. 12 points for each port were used for a 24 point sampling traverse i n accordance with EPA Method 1, "Sample and Velocity Traverses for Stationary Sources". A diagram of the sampling port locations is presented as Figure 2.1-1. A small 1/2" porthole was drilled six inches below the sampling port to accommodate the CEM sample probe.
SECTION 3
METHODOLOGY
e
SECTION 3
METHODOLOGY
3.0 Introduction The testing and analytical procedures used for the various parameters are based on U. S. Environmental Protection Agency (EPA), and California Air Resources Board (CARB) approved and published methods. The EPA methodology was used primarily to serve the client requirement for a "national pooled source test".
3. 1 Test Methods
Location and Number of Traverse Points (EPA Method 1 )
The procedures specified by EPA Method 1, "Sample and Velocity Traverses for Stationary Sources", were followed to determine the number and location of the traverse points to be used for sampling the source. The number of straight run stack diameters (equivalent diameters) upstream and downstream from the sample ports were measured and used to determine the exact number of traverse points required.
Parallel, or non-cyclonic, gas stream flow in the stack was verified using an S-type Pitot tube connected to an inclined oil manometer. The manometer has 0.01 inch
gradations on the inclined scale and 0.10 inch gradations on the vertical scale. The Pitot
tube was rotated so the planes of the face openings were perpendicular to the stack cross-
sectional plane. This is referred to as the 0 degree reference position. A zero manometer reading obtained i n this position indicates no cyclonic flow. If the manometer did not read zero, the Pitot tube was rotated until a zero reading was obtained. The angle of rotation was measured to the nearest degree. All traverse points were examined in this fashion. Since the average of the absolute values of the rotation angles were less than 20 degrees, the sampling location was deemed acceptable.
Velocitv and TemDerature Traverse (EPA METHOD 21
The procedures delineated by EPA Method 2, "Determination of Stack Gas Velocity and Volumetric Flow Rate (S-Type Pitot Tube)", were followed to determine the stack gas velocity and volumetric flow rate. A preliminary velocity and temperature 3- 1
The wet impingement sampling train was comprised of a quartz-glass nozzle and a
heated probe connected to four chilled Greenburg-Smith design impingers followed by a
vacuum pump, and a calibrated dry gas meter. The nozzle had a sharp, tapered leading
edge. The angle of taper was < 30 on the outer side to preserve a constant internal diameter. The nozzle was of the button hook design, permanently and uniquely identified, then inspected and calibrated before use.
0
The probe liner was made of quartz-glass and was equipped with a heating system capable of maintaining the temperature of the extracted gas i n the range of 248 k25"F at the probe exit. An S-type Pitot tube and a type K thermocouple were attached to the probe to constantly monitor stack gas velocity and temperature as described in EPA Method 2.
The filter was fiberglass, without organic binder, and exhibiting a collection efficiency of 99.95% (<0.05% penetration) for 0.3-micron dioctyl phthalate. The filter was visually inspected against light for flaws and pinhole leaks, numbered, and tared to constant weight prior to testing. The filter holder was constructed of glass, with a glass frit filter support and was installed in a heated box at the exit end of the sampling probe.
0 The four impingers were connected in series with leak free glass U-tubes. The first, third, and fourth impingers were modified Greenburg-Smith design with the tip replaced with 112 inch ID glass tube extending about 112 inch from the flask bottom. The first two impingers each contained 100 mls of distilled, deionized water. The third impinger was empty. The fourth impinger contained approximately 400 grams of indicating silica gel.
A leak check of the entire train assembly was conducted prior to and after each test
run. The train was operated in an ice bath i n an effort to maintain an exit gas temperature
of 68°F or less from the last impinger. The train was operated isokinetically with a
sampling time of 3 minutes per traverse point for a total sample time of 72 minutes and a minimum volume of at least 45 dry standard cubic feet.
e
The train recovery was conducted in a clean and wind-free environment to minimize contamination of, or loss of sample. The probe and filter holder were allowed to cool so they could be safely handled. When the assembly was cool, the nozzle and probe tip were capped off. HPLC grade acetone with a predetermined value of 50.001 % residue was used as a probe wash solution. A portion of the acetone taken into the field
3-
e
a
was saved and analyzed as an acetone blank. The liquid gain in the collection impingers was determined volumetrically and the weight gain in the silica gel impinger was determined gravimetrically. The total moisture gain was used to calculate the moisture content of the stack gas.
The recovered fractions were reported as follows: Front Half comprised of the filter catch, and probe catch (nozzle included); and Back Half comprised of the impinger catch, and the solvent extract ( methylene chloride extraction of impinger water). The sample fractions, as applicable, were stored in leak free polyethylene bottles (water) and borosilicate bottles (solvents) with Teflon lined screw caps.
The analysis consisted of particulate weight determination by drydown of each fraction. The respective acetone rinses were dried down at ambient conditions. The water fractions were extracted with methylene chloride. After extraction, the aqueous fraction was dried down at 105 k2"C while the organic fraction was dried down at ambient conditions. The dried fractions were desiccated and weighed to a constant weight. The total weight of all the fractions less the acetone blank and solvent extract weight was reported as total particulate.
Condensible PM (EPA Draft Method 202)
The emission rate of condensible particulate matter was determined using EPA Draft Method 202, "Determination of Condensible Emissions from Stationary Sources", which is very similar to EPA Method 5. The major method modification was an additional
impinger charged with 100 ml of deionized water and a gaseous nitrogen purge of the
impinger train and final collected contents. The post-test nitrogen purge was set at a rate
of 20 Ipm for one hour. Analysis of the organic and inorganic fractions of the impinger
contents allowed the determination of emissions which condense at temperatures below filter temperature.
The filter and probe catches (Front Half) were determined according to EPA 5 procedures. The particulate matter i n the impinger catch (Back Half) was also analyzed. The back half catch was included as particulate matter. The total back half weight was the
sum of the weight of the impinger catch (organic and inorganic) and the weight of the back
half acetone rinse. The impinger liquid was extracted with 75 ml portions of methylene chloride. The organic and aqueous phases were fully separated and drained into separate
3-
The sampling train consisted of a glass nozzle, a heated quartz lined probe, a heated Teflon coated glass fiber filter, Teflon sample line, glass condenser, XAD-2 sorbent module, and the impinger train. Heated components were maintained at 248 +25 O F. The condenser is designed to provide sufficient cooling of the sample gas to maintain a gas exit temperature of no more than 68 OF. The impinger train consisted of four impingers in series; the first two each contained 100 mls of deionized, distilled water, the third impinger was dry, and the fourth contained approximately 400 grams of indicating silica gel. A thermocouple was located at the outlet of the silica gel impinger to measure the temperature of the exiting gases. Sealant grease was not used in any portion of the sampling train. Train components were covered with hexane rinsed foil before and after each test run.
All recovery solvents were nanograde quality, distilled-in-glass, and stored in their original containers. The sample recovery involved at least three sequential brushings and rinsings with acetone and methylene chloride until all train components were visually clean.
The analytical method is isotope dilution mass spectrometry combined with high resolution gas chromatography. The analysis included the addition of internal standards in known quantities. A matrix specific extraction was performed, followed by clean-up and analysis of the extract for PAH using high-resolution capillary column gas chromatography coupled with high resolution mass spectrometry (HRGUHRMS).
The analysis was performed by ALTA Laboratories of El Dorado Hills, California. The PAH and cresol compounds for analysis included: POLYCYCLIC AROMATIC HYDROCARBONS & CRESOLS
Acenaphthene Acenaph thylene Anthracene Benz-a-Anthracene Dibenzo-a,h-Anthracene Chrysene Fluoranthene Benzo-b-Fluoranthene 2-Methylphenol 4-Methylphenol
Benzo-k-Fluoranthene Fluorene Naphthalene Phenanthrene Pyrene Benzo-a-Pyrene Indeno- 1,2,3-cd-Pyrene Benzo-ghi-Perylene 3-Methylphenol
Formaldehvde Sampline (EPA Method 001 1)
The formaldehyde emission rate was determined using the procedures delineated in EPA Method 001 1, "Sampling for Aldehyde and Ketone Emissions from Stationary Sources". The gas sample was extracted isokinetically from the source and collected in 2,4-dinitrophenyl-hydrazine (DNPH) solution. Full size (1000 ml) impingers were used and contained 100 ml each of DNPH in the first and second impingers. A minimum volume of 45 DSCF was collected. Three test runs were conducted.
The method is based on the formation of the formaldehyde dinitrophenylhydrazone
derivative. The derivative is extracted, solvent-exchanged, concentrated, and then analyzed by high performance liquid chromatography (HPLC). The samples were analyzed for formaldehyde content as specified by EPA Method 001 1. The samples underwent a solvent extraction with a 70/30 mix of hexane and methylene chloride. A gradient HPLC system complete with column supplies, mobile phase reservoir, high pressure pump and syringes, and all required accessories, including an injection valve, and a compatible data recording system was used to measure peak areas and retention time. The analysis was performed by Atmospheric Assessments Association (AtmAA), of Chatsworth. California.
Methane. Benzene. Toluene. Ethvlbenzene and Xylene (BTEX) (EPA Method 18)
Sampling for benzene, toluene, ethylbenzene and xylene was conducted in accordance with EPA Method 18, "Measurement of Gaseous Organic Compound Emissions by Gas Chromatography". Samples were collected over a period of thirty minutes each. Each sample was collected into an evacuated ten liter Tedlar bag. The Tedlar bag was cleaned and leak checked prior to and after the testing and analysis. The Tedlar bag was equipped with a minimum length 1 / 8 " ID Teflon tube and a 118" ID stainless steel probe. The bag was mounted in a rigid leak free vessel. Sample was drawn into the bag by withdrawing the air from the vessel containing the Tedlar bag according to the procedures of the method.
The samples were subjected to gas chromatographic (GC) analysis using an appropriate column and detector. Gas chromatography with electron capture detection was used. The chromatographic columns were suitable for the separation of methane, benzene,
3-
