Characterization of Air Emissions from Open Burning at the Radford Army Ammunition Plant Johanna Aurell1, Brian Gullett2* 1 University of Dayton Research Institute 2 U.S. EPA, Office of Research and Development Research Triangle Park, North Carolina June 20, 2017 i Acknowledgment Technical support was provided by Bill Mitchell, Dale Greenwell, and Dennis Tabor (EPA/ORD). Flight operations and range safety were handled by Ved Chirayath and David Satterfield (NASA Ames). Abstract 5 10 The Department of the Army (PD Joint Services, Picatinny Arsenal) commissioned NASA-Ames to fly their unmanned aerial vehicle (UAV), a hexacopter, into the plumes from open burning of propellant and manufacturing discards at the Radford Army Ammunition Plant while carrying a gas and particle sensor system designed and operated by the EPA Office of Research and Development (ORD). Over a 2-week period the NASA/ORD team sampled 33 plumes, determining emissions factors for particulate matter, metals, chloride, perchlorate, volatile organic compounds, chlorinated dioxins/furans, and nitrogen-based organics. Results show agreement with published emission factors and good reproducibility (e.g., 11% relative standard deviation for PM2.5). The UAS/sampler presents a significant advance in emission characterization capabilities for open area sources, safely and effectively making measurements heretofore deemed too hazardous for personnel or beyond the reach of land-based samplers. ii Table of Contents 1. Introduction........................................................................................................1 15 2 20 1.1 Brief .................................................................................................................................. 1 1.2 Objective .......................................................................................................................... 1 Materials and Methods .....................................................................................1 2.1 Test Site Location and Description .................................................................................. 1 2.2 Test Ordnance and Test Schedule .................................................................................... 2 2.2.1 MK-90 ....................................................................................................................... 3 2.2.2 Skid Waste ................................................................................................................ 3 2.3 25 2.3.1 Target Analytes and Collected Target Analytes ....................................................... 4 2.3.2 Unmanned Aerial Vehicle Based Sampling Method ................................................ 5 2.4 30 3 40 Emission Sampling and Analytical Methods ................................................................... 7 2.4.1 CO2............................................................................................................................ 7 2.4.2 CO ............................................................................................................................. 8 2.4.3 PM and Elements ...................................................................................................... 8 2.4.4 Chromium(VI) .......................................................................................................... 9 2.4.5 VOCs......................................................................................................................... 9 2.4.6 Energetics ................................................................................................................ 10 2.4.7 HCl, Perchlorate, and Chlorate ............................................................................... 11 2.4.8 PCDD/PCDF ........................................................................................................... 11 2.5 35 Testing Procedures ........................................................................................................... 4 Calculations .................................................................................................................... 12 2.5.1 Converting from mass/mass Carbon to mass/mass initial source ........................... 12 2.5.2 PCDD/PCDF Toxic Equivalent Calculations ......................................................... 12 2.5.3 Data Variability....................................................................................................... 13 Results and Discussion ....................................................................................15 3.1 PM .................................................................................................................................. 15 3.2 Elements/Metals ............................................................................................................. 15 3.2.1 Elements/Metals ...................................................................................................... 15 I 3.2.2 45 Chromium(VI) ........................................................................................................ 18 3.3 HCl, chlorate, and perchlorate........................................................................................ 19 3.4 PCDD/PCDF .................................................................................................................. 20 3.5 VOCs .............................................................................................................................. 21 3.6 Energetics ....................................................................................................................... 29 4 Conclusions .......................................................................................................30 5 References .........................................................................................................31 50 II List of Figures Figure 2-1. Overhead View of RFAAP Burn Pan Site. .................................................................. 2 Figure 2-2. Composition of the two types of skid wastes tested, type 1 (left, total mass 3,254 lbs.) and type 2 (right, total mass 1,589 lbs.). ......................................................................................... 3 55 Figure 2-3. UAV-Based Sampling Method .................................................................................... 5 Figure 2-4. NASA’s UAV. ............................................................................................................. 6 Figure 2-5. Kolibri Instrumentation, Oden and Balder in foreground and Tor and Loke in background. ..................................................................................................................................... 7 60 Figure 3-1. Comparison of PCDD/PCDF (Dioxin) emission factors from a) skid waste and forest burns [20], and b) emission factor derived from this study (EF) and emission factor used today by RFAAP (RFAAP EF). ............................................................................................................. 21 List of Tables Table 2-1. Test schedule, amount of total pan load and amount of waste burned per test day. ..... 2 65 Table 2-2. Constituents in each burn pan of “MK-90” burns. ........................................................ 3 Table 2-3. Skid waste carbon and metal fraction. ........................................................................... 3 Table 2-4. Target Analytes. ............................................................................................................ 4 Table 2-5. Collected Target Analytes from MK-90 and Skid Waste. ............................................ 5 Table 2-6. Sampling Instrumentation used during each test day. ................................................... 7 70 Table 2-7. Elements determined using XRF. .................................................................................. 8 Table 2-8. VOCs analyzed from Carbotrap 300 ........................................................................... 10 Table 2-9. The 2005 World Health Organization PCDD/PCDF Toxic Equivalent Factors for mammals/humans.[17] .................................................................................................................. 13 Table 3-1. PM2.5 emission factors in g/kg initial source and lb/lb initial source. ......................... 15 75 Table 3-2. Element emission factors in PM2.5 fraction in mg/kg initial source and mg/kg waste.a ....................................................................................................................................................... 16 Table 3-3. Metal emission factors in PM2.5 fraction in lb/lb initial source and lb/lb waste.a ........ 17 Table 3-4. Comparison of EFs derived in this project with EFs used by RFAAP’s HHRA. ....... 18 Table 3-5. Cr(VI) emission factors. .............................................................................................. 18 80 Table 3-6. HCl, chlorate, and perchlorate emission factors from skid waste type 1. ................... 19 Table 3-7. PCDD/PCDF results. ................................................................................................... 20 III Table 3-8. VOC Emission Factors in lb/lb waste from skid waste type 2. ................................... 21 Table 3-9. VOC Emission Factors in mg/kg waste from skid waste type 2. ................................ 24 Table 3-10. VOC Emission Factors in lb/lb initial source from skid waste type 2. ..................... 25 85 Table 3-11. VOC Emission Factors in mg/kg initial source. ........................................................ 27 Table 3-12. Energetics based on method detection limit. ............................................................. 30 List of Appendices Appendix A: Element emission factors 90 Appendix B: PCDD/PCDF emission factors List of Acronyms CO CO2 Cr(VI) DOD DQI EF EPA FOD GC GPS HCl HMX HPLC IC ICP LC LRGC LRMS MCE MK-90 NASA NC NDIR NG Carbon monoxide Carbon dioxide Chromium VI U.S. Department of Defense Data Quality Indicator Emission Factor U. S. Environmental Protection Agency Foreign object debris Gas chromatography Global positioning system Hydrogen chloride Octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine High-Performance Liquid Chromatography Ion chromatography Inductively coupled plasma Liquid chromatography Low resolution gas chromatography Low resolution mass spectrometer Mixed cellulose ester MK-90 rocket motors National Aeronautics and Space Administration Nitrocellulose Non-dispersive infrared Nitroglycerine IV NIST NO NO2 NRE OB/ OD OBG OD ORD PM2.5 PUF QA QAPP RDX RFAAP RPD SD SDS SIM SRM SVOC UAV UDRI USB VOCs XRF National Institute for Standards and Technology Nitrogen oxide Nitrogen dioxide New river energetics Open burning/Open detonation Open burning ground Outer diameter Office of Research and Development Particulate matter equal to and less than 2.5 µm Polyurethane foam Quality Assurance Quality assurance project plan Research Department Formula X, 1,3,5-Trinitroperhydro-1,3,5-triazine Radford Army Ammunition Plant Relative percent difference Secure digital Safety Data Sheets Selective ion monitoring Standard reference material Semivolatile organic compounds Unmanned aerial vehicle University of Dayton Research Institute Universal serial bus Volatile organic compounds x-ray fluorescence spectrometry V 1. 95 100 105 110 115 Introduction 1.1 Brief The Radford Army Ammunition Plant (RFAAP) conducts on-site disposal of a variety of hazardous energetic wastes via open burn pans located at the facility's open burning ground (OBG). Data on potential combustion emissions and their emission factors are available only from small laboratory and pilot scale simulations and their relevance to the RFAAP’s scenario is uncertain. To resolve this issue, the RFAAP asked the U.S. Environmental Protection Agency’s (EPA) Office of Research and Development (ORD) to perform direct sampling and quantification of the RFAAP's OBG emissions. ORD has considerable experience sampling emissions from open burning and open detonation (OB/OD) of military ordnance and static firing of rocket motors (for example, see Aurell et al. [1]). Since 2010, ORD has worked with the Department of Defense’s (DoD’s) Joint Munitions Command (and their predecessor, the Defense Ammunition Center), the Army Corps of Engineers, and the Defence Research and Development Canada -Valcartier to sample OB/OD emissions at three sites in the US and Canada. ORD has developed a suite of technologies for sampling an array of OB/OD emission constituents from both aerial and ground-based sampling platforms. These sampling methods have been developed over the last five years and include novel methods employing small sensors and samplers, necessitated by the challenge of sampling within a plume located several hundred feet in the open air. To transport ORD’s emission sensors/samplers into the plumes, RFAAP entered into an Interagency Agreement with the National Aeronautics and Space Agency, Ames Research Center (NASA Ames) for them to pilot their hexacopter unmanned aerial vehicle (UAV). 1.2 Objective 120 125 The objective of this work was to characterize and quantify emissions from open burning of dry propellant burns (MK-90 rocket motors) and so-called “skid burns”, which are a combination of process wastes from onsite production operations. This skid waste is generally a combination of energetic material, soil, gravel, and other foreign object debris (FOD). Skid burns are what the facility refers to as "assisted burns," where the materials are placed on wooden skids, and nested with dunnage and diesel fuel to promote burning. Quantification of the emissions includes determination of emission factors relating the amount of compound emitted to the amount present in the original material. 2 Materials and Methods 2.1 Test Site Location and Description The sampling was conducted at the Radford Army Ammunition Plant (RFAAP) in the mountains of southwest Virginia, approximately five miles northeast of the city of Radford, Virginia. 1 130 RFAAP lies along the New River in the relatively narrow northeastern corner of the valley. Approximate GPS coordinates are 37.1925 N, 80.5233 W. Figure 2-1 shows an overview of the RFAAP burn pan site. Figure 2-1. Overhead View of RFAAP Burn Pan Site. 135 2.2 Test Ordnance and Test Schedule Two fuel sources were sampled: dry propellant burns (MK-90) and skid burns (two types). The test schedule is shown in Table 2-1. The composition of these fuel sources, particularly metals, is critical toward assessing the environmental fate of the constituents. Knowledge of the carbon content of the fuel is required for determination of emission factors, as explained in 2.5.1, below. 140 Table 2-1. Test schedule, amount of total pan load and amount of waste burned per test day. Test Date Fuel 09/27/2016 09/28/2016 09/29/2016 09/30/2016 10/03/2016 10/04/2016 10/05/2016 10/06/2016 MK-90 Skid waste: Type 1 MK-90 Skid waste: Type 2 MK-90 Skid waste: Type 1 MK-90 Skid waste: Type 2 Amount of burn pans 5 3 5 2 5 3 5 2 Amount of Total pan load lb (kg) 3,000 (1,364) 3,254 (1,479) 3,000 (1,364) 1,589 (722) 3,000 (1,364) 3,254 (1,479) 3,000 (1,364) 1,589 (722) Amount of Total waste lb (kg) 3,000 (1,364) 1,620 (736) 3,000 (1,364) 500 (227) 3,000 (1,364) 1,620 (736) 3,000 (1,364) 500 (227) 2 2.2.1 MK-90 145 150 The test MK-90 composition was constant for all burn tests as shown in Table 2-2. Data were derived from two sources in order to complete the carbon composition. Safety Data Sheet (SDS) compositional data were used to supplement RFAAP laboratory analyzed composition data where components such as nitrocellulose were missing (see footnote “a” in Table 2-2). Each burn pan charge was comprised of 99% MK-90 and 1% NRE contaminated waste, by weight, as shown in Table 2-2. Table 2-2. Constituents in each burn pan of “MK-90” burns. REDACTED DUE TO TRADE SECRET PROTECTION 155 2.2.2 Skid Waste 160 Two different types of skid waste compositions were tested as shown in Figure 2-2. The main difference between the two skid waste types were the chlorine, lead, copper, and chrome fractions. Skid waste type 1 was designed to be a high chlorine burn and skid waste type 2 was a high metals burn. The majority of the carbon in the skid waste originated from the wood pallets (Table 2-3). Both skid waste types contained the same number of wood pallets, however, skid waste type 2 contained 26% more carbon than skid waste type 1 due to a higher mass fraction of pallets (less waste mass in type 2). Skid waste, type 1: 9/28/2016 and 10/04/2016 Cardboard, 0.28% Skid waste, type 2: 9/30/2016 and 10/06/2016 Pit #1, 4.3% Cardboard, 0.38% Pit #2, 13.0% Pit #4, 5.9% Pit #5, 11.8% Pit #3, 4.3% Pit #6, 11.8% Skids (pallets), 46.1% Grucci whistles, 4.3% Skids (pallets), 62.9% NRE Contaminated, 1.9% MCA-LAP Tracer slum, 13.0% NRE 1 filters, 2.8% Diesel, 3.8% NRE Contaminated, 7.1% 165 Diesel, 5.2% NRE tape, 0.92% Figure 2-2. Composition of the two types of skid wastes tested, type 1 (left, total mass 3,254 lbs.) and type 2 (right, total mass 1,589 lbs.). Table 2-3. Skid waste carbon and metal fraction. Waste type/ Test Dates Composition Carbon Fraction of each component Carbon fraction in burn pan 3 Waste type/ Test Dates Composition Skid waste Type 1 09/28/2016 and 10/04/2016 Pallets 46% Cardboard 0.28% Diesel 3.8% Pit #1 4.3% Pit #2 13% Pit #3 4.3% Grucci whistles 4.3% MCA-LAP Tracer slum 13% NRE 1 filters 2.8% NRE tape 0.92% NRE Contaminated 7.1% Total Carbon fraction Skid waste Type 2 09/30/2016 and 10/06/2016 Pallets 63% Cardboard 0.38% Diesel 5.2% Pit #4 5.9% Pit #5 11.8% Pit #6 11.8% NRE Contaminated 1.9% Total Carbon Fraction Carbon Fraction of each component 0.502a 0.46b 0.86b 0.017d 0.046d 0.41d 0.16d 0.0003d 0.013d 0 0.046d 0.502a 0.46b 0.86c 0.052d 0.038d 0.056d 0.046d Carbon fraction in burn pan 0.23 0.0013 0.033 0.00074 0.0059 0.0018 0 0.000043 0.00035 0.00016 0.0032 0.28 0.32 0.0017 0.045 0.0031 0.0045 0.0066 0.00086 0.38 a [2] [3] c Calculated using molecular formula C12H23 and density 0.832 kg/L. d Analytical measured data from BAE. b 170 2.3 Testing Procedures 2.3.1 Target Analytes and Collected Target Analytes 175 The target analytes are listed in Table 2-4. The full list of target VOCs are listed in Chapter 2.4.5. CO2 and CO were successfully measured continuously through all burns. The total number of target analyte samples collected for each type of waste are shown in Table 2-5. Table 2-4. Target Analytes. Analyte Instrument/Method Frequency CO2 CO PM2.5a Nitrocellulose Nitroaromatics PCDD/PCDF Non-dispersive infrared Electrochemical cell Impactor, Teflon filter Glass fiber filter Glass fiber filter Glass fiber filter and PUFb Continuous Continuous Batch Batch Batch Batch 4 Elements Cr(VI) HCl Perchlorate/chlorate VOCs Teflon filter from PM2.5 batch filter Bicarbonated-impregnated MCEc filter Na2CO3 coated quartz filter Quartz filter Carbotrap 300 Batch Batch Batch Batch Batch a 180 Fine particles in the ambient air with particles less than or equal to 2.5 μm in diameter. PUF – polyurethane foam plug. c MCE – mixed cellulose ester. b Table 2-5. Collected Target Analytes from MK-90 and Skid Waste. Analyte PM2.5 Nitrocellulose Nitroaromatics PCDD/PCDF Elements Cr(VI) HCl Perchlorate/chlorate VOCs MK-90 Skid waste Total 5 2 4 0 5 5 0 0 0 2 0 0 4 2 3 6 6 4 7 2 4 4 7 8 6 6 4 185 2.3.2 Unmanned Aerial Vehicle Based Sampling Method 190 Figure 2-3 shows the sampling instrumentation attached to the bottom of the UAV. This combined system was used for collecting air emissions from propellant plumes. Figure 2-3. UAV-Based Sampling Method 5 2.3.2.1 Unmanned Aerial Vehicle – UAV 195 200 Aerial sampling was conducted by a UAV operated by NASA Ames. NASA used a DJI Matrice M600 UAV (Figure 2-4). It is a 6-rotor hexacopter with a 9.1 kg weight and a 15.1 kg maximum acceptable gross take-off weight. Its maximum loaded flight time was approximately 13.5 min whereupon the remaining battery charge was 40%. The UAV can be controlled automatically or by pilot-in-command modes and provides the operator a GPS display screen of location in real time with a 2.4 GHz telemetry system. The M600 has an inertial measurement unit and GPS with a return to base function at a preset charge threshold. Figure 2-4. NASA’s UAV. 2.3.2.2 Kolibri – Sampling System 205 210 215 EPA/ORD’s sampling system called the “Kolibri” has been developed specifically for sample collection of plumes from open combustion sources. There are two configurations of the Kolibri primarily relating to the different sizes of the pumps needed for specific analytes. There are duplicate models of both Kolibris configurations for redundancy, referred to as “Oden” and “Balder” for the smaller unit and “Tor” and “Loke” for the larger unit (Figure 2-5). Because of payload limitations on the UAV, it was not possible to sample all of the target analytes with all of the pumps on a single platform. In addition, one pump has to be used for multiple analytes (PM2.5 or Total PM, Nitrocellulose or Nitroaromatics) and these can only be sampled separately. Hence, the full suite of analytes could only be collected using both Kolibris with sampler variations on each one (Table 2-6). In addition, energetics and VOCs required composite samples comprised of emission sampling from plumes of multiple burns. Because each of these samples has to be collected separately with composite samples, the number of repeat samples was limited. The Kolibri is capable of plotting real time CO2 and CO data, displaying sampling time and VOC sampling volume, while performing real time calculations to estimate the total amount of gaseous carbon sampled for the energetic sample. 220 6 CO2 and CO inlet Unit # 3 and 4 PM2.5 Unit # 1 and 2 VOC Energetics CO2 and CO inlet PM2.5 Figure 2-5. Kolibri Instrumentation, Oden and Balder in foreground and Tor and Loke in background. 225 Table 2-6. Sampling Instrumentation used during each test day. Test Date Ordnance Kolibri Unit Analytes Collected 09/27/2016 MK-90 Unit 4: Loke Nitroaromatics/PM2.5/Metals 09/29/2016 MK-90 Unit 4: Loke Nitrocellulose/Cr(VI) 10/03/2016 MK-90 Unit 4: Loke Nitroaromatics/Cr(VI) 10/05/2016 09/28/2016 10/04/2016 09/30/2016 MK-90 Unit 4: Loke Nitrocellulose/PM2.5 /Metals Skid waste Unit 4: Loke PCDD/PCDF/ HCl/Perchlorate/Chlorate Skid waste Unit 2: Balder VOCs/Cr(VI) 10/06/2016 Skid waste Unit 1: Oden VOCs/Cr(VI) 10/06/2016 Skid waste Unit 1: Oden VOCs/PM2.5/Metals 2.4 Emission Sampling and Analytical Methods 2.4.1 CO2 230 235 The system CO2 sensor (DX62210/DX6220 OEM Model, RMT Ltd, Moscow, Russia) measured CO2 concentration by means of non-dispersive infrared absorption (NDIR). The DX62210/DX6220 CO2 concentration was recorded on a standard secure digital (SD) card at a rate of one sample per second (1 Hz). The DX62210/DX6220 was calibrated for CO2 and checked for drift on a daily basis in accordance with EPA Method 3A [4]. The gas cylinders used for calibration were certified by the suppliers and traceable to National Institute of Standards and Technology (NIST) standards. A precision dilution calibrator Serinus Cal 2000 (American ECOTECH L.C., Warren, RI, USA) was used to dilute the high-level span gases for acquiring the mid-point concentrations for the DX62210/DX6220 calibration curves. The daily CO2 system drift for Unit 4 (Loke) varied from -4.6% to -0.4% of the full span and +1.0% for 7 240 Unit 2 (Balder), which is within the 5% acceptance limit of the sensor. Unit 1 (Oden) did not have a long enough warm up period before calibration therefore the drift of 7.9% was slightly outside acceptance limit, for this reason, the post-calibration curve was used for calculations as opposed to the pre-calibration curve. 2.4.2 CO 245 250 The CO sensor (e2V EC4-500-CO) was an electrochemical gas sensor (SGX Sensortech Ltd, High Wycombe, Buckinghamshire United Kingdom) which measured CO concentration by means of an electrochemical cell through CO oxidation and changing impedance. The sensor was calibrated for CO on a daily basis in accordance with U.S. EPA Method 3A[4]. The e2V CO concentration was recorded on a SD card at a rate of one sample per second (1 Hz). All gas cylinders used for calibration are certified by the suppliers and traceable to NIST standards. A precision dilution calibrator Serinus Cal 2000 (American ECOTECH L.C., Warren, RI, USA) was used to dilute the high-level span gases for acquiring the mid-point concentrations for the e2V EC4-500-CO calibration curves. The daily CO system drift for Unit 4 (Loke) varied from 8.4% to 2.8% and -1.2% for Unit 2 (Balder) and -4.5% for Unit 1 (Oden), which is within the 10% acceptance limit of the sensor. 2.4.3 PM and Elements 255 260 265 270 PM2.5 was sampled with SKC impactors (761-203B) using 37 mm tared Teflon filter (obtained from Chester LabNet) with a pore size of 2.0 µm via a constant micro air pump (C120CNSN, Sensidyne, LP, St. Petersburg, FL, USA) of 10 L/min. Total PM was sampled using cassette with a 37 mm tared Teflon filter (Chester LabNet) with a constant air pump (C120CNSN, Sensidyne, LP, St. Petersburg, FL, USA). PM were measured gravimetrically following the procedures described in 40 CFR Part 50 [5]. The constant flow pump was calibrated daily with a Gilibrator Air Flow Calibration System (Sensidyne LP, St. Petersburg, FL, USA). The plume samples PM2.5 concentrations were more than 100 times higher than the collected ambient air background sample. Elements were determined by x-ray fluorescence spectrometry (XRF) analysis of the Teflon PM2.5 and Total PM filters using EPA Compendium Method I0-3.3 [6]. The elements analyzed using XRF are stated in Table 2-7. Chester LabNet evaluated precision with a multi-element quality control standard (QS285) and accuracy using NIST standard reference materials (SRMs): SRM 1832, SRM 1833 and SRM 2783. The SRMs used for quality assurance/quality control (QA/QC) had a recovery of 91.9-108.6%, which is within the 80-120% acceptance criteria of the method. The plume samples’ element concentrations were at least 4 times higher than the ambient air background concentration. Table 2-7. Elements determined using XRF. 8 Elements Aluminum (Al) Antimony (Sb)* Arsenic (As)* Barium (Ba) Bromine (Br) Cadmium (Cd)* Calcium (Ca) Chlorine (Cl) Chromium (Cr)* Cobalt (Co)* * Copper (Cu) Gallium (Ga) Germanium (Ge) Indium (In) Iron (Fe) Lanthanum (La) Lead (Pb)* Magnesium (Mg) Manganese (Mn)* Mercury (Hg)* Molybdenum (Mo) Nickel (Ni)* Palladium (Pd) Phosphorus (P) Potassium (K) Rubidium (Rb) Selenium (Se)* Silicon (Si) Silver (Ag) Sodium (Na) Strontium (Sr) Sulfur (S) Tin (Sn) Titanium (Ti) Vanadium (V) Yttrium (Y) Zink (Zn) Zirconium (Zr) On U.S. EPA’s list of hazardous air pollutants [7]. 275 2.4.4 Chromium(VI) 280 Chromium(VI) (Cr(VI)) was sampled on a bicarbonate-impregnated “acid hardened” cellulose filter via a constant micro air pump (C120CNSN, Sensidyne, LP, St. Petersburg, FL, USA) of 10 L/min. Cr(VI) was determined using a proprietary method (ChesterLabNet, Tigard, OR) based on an EPA standard procedure [8]. The control sample had recoveries of 97.6 to 101.0% which is within the acceptance limits for the method 75-125%. No detectable levels of Cr(VI) were found in the ambient air background collected sample. 2.4.5 VOCs 285 290 295 VOCs was sampled using Carbotrap 300 stainless steel TD Tube (Supelco Inc., Bellefonte, PA, USA) via a constant micro air pump with an air flow rate of 0.185 L/min (3A120CNSN, Sensidyne, LP, St. Petersburg, FL, USA) in accordance with U.S. EPA Method TO-17 [9]. The Carbotrap 300 tubes were analyzed by ALS Simi Valley for VOCs by thermal desorption GC/MS according to U.S. EPA Method TO-17 [9]. The target VOCs analyzed from Carbopack 300 are stated in Table 2-8. The surrogate spikes used for the QA/QC had recoveries of 85-107% for all samples, which is within the accuracy of the method 70-140%. Eight (Trichlorofluoromethane, methylene chloride, carbon disulfide, trichloroethene, 1,1,2trichloroethane, toluene, 1,2-dibromoethane, bromoform) of sixty-one VOCs had recoveries slightly outside the acceptance limits for the laboratory control sample. The other 53 VOCs had recoveries of 99-118%, which is within the acceptance limit of the method 52-135%. The VOC method blank showed all non-detectable levels of VOCs except for carbon disulfide. The VOC trip blank showed detectable levels of ethanol, acetonitrile, and acetone. The VOC plume sample levels were 2-14, 22-53, and 4-35 times higher for ethanol, acetonitrile, and acetone, respectively, than the trip blank and ambient background levels. The VOC plume samples were corrected for the trip blank concentrations as well as corrected for ambient air background 9 300 concentrations. The constant flow pump was calibrated daily with a Gilibrator Air Flow Calibration System (Sensidyne LP, St. Peterburg, FL, USA). Table 2-8. VOCs analyzed from Carbotrap 300 VOCs 1,1,1-Trichloroethane* 1,1,2,2-Tetrachloroethane* 1,1,2-Trichloroethane* 1,1-Dichloroethane 1,1-Dichloroethene 1,2,4-Trichlorobenzene* 1,2,4-Trimethylbenzene 1,2-Dibromo-3-chloropropane 1,2-Dibromoethane 1,2-Dichloro-1,1,2,2-tetrafluoroethane (CFC 114) 1,2-Dichlorobenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5-Trimethylbenzene 1,3-Butadiene* 1,3-Dichlorobenzene 1,4-Dichlorobenzene* 1,4-Dioxane 2,2,4-Trimethylpentane* (Isooctane) 2-Butanone (MEK)* * 2-Hexanone 2-Propanol (Isopropyl Alcohol) 4-Methyl-2-pentanone Acetone Acetonitrile* Benzene* Bromodichloromethane Bromoform* Carbon Disulfide* Ethanol Ethylbenzene* Hexachlorobutadiene* m,p-Xylenes* Methyl tert-Butyl Ether Methylene Chloride* Naphthalene* n-Heptane n-Hexane Carbon Tetrachloride* n-Octane Chlorobenzene* Chloroethane Chloroform* Chloromethane* cis-1,2-Dichloroethene cis-1,3-Dichloropropene* Cumene* Cyclohexane Dibromochloromethane Dichlorodifluoromethane (CFC 12) o-Xylene* Styrene* Tetrachloroethene Tetrahydrofuran (THF) Toluene* trans-1,2-Dichloroethene trans-1,3-Dichloropropene* Trichloroethene Trichlorofluoromethane Trichlorotrifluoroethane Vinyl Chloride* On U.S. EPA’s list of hazardous air pollutants [7]. 305 2.4.6 Energetics 310 315 Nitroaromatics/Nitrocellulose were sampled using two 15 cm glass fiber filters (Fisher Scientific) with a nominal rate of 500 L/min. Energetics were sampled using a low voltage MINIjammer brushless blower (AMTEK, USA). The flow rate was measured by a 0-622 Pa Model 265 pressure differential transducer (Setra, USA) across a Herschel Standard Venturi tube (EPA in-house made). The Venturi tube is specially designed to meet the desired sampling rate for the target compound. The voltage equivalent to this pressure differential is recorded on the onboard Teensy USB microcontroller board, which was calibrated with a Roots meter (Model 5M, Dresser Measurement, USA) in the U.S. EPA metrology laboratory before sampling effort. The energetics samples were analyzed by an outside laboratory using analytical methods U.S. EPA Method 8330b [10] for nitroaromatics and the nitrocellulose by U.S EPA Method 353.2 [11] which is a nitrate-nitrite colorimetric method. The surrogate spikes used for the 10 320 325 330 335 nitroaromatics QA/QC had recoveries of 99.9-104% for all samples, which is within the accuracy of the method 70-130%. The laboratory control spike recoveries for nitroaromatics were between 99.5% and 100%, which is within the accuracy of the method 70-150%. The laboratory control spike recovery for nitrocellulose was 108%, which is within the accuracy of the method 40-120%. Nitroaromatics and nitrocellulose were not detected in the ambient air background sample. 2.4.7 HCl, Perchlorate, and Chlorate HCl was sample using an alkali-impregnated filter following a solid perchlorate and chloride filter (ISO Method 21438-2) [12]. The sampling was conducted at a flow rate of 2 L/min using a constant micro air pump (C120CNSN, Sensidyne, LP, St. Petersburg, FL, USA). The constant flow pump was calibrated daily with a Gilibrator Air Flow Calibration System (Sensidyne LP, St. Petersburg, FL, USA). Perchlorate salts were captured as a solid on the filter, which assumes no perchloric acid formation [13]. Samples were analyzed at ALS, NY. The alkali-impregnated filter was analyzed for HCl by ion chromatography methods specified in U.S. EPA Method 26 [14]. The laboratory control spike recovery for perchlorate and chlorate was 100% and 115%, respectively which is within the accuracy of the methods 40-120%. The laboratory control spike recovery for chloride was 107%, which is within the acceptance limit of the method 90-110%. Chlorate, perchlorate, or HCl were not detected in the ambient air background sample. 2.4.8 PCDD/PCDF 350 PCDD/PCDF were sampled as for energetics (see 2.4.6) but with the addition of a polyurethane foam plug (PUF) following the glass fiber filter. PCDD/PCDF samples were cleaned up and analyzed using an isotope dilution method based on U.S. EPA Method 23 [15]. Concentrations were determined using high resolution gas chromatography/high resolution mass spectrometry (HRGC/HRMS) with a Hewlett-Packard gas chromatograph 6890 Series coupled to a Micromass Premier mass spectrometer (Waters Corp., Milford, MA, USA). U.S. EPA Method 8290 [16] was used for analysis of tetra- through octa-CDDs/Fs. The laboratory control spike recoveries were within the acceptable 40-130% range for Tetra to Hexa PCDD/PCDF and 25-130% for Hepta to Octa PCDD/PCDF for most of the congeners. The HpCDF recovery was slightly outside the acceptance criteria for three of the four samples (13-23%), PentaCDD was outside the acceptance criteria in two of the four samples (155% and 178%). The collected plume samples had 10-250 and 700- >10,000 times higher levels of Total and TEQ PCDDs/PCDFs, respectively, than the collected ambient background sample. 355 The 2005 World Health Organization (WHO) toxic equivalent factors (TEFs) [17] were used to determine the PCDD/PCDF toxic equivalent (TEQ) emission factors (see Chapter 2.5.2 for calculations). Some of the seventeen TEF-weighted PCDD/PCDF congeners were undetected. The congeners that were not detected (ND) were considered as zero mass for the reported text calculations, however Appendix B shows both ND = 0 and ND = limit of detection mass value. 340 345 11 2.5 Calculations 2.5.1 Converting from mass/mass Carbon to mass/mass initial source 360 365 370 The emission ratio of each analyte/species of interest was calculated from the ratio of background-corrected pollutant concentrations to background-corrected carbon dioxide (CO2) and carbon monoxide (CO) concentrations. Emissions factors were calculated using these emissions ratios following the carbon balance method [18], and presented as mass pollutant per mass of charge weight. For the two skid waste types, the charge weight was expressed both as 1) the total initial weight of the waste plus the supplemental pallet and diesel fuel (“mass pollutant/mass initial source”) as well as 2) the weight of the RFAAP waste alone (“mass pollutant/mass waste”). For the MK-90s the charge weight was the total mass of initial MK-90 source material in the pan, resulting in emission factors expressed as “mass pollutant/mass initial source” which is the same meaning as “mass pollutant/mass waste” since no supplemental fuels were added to the waste, Equations 2-1 to 2-4. Emission factors determined here are compared with the emission factors used in the RFAAP Human Health Risk Assessment document, specifically Table 2-13 [citation?]. 𝐸𝐹𝑖 = 𝑓𝑐 × 375 where: EFi = fc = Analytei = ΣCj = 380 𝐴𝑛𝑙𝑦𝑡𝑒𝑖 ∑ 𝐶𝑗 Emission factor of target analyte i in terms of mass pollutant per mass initial source mass fraction of carbon in the initial source the mass emission ratio of species i, the background corrected mass concentration of carbon in major carbon emissions species j (carbon calculated from ΔCO2 and ΔCO). 𝐼𝑊 𝐸𝐹𝑊𝑎𝑠𝑡𝑒 = 𝐸𝐹𝑖 × 𝐼𝑊+𝑆𝐹 385 where: EFWaste = IW = SF = IW/(IW+SF) Equation 2-1 Equation 2-2 Emission factor of target analyte i in terms of mass pollutant per mass waste Initial weight of waste Supplement fuel (pallet, cardboard, and diesel) = 2.01 and 3.18 for skid waste type 1 and 2, respectively The majority of the carbon emissions were emitted as CO2 and CO. With this assumption, CO2 and CO are the only carbon-containing compounds that were required to be measured. 390 2.5.2 PCDD/PCDF Toxic Equivalent Calculations 12 PCDDs and PCDFs include 75 and 135 congeners, respectively. Of these 210 congeners 17 are toxic and have been assigned toxic equivalency factor (TEF) values (Table 2-9). The TEQ value is obtained by multiplying the concentration of a PCDD/PCDF congener by its TEF-value and summing the result for all 17 toxic congeners. 395 Table 2-9. The 2005 World Health Organization PCDD/PCDF Toxic Equivalent Factors for mammals/humans.[17] PCDDs TEF PCDFs TEF 2,3,7,8 - TCDD 1,2,3,7,8 - PeCDD 1,2,3,4,7,8 - HxCDD 1,2,3,6,7,8 - HxCDD 1,2,3,7,8,9 - HxCDD 1,2,3,4,6,7,8 - HpCDD 1,2,3,4,6,7,8,9 - OCDD 1 1 0.1 0.1 0.1 0.01 0.0003 2,3,7,8 - TCDF 1,2,3,7,8 - PeCDF 2,3,4,7,8 - PeCDF 1,2,3,4,7,8 - HxCDF 1,2,3,6,7,8 - HxCDF 1,2,3,7,8,9 - HxCDF 2,3,4,6,7,8 - HxCDF 1,2,3,4,6,7,8 - HpCDF 1,2,3,4,7,8,9 - HpCDF 1,2,3,4,6,7,8,9 - OCDF 0.1 0.03 0.3 0.1 0.1 0.1 0.1 0.01 0.01 0.0003 2.5.3 Data Variability 400 Standard deviation, as well as the relative standard deviation (RSD), were used for showing the measure of dispersion of three or more data values, see Equations 2-5 and 2-6. RSD indicates how precise the data is, for example a RSD of 50% indicates that the data is more spread out than a RSD of 20%. ∑(𝑥−𝑥̅ )2 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = √ 405 (𝑛−1) Equation 2-5 where: x = each sample value, x̅ = mean value of samples, n = number of samples 𝑅𝑆𝐷 (%) = 100 × 410 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑆𝑎𝑚𝑝𝑙𝑒 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 Equation 2-6 The relative percent difference (RPD) was used as a quality indicator when only two data values (duplicate samples) were obtained, Equation 2-7. RPD indicates how precise the data is, for example a RPD of 20% indicates that the data is more precise than a RPD of 50%. 13 𝑅𝑃𝐷 (%) = 100 × 415 𝑥−𝑦 ( 𝑥+𝑦 ) 2 Equation 2-7 where: x = sample number one, y = sample number two 14 3 420 425 Results and Discussion 3.1 PM The PM2.5 emissions are reported in Table 3-1. PM2.5 emissions were higher from the MK-90 than from the skid waste (Table 3-1). The MK-90 PM2.5 emission factor (15.5 g/kg initial source) is similar to those from static firing of CRV-7 (16 g/kg initial source) and MK-58 (34 g/kg initial source) rocket motors [19] and lower than static firing of Sparrow rocket motors (120 g/kg initial source) [1]. The HHRA document lists no PM emission factors, precluding comparison of these site-sampled values. Table 3-1. PM2.5 emission factors in g/kg initial source and lb/lb initial source. Average Stand. Dev.b RSDc RPDd Average Stand. Dev.b Average Average Unit g/kg initial source g/kg initial source % % lb/lb initial source lb/lb initial source g/kg waste lb/lb waste MK-90 na = 5 15.5 1.73 11 N/Ae 0.0155 0.0017 15.5 0.0155 PM2.5 Skid Waste - Type 2 na = 2 2.3 N/Ae N/Ae 9.8 0.0023 N/Ae 7.3 0.0073 a Number of samples collected. Stand. Dev. – standard deviation, calculated only if n ≥ 3. c RSD – relative standard deviation, calculated only if n ≥ 3. d RPD – relative percent difference, calculated only if n = 2. e N/A – not applicable. 430 b 435 3.2 Elements/Metals 3.2.1 Elements/Metals 440 Sixteen metals/elements were detected above instrument limits for one or both of the ordnance sources (Table 3-2). Lead (Pb) and copper (Cu) had the highest emission factors from the MK90 burns of all the metals analyzed, 0.0102 and 0.00307 lb/lb initial source, respectively (Tables 3-2 to 3-4). Pb, chloride (Cl), potassium (K), Cu, and zinc (Zn) had the highest element emission factors for the “high metal” skid waste. The average standard deviation for the MK-90 metal/element emission factors was 29%. The average relative percent difference for the skid waste emission factors (only two samples were taken) was 55%. These relatively low values validate the precision of the sampling method, particularly given the small number (less than 15 445 five) of samples. All element values from the XRF analyses for each collected sample are shown in Appendix A. Table 3-2. Element emission factors in PM2.5 fraction in mg/kg initial source and mg/kg waste.a Skid waste – Type 2 MK-90 Element b n Average Stand. Dev. mg/kg initial source Pb 5 Cu Cl 455 RSD d b n % Average Average RPDe mg/kg initial source mg/kg waste % 2,158 40 1,103 11 2 678.9 5 10,186 3,073 380 12 2 17.4 55.4 92 5 30 24 80 2 80.4 255.5 24 Ca 5 28 5.8 20 2 2.17 6.91 20 K 5 25 5.2 20 2 43.4 138.0 1.9 As 4 21 5.3 25 2 1.45 4.62 62 Fe 5 16 3.3 21 2 0.53 1.70 129 Br 5 15 2.5 17 2 1.53 4.86 45 Ge 5 11 2.7 24 2 0.66 2.09 57 Y 5 11 2.8 26 2 0.80 2.53 46 Rb 5 8 1.6 20 2 0.81 2.57 41 Ba 4 6.4 0.42 6.6 2 0.24 0.75 36 Al 3 7.3 f 5.9 80 0 ND g g N/Ah Cd 5 2.0 1.2 59 1 0.19 0.62 N/Ah Cr 4 1.4 0.21 1 0.038f 0.12f N/Ah g 450 c h 15 ND h ND Zn 5 N/A N/A 2 7.6 24.1 121 a Element concentrations were 22 times higher than the ambient air levels except for Cr which was four times higher than the ambient levels. All element values from XRF analyses are presented in Appendix A. b Number of samples collected with detectable levels. c Stand. Dev. – standard deviation, calculated only if n ≥ 3 d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f Results less than three times the uncertainty level of the analyses. g ND – not detected. h N/A – not applicable. 460 465 470 16 Table 3-3. Metal emission factors in PM2.5 fraction in lb/lb initial source and lb/lb waste.a Skid waste - Type 2 MK-90 Element nb Average Stand. Dev.c 485 490 Average Average lb/lb waste 2.16E-03 RPDe 1.10E-03 11 2 5 1.02E-02 3.07E-03 3.80E-04 12 2 1.74E-05 5.54E-05 92 5 2.97E-05 2.37E-05 80 2 8.04E-05 2.56E-04 24 Ca 5 2.84E-05 5.80E-06 20 2 2.17E-05 6.91E-06 20 K 5 2.53E-05 5.17E-06 20 2 4.34E-05 1.38E-04 1.9 As 4 2.08E-05 5.29E-06 25 2 1.45E-06 4.62E-06 62 Fe 5 1.60E-05 3.32E-06 21 2 5.34E-07 1.70E-06 129 Br 5 1.47E-05 2.49E-06 17 2 1.53E-06 4.86E-06 45 Ge 5 1.11E-05 2.71E-06 24 2 6.59E-07 2.09E-06 57 Rb 5 8.41E-06 1.64E-06 20 2 8.08E-07 2.57E-06 41 Y 5 1.07E-05 2.78E-06 26 2 7.95E-07 2.53E-06 46 Ba 4 6.36E-06 4.19E-07 6.6 2 2.37E-07 7.53E-07 36 Al 3 7.32E-06 f 5.89E-06 80 0 g ND (6.11E-05) ND Cd 5 1.99E-06 1.18E-06 59 1 1.94E-07 6.18E-07 N/Ah Cr 4 1.40E-06 2.06E-07 15 1 3.79E-08f 1.21E-07f N/Ah Pb 5 Cu Cl g 480 nb lb/lb initial source 6.79E-04 lb/lb initial source 475 RSDd % h g % 40 N/Ah h Zn 0 ND (4.73E-07) N/A N/A 2 7.58E-06 2.41E-05 121 a Elements levels were 22 times higher than the ambient air levels except for Cr which was four times higher than the ambient levels. All element values from XRF analyses are presented in Appendix A b Number of samples collected with detectable levels. c Stand. Dev. – standard deviation, calculated only if n ≥ 3 d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f Results less than three times the uncertainty level of the analyses. g ND – not detected, method detection limit within parentheses. h N/A – not applicable. The sampled emission factors were compared with the assumed emission factors used in the RFAAP EFs listed in the HHRA (Table 3-4) [reference]. Of the twelve metals that overlapped for the MK-90s, seven sampled emission factors were lower than the RFAAP EFs and four emission factors were higher than the RFAAP EF (As, Cd, Pb, and Ag). One metal, Hg, was reported as ND so its ratio (<2.2) is not clearly greater or less than unity. For the twelve metals from the skid waste burns, emission factors for ten metals were less than estimated in the HHRA. Two metals, As and Pb, were above unity. 495 17 Table 3-4. Comparison of EFs derived in this project with EFs used by RFAAP’s HHRA. MK-90 Element EF RFAAP EF lb/lb initial source Al 7.32E-06a Sb a EF/RFAAP EF 0.00073 EF Ratio 5.36E-02 EF/RFAAP EF <0.0011 0.41 b ND (<2.14E-07) 5.62E-06 <0.038 lb/lb waste NDb (<6.11E-05) 5.62E-06 As 2.32E-06 2.08E-05 5.54E-07 37.5 4.62E-06 5.54E-07 8.3 Ba 6.36E-06 8.80E-07 0.072 7.53E-07 8.80E-05 0.0086 Cd 1.99E-06 1.32E-05 1.5 6.18E-07 1.32E-06 0.47 f 1.20E-05 0.010 1.1 1.40E-06 1.20E-05 0.12 Pb 1.02E-02 2.06E-03 5.0 Hg NDb (<1.65E-06) 7.38E-07 Ni NDb (<3.32E-07) 1.98E-05 Se 9.38E-07a 1.56E-06 Cr 1.21E-07 2.16E-03 2.06E-03 <2.2 b ND (<1.65E-07) 7.38E-07 <0.22 <0.017 8.19E-09a 1.98E-05 0.00041 0.60 NDb (<6.68E-08) 1.56E-06 <0.043 2.12E-07 0.97 7.55E-05 0.32 a a 1.27E-06 2.12E-07 6.0 2.06E-07 b ND (<4.73E-07) Zn 7.55E-05 <0.0063 2.41E-05 a Results less than three times the uncertainty level of the analyses. b ND – not detected, detection limit within parentheses. Ag 500 1.00E-02 Ratio Skid waste RFAAP EF 3.2.2 Chromium(VI) 505 510 The Cr(VI) emission factors are reported in Table 3-5. Analysis of the PM2.5 solids showed that the percentage of Cr(VI) to total Cr in the emissions was 28% and 14% for the MK-90 and skid waste, respectively. Table 3-4 indicates that the total Cr emission factor from sampling was less than used in the HHRA for both MK-90 (12% of the HHRA emission factor) and skid waste (1% of the HHRA emission factor). Table 3-5. Cr(VI) emission factors. Average Stand. Dev.b RSDc Average Stand. Dev.b Average Average a c Unit mg/kg initial source mg/kg initial source % lb/lb initial source lb/lb initial source mg/kg waste lb/lb waste MK 90 na = 5 0.39 0.13 34 3.95E-07 1.34E-07 0.39 3.95E-07 Cr(VI) Skid Waste -Type 2 na = 1 0.0053 N/Ad N/Ad 5.31E-09 N/Ad 0.017 1.69E-08 Number of samples collected with detectable levels. b Stand. Dev. – standard deviation, RSD – relative standard deviation, calculated only if n ≥ 3. d N/A – not applicable. 18 515 Cr(VI) was detected in all five MK-90 samples collected but only in one of the three samples collected from the skid waste type 2 (Table 3-2). The collection time for the three Cr(VI) skid waste samples was approximately the same but the amount of carbon collected was approximately two times higher in the detected sample than the two with no detectable levels. This simply indicates a greater plume sampling efficiency (collection of oxidized carbon) during the one detectable sample. 520 3.3 HCl, chlorate, and perchlorate 525 No chlorate or perchlorate compounds were detected in any of the six samples collected from skid waste type 1 which was the “high Cl” waste (Table 3-6). The HCl emissions (0.000229 lb/lb initial source) from the skid waste were over 100 times lower than those emitted from static firing (versus open burning) of MK-58 (0.030 lb/lb initial source) and CRV-7 rocket motors (0.086 lb/lb initial source) [19]. Three of the six collected HCl samples were under the method reporting limit (no detectable levels of chloride). Table 3-6. HCl, chlorate, and perchlorate emission factors from skid waste type 1. Unit Average Stand. Dev.d RSDe Average Stand. Dev.d Average Stand. Dev.d Average Stand. Dev.d Average Stand. Dev.d Average Stand. Dev.d 530 535 mg/kg initial source mg/kg initial source % mg/kg waste mg/kg waste lb/lb initial source lb/lb initial source lb/lb waste lb/lb waste % into air from initial sourcef % into air from initial sourcef % into air from wastef % into air from wastef HCl na = 3 229 135 59 459 272 2.29E-04 1.35E-04 4.59E-04 2.72E-04 8.4 5.0 26.8 15.9 Skid Waste -Type 1 Chlorate Perchlorate na = 0 na = 0 ND (0.054)b ND (0.054)b N/Ac N/Ac N/Ac N/Ac ND (0.11)b ND (0.11)b c N/A N/Ac ND (5.40E-08)b ND (5.40E-08)b N/Ac N/Ac ND (1.08E-07)b ND (1.08E-07)b N/Ac N/Ac N/Ac N/Ac c N/A N/Ac N/Ac N/Ac N/Ac N/Ac a Number of samples collected with detectable levels. ND – not detected, detection limit within parentheses. c N/A – not applicable. d Stand. Dev. – standard deviation. e RSD – relative standard deviation. f percent of Cl in skid waste going into air as HCl. b 19 3.4 PCDD/PCDF 540 545 The PCDD/PCDF emission factor from the Type 1, high Cl skid waste (1.77±1.59 ng TEQ/kg waste) was in the same range as emission factors from prescribed forest burns (1.55±1.65 ng TEQ/kg biomass [20]) and much lower than from open burning of municipal solid waste (1,765±1,474 ng TEQ/kg waste [21]). The sampled emission factor was less than 0.1% of the value used in the HHRA. Values are shown in Table 3-7 and Figure 3-1. Emission factors for each homologue group and each TEF-weighted congener are shown in Appendix B, Tables B-1 to B-6. The MK-90s were not sampled for PCDD/PCDF. Table 3-7. PCDD/PCDF results. Skid waste – Type 1 Unit PCDD Total PCDF Total PCDD/PCDF Total PCDD TEQa PCDF TEQa PCDD/PCDF TEQ SUMa PCDD Total PCDF Total PCDD/PCDF Total PCDD TEQa PCDF TEQa PCDD/PCDF TEQ SUMa ng/kg initial source ng/kg initial source ng/kg initial source ng TEQ/kg initial source ng TEQ/kg initial source ng TEQ/kg initial source ng/kg waste ng/kg waste ng/kg waste ng TEQ/kg waste ng TEQ/kg waste ng TEQ/kg waste Average Stand. Dev. RSD EF RFAAP 13.2 33.4 46.6 0.10 0.79 0.88 26.5 67.1 93.6 0.19 1.58 1.77 8.6 37.5 41.1 0.15 0.71 0.79 17.4 75.3 82.6 0.30 1.43 1.59 66% 112% 88% 158% 90% 90% 66% 112% 88% 158% 90% 90% NVb NVb NVb NVb NVb NVb 105.7 105000 105000 17.8 9940 9950 Ratio EF/EF RFAAP 0.25 0.00064 0.00089 0.0107 0.00016 0.00018 a Not detected congeners set to zero. Appendix B shows data with not detected congeners set to the limit of detection. b NV = no value. 550 20 4.00 PCDD/PCDF ng WHO-TEQ/kg waste ng WHO-TEQ/kg biomass 3.50 Dioxin Comparision b) PCDD/PCDF lb WHO-TEQ/lb waste a) 3.00 2.50 2.00 1.50 1.00 0.50 Skid waste 1.2E-08 1E-08 8E-09 6E-09 4E-09 2E-09 0.00 Skid waste 0 Forest burns EF RFAAP EF Figure 3-1. Comparison of PCDD/PCDF (Dioxin) emission factors from a) skid waste and forest burns [20], and b) emission factor derived from this study (EF) and emission factor used today by RFAAP (RFAAP EF). 555 560 565 3.5 VOCs VOC sampling was prioritized only for the type 2 skid waste due to project time limitations. All VOCs analyzed are presented in Tables 3-8 to 3-11. Toluene (3.26E-4 lb/lb waste), benzene (3.11-04 lb/lb waste), naphthalene (1.45E-04 lb/lb waste), methylene chloride (1.26E-04 lb/lb waste), styrene (5.07E-05 lb/lb waste), and xylenes (5.73E-05 lb/lb waste) were the most abundant VOCs emitted from skid waste type 2, all on EPA’s list of hazardous air pollutants [7]. These emission values compare to emissions from static fire of rocket motors: toluene 4.5E-04 lb/lb waste, naphthalene 9.2E-06 lb/lb waste, and xylenes 1.2E-03 lb/lb waste [1]. Of the 26 compounds common between sampled and detectable VOC emissions at Radford and the HHRA, 25 of the VOCs were less than the HHRA emission factor (Table 3-8). Only chloromethane was found at RFAAP to be higher (2.3 times) the HHRA emission factor. Table 3-8. VOC Emission Factors in lb/lb waste from skid waste type 2. Averageb Stand. Dev.c RSDd RPDe Reference Ratio % % lb/lb waste EF/ RFAAP EF na Compound lb/lb waste 1,1,1-Trichloroethanef 1,1,2,2-Tetrachloroethane f 0 ND (8.04E-08) 1.00E-04 0 ND (9.38E-08) 1.04E-04 1,1,2-Trichloroethanef 1 1.11E-06 1.15E-04 1,1-Dichloroethane 0 ND (3.95E-08) 2.92E-05 1,1-Dichloroethene 0 ND (1.14E-07) 4.94E-05 0.010 21 Averageb Stand. Dev.c RSDd RPDe Reference % % lb/lb waste na Ratio 3.28E-06 EF/ RFAAP EF <0.084 5.09E-04 0.053 ND (1.14E-07) 3.28E-06 <0.035 1.01E-07 4.31E-05 0.002 1 1.34E-06 4.31E-05 0.031 4 7.28E-06 4.13E-06 57 4.31E-05 0.169 4 1.97E-05 5.32E-06 27 0.453 1,3-Dichlorobenzene 1 1.14E-07 4.35E-05 NVg 1,4-Dichlorobenzene 1 1.73E-07 0.053 1,4-Dioxane 2,2,4-Trimethylpentane (Isooctane) 2-Butanone (MEK) 2 4 6.93E-07 7.21E-07 3.28E-06 NVg 7.11E-07 99 NVg 4 1.02E-05 6.02E-06 59 NVg 2-Hexanone 1 6.43E-06 NVg 2-Propanol (Isopropyl Alcohol) 1 3.95E-06 NVg 4-Methyl-2-pentanone 4 1.47E-06 1.60E-06 109 NVg Acetone 4 4.47E-05 2.70E-05 35 7.44E-04 Acetonitrilef 4 2.69E-05 1.58E-05 56 NVg Benzenef 4 3.11E-04 1.85E-04 59 9.69E-04 Bromodichloromethane 0 ND (6.37E-08) 9.69E-04 Bromoform 0 ND (9.38E-08) NVg 1 1.07E-06 4 1.09E-06 1 1.71E-06 3 2.35E-06 1.68E-06 Chloroform 3 2.23E-07 Chloromethanef 4 7.58E-06 0 Compound lb/lb waste 1,2,4-Trichlorobenzenef 0 1,2,4-Trimethylbenzene 4 2.72E-05 1,2-Dibromo-3-chloropropane 0 ND (1.41E-07) 1,2-Dibromoethane 1,2-Dichloro-1,1,2,2tetrafluoroethane (CFC 114) 1,2-Dichlorobenzene 0 3 ND (6.57E-08) 1.46E-07 0 1,2-Dichloroethane 1 1,2-Dichloropropane 1,3,5-Trimethylbenzene 1,3-Butadiene f Carbon Disulfide f ND (2.75E-07) 1.53E-05 56 1.51E-07 103 71 0.060 0.321 3.25E-06 0.329 3.25E-06 0.335 3.25E-06 0.526 71 3.25E-06 0.723 1.55E-07 70 3.25E-06 0.069 6.64E-06 88 2.332 ND (6.23E-08) 3.25E-06 NVg 0 ND (7.37E-08) NVg Cumenef 4 3.75E-06 Cyclohexane 1 8.71E-06 Dibromochloromethane Dichlorodifluoromethane (CFC 12) Ethanol 0 3 ND (4.56E-08) 6.72E-06 5.64E-06 84 NVg 4 1.06E-05 7.98E-06 80 NVg Ethylbenzenef 4 2.08E-05 1.00E-05 48 4.53E-05 Carbon Tetrachloridef Chlorobenzene f Chloroethane f cis-1,2-Dichloroethene cis-1,3-Dichloropropene f 1.15E-06 2.41E-06 106 64 NVg 2.67E-05 NVg 0.326 0.459 22 Averageb Stand. Dev.c RSDd RPDe Reference Ratio % % lb/lb waste EF/ RFAAP EF na Compound lb/lb waste NVg 46 NVg 0 ND (2.01E-07) m,p-Xylenesf 4 4.11E-05 Methyl tert-Butyl Ether 0 ND (4.69E-08) Methylene Chloridef 4 1.26E-04 2.37E-04 189 1.17E-03 0.108 4 1.45E-04 8.23E-05 57 7.87E-04 0.184 Naphthalene f 1.91E-05 NVg g n-Heptane 4 4.70E-06 1.85E-06 39 NV n-Hexane 4 1.63E-05 2.94E-05 180 2.56E-05 n-Octane 1.56E-05 6.08E-06 39 NV 4 1.61E-05 8.53E-06 53 NVg Styrenef 4 5.07E-05 3.15E-05 62 Tetrachloroethene 2 6.11E-07 5.56E-05 NVg Tetrahydrofuran (THF) 3 7.30E-07 2.04E-07 28 NVg Toluenef 4 3.26E-04 4.10E-04 126 trans-1,2-Dichloroethene 0 ND (8.04E-08) 4.75E-04 NVg trans-1,3-Dichloropropene 0 ND (7.37E-08) NVg Trichloroethene 1 2.81E-07 Trichlorofluoromethane 4 2.48E-06 1.91E-06 77 6.59E-05 NVg Trichlorotrifluoroethane 4 1.00E-06 1.11E-06 111 NVg Vinyl Chloridef 0 ND (9.38E-08) Xylenes 4 5.73E-05 2.75E-05 a Number of samples with detectable levels out of 4 samples. b ND – not detected. Detection limit within parentheses. c Stand. Dev. – standard deviation, calculated only if n ≥ 3. d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f On U.S. EPA’s list of hazardous air pollutants [7] g NV = no value. 48 f 185 0.637 g 4 o-Xylene 570 N/A Hexachlorobutadienef 0.912 0.686 0.004 9.28E-05 4.52E-04 0.127 575 580 23 Table 3-9. VOC Emission Factors in mg/kg waste from skid waste type 2. na Compound 1,1,1-Trichloroethane f 0 1,1,2,2-Tetrachloroethane f 0 1,1,2-Trichloroethanef 1 1,1-Dichloroethane 0 1,1-Dichloroethene 0 f 0 1,2,4-Trimethylbenzene 4 1,2-Dibromo-3-chloropropane 0 1,2-Dibromoethane 0 1,2-Dichloro-1,1,2,2-tetrafluoroethane (CFC 114) 3 1,2-Dichlorobenzene 0 1,2-Dichloroethane 1 1,2-Dichloropropane 1 1,3,5-Trimethylbenzene 4 1,2,4-Trichlorobenzene 1,3-Butadiene f 4 1,3-Dichlorobenzene 1 1,4-Dichlorobenzene 1 1,4-Dioxane 2 2,2,4-Trimethylpentane (Isooctane) 4 2-Butanone (MEK) 4 2-Hexanone 1 2-Propanol (Isopropyl Alcohol) 1 4-Methyl-2-pentanone 4 Acetone 4 Acetonitrilef 4 Benzene f 4 Bromodichloromethane 0 Bromoform 0 Carbon Disulfide f 0 Carbon Tetrachloridef Chlorobenzene 4 f 1 Chloroethane 3 f Chloroform 3 Chloromethanef 4 cis-1,2-Dichloroethene cis-1,3-Dichloropropene Cumene f 0 f 0 4 Cyclohexane 1 Dibromochloromethane 0 Averageb Stand. Dev.c mg/kg waste ND (0.080) ND (0.094) 1.11 ND (0.040) ND (0.11) ND (0.28) 27.17 ND (0.14) ND (0.066) 0.15 ND (0.11) 0.1 1.34 7.28 19.67 0.11 0.17 0.69 0.72 10.24 6.43 3.95 1.47 44.7 26.9 310.88 ND (0.064) ND (0.094) 1.07 1.09 1.71 2.35 0.22 7.58 ND (0.062) ND (0.074) 3.75 8.71 ND (0.046) RSDd RPDe % % 15.31 56 0.15 103 4.13 5.32 57 27 71 0.71 6.02 99 1.6 26.95 15.8 184.78 109 1.15 106 1.68 0.16 6.64 71 2.41 64 59 60 59 59 70 88 24 na Compound Dichlorodifluoromethane (CFC 12) 3 Ethanol 4 Ethylbenzene f 4 Hexachlorobutadienef 0 m,p-Xylenesf 4 Methyl tert-Butyl Ether Methylene Chloride 0 4 Naphthalenef 4 n-Heptane 4 n-Hexane 4 n-Octane 4 o-Xylene f 4 Styrenef 4 Tetrachloroethene 2 Tetrahydrofuran (THF) 3 Toluene 585 f f 4 trans-1,2-Dichloroethene 0 trans-1,3-Dichloropropene 0 Trichloroethene 1 Trichlorofluoromethane 4 Trichlorotrifluoroethane 4 Vinyl Chloridef 0 a Number of samples with detectable levels out of 4 samples. b ND – not detected. Detection limit within parentheses. c Stand. Dev. – standard deviation, calculated only if n ≥ 3. d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f On U.S. EPA’s list of hazardous air pollutants [7] Averageb Stand. Dev.c RSDd RPDe % % mg/kg waste 6.72 10.63 20.81 ND (0.20) 41.14 ND (0.047) 125.62 144.54 4.7 16.35 15.62 16.12 50.71 0.61 0.73 326.46 ND (0.080) ND (0.074) 0.28 2.48 1 ND (0.094) 5.64 7.98 10.04 84 19.07 46 237.46 82.32 1.85 29.36 6.08 8.53 31.49 189 75 48 57 39 180 39 53 62 185 0.2 409.87 126 1.91 1.11 111 28 77 590 Table 3-10. VOC Emission Factors in lb/lb initial source from skid waste type 2. na Compound 1,1,1-Trichloroethane f Averageb lb/lb initial source 0 ND (2.53E-08) 0 ND (2.95E-08) 1 3.48E-07 1,1-Dichloroethane 0 ND (1.24E-08) 1,1-Dichloroethene 0 ND (3.58E-08) 1,2,4-Trichlorobenzenef 0 ND (8.64E-08) 1,2,4-Trimethylbenzene 4 8.55E-06 1,1,2,2-Tetrachloroethanef 1,1,2-Trichloroethane f Stand. Dev.c 4.82E-06 RSDd RPDe % % 56 25 na Compound Averageb Stand. Dev.c lb/lb initial source 1,2-Dibromo-3-chloropropane 0 ND (4.43E-08) 1,2-Dibromoethane 0 ND (2.07E-08) 1,2-Dichloro-1,1,2,2-tetrafluoroethane (CFC 114) 3 4.60E-08 1,2-Dichlorobenzene 0 ND (3.58E-08) 1,2-Dichloroethane 1 3.16E-08 1,2-Dichloropropane 1 4.22E-07 1,3,5-Trimethylbenzene 4 1,3-Butadienef 4 1,3-Dichlorobenzene 1 3.58E-08 1,4-Dichlorobenzene 1 5.45E-08 1,4-Dioxane 2 2.18E-07 2,2,4-Trimethylpentane (Isooctane) 4 2-Butanone (MEK) RSDd RPDe % % 4.74E-08 103 2.29E-06 1.30E-06 57 6.19E-06 1.67E-06 27 2.27E-07 2.24E-07 99 4 3.22E-06 1.89E-06 59 2-Hexanone 1 2.02E-06 2-Propanol (Isopropyl Alcohol) 1 1.24E-06 4-Methyl-2-pentanone 4 4.64E-07 5.04E-07 109 Acetone 4 1.78E-05 6.16E-06 35 4 1.10E-05 6.20E-06 56 4 9.78E-05 5.81E-05 59 Bromodichloromethane 0 ND (2.00E-08) Bromoform 0 ND (2.95E-08) 1 3.37E-07 4 3.43E-07 3.63E-07 106 1 5.37E-07 Chloroethane 3 7.40E-07 5.28E-07 71 Chloroformf 3 7.02E-08 4.89E-08 70 4 2.38E-06 2.09E-06 88 0 ND (1.96E-08) 7.58E-07 64 Acetonitrile Benzene f f Carbon Disulfide f Carbon Tetrachloride Chlorobenzene f f Chloromethane f cis-1,2-Dichloroethene cis-1,3-Dichloropropene f 71 0 ND (2.32E-08) Cumenef 4 1.18E-06 Cyclohexane 1 2.74E-06 Dibromochloromethane 0 ND (1.43E-08) Dichlorodifluoromethane (CFC 12) 3 2.11E-06 1.77E-06 84 Ethanol 4 3.56E-06 2.85E-06 80 Ethylbenzenef 4 6.55E-06 3.16E-06 48 0 ND (6.32E-08) m,p-Xylenes 4 1.29E-05 6.00E-06 46 Methyl tert-Butyl Ether 0 ND (1.48E-08) Methylene Chloridef 4 3.95E-05 7.47E-05 189 4 4.55E-05 2.59E-05 57 Hexachlorobutadiene f f Naphthalene f 26 na Compound Stand. Dev.c lb/lb initial source RSDd RPDe % % n-Heptane 4 1.48E-06 5.81E-07 39 n-Hexane 4 5.14E-06 9.24E-06 180 n-Octane 4 4.92E-06 1.91E-06 39 4 5.07E-06 2.68E-06 53 4 1.60E-05 9.91E-06 62 Tetrachloroethene 2 1.92E-07 Tetrahydrofuran (THF) 3 2.30E-07 6.41E-08 28 Toluenef 4 1.03E-04 1.29E-04 126 trans-1,2-Dichloroethene 0 ND (2.53E-08) trans-1,3-Dichloropropene 0 ND (2.32E-08) Trichloroethene 1 8.85E-08 Trichlorofluoromethane 4 7.80E-07 6.02E-07 77 Trichlorotrifluoroethane 4 3.15E-07 3.50E-07 111 o-Xylene Styrene f f f 595 Averageb Vinyl Chloride 0 a Number of samples with detectable levels out of 4 samples. b ND – not detected. Detection limit within parentheses. c Stand. Dev. – standard deviation, calculated only if n ≥ 3. d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f On U.S. EPA’s list of hazardous air pollutants [7]. ND (2.95E-08) Table 3-11. VOC Emission Factors in mg/kg initial source. na Compound 1,1,1-Trichloroethane f 1,1,2,2-Tetrachloroethane f Averageb mg/kg initial source 0 ND (0.025) 0 ND (0.030) 1,1,2-Trichloroethanef 1 0.35 1,1-Dichloroethane 0 ND (0.012) 1,1-Dichloroethene 0 ND (0.036) f 0 ND (0.086) 1,2,4-Trimethylbenzene 4 8.55 1,2-Dibromo-3-chloropropane 0 ND (0.044) 1,2-Dibromoethane 0 ND (0.021) 1,2-Dichloro-1,1,2,2-tetrafluoroethane (CFC 114) 3 0.046 1,2-Dichlorobenzene 0 ND (0.036) 1,2-Dichloroethane 1 0.03 1,2-Dichloropropane 1 0.42 1,3,5-Trimethylbenzene 4 1,2,4-Trichlorobenzene 1,3-Butadiene f 1,3-Dichlorobenzene Stand. Dev.c RSDd RPDe % % 4.82 56 0.047 103 2.29 1.3 57 4 6.19 1.67 27 1 0.04 27 na Compound Averageb Stand. Dev.c mg/kg initial source RSDd RPDe % % 1,4-Dichlorobenzene 1 0.05 1,4-Dioxane 2 0.22 2,2,4-Trimethylpentane (Isooctane) 4 0.23 0.22 99 2-Butanone (MEK) 4 3.22 1.89 59 2-Hexanone 1 2.02 2-Propanol (Isopropyl Alcohol) 1 1.24 4-Methyl-2-pentanone 4 0.46 0.5 109 Acetone 4 14.06 8.48 60 4 8.46 4.97 59 4 97.8 58.13 59 Bromodichloromethane 0 ND (0.020) Bromoform 0 ND (0.030) 0 ND (0.17) 4 0.34 0.36 106 1 0.54 Chloroethane 3 0.74 0.53 71 Chloroformf 3 0.07 0.05 70 4 2.38 2.09 88 0 ND (0.020) 0.76 64 Acetonitrile Benzene f f Carbon Disulfide f Carbon Tetrachloride Chlorobenzene f f Chloromethane f cis-1,2-Dichloroethene cis-1,3-Dichloropropene f 71 0 ND (0.023) Cumenef 4 1.18 Cyclohexane 1 2.74 Dibromochloromethane 0 ND (0.014) Dichlorodifluoromethane (CFC 12) 3 2.11 1.77 84 Ethanol 4 3.34 2.51 75 Ethylbenzenef 4 6.55 3.16 48 0 ND (0.063) m,p-Xylenes 4 12.94 6 46 Methyl tert-Butyl Ether 0 ND (0.015) Methylene Chloridef 4 39.52 74.71 189 4 45.47 25.9 57 n-Heptane 4 1.48 0.58 39 n-Hexane 4 5.14 9.24 180 n-Octane 4 4.92 1.91 39 4 5.07 2.68 53 4 15.95 9.91 62 Tetrachloroethene 2 0.19 Tetrahydrofuran (THF) 3 0.23 0.06 28 Toluenef 4 102.71 128.94 126 trans-1,2-Dichloroethene 0 ND (0.025) Hexachlorobutadiene f f Naphthalene o-Xylene Styrene f f f 185 28 na Compound 605 Stand. Dev.c mg/kg initial source RSDd RPDe % % trans-1,3-Dichloropropene 0 ND (0.023) Trichloroethene 1 0.09 Trichlorofluoromethane 4 0.78 0.6 77 Trichlorotrifluoroethane 4 0.32 0.35 111 f 600 Averageb Vinyl Chloride 0 a Number of samples with detectable levels out of 4 samples. b ND – not detected. Detection limit within parentheses. c Stand. Dev. – standard deviation, calculated only if n ≥ 3. d RSD – relative standard deviation, calculated only if n ≥ 3. e RPD – relative percent difference, calculated only if n = 2. f On U.S. EPA’s list of hazardous air pollutants [7]. ND (0.030) 3.6 Energetics 610 None of the energetics and nitroaromatic compounds for the MK-90 rocket motors exceeded the analytical method detection limit (Table 3-12). Energetics were not sampled for the skid waste due to time limitations. The ratio of the method detection limit (for the sampled emission factor) to that of the HHRA emission factor resulted in eight overlapping compounds to be less than 1.1. 615 620 625 29 Table 3-12. Energetics based on method detection limit. Energetics MK-90 mg/kg initial source < 51 MK-90 lb/lb initial source < 5.1E-05 RFAAP EF lb/lb initial source NVa Ratio EF/RFAAP EF 1,3,5-Trinitrobenzeneb < 1.1 < 1.1E-06 2.28E-05 <0.048 1,3-Dinitrobenzene < 1.1 < 1.1E-06 8.19E-06 <0.13 2,4,6-Trinitrotoluene < 1.1 < 1.1E-06 3.48E-05 <0.032 2,4-Dinitrotoluene < 1.1 < 1.1E-06 1.05E-04 <0.010 2,6-Dinitrotoluene < 1.1 < 1.1E-06 9.81E-07 <1.1 Nitrocellulose (n=2) 630 2-Amino-4,6-Dinitrotoluene < 1.1 < 1.1E-06 NV 2-Nitrotoluene < 1.1 < 1.1E-06 NVa 3,5-DNA < 1.1 < 1.1E-06 NVa 3-Nitrotoluene < 1.1 < 1.1E-06 NVa 4-Amino-2,6-Dinitrotoluene < 1.1 < 1.1E-06 NVa 4-Nitrotoluene < 1.1 < 1.1E-06 NVa HMX < 1.1 < 1.1E-06 2.16E-05 <0.051 Nitrobenzene < 1.1 < 1.1E-06 3.28E-06 <0.34 Nitroglycerin < 1.1 < 1.1E-06 3.07E-06 <0.36 PETN < 2.7 < 2.7E-06 NVa RDX < 1.1 < 1.1E-06 NVa < 1.1 < 1.1E-06 Tetryl a NV = no value. b Four samples for all energetics except nitrocellulose. NVa 4 635 640 645 a Conclusions Aerial sampling methods for emission quantification of demilitarization efforts have only been comprehensively in use since their first deployment in 2010. The logistical challenges experienced in these earlier efforts and recent developments in UAV and sensor technology prompted EPA’s Office of Research and Development to create a new system applicable for sampling open demilitarization plumes. Working with pilots and a hexacopter from NASA Ames, EPA/ORD demonstrated the first comprehensive test of a UAV-borne emission sampler at RFAAP’s open burning grounds. Plume sampling of open burns of MK-90 rocket motors and skid waste was successfully accomplished with the UAV/Kolibri system based on the number of plumes sampled (100%), the repeatability of the emission factors, and the comparability of the emission factors with previous aerial sampling methods. Emissions were sampled for PM, elements including metals, particularly Cr(VI), VOCs, dioxins, and nitroaromatics. PM2.5 emission factors for MK-90s were within the range of three other previously-documented sources. The majority of the metal emission factors, 17 of 24, were lower than those emission factors used in the HHRA. Cr(VI) emissions were 28% and 14% of the total Cr emitted from the burns of the MK-90 and skid waste, respectively. Emission factors 30 650 were compared with other recently sampled, aerial emission data and found to be consistent or, in some cases (for example, HCl) found to be considerably lower. Chlorate and perchlorate emission were below detection limits. Dioxin emissions were less than 0.1% of the emission factor found in the HHRA for skid waste and were similar to those values typically reported from prescribed forest or biomass burns. Residual energetics and nitroaromatics for the MK-90s were below the detection limit. Of the 26 compounds in common between detectable VOC emissions from Radford’s skid waste and the listed HHRA emission factors, 25 of the VOCs were less than the HHRA emission factor. 655 5 660 1 Aurell, J.; Gullett, B.K.; Tabor, D.; Williams, R.K.; Mitchell, W.; Kemme, M.R. Aerostat-based sampling of emissions from open burning and open detonation of military ordnance. Journal of Hazardous Materials. 284:108-120; 2015 2 Ragland, K.W.; Aerts, D.J.; Baker, A.J. Properties of wood for combustion analysis. Bioresource Technology. 37:161-168; 1991 3 Aurell, J.; Gullett, B. Characterization of Emissions from Open Burning of Meals Readyto-Eat and their Paperboard Packaging. EPA 600/R-16/220. U.S. EPA. 2016. https://nepis.epa.gov/Exe/ZyPDF.cgi/P100PZ6F.PDF?Dockey=P100PZ6F.PDF Accessed April 5, 2017 4 U.S. EPA Method 3A. Determination of oxygen and carbon dioxide concentrations in emissions from stationary sources (instrumental analyzer procedure). 1989. http://www.epa.gov/ttn/emc/promgate/m-03a.pdf Accessed May 5, 2014 5 40 CFR Part 50, Appendix L. Reference method for the determination of particulate matter as PM2.5 in the Atmosphere. 1987. https://www.gpo.gov/fdsys/pkg/CFR-2014title40-vol2/pdf/CFR-2014-title40-vol2-part50-appL.pdf Accessed November 22, 2016 6 U.S. EPA Compendium Method IO-3.3. Determination of metals in ambient particulate matter using X-Ray Fluorescence (XRF) Spectroscopy. 1999. http://www.epa.gov/ttnamti1/files/ambient/inorganic/mthd-3-3.pdf Accessed May 5, 2014 7 U.S. EPA Hazardous Air Pollution List. Clean Air Act: Title 42 - The public health and welfare. U.S. Government Printing Office. 2008. http://www.gpo.gov/fdsys/pkg/USCODE-2008-title42/pdf/USCODE-2008-title42chap85.pdf Accessed May 5 2014 8 U.S. EPA SOP. Standard Operating Procedure for the Determination of Hexavalent Chromium In Ambient Air Analyzed By Ion Chromatography (IC). 2006. https://www3.epa.gov/ttnamti1/files/ambient/airtox/hexchromsop.pdf Accessed April 4, 2017 665 670 675 680 References 31 9 U.S. EPA Method TO-17. Determination of Volatile Organic Compounds in Ambient Air Using Active Sampling Onto Sorbent Tubes. 1997. http://www.epa.gov/ttnamti1/files/ambient/airtox/to-17r.pdf Accessed July 25, 2013 10 U.S. EPA Method 8330B. Nitroaromatics, nitramines, and nitrate esters by high performance liquid chromatograph (HPLC). 2006. https://www.epa.gov/sites/production/files/2015-07/documents/epa-8330b.pdf Accessed July 18, 2016 11 U.S. EPA Method 353.2. Determination of Nitrate-Nitrite Nitrogen by automated colorimetry. 1993. https://www.epa.gov/sites/production/files/201508/documents/method_353-2_1993.pdf Acessed July 18, 2016 12 International standard ISO 21438-2:2009. Workplace atmospheres — Determination of inorganic acids by ion chromatography — Part 2: Volatile acids, except hydrofluoric acid (hydrochloric acid, hydrobromic acid and nitric acid). 2009. 13 Agency for Toxic Substances and Disease Registry. Perchlorate: Potential for human exposure. Chapter 6: tp162-c6. https://www.atsdr.cdc.gov/ToxProfiles/tp162-c6.pdf Accessed May 5, 2017 14 U.S. EPA Method 26. Determination of Hydrogen Halide and Halogen Emissions from Stationary Sources Non-Isokinetic Method. https://www3.epa.gov/ttnemc01/promgate/m26.pdf Accessed July 15, 2016 15 U.S. EPA Method 23. Determination of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans from stationary sources. 40 CFR Part 60, Appendix A. 1991. http://www.epa.gov/ttn/emc/promgate/m-23.pdf Accessed November 10, 2015 16 U.S. EPA Method 8290A. Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) by high-resolution gas chromatography/highresolution mass spectrometry (HRGC/HRMS). 2007. http://www.epa.gov/osw/hazard/testmethods/sw846/pdfs/8290a.pdf Accessed November 21, 2012 17 Van den Berg, M.; Birnbaum, L.S.; Denison, M.; De Vito, M.; Farland, W.; Feeley, M.; Fiedler, H.; Hakansson, H.; Hanberg, A.; Haws, L.; Rose, M.; Safe, S.; Schrenk, D.; Tohyama, C.; Tritscher, A.; Tuomisto, J.; Tysklind, M.; Walker, N.; Peterson, R.E. The 2005 World Health Organization reevaluation of human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci. 93:223-241; 2006 18 Burling, I.R.; Yokelson, R.J.; Griffith, D.W.T.; Johnson, T.J.; Veres, P.; Roberts, J.M.; Warneke, C.; Urbanski, S.P.; Reardon, J.; Weise, D.R.; Hao, W.M.; de Gouw, J. Laboratory measurements of trace gas emissions from biomass burning of fuel types from the southeastern and southwestern United States. Atmos Chem Phys. 10:11115-11130; 2010 19 Gullett, B.K.; Aurell, J.; Williams, R. Characterization of Air Emissions from Open Burning and Open Detonation of Gun Propellants and Ammunition. SERDP WP-2233. 2016. https://www.serdp-estcp.org/index.php/Program-Areas/Weapons-Systems-and- 685 690 695 700 705 710 715 720 32 725 730 Platforms/Energetic-Materials-and-Munitions/Munitions-Emissions/WP-2233/WP-2233TR Accessed March 29, 2017 20 Aurell, J.; Gullett, B.K. Emission Factors from Aerial and Ground Measurements of Field and Laboratory Forest Burns in the Southeastern US: PM2.5, Black and Brown Carbon, VOC, and PCDD/PCDF. Environmental Science & Technology. 47:8443-8452; 2013 21 Aurell, J.; Gullett, B.K.; Yamamoto, D. Emissions from Open Burning of Simulated Military Waste from Forward Operating Bases. Environmental Science & Technology. 46:11004-11012; 2012 33 735 Appendices 740 Appendix A: Element emission factors Table A-1. Elements analyzed for each sample collected in mg/kg initial source.a MK90 Date 09/27/16 MK90 09/27/16 MK90 10/05/16 MK90 10/05/16 MK90 10/05/16 Skid waste 10/06/16 Skid waste 10/06/16 Element Unit Burn 1 Burn 2,3 Burn 1 Burn 2 Burn 3 Burn 1 Burn 1 Na mg/kg initial source 8.58E+02 9.24E+02 1.06E+03 1.05E+03 6.66E+02 2.77E+01 4.32E+01 Na Unc. mg/kg initial source 1.16E+02 1.37E+02 1.62E+02 1.68E+02 1.32E+02 1.40E+01 1.61E+01 Mg mg/kg initial source 1.40E+02 1.66E+02 1.96E+02 1.86E+02 1.25E+02 1.91E+00 2.92E+00 Mg Unc. mg/kg initial source 1.99E+01 2.56E+01 2.89E+01 3.03E+01 2.45E+01 1.29E+00 1.43E+00 Al mg/kg initial source 1.54E+00 ND ND 1.33E+01 7.11E+00 ND ND Al Unc. mg/kg initial source 4.13E+00 5.50E+00 6.43E+00 6.72E+00 5.62E+00 6.11E-01 6.50E-01 Si mg/kg initial source 1.56E+02 1.22E+02 1.66E+02 1.72E+02 1.39E+02 1.90E+01 2.27E+01 Si Unc. mg/kg initial source 1.02E+01 9.31E+00 1.18E+01 1.21E+01 9.87E+00 1.33E+00 1.52E+00 P mg/kg initial source 3.82E+00 2.20E+00 4.06E+00 5.93E+00 5.20E+00 4.30E-01 7.16E-01 P Unc. mg/kg initial source 1.85E+00 2.30E+00 2.89E+00 2.96E+00 2.39E+00 2.41E-01 2.80E-01 S mg/kg initial source ND ND ND ND ND ND ND S Unc. mg/kg initial source 1.72E+02 1.49E+02 3.99E+01 1.93E+02 3.06E+01 1.76E+00 2.40E+00 Cl mg/kg initial source 4.38E+01 6.31E+01 8.86E+00 2.46E+01 7.98E+00 7.08E+01 9.00E+01 Cl Unc. mg/kg initial source 6.64E+00 9.41E+00 6.73E+00 7.07E+00 5.74E+00 3.70E+00 4.69E+00 K mg/kg initial source 3.34E+01 2.58E+01 2.00E+01 2.58E+01 2.18E+01 4.30E+01 4.38E+01 K Unc. mg/kg initial source 2.37E+00 2.70E+00 1.83E+00 2.13E+00 1.75E+00 2.23E+00 2.28E+00 Ca mg/kg initial source 3.74E+01 2.19E+01 2.82E+01 2.96E+01 2.51E+01 2.39E+00 1.96E+00 Ca Unc. mg/kg initial source 2.42E+00 2.20E+00 2.13E+00 2.13E+00 1.79E+00 2.21E-01 2.22E-01 Ti mg/kg initial source 1.89E+00 ND 1.61E+00 9.88E-01 1.82E+00 2.48E-01 1.24E-01 Ti Unc. mg/kg initial source 4.37E-01 6.97E-01 4.64E-01 4.72E-01 4.02E-01 5.34E-02 5.73E-02 V mg/kg initial source ND 1.99E-01 3.25E-01 2.36E-01 ND 3.34E-02 ND V Unc. mg/kg initial source 4.37E-01 4.98E-01 3.83E-01 4.72E-01 3.32E-01 3.34E-02 4.10E-02 Cr mg/kg initial source 1.27E+00 4.98E-01 1.47E+00 1.66E+00 1.20E+00 2.67E-02 4.92E-02 Cr Unc. mg/kg initial source 3.06E-01 5.97E-01 3.25E-01 3.93E-01 2.82E-01 4.01E-02 4.10E-02 Mn mg/kg initial source 5.25E-01 ND ND 1.57E-01 ND ND ND Mn Unc. mg/kg initial source 5.25E-01 9.96E-01 4.10E-01 5.16E-01 3.55E-01 5.34E-02 7.37E-02 Fe mg/kg initial source 1.62E+01 1.44E+01 1.53E+01 2.15E+01 1.27E+01 8.79E-01 1.89E-01 Fe Unc. mg/kg initial source 1.14E+00 1.41E+00 9.80E-01 1.34E+00 8.25E-01 9.43E-02 7.37E-02 Co mg/kg initial source ND ND ND ND ND ND ND Co Unc. mg/kg initial source 3.94E-01 5.97E-01 3.25E-01 3.93E-01 3.08E-01 3.34E-02 4.10E-02 Appendix A 1 MK90 Date 09/27/16 Element Unit Burn 1 Ni mg/kg initial source ND Ni Unc. mg/kg initial source Cu MK90 09/27/16 Burn 2,3 MK90 10/05/16 MK90 10/05/16 MK90 10/05/16 Skid waste 10/06/16 Skid waste 10/06/16 Burn 1 Burn 2 Burn 3 Burn 1 ND ND ND ND ND 8.19E-03 4.37E-01 6.97E-01 3.56E-01 4.32E-01 3.32E-01 3.34E-02 3.28E-02 mg/kg initial source 2.99E+03 2.55E+03 3.40E+03 3.48E+03 2.95E+03 2.54E+01 9.44E+00 Cu Unc. mg/kg initial source 1.50E+02 1.27E+02 1.70E+02 1.74E+02 1.47E+02 1.28E+00 4.85E-01 Zn mg/kg initial source ND ND ND ND ND 3.00E+00 1.22E+01 Zn Unc. mg/kg initial source 5.25E-01 7.97E-01 5.46E-01 6.34E-01 4.73E-01 1.68E-01 6.26E-01 Ga mg/kg initial source 9.71E+00 5.30E+00 2.70E+00 3.79E+00 ND ND ND Ga Unc. mg/kg initial source 2.86E+00 3.30E+00 3.05E+00 3.12E+00 2.55E+00 1.94E-01 2.55E-01 Ge mg/kg initial source 1.08E+01 6.81E+00 1.29E+01 1.12E+01 1.39E+01 4.70E-01 8.48E-01 Ge Unc. mg/kg initial source 1.14E+00 1.19E+00 1.28E+00 1.30E+00 1.16E+00 6.68E-02 9.83E-02 As mg/kg initial source 1.35E+01 2.20E+00 2.09E+01 2.27E+01 2.61E+01 1.01E+00 1.90E+00 As Unc. mg/kg initial source 6.16E+00 6.81E+00 6.95E+00 7.11E+00 5.95E+00 4.09E-01 5.77E-01 Se mg/kg initial source ND ND ND 1.26E+00 6.14E-01 ND ND Se Unc. mg/kg initial source 1.10E+00 1.19E+00 1.20E+00 1.22E+00 1.02E+00 6.68E-02 9.01E-02 Br mg/kg initial source 1.49E+01 1.05E+01 1.69E+01 1.61E+01 1.53E+01 1.19E+00 1.87E+00 Br Unc. mg/kg initial source 1.06E+00 9.96E-01 1.20E+00 1.18E+00 1.06E+00 8.01E-02 1.24E-01 Rb mg/kg initial source 7.34E+00 8.40E+00 1.02E+01 9.84E+00 6.28E+00 6.44E-01 9.72E-01 Rb Unc mg/kg initial source 8.80E-01 8.96E-01 1.01E+00 1.03E+00 8.02E-01 6.01E-02 8.19E-02 Sr mg/kg initial source 1.54E+00 2.00E+00 ND 9.88E-01 2.17E+00 2.67E-02 ND Sr Unc. mg/kg initial source 6.18E-01 7.97E-01 7.36E-01 7.13E-01 5.67E-01 4.01E-02 5.73E-02 Y mg/kg initial source 1.44E+01 7.41E+00 1.26E+01 9.76E+00 9.31E+00 6.11E-01 9.80E-01 Y Unc. mg/kg initial source 1.80E+00 1.71E+00 1.94E+00 1.90E+00 1.61E+00 1.01E-01 1.48E-01 Zr mg/kg initial source ND ND ND ND ND ND ND Zr Unc. mg/kg initial source 7.05E-01 9.96E-01 7.63E-01 8.31E-01 6.61E-01 5.34E-02 6.55E-02 Mo mg/kg initial source 1.14E+00 6.97E-01 5.42E-02 1.34E+00 1.06E+00 3.34E-02 ND Mo Unc. mg/kg initial source 7.05E-01 1.19E+00 7.36E-01 8.31E-01 6.37E-01 6.68E-02 8.19E-02 Pd mg/kg initial source 2.15E+00 9.96E-02 1.36E-01 ND 1.91E+00 8.76E-02 ND Pd Unc. mg/kg initial source 1.36E+00 2.50E+00 1.12E+00 1.34E+00 9.46E-01 1.54E-01 1.98E-01 Ag mg/kg initial source 1.27E+00 ND ND ND ND ND 2.06E-01 Ag Unc. mg/kg initial source 1.32E+00 2.50E+00 1.06E+00 1.34E+00 9.22E-01 1.54E-01 1.89E-01 Cd mg/kg initial source 1.23E+00 3.10E+00 3.27E+00 1.82E+00 5.43E-01 1.94E-01 ND Cd Unc. mg/kg initial source 1.32E+00 2.50E+00 1.12E+00 1.34E+00 9.46E-01 1.48E-01 1.89E-01 Appendix A Burn 1 2 MK90 Date 09/27/16 MK90 09/27/16 MK90 10/05/16 MK90 10/05/16 MK90 10/05/16 Skid waste 10/06/16 10/06/16 Element Unit Burn 1 Burn 2,3 Burn 1 Burn 2 Burn 3 In mg/kg initial source 2.37E+00 1.71E+00 2.10E+00 1.97E-01 1.98E+00 ND 1.64E-02 In Unc. mg/kg initial source 1.36E+00 2.70E+00 1.15E+00 1.46E+00 9.69E-01 1.61E-01 2.06E-01 Sn mg/kg initial source ND 7.97E-01 7.71E+00 1.66E+00 2.82E-01 8.01E-02 ND Sn Unc. mg/kg initial source 1.58E+00 3.30E+00 1.36E+00 1.66E+00 1.11E+00 2.01E-01 2.47E-01 Sb mg/kg initial source ND ND 3.27E+00 ND 1.37E+00 ND ND Sb Unc. mg/kg initial source 1.67E+00 3.50E+00 1.36E+00 1.74E+00 1.18E+00 2.14E-01 2.63E-01 Ba mg/kg initial source 5.84E+00 8.96E-01 6.68E+00 6.72E+00 6.19E+00 1.94E-01 2.80E-01 Ba Unc. mg/kg initial source 1.41E+00 2.30E+00 1.55E+00 1.66E+00 1.32E+00 1.48E-01 1.73E-01 La mg/kg initial source 3.87E+00 1.31E+00 5.53E+00 6.16E+00 4.37E+00 2.00E-02 1.57E-01 La Unc. mg/kg initial source 9.68E-01 1.31E+00 1.04E+00 1.15E+00 8.72E-01 9.43E-02 1.16E-01 Hg mg/kg initial source ND ND ND ND ND ND ND Hg Unc. mg/kg initial source 1.80E+00 2.00E+00 2.04E+00 2.01E+00 1.65E+00 1.21E-01 1.65E-01 Pb mg/kg initial source 1.00E+04 8.77E+03 1.15E+04 1.11E+04 9.57E+03 5.42E+02 8.16E+02 Pb Unc. mg/kg initial source 5.02E+02 4.39E+02 5.74E+02 5.55E+02 4.79E+02 2.71E+01 4.08E+01 a Burn 1 Skid waste Burn 1 Yellow box with red text = less than three times the uncertainty level. ND = not detected. Unc. = Uncertainty level 745 Appendix A 3 Appendix B: PCDD/PCDF emission factors Table B-1. PCDD/PCDF total emission factors from skid waste. Homologue na TeCDD Total PeCDD Total HxCDD Total HpCDD Total OCDD 0 1 3 4 4 TeCDF Total PeCDF Total HxCDF Total HpCDF Total OCDF 4 3 2 2 4 PCDD Total PCDF Total PCDD/PCDF Total Skid Waste -Type 1 Average Stand. Dev.b RSDb ng/kg initial source % d ND 0.14 1.25 1.33 107 3.71 2.07 56 8.49 5.32 63 25.51 8.51 0.85 1.26 0.45 30.19 7.30 0.17 37 13.17 33.41 46.58 8.66 37.48 41.13 66 112 88 RPDc % 118 86 70 64 a 750 Number of samples with detectable levels. b Stand. Dev. = standard deviation, RSD = relative standard deviation calculated when n = 3 or more. c RPD = relative percent difference, calculated when n=2. d ND = not detected. 755 760 Appendix B 1 765 Table B-2. PCDD/PCDF TEQ emission factors from skid waste, ND = 0. Skid Waste -Type 1 Stand. Dev.b ND=0 ng TEQ/kg initial source ND Average Homologue na RSDb RPDc % % 2,3,7,8 - TCDD 0 1,2,3,7,8 - PeCDD 1 0.208 1,2,3,4,7,8 - HxCDD 0 ND 1,2,3,6,7,8 - HxCDD 1 0.037 1,2,3,7,8,9 - HxCDD 1 0.025 1,2,3,4,6,7,8 - HpCDD 4 0.025 0.015 60 1,2,3,4,6,7,8,9 - OCDD 4 0.0025 0.0016 64 2,3,7,8 - TCDF 4 0.371 0.389 105 1,2,3,7,8 - PeCDF 2 0.045 2,3,4,7,8 - PeCDF 3 0.503 1,2,3,4,7,8 - HxCDF 2 0.024 1,2,3,6,7,8 - HxCDF 1 0.017 1,2,3,7,8,9 - HxCDF 0 ND 2,3,4,6,7,8 - HxCDF 0 ND 1,2,3,4,6,7,8 - HpCDF 0 ND 1,2,3,4,7,8,9 - HpCDF 0 ND 1,2,3,4,6,7,8,9 - OCDF 3 0.000145 0.000046 31 PCDD TEQ Total 0.10 0.15 158 PCDF TEQ Total 0.79 0.71 90 PCDD/PCDF TEQ Total 0.88 0.79 90 31 0.285 57 64 a Number of samples with detectable levels. b Stand. Dev. = standard deviation, RSD = relative standard deviation calculated when n = 3 or more. c RPD = relative percent difference, calculated when n=2. d ND = not detected. 770 775 Appendix B 2 780 Table B-3. PCDD/PCDF TEQ emission factors from skid waste, ND = LOD. Skid Waste -Type 1 Homologue 2,3,7,8 - TCDD Average Stand. Dev.a ND=LODb ng TEQ/kg initial source 0.141 0.0591 RSDa % 42 1,2,3,7,8 - PeCDD 0.152 0.0393 26 1,2,3,4,7,8 - HxCDD 0.010 0.00119 12 1,2,3,6,7,8 - HxCDD 0.019 0.0124 65 1,2,3,7,8,9 - HxCDD 0.014 0.00709 49 1,2,3,4,6,7,8 - HpCDD 0.025 0.0152 60 1,2,3,4,6,7,8,9 - OCDD 0.0025 0.00163 64 2,3,7,8 - TCDF 0.371 0.389 105 1,2,3,7,8 - PeCDF 0.025 0.0244 98 2,3,4,7,8 - PeCDF 0.390 0.324 83 1,2,3,4,7,8 - HxCDF 0.017 0.0105 61 1,2,3,6,7,8 - HxCDF 0.013 0.00232 17 1,2,3,7,8,9 - HxCDF 0.014 0.000949 7.0 2,3,4,6,7,8 - HxCDF 0.012 0.000806 6.6 1,2,3,4,6,7,8 - HpCDF 0.0022 0.00132 61 1,2,3,4,7,8,9 - HpCDF 0.0026 0.00158 61 1,2,3,4,6,7,8,9 - OCDF 0.00014 0.000037 26 PCDD TEQ Total 0.36 0.10 27 PCDF TEQ Total 0.85 0.69 81 PCDD/PCDF TEQ Total 1.21 0.69 57 a Stand. Dev. = standard deviation, RSD = relative standard deviation. b ND = not detected, LOD = limit of detection. 785 790 Appendix B 3 Table B-4. PCDD/PCDF total emission factors from skid waste. Homologue na TeCDD Total PeCDD Total HxCDD Total HpCDD Total OCDD 0 1 3 4 4 TeCDF Total PeCDF Total HxCDF Total HpCDF Total OCDF 4 3 2 2 4 PCDD Total PCDF Total PCDD/PCDF Total Skid Waste -Type 1 Average Stand. Dev.b RSDb ng/kg waste % d ND 0.28 2.51 2.68 107 7.45 4.17 56 17.06 10.68 63 51.25 17.10 1.71 2.53 0.91 60.63 14.67 0.34 37 26.5 67.1 93.6 17.4 75.3 82.6 66 112 88 RPDc % 118 86 70 64 a 795 Number of samples with detectable levels. b Stand. Dev. = standard deviation, RSD = relative standard deviation calculated when n = 3 or more. c RPD = relative percent difference, calculated when n=2. d ND = not detected. 800 805 Appendix B 4 810 Table B-5. PCDD/PCDF TEQ emission factors from skid waste, ND = 0. Skid Waste -Type 1 Stand. Dev.b ND=0 ng TEQ/kg waste ND Average Homologue na 2,3,7,8 - TCDD 0 1,2,3,7,8 - PeCDD 1 1,2,3,4,7,8 - HxCDD 0 1,2,3,6,7,8 - HxCDD 1 1,2,3,7,8,9 - HxCDD 1 1,2,3,4,6,7,8 - HpCDD 4 1,2,3,4,6,7,8,9 - OCDD 4 2,3,7,8 - TCDF 4 1,2,3,7,8 - PeCDF 2 2,3,4,7,8 - PeCDF 3 1,2,3,4,7,8 - HxCDF 2 1,2,3,6,7,8 - HxCDF 1 1,2,3,7,8,9 - HxCDF 0 2,3,4,6,7,8 - HxCDF 0 1,2,3,4,6,7,8 - HpCDF 0 1,2,3,4,7,8,9 - HpCDF 0 1,2,3,4,6,7,8,9 - OCDF 3 PCDD TEQ Total PCDF TEQ Total PCDD/PCDF TEQ Total 0.417 ND 0.075 0.050 0.051 0.005 RSDb RPDc % % 0.030 0.003 60 0.781 105 64 0.745 0.091 1.011 0.049 0.033 ND ND ND ND 0.000291 0.000091 31 0.19 1.58 1.77 0.30 1.43 1.59 158 31 0.572 57 64 90 90 a Number of samples with detectable levels. b Stand. Dev. = standard deviation, RSD = relative standard deviation calculated when n = 3 or more. c RPD = relative percent difference, calculated when n=2. d ND = not detected. 815 820 Appendix B 5 Table B-6. PCDD/PCDF TEQ emission factors from skid waste, ND = LOD. Skid Waste -Type 1 Homologue 2,3,7,8 - TCDD 1,2,3,7,8 - PeCDD 1,2,3,4,7,8 - HxCDD 1,2,3,6,7,8 - HxCDD 1,2,3,7,8,9 - HxCDD 1,2,3,4,6,7,8 - HpCDD 1,2,3,4,6,7,8,9 - OCDD 2,3,7,8 - TCDF 1,2,3,7,8 - PeCDF 2,3,4,7,8 - PeCDF 1,2,3,4,7,8 - HxCDF 1,2,3,6,7,8 - HxCDF 1,2,3,7,8,9 - HxCDF 2,3,4,6,7,8 - HxCDF 1,2,3,4,6,7,8 - HpCDF 1,2,3,4,7,8,9 - HpCDF 1,2,3,4,6,7,8,9 - OCDF PCDD TEQ Total PCDF TEQ Total PCDD/PCDF TEQ Total Average Stand. Dev.a ND=LODb ng TEQ/kg waste RSDa % 0.283 0.306 0.020 0.038 0.029 0.051 0.0051 0.119 0.079 0.0024 0.025 0.014 0.030 0.0033 42 0.745 0.050 0.784 0.034 0.027 0.027 0.024 0.0043 0.0052 0.00029 0.781 0.049 0.651 0.021 0.0047 0.0019 0.0016 0.0026 0.0032 0.000075 105 0.73 1.70 2.43 0.20 1.38 1.38 27 26 12 65 49 60 64 98 83 61 17 7.0 6.6 61 61 26 81 57 a 825 Stand. Dev. = standard deviation, RSD = relative standard deviation. b ND = not detected, LOD = limit of detection. Appendix B 6