Combustion Products of Plastics as Indicators for Refuse Burning in the Atmosphere

B E R N D R . T . S I M O N E I T , * ,, P A T R I C I A M . M E D E I R O S , , A N D BORYS M. DIDYK Environmental and Petroleum Research Group, College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331, Environmental Sciences Graduate Program, Oregon State University, Corvallis, Oregon 97331, and Refiner a Aconcagua, ENAP Refiner as SA, Avenida Borgon o, 25777 Conco n, Chile

Despite all of the economic problems and environmental discussions on the dangers and hazards of plastic materials, plastic production worldwide is growing at a rate of about 5% per year. Increasing techniques for recycling polymeric materials have been developed during the last few years; however, a large fraction of plastics are still being discarded in landfills or subjected to intentional or incidental open-fire burning. To identify specific tracer compounds generated during such open-fire combustion, both smoke particles from burning and plastic materials from shopping bags, roadside trash, and landfill garbage were extracted for gas chromatography-mass spectrometry analyses. Samples were collected in Conco n, Chile, an area frequently affected by wildfire incidents and garbage burning, and the United States for comparison. Atmospheric samples from various aerosol sampling programs are also presented as supportive data. The major components of plastic extracts were even-carbon-chain n-alkanes (C16C40), the plasticizer di-2-ethylhexyl phthalate, and the antioxidants and lubricants/antiadhesives Irganox 1076, Irgafos 168, and its oxidation product tris(2,4-di-tertbutylphenyl)phosphate. Major compounds in smoke from burning plastics include the non-source-specific n-alkanes (mainly even predominance), terephthalic acid, phthalates, and 4-hydroxybenzoic acid, with minor amounts of polycyclic aromatic hydrocarbons (including triphenylbenzenes) and tris(2,4-di-tert-butylphenyl)phosphate. 1,3,5Triphenylbenzene and tris(2,4-di-tert-butylphenyl)phosphate were found in detectable amounts in atmospheric samples where plastics and refuse were burned in open fires, and thus we propose these two compounds as specific tracers for the open-burning of plastics.

plastics has reached levels of 150 million tons per year, and plastic consumption is growing a rate of approximately 5% annually (1). Plastics are employed in a very broad range of applications such as films, wrapping materials, shopping and garbage bags, fluid containers, clothing, toys, household and industrial products, and building materials (1, 2). Although increasing volumes of polymeric materials are currently being recycled (2-4), a fraction of the plastic materials used worldwide is discarded and becomes incorporated into refuse or urban and roadside litter (3). Thus plastic materials have become ubiquitous components of urban litter and domestic trash, constituting a significant proportion of the rejected materials disposed of by society (4). Plastics are readily combustible and under open-fire conditions generate black smoke, decomposition, and volatilization products, which become incorporated into the ambient environment (5-9), resulting in human and environmental exposure. Garbage, industrial, and urban refuse removal and their disposal are regulated to various degrees in different parts of the world (2-4). Current practices range from recycling and controlled landfill disposal to intentional or incidental open-fire burning. When garbage or refuse is burned under open-fire conditions, the plastics that they contain contribute to smoke generation and inject plastic decomposition compounds into the smoke (3, 6-8). The purpose of this work is to identify specific tracer compounds generated during open-fire plastic combustion to be used as plastic combustion markers and as source indicators for garbage and refuse burning. For this work, a particular location was selected in Conco n, Chile, where wind-blown plastic litter accumulates on bushes and roadsides, particularly in the vicinity of a local landfill (La Jarilla). These are areas frequently affected by wildfire incidents and intentional or incidental garbage burning. In the immediate vicinity of the La Jarilla landfill, material recycling activities are undertaken, such as metals and electrical cable recoveries, which involve accumulation of the recovered materials and elimination of the plastic insulation covers by open-fire combustion. For comparison, a composite sample of plastic shopping bags from the United States was also extracted and burned.

Background Information
Types of Plastic Materials Found in Garbage and Trash. Plastic materials are widely used, and their broad applicability in consumer goods, household materials, packaging, and bag materials results in the incorporation of plastic refuse into garbage and litter. The principal types of plastic materials found in refuse are high density polyethylene (HDPE), used in trash bags, milk jugs, and shopping bags, low density polyethylene (LDPE), used for bags, food wrap, and plastic film, vinyl and poly(vinyl chloride) (PVC), used for bottles, packaging, and containers, poly(ethylene terephthalate) (PETE), used predominantly in beverage bottles and similar containers, polystyrene (PS), the spongy, light material used in supermarkets for meat, egg, and miscellaneous product trays, hot beverage cups, and thermally insulated take-home boxes, and polypropylene (PP), used for yogurt containers, straws, diapers, wrapping films, margarine tubs, and special bags. All of these plastics contact food and other products in a predominantly single-use event, after which they are discarded as garbage and/or litter. By far the major environmental impact is generated by plastic bags, made

Introduction
Plastics are versatile polymeric materials produced on a massive scale and used worldwide. Global production of
* Corresponding author phone: (541)737-2155; fax: (541)737-2064; e-mail: simoneit@coas.oregonstate.edu. Environmental and Petroleum Research Group, Oregon State University. Environmental Sciences Graduate Program, Oregon State University. ENAP Refiner as SA.

TABLE 1. Plastic Samples Collected for Extraction and Open-Fire Self-Combustion Source Testsa
extracts sample type/locale 1. shopping and garbage bags, unused (Chile) 2. Route 60-Ch, roadside plastic trash (Chile) 3. La Jarilla landfill, plastics from garbage (Chile) 4. shopping bags, unused (USA) 5. shopping bags, unused (China) 6. shopping bags, unused (Brazil) plastic formulation/mixture PE PE 17.3%, PET 29.7%, PVC 39.3%, PS 2.9%, unidentified plastics 10.8% 3-12% of garbage is plastic, hand-picked PE, PET, PS, and PVC (used same proportion as 2) PE PE PE burn testsb sample type/locale 1. PE (Chile) 2. PE, PET, PVC, PS (Chile) 3. La Jarilla 3.2% of total garbage is plastic, burned same proportion as 2 (Chile) 4. PE (USA)
a

extract yield (mg hexane-soluble g-1 polymer) 43.1 40.3 28.8 35.3 12.8 39.4

plastic formulation/mixture 41% combusted, 59% residue 79.5% combusted, 20.5% residue 85% combusted, 15% residue 27% combusted, 73% residue

extract yield (mg hexane-soluble g-1 PM) 17.5 5.4 9.1 51.2

elemental carbonc (mg g-1 PM) 68.1 22.7 60.7 6.9

PE ) polyethylene (high density), PET ) polyethylene terephthalate, PVC ) polyvinyl chloride, PS ) polystyrene, PM ) particulate matter, equivalent to total suspended particles (TSPs). b Extract yield is from particulate matter in smoke from 1 g of plastic burned. c Elemental carbon (EC or black carbon) determined commercially (ref 28).

predominantly of polyethylene (HDPE and LDPE) with global usage somewhere between 500 and 1000 billion new shopping bags per year, with countries such as Australia using 6.9 billion shopping bags annually (10) and the U.K. using ca. 17 billion bags per year (11). Introduced in the 1970s, plastic bags have overtaken the shopping market, and today almost everyone carries shopping and food in plastic trays, containers, and bags. All of these plastic packaging materials, after occasional reuse, get discarded in trash, garbage, or litter. Even if recycled, most of the recovered plastic winds up in trash as garbage bags or other disposable plastic materials (11). Plastics Content of Trash and the Evolution of Plastic Usage by Type and Volumes. Plastic materials have been found in all contemporary garbage and litter composition studies and represent the plastic use and disposal patterns of the society. Plastic content in garbage has been found to vary seasonally and with the area where the refuse is generated. The plastic content of garbage is lower in the summer 6.4% (w/w), increasing to 12% (w/w) during the winter holiday season in Calgary, Alberta (12). Yearly average plastic contents of 9.3-10.4% (w/w) and broad compositions of plastic types, where plastic bags and plastic film represent from 47% to 51% of the total plastic content, have been reported for cities such as Seattle, WA, and Vancouver, BC, Canada, respectively (13, 14). Litter has a similar compositional variability with total plastic contents ranging from ca. 6% to 11% (w/w) but with higher recyclable plastic levels (2.3-3.1% w/w) than those of garbage (0.6-1.1% w/w) (12, 15). Compounds and Additives Used in Plastic Materials Manufacture. Plastics are manufactured by polymerization, polycondensation, or polyaddition reactions where monomeric molecules are joined sequentially under controlled conditions to produce high-molecular-weight polymers whose basic properties are defined by their composition, molecular weight distribution, and their degree of branching or cross-linking. To control the polymerization process, a broad range of structurally specific proprietary chemical compounds is used for polymerization initiation, breaking, and cross-linking reactions (peroxides, Ziegler-Natta, and metallocene catalysts). The polymerized materials are admixed with proprietary antioxidants (sterically hindered

phenols, organophosphites), UV and light stability improvers (hindered amines and piperidyl esters), antistatic agents (ethoxylated amines), impact modifiers (methacrylatebutadiene-styrene compounds), heat stabilizers (methyl tin mercaptides), lubricants (esters), biostabilizers (arsine, thiazoline, and phenol compounds), and plasticizers used to modify the plasticity, softness, and pliability of plastics (phthalates and esters). World production of plastic additives is on the order of 18 billion pounds per year with plasticizers representing a 60% of the total amount (16). The plastic material itself is environmentally quite stable, but the additives, their reaction and degradation products incorporated into the polymeric material, can be released into the environment as well as into the contacting fluids, products, or food (17, 18). The main environmental concerns associated with additives used for plastics are related to their (a) potential ecotoxic effects, (b) mobility under conditions of use, (c) capacity to accumulate in the environment or bioaccumulate in organisms, and (d) generation or release of hazardous substances during disposal procedures or under normal geo-environmental conditions. Additives are released from plastics by leaching and contact transference (17). Diethylhexyl phthalate (DEHP) is a compound of special concern among the plasticizers used in plastic manufacture because it has been described as a probable human carcinogen by the U. S. Environmental Protection Agency, which has set exposure limits for DEHP in water (19-22). DEHP has also been described as a potential endocrine disruptor to numerous organisms (23-25) and is believed to be harmful by inhalation, generating possible health risks and irreversible effects (22, 26, 27). Atmospheric Source Models. Atmospheric source correlation models do not include garbage burning contributions due to the lack of proven source indicators or markers. Plastic combustion under open-fire conditions can generate smoke and airborne particulate matter containing plastic depolymerization products, volatilized additives, and their decomposition products, generating an additional source of airborne contaminants. These have so far not been quantified or incorporated in source-contaminant correlations in ambient aerosol emission inventories.

FIGURE 1. GC-MS data for the total extracts from plastics: (a) total ion current (TIC) trace of extract from new polyethylene bags (Chile), (b) m/z 85, key ion for n-alkanes of extract from roadside plastic trash, (c) TIC trace of extract from landfill plastic, (d) TIC trace of extract (trimethylsilyl (TMS) derivatives) from landfill plastics, (e) TIC trace of total extract from new polyethylene bags (USA), and (f) TIC trace of total extract from new polyethylene bags (Brazil). Numbers refer to the carbon-chain lengths of the alkanes: b ) n-alkane, i:1 ) olefin, 4P ) dibutyl phthalate. Roman numerals refer to compounds in Chart 1: I ) diethylhexyl phthalate, II ) octadecyl 3-(3,5-di-tert-butyl4-hydroxyphenyl)propionate (Irganox 1076), III ) tris(2,4-di-tert-butylphenyl)phosphite (Irgafos 168), IV ) tris(2,4-di-tert-butylphenyl)phosphate. Other compounds are labeled.

Experimental Methods
Aerosol Samples. Atmospheric particles were collected on quartz fiber filters using high-volume air samplers during various urban and regional aerosol sampling programs (e.g., ACE-Asia in Gosan, South Korea, and Sapporo, Japan, and Municipal Air Quality Monitoring in Santiago, Chile). The Gosan and Sapporo samples are total suspended particles (TSPs) collected with high-volume air samplers (Kimoto AS 810), and the Santiago samples are PM10 collected using an Anderson high-volume sampler. The organic compound compositions have been reported (e.g., refs 5 and 9), and key examples are shown here as ancillary and supportive data. Aerosols from several locales were examined to detect the plastic combustion markers and to assess continental similarities. Source Test Samples. Two environmentally relevant types of refuse were sampled by hand-picking for plastic materials: roadside trash and landfill garbage. The types of plastic products were identified and quantified based on the plastic identification code embossed or imprinted on the recovered piece or alternatively identified as a fragment from a larger

piece. In both cases, the most ubiquitous plastic materials were plastic bags made of polyethylene (PE). Unused plastic grocery bags from numerous countries were also analyzed. This limited survey is adequate for this study because global industry uses generally the same precursors and additives, which are technology-controlled. The plastic samples (Table 1) were shredded, extracted with n-hexane in a Soxhlet apparatus, and blown down with purified nitrogen to constant weight to obtain the n-hexane-soluble compounds present in the plastic material. Smoke samples (Table 1) were taken by high-volume filtration of smoke from low-temperature, open-fire combustion of the plastics. The smoke was generated by placing ca. 20 g of the shredded plastic sample on a precleaned stainless steel tray, igniting the plastic with a methane torch and leaving it to burn down until the fire went out. After the fire was started, the black smoke generated by the selfcombustion was collected on a precleaned, preweighed quartz fiber filter with a Hi-Vol air sampler. The filters were dried to a constant weight in a desiccator, the collection area was measured, and the mass of the particles was determined

TABLE 2. Extractable Organic Compounds from Plastics and Tracers in Smoke from the Burning of Plasticsa
plastic surface extractsb Chile compounds hexadecane heptadecane octadecane nonadecane eicosane heneicosane docosane tricosane tetracosane pentacosane hexacosane heptacosane octacosane nonacosane triacontane hentriacontane dotriacontane tritriacontane tetratriacontane pentatriacontane hexatriacontane heptatriacontane octatriacontane nonatriacontane tetracontane CPId tridecanal tetradecanal pentadecanal hexadecanal heptadecanal octadecanal nonadecanal eicosanal heneicosanal docosanal tricosanal tetracosanal pentacosanal hexacosanal heptacosanal octacosanal nonacosanal triacontanal hentriacontanal dotriacontanal tritriacontanal dibutyl phthalatese di(ethylhexyl) phthalate Irganox 1076 Irgafos 168 tris(2,4-di-tert-butyl-phenyl)phosphate di(ethylhexyl) adipate 1,2,4-triphenylbenzene 1,3,5-triphenylbenzene anthracene phenanthrene fluoranthene pyrene benzo[ghi]fluoranthene cyclopenta[cd]pyrene benz[a]anthracene chrysene benzo[j and k]fluoranthenes composition C16H34 C17H36 C18H38 C19H40 C20H42 C21H44 C22H46 C23H48 C24H50 C25H52 C26H54 C27H56 C28H58 C29H60 C30H62 C31H64 C32H66 C33H68 C34H70 C35H72 C36H74 C37H76 C38H78 C39H80 C40H82 bags new roadside litter landfill trash 12.4 7.6 49.5 9.2 83.1 7.2 116.9 8.1 154.5 10.4 159.6 9.0 157.8 18.5 145.1 16.1 115.2 2.8 75.8 nd 53.0 nd 36.6 nd 12.2 0.10 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd USA bags new 89.3 nd 140.1 nd 123.2 nd 87.2 nd 56.3 nd 45.9 nd 36.4 nd 28.1 nd 21.5 nd 15.6 nd 14.8 nd 7.2 nd 3.1 bags new 29.2 8.7 8.1 6.8 9.2 5.8 10.6 8.9 21.5 nd 22.4 nd nd nd 81.7 46.4 nd nd 28.2 15.8 27.5 13.2 10.5 5.3 2.6 0.74 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 47.0 1801.0 nd nd 33.9 nd 0.8 63.2 1.1 19.8 17.3 10.4 16.1 3.0 37.4 62.8 316.9 open burn smoke PMc Chile roadside litter 30.5 5.8 5.0 6.5 6.5 4.9 12.3 13.9 20.8 24.4 75.4 37.0 32.7 18.6 103.1 22.4 49.3 37.1 31.8 22.5 11.8 nd nd nd nd 0.75 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 158.9 4851.5 nd nd 9.3 67.6 33.9 208.3 1.5 14.0 27.8 16.6 48.0 5.2 65.9 128.7 360.8 landfill trash 28.7 8.4 7.1 13.5 14.2 12.6 16.9 13.2 12.6 13.5 41.8 33.2 68.6 42.0 127.6 89.3 93.7 59.9 75.3 37.9 43.8 25.2 12.6 3.8 3.6 0.97 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 39.0 2255.1 nd nd 174.1 28.5 nd 56.8 0.9 20.2 41.7 35.4 49.5 5.1 41.5 66.3 206.2 USA bags new nd nd nd nd 65.0 37.1 66.1 35.3 55.2 24.7 47.6 22.7 71.5 27.2 68.4 53.4 96.3 76.4 100.2 33.7 55.8 nd 37.8 nd 21.0 0.68 16.2 22.6 28.0 31.1 39.3 39.7 38.8 31.7 23.5 22.3 20.0 17.0 21.4 21.3 20.5 21.0 37.6 36.0 35.4 32.4 23.4 2.4 1.6 nd nd 126.0 nd nd 0.2 nd nd 1.8 2.0 nd nd nd nd nd

n-alkanes 16.9 1.1 nd 1.7 32.7 9.6 nd 12.2 55.9 50.5 nd 18.6 74.7 74.2 nd 15.4 93.5 117.1 nd 14.9 118.4 113.7 nd 19.0 110.8 112.1 nd 121.2 101.5 124.8 nd 39.3 80.4 112.0 nd 15.3 55.9 93.1 nd 4.1 29.6 70.5 nd nd 12.0 43.6 nd nd 5.2 21.9 0.47
nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

C13H26O C14H28O C15H30O C16H32O C17H34O C18H36O C19H38O C20H40O C21H42O C22H44O C23H46O C24H48O C25H50O C26H52O C27H54O C28H56O C29H58O C30H60O C31H62O C32H64O C33H66O C16H22O4 C24H38O4 C35H62O3 C42H63O3P C42H63O4P C22H42O4 C24H18 C24H18 C14H10 C14H10 C16H10 C16H10 C18H10 C18H10 C18H12 C18H12 C20H12

n-Alkanals nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 3.2 nd nd nd 819.1 nd nd nd nd nd nd nd nd nd nd nd nd

Plasticizers/Antioxidants nd nd 24.2 203.0 2164.7 230.9 1.7 nd nd 255.8 nd nd 90.6 1795.7 711.4 nd nd nd

Polycyclic Aromatic Hydrocarbons nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd

TABLE 2 (Continued)a
plastic surface extractsb Chile compounds benzo[e]pyrene benzo[a]pyrene perylene indenopyrenes benzo[ghi]perylene coronene 4-hydroxybenzoic acid 1,2-benzenedicarboxylic acid 1,3-benzenedicarboxylic acid 1,4-benzenedicarboxylic acid juvabione manool squalene palmitic acid methyl palmitate stearic acid methyl stearate 1-monopalmitin 1-monostearin cholesterol 7-hydroxycholesterol levoglucosan R- + -glucose sucrose erucamide total compounds
a
1/ 2[(oddC17-C39)/(evenC18-C40)

open burn smoke PMc USA Chile bags new 105.4 42.9 8.5 239.7 90.3 17.5 3.0 nd 2.8 907.1 nd nd nd 6.0 15.6 4.6 25.5 nd nd nd nd 28.1 7.9 4.2 nd 4302.2 roadside litter 116.3 70.5 12.5 173.3 89.8 23.2 47.8 nd 35.4 5032.6 407.1 174.0 nd 21.1 8.1 5.9 10.2 nd nd nd nd 23.9 12.4 94.3 nd 12928.4 landfill trash 64.1 35.5 6.2 100.3 51.0 10.2 3.3 nd 4.8 176.3 6.3 46.7 nd 35.2 6.7 64.0 15.0 4.1 6.3 6.2 nd 27.6 48.4 35.1 nd 4672.8 USA bags new nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 1708.8 bags new

composition C20H12 C20H12 C20H12 C22H12 C22H12 C24H12

bags new

roadside litter

landfill trash

Polycyclic Aromatic Hydrocarbons, Continued nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 12.9 nd nd nd nd nd nd nd nd nd nd nd nd 1351.4 Other Compounds nd nd nd nd nd nd nd nd nd nd nd nd nd 34.9 3.4 2.0 12.5 16.9 nd nd 27.7 26.9 nd nd nd nd nd 57.4 nd 0.9 nd nd nd nd nd nd nd nd 5209.8 2366.4 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 72.9 1563.7

C7H6O3 C8H6O4 C8H6O4 C8H6O4 C16H26O3 C20H34O C30H50 C16H32O2 C17H34O2 C18H36O2 C19H38O2 C19H38O4 C21H42O4 C27H46O C27H46O2 C6H10O5 C6H12O6 C12H22O11 C22H43NO

nd ) not detected.

Concentrations are g g-1 plastic. c Concentrations are ng mg-1 smoke particles. d CPI (carbon preference index) ) + (oddC17-C39)/(evenC16-C38)]. en-Butyl and iso-butyl phthalates are given as a total.

gravimetrically. After combustion, the remaining matter was weighed to establish the uncombusted plastic residue. Aliquots of filters were analyzed for elemental carbon by a commercial laboratory using the laser combustion method (28). Extraction and Derivatization. Various samples of plastics were extracted with hexane for analysis of the surface organic matter. The samples of smoke and aerosol filter aliquots were sonicated three times for 15 min each with dichloromethane/ methanol (2:1; v/v). The solvent extract was filtered through quartz wool packed in a Pasteur pipet, concentrated by use of a rotary evaporator, and then blown down with dry nitrogen gas. Total extracts were analyzed directly, and aliquots were reacted with N,O-bis-(trimethylsilyl)trifluoroacetamide (BSTFA) with 1% trimethylsilyl chloride and pyridine for 3 h at 70 C to derivatize polar compounds. This procedure derivatizes COOH and OH groups to the corresponding trimethylsilyl (TMS) esters and ethers, respectively. Gas Chromatography-Mass Spectrometry. Gas chromatography-mass spectrometry (GC-MS) analyses of the total extracts and derivatized aliquots were performed on a Hewlett-Packard model 6890 GC coupled to a HewlettPackard model 5973 mass selective detector (MSD). Separation was achieved on a fused silica capillary column coated with DB-5 (30 m 0.25 mm i.d., 0.25-m film thickness). The GC operating conditions were as follows: The temperature was held at 50 C for 2 min and increased from 50 to 300 C at a rate of 6 C min-1 with a final isothermal hold at 300 C for 20 min. Helium was used as the carrier gas. The sample was injected splitless with the injector temperature at 300 C. The silylated extracts were diluted (1:1) with n-hexane prior to injection. The mass spectrometer was operated in the electron impact mode at 70 eV and scanned

from 50 to 650 Da. Data were acquired and processed with Chemstation software. Individual compounds were identified by comparison of mass spectra with literature and library data, comparison with authentic standards, and interpretation of mass spectrometric fragmentation patterns. GC-MS response factors were determined using authentic standards. 1,3,5-Triphenylbenzene, octadecyl 3-(3,5-di-tert-butyl-4hydroxyphenyl)propionate (Irganox 1076, also known as octadecyl 3,5-di-tert-butyl-4-hydroxyhydrocinnamate), and tris(2,4-di-tert-butylphenyl)phosphite (Irgafos 168) are available from Sigma-Aldrich, Co. The standard tris(2,4-di-tertbutylphenyl)phosphate was prepared from the antioxidant precursor (Irgafos 168) by reaction with hydrogen peroxide (15%) followed by extraction into dichloromethane for GCMS analysis (yield 100% conversion). A comment should be made about the MS characteristics of the four phthalates with molecular weight 390 da (C24H38O4) known to us. Di-(2-ethylhexyl) phthalate (DEHP, also known commercially as DOP, octoil, and dioctyl phthalate) has a characteristic mass spectrum: M+ 390 (0.5), m/z 279 (17), 167 (39), 150 (12), 149 (100), 113 (9), 71 (17), 57 (24). (Note that the ions are listed as m/z with relative intensities following in parentheses.) All four isomers have the base peak at m/z 149 (C8H5O3), but the distinguishing feature is the m/z 167 ion (C8H7O4) for DEHP, which is much lower in intensity for the other isomers. The mass spectra of the others can be summarized as follows: di-n-octyl phthalate M+ 390 (0.5), m/z 279 (12), 261 (2), 167 (1.5), 150 (10), 149 (100), 71 (5), 57 (7); decyl hexyl phthalate M+ 390 (0.5), m/z 307 (5), 251 (10), 233 (2), 167 (1.5), 150 (10), 149 (100), 69 (2), 55 (5); di-iso-octyl phthalate (Genomoil 100) M+ 390 (0.5), m/z 307 (14), 167 (2), 150 (11), 149 (100), 71 (3), 57 (8).

CHART 1. Chemical Structures Cited

Results and Discussion

The plastic materials collected for both extraction and combustion source testing are described in Table 1. The hexane-soluble extracts ranged from 29 to 43 mg g-1 plastic polymer, where the highest value is for new polyethylene. This is as expected; new plastic has not lost the volatiles as is the case for plastic exposed to sun heat or contact desorption. The smoke particles (PM) from open-burning of these plastics contained extractable compounds ranging from 5.4 to 17.5 mg g-1 of particulate matter, and the unburned residues (ash) varied from 15% to 59%. Polyethylene had the highest PM extract and residue. Elemental carbon (EC or black carbon) emissions varied depending on burn temperature and ranged from 7 to 68 mg g-1 of particulate matter (Table 1). Organic Tracers of Plastics. The GC-MS data for the hexane extracts of the plastics analyzed are shown in Figure 1, and the organic compounds identified are listed in Table 2. The major components for all samples are plasticizers/ antioxidants and even-carbon-chain n-alkanes ranging from C16 to C40, with a carbon number maximum (Cmax) at 26. These even n-alkanes are low-molecular-weight oligomers resulting from polymerization of ethylene, and they probably extend to beyond C40. The dominant plasticizer is di-2-ethylhexyl phthalate (DEHP, I, chemical structures are given in Chart 1), with minor amounts of di-n-butyl, di-iso-butyl, n-butylcyclohexyl, and dicyclohexyl phthalates. The dominant antioxidants and lubricants/antiadhesives are Irganox 1076 (octadecyl 3-(3,5di-tert-butyl-4-hydroxyphenyl)propionate, II), Irgafos 168 (tris(2,4-di-tert-butylphenyl)phosphite, III), and its oxidation product tris(2,4-di-tert-butylphenyl)phosphate (IV). Their mass spectra are shown in Figure 2. Compound II has a Kovats index of 3615 and shows the molecular ion at m/z 530 as a base peak with minor fragments at m/z 515 (M-15), 219, and 57. Compound III has a Kovats index of 3440 and a weak molecular ion at m/z 646 and eliminates a di-tert-butylphenoxy moiety to the base peak at m/z 441, with minor fragments at m/z 308, 191, 147, and 57. Tris(2,4-di-tertbutylphenyl)phosphate (IV) has a Kovats index of 3624, very close to that of compound II, and exhibits a strong molecular ion at m/z 662, loss of CH3 to the base peak at m/z 647, a

FIGURE 2. Mass spectra of standards: (a) octadecyl 3-(3,5-di-tertbutyl-4-hydroxyphenyl)propionate (Irganox 1076, II), (b) tris(2,4-ditert-butylphenyl)phosphite (Irgafos 168, III), and (c) tris(2,4-di-tertbutylphenyl)phosphate (IV). key ion at m/z 316 (no isotope peaks), and minor fragments at m/z 191 and 57. Thus, the key ions for compound IV are m/z 647 or the molecular ion at m/z 662, and it elutes just prior to coronene. Diethylhexyl phthalate and tris(2,4-ditert-butylphenyl)phosphate are also found in plastics used in the United States, China, Nigeria, and Europe, so they appear to be global as components of plastics. However, there are numerous other minor plasticizers and antioxidants in the myriad plastic formulations (e.g., fatty acid isopropyl esters, erucamide, Citroflex A-4 also known as tributyl acetylcitrate, etc.). The minor lipid compounds on the plastics from the landfill, such as palmitic and stearic acids and their methyl esters, 1-monopalmitin, 1-monostearin, squalene, and cholesterol (Table 2), are absorbed fat residues from food packaging materials in garbage. However, 1-monopalmitin, 1-monostearin, and erucamide are also used as rolling lubricants/antiadhesives in plastics manufacture and thus are surface components on some plastics. 7-Hydroxycholesterol (V), which does not occur in fats, is interpreted to be derived from the aerial oxidation of the lipid cholesterol. No polycyclic aromatic hydrocarbons (PAHs) are detectable in the extracts. Organic Tracers in Smoke from Burning Plastics. The GC-MS data for the total extracts of the smoke particulate matter from open-burning of plastics are shown in Figure 3, and the organic tracers identified are listed in Table 2. The

FIGURE 3. GC-MS data for the total extracts from smoke particles of burning plastics: (a) TIC trace of extract from new polyethylene bags (Chile) burn test, (b) TIC trace of extract from landfill plastics burn test, (c) TIC trace of derivatized (TMS) extract from landfill plastics burn test, (d) TIC trace of extract from new polyethylene bags (USA) burn test, (e) m/z 85, key ion for n-alkanes from extract of landfill plastics burn test, and (f) m/z 202, 228, 252, 276, 300, and 306, key ions for selected PAHs and triphenylbenzenes, for extract from roadside plastic litter burn test. Numbers and labels as in Figure 1: 8A ) diethylhexyl adipate, ] ) n-alkanal, Fl ) fluoranthene, Py ) pyrene, AcP ) acephenanthrylene, BzA ) benz[a]anthracene, Chry ) chrysene, Bzfl ) benzofluoranthenes (j and k), Bep ) benzo[e]pyrene, Bap ) benzo[a]pyrene, Per ) perylene, 135-TPB (VI) ) 1,3,5-triphenylbenzene, 124-TPB (VII) ) 1,2,4-triphenylbenzene, Infl ) indeno[1,2,3cd]fluoranthene, Inpy ) indeno[1,2,3-cd]pyrene, Bzper ) benzo[ghi]perylene, and Cor ) coronene. major components are terephthalic acid (1,4-benzenedicarboxylic acid, Kovats index of 1795) and 4-hydroxybenzoic acid, and the plasticizers dibutyl phthalates, diethylhexyl adipate, and diethylhexyl phthalate. Minor specific tracers are the n-alkanes, still with an even-carbon-number predominance (CPI ) 0.10-0.97) and range from C16 to C40+, the PAHs 1,3,5-triphenylbenzene (VI, Kovats index of 3007) and 1,2,4-triphenylbenzene (VII, Kovats index 2630), and the antioxidant product tris(2,4-di-tert-butylphenyl)phosphate (IV). The decrease in the n-alkane CPIs is due to cracking reactions during burning of the polyethylene moieties, which generate both odd- and even-carbon-numbered homologues, diluting the directly volatilized even alkanes (Figure 3e). Some plastics also emit significant amounts of carbonyls (i.e., n-alkanals, C11-C31, Figure 3d) in the smoke. The n-alkanals have no carbon number predominance (CPI ) 1.0), indicating an origin from incomplete combustion and thermal cracking of the polymer. It is obvious that different temperatures, aeration, and fuel mixtures will emit variable compound mixtures and concentrations. Burning of plastics generates PAHs, and the minor compounds range from phenanthrene to coronene (Table 2) with significant amounts of the two triphenylbenzenes (Figure 3f) (6-8). The key ion for triphenylbenzenes is m/z 306, the molecular weight, and the symmetric 1,3,5-isomer is always dominant. The formation of these compounds is not by the conventional zigzag mechanism as for the normal PAH (29), where the parent PAH forms by acetylene radical recombination in ortho-ring (4-carbon, e.g., naphthalene to phenanthrene) and peri-ring (2-carbon, e.g., phenanthrene to pyrene) closures (30). Thus, we propose an origin from dehydration/cyclization of the polyoxyethylene moieties of some antioxidants. The plasticizers and antioxidants are volatilized directly into the smoke by steam stripping (cf., compounds I, IV, VI, and VII in Figure 3) as described for biomass burning (e.g., ref 31). Terephthalic acid is a major pyrolysis product from poly(ethylene terephthalate) (PET). 4-Hydroxybenzoic acid is a pyrolysis product from the antioxidants; i.e., the alkylphenol moieties oxidize to this compound due to

FIGURE 4. GC-MS data of total extracts from aerosol particles showing the plastic burn tracers: (a) TIC trace for derivatized (TMS) extract of Santiago, Chile, aerosol (April, 1999, ref 5), (b) m/z 202, 228, 252, 276, 300, and 306 key ion plot for selected PAHs and triphenylbenzenes for Santiago, Chile, aerosol (April, 1999, ref 5), (c) TIC trace for derivatized (TMS) extract of Gosan Island, South Korea, aerosol (April 2001, ref 9), (d) m/z 202, 228, 252, 276, 300, and 306 key ion plot for Gosan Island aerosol (ref 9), (e) TIC trace for derivatized (TMS) extract of Sapporo, Japan, aerosol (April, 2001, ref 9), and (f) m/z 202, 228, 252, 276, 300, and 306 key ion plot for Sapporo aerosol (ref 9). Numbers and labels as in Figures 1 and 3: 4 ) n-alkanoic acid, O ) n-alkanol, DHA ) dehydroabietic acid, G ) galactosan, M ) mannosan, Lf ) 1,6-anhydro--glucofuranose, C ) cholesterol, and UCM ) unresolved complex mixture.

TABLE 3. Tracers from Burning Plastics in Ambient Atmospheres (ng m-3)a

sample locale April 1997 April 1998 April 1999 November 2000 ambient (ref 33) tunnel (ref 34) ambient, suburban (ref 35) April 2001 (Asian dust, ref 9) April 1, 2001 (Asian dust, ref 9) September 2001 (ref 9) triphenylbenzenes 19 44 17 7 nd nr nd 0.28 2 0.06
b

diethylhexyl phthalate

terephthalic acid 3.2 0.5 21 1.1 5.4 (0.9-17) 136 mg L-1 fuel 0.1 11 6.3 0.2

Irganox 1076b 11 7 132 16 nr nr nd nd nd nd

Irgafos 168 oxidation productc 6 4 48 3.8 nr nr nd tr tr nd

Santiago, Chile
613 609 6920 880

Los Angeles, CA
nr nr

Corvallis, OR
nd

Gosan Island, South Korea

6

Sapporo, Japan
34 42

a nd ) not detected, nr ) not reported, tr ) trace. -butylphenyl)phosphate (IV).

Octadecyl 3-(3',5'-di-tert -butyl-4'-hydroxyphenyl)propionate (II). c Tris(2,4-di-tert

incomplete combustion. However, these latter two tracers are emitted by numerous other burning/combustion processes and are therefore not source-specific. The minor polar compounds, such as levoglucosan, palmitic and stearic acids, cholesterol, R- and -glucose, and sucrose, are volatilized directly or pyrolized from impurities and residues on the used plastic bags. Levoglucosan is probably derived from the cellulose of paper wrap and labels (31, 32). Tracers from Burning Plastics in the Atmosphere. The specific key organic tracers for burning of plastics found in atmospheric particle samples are 1,3,5-triphenylbenzene (VI, e.g., Figure 4) and traces of tris(2,4-di-tert-butylphenyl)phosphate (IV), which occur at significant levels in regions where plastic and refuse are burned in open fires. The levels of 1,3,5-triphenylbenzene (Table 3) in Santiago, Chile, vary from 17 to 44 ng m-3, Sapporo, Japan, 0.06-2.0 ng m-3, and during an Asian dust event on Gosan Island, South Korea, 0.28 ng m-3 (9). It is not detectable in aerosol particulate matter from Los Angeles, CA, or Corvallis, OR. Both diethylhexyl phthalate and terephthalic acid are also detectable in aerosols (Table 3), but they have numerous other sources and can therefore be utilized only in part as confirming tracers. The minor specific tracer, which is also detectable in the Santiago and Gosan aerosols, is tris(2,4-di-tertbutylphenyl)phosphate (IV). It is readily detectable by key ion search. Therefore, 1,3,5-triphenylbenzene is the tracer of greatest utility specific for open-burning of plastics, especially when coupled with the presence of the antioxidant product IV. 1,3,5-Triphenylbenzene has also been detected in particles from solid waste incinerators burning plastics (6, 7). Thus, it should prove to be a useful tracer for the burning of plastic products in domestic waste and litter. Among the five cities compared in this study, the aerosol particulate matter of Santiago has the highest plastic marker content, suggesting that air contamination has a significantly higher plastic combustion load than those of the other four cities. This is most likely associated with the higher contribution of garbage, litter, and plastic related combustion incidents and differences in garbage disposal procedures, littering levels, wind-blown plastic dispersal, and ultimately practices of plastic refuse combustion for energy generation. Some of the other comparative cities operate garbage segregation programs and have more stringent littering and disposal control procedures. Thus, the measurement of these plastic combustion markers in air basins and urban environments should be a good assessment tool for the contribution of garbage and litter combustion to airborne contamination and provide further information to improve emission inventories and air quality management.

Acknowledgments
We thank Hernan N. Hinojosa and Mauro D. Muzio for laboratory assistance in plastic sampling and combustion experiments. P.M.M. thanks the Brazilian government (Conselho Nacional de Desenvolvimento Cient fico e Tecnolo gico) for financial support (Grant No. 200330/01-2). The reviewers are gratefully acknowledged for the useful comments that improved this paper.

Literature Cited
(1) Braun, D. Poly(vinyl chloride) on the way from 19th century to the 21st century. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 578-586. (2) Plastics Institute of America. Secondary Reclamation of Plastics Waste; Technomic Publishing Co.: Lancaster, PA, 1989. (3) U.S. Environmental Protection Agency (EPA). Characterization of Municipal Solid Waste in the United States: 1992 Update; Report No. EPA/530-S92-019; U.S. EPA: Washington, DC, 1992, 428 p.

(4) Resource Recovery of Municipal Solid Wastes; Knox, P. J., Ed; American Institute of Chemical Engineers: New York, 1988. (5) Didyk, B. M.; Simoneit, B. R. T.; Pezoa, L. A.; Riveros, M. L.; Flores, A. A. Urban aerosol particles of Santiago, Chile: Organic content and molecular characterization. Atmos. Environ. 2000, 34, 1167-1179. (6) Tong, H. Y.; Shore, D. L.; Karasek, F. W.; Helland, P.; Jellum, E. Identification of organic compounds obtained from incineration of municipal waste by high-performance liquid chromatographic fractionation and gas chromatography-mass spectrometry. J. Chromatogr. 1984, 285, 423-441. (7) Karasek, F. W.; Tong, H. Y. Semipreparative high-performance liquid chromatographic analysis of complex organic mixtures. J. Chromatogr. 1985, 332, 169-179. (8) Fu, J. M.; Sheng, G. Y.; Chen, Y.; Wang, X. M.; Min, Y. S.; Peng, P. A.; Lee, S. C.; Chan, L. Y.; Wang, Z. S. Preliminary study of organic pollutants in air of Guangzhou, Hong Kong, and Macao. In Molecular Markers in Environmental Geochemistry; Eganhouse, R. P., Ed; ACS Symposium Series 671; American Chemical Society: Washington, DC, 1997, pp 164-176. (9) Simoneit, B. R. T.; Kobayashi, M.; Mochida, M.; Kawamura, K.; Lee, M. H.; Lim, H.-J.; Turpin, B. J.; Komazaki, Y. Composition and major sources of organic compounds of aerosol particulate matter sampled during the ACE-Asia campaign. J. Geophys. Res. 2004, 109, D19S10, http://dx.doi.org/10.1029/2004JD004598. (10) National Packaging Covenant Council. Plastic Shopping Bags in Australia: Plastic Bag Working Group Report; Australian Government, Department of the Environment and Heritage: Canberra, Australia, 2002, p 6. (11) Williams, C. Battle of the Bag. New Sci. 2004, 30-33. (12) Pitzer, D.; Willets, S.; Magdich, P. Garbage is seasonal: Is your waste a composition study? In Proceedings of the Solid Waste Association of North America Pacific Northwest Regional Symposium, Burnaby, BC, April 13-14, 2000; SWANA: Silver Spring, MD. (13) Cascadia Consulting Group, Inc. 2002 Residential Waste Stream Composition Study: Seattle Public Utilities Report; Cascadia Consulting Group, Inc.: Seattle, WA, 2003. (14) Technology Resources, Inc. Solid Waste Composition Study: Study for Greater Vancouver Regional District; Technology Resources Inc.: North Vancouver, BC, Canada, 2005. (15) Cascadia Consulting Group, Inc. 2000-2001 Litter Composition Study: Seattle Public Utilities Report; Cascadia Consulting Group, Inc.: Seattle, WA, 2001. (16) Tullo, A. H. Plastic additives. Chem. Eng. News 2003, 81 (46), 36-42. (17) Piringer, O. G. Evaluation of plastics for food packaging. Food Addit. Contam. 1994, 11, 221-230. (18) Helmroth, L. E. Release of additives from packaging plastics. Ph.D. Thesis, Wageningen University, The Netherlands, 2002. (19) Staples, C. A.; Peterson, D. R.; Parkerton, T. F.; Adams, W. J. The environmental fate of phthalates: A literature review. Chemosphere 1997, 35, 667-749. (20) Cadogan, D. F. The Health and Environmental Effects of Phthalates, PVC99; In Proceedings of the 7th International PVC Conference, Brighton, U.K., April 20-22, 1999; Maney Publishing: Leeds, U.K., 1999. (21) Suggatt, R. H.; Foote, K. Comprehensive Review of Acute Aquatic Toxicity Data on Phthalates. Final Report; Contract SRC TR81537; Syracuse Research Corporation: Syracuse, NY, 1981. (22) Jobling, S.; Reynolds, R.; White, R.; Parker, M. G.; Sumpter, J. P. A variety of environmentally persistent chemicals, including some phthalate plasticizers, are weakly estrogenic. Environ. Health Perspect. 1995, 103, 582-587. (23) Rhodes, J. E.; Adams, W. J.; Biddinger, G. R.; Robillard, K. A.; Gorsuch, J. W. Chronic toxicity of 14 phthalates to Daphnia magna and rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 1995, 14, 1967-1976. (24) Streufert, J. M.; Jones, J. R.; Sanders, H. O. Toxicity and biological effects of phthalates on midges (Chironomus plumosus). TransMO Acad. Sci. 1980, 14, 33-40. (25) Gray, L. E.; Wolf, C.; Lambright, C.; Mann, P.; Price, M.; Cooper, R. L.; Ostby, J. Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,pDDE, and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169, and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol. Ind. Health 1999, 15, 94-118.

(26) Zacharewski, T. R.; Meek, M. D.; Clemons, J. H.; Wu, Z. F.; Fielden, M. R.; Matthews, J. B. Examination of the in-vitro and in-vivo estrogenic activities of eight commercial phthalates. Toxicol. Sci. 1998, 46, 292-293. (27) Thure n, A.; Larsson, P. Phthalate esters in the Swedish atmosphere. Environ. Sci. Technol. 1990, 24, 554-559. (28) Johnson, R. I.; Shaw, J. J.; Cary, R. A.; Huntzicker, J. J. An automated thermal-optical method for the analysis of carbonaceous aerosol. In Atmospheric Aerosol: Source/Air Quality Relationships; Macias, E. S., Hopke, P. K., Eds.; ACS Symposium Series 167, American Chemical Society: Washington, DC, 1981, 223-233. (29) Simoneit, B. R. T.; Fetzer, J. C. High molecular weight polycyclic aromatic hydrocarbons in hydrothermal petroleum from the Gulf of California and Northeast Pacific Ocean. Org. Geochem. 1996, 24, 1065-1077. (30) Stein, S. On the high-temperature equilibra of polycyclic aromatic hydrocarbons. J. Phys. Chem. 1986, 83, 566-571. (31) Simoneit, B. R. T. Biomass burningsA review of organic tracers for smoke from incomplete combustion. Appl. Geochem. 2002, 17, 129-162.

(32) Simoneit, B. R. T.; Schauer, J. J.; Nolte, C. G.; Oros, D. R.; Elias, V. O.; Fraser, M. P.; Rogge, W. F.; Cass, G. R. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles. Atmos. Environ. 1999, 33, 173, 182. (33) Fraser, M. P.; Cass, G. R.; Simoneit, B. R. T. Air quality model evaluation data for organics: 6. C3-C24 organic acids. Environ. Sci. Technol. 2003, 37, 446-453. (34) Fraser, M. P.; Cass, G. R.; Simoneit, B. R. T. Gas-phase and particles-phase organic compounds emitted from motor vehicle traffic in a Los Angeles roadway tunnel. Environ. Sci. Technol. 1988, 22, 2051-2060. (35) Simoneit, B. R. T. Organic matter of the tropospheresIII. Characterization and sources of petroleum and pyrogenic residues in aerosols over the western United States. Atmos. Environ. 1984, 18, 51-67.

Received for review April 21, 2005. Revised manuscript received July 1, 2005. Accepted July 6, 2005. ES050767X