Document z88NgQOjBzvRL3XY0EKYY0p0

700182 Automobile Emissions -- A Study in Environmental Benefits versus Technological Costs P. S. Myers 1969 SAE President Dept, of Mechanical Engineering, University of Wisconsin Automobile Division Council of Institution of Mechanical Engineers St. Mary's Hali, Coventry, England November 4, 1969 Science Academy of the Socialist Republic of Roumania Bucharest, Roumania November 20, 1969 SOCIETY OF AUTOMOTIVE ENGINEERS, INC. V 700182 Automobile Emissions -- A Study in Environmental Benefits versus Technological Costs P. S. Myers 1969 SAE President Dept, of Mechanical Engineering, University of Wisconsin THE PROBLEM of air pollution has existed ever since our ancestors sat coughing around a smoky fire in a recessed cave. In fact, there probably never was an unpolluted atmo sphere, since decaying vegetation and animal matter, forest fires, dust storms, volcanoes, and other natural phenomena surely have emitted gaseous and particulate matter ever since the world began. Fortunately, our atmosphere is continually being cleansed by natural means--for example, removal of particulates by settling and by precipitation and consump tion of carbon dioxide by growing plants. Thus, from the. beginning of the world the condition and concentration of many of the components of our atmosphere has been influ enced (and perhaps determined) by the relative rate of these additions and removals. In the past the rate of these additions has been determined primarily by natural events. Consequently, over most of geological time the condition of the atmosphere has not changed rapidly, except for major isolated events such as volcanic eruptions, etc. Inside the last century, however, man-caused atmospheric additions have become increasingly important--possibly even to the extent of almost determin ing the trace constituents of the atmosphere in many places. The increase in man-caused additions has resulted from dramatic increases in both population and industrialization in the past few centuries. Fig. 1 (1)* shows the population increase as a plot of popu lation versus time--note the logarithmic scale as well as the extrapolation to the year 2000. The number of "contribu tors" to the atmosphere is increasing at an increasing rate! At the same time, especially in the United States but also in the world as a whole, the "contribution" per person has also increased dramatically. Fig. 2 presents a plot made by -Magill (2) showing one "contribution" criteria: energy con sumption per capita in the United States as a function of time. Fig. 3 shows the energy consumption for the world as a whole (3). You will recognize, of course, that energy produc tion "contributes" to atmospheric pollution both directly and indirectly. Not only are our number of "contributors" increas ing dramatically, our "contribution" per person to the atmo sphere is increasing dramatically also. As a result of these trends, there is well-founded concernand considerable evidence-that these dramatic increases in man-caused additions to our atmosphere are markedly chang- *Numbers in parentheses designate References at end of paper. ABSTRACT Recent increase in man-caused additions to the atmosphere are markedly changing its quality. With rising population, these additions will increase in the future. This paper dis cusses tire magnitude of the overall problem in relation to mass rate balances for particulate matter,carbon dioxide, ni trogen oxides, sulfur dioxide, lead, and hydrocarbons. Pollu tion from automobiles is then discussed specifically, along with some possible solutions (both with the existing recip rocating engine and alternate power sources), the effect of controls, and the long-term reduction of emissions. Methods of cooperation between governmen^manufac turers, and the public are suggested. DUP050311936 Fig. 1 - Population growth * vori4 1*et eoftMmplIcn | 2.(S.rf2 ewuitfllM* lotal(BCrartMOl q ommql twnimption 0,1*10 J u (30x10*4 par eeptio) -- tipKltdOHtMlemimptiantySOOAO&SxlO^j gw 4800*10*4 p* *pKq) 2 MOO 1600 YEAR Fig. 3 - Annual fuel consumption for world TIME OF DAY Fig. 4 - Diurnal variation patterns of NOx Fig. 2 - Energy use per capita in United States, 1875-197S ing its quality. This concern is highlighted by evidence such as recurrent "photochemical smog" in Los Angeles, the deaths of 4000 people as a result of the 1952 London Fog, or 20 people in the small town of Donora, Pa., in 1948, the "Yo kohama asthma" experienced by American Troops stationed in Japan after World War II, and the rapid growth in the phe nomena of lung deterioration called emphysema. About 20% of the energy consumed in the United States"is" used in automobiles (3). Thus, the automobile has been, and is, still under severe attack as a "polluter." Some people feel the criticism is too harsh and that significant advances have been made (4); others feel that the situation is urgent and immediate action is required (5). However, we, as automobile engineers, should not be too discouraged, for it has been shown that pigeons (6) also contribute significantly to air borne particulate matter in New York City. In any event/we., as engineers concerned with technical and technological .feasi bilities and relative costs, have a special interest and role to play in the problem of air pollution--so let's try to assess the magnitude and urgency of the problem. After having done this, let's next look at the automobile as a source of pollution and what can and is being done about it. AIR POLLUTION-URGENCY AND MAGNITUDE OF PROBLEM Air pollution is a very complex problem and much of the data required for realistic evaluation are missing. First of all, air pollution varies with time at a fixed point (both daily and annually) and has long-term trends as well. Furthermore, these.trends may vary with time in a different way at different locations as shown in Fig. 4 (7). Thus, one must be concerned with the maximum value of a pollutant as well as time and space averages. At the same time, at the low concentrations found in the atmosphere, the effects of many pollutants on health is not well established. Fundamentally, there is a conflict today between man's technological and biological needs and because of insufficient . data, judgments and evaluations are involved. MASS RATE BALANCES - Our atmosphere--like any lifesustaining environment-is self-cleansing with contaminants continually entering and leaving the atmosphere. Thus, the concentration of many of our pollutants is affected by the mass rate of addition and removals; in a simple sort of way it is like a pail of water with a hole in it--under steady-state con ditions the level of water is determined by the rate of addition and removal of water, but during times of change' the size of the pail is of interest also. Let us then look at the mass of the pollutant in our atmosphere as well as the mass rate of addi tions to and removal from the atmosphere. In looking at the mass rate of additions, we recognize that they can arise from natural or from man-caused events. How ever, the rate of removal is primarily due to natural events. Since man-caused events are of fairly recent origin, we will particularly note the relative size of the man-caused additions in comparison with the size of the natural additions. On a global basis, the total mass of the earth's atmosphere is estimated as 5.7 X 10^ tons of major components plus selected trace constituents as shown in Table 1. However, my friends in meteorology tell me that the rate of atmospheric mixing between the northern and southern hemisphere is very DUP050311937 Table 1 - Mass and Lifetime ofSelected Atmospheric Constituents Lifetime Mass in Entire Atmosphere, tons X1015 Approx. Mass in 30-60 deg Atmosphere, tons X1015 Approx. Mass in Los Angeles Basin, tons X 1010 - Years Years Days Years Days to years Days N2 4.25 02 1.3 tons. X 10 C02 2800 CO NOx 0.6 0.009 CH4 5 tons X107 Particulate matter S02 15.5 0.25 0-55 0.17 tons X 109 - - 0.002 - tons X 107 3 0.05 2.66 0.82 tons X IO4 " -' 4 1 - tons X IO4 0.6 0.3 slow, in fact, they suggest that the portion of the atmosphere . lying roughly between 30-60 N latitude and below 30,000 ft (a little less than 20% of the. total)* be considered as an iso lated system except for very long times. Table 1 also shows the order of magnitude of lifetime in the atmosphere of selected trace constituents. It seems logical to consider the 30-60 N portion of the atmosphere as a system for those constituents having a lifetime measured in days and the entire atmosphere for those measured in years. Aside from obtaining the numbers, determining mass rate additions to the atmosphere poses problems. Much of the world's industry is located in the 30-60 region of the north ern hemisphere; eonsequently, mueh-of-the man-caused addi- tions probably originate within this section of the atmosphere even though they are normally given on a world-wide basis. However, natural additions and removals correspond more nearly to a worldwide basis. Some of the contaminants origi nating over the ocean may be entirely destroyed over the ocean and thus have but little effect on mankind. Very little precision is claimed for any of the numbers shown in the mass rate balances. It will be impractical to identify the source of every esti mate. The Air Conservation Commission (8), Sittig (9), Jaffee (10)? Magill (2), National Conference on Air Pollution (53), Erikksson (11), Kehoe (12), Stopps (13), California Depart ment of Public Health (14), U. S. Department of Health, Ed ucation and Welfare (15), Hovey (i6), Altshuler (17) Robin son (18), as well as many other sources, have been used. PARTICULATE MATTER - Natural events, such as meteoric dust, volcanic eruptions, dust storms, natural fires, etc., all contribute to particulate matter. However, human activities *As will become more evident later on, mass rate additions are known with very little precision. In fact, in some cases, they are really estimates. We will indicate this lack of preci sion by rounding of numbers. ADDITIONS (tons/year) 30-60* ATMOSPHERE NATURAL EVENTS -IO9 Ph i 5oxl5S------------------------- Noturol Put_____________ 6KI07 Malaoric DuIt _______ IX10* PorllcuJat* MAN-CAUSED EVENTS -IO InduttrtoT Out a Ah SxlOT tana 3xl07 Spacs Hiding_______ xIO Solid Wal Qippaial 30 microgram cu. malar X 10s Sulphur Coffipoundp 6 x IOT AutomobHpt______ ax io * REMOVALS Gravitation Impaction Ovotltt Nuatot Washing Fig. 5 - Mass rate balance for particulate matter contribute directly to particulate pollution and may cause in creased natural event.particulate pollution. Fig. 5 presents a mass rate balance for particulate matter. The inclusion of the gas sulfur dioxide jn the particulate mass balance needs explanation. A large percentage of the sulfur dioxide in the air oxidizes to sulfur trioxide which forms sulfuric acid mist. The sulfuric acid in the aerosols re acts with other materials in the air and forms, among other things, ammonium and calcium sulfate. The available evidence suggests that a substantial portion of the atmospheric sulfur dioxide is directly neutralized by ammonia, calcite dust, or one of the many other airborne alkalies and is then rapidly oxidized to the corresponding sulfates (19). Fog is included as a matter of interest rather than concern with it as a pollutant. Fig. 5 indicates that the man-caused additions are approxi mately equal to the natural additions and that both are larger than the mass in the atmosphere. This is in agreement with the estimate of Junge (20) that the mean life time of particu lates varies from a few days to a few years depending on par ticle size. DUP050311938 4 1 Fig. 6 - Self-contained dust dome of a laige indus trial city showing how air circulation patterns in tensify air pollution Remembering that the Los Angeles basin is not a fixed-mass atmospheric system (but nevertheless assuming a fixed-mass system), it is interesting to compare mass rates for that area. It is estimated (21) that there are 170,000 tons per year of particulate matter (including sulfur dioxide) descending upon tiie Los Angeles area per year. The mass of dust in the air i over Los Angeles is estimated at 6000 tons, that is, the annual ii mass input rate is some 30 times the mass in the atmosphere. There is considerable evidence that suggests that the density of particulate matter is increased over industrial areas by a self-contained dust dome as postulated by Bryson (22) and illustrated in Fig. 6. All the evidence indicates that only urban areas 6 miles or more in diameter have circulation patterns that sustain a dust dome. The dust dome is caused by the energy released by fossil and nuclear fuels in a city, by the increased absorption of solar radiation, and by the decreased Fig. 7 - Changes in dust concentration, COj concentration and world temperature with time. energy going into evaporation because the land is covered by pavement rather than vegetation. All of these combine to cause a chimney effect over the city with consequent air in flows as shown in Fig. 6. ............ .. _ ........................... ... Particulate matter has the obvious effects of decreasing visibility and increasing soiling and corrosion. From a health standpoint, it is estimated that an individual breathes about earth, that is, the greenhouse effect. Reasons for this are shown in Fig. 7 where the rising mean temperatures corre spond to rising C02 concentrations--and rising sunspot num- ber also!.......... " ~ -- ` ......................... -- Fig. 8 presents comparative data on tonnages of CC>2 in volved. Note the logarithmic scale. The total amount of CC>2 1 mg suspended matter per day during times of heavy pollu tion (23). The larger particles are undoubtedly collected by the mucous lining, but the smaller particles usually reach the deeper structure of the lungs. There is also considerable concern that increased particulate added to date by combustion is not large: say, 10-15% of that present in the atmosphere. It does appear, however, that ap proximately one-half of the CO2 added by combustion is cur rently being stored in the atmosphere. Since the total CO2 concentrations in the atmosphere may cause significant weather modifications. Bryson (24) has made the plot shown that could be added, if all fossil reserves were burned, is some 5-10 times the CO2 in the atmosphere, there is an ultimate in Fig. 7 illustrating the changes in CC>2 concentration and potential for significant changes of the atmospheric concen dust concentration from 1880 to 1960. He suggests that the tration of C02. data indicate that since 1930 the increase of dust has begun to dominate the carbon dioxide effect, terminating the warming period and producing a decline in world temperatures. To summarize, man-caused particulate mass rates are begin ning to be of the same order of magnitude of those caused by natural events and on an annual basis both are larger than the Fig. 9 shows a mass rate balance for C02- The primary source of C02 from natural events is decay of organic mate rial, although respiration amounts are large and will increase as the population increases. Note that combustion is the. pri mary source of man-caused C02 and that the natural and man- storage in the atmosphere. CARBON DIOXIDE - Carbon dioxide is not normally con caused amounts are of the same order of magnitude. One big removal source is, of course, photosynthesis by sidered as a pollutant--it is essential for plant life. There has, plants. Another big source is the ocean, but absorption rate however, been increasing concern about its potential for modi data are lacking. The potential is there--the ocean could hold fying our climate (25) by modifying the energy balance of the some 50 times the amount of the C02 in the atmosphere. DUP050311939  ' s' --------------------------. '-r-i-----------r-r.---------J-- ADDED TO DATE-COMBUSTION | gg 1880-290 PPM In ATMOSPHERE | |S " ^ 1969-325 PPM in ATMOSPHERE ............. . 1880rI860 STORAGE | ESTIMATED~ADDD by 2010 ADDITIONS -Kini/yr 1 s S' = f? S S g-S-2 o- | 3 5 ** 3 3P f * 2 2 POTENTIAL-COMBUSTION of ALL FOSSIL RESERVES -------- * *- ... ---- ------- -----L. .. 1.....1. | 1 .. Fig. 8 - Tons of CO2 in and added to atmosphere ADDITIONS (lons/year) ENTIRE ATMOSPHERE REMOVALS NATURAL EVENTS Total-2*K>,r Volconoo*_____________ I* 10 Otcny of Orqonic Mower 2*10" Respiration tx I0 MAN-CAUSED EVENTS Total -- 2 x 10 Combustion 2*10 Industrial 2 x I0 C02 3 x I012 ions 329 PPM Photosynthesis 2 x iq" Absorption in Ocean Weathering of Rocks Fig. 9 - Mass rate balance for CO2 However, the oceans circulate slowiy (26) and CO2 can be stored (and released) by carbonates (27) in the sea water. Thus, the best estimates are that the CO2 concentration will continue to increase and could become large enough to cause 'Climatic changes' CARBON MONOXIDE - Aside from CO2, carbon monoxide Fig. 10 - Correlation between carbon monoxide concentration and traffic patterns during day ADDITIONS (Ions/year) ENTIRE ATMOSPHERE REMOVALS NATURAL EVENTS Toiol ~2 x 10 MAN-CAUSED EVENTS Toiol - 2x10 Automobiles 2X108 Industry______________ 4 x 10* Poxer Plant!_________ 2 x 106 Space Heating_______ 4 x 10* Solid Waste Dlepoeol ... 2 x 10s Carbon Monoxide 6x 10 tone 0.1 PPM To COg in Atmosphere Dissolved in Water Consumption by Vegetation Unknown Sinks Fig. 11 - Mass rate balance for CO is the most abundant and widely distributed pollutant. Very little CO is due to natural events, almost all is from man-caused events. Approximately 90% of man-caused CO is from auto mobiles. Thus, it is not surprising that the correlation between carbon monoxide and traffic is excellent, as shown in Fig. 10 (28). However, some carbon monoxide does come from natural events such as marsh gas, coal mines, seed germination, lightning storms, injured vegetation, forest fires,-etc. (10). Destruction of carbon monoxide is almost entirely by natural events, all poorly understood and possibly unknown. Jaffe (10) states, "The removal process or processes cannot, at the present time, be identified with certainty." The dura tion of CD in the atmosphere is equally uncertain and has been estimated as low as 0.3 years (20) and as high as 5 years (29). An estimated mass rate balance is shown in Fig. 11 for carbon monoxide. Because of the uncertainties, no mass rate of de struction is shown. The mass rate additions are beginning to approach the order of magnitude of the mass in the atmosphere. On a local basis the figures are even more striking. In Los Angeles, for example, about 4 X 10^ tons per year of CO are added to the atmosphere (21). If the Los Angeles basin is again assumed to be a closed system (during temperature in versions, this might be a better assumption than first thought) the Los Angeles atmosphere would have of the order'of mag nitude of 4 X 104 tonsofCO in it, that is, the annual additions would be some 100 times the mass of the carbon monoxide in the Los Angeles atmosphere. The Air Conservation Commis sion (8) found that in Los Angeles the time-and-space-averaged concentration of carbon monoxide increased from about 7 ppm to about 11 ppm from 1957 to 1963, which is not sur prising even considering that Los Angeles is not a closed sys tem. Since 1966, when control devices on automobiles were instituted, the concentration of CO in Los Angeles has been slowly dropping (30). Let us next look at carbon monoxide from a health stand point. Carbon monoxide is dangerous because it has a strong affinity for hemoglobin, which carries oxygen to body tissues. Thus the effect of carbon monoxide is to deprive the body tis- DUP050311940 I!a WMIt cciuipgl' ..SiTii 400 PPM . 6 PERCENTCAR80XY HEMO lGOBtN . 0.1 HOURS EXPOSURE 100 Fig. 12 - Effect ofCO concentration and time on caiboxy hemoglobin sues of oxygen. Most American scientists believe that carbon monoxide is not a cumulative poison (31), with the blood being cleared of about half of the carbon monoxide in 3-4 hr for healthy subjects. However, some of our European col leagues (32) feel its effects maybe cumulative and not com pletely reversible. The effects of carbon monoxide involve both concentration and time. The effects are best shown as a plot of percent carboxy hemoglobin versus time as shown in Fig. 12 (33). For comparison, the highest average of recent carbon monoxide measurements in urban areas (28) was 38 ppm for 1 hr, 27 ppm for 8 hr, and 17 for 24 hr. None of these concentrations should give noticeable effects according to Fig. 12. However, short-term concentrations considerably higher than these have been reported numerous times.and places- For example, con-centrations in moving vehicles (34) of as high as 75 ppm for periods of at least 5 min with a median of 60 ppm weie found. Also some people would feel that Fig. 12 is too conservative. For example, the California Department of Public Health (35) indicated that exposure of 30 ppm for 8 hr may be a serious risk to the health of sensitive people. To summarize, carbon monoxide is primarily a man-caused pollutant, the removal mechanism is not known, present con centrations are clearly not a hazard to healthy people and probably in only a few cases present a health hazard to sensi tive people. In addition, many people, when smoking, volun tarily expose themselves to levels many orders of magnitude (42,000 ppm) higher than those under discussion (36). OXIDES OF NITROGEN - In addition to the molecular nitrogen found in the air, nitrogen is present as NO, NO2. NH3, N2O, NjO^, N7Oj , and possibly other nitrogen com pounds. However, of all of these compounds the only pollu tants that are significantly man-caused are NO and NO2. tion of nitrogen compounds. For example, the mass rates of NH-j circulation are some 10 times those of the two oxides of nitrogen. Nitric oxide is formed at high combustion temperatures with the exact amount primarily dependent upon the temperature, oxygen concentration, and time. In the atmosphere nitric oxide reacts with oxygen and is rapidly converted to nitrogen dioxide. This reaction can be greatly accelerated by the pres ence of sunlight and organic material in the air. Most atmospheric determinations of oxides of nitrogen com bine together nitric oxide and nitrogen dioxide and designate the combination as NOx- Typical urban atmospheric concen trations range 0.02-0.9 ppm (37), although concentrations as high as 3.9 ppm have been recorded in Lbs Angeles. By way of comparison, tobacco smoke contains 250 ppm (31, 36). In addition to causing a reduction in atmospheric visibility, nitrogen dioxide also has an affinity for hemoglobin. How ever, because it forms acid in the lungs, it is considerably more toxic than carbon monoxide for the same concentrations. For example, a tentative standard of 5 ppm for an 8 hr working day has been set and some people think this is too high. A mass rate balance for oxides of nitrogen is shown in Fig. 13. The natural event addition of 2 X 10 tons,by biological reactions, was justified by Robinson (18) as necessary to have a reasonable residence time in the atmosphere. This amount is surprisingly large with respect to man-caused additions and is quite large with respect to the mass in the atmosphere. This gives mass rate additions some 100 times the mass in the atmo sphere. A local mass rate balance around Los Angeles (using our inaccurate closed-system assumption) gives a different picture. The concentration in the Los Angeles atmosphere is some 100 times higher than the world average, giving a mass in the Los Thus, we shall limit our attention to the so-called oxides of nitrogen. It should be realized, however, that these two oxides of nitrogen represent only a small fraction of the total circula Angeles atmosphere of 1 X 10^ tons--this is about the same as ' the natural event biological reaction contributions (assuming uniform distribution). On the other hand, if the man-caused oxides of nitrogen were uniformly pr&duced throughout the DUP050311941 7 ADDITIONS (tons/year) NATURAL EVENTS-2XI08 Biologicol Reactions 2 x 108 MAN-CAUSED EVENTS Total ~5x I0t Automobiles_____________ 2 X IO7 Industry__________ 2 x I07 Power Plants__________ 8 x 106 Space tieolim________ 2 x 10 Solid Waste Oisposal 2 X I09 30-60* ATMOSPHERE Oxides of Nitrogen Z x 106 tons I1 PPB REMOVALS Removal os Nitrates by Precipitation Vegetation Fig. 13 - Mass rate balance for N0X ADDITIONS (tons/year) NATURAL-Negligible Volcanoes MAN-CAUSED EVENTS Total--lx 10 Automobiles__________ 2 x 106 Indmlry l x I07 Space Heating 6x106 Solid Waste Disposal 3 X 106 ' Power Plants 5 x lot 30-60* ATMOSPHERE REMOVALS Sulphur Dioxide 5 xio9 tons Z PPB Precipitation CrovitoHon - Fig. 14 - Mass rate balance for SOj world, there should only be some 3X10^ tons/year produced in Los Angeles-in actuality, there are 3.5 X, 105 tons/year (21) produced in Los Angeles--a factor of approximately 1000. It is not surprising that the concentrations of oxides of nitrogen are higher in Los Angeles than the average concentration for the world. ' In summary, on a global basis it appears that man is con tributing but little to the concentrations of oxides of nitrogen. However, on a local basis where man-caused contributions are grouped together, concentrations can be hundreds of times . greater than the average atmospheric concentration, giving severe local problems. SULFUR DIOXIDE - Atmospheric sulfur is primarily in three forms: SO2, I^S, and sulfates. H2S comes primarily ADDITIONS (tons /year) NATURAL EVENTS Putt Voleonoo Combustion S>o Spray MAN-CAUSEp EVENTS* Automobile Exhaui 2 x I0 Pewlddw_______ Cool Burnino______ incineration______ Indmlry Point Attrition Storoge Battery 4 x 108 30-60* ATMOSPHERE REMOVAL -'Prgclpilotton Fig. 15 - Mass rate balance for lead from natural sources; S02 comes primarily from man-caused events; the sulfates come primarily from sea spray and from oxidation of S02- Thus we shall consider only S02, even. though it represents only a small fraction of the mass rate of the total sulfur mass rate balance. It should be appreciated, however, that the great majority of the sulfur in the atmosphere comes from natural sources. For example, estimates of amounts from natural sources ranging from 75% (8) to as high as 89% (11) of total sulfur mass rates have been given. Fig. 14 presents a mass rate balance for S02- It is estimated that 70% of S02 comes from the combustion- of coal. It is clear from Fig. 14 that mass rates are high with respect to the mass in the atmosphere. With regard to health, the peaks of S02 concentrations rarely exceed 11 ppm (38)--this is about the bottom of the range frequently used for health observations in the labora tory. For example, a concentration of 0.6 ppm will produce no detectable response in healthy human beings; most people can detect 5 ppm and most people find 10 ppm quite un pleasant. Note that concentrations up to 3.2 ppm have been measured in industrial sections of cities that use a great deal of solid fuel (37). In addition, there is considerable doubt as to . the accuracy of the measuring instruments used for sulfur dioxide (38). The situation is further complicated by the interaction and increased toxicity, insofar as health is con cerned, when sulfur dioxide, high air temperature, high air humidity, aerosols, etc.,are simultaneously present. In summary, the available evidence indicates to me that on a global basis man-caused contributions are not of concern. On a local basis, however, severe problems have existed and will undoubtedly continue to exist in the future. LEAD - Lead is one of the most widely used of metals, with storage batteries being the largest consumer. It is this wide spread use that makes evaluation of lead as an atmospheric pollutant very difficult. A schematic mass rate balance for lead is shown in Fig. 15, but few numbers are given. In the first place, I was unable to find reasonable estimates. Also, as will be explained later, as I studied the problem more I was not too certain the numbers would be meaningful anyway. Concern with atmospheric lead arises primarily from its use as a gasoline additive. Tests have shown fairly conclusively that lead concentrations in the urban atmosphere (as well as carbon monoxide concentrations) follow traffic density pat terns (39). The reason for this is shown in Fig. 16. Fig. 16 shows that approximately 70% of the lead used in your car is emitted from the tail pipe with 30% settling almost imme diately to the ground and the other 40% becoming airborne and causing concern. Airborne lead and its effect on human beings must be con- DUP050311 942 8 IOQ% LEAP CAR Approximately 40% Airborne 30% Immediate Settling' 30% Oil Drainage Pius Storage in Car >. 16 Schematic lead balance for an automobile ADDITIONS AIRBORNE -JJ15-003 35-55% EOioI.d 35-50% Swallowed 7-13 % Absorbed FOOD and WATER /Hoy 5-10% Reaches Blood 25-35 mg/day SMOKING JS mg/cigarette STOHAGE Bon.-200-4000i?r^- Soft Tissue*10*280 K)0 gm REMOVALS EXHALED AIR 35-55% of Inhaled Lead URINE 0I-j O4 mg/doy FECES J-.4 mg/day PERSPIRATION Eetimated same concentration at urine ADDITIONS (tons/year) ENTIRE ATMOSPHERE - NATURAL EVENTS-3 x10s ' 21*_____________ __ S XIOB Vegetation 2 * 10s ____________ MAN-CAUSED EVNTS~9x|07 Aulomobile Exhaust_______ 4 X 10* Industry _____________ Hydrocarbons tons CH4-5x I0* 1.5 PPM ' no data on Combustion of Non-Aulomolrva Fuel moss of'other I x I07 " hydrocarbons Evaporation 8t Transfer I * ICT incinerators _______ ' z * ioT ' - a f miu i&i RE"OVAL Oxidation Precipitation Fig. 18 - Mass rate balance for hydrocarbons major concern is with the reactive hydrocarbons rather than simply total hydrocarbon emission. Fig. 17 - Mass rate balance foi lead for an individual AIR POLLUTION FROM AUTOMOBILES I hope that I have convinced you in the previous discussion sidered in the proper context, however. Fig. 17 shows an ap that air pollution is a current problem and that you either as proximate mass rate for a human being. Inspection of Fig. 17 an individual or a professional engineer, will be concerned shows that the airborne lead is small in mass in comparison with the problem and its solution. In addition, we automo with that taken in via food and water, but that a higher per tive engineers have a special interest and contribution to make centage of the airborne lead reaches the blood. It appears that since our product is a major contributor to the problem. the body has over a period of time, adapted to certain lead Table 2 presents a summary of air pollution from all sources levels determined primarily by "natural" events and that the in the United States in 1966. Note that the automobile con airborne lead from automobiles is an addition to this relatively tributed about 60% of the total mass of air pollutants. Table 3 \ large quantity. There is general agreement that there is no shows how this 60% was distributed between the different evidence to date of ill effects due to the present levels of lead pollutants. Let us look at the source in the automobile of f in the air. these different pollutants. HYDROCARBONS - Hydrocarbons are known to be a par One obvious source is blowby past the rings. This source 4 ticipant in photochemical smog and suspected as a contributor was estimated as containing from 25-33% of the contaminants to cancer. Smog was reported in Los Angeles during the days (40,41). However, all cars in the United States are equipped of its early exploration by the white man. Also, it has been with PCV valves and thus this source has been fairly well elimi inferred (57) that "The nature and concentration of or nated--at least in well-maintained vehicles (42). ganic vapors of an atmosphere are among the prime deter Fig. 19 presents information showing the realtive magnitude minants of its sensory quality of freshness." Thus, it seems of emissions from the carburetor, gas tank, and exhaust. The certain that natural events contribute significantly to "air source of emissions from the tank and carburetor are fairly quality" insofar as hydrocarbons are concerned. It must-be obvious and plans for their control are well underway, so let's remembered, of course, that the term hydrocarbon covers concentrate on understanding the source of the emissions in literally hundreds of compounds as opposed to previously disr cussed pollutants. It should also be realized that particulate matter may contain hydrocarbons. Fig. 18 summarizes the data that were found. Methane is the largest single source-it is thought to come primarily from bacterial decomposition which as a source seems relatively stable with time. Methane also has a relatively long lifetime. Vegetation contributes hydrocarbons also. The. automobile is the biggest single contributor to man-caused hydrocarbon pollution, although the contribution of incinerators is sur prisingly large. Man-caused events are not large with respect to methane emissions, but they are large with respect to rfatural sources of heavier hydrocarbons. In addition, in areas where photochemical air pollution is a serious problem, the the exhaust. It should be realized that the exhaust contami nants have both quantity (Fig. 19) and quality (Fig. 20); that is, smog forming potential. By either standard, the exhaust is by far the largest contributor. OXIDES OF NITROGEN - Fig. 21 (43) shows a plot of NO concentration versus the extent of expansion of the products of combustion in the cylinder. The curves shown are for dif ferent equivalence ratios and show the NO concentration that would be present if thermodynamic equilibrium were main tained during expansion. Also shown on the curves are mea sured exhaust levels of NO concentration at two different equivalence ratios. It is clear from Fig. 21 that the measured levels correspond more nearly to the NO concentrations pres ent at the beginning of expansion, rather than the NO concen- DUP050311943 9 Table 2 -Total United States Air Pollution, 1966 Source 10 tons/year % of Total Industry Poweiplants Motor vehicles Space heating Refuse disposal 23 20 86 8 5 16.8 14.1 60.6 5.6 3.5 Table3 - Motor Vehicle Pollution Pollutant . 106 tons/year Carbon monoxide Oxides of nitrogen Hydrocarbons Sulfur oxides Lead compounds Particulates 66.0 6.0 12.0 1.0 0.19 1.0 EMISSION SOURCES AT VARIOUS FUEL VOLATILITY (RVP) Fig. 20 - Relative magnitude of smog-forming potential of emissions tram a car IOOOC9000 COMPRESSION RATIO: 10= I FUEL: ISO-OCTANE <t> FUEL-AIR EQUIVALENCE RATIO EMISSION SOURCES AT VARIOUS FUEL VOLATILITY CRVR) Fig. 19 - Relative magnitude of mass of emissions from a car 0.2 0.4 0.6 O.B EXTENT OF EXPANSION 1.0 Fig. 21 - Variation of equilibrium NO concentration throughout expan sion . tration at the end of expansion. In other words, apparently . there is nonequilibrium during expansion insofar as NO is con cerned. One concludes from this that the NO concentration' measured in the exhaust should correlate reasonably well with the peak cycle temperature, recognizing that the peak tem perature varies with location in the combustion chamber. Fig. 22 shows a plot of peak cycle temperature versus spark advance as measured experimentally. Also plotted in Fig. 22 are exhaust NO concentrations in parts per million as a func tion of spark advance (NO data from Ref. 44). The NO con centrations at approximately stoichometric air-fuel ratios fol low the same trend as the peak temperature curve. The lowei set of data in the figure are for a rich mixture--they do not follow the same trend as the peak temperature curve. If the NO is primarily related to peak combustion tempera ture (and to air-fuel ratio as will be shown later), it may be In- 5000 4000 3000 2000 1000 0 Fig. 22 - Correlation between peak cycle temperature and NO concen* tration * N O -PPM DUP050311944 .4500 10 NDIR ANALYZER DATA MANIFOLD VAC. 8*H6. SPEED- 2000 RPM SPARK TIMING 30" BTOC AIR-FUEL RATIO IS.8:| EXHAUST FROM CYL5 500 OI . 8 16 24 32 40 4S DISTANCE FROM HEAD-MANIFOLD SURFACE (INCHES ) Fig. 23 - NO and HC concentration profile of exbaust from cylinder ferred that it is formed primarily in the bulk gases as opposed to being formed in a quench zone next to the cold combustion chamber walls. This deduction is confirmed by the data shown in Fig. 23 (44) where both NO and hydrocarbon con centrations are plotted as a function of distance from the head-manifold surface, that is, as distance downstream from the exhaust valve. For three-quarters of the cycle there is no flow in the exhaust line. When exhaust does occur, the first gases to come out will be those adjacent to the exhaust valve, which in turn will be followed by the large bulk of the gases, which in turn will be followed by the quench zone gases scraped from the cylinder walls as the piston moves up on the exhaust stroke. If the deduction that the NO is Formed pri marily in the bulk gases Is correct, the concentration of NO as a function of distance down the exhaust pipe should increase, reach a maximum, and then decrease. Thus, the data shown in Fig. 23 are in agreement with the hypothesis that NO is formed in the bulk gases. Inasmuch as the NO does seem to be formed in the bulk gases and at nearly the peak cycle temperature, let us next see how under equilibrium conditions the NO concentration varies with temperature, pressure, and fuel-air ratio. Fig. 24 (45) presents a plot resulting from equilibrium calculations. It shows NO concentration versus equivalence ratio with lines of constant temperature at two different pressures. A rich mix ture is shown in Fig. 24 as abscissa values greater than 1.0 and a lean mixture as abscissa values less than 1.0. Following a line of constant temperature, NO increases rapidly as we go from rich to lean. This occurs because of the increased oxygen con centration. As one would expect, NO increases rapidly withtemperature at a given fuel-air ratio. It can also be seen in Fig. 24 that pressure has a significant effect on NO concen tration at rich mixtures, but very little effect at lean mixtures. For rich mixtures an increase in pressure decreases the amount of NO formed under equilibrium conditions. Fig. 25 presents a summary of experimentally measured NO concentrations for both spark ignition and diesel engines. In F-A .RATIO/ STOICHIOMETRIC F-ARATIO Fig. 24 - Calculated NO concentration versus equivalence ratio o xr o, z EQUILIBRIUM COMPUTATIONS BY NEWHALL CONSTANT VOLUME COMBUSTION B,NO FIXED AT PEAK CYCLE TEMPERATURE ~. PREMIXED SI,CRaB, WOT PEAK PRESSURE 10* ATOC.IOOORPM -HETEROGENEOUS SI,CR=8, WOT EAK PRESSURE 10" ATDC, IOOORPM DIRECT INJECTION. 2 CYCLE. Cl DIRECT INJECTION. 4 CYCLE.CI, 1250 RPM ` -- CR-IS, INJ-20" PRECHAMBER Cl. TURBOCHARGED 1200 RPM, INJI5" 10 20 30 40 SO 60 70 A/F 80 90 Fig. 25 - NO concentration versus air-fuet ratio IOO Fig. 25 NO concentration is plotted as a function of air-fuel ratio. The first curve in Fig. 25 is a computed curve showing the amount of NO that would be present in the exhaust if an ideal, constant volume cycle were followed and if the NO con centration in the exhaust was that formed at top-center, that is, at the peak cycle temperature. It can be seen that for tills situation NO reaches a maximum at an air-fuel ratio of around 18 or 19. Two experimental curves for spark-ignition engines are shown. Both curves were taken with the spark timing adjusted as the air-fuel ratio was changed so that the peak pressure al ways occurred at 10 deg ate. The reason for doing this is that in the engine, peak temperature changes with a change in airfuel ratio for two reasons. In the first place, peak temperature will change with a change in air-fuel ratio because of more or less excess air to be heated up, just as in a burner. However, in the engine, in addition the flame speed varies with a change in air-fuel ratio and, as a consequence, peak pressure will occur DUP050311945 11 Fig. 26 -Effect of atmospheric nitric oxide concentration on hydro carbon oxidation EXTENT OP EXPANSION * REF It Fig. 27 - Variation of theoretical CO concentration throughout expan sion at different'cylinder volumes with a consequent change in peak temperature. One of the curves is for a premixed mix ture and the other curve is for a heterogeneous mixture such as is undoubtedly present in a practical multicylinder sparkignition engine. In comparing the two curves (premixed versus hetero geneous) it can be seen that the heterogeneous mixture has higher NO concentrations than does the premixed mixture. This can be explained by the nonlinear averaging that occurs. For example, if the average mixture was rich (say an air-fuel ratio of 12), the heterogeneous mixture would have local airfuel ratios both richer and leaner than 12. The richer mixtures would contribute but little NO, while the lean mixtures would contribute significantly greater amounts of NO. Thus, on the average, the amount of NO would be larger than that obtained with a premixed mixture. A similar explanation can be given for the diesel data shown in Fig. 25/ It is believed that even though the overall mixture is quite lean in diesel engines, combustion actually takes place at an air-fuel ratio richer than stoichometric. Thus, the mea sured NO concentrations would correspond to the air-fuel ratio at which combustion took place rather than the overall air-fuel ratio. This explanation is consistent with the data shown in Fig. 25. After consideration of all of the data presented, it can be concluded that the recombination reactions which destroy NO have relatively slow reaction rates. The four most im portant destruction reactions are: 1. NO + M = N + O + M. 2. NO + N = N2 + O. 3. NO+O = 02 + N. 4. NO + NO = N20+0 = N2 + 02. Of these, four reactions, number 1 is the slowest, while num bers 2 and 3 are the fastest. All of them are quite slow, thus explaining the fact that very little recombination takes place during expansion. Before leaving the subject of NO, it should be mentioned that NO is considered to take part in the destruction of hydro carbons and may be necessary for the elimination of hydro carbons from the atmosphere. This is shown in Fig. 26 (46) where the rate of propylene oxidation is plotted as a function of NO concentration for different initial concentrations of propylene-. It can be seen that the rate of destruction of pro pylene is related to the NO concentration. CARBON MONOXIDE - The next partially oxidized product I wish to talk about is carbon monoxide. In an attempt to deduce where and how carbon monoxide is formed, the re sults of theoretical computations are presented in Fig. 27 (43). Fig. 27 shows CO concentration as a function of the extent of expansion of the gases for an ideal cycle for three different equivalence ratios. Experimentally measured exhaust gas con centrations of CO are shown for two equivalence ratios. Again, the data show that the measured concentrations of CO correspond more nearly to the equilibrium concentrations of CO at peak cycle temperature than those existing at the end of expansion. Thus, as in the case of NO, we deduce that CO is formed in the bulk gases and in part, at least, as a result of frozen equilibrium, that is, a slow rate of destruction of some of the products. Fig. 28 is a summary plot of calculated and experimental data showing mole percent CO versus air-fuel ratio. The upper solid curve is tire CO concentration that would be computed at the peak cycle temperature of an idea! constant volume cycle. The lower solid curve is the computed concentration, assuming equilibrium during the expansion to 1400 R. The experimental data points were taken holding the peak pressure constant at 10 deg ate so that any tempera- DUP050311946 , 12 ture variations that took place were due to the air-fuel ratio effects directly, rather than to variations in flame speed. Fig. 28 clearly shows that, particularly at rich mixtures, mea sured concentrations of CO are closer to the peak cycle con centrations than to those existing if equilibrium occurred all during the expansion process. Interestingly enough, as one approaches stoichometric or leaner mixtures, it appears that measured concentrations of CO more nearly approach the concentrations that would be expected if equilibrium occurred during all of the expansion process. As a matter of interest, Newhall (47) has set up a model of the expansion process using 30 reaction rate equa tions. He predicts the same trend as is shown in Fig. 28, that is, CO is further from equilibrium at rich mixtures. Since we do apparently have nonequilibrium insofar as the concentration of CO is concerned during the expansion pro cess, let us look at the destruction reactions involved. The primary destruction reaction for CO is: CO + OH = C02 + H This reaction is in equilibrium, that is, its reaction rate is' relatively rapid. It appears that the CO concentration is high because the atomic hydrogen concentrations are high, that is, the rate of destruction of hydrogen appears, to be low. The two reactions that cause hydrogen destruction are: 1. H + H = H2 2. H + OH = H20 Apparently we have a peculiar situation where the CO con centrations are high because the atomic hydrogen concentra tions are high. HYDROCARBONS - Turning next to hydrocarbons in the exhuasi, theoretical computations show that no hydrocarbons Fig. 29 - Variation in HC concentration during exhaust stroke should exist if the system is in thermodynamic equilibrium. Thus it is believed that they are formed in the quench area, that is, in the cold zone immediately adjacent to the cold combustion chamber walls. This hypothesis is confirmed by data shown in Fig. 29 (48) showing hydrocarbon concentra tion as a function of the time in the cycle. The bulk gases will be exhausted primarily during the middle portion of the ex haust period. During this time, it is experimentally observed that the hydrocarbon concentration is low. On the other hand, the first and last part of the exhaust process which is primarily when the quench gases are being exhausted shows high concentration of hydrocarbons. Fig. 23 would also con firm this deduction. Fig. 30 (48) shows engine data also confirming the quench theory. To obtain the data shown in Fig. 30, a sampling valve was used at different flow rates. The thought here is that at very low rates only the quench gases will be drawn through the sampling valve, while at high flow rates one will obtain in creasing quantities of the bulk gases. It can be seen in Fig. 30 that at low sampling rates, high hydrocarbon concentrations are obtained, thus supporting the quench theory. In looking at hydrocarbon data from real engines, however, one must recognize that oxidation does take place in the ex haust process and exhaust system. Fig. 31 (45) shows mea sured hydrocarbon concentration as a function of spark ad vance for,two different situations. In the first situation, called "exhaust cooling off," a copper cooling coil was placed in the exhaust system adjacent to the exhaust valve. However, no water was passed through the copper cooling coil and the data as shown were obtained. It will be remembered that as the spark is retarded and combustion occurs later in the cycle. DUP050311947 13 * SAMPLING RATE t J. IN. PER MIN - STPI Fig. 30 - Hydrocarbon component concentrations near combustion chamber wall there will be an increased quench area and one would antici Fig. 31 - Exhaust temperature and HC concentration versus spark timing with and without cooling pate, therefore, an increased concentration of hydrocarbons in the exhaust. However, the data indicate a small increase, followed by a decrease. The second set of data were taken with the water flowing through the copper cooling coil located immediately adjacent to the exhaust valve. It can be seen in Fig. 31 that the data obtained are more consistent with quench theory; that is, as the spark is retarded, the concentration of hydrocarbons in the exhaust gas increases. From this it can be concluded that there is significant destruction of hydrocarbons in the exhaust system, and that what finally comes out of the exhaust system is the hydrocarbons formed less those destroyed in the exhaust system. One obvious combination would be to run rich, which would form a considerable amount of unburned hydrocarbons and CO and then destroy the hydrocarbons and CO in the exhaust system. This thought could be expanded to include two-stage combustion with additional air added between stages. Either technique would probably involve sacrifices in fuel consump tion. ............ -- Another possibility for elimination of NO in the exhaust gases would be to speed up the destruction of NO during the expansion process. This would require a catalyst that would speed up the previously quoted NO reduction reactions so that equilibrium would exist all during the expansion process. SOLUTIONS TO PROBLEM OF EXHAUST EMISSIONS While finding such a catalyst is not an impossibility, the author Is unaware of any such catalyst. Yet another possibility is to reduce the NO in the exhaust RECIPROCATING ENGINES - system. In general, this is difficult because of the slow reduc Oxides ofNitrogen - Now that I have presented the source of tion of NO. the different emittants from the exhaust of the automobile, Another possibility is the reduction of peak flame tempera what solutions can be found to the problem? Looking first at tures by retarding the spark. While this procedure is obviously NO, one obvious solution would be to prevent the formation bad from an economy standpoint, it does raise the exhaust of NO. One suggested way to do this is to add an inert gas, for gas temperature which helps to destroy hydrocarbons in the example, recycle some of the exhaust. This would decrease exhaust. the peak cycle temperature without increasing the oxygen Carbon Monoxide - Let us look next at the elimination of concentration. carbon monoxide. Again a possible procedure would be to Another obvious way of preventing the formation of NO is add an inert gas. Preliminary calculations indicate, however, to burn a mixture that is either quite rich or quite lean. Burn that equilibrium persists during the early portion of the ex ing a mixture that is quite rich is undesirable in the sense that pansion stroke. Consequently, rather large amounts of inert it increases the amount of CO and unburned hydrocarbons. gas would have to be added in order to obtain the needed Looking at the lean side, it is difficult to burn a mixture lean temperature effect. Thus, the addition of an inert gas does enough to decrease the NO to the desired quantity. not seem to be a profitable approach, < DUP050311 948 6*1960 - No Controls 14 1971 - Tentative Calif. SM. yl97X - Exhaust Recirculation? 41972 - Tentative Calif. Stt. *1974 - Tentative Call!. SM. Stirling. Gas Turbine, Steam, etc. JSri?7X * Exhaust Reactors _L RELATIVE POWERPLANT COST Fig. 32 - Relative cost of reducing NOx (California vehicles) Fig. 33 - Relative cost of reducing CO As can be seen in Fig. 28, CO1 is primarily a function of air- fuel ratio. If the engine can operate at a mixture leaner than stoichometric, it would appear that CO could be reduced to manageable levels. Also, as in the case of NO, an equilibrium catalyst might be used to achieve equilibrium all during the expansion process. However, at richer than stoichiometric mixtures CO would still be high. It will be remembered from the previous discussion that the equilibrium catalyst would be needed for the reactions involving the destruction of the hy drogen atom. Again the author knows of no such catalyst. It is also possible to destroy carbon monoxide in the exhaust system by further oxidation. Hydrocarbons - Turning next to hydrocarbons in the ex Fig. 34 - Relative cost of reducing HC haust, we could in theory minimize their formation by using a stratified charge engine. In a stratified charge engine the area While there is considerable disagreement as to exact num of the quench zone next to the cold combustion chamber bers, there is complete agreement that air pollution controls wall would be automatically reduced at reduced load. How are going to cost money. Figs. 32-34 show the progress al ever, a new quench zone at the interface between the com ready made and estimated costs of additional progress by both bustible mixture and the air would be introduced and there reciprocating and other types of engines (33). could be considerable mixing with a quenching effect at this EFFECT OF CURRENT AND PROPOSED CONTROLS ON interface. Thus most people in the United States feel that a - AIR POLLUTION-As our 4echnology has progressed our stratified charge engine will not give the required reduction in standards for pollution for automobiles have become mote exhaust gas emissions. stringent. The extent of progress is shown in Fig. 35 where One approach is to complete the oxidation of hydrocarbons the estimated emissions from untreated cars is shown as well in the exhaust. This is the basic approach that is currently as the progressive decrease in emittants with time. The de being followed in the United States; that is, to increase the crease in hydrocarbon emission is 85%--in carbon monoxide rate of exhaust destruction in the exhaust either by increasing it is 68%. These are significant reductions, but I suspect the the amount of oxygen present in the exhaust, or by maintain general public does not appreciate the tremendous technical ing the exhaust at high temperatures with the use of insulated effort required to achieve these reductions or the magnitude or heat conserving manifolds or by both. In general, retarded of the further effort required to bring about further reduc spark timing gives a higher exhaust temperature and is being tions. used as a means of increasing exhaust temperatures. Increased The 1970 and 1971 standards were switched from exhaust engine operating temperatures are an aid to minimizing quench concentration standards to a mass-per-vehicle-mile standard. gases and maximizing exhaust temperatures. Combustion It appears that future standards will continue to use mass per chamber design for minimum quench area is now universal. mile. While proposed future standards may become more Competing Power Sources - Since hydrocarbons are thought stringent, they will also probably place more emphasis on to be formed in the quench zone, engines which do not startup emissions with other driving cycle modifications aimed "scrape off' the quench zone will undoubtedly have hydro at more nearly representing driver patterns. carbon emissions which are relatively quite low. The situation Fig. 36 presents Heinen's (4) estimate of the present and is not as clearly favorable with regard to carbon monoxide and future situation so far as automotive exhaust emissions in the oxides of nitrogen since they are formed in the bulk gases. It United States are concerned. The effect of crankcase emission is true that in steady flow combustion there is much more latitude in choice of overall fuel air ratios and consequently a control in the early 1960's is evident as well as the effect of exhaust emission controls in the late 1960's. Fig. 36 empha wider choice of solutions. sizes the time lag between the introduction of a control device DUP050311949 IS Fig. 35 - Reduction in automotive emissions with time Automotive Atmospheric Improve ments in United States Fig. 37 - Example of reductions in HC obtained in laboratory experi ments 2500 a2000 0.1500 Z *000 500 0 Air -Injection Du Pont Reactor 10 20 30 40 % AROMATICS IN FUEL Fig. 38 - Reduction in NO^ using experimental thermactor Fig. 36 - Atmospheric improvements in United States as result of auto motive emissions control on new cars and the impact of this control on total emissions because of the many uncontrolled cars on the road. LONG TERM REDUCTION IN EMISSIONS - Although the general public is probably unaware of its magnitude, much ef fort has been expended in attempting to reduce pollution from the IC engine below any currently proposed standards. The main thrust of the efforts insofar as hydrocarbons and carbon monoxide are concerned has been either to operate very lean or to keep the exhaust gases hot and complete oxidation in the exhaust system. As an example of a laboratory experi ment and dearly not as a technique ready for production. Fig. 37 shows the combined effect of lean mixture, heated ex haust, and the use of back pressure (49). Another laboratory approach is the use of a heat-conserving exhaust manifold (50). Figs. 38-40 show results from this approach which is equally effective when using olefins; There are obvious ma terial problems using this approach. The above two approaches are' Only samples of the intensive effort aimed at reducing automotive pollutants. Ultimately, as exemplified by Figs. 32-34, it will be a simple economic question--what is the cost-benefit ratio, for further reductions and will adaptations of the IC engine be the most economic standard * 0*2 0 1 Air Inleclion Lean-Modified Spark DuPont Reactor , % AROMATICS IN FUEL Fig. 39 - Reduction in CO using experimental thermactor 1000 900 ^ 800 roo g 600 o 03 500 OC g 400 O mQ 300 X 200 100 0 Standard Lean-Modified Spark Air Injection DuPont Reactor % AROMATICS IN FUEL Fig. 40 - Reduction in HC using experimental thermactor DUP050311950 16 way to achieve these reductions in comparison with steam en man-caused additions of carbon monoxide, oxides of nitrogen, gines, gas turbines, Stirling engines, batteries, fuel cells, etc. hydrocarbons, and airborne lead. 6. While there is Some disagreement as to the exact time it THE ENGINEER AND AIR POLLUTION will occur, there is universal agreement that at some time in the future the growth of the automobile population will ex You have now had a brief exposure to the major elements of the problem--projected population growth, atmospheric contaminants, the source of emittants from automobiles, the effect of current and proposed controls and potentialities, and at least order of magnitude cost estimates. At this point, it would be simple for us as engineers to shrug our shoulders, argue that science is morally neutral and return to our com puters. However, as stated by Solandt (51), "Even if science is morally neutral, it is then technology or the application of science that raises moral, social and economic issues." As stated in a different way by one of the fop United States auto motive engineers, ``our industry recognizes that the wishes of the individual customer are increasingly in conflict with the needs of society as a whole and that the design of our product is increasingly affected by this conflict." This is clearly the case with pollution--an individual customer will not volun tarily pay extra for a car with emissions control even though the needs of society as a whole may be for emission controls. This conflict between individual and society needs has been present in the behavioral field since Cain and Abel. The con flict between the two is more recent in the technological field, but in my opinion will increase and not decrease in the future. The conflict involves social benefit versus technological costs (52). Thus, we as engineers will be "in the middle," so per haps we should "get used to it." I think if in the future we are to truly optimize our engineering design we will inevitably be struggling with such decisions, so let's make an attempt at this one. To assemble our "facts" as I see them: ceed the effect of present and proposed controls and that if no further action is taken mass rates of addition of pollutants from automobiles will rise again. 7. There is a time lag of 3-5 years from the establishment of probable technological feasibility until a control technique in volving significant modifications can be placed in universal mass production. 8. There is an additional time lag of 4-7 years before a suf ficient fraction of the car population are equipped with the control device so that noticeable reductions occur. From these "facts" it seems clear to me that the heavy em phasis placed on solving the emission problem by both manu facturers and government is justified. I would observe that I think the manufacturers would be well advised to do much more in the way of informing the public about the problem, its complexities and what they, the manufacturers, are doing in an attempt to solve the problem. However, we have still not faced squarely up to the prob lem--how much should we spend for pollutant control and what should'be our standards for air quality? I think the ap proach suggested by Cassell (38) is sound. He suggests that air pollution control be based on the concept of "the control of pollutant emissions to the greatest degree feasible employ ing maximum technological capabilities." The statements "feasible" and "maximum technological capability" have built into them continued progress whereas fixed standards are notably difficult to change. Furthermore, if a fixed standard is not feasible for technological or economic reasons, it is not enforced. However, the urgencies arid necessities oFthe problem must 1. Man-caused additions are, or will be, large in the near future with respect to natural contributions for particulate matter, carbon monoxide, oxides of nitrogen, and nonmethane also be taken into account in evaluating the "greatest degree feasible employing maximum technological capabilities." This is illustrated in Fig. 41*. (CO2 curve from Ref. 55.) You hydrocarbons. They are adding appreciably, especially on a local basis, to our environmental levels of sulfur dioxide and lead. 2. There are fragmentary data indicating a potential for ef fects on weather by atmospheric pollutants, expecially CO2 and particulates. 3. There are no data indicating that pollutants at present levels have a deleterious effect on healthy persons. This state ment could be questioned in the specific case of NOx in the may quarrel with the location of the caution and catastropic regions in Fig. 41, but they clearly exist. In addition to the ideas illustrated by Fig. 41, information on the length of time it will take the atmosphere to clear itself is also needed. Perhaps we should talk about the half life of pollutants! In any event, I think the concept is clear--"the greatest degree feasible employing maximum technological capabilities" must be interpreted differently when you are in the caution or catastropic region as compared to when you are well be Los Angeles basin where levels are sometimes briefly above those recommended by toxicologists. In addition, people vol untarily expose themselves to situations having greater po low the caution region. The same concept can and should be applied to local situa tions. For example. Figs- 42 and 43 show data for the Los tential effect. Nevertheless, at present levels there is evidence Angeles basin (56). It is unfortunate that data on "alerts" are of some deleterious effects on those having marginal health, not available from 1940 on. It is interesting to speculate especially under adverse conditions. 4. Large population growths and nonlinear relationships whether the very low number of "alerts" for CO in 1968 is truly a result of the downtrend in the CO emitted curve or is a between population density and pollutants and their effects must be recognized. 5. The automobile provides a significant portion of the *Ttus method of plotting was suggested by Dr. N. J. Beck of White Motor Corp. DUP050311951 17 CATASTROPHIC REGION CAUTION REGION 6 *' z a s s u mp t io n s : w o r l d c o , p r o d u c t io n is S TIMES US COj PRODUCTION s" ONE HALF MAN-CAUSED C02 REMAINS IN ATMOSPHERE. Ol__________ I__________ I__________ 1__________ IBSO IS90 1930 1970 ZOIO YEAR Kg. 41 - Plot of COj on worldwide basis illustrating different control urgencies Fig: 43 - Plot of NO in Los Angeles Basin illustrating different control urgencies cases): 1. Government employees. 2. Advisory groups set up by government officials. 3. Industry groups or associations. 4. Appropriate professional societies. Fig. 42 - Plots of CO in Los Angeles Basin iUustrating,different control, urgencies result of atmospheric conditions. Again. I think the principle is clear--assuming the alerts represent highly unsatisfactory or dangerous conditions, your interpretation of the "greatest de gree feasible employing maximum technological capabilities" becomes more stringent as the number of alerts increase. With this understanding I think it is up to us as professional engi neers to see that our engines are controlled to produce "the minimum emissions feasible employing maximum technologi cal capabilities." Several procedures can be followed in "setting the num bers" that represent "the greatest degree feasible employing maximum technological capabilities." I do not have in mind here the enabling legislation that is probably inevitable and necessary--that is unquestionably a perogative of government. What I have in mind is the process by which reasonable num bers are arrived at which, while they may ultima ;ly be incor porated into legislation, are set initially and primarily to rep resent "the greatest degree feasible using maximum technological capabilities." The groups that might be involved in establishing such num bers are (assuming appropriate technical qualifications in ail In considering the problem, it seems axiomatic to me that none of us are unbiased, try as hard as we may. Thus I con clude that an "unbiased" decision is best reached by a widely representative group whose individual bias in total add up to approximately zero and who are individually flexible and honest enough to recognize the possibility of bias and the necessity for compromise. Thus, 1 would feel that either two or four would have the most potential for reaching unbiased decisions. Personally, I would favor the use of appropriate technical societies--possibly because of my own bias, but more ' fundamentally because I think this offers the best opportunity for obtaining qualified individuals with the minimum of politi cal and economic pressures. I predict that the procedure fol lowed in the setting of performance level standards for tech nological matters will markedly influence the type of society we enjoy in the future. REFERENCES 1. P. S. Myers, "The Changing Criteria For Optimizing En gineering Designs." SAE. Journal, Vol. 77, December 1969. 2. P. L. Magili, F. R. Holdan, and C. A. Ackley, "Air Pol lution Handbook." New York: McGraw-Hill Book Co., Inc., 1956. 3. R. P. Hammond, "Low Cost Energy-A New Dimen sion." Science Journal, Vol. 5, January 1969, pp. 34-42. 4. Charles M. Heinen, "We've Done the Job--What's Next.' Paper 690539 presented at SAE Metropolitan Section, New York, April 1969. 5. A1 J. Sweet, B. J. Steigerwald, and J. H. Ludwig, "The Need for a Pollution-Free Vehicle." Journal of Air Pollution Control \ssociation, Vol. 18, February 1968, pp. 111-113. 6. M. M. Braverman, C. Theophil, F^Masciello, and DUP050311952 18 C. Smith, "The Contribution to Air Pollution by Pigeons." on Lung Cancer Research." Workshop I, Appendix I, pp. Journal of Air Pollution Control Association, Vol. 12, Decem 56-71, American Cancer Society, New York, 1957. ber, 1962, pp. 570-571. 24. R. A. Bryson, "Climatic Effects of Atmospheric Pollu 7. Lynn and McMullen, "Air Pollution in Six Major U. S. tion." Paper presented at National Meeting of American Asso Cities as Measured by the Continuous Air Monitoring Pro ciation for the Advancement of Science, December 1968, gram." Journal of Air Pollution Control Association, Vol. 16, Dallas, Texas. April 1966, pp. 186-190. 25. N. Eichhorn, "Implications of Rising Carbon Dioxide 8. Air Conservation Commission, "Air Conservation." Content of the Atmosphere." Report of Meeting on March Publication No. 80, American Association for the Advance- 12,1963, sponsored by Conservation Foundation; 30 E. 40th ment of Science, Library of Congress Catalog Card 65-28248, St., New York. 1965. 26. G.N. Plass, "Proceedings of the Conference of Recent 9. M. Sittig, "Air Pollution Control Processes and Equip Research in Climatology," pp. 81-92. Edited by H. Craig, ment." Chemical Process Review, No. 24, Library of Congress University of California, La Jolla, 1957. Card Catalog 68-28248, 1965. ' 27. P. Groen, "The Waters of the Sea." New York: Van 10. Louis S. Jaffe, "Ambient Carbon Monoxide and Its Nostrand, 1957. Fate in the Atmosphere." Journal of Air Pollution Control 28. J. J. Coiucci and C. R. Begeman, "Carbon Monoxide in Association, Vol. 18, August 1968, pp. 534-540. 11. E. Erikksson, "The Yearly Circulation of Sulphur in Nature." Journal of Geophysical Research, Vol. 68, July 1, Detroit, New York, and Los Angeles Air." Proceedings Air Pollution Control Association, 1967, paper 67-182. 29. R. C. Robbins, K. M. Borg, and E. Robinson, "Carbon 1963, pp. 4001-4007. 12. R. A. Kehoe, "Toxiological Appraisal of Lead in Rela Monoxide in the Atmosphere." Journal of Air Pollution Con trol Association, Yol. 18, February 1968, pp. 106-110. tion to the Tolerable Concentration in the Ambient Air." 30. L. J. Fuller and W. J. Hamming, personal communica Kettering Laboratory, University of Cincinnati, April 1969. tion, 1969. 13. G. J. Stopps, "Symposia on Air Quality-Lead." Pro 31. H. E. Stokinger, "Effects of Air Pollution on Animals." ceedings of the Symposium on Air Quality Criteria, New York, Air Pollution, edited by A. C. Stern, Vol. 1, pp. 282-334, New June 1968, Journal of Occupational Medicine, Vol. 10, Sep York: Academic Press, 1962. tember 1968, pp. 118-128. . 32. H. Heiman, "Effects on Human Health--Air Pollution." 14. California Department of Public Health, "Lead in the Environment and its Effects of Humans." Berkeley, Cali WHO Monograph No. 46, pp. 159-220 (World Health Organi zation, Geneva, 1961). fornia, 1967. 33. L. R. Hafstad, remarks during 1969 General Motors 15. U. S. Department of Health, Education and Welfare, "Progress of Power" Show, General Motors Technical Cen ter, "Survey of Lead in the Atmosphere of Three Urban Com May 1969. munities." January 1965. 34. R. M. Brice and J. F. Roesler, "The Exposure to Carbon 16. H. H. Hovey, A. Risman, and J. F. Cunnan, "The De- Monoxide of Occupants of Vehicles Moving in Heavy Traffic." -velopment of Ajr Contaminant Ejnission Tables for Non- _ . . Journal of Air Pollution Control Association, Vol. 16, Novem Process Emissions." Journal of Air Pollution Control Asso ber 1966, pp. 597-600. ciation, Vol. 16, July 1969, pp. 362-366. 35. California State Department of Public Health, "Tech 17. A. P. Altshuler, G. C. Ortman, B. E. Saltzman, and R. E, nical Report of California Standards for Ambient Air Quality Neligan, "Continuous Motoring of Methane and Other Hydro and Motor Vehicle Exhaust." Berkeley, California, 1960. carbons in Urban Atmospheres." Journal of Air Pollution 36. P. H. Abelson, "A Damaging Source of Air Pollution." Control Association, Vol. 16, February 1966, pp. 87-91. Editorial in Science, Vol. 158, December 1967. 18. E. Robinson and R. C. Robbins, "Sources, Abundance 37. B. D. Tebbens, Paper in Air Pollution, edited by A. C. and Fate of Gaseous Atmospheric Pollutants." Stanford Re Stem, Vol. 1, pp. 23-40: New York: Academic Press, 1962. search Institute, February 1968, report prepared for Ameri 38. E. J. Cassell, "Are We Ready for Ambient Air Quality can Petroleum Institute. Standards?" Journal of Air Pollution Control Association, 19. A. P. Van Den Heuvel, and B. J. Mason, "The Forma Vol. 18, December 1968, pp. 799-802. tion of Ammonium Sulfate in Water Droplets Exposed to 39. R. S. Brief, A. R. Jones, and J. D. Yoder, "Lead, Carbon Gaseous Sulphur Dioxide and Ammonia." .Quarterly Journal, Monoxide and Traffic-A Correlation Study." Journal of Air Royal Meteorological Society, Vol. 89 (1963), pp. 271-275. Pollution Control Association, Vol. 30, October 1960, pp. 20. C. E. Junge, "Air Chemistry and Radioactivity." New 384-388. York: Academic Press, 1963, pp. 96-99. 40. A. H. Rose, Jr. and R. C. Stahman, `The Role of En 21. J. R. Goldsmith, "Los Angeles Smog." Science Journal. gine Blowby in Air Pollution." Journal of Air Pollution Con Vol. 5, March 1969, pp. 44-49. trol Association, Vol. II, March 1961, pp. 114-117. 22. R. A. Bryson. "Dust Dome." LTniversity-lndustry Re 41. H. C. McKee, and K. D. Mills, "A Comparison of Vari search--Research Newsletter, University of Wisconsin, Madi son, May 1969. 23. A. J. Haagen-Smit, "Proceedings of the First Workshop ous Sources of Automotive Emissions." Journal of Air Pollu tion Control Association, Vol. 11, November 1961, pp. 516522. DUP050311953 19 42. W. C. Cote, "The Performance of Positive Crankcase Ventillation Systems on Automobiles in Cincinnati, OMo, Re port No. 2." Journal of Air Pollution Control Association, Vol. 17, March 1967, pp. 139-141. 43. E. S. Starkman and H. K. Newhall, "Characteristics of the Expansion of Reactive Gas Mixtures as Occurring in In ternal Combustion Engine Cycles." SAJE Transactions, Vol. 74(1966), paper 650509. 44. T. A. Huls and H. A. Nikol, "Influence of Engine Vari ables on Exhaust Oxides of Nitrogen Concentrations from a Multi-Cylinder Engine." SAE Transactions, Vol. 76, paper 670482. 45. T. A. Huls, P. S. Myers, and O. A. Uyehara, "Spark Ig nition Engine Operation and Design for Minimum Exhaust Emissions." SAE Transactions, Vol. 75 (1967), Section, 2 pp. 699-719, paper 660405. 46. J. D. Caplan, "Smog Chemistry Points the Way to Ra-. tional Vehicle Emission Control." SAE Transactions, Vol. 74 (1966), paper 650641. 47. H. K. Newhall, "The Kinetics of Engine-Generated Nitrogen Oxides and Carbon Monoxide." 12th International Combustion Symposium, August 1969, pp. 603-613. 48. W. Daniel, "Engine Variable Effects on Exhaust Hydro carbon Compositon--A Single-Engine Study with Propane as the Fuel." SAE Transactions, Vol. 76, Section 1, pp. 774-795, paper 670124. 49. L. Eltinge, F. J. Marsee, and J. A. Warren, "Potentiali ties of Further Emissions Reduction by Engine Modification." SAE Transactions, Vol. 77 (1968), paper 680123. 50. E. N. Cantwell, I. T. Rosenlund, W. J. Barth-, F. L. Kinnear, and S. W. Ross, "A Progress Report on the Development of Exhaust Manifold Reactors." Paper 690139 presented at SAE International Automotive Engineering Congress, Detroit, January 1969. 51. O. M. Solandt, "The Control of Technology." Editorial in Science, Vol. 165, Aug. 1,1965. 52. C. Starr, "Social Benefit Versus Technological Risk." Science, Vol. 165, September 1969, pp. 1232-1238. 53. National Conference on Air Pollution, Department of Health, Education and Welfare, Washington, D. C., No vember 1958. 54. T. J. Chow and C. C. Patterson, "The Occurrence and Significance of Lead Isotopes in Pelagic Sediments." Geochem. Cosmochem. Acta, Vol. 26 (1962), pp. 263-368. 55. F. A. Rohrman, B. J. Steigerwald, and J. H. Ludwig, "Industrial Emissions of Carbon Dioxide in the United States, A Projection." Science, Vol. 156, May 19,1967, pp. 931-932. 56. W. J. Hamming, R. D. MacPhee, and J. R. Taylor, "Contaminent Concentrations in the Atmosphere of Los Angeles County." Journal of Air Pollution Association, Vol. 10, Feb ruary 1960, p. 7. 57. Amos Turk and Claude J. D'Angio, "Composition of Natural Air." Journal of Air Pollution Control Associ ation, Vol. 12, January 1962, pp. 29-33. This paper is subject to revision. Statements and opinions advanced in papers or discussion are the author's and are his responsibility, not the Society's; however, the paper has Society of Automotive Engineers, Inc. TWO PEN Mtrvvamia viA2A. n e w vow*, n . r. tooai been edited by SAE for uniform styling and format. Discussion will be printed with the paper if it is published in SAE Transactions. Po t permission to publish this paper in full or in part, contact the SAE Publications Division and the authors. 20 page booklet. Printed in U.S.A. DUP050311954