Click for next page ( 2


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
1 POLYCYCLIC AROMATIC HYDROCARBONS FROM MOBILE SOURCES AND THEIR ATMOSPHERIC CONCENTRATIONS Exhaust products of fuel combustion from stationary or mobile sources that have been identified as hazardous to humans are commonly targeted for abatement or regulatory control. A variety of control techniques--e."., particle collectors, gaseous-emission scrubbing devices, catalytically equipped exhaust systems, and "scavenger" fuel additives--have been used to convert the unburned and partially burned hydrocarbons, including polycyclic aromatic hydrocarbons (PAHs), in exhaust to less hazardous chemicals. This chapter discusses the annual consumption of fuels in various types of vehicles, sampling, PAR emission from mobile sources, and future control technologies . FUEL CONSUMPTION IN THE UNITED STATES The important fuels consumed in this country are listed in Table 1-1 with estimates of annual consumption figures for 1979, the latest year for which all data are available. The major energy source, of course, is crude oil. Table 1-2 lists the uses of the major crude-oil fractions for 1979.60~61 The U.S. consumption of crude oil is decreasing. In addition, important changes in how oil is used are possible within the next two decades. For example, gasoline consumption currently far exceeds the con- sumption of diesel fuel. Owing to the increased fuel mileage of gasoline- fueled vehicles, the increasing use of diesel-fueled vehicles, and overall efforts at energy conservation, it is possible that diesel-fuel consumption could outstrip gasoline consumption in two decades. TYPES OF MOBILE SOURCES AND THEIR RELATIVE IMPORTANCE The term "mobile source" represents a broad range of vehicle classi- fications with considerable differences in miles traveled, amount and type of fuel consumed, exhaust emission rate, and location of fuel use. In addition, the emission from any particular category may change considerably from one year to the next with technologic advances in engine design and emission-control techniques. Current estimates of miles traveled and fuel cons::pi~og fig Etch mg~ile-source category are listed in Table 1-3. ~ ~ 0, , ,82, The present status and projected changes in relative importance of each of the categories are discussed below. Light-duty passenger cars with spark-ignition engines account for most of the motor-vehicle mileage accumulated in this country. To meet the exhaust-emission standards for gaseous hydrocarbon (HC) and carbon 1-1

OCR for page 1
monoxide (CO), most 1975 and later model-year spark-ignition passenger cars have been equipped with oxidation catalys ts on the exhaust system. The catalysts are poisoned by lead in the fuel and therefore require unleaded fuel. The use of unleaded fuel in catalyst-equipped cars and lower lead concentrations in leaded fuel have resulted in considerable decreases in rates of emission of lead, HC, CO, and particulate material from passenger cars.~51 The transition to catalyst-equipped cars has continued, and older non-catalyst-equipped cars are continually being removed from service, so the present (1982) mileage of catalyst-equipped cars is now greater than the mileage of noncatalyst cars. By the early l990s, more than 95% of the gasoline-fueled passenger-car mileage will be attributed to catalyst-equipped vehicles. Beginning with the 1981 model year, most new passenger cars have three-way catalysts capable of reducing emission of HC, CO, and nitrogen oxides (NOx). More than 50: of the catalyst-equipped passenger cars will have three-way catalysts by the mid-199Os. The exact mix will depend heavily on future NOx emission standards. Three-way catalysts result in significantly lower emission of CO, NOx, gaseous HC, and particulate material than the original oxidation catalysts.51 Increasingly stringent federal fuel-economy standards (Figure 1-1) are in effect through the 1985 model year for passenger cars.66 Coupled with oil shortages and the goal of decreased U.S. dependence on foreign oil, the fuel-economy standards will result in an approximate doubling of new-car fuel economy between 1974 and 1985. The goal of improved fuel economy is being attained by a decrease in vehicle weight, the use of more fuel-efficient engines, and an increase in the use of diesel engines in light-duty vehicles (both in passenger cars and in light- and medium-duty trucks). Diesel-engine vehicles achieve about 25% higher fuel mileage than their counterparts among spark-ignition-engine vehicles (a somewhat smaller improvement if the volume of crude oil or the energy content of the fuel is used as a basis). The cost advantage enjoyed by diesel fuel over gasoline has largely dissipated in the last few years and could even turn into a cost penalty as the demand for diesel fuel increases and the demand for gasoline decreases.14 It has been projected that 25% of the passenger-car fleet could be diesel-powered by the mid-1990s, but many factors will affect the actual rate of approach to that proportion and the percentage ultimately attained.2~25~28~44 Buyers' demand for diesel-powered passenger cars and light-duty trucks has been strong since about 1979, but there is considerable concern about possible health effects and urban visibility degradation associated with emission of particles in diesel exhaust. Diesel particulate-emission rates are about two orders of magnitude greater than those associated with catalyst-equipped spark-ignition vehicles.25~51~9 Comparative emission factors are discussed later, but it is evident from Table 1-3 that there is already extensive use of diesel-engine vehicles in this country, and it will be well into the l990s, if ever, before light-duty diesel particulate emission becomes equivalent in tonnage to the particulate material from heavy-duty diesel-eng~ne vehicles nationwide.12,13,44 1-2

OCR for page 1
Medium-duty trucks (gross vehicle weight, 8,500-33,000 lb) are now equipped with spark-ignition engines, but are also undergoing diesel- ization rapidly. Heavy-duty trucks (~33,000 lb) are already more than 90: diesels. In 1980, about 15% of new trucks sold were diesel-engine vehicles, and this figure could grow to 50% by the year 2000. In addition, the total number of trucks in the United States is expected to increase by 5% per year until beyond 2000.44~46~60 Approximately 90% of all commercial buses are powered by diesel engines; school buses are still powered by spark-ignition engines. Trains and ships are powered by diesel engines, as is most industrial equipments Private boats and planes are powered predominantly by spark-ignition engines. Commercial aircraft are powered by gas-turbine engines that use jet fuel. Although information on engine types used in military vehicles is not readily available, there is some information in Table 1-3 on military fuel consumption. Emission from military vehicles is not a major source of atmospheric PAHs, except possibly in particular areas. PAR EMISSION FROM MOBILE SOURCES Internal-combustion engines emit gases, liquids, and solids from the exhaust system as products of the incomplete combustion of the fuel and as noncombusted fuel, lubricants, and fuel additives. Chemical processes also occur in the exhaust system, especially in the catalytic emission- control devices. Some reactions continue after the exhaust is released into the atmosphere. The temperatures in the combustion chamber and in the exhaust system and the volume flow rates depend directly on engine design, size, operating speed, and working load. These factors are important in the formation of PAHs and in the amounts of PAHs that are emitted into the atmosphere. The combustion process in a spark-ignition engine takes place with near-stoichiometric amounts of oxygen at temperatures in the vicinity of 3500C. In diesel engines, there is an excess of oxygen with combustion temperatures in the vicinity of 2000C. Exhaus t temperatures for spark-ignition engines are commonly between 400 and 600C, but diesel exhaust is typically at 200-400C (except at high load factors). Oxidation catalysts typically must be at 400C or higher before becoming active, so current types of oxidation or three-way catalysts do not function efficiently on diesel vehicles.29~43 Many of the PAHs have boiling points of 200-300C and are sufficiently volatile to exist predominantly in the gas phase at temperatures above 200C. Even at room temperature, some of the more volatile PAHs are distributed between the vapor and particle-adsorbed phases.52 1-3

OCR for page 1
In the open air, vehicular exhaust is diluted by a factor of about 1,000 in the first few seconds, so cooling to near-ambient temperature is quite rapid. But condensation of PAHs, by adsorption on existing particles, can occur many feet behind a vehicle, thus allowing some mixing of exhaust plumes from different emission sources. Normally, black "elemental carbon" particles,* also products of incomplete fuel combustion, act as condensation nuclei for the condensation of vapor-phase organic chemicals, such as aliphatic compounds, aromatic compounds (including the PAHs), aldehydes, ketones, acids, and heterocycles. The exhaust is the major source of the PAHs; another possible source of PAHs is engine oil, because it can act as a sink for them. It has been estimated that crankcase oil collects 10 times as much PAN per mile traveled as is released from the exhaust system. 70 In the case of vehicles in which volatile emission from the crankcase is not controlled, it could be a significant source of PAHs in the atmosphere, but quantitative assessment is not now possible. Studies of particle-size distribution of spark-ignition and diesel exhaust particulate material show mass-median aerodynamic diameters of 0.1-0.25 ~m.20~36~63 More than 90% of the mass is in particles less than 1 Am in diameter.72 Larger particles presumably result from deposition of particulate material on and later release from the walls of the exhaust system. Resuspended road dust, roadbed material, and tire particles result in particle sizes of about 8 Am in median diameter, which can account for as much as 10t of the measured vehicular respirable particulate mass in near-road measurements.68~72 Spark-ignition vehicles with oxidation catalysts emit particulate material that is mostly aqueous sulfuric acid droplets with organic compounds presumably adsorbed on droplet surfaces. Particle median diameters tend to be somewhat less than 0.1 ~m.59 Diesel particulate material is mostly elemental carbon. The primary particles are spherules 0.015-0.03 Am in diameter that agglomerate at high temperatures to irregular clusters and chains. These clusters, about 0.15 Am in diameter and containing up to 4,000 spherules, act as carriers of the PAHs and other adsorbed species.86 The PAHs are adsorbed on the surface of the carbon and into the minute pores between the spherules. The small particle size results in long atmospheric residence times and in deposition in alveolar regions of the lung.3~13~15~58~64 (See discussions in Chapter 3 on particles in the atmosphere and in Chapter 5 on relations of deposition of PAHs and particles.) *This material is not truly elemental carbon, nor is it graphitic carbon. No term has been found or accepted that properly describes the material.92 1-4 . - ,.

OCR for page 1
Exhaust-emission standards from EPA for HO, CO, and NOX have resulted in 84, 79, and 56% reductions, respectively, in 50, 000-mi emission from spark-ignition passenger cars, as shown in Table 1-4.84 Particulate emission and lead emission have also decreased as a result of the use of catalysts and the decrease in lead concentrations in leaded fuel. Figure 1-2 shows the decrease in urban CO concentrations since 1973, 3 and Figure 1-3 shows the decrease in traffic-average lead emission rates as measured in highway tunnels.72 SAMPLING OF EXHAUST FROM MOBILE SOURCES Techniques for sampling exhaust from mobile sources have been thoroughly described elsewhere and are not reviewed here except as pertinent to the analysis of pAHS.4,21,53,77 Exhaust-particle sampling in this country commonly involves the use of dilution tunnels. The dilution tunnel represents a laboratory attempt to simulate the normal atmospheric dilution and cooling of the exhaust. Atmospheric dilution is by about 1,000:1 in the first few seconds, whereas typical laboratory dilutions are between 5:1 and 20:1.19~24 Exhaust from a vehicle tailpipe is mixed with particle-free, temperature- and humidity- controlled air in a tunnel that is typically 8-16 in. (20-40 cm) in diameter. Downstream from the exhaust inlet, a constant fraction of the diluted exhaust is pumped through a high-efficiency filter to collect exhaust particles. The weight gain of the filter is a measure of total particulate emission. Adsorbed organic matter, including PAHs, is isolated from the carbon particles by solvent extraction or other techniques. The organic extract material can then be analyzed in many ways, including high-performance liquid chromatography (HPLC)~ gas chromatography (GO), mass spectrometry (MS or GC/MS), and bioassays.5~4 ~ 3~77379 The Salmonella assay has become a commonly used test in almost all laboratories working with vehicle-exhaust particulate material. A technique applied more commonly in Europe uses low-temperature condensers (as many as three in series, at successively lower temperatures) followed by filtration.35~54~55 This approach is used either on the undiluted exhaust or on the exhaust from a dilution tunnel. As mentioned previously, PAHs with high volatility (molecular weight <250) can be distributed in the vapor and condensed phases. Therefore, filter-only sampling from a dilution tunnel misses some of the more volatile PAHs. By combining the PAHs collected in each of the condensers with the PAHs collected on the filter, one obtains a better quantitative assessment of PAR emission. Results from filter-only dilution-tunnel studies are used to provide a qualitative description of PAHs in exhaust, and condenser-study results are used to provide quantitative emission rates. In any attempt to sample a chemical system, it is necessary to show that the sampling process does not alter the chemical concentrations of the 1-5

OCR for page 1
mixture. The production and destruction of chemicals during sampling are called "artifacts" of the sampling process. Sampling artifacts are important in a discussion of the production of the nitro-PAHs. QUALITATIVE DESCRIPTION OF EMISSION FROM MOBILE SOURCES The combustion of gasoline or diesel fuel in air yields water and carbon dioxide as the principal combustion products. Nitrogen oxides result from the high-temperature reaction of nitrogen in the air and from combustion of nitrogen-containing compounds in the fuel and lubricant.l,41 Carbon monoxide, gas-phase hydrocarbons, elemental carbon, and particle-adsorbed organic material are formed as products of the incomplete combustion pro- cess. Fuel and lubricant additives and impurities and their combustion products are also found in exhaust. For example, sulfur-containing organic compounds in the fuel are combusted to gaseous sulfur dioxide, some of which can be further oxidized to sul furic ac id in the combustion chamber or in the oxidation catalyst and give rise to sulfuric acid in the particulate material. The components detected as gas-phase hydrocarbons are listed in Table 1-5 (from a study of on-road gaseous organic-compound emission). The quantitative emission rates have not been determined.37~38 Diesel-exhaust particulate material has been the subject of extensive study in the last 5 yr. It is typically about 25% extractable into organic solvents, although different vehicles may have extractable fractions of 10-90%, depending to some extent on operating conditions. More than half the extractable material is aliphatic hydrocarbons of 14-35 carbon atoms and alkyl-substituted benzenes and naphthalenes.4~42~78 The remaining extractable mass is PAHs and oxidized derivatives of the PAHs, such as ketones, carboxaldehydes, acid anhydrides, hydroxy compounds, quinones, nitrates, and carboxylic acids. There are also heterocyclic compounds containing sulfur, nitrogen, and oxygen atoms within the aromatic ring. The alkyl-substituted PAHs and PAH derivatives tend to be more abundant than the parent PAN compounds. The particulate-extract HPLC eluent can be separated into nonpolar, moderately polar, and highly polar fractions. The fractions can then be further analyzed by GC/MS. Table 1-6 lists the results of such an analysis of the nonpolar and moderately polar fractions of a particulate extract from an Oldsmobile diesel vehicle, including the approximate extract concentra- tions for this particular vehicle.78 The highly polar fraction has not been fully characterized. It contains the PAH carboxylic acids, acid anhydrides, and probably sulfonates and other highly polar species.78~94 Most (75%) of the direct bacterial mutagenicity resides in the moderately polar fraction (see discussions of Salmonella strains in Chapter 4' 10,71,73,74,76,80,81,88 The remaining direct mutagenicity is in the highly polar fraction. These aspects are discussed further in Chapter 4. 1-6

OCR for page 1
Over 50 chromatographic peaks of nitro-PAH compounds have been identified in diesel particulate extracts, as listed in Table 1-7 22,40,47,65,75,93 1-Nitropyrene is the most abundant of the nitro-PAHs, ranging from 25 to 2,000 ppm in the vehicle extracts studied. The other nitro-PAHs are present at concentrations from below the parts- per-million range to a few parts per million. The nitropyrenes have been studied in greater detail. They are released in diesel and gasoline exhaust (according to particulate extracts) at approximately 8.0 and 0.30 ~g/mi, respectively. The latter value was obtained with leaded gasoline; with unleaded fuel, the rate was 0.20 ~g/mi.50 Gibson26 has determined the concentration of 1-nitropyrene in ambient particulate extracts obtained from suburban areas in Michigan to be 0.016-0.030 ng/m3 of air (corresponding to 0.2-0.6 ng/mg of particles. Gibson has also observed that catalytic converters greatly reduce the concentration of nitropyrenes. The nitropyrene concentration in extracts from particles obtained from the emission of a wood-burning fireplace was less than 0.1 ng/mg of particles.26 1-Nitropyrene has been the only nitro-PAH detected in spark-ignition particulate extracts.51~95 On-road heavy-duty diesel and light-duty spark-ignition vehicles have recently been found to have very low 1-nitropyrene particulate extract concentrations, which thus account for very small fractions of the on-road direct bacterial mutagenicity associated with these vehicle categories.31 QUANTITATIVE DESCRIPTION OF PAH EMISSION FROM MOBILE SOURCES The work of Grimmer and co-workers, 33~34 of Kraft and Lies,48 and more recently of Zweidinger and colleagues51395 can be used to derive typical rates of emission of many of the PAHs from the different categories of mobile sources. Because a relatively small number of vehicles have been used to measure these emission rates, the uncertainty in the derived vehicle-category emission factors is quite large--probably at least a factor of 2 and possibly even larger. Table 1-S lists the best current measurements of rates of emission of numerous PAHs and their derivatives for spark-ignition vehicles ~ light-duty with and without catalysts and heav - duty) and for light-duty and heavy-duty diesel vehicles.6~6~~39~90~9{ For the other categories of mobile sources, the estimate of total PAH emission can be based on the heavy-duty spark-ignition or heavy-duty diesel emission rate per gallon of fuel consumed and the total fuel consumption of the category (railroads, aircraft, etc.~. When more than one value for a particular PAH emission rate is available for a source category, the micrograms-per-gallon-of-fuel figures are averaged. Kraft and Lies found a very similar distribution of the PAHs for diesel and gasoline vehicles. Owing to the paucity of emission-rate measurements, we used this observation to derive emission factors for vehicle categories when measurements are lacking. Table 1-9 lists the resulting emission-rate estimates for the

OCR for page 1
different vehicle categories. estimates were: The ane~mptions made in deriving these That the PAH distributions for both spark-ignition and diesel vehicles are the same as the distributions for average light-duty spark- ignition noncatalyst vehicles. O That the measured BaP emission rates for the oxidation-catalyst spark-ignition vehicles and for the three-way-catalyst classes represent the reductions in all PAH emission rates, compared with the noncataly~t values. O That the heavy-duty spark-ignition class has the same fuel- specific emission rates (in micrograms per gallon) as the light-duty spark- ignition noncataly~t vehicles. . That the heavy-duty diesel clans has the same fuel-~pecific emission rates as the light-duty diesel class. Within the limits of those assumptions, we have a complete list of PAH emission factors for each of the vehicle categories in teems of micrograms per gallon of fuel consumed. With typical fuel-economy values for each class, one can calculate the micrograms per mile for each class. These results are also listed in Table 1-9. The 1-nitropyrene values are those from actual experimental measurements, unless a derived value was higher than the measured value. Therefore, the resulting t-nitropyrene emission rate should be considered an upper limit. Fuel-specific PAH emission rates can be combined with the total fuel-consumption values in Table 1-3. That yields- a total emission tonnage for each mobile-source category for each PAH and PAH derivative in Table 1-9. The PAHe released from mobile sources in 1979 according to these estimates are listed in Table 1-10. The total BaP emission from all mobile sources is estimated to be 43 metric tons. This encompasses all mobile sources, whereas the motor-vehicle contribution is 27 metric tons (about 63: of the totat) exclusive of the railroad, aircraft, ship, farm, military, and other contributions. Motor-vehicle BaP emission was estimated in 1972 at about 20 metric tons/yr.62 The calculated mobile-source emission of 1-nitropyrene is 17 metric tons, of which 30X is calculated to be contributed by motor vehicles. The non-motor-vehicle categories tend to be less relevant to polluted-air concentrations, because they are used away from urban areas (railroads, ships, farm machinery) or because their emission is dispersed above the boundary layer (aircraft). In addition, the various motor-vehicle categories are uset to various extents in urban areas. Passenger-car use is 60X urban, light-truck use 55: urban, and heavy-truck use only 20X urban.60 The urban fraction of the total motor-vehicle PAH emission is calculated to be 63Z for 1979, owing mainly to the dominance of noncatalyst-pa~senger-car emission of PAHs. The relative contributions of each of the mobile-source categories to PAH and 1-nitro- pyrene emission are listed in Table I-11. The 70X contribution of the non-motor-vehicle sources to the 1-nitropyrene emission may be an artifact of the method used for calculating emission rates, inasmuch as 1-nitro- pyrene from sources other than passenger cars and trucks has not been investigated. 1-S

OCR for page 1
By far the largest single contribution to PAR emission (from mobile sources) is that from noncatalyst spark-ignition passenger cars, which will soon be supplanted by vehicles equipped with oxidation catalysts and three-way catalysts. The second most important category for motor-vehicle PAN emission consists of spark-ignit ion light trucks. As much as half the spark-ignition light trucks (all those under 8,500 lb) will have catalysts by the year 2000. The relative importance of the heavy-truck diesels can be expected to increase with that of the light-truck diesel and passenger-car diesel categories. Comparison of the estimated BaP emission factors in Table 1-9 for each of the motor-vehicle categories with values reported in the literature indicates that the present estimates tend to be on the high side of what might be expected for fleet-average values. (For the purpose of the estimates in this report, overestimates are obviously preferable to underestimates.) For example, light-duty diesel BaP emission rates range from less than 1 ~g/mi to more than 20 ~g/mi, with mean values reported in the vicinity of 3-4 ~g/mi. The present estimate is 13 ~g/mi for light-duty diesels. The few measurements of BaP emission rates for heavy-duty diesels that have been reportedl8 indicate that the 54~ g/mi value in Table 1-9 may be too high by as much as an order of magnitude. The reason for this discrepancy is not apparent, but it may reflect a real difference between the four-stroke indirect-injection light-duty diesel and the two- or four-stroke direct-injection heavy-duty diesel. If the lower emission rates are correct, the role of heavy-duty diesel emission is considerably less than portrayed in later sections of this report. ~ It is now possible to use the emission rates in Table 1-9 and the projections previously described to estimate future rates of emission from motor vehicles. Using the current BaP emission rates, we have calculated the motor-vehicle BaP emission for the year 2000 and listed the results in Table 1-12. The 24 metric tons of BaP represents an 11% decrease from the 1979 value of 27 metric tons and reflects the benefit of catalyst-equipped spark-ignition passenger cars over their noncatalyst counterparts, which is partially offset by the incursion of diesel vehicles. In the year 2000, without further particulate-emission controls, diesel vehicles will account for 407 of the mileage, 50t of the fuel consumption, and 80% of the total motor-vehicle BaP emission, according to this estimate. If the present distribution of motor-vehicle use between urban and rural areas will still be valid in the year 2000, it can be estimated that about 40: of the BaP from motor vehicles will be released in urban areas in the year 2000, compared with 63Z in 1979. Thus, the BaP tonnage nationwide will decrease slightly and there will be a shift to more rural emission and away from urban areas. The total-tonnage estimates just described do not assess directly the problem of human exposure to air pollutants. In this regard, emission rates are not the sole important quantities. What is needed is an estimate of atmospheric concentrations in the air inhaled by people. The results of atmospheric-dispersion modeling by Ingalls and Garbe can be used to 1-9

OCR for page 1
calculate atmospheric concentrations resulting from motor-vehicle emission in many of the typical urban-exposure situations.45 The model was constructed on the basis of a hypothetical 1-g/mi traffic-weighted emission rate for 1980 vehicle distributions. To calculate atmospheric concentra- tions for an emission component with other than a 1-g/mi emission rate, one need only multiply the Ingalls and Garbe exposure-concentration factor by the actual traffic-weighted emission rate in grams per mile. Table 1-13 lists the exposure conditions modeled by Ingalls and Garbe and the 1-g/mi exposure-concentration factors derived. One can calculate exposure to BaP on the basis of the data in Tables 1-13 through 1-15. In the calculation, the effect of binding of BaP to particles on deposition and absorption is not considered. Because retention of particle-bound BaP, and thus absorption, depends heavily on particle size and because particle size varies widely, we have characterized exposure on the assumption of complete retention. We assume that 90% of the BaP is bound to particles less than 1 Am in diameter.72 To use these results for calculation of BaP exposures, one uses Table 1-14 to derive the traffic-weighted BaP emission rate for 1979. The same data for the year 2000 are listed in Table 1-15. The 1979 exposure concentration of BaP in a typical roadway tunnel is O .017 ~g/m3 ~ ( 15.3 ~g/mi) ( 10-6 grog) ( 1,123 g/m3 per g/mi) ~ . A person exposed to a concentration of 0.017 ~g/m3 for 2 min while breathing at the rate of 15 m3/d would inhale 0.4 ng of BaP ~ (0.017 ~g/m3) (15 m3/d) (1/24~1/60~2 min)~103 ng/~g)~. The total daily exposure of a person can be calculated by summing over each of the exposure situations experienced in the course of the day. The result of these calculations is a degree of exposure by inhalation. The dose of BaP to the body would be less and would depend on the fraction of the BaP-laden particles that is deposited in the body. This fraction is highly uncertain and depends on particle size, shape, and hygroscopicity and on BaP loading per particle, which also depends on particle size. The fraction of BaP deposited is probably about 20-50X of the BaP inhaled. This has been done in Table 1-16 for a person living in a suburb (1,000 m from an expressway) with a 1-h commute to a job at street level in a central-city street canyon. These are rather severe conditions and result in higher exposures than would be expected for the average urban dweller. These conditions lead to a calculated inhalation of 20 ng of BaP. Had the person stayed home all day, the exposure would have resulted in a 3.0-ng inhalation. For comparison, the BaP inhalation from one cigarette is 20 ng.62 Smoking 1 cigarette/d has the effect of being exposed to over 15 ng/m3 for the entire day (see discussion on BaP exposure from smoking in Appendix C)--an inhalation of over 330 ng of BaP. For the traffic composition projected to exist in the year 2000, the calculated workday exposure of the person is 9.1 ng of BaP, a 557` decrease from the 1979 value. This shows again the decrease in urban exposure at the expense of an increase in rural exposure. The 9.1-ng BaP inhalation when combined with the 15-m3/d inhalation rate gives a calculated average daily atmospheric concentration of 0.6 ng/m3 in the year 2000. Calculations of BaP total motor-vehicle emission tonnage, urban fractions, and inhalation exposures for the year 2000 have assumed no changes from the present fuel-specific emission rates for the various motor-vehicle categories. Even modest particulate-emission controls (50t 1-10

OCR for page 1
reductions) for diesel cars and trucks and for spark-ignition light trucks would result in significantly greater reductions in BaP (and other PAR) exposures than those calculated here. This of course assumes that compliance (i.e., not removing or poisoning catalytic converters) approaches 100%. The benefit of such controls on the basis of BaP exposures will need to be assessed. FUTURE CONTROL TECHNOLOGIES The need for control of emission from light-duty spark-ignition vehicles seems to be moot, with all new vehicles being sold now having catalyst systems. Use of catalysts on heavy-duty spark-ignition vehicles, if feasible, would be expected to result in PAR reductions comparable with those observed for light-duty spark-ignition vehicles. Control of diesel particulate material has received much attention recently. The light-duty diesel particulate-emission standard of 0.6 g/mi that went into effect for the 1982 model year was achieved by most diesel manufacturers through engine modifications. The 0.2-g/mi standard proposed for 1985 would not be as readily attained, at least for the larger vehicles. Currently, only diesels of less than 2,600 lb could meet a 0.2-g/mi particulate-emission standard without exceeding the gaseous-emission standards.85 Variations in diesel fuel appear to be inadequate to allow attainment of the standard for all but the smaller diesel vehicles. A number of diesel-particle control techniques are under investigation. In general, these entail after-treatment devices designed to collect particles from the exhaust stream and to oxidize the collected material periodically. Diesel-particle control devices are being developed by several companies and are being tested by automobile manufacturers. Texaco has reported results on an alumina-coated metal-wool diesel-particle filter that achieved a collection efficiency of about 70% without increasing the backpressure enough to sacrifice fuel economy and performance. The col- lected particulate material must then be removed by combustion.57~69~87 Johnson-Matthey is developing a wire-mesh particle trap that is coated with a catalytic material to initiate the combustion of the collected soot. Efficiencies exceeding 50% have been achieved during 50,000-mi accumulation with regeneration every 300-1,000 mi. Corning has developed a ceramic- honeycomb monolithic particle filter that can be coated with a catalyst material Lo assist in soot combustion.87 Gorse et al.29 and Williams8 recently reported emission characterization studies that used some of the above-mentioned control devices. The ceramic trap removes more than 90% of the elemental carbon particles and about 50% (with the catalyst coating) of the particulate organic material and can result in an order-of- magnitude decrease in the emission of bacterial mutagens per mile of travel The wire-mesh catalyst trap removes more than 90t of the particulate organic material and 30% of the elemental carbon. Some of the catalyst-coated traps can produce very high sulfate emission rates, especially during regeneration. In general, there appears to be some hope of success for diesel-particle control, but the devices tested so far need to be tested for durability, 1-11

OCR for page 1
90. 91. 92. 93. 94. 95. Williams, R. L., and D. P. Chock. Characterization of diesel particulate exposure, pp. 3-33. In W. E. Pepelko, R. M. Banner, and N. A. Clarke, Eda. Health Effects of Diesel Engine Emissions: Proceedings of an International Symposium. EPA-600/9-80-057a. Cincinnati, Ohio: U.S. Environmental Protection Agency, Office of Research and Development, 1980. Williams, R. L., and S. J. Swarin. Benzo~a~pyrene Emissions from Gaso 1 ine and Die se 1 Automob iles . SAE Technical Paper 790419 . Warrendale, Pa.: Society of Automotive Engineers, Inc., 1979. 8 pp. Wolff, G. T., and R. L. Klimisch, Ede. pheric Life Cycle, pp. v-vi. New York: Plenum Press, 1982. Ku, X. B., J. P. Nachtman, Z. L. Jin, E. T. Wei, S. Rappaport, and A. L. Burlingame. Isolation and identification of mutagenic nitro arenes in diesel-exhaust particulates, pp. 556-558. In EPA Diesel Emissions Symposium, October 5-7. Raleigh, N.C.: U.S. Environ mental Protection Agency, 1981. Yu, M-L ., and R. A. Hites . Identif ication of organic compounds on diesel engine soot. Anal . Chem. 3: 951-954, 1981. Zweidinger, R. B. Emission factors from diesel and gasoline powered vehicles; correlation with the Ames test, pp. 95-lO8. In EPA Diesel Emissions Symposium, October 5-7. Raleigh, N.C.: U.S. Environmental Protection Agency, 1981. 1-42 Particulate Carbon: Atmos

OCR for page 1
2 POLYCYCLIC AROMATIC HYDROCARBONS FROM NATURAL AND STATIONARY ANTHROPOGENIC SOURCES AND THEIR ATMOSPHERIC CONCENTRATIONS Although the emphasis of this report is on the identification of the polycyclic aromatic hydrocarbons (PAHs) emitted from motor vehicles, PAHs are ubiquitous substances. They are found in terrestrial and aquatic plants, in soils and bottom sediments, in fresh and marine waters, in emission from volcanoes and naturally occurring forest fires, and in the products of numerous human activities. The anthropogenic sources vary widely--major oil spills and the inestimable minor spills of petroleum products, emission from coal- and gas-fired boilers and electric-power generating plants, space heaters (especially in individual residences), municipal and industrial incinerators, and all sorts of industrial processes. It is not possible to list all the sources or to count or measure the PAHs produced by them. The various PAH compounds and the amounts emitted into the environment from each of the sources result in a complexity that makes it difficult to trace and identify the major contribu t ing source s . PAH COMPOUNDS IN PETROLEUM AND FOSSIL-FUEL PRODUCTS The carcinogenic potential of petroleum hydrocarbons was examined in a critical review of world literature in the period 1960-1978 by gingham et al.3 Although the carcinogenic potential of some samples of petroleum and other fossil-fuel material tested in experimental animals could be associated with the presence of benzota~pyrene (BaP), others without benzota~pyrene were also carcinogenic. Thus, the authors suggested that benzo~a~pyrene may not be the most prevalent or important component in the samples and recommended further chemical analyses of a variety of petroleum samples to determine the profile of PAH compounds in them. The review included references to the carcinogenicity of high-boiling-point (above 260C) petroleum fractions, res idues, and products and to the occurrence of cancer in workers in refineries and industries in which these materials are used. The review did not discuss the carcinogenicity of pure PAHs or studies of environmental pollution from general sources, as are covered in this report. PAH COMPOUNDS IN CRUDE OILS COAL AND OIL-SHALE DERIVATIVES , , Mutagenicity testing (with standard S. typhimurium procedures) by Guerin _ al.13 Of several crude oils and shale- and coal-derived petroleum substitutes showed the petroleum-substitute mutagenicities to be equal to or 10-100 times greater than those of petroleum products. The nonpolar neutral constituents generally were found to contribute over half the mutagenicity. Those findings give added incentive for identifing and characterizing the individual PAHs in these products. 2-1

OCR for page 1
Crude Of 1 In their review of the carcinogenicity of petroleum hydrocarbons, gingham et al.3 cited several of the early researchers' work (1931) on the carcinogenicity of crude-oil fractions from various sources when applied to the skin of mice. The distillate fractions in the 300-400C range were later found to be more carcinogenic than whole crude oil in mice and rabbits. Although the review cited studies by Hueper in 1965 that failed to produce cancer within the normal life span when undiluted crude oil was appl fed to C57BL mice, it cited Barr-Nea and Wolman in 1972, who found papillomas after 5-7 ma of topical application of acetone-extracted crude oil. Numerous other studies cited in the review described the carcinogenic potential of crude oils from various locations when tested in animals. The review by gingham et al. cited references that gave the content of BaP: 40, 1,320, and 1,660 AWL, respectively, in Persian Gulf, Libyan, and Venezuelan petroleum (Graf and Winter, 1968) and 1,000 and 2 , 800 ~g/kg in South Louisiana and Kuwait crude oils, respectively (Panceron and Brown, 1975) . Several studies cited by McKay and Latham22 reported qualitative findings of anthracenes, phenanthrenes, benzophenanthrenes, fluorenes, chrysenes, pyrenes, perylenes, and coronene in virgin petroleum. In the process of cracking of petroleum distillates, the high-temperature hydroconversions formed ring systems, such as benzocoronenes, dibenzocoronenes, and tribenzocoronenes. The authors identified seven polynuclear aromatic compounds not previously found in virgin petroleum distillates (temperature, 335-550C): 1.12,2.3- dibenzoperylene, 1,12-o-phenyleneperylene, pyrenotl.3:10'.2'jpyrene, 2.3,10.11-dibenzo- perylene, 1,2,4,5-dibenzopyrene, benzote~pyrene (BeP), and benzo~gichrysene. Similar quantitative evidence of the presence of pyrene, BaP, BeP, chrysene, and 1,23-benzoperylene was reported by Coleman et al.7 in fluorescence emission and fluorescence excitation analysis of Prudhoe Bay crude oil. Coal and Oil-Shale Derivatives Coal gasification has been used to produce clean fuels in many countries since 1880. The measurement of individual PAHs in the synthetic oils produced by coal liquefaction and in natural crude oil remains a difficult problem. The samples often are from small-scale processes with questionable resemblance to the products of eventual commercial-scale operations. Guerin et _ .12 analyzed fractions of two coal-derived crude oils (synthoil from catalytic hydrogenation of coal, synthoil C, and syncrude from pyrolysis of coal, sync rude D), shale-derived crude oil shale B. and a petroleum mix (crude-oil mixture from California, Canada, Alaska, Iran, Louisiana- Mississippi, and Arabian Light). The results of the chromatographic analyses are shown in Table 2-1. The coal-derived c rude s had larger quantities and a greater variety of PAHs than the petroleum sample, and 2-2

OCR for page 1
shale B had les s than either of them. The summat ion of the PAHs produced the following totals: synthoil C, 135 mg/g; syncrude D, 132 mg/g; shale B. 36 mg/g; and petroleum mix A, 58 mg/g. The increased interest in greater use of petroleum substitutes from oil shale and coal has raised concern about the health hazards associated with these fuels. This concern led Buchanan et al.5 to study the mutagenicity of subfractions from several fossil fuels, princ ipal ly the primary aromas ic amines (PAAs ~ . Aminofluorene and aminoanthracene were among the PAAs identified in the subfractions, and azabenzofluorene, azabenzopyrene, and azaanthanthren~ were identified in the azaarene group. They mentioned that the mutagenic activity of PAAs is greater than that of azaarenes. Katz and Oganl7 identified chromatographic peaks of benz~aJanthracene, dibenz~ah]- anthracene, and benzotghi~perylene in coal liquid (EPA Chemical Repository Samples CRM-1-3~. Earlier, White et al.,49 during the process of development of gas-chromatographic analysis, used a coal-liquefaction sample from the Synthoil Process Development Unit, Bruceton, Pennsylvania, and identified fluorene, 9,10-dihydrophenanthrene, 1,2,3,4,5,6,7,8-octahydro- phenanthrene, 1, 2,3,4-tetrahydrophenanthrene, phenanthrene, anthracene, fluoranthene, and pyrene. Nichols et al.34 studied raw gases from fixed-bed reactors fueled by different coal or vegetative fuels and collected in a stainless-steel cooler-condenser during the run. The gas-chromatographic analyses of the condensate are reported in Table 2-2 as micrograms per gram of solid feed to the gasifier unit. USED ENGINE OIL The 1972 NRC study26 on particulate polycyclic organic matter (POM) reported findings in the literature that BaP emission increased as the vehicle aged and oil consumption increased from 1,600 mi/qt to 200 mi/qt and that the BaP preferentially concentrated in the crankcase. The report did not list any other POM in used oil. There are numerous analytic ~roblems in isolating and analyzing for PAHs in used engine oil. Lee et al.2 sampled o~! taken from the oil pans of four randomly selected 4-, 6-, and 8-cylinder automobiles. The qualitative results of high-resolution (capillary) gas chromatography showed peaks for fluorene, phenanthrene, anthracene, 4-5-methylene, 9-methylphenanthrene, fluoranthene, pyrene' 1-methylpyrene, triphenylene, chrysene, BaP, BeP, perylene, and dibenz~acianthracene. Peake and Parker35 estimated that 500 million gallons of used motor oil is not reclaimed, but is haphazardly discharged into sewers or onto wasteland each year. In analysis of motor oil by gas chromatography-mass spectrometry, they found a predominance of alkyl-substituted aromatic compounds and 11 alkylfluorene isomers. Table 2-3 lists identified compounds and amounts per mil liliter of oil, based on the detector response to perdeuteroanthracene. 2-3

OCR for page 1
PETROLEUM AND OTHER FOSSIL-FUEL COMBUSTION FOR HEAT AND POWER GENERATION The data presented below show the individual PAHs present in emission from several sources of combustion effluents. Compounds known to be carcinogenic are identified in the tabulations of some of the data. A 1980 NRC study30 recommended that future research should continue to monitor such emission and measure (in a mass-balance study) the contribution of known carcinogenic or mutagenic compounds to the environment. There is a need for specific-site and broad-scale mass-balance studies of the release of PAHs into the atmosphere. The highest PAH emission rates in heat- and power-generation categories given in the 1967 review by Hangebrauck et al.14 were associated with small, domestic, coal-fired furnaces used to heat single-family homes. The emission from oil-burning was generally much lower than that from coal-burning and slightly higher than that from gas-fired units. The emission rates for 10 PAH compounds from under-feed stokers and hand-stoked coal furnaces were higher by several orders of magnitude than those from coal-fired power-plant units, as shown in Tables 2-4 and 2-5, respectively. In Table 2-6, the emission rates for inte~=ediate-sized coal, oil, or gas units using different firing methods show that the under-feed coal-stoker units emit all 10 PAHs at the higher rates. The EPA Industrial Environment Research Laboratory, Research Triangle Park, N.C. (IERL-RTP), developed the source-assessment sampling system (SASS) train for collection of gaseous, particulate, and volatile exhaust matter. The SASS was used to determine emission data on 74 inorganic trace elements and 21 PAHs in the effluent of 11 industrial coal-stoker-fired boilers. These 11 units represent a wide range of designs, which reportedly have changed very little over the last 20 yr. Three units were spreader stokers with reinfection from the dust collector (Table 2-7), three were without reinfection (Table 2-7), and five were mass-fired over-feed stokers. The study was done by the American Boiler Manufacturers' Association under the joint sponsorship of the U.S. Department of Energy and the EPA by Burlingame et al.6 There were 23 SASS tests conducted, and the emission data were presented in three units: nanograms per joule of energy input, micrograms per dry standard cubic meter of flue-gas samples, and micrograms per kilogram of fuel input. The PAH totals of all the SASS tests were reported, but only the average emission in micrograms per standard cubic meter of effluent and in micrograms per kilogram of fuel is listed in the table. PAHs FROM COKE PRODUCTION The EPA's July 1981 report46 on background information for a proposed standard on coke-oven emission gives a comprehensive overview of emission and locations of coke plants, as well as a cancer risk assessment. Only some of the literature showing the PAH compounds found in the emission from coke ovens and the relationship to particle size is discussed here. 2-4

OCR for page 1
Lao et al.l9 reported the results of sampling emission from coke ovens in the steel industry. Two samples were collected on glass-fiber filters and two on 0. 8~ m-pore silver membrane filters, extracted for 24 h in Soxhlet extractors, and measured in GC/MS and GC/FID systems. The results are given in Table 2-8 in micrograms per gram of extract. Some of the early studies on carcinogenicity of coke-oven effluent (as early as 1875) were reviewed by Hoffmann and Wynderl5 in 1976. The relative risk of developing lung cancer for men employed 5 yr (1951-1955) or more at the full topside of the ovens from exposure to coke-oven effluent was 6.9 times the expected (the prediction was 2.5 times greater than for the general population); an unexpected finding of kidney-cancer incidence 7.5 times greater than in the general population was also reported. Several reports have discussed the PAR emission from coke ovens in other countries: Norway,4 Finland,41 Czechoslovakia,21 Canada, 37 and Brazil.24 In studying concentrations of PAHs on particles of the various sizes, Bjorseth found the greatest amount on the particles of 0.9-3 Am and only about 1% of the total on particles larger than 7 Am (see Table 2-9~. The particles were collected at the top of a coke-oven battery, fractionated according to particle size by a Lundgren impactor, and analyzed in a alass-capillarY GC/MS system. Miguel and Rubenich24 gave the concentrations of BaP found on particles in various size ranges (collected with an eight-stage low-pressure impactor that separated by aerodynamic diameter) from an urban automobile traffic tunnel, from ambient air, and from the bench (push side) of a steel-mill coke oven (see Figure 2-1~. ~ , _ COAL MINING The PAHs found in respirable coal-dust samples were identified as phenanthrene, pyrene, benzotghi~fluoranthene, chrysene, perylene, benzoperylene, benzochrysene, and dibenzoperylene. These compounds were more common in coal dust in the mine than in dust from other locations. The OTA report39 of 1979 stated that the long-term exposure of coal miners to these suspected carcinogenic compounds had not been analyzed. Shultz et al.40 reported finding 13 PAHs in the respirable fraction of mine dusts. WOOD-BURNING FOR HEAT AND POWER GENERATION As the prices for home heat ing with electricity ~ gas ~ and oil increase and fuel availability is threatened, many home owners are using wood or coal stoves as a supplement for space heating. Duncan et al.ll estimated an increase of 40,000 wood-burning stoves in 201 counties of the TVA power-distribution area from 1974 to 1976. According to DeAngelis et _ .,10 the U.S. Bureau of the Census data showed that 452,000 new homes had fireplaces and that 550~000 wood-burning stoves were shipped by manufacturers in 1975. Owing to the difficulty in achieving controlled 2-5

OCR for page 1
combustion in fireplaces and wood- and coal-burning stoves, there is often not an efficient burn; consequently, there is a need for more frequent cleaning of chimneys. Chimney-cleaning equipment is being sold for use by individual homeowners, and the occupation of chimneysweep has become prominent once again. Hazards of exposure to the particulate matter in chimney-cleaning are recognized to be associated with not only skin exposure, but also inhalation. The carcinogenic health hazards associated with chimney-cleaning were reviewed by Bagchi and Zimmerman.1 The number of housing units burning wood was estimated by DeAngelis et al.,9 using 1970 U.S. Census of Housing data in conjunction with the 1976 Housing Survey. The state-by-state tabulation showed totals of 912,000 units burning wood as the primary source of heat and 35, 467, 900 burning wood for auxiliary or aesthetic purposes, wi th an estimated consumption of 5,122,000 metric tons a year for the primary units and 11,500,000 metric tons for auxiliary or aesthetic units. The range of POM emission from wood stoves and fireplaces was 0.01-0.4 and 0.02-0.04 g/kg, respectively. In 1981, Peters, a coauthor of the above work, estimated the annual emission of POM into the ambient air from primary heating units at 1,383 metric tons, from auxiliary units at 2,376 metric tons, and from fireplaces at 78 metric tons, for a total of 3,837 metric tons. The NRC report Indoor Pollutants29 assessed some of the sources of pollutants indoors and their effect on air quality. When wood stoves were in use, the BaP concentration monitored over 24 h indoors was 5 times higher than when stoves were not in use. In assessing the impact of wood-combustion emission on the environment, the U.S. Department of Energy, 44 in 1979, stated that the ma jar pollutants of concern from residential wood-combustion devices were unburned combustibles, carbon monoxide, particles , and hydrocarbons. Owing to the inefficient combustion in many home-heating units, large quantities of all of them are emitted. According to the 1980 Department of Energy report45 on health effects of residential wood combustion, the emission from such combustion is a major environmental problem affecting local air quality. POM is the most important group of organic compounds among the noncriteria substances emitted. There is no federal regulation of atmospheric emission applicable to residential space-heating units. In the preliminary EPA assessment of wood-fired residential combustion equipment, it was stated that the emission of organic substances, including POM, is relatively high, owing to the use of large pieces of fuel, highly resinous fuel, uneven fuel distribution, and hand-feeding in batches. The emphasis on use of wood-burning, air-tight stoves may greatly increase the magnitude of the emission problem. The PAHs emitted from wood-burning in fireplaces and baffled and nonbaffled stoves are shown in Table 2-10. In a study of modifications of combustion to reduce emission from residential wood-burning, EPA47 suggested several techniques to reduce the gaseous components in the emission. The following compounds were identified in the emission from wood-burning stoves that burned oak or green pine in either the up-, down-, side-, or high-turbulence-draft delivery systems. The up-draft system had the lowest total-particle and gaseous emission rates. 2-6

OCR for page 1
Nap~thalene Acenaphthene Acenaphthylene Fluorene Phenanthrene Anthracene Methylanthracene/methylfluoranthenes Fluoranthene Pyrene Methylpyrenes/methylfluoranthenes Benz~aJanthracene Chrysene Methylchrysenes Dimethylbenz~aJanthracene Benzofluoranthenes Benzo~e~pyrene Benzota~pyrene Perylene Indeno[1,2,3-cd~pyrene Benzotghi~perylene Coronene Dibenzo~ahipyrene PAH emission rates under controlled burn conditions were characterized by Rubble et al.,16 and their results are shown in Table 2-11. The relationship of ambient air concentrations of PAHs to the sources of wood-burning has been studied by several investigators. In some circumstances, wood-burning can be the largest combustion source for various atmospheric pollutants in urban areas. For example, the contribution of wood-burning emission products to atmospheric PAHs in Telluride, Colorado, has been estimated by Murphy et al.25 Telluride is a small community in a valley with poor ventilation and with large temperature inversions. It depends heavily on wood-burning as a residential heating source and has only light automobile traffic. In this community in 1980, the BaP concentration in the air reached 7.4 ng/m3, which exceeds several times over that which is found in a number of U.S. metropolitan areas, such as Los Angeles, and Telluride is not unique in this respect. The mechanisms of PAH formation during the combustion of wood are poorly understood. It is known that wood contains substantial amounts of alkylbenzene derivatives, which contribute to the foundation of the PAHs. The latter reactions depend markedly on temperature: no important formation of hydrocarbons occurs below 450C. Approximately 75 organic compounds have been identified in flue-gas samples of the identified organic substances, and the PAHs make up about 35% of the mass. The PAHs that are produced during the pyrolysis of wood and are found in the smoke include anthracene, phenanthrene, dibenz~aj~anthracene, dibenz~ah~anthracene, fluoranthene, benzotghi~fluoranthene, benzotb~fluoranthene, benzotciphen- anthrene, benzotghi~perylene, pyrene, BaP, BeP, 3-methylcholanthrene, dibenzotcgicarbazole, dibenzotai~carbazole, cyclopentatcd~pyrene, and some methylated substances. A detailed study of the effects of wood type, degree of seasoning (e.g., the moisture content), and type of wood stove on the emission of PAHs in smoke has been conducted. 36 A fireplace, a baffled and a nonbaffled wood stove, and seasoned (4-5: moisture) and green (27-30% moisture) oak and pine were used. The results are indicated in Table 2-13. Wood smoke generated in a fireplace contained much less PAH than that formed in wood stoves. No difference in relative distribution of particular PAHs was noted when green and seasoned wood were burned in the f ireplace (data not shown). Furthermore, neither wood type nor extent of baffling significantly influenced the pattern of PAH emission in the wood smoke. 2-7

OCR for page 1
Peters36 has compared the emission of PAHs from several residential combustion sources as a function of thermal efficiency (Table 2-14~. Wood-fired heating resulted in a much higher output of PAHs than did coal-, oil-, or gas-fired furnaces; i.e., the mass of PAHs emitted per joule was 10, 5,000, and 30,000 times greater, respectively. REFUSE BURNING AND INC INERATION Before the passage of the Solid Waste Disposal Act in 1965, numerous municipal dumps practiced open or uncontrolled burning throughout the United States. Owing to the standards established under this Act, most of the large incinerators then in operation were shut down, largely because upgrading them would have been expensive. Congress passed the Resource~Recovery Act in 1970 and the Resource Conservation and Recovery Act in 1976. As a consequence of the passage of these acts, numerous demonstration projects for solid-waste disposal and energy and resource recovery were initiated. The concluding remarks of the 1981 NRC report The Recovery of Energy and Materials from Solid Waste27 stated "that the technologies for energy recovery were still under development and that the most highly developed and least risky was mass burning, but that other technologies were being tested." The solid waste from residential, commercial, and institutional sources amounted to 130 million metric tons in 1976 and is projected to increase to 180 million tons by 1985.48 In EPA's 1971 review42 of the literature on municipal incineration, the findings of Hangebrauck et al.14 (Table 2-15) showed the emission of 10 PAHs from municipal and commercial units burning wastes from households, grocery stores, and restaurants. In 1976, Davies et al.8 reported that the measured emission of PAHs was similar to that found by Hangebrauck. They used a modern, continuous-feed municipal refuse incinerator rated at 9.14 tone of refuse per hour with a water-spray cooling tower (cooled to 250-300C) and an electrostatic precipitator in the flue-gas stream. The analytic results on samples collected after the electrostatic precipitator are given in Table 2-16. The findings by several investigators on the emission of chlorinated dibenzo-p-dioxins in waste incineration have caused some concern about the buman-health implications of this group of compounds, but discussion of dioxin emis s ion is beyond the scope of this report. METAL PROCESSING The stack gases from a smelter processing lead from batteries were sampled. The PAHs and their concentrations found in four samples are shown in Table 2-17. The polymeric organic battery casings were included in the process and presumed to be the contributing source of the organic emission. Lao and Thomasl8 identified several PAR compounds in the 2-8

OCR for page 1
particles collected on glass-fiber filters in the exhaust flue from the "pot room" of a nonferrous-metal production room. Particles from iron foundries in Finland were analyzed for PAHs by Schimberg,38 and the following were identified: phenanthrene, anthracene, fluoranthene, pyrene, benzo~a~fluorene, benzoiciphenanthrene, benzofluoranthenes, BeP, BaP, perylene, o-phenylenepyrene, dibenzanthracenes, benzochrysenes, and benzotghi] perylene. NATURAL SOURCE S The mechanism whereby complex mixtures of PAHs occur in natural deposits of peat, coal, crude oil, and shale oil is unknown. Neff33 discussed the sources of PAHs in the aquatic environment and their possible biosynthesis by bacteria, yeasts, and higher plants. Although some publications apparently provide evidence of biosynthesis, others refute it. Until relatively recently, it was assumed that PAHs are formed only by pyrolysis of organic material. However, the finding of 1-methyl-7-isopropylphenanthrene (retene) in pine tar and the isolation of PAHs directly from plant material have helped to keep alive the uncertainty of PAN biosynthesis in plants. FOREST FIRES Forest fires are sporadic and sometimes uncontrollable occurrences that apparently contribute significantly to the PAHs found in the atmosphere. Some laboratory data are given below on the amount of PAN emission. The discussion in Chapter 3 of physical removal of PAHs from the atmosphere mentions forest fires as a source of atmospheric PAHs. This source is mentioned here, because it has the potential for being a significant contributor; however, data are insufficient for an assessment of the impact on atmospheric quality. The 1976 NRC report Air Quality and Smoke from Urban and Forest Fires28 described "prescribed forest fires" as fires set to reduce the amount of secondary plant or roughage undergrowth in the hope of reducing the incidence of wild fires, for disease control, and for other management purposes. In the 1976 report, it was noted that wild forest fires consume about 3 times as much fuel as prescribed fires and produce about 3 times as much particulate matter per ton of fuel burned as prescribed fires. On the average, prescribed fires consume 3 tons of fuel per acre (range, 1-10 tons/acre) and wild fires 9 tons/acre (range, 1-50 tons/acre). Emission from forest fires varies widely, owing to the variety of fuels, fire type (heading fires, with the fire line moving with the wind, or backing fires, with the fire line moving against the wind), fire intensity, and combustion phase (flaming vs. smoldering). The results of a screening experiment reported by McMahon and Tsoukalas,23 using slash pine needle litter as fuel, are shown in Table 2-18. The PAHs present in the particles show that heading fires produce higher total amounts of particles, but smaller total amounts of PAHs. 2-9

OCR for page 1
AQUATIC ENVIRONMENTS (FRESHWATER AND MARINE) In its twenty-fifth report to Congress, the Committee on Governmental Operations43 considered groundwater destruction one of the most serious environmental problems of the 1980s. It estimated that 50 billion gallons of water are placed in industrial surface-water impoundments throughout the country each year. The report stated that tens of millions of Americans obtain their water from private wells, because no public water supply is available. A high percentage of PAHs is removed from municipal water by flocculation and sedimentation of suspended particulate matter. The chlorination process removes a large percentage of residual PAN contamination (the amounts removed by this process are discussed elsewhere in this report). We do not have national survey information on the incidence of well-water contamination by PAHs, and this route of human exposure has not been assessed. But it is of potential importance, in that most well-owners do not treat the water before use, on the assumption that it is uncontaminated. Owing to the large number of industrial wastewater impoundments that are unlined (no plastic liner) and are directly over groundwater sources, there is a need for investigation to determine whether there is PAR contamination. (See the section of this report on human exposure for discussion of PAHs in drinking water.) In the comprehensive review by Neff33 on PAHs-in the aquatic environment, the following sources of pollution were listed: industrial and domestic sewage effluent, surface runoff from land, deposition of airborne particles, and spillage of petroleum and petroleum products into water bodies. Specific PAHs from these sources are discussed earlier in this section; but Neff's work includes more detail on the effects on the aquatic environment. The 1975 NRC report Assessing Potential Ocean Pollutants32 discussed the effects of several sources of pollution, e.g., ocean discharge of dredge spoil, municipal sewage sludge, petroleum discharge, and spills. Although that report did not discuss the effects specifically of PAHs, the sources of pollution described in it have been identified as sources of PAHs. An NRC committee is in the process of reassessing the pollutants and their effects on the marine environment. In 1974, owing to the increasing amounts of petroleum being transported over the world's waterways, the need for a comprehensive, international marine environmental monitoring assessment program led to a symposium and workshop on marine pollution monitoring held at the National Bureau of Standards in Gaithersburg, Maryland. At the symposium, deep concern was expres sed about the amount of petroleum pollution observed in the marine environment and its effects and ultimate fate. Among those expressing concern were representative of several national and international organizations with interests and oversight for pollution problems in the marine environment, such as the Special 2-10