United States Environmental Protection Agency Air and Energy Engineering Research Laboratory Research Triangle Park NC 27711 Research and Development EPA/600/S2-89/054 Jan. 1990 4>EPA Project Summary Characterization of Emissions from the Simulated Open Burning of Scrap Tires Jeffrey V. Ryan Discarded automobile tires have become a serious health concern, largely because the growing number of stockpile fires has focused attention on the potentially harmful products of incomplete combustion (PICs) emitted into the atmosphere from burning scrap tires. This small- scale combustion study was designed to collect, identify, and quantify products emitted during the simulated open combustion of scrap tires. Fixed combustion gas, volatile and semi-volatile organic, participate, and airborne metals data were collected under two burn conditions that varied by the size of tire material. Burn rates, varied by material size, were used to estimate potential emissions of identified products. Total estimated emissions of semi- volatile organics ranged from 10 to 50 g/kg of tire material burned. Mono and poly aromatic hydrocarbons were the main emission products identified. Benzo(a)pyrene (BAP) In participate extracts is of particular concern because it is a known carcinogen. The presence of zinc in gaseous particulate collection was also verified and quantified. Several trends were evaluated relating emission products with burn rates. This Project Summary was developed by EPA's Air and Energy Engineering Research Laboratory, Research Triangle Park, NC, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Approximately 240 million vehicle tires are discarded annually. Although viable methods for reclamation exist, less than 25% are re-used or re-processed. The remaining 170 million scrap tires are discarded in landfills, above-ground stockpiles, or illegal dumps. Many landfills are refusing to accept tires because they present not only disposal but also health-related problems. After burial, tires often float to the surface and become partially filled with water. Cutting the tire in half or in pieces can reduce this tendency. It is very costly to cut or shred tires into a condition suitable for landfill; in any event, many sites lack the necessary equipment. Steel-belted radials which comprise the majority of the nation's discarded tires, are particularly difficult to cut and/or shred. Often, they are simply stockpiled or illegally dumped. These stockpiles and dumps can become a breeding ground for many insects, especially mosquitos, where water collects in the tires and creates an ideal breeding habitat. The introduction and spread of several mosquito species has been directly attributed to the presence of refuse tires. The growing incidence of tire fires creates another potential health hazard. More tire stockpiles and illegal dumps are coming into existence, and with them the occurrence of tire fires. These fires, often started by arson, generate a huge amount of heat making them extremely difficult to extinguish. Some of these tire fires have continued for months. For example, the Rhinehart tire fire in Winchester, Virginia, burned for nearly 9 ------- months, potentially exuding large quantities of harmful compounds. The EPA's Control Technology Center (CTC) received numerous requests from state and local agencies nationwide for information pertaining to tire fires and their effects. Because very little information on the open burning of scrap tires was available, the steering committee felt a study investigating this potential problem was warranted. Guided by the Combustion Research Branch (CRB) of EPA's Air and Energy Engineering Research Laboratory (AEERL), Acurex conducted a study which identified and quantified organic and inorganic emission products produced during the simulated open combustion of scrap tires. Experimental Approach The project consisted of a parametric study to collect organic and inorganic emissions from the simulated open combustion of scrap tires. Small quantities (10-20 Ib, 4.5-9.0 kg) of scrap- tire material were burned under two controlled conditions determined by the size of the material. The conditions were evaluated in duplicate on successive days. An existing burn hut used for similar projects was modified to accommodate this task. A separate outbuilding housed the required organic and particulate sampling equipment. CRB's Hazardous Air Pollutants Mobile Laboratory (HAPML) was used to monitor fixed combustion gases. Organics were collected using the Volatile Organic Sampling Train (VOST) and a semi- volatile collection system using XAD-2 and particulate filters. Particulate was also collected to assess airborne metals. The organic constituents were analyzed both qualitatively and quantitatively by GC/MS, GC/FID, HPLC, and gravimetric methodologies. Representative scrap truck and passenger car bias ply tires were obtained from local tire dealers. Bias ply tires were chosen due to difficulties encountered attempting to cut steel belted radial tires. The tires were cut into two sizes in order to vary initial surface area and therefore combustion rate. The "CHUNK" condition represented a quarter to a sixth of an entire tire, while tire material cut into 5.08 cm (2 in.) squares represented the "SHRED" condition. Tire material approximating the equivalent weight of two passenger car tires (30 Ibs, 13.6 kg) was placed in a small burn pit and combusted. A platform scale under the burn pit monitored the weight of tire material combusted. An air system designed to deliver nominally 1,200 cfm (34.0 ms/min) was used to simulate rapid dilution and cooling of combustion products. An insulated duct transported the gaseous sample to an adjacent sampling shed for collection of volatile and semi- volatile organics, particulate, and fixed combustion gases. Common combustion gases, 02, CO, C02, S02, and total hydrocarbons (THC) were monitored continuously throughout the burn period. Volatile organics were collected using the Volatile Organic Sampling Train (VOST). Semi-volatile organics and particulate were collected using applicable modified sampling systems. During the sampling, tire material weight differentials were recorded to determine burn rates. The VOST samples collected were analyzed by gas chromatography/mass spectrometry (GC/MS). Compounds were identified through spectral library searches and matching, along with investigator interpretation. The identified compounds were quantitated using the system response to toluene. Individual compound responses were not determined. Semi-volatile organics were collected on both particulate filters as well as the XAD-2 organic sorbent material. The organics were retrieved from the collection media by soxhlet extraction with dichloromethane. Both the particulate extracts and the XAD-2 extracts were analyzed for total chromatographable organics (TCO) and GRAV—organic compounds with boiling points of 100-300°C, and > 300°C, respectively. The XAD-2 extracts were analyzed by GC/MS to identify compounds found in the TCO range. Again, compounds were identified using library spectral matching and investigator interpretation. Identified compounds were quantitated using the same response used for TCO quantitation. Individual identified compound responses were not determined. The organic extracts were also analyzed for polycyclic aromatic hydrocarbons (PAHs). Separately collected particulate matter was also quantitated for specific metals common in tire ash residues. Data and Results The size of tire material was varied to change combustion conditions and gain insight into the mechanisms governing burn rate. Burn rates were calculated by dividing the amount of tire material burned over a given period by that time in minutes and normalizing to a mass per hour basis. A higher burn rate was observed during CHUNK tests than in SHRED tests. Roughly double the amount of tire material was combusted during the same time. In both burn conditions, an initial high burn rate was observed and led to a gradual leveling off until the burn rate was steady. Combustion gases were monitored continuously throughout the test. High emissions of CO, S02, and THC were observed at high burn rates. The identification of unknown volatile organics using the MS proved to be highly successful. More than 50 compounds were identified from the VOST samples collected. Most of these compounds are aliphatically, olefinically or acetylenically substituted aromatics. Cyclic and chained alkanes, alkenes and dienes were also identified. Several sulfonated and nitrogenated compounds were also identified in samples. Thiophene, substituted thiophenes, isocyanobenzene and benzodiazine were isolated in multiple samples. Following quantitation of identified compounds, average gaseous concentration and emissions were estimated.Of particular interest are the emission rate estimations. The estimated emissions are based on several variables and were obtained by relating the amount of tire material combusted during the sampling period with the average gaseous concentration. It was assumed that the dilution air added to the burn hut was constant. It is difficult to isolate consistent trends in the types and amounts of specific volatile organic compounds formed during the combustion of scrap tires under varied burn rates. A trend appears that, as burn rate decreases, the amount of organics potentially emitted tends to increase with respect to the amount of tire material combusted. As with the volatile organic analyses, the MS analysis of the XAD-2 extracts identified the same types of compounds. Table 1 lists the 60 compounds identified and indicates that substituted mono and poly aromatics were the predominant products of incomplete combustion. Table 2 presents the average gaseous concentration and estimated emissions of the quantitated identified compounds from the four test conditions. The estimated emissions of many compounds increase with decreased burn rate. Table 3 summarizes the semi-volatile organic emission data from the four test conditions. The total organics from respective boiling-point-based analyses are presented for each component of the sampling media. The estimated emis- ------- Table 1. Compounds Identified by GC/MS from XAD-2 Extracts Compound Identified Formula Compound Identified Formula Compound Identified Formula Methyl Benzene Cyclopentanone Dimethyl Cyclohexene Ethenyl Cyclohexene Ethyl Benzene Dimethyl Benzene Hexanenitrile Ethynyl Benzene Styrene Nonane Propenyl Cyclohexane Methylethyl Benzene Propyl Benzene Benzaldehyde Trimethyl Benzene Phenol Cyanobenzene Propenyl Benzene Methyl, Ethenyl Benzene Methyl, Methylethyl Benzene C7H8 CgH12 £-8^70 CgHg CgH20 CgH,2 C7H60 CgH72 C6H60 C7H5N Limonene Dihydro Indene Hydroxy Benzaldehyde Indene Tetramethyl Benzene Ethyl, Dimethyl Benzene Methyl Phenol Methyl Benzaldehyde Methyl, (Methylethyl) Benzene Propenyl, Methyl Benzene Undecane (Dimethylpropyl) Benzene Dimethyl, (Methylethyl) Benzene Butynyl Benzene Methyl Indene Azulene Naphthalene Benzo[b]th!ophene Benzisothiazole Hexahydro Azepinone C7H602 CgHg C7H80 CgHaO C,jH24 CgHgS C7H5NS CgH^NO Dihydro, Methyl Naphthalene Butyl,Trimethyl Benzene Methyl Naphthalene Biphenyl Dimethyl Naphthalene Dihdro Acenaphthalene Acenaphthalene (Dimethyl, Hexenyl), Methyl Benzene Pentadecane 1,1' Biphenyl,Methyl Isocyano Naphthalene Naphthalenecarboxaldehyde Propenyl Naphthalene Trimethyl Naphthalene 1H Fluorene Dimethyl Biphenyl Dibenzothiophene Phenanthrene 9H Fluorene, Methylene Phenylnaphthalene C12H12 C12H10 C12Hg C13H12 C,,H7/V C,,H80 C12H10S C14H-,Q C 14^10 sions data presented show that 12-50 g of semi-volatile organics can be emitted for every kilogram of tire burned. A trend does seem evident when relating the amount of organics emitted to burn rates. It appears that, as burn rate decreases, the amount of organics emitted, particularly in the TCO range, increases. There did not seem to be any significant variation in GRAV range organic emissions when related to burn rate. A PAH analysis was performed on the XAD- 2 and particulate filter extractions. Of particular importance is the presence of benzo(a)pyrene (BAP) in all particulate portions. BAP is a reported carcinogen and a highly scrutinized compound when evaluating combustion processes. A separate particulate collection system was operated to verify and quantitate the presence of metals collected from ambient emissions. The results of the analyses were marginal for many of the targeted metals. Many of the results were at or near instrument detection levels. It did seem evident that lead and particularly zinc were found to be collected from gaseous emissions. Conclusions The primary goal of this project was to characterize potentially harmful emissions from the simulated open burning of scrap tires. The simulation was necessarily crude, because it woujd be extremely difficult to match the burning of the equivalent of two tires with a 6 million tire, full-scale, stockpile fire. Neverthe- less, the study allowed the investigators to identify and measure gaseous emissions and directly relate this information to a mass burn rate. This task was accomplished by accurately measuring dilution volumes, sample volumes, and weights of tire material combusted. It is not known how well the concentra- tions obtained from this study represent those at an actual tire fire. The dilution air added to the burn hut was used not only to control known volumes introduced, but also to simulate ambient condition's.The same types of compounds identified during this study are probably emitted during an actual fire, but whether the average gaseous concentrations and estimated emissions are comparable is uncertain. A comparison with limited data collected at the Winchester, Virginia, fire by NIOSH, indicates that reasonable agreement exists within several measurement areas. Many of the same compounds were identified in actual plume samples. Particularly good agreement exists in PAH plume measurements. NIOSH reported ambient concentrations of total PAHs are gener- ally within the same order of magnitude as average gaseous concentrations obtained during testing. Measurements of CO and metals also indicate similar agreement. Both the lead and zinc measurements show similar values both in concentration and relative concentration between the two metals. It may be reasonable to assume that the estimates obtained during this study may be within an order of magnitude of emissions realized from actual stock-pile fires. The results of this study pose a variety of pertinent topics and questions regarding tire fires. How far does the particulate from a stockpile fire carry? Are evacuation procedures for populations near stockpile fires sufficient? Is it good policy to continue to let tire fires burn themselves out? A greater potential for harmful organic emissions seems to exist at lower burn rates; thus a smoldering tire fire may be more harmful than one that is burning out of control. The identification of significant quantities of benzo(a)pyrene in the particulate extracts warrants serious concern. High emissions of other noxious compounds, particularly benzene, with concentrations often exceeding 1 ppm, suggest that uncontrolled scrap tire combustion poses significant health risks. ------- Table 2, Quantitation and Emission Summary of Compounds Identified in Day 1 CHUNK Condition Vol Samp = 12,98 m3 Burn Rate =2.3 kg/h Compound Identified Methyl Benzene Ethyl Benzene Dimethyl Benzene Ethynyl Benzene Styrene Methylethyl Benzene Propyl Benzene Benzatdehyde Trimethyl Benzene Phenol Cyanobenzene Trimethyl Benzene Methyl, Methylethyl Benzene Limonene Indene Tetramethyl Benzene Ethyl, Dimethyl Benzene Methyl Benzatdehyde Ethyl, DimeVtyl Benzene Propenyl, Methyl Benzene Methyl Indene Methyl Indene Naphthalene Benzo[B]thiophene Benzisothiazole Hexahydro Azepinone 2-Methyl Naphthalene 1 -Methyl Naphthalene Biphenyl Dimethyl Naphthalene Acenaphthalene 1,1' Biphenyl, Methyl Isocyano Naphthalene Propenyl Naphthalene Trimethyl Naphthalene 1H Fluorene Phenanthrene Totals Average Gaseous Concentra- tion (mg/m3) 0,716 0.074 0.000 0,185 0.419 0,000 0.000 0.111 0.000 0.190 0.095 0.127 0.000 0.000 0.325 0.000 0,000 0,000 0.000 0.000 0.000 0.000 1,230 0.099 0.000 0.000 0.164 0.124 0.083 0.000 0.390 0.025 0.021 0.053 0,000 0.087 0,152 4,670 Estimated Emissions (mg/kg TIRE) 634.8 65.8 0.0 163.6 371.5 0,0 0.0 98.6 0.0 168.3 84.3 112.8 0.0 0.0 288.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1,090.4 87.6 0.0 0.0 145.6 109.9 73.5 0.0 345.6 22,1 18.7 46.6 0.0 76,9 135.0 4,140.4 XAD-2 Extracts' •* Day 2 CHUNK Condition Vcrf Samp = 12.76 mg/m3 Burn Rate = 1.7 kg/h Average Gaseous Concentra- tion (mg/m3) 1.500 0.289 0.513 0.325 0.795 0.081 0.000 0.327 0.000 0.473 0.305 0.256 0.215 0.094 0.602 0.000 0,000 0,000 0.000 0.000 0.188 0.000 1.931 0,000 0.000 0.126 0.466 0.463 0.290 0.069 0.773 0.000 0.000 0.000 0.000 0.288 0.194 12,672.8 Estimated Emissions (mg/kg TIRE) 1,799.5 347.1 615.3 390.0 953.3 97.4 0.0 392.1 0.0 566.9 365.5 307.6 258.1 113.2 722.0 0.0 0.0 0.0 0.0 0.0 225.6 0.0 2,315.9 0.0 0.0 151.4 559.4 555.3 347.8 83.0 927.0 0.0 00 0.0 0.0 346.0 233.2 9.367 Day 1 SHRED Condition Vol Samp = 13.77 mg/m3 Burn Rate = 1.1 kg/h Average Gaseous Concentra- tion (mg/m3) 0.731 0.192 0.554 0.078 0.332 0.142 0.136 0.343 0.156 0.360 0.307 0.195 0.926 1.402 0.156 0.094 0.000 0,089 0.162 0.537 0.223 0.000 0.516 0.000 0.179 0.345 0.202 0.122 0.180 0.183 0.217 0.000 0.000 0.000 0.169 0,141 0.000 17,362.2 Estimated Emissions (mg/kg TIRE) 1,354.9 355.0 1,026.1 144.2 614.6 263.5 251.2 635.3 289.9 667.2 569,3 361.5 1,716.1 2,599.2 289.9 174.8 0.0 164.7 300.0 996,2 413.7 0.0 957.2 0.0 330.9 638.9 373.9 225.7 333.6 338.8 402.2 0.0 0.0 0.0 312.9 260.8 0.0 9.633 Day 2 SHRED Condition Vol Samp = 12.43 mg/m3 Burn Rate = 1.3 kg/h Average Gaseous Concentra- tion (mg/m3) 0.911 0.203 0.532 0.150 0.433 0.122 0.118 0.000 0.351 0.470 0.292 0.199 0.704 1.316 0.251 0.000 0.154 0.000 0.000 0.000 0.114 0.132 0.843 0.000 0.000 0.555 0.313 0.146 0.208 0.000 0.430 0.000 0.000 0.000 0.203 0.230 0.251 15,108.7 Estimated Emissions (mg/kg TIRE) 1,429,0 318.3 834.2 235,2 679.8 191.0 184.6 0.0 549.9 736.8 458.6 312.2 1,104.9 2,064.5 393.9 0.0 241,8 0.0 0.0 0.0 179.2 207,3 1,322.8 0.0 0,0 869.9 490.5 229,7 326.3 0.0 673,8 0.0 0.0 0.0 319.1 361.2 394.2 10,566 1 Concentrations determined using system response to TCO calibration mix. 2 Average gaseous concentrations and estimated emissions are based on controlled dilution of pit emissions. It is not known how well this dilution represents ambient air exchange under actual conditions. ------- Table 3. Organic Emission Summary Average Gaseous Concentration Sample Portion Sample Time (min) Weight Change (kg) Avg Rate Bum (kglh) Volume Samp (m3) TCO (mg/m3) GRAV (mglm3) Estimated Emissions TCO (mglkg GRAV TIRE) XAD-2 Filter 1 Filter 2 Filter Total KAD-2 Filter 1 Filter 2 Filter Total XAO-2 Filter 140 11 129 140 152 42 110 152 182 182 5.4 1.7 3.7 5.4 4.3 2.4 1.8 4.3 3.2 3.2 2,3 9,4 1.7 2.3 1.7 3.5 1.0 1.7 1.1 1,1 Day 1 CHUNK Condition 12.98 0.93 12.05 12.98 6.067 0.043 0.066 0.065 Day 2 CHUNK Condition 12.78 3.3 9.45 12.76 11.398 0,018 0.53 0.398 Day 1 SHRED Condition 13.77 13.77 20.658 0.141 Day 2 SHRED Condition 0.698 33.634 •J.S23 6,888 Total 0,749 18.498 10.219 12.359 Total 0.89 6.822 Total 5,379 9 79 58 5,436 13,671 11 1,081 477 14,148 38,292 261 38,554 619 7,298 5,785 6,106 6,725 898 10,776 20,837 14,824 15,722 1,SSO 12,646 14,295 12,161 25,870 52,849 XAD-2 Filter 183 183 4.0 4.0 1.3 1.3 12.43 12.43 15.65 0,173 1,099 8.77? Total 24,546 271 24,819 1,724 13,757 15,481 40,299 Results from the airborne metals portion of the study were inconclusive. Maximum values were presented, often based on detection levels. Emissions of lead and zinc may reach significant quantities. Reported chemical analysis of tire ash residues reveals that zinc comprises nearly 50 percent of the total residue. Evidently, the other metals known to be contained in tires remain in the ash residue. Although no attempt was made to analyze ash residue, significant quantities of metals present in the ash could potentially be leached out into groundwater systems, posing another major problem. The values obtained by the on-line analyzers for normal combustion gases showed that as burn rate increased, the amount of CO, SO2, and unburned hydrocarbons also increased. High burn rate conditions were not fully evaluated, so greater quantities of these gases, particularly SOg, may be emitted during a stockpile fire. Tires contain a significant amount of sulfur, so high emissions of SOa, while likely only a minor contributor to the acid rain problem, could have significant local consequences. This study was designed to identify the potential chemical hazards from tire fires on a small-scale, simulation basis. The study reveals the potential for the emission of great amounts of organic compounds, primarily aromatics, some of which may be extremely harmful. Although the estimates of average gaseous concentrations and emissions are crude, the trends presented for burn rate may be helpful in directing further research and control efforts. That the "SHRED" condition resulted in a lower burn rate indicates that the gaps between the tire material provide the major avenue of oxygen transport. Oxygen transport appears to be a major if not the con- trolling mechanism for sustaining the combustion process. This fact could have advantageous implications for those attempting to combat tire fires. It may be possible to fill the gaps between tires with a foam inhibitor, potentially suffo- cating the fire from within. ------- J. V. Ryan is with Acurex Corp., Research Triangle Park, NC 27709. Paul lit Lemieux is the EPA Project Officer (see below). The complete report, entitled "Characterization of Emissions from the Simulated Open Burning of Scrap Tires," (Order No. PB 90-126 0041 AS; Cost: $17.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Air and Energy Engineering Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC 27711 United States Center for Environmental Research Environmental Protection Information Agency Cincinnati OH 45268 Official Business Penalty for Private Use $300 EPA/600/S2-89/054 ------- |