Tobacco Documents Online

ENVIRONMENTAL RESEARCH 13, 1-35(1977) H K021"74Q84 Exposure to Pollutants in Enclosed "Living Spaces" THEODOR D. STERLING AND DIANA M. KOBAYASHI Faculty of Interdisciplinary Studies, Simon Fraser University, Burnaby, B.C., Canada Received October 10, 1975 , INTRODUCTION Pollution of domestic premises, public buildings, and transport vehicles, is linked by problems peculiar to enclosures. Enclosures afford protection from toxic substances. On the other hand, they may entrap pollutants inside that have seeped in from the outside or have been generated inside, as enclosed spaces almost always contain sources of pollution of their own. Studies on enclosed environments are grouped- for our purposes into four categories. Each category will be discussed separately: pollutants in artificially- sealed environments, pollutants in domestic premises, pollutants in public build- ings, and pollutants in transportation related enclosures. Pollution levels reported by different studies are summarized in a series of appended tables. The information available about pollution in enclosed spaces is sparse but suffi- cient to indicate the magnitude of possible exposure to inhabitants. Evaluation of existing studies leads inevitably to one conclusion: A building does not protect its inhabitants from pollution. To the contrary. The body burden of toxic vapors and dusts in the -" inside" may very well exceed the burden of pollution in the " out- side." POLLUTION IN ARTIFICIALLY- SEALED (SUBMARINE) ENVIRONMENTS Studies of sealed environments, especially of submarines, are an important source of information about pollutants in enclosed spaces. Contaminants gener- ated outside do not penetrate the isolating structure. The types and amounts of pollutants generated within the enclosed environment can be determined with good accuracy and their source can be established. At the same time, studies of these artificially sealed environments have to contend with unique variables: oxygen must be provided and carbon cdioxide must be removed or reconverted into oxygen, a pollutant-removal system usually is installed, ample machinery is usually present in addition to the equipment required to maintain a breathable atmosphere, and the structures are usually pressurized. CARBON MONOXIDE Because of the rapid buildup of carbon monoxide, burners (actually non- specific incinerators) must be utilized at all times. Even so CO averages 50 ppm during periods of submergence. There are numerous sources of CO production, including heating, cooking, oxidation of oils and lubricants, smoking, and aging of paints (Schulte, 1961, 1964). These results are confirmed by Ebersole (1960) in his report on an early record-breaking 60-day dive of the USS Seawolf in 1958, and by Hine (1964). Although average CO concentrations of around 50 ppm are reported in studies conducted by Hine, Ebersole, and Schulte, a report by Alvis (1952) of I All rights of reproduction in any form reserved. ISSN 0013-9351 Copyright © 1477 by Academic Press. Inc. 0 ER-Srr'? 003663

2 STERLING AND ROBA1'ASHI N !{E2i70085 mass CO poisoning of a crew indicates that dangerously high peaks of CO are possible during submarine cruises. At one time, cigarette smoke was thought to be a major source of CO (Anderson, 1964). Schulte (1960, 1964), however, points out that when additional CO detection equipment was installed in newer submarines. inaccessible reactor compartments contained the highest concentrations of carbon monoxide. Further measures were made of areas housing equipment. It was de- termined that most CO production was caused by the oxidation of oils, lubricants. and paints, especially on steam pipes. Living areas, the only areas where smoking was allowed, contained much less CO than other areas of the ship by a factor of 3 in comparison to the reactor room (Ebersole, 1960). HYDROCARBONS Extremely high concentrations (500 ppm) of aromatic hydrocarbons (also refer- red to as "undifferentiated" hydrocarbons) have been reported by Schulte (1964) coincidental to cooking episodes. Hine (1964) also lists cooking as a major source of hydrocarbon production. Besides cooking, aromatic hydrocarbon sources are: paints, varnishes, lacquers, paint thinners, solvents, cleaning fluids, artificial leather, linoleum, asphalt tiles, rubber and plastic cement, and bonding com- pounds (Ebersole, 1960). Many substances give off hydrocarbons for many months at a slow logarithmic rate (Schulte, 1964); Anderson and Saunders, 1964; Ebersole, 1960). Although water-base paints are used whenever possible, Siegel (1961) has found that a 72 hour limit of painting must still be imposed before submergence. MISCELLANEOUS CONTAMINANTS Cooking is responsible for most of the aerosol production (Ebersole, 1960; Schulte, 1964). Oxides of nitrogen are formed on submarines by electric arcing of armatures and by short circuits (Schulte, 1964). Sources of ozone are the same as oxides of nitrogen (Schulte, 1964; Ebersole, 1960). SO2 and hydrogen sulfide are produced by the bacterial action of sanitary tanks (Schulte, 1964). Freon is easily broken down under heat (often in CO burners) to chlorine, fluorine, hydro- gen chloride, hydrogen fluoride, and phosgene. There is not a good means of removing this gas once it has leaked from cooling and refrigeration systems (Eber- sole, 1960; Hine, 1964; SSchulte, 1964). Halogens are produced when freon is oxidized in CO burners (Schulte, 1964). Ammonia is produced by the action of CO2, scrubbers and as a biological end product in sanitary tanks (Schulte, 1961; Hine 1964). Mercury vapor comes from meters and gauges (Schulte, 1961; Siegel, 1961). Methyl alcohol comes from mimeograph equipment (Ebersole, 1961). Tri-aryl phosphate comes from hydraulic fluids (Siegel, 1961). Formaldehyde is a product of oxidation of methyl alcohol (Schulte, 1964). Radiation was and occasionally still may be a source of hazard on submarines (Schulte, 1964; Ebersole, 1957). Among other sources of pollutants mentioned by Hine (1964) are shoe polish, insect sprays, lighter fluids, shaving soap, and hair tonics. Kitzes (1958) expands the list of pollutants present in cabin-type environments by adding a few more sources found in aircraft, including anti-icing fluids, fire extinguishants, cargo, fuels, oils, and selenium in rectifiers. The air also contains by-products of metabolic activity: CO2, water, feces, flatus, urine, breath, sweat, glandular secretions, organic dust t r MRSTER--55pp 003684

H i{I217(t48E EXPOSURE TO POLLUTANTS 3 particles from hair, skin, mucus, and dead cells. From breath come acetic acid, acetone, volatile oils, methane, and hydrogen sulfide. From urine come volatiles, ammonia, etherial sulfates, and from sweat and glandular secretions come urea, and lactic and other organic acids (Hine, 1964; Gorban, 1964). The levels cited in these studies are not necessarily accurate reflections of all submarine exposures. Peak levels may be much higher than those given. Further, Schulte (1961) mentions occasional failures of pollution control devices which cause great increases in pollutants. POLLUTANTS IN DOMESTIC PREMISES The amount and type of pollutant present in homes depends on the number of persons occupying the premises, kinds and duration of operation of appliances, methods of cooking and heating, types of chemicals and paints present, flow of air, filtration devices used, penetrability of the structure, amount of outdoor air pollution and air movement, temperatures and humidity inside and outside the home, and activities within the structure. In addition, most sources of pollution isolated in completely enclosed environments also exist in the home. INDOOR-OUTDOOR COMPARISONS Biersteker et al. (1965) furnish gross estimates of the relation of indoor smoke and SO2 levels to outside concentrations. Eight hundred paired samples were taken in 60 different bungalows, high-rise apartments, and flats, over a 7-day period. Suspended particulate levels were about 80% of those outdoors, while SO2 levels were about 20% of outdoor levels. Much more exact measures were ob- tained by Yocom et al (1971a) for suspended particulates, soiling particulates, CO and SO2 in both gas- and coal-heated medium-size, single-family dwellings. Total indoor levels of CO and SO2 almost constantly exceeded outdoor concentration levels. In coal-heated homes, SO2 and CO levels closely followed outdoor levels until the coal furnaces were activated. Then, extremely high levels of pollutants were found at each period of furnace activation, greatly exceeding outdoor levels of the same substance. Indoor particulate matter was less in overall amount but with a greater organic enrichment factor than outdoor particulate matter. Lead was also detected (0.47 to 1.75 µg/m3). Yocom et al. noted significant seasonal variations. Particulates were less in amount during the winter when entrance of outside air was impeded by closed windows. On the other hand, indoor levels of CO increased during the winter. A number of other studies concur with the findings of Yocom et al. and provide additional information. Overall dust levels in single-family dwellings were studied by Lefcoe and Inculet (1971, 1975) under varying conditions, with and without air filtration. Particulate matter up to 5µm was continuously monitored. Increases in dust levels occurred when windows were opened, when filters were not used, and when there was a good deal of activity in the home. Although outdoor increases were generally higher, outdoor increases were reflected indoors. SO2, NO2, and 03 levels were also tested. The maximum amounts were <0.06 pphm for SO2, <0.1 pphm for NO2 and < 1.96 pphm for O3. There were no significant differences between indoor and outdoor measures noted. Some idea of the influence of outdoor pollution levels on indoor environments MRSTER-5PP 003685

HK1217U087 - 4 STERLING AND KOBAYASHI can be surmised from urban-rural, indoor-outdoor pollution differences. Godin et al. (1972) studied CO levels. At a semirural farm, outdoor CO values were 1.0 ± 0.8 ppm. Values were double at a suburban home (outdoors: 2.0 ± 1.4 ppm; indoors: 1.9 ± 1.3 ppm). Schaefer et al. (1972) correlated particulate fallout in homes with their geographic locations. Homes in cities showed the highest amounts of sedimentation, and those in rural areas showed the lowest amounts. Although there were differences from room to room, kitchens, in general, were shown to have the highest amounts of particulate matter. Jacobs et al. (1962) similarly found that indoor particulate concentrations were similar to outdoor concentrations, although more submicron particles were found indoors than out- doors. SOURCES OF INDOOR POLLUTION Pollutants may be indoor-generated or they may originate from the outside. Furthermore, once present they may build up over time. As part of Yocom's study CO levels were monitored in an unoccupied house. CO levels increased more slowly inside than out, but, once built up, indoor levels remained higher for a longer period than did outdoor levels. Thus, domestic premises have a tendency to entrap gaseous pollutants. Garages attached to homes may also entrap pollutants, allowing them to seep into the home. In one of the homes tested by Yocom et al. (1971 a) the attached garage proved to be a greater source of CO than even the gas stove. Cracks in structures, in addition to doors and windows, permit this entrance of pollutants and the possible subsequent entrapment of pollutants. Although applied to an unusual case, the possible prolonged penetration of pollutants was strikingly demonstrated by Megaw (1962). In October 1957, a cloud of nuclear fission prod- ucts was accidentally released near Windscale, England, permitting a test of the amount of 23'I found within contaminated houses. Although 1311 levels indoors were found to be much lower than 1311 levels outdoors, deposits on roofs and in crevices suggested that seepage over time was likely to occur. Megaw concluded that over time, amounts of 1311 trapped on roofs and in crevices could constitute a health hazard. Further information regarding the sources of indoor-generated CO comes from a number of surveys. Goldsmith (1970) estimates the number of persons suffering from household exposure to CO in the United States to be 100,000 per year. While the exact number of persons exposed is not known, Goldsmith's conclusions, nonetheless, highlight the fact that exposure to elevated amounts of CO may be affecting a large portion of the U.S. population. The extent of indoor CO pollution may be assessed also from Kahn et al. (1974) who found that the COHb content of blood donors increased during winter months, despite reduction in the ambient CO level. Kahn points out that there is reduced traffic in the winter months and concludes that indoor emissions were the largest contributing factor to increased COHb levels during these winter periods. One recent study by the National As- sociation for Sanitarians included investigations of the homes of 300 cases of suspected CO poisoning. Over 90% of the homes were positive for CO (Amiro, 1969). While there was no tabulation of overall average levels given, emissions of

HKO217o088 EXPOSURE TO POLLUTANTS 5 200 to 300 ppm and above of CO were reported due to space heaters and stoves. Yates (1967) reported results from a survey including homes and appliances of CO poisoned patients referred to hospitals and randomly selected residents of houses and mobile homes. Thirty-three percent of these residential inspections found excesses of CO. All hospital referrals came from homes in which excessive CO was found. Yates also found that, in a random inspection of nursing homes, 23 % of appliances used were found to be contributing to the excessive levels of CO. Emissions from these appliances ranged from 10 to over 2500 ppm. Similar results were found in a study conducted in Japan by Tanaka et al. (1971), which measured oxygen, CO, and CO2 present when gas cooking stoves were in use. The amount of CO was greatest when pans were over the flames, impeding oxygen flow. This exposure to CO constitutes higher exposures for the cook than for the rest of the family. This may also be true for small children, who tend to stay near their mothers. Wade et al. (1975) extended the work done on gas stove emissions. A study of four homes clearly demonstrated the contribution of gas stoves to elevated levels of NO, NO2, and CO. These levels exceeded those found outdoors during periods of stove use. A similar study by DeRouane et al. (1974), which also included a gas water heater, found overall NOx levels to be quite high (up to 1000--2000µg/m3) coincidental to the running of gas appliances. A number of studies were done in regions with extremely low outdoor pollution. A study was conducted in Nigerian homes by Sofoluwe (1968). The structures tested were in lower-income sections of the city and pollution was generated mainly by cooking devices. Wood, oil, coal, and gas were used for cooking and heating. No one used electricity. While the duration of pollution within structures was short-lived, the exposure of individuals was found to be very high. For CO the range was 100 to 3000 ppm with a mean of 940.2 ppm; for NOZ the range was 0.5 to 50 ppm with a mean of 8.6 ppm; for SO2 the range was 5 to 100 ppm with a mean of 37.8 ppm and for benzene soluble hydrocarbons the range was 25 to 200 ppm with a mean of 56 ppm. (In one home, CO concentrations reached 3000 ppm and in another, benzene reached a level of 200 ppm. Some extreme values observed were due to peaks of very short duration occurring while fires were being lit or when particular woods were used [like eucalyptus].) Standard indicator tubes were utilized to determine pollutant concentrations. As the method is not strictly quantitative, the figures given by Sofoluwe may not be entirely accurate. That the pollutants detected were very high, however, re- mains as fact. Corroboration of Sofoluwe's findings comes from Cleary and Blackburn (1968). Their findings for native huts in New Guinea, however, while determined by similar measurement methods, show much lower concentrations of pollutants. Mean concentrations of smoke density, aldehydes, and CO for the Eastern Highlands were 666fcg/m3, 3.8 ppm, and 150 ppm, respectively. As in the Sofoluwe study, pollutants tested reached very high levels. While the ranges found by Sofoluwe and Cleary in developing countries repre- sent very high exposure to indoor pollutants they may not differ greatly from those found elsewhere. In all communities, whether in developed or developing coun- tries, many homes are heated with cheaper fuels. Cooking is often done with gas . HASTER-SPP 0038370' _

N KO217U089 _ 6 STERLING AND KOBAYASHI or oil, or even with coal, and inefficient appliances and fireplaces are frequently used. The risk of exposure to elevated levels of pollutants occurs in nearly all countries. Although cooking and heating indoors are the most prominent polluters, numerous other sources of pollution exist. Bridbord et al. (1975) mention aerosols and solvents as sources of halogenated hydrocarbons. Minerals such as asbestos and beryllium can be brought into the domestic indoor environment on the clo- thing of factory workers (Lieben et al., 1969; Selikoffet al., 1972). In Selikoffet al. study the levels of asbestos in the workmen's homes were found to exceed background levels. Further studies show that home repair utilizing spackling, patching, and taping compounds may increase exposure to minerals, primarily asbestos, but also quartz and talc (Rohl et al., 1975). Talc from other sources, e.g., talcum powder, may pollute the indoor air as well (Castleman and Fritsch, 1973). Finally, home furnishings (curtains, carpets, etc.) have been found to contribute fibers to the indoor domestic atmosphere (Corn, 1974). TOBACCO-INDUCED POLLUTION IN DOMESTIC PREMISES Unfortunately, most studies have concentrated on public enclosures and only a few on household tobacco smoke pollution. Therefore, relatively little direct in- formation is available of tobacco-induced pollution in the home. Biersteker et al. (1965) obtained information on the smoking habits (i.e., smoker, non smoker, light, heavy, etc.) of occupants in 60 Rotterdam homes. CO and SO2 levels were related to the year of construction of the home, type of heating (gas, oil, or coal), and smoking habits. Of all these measures, smoking habits correlated highest with the levels of indoor pollution. However, it is dif- ficult to separate the effects of smoking from the effects of heating, cooking, and other activities in this study. In another study of a middle-class, single-family dwelling, Lefcoe and Inculet (1971) report that during the period in which a cigarette was smoked, dust levels rose, then returned to normal within an hour. However, these increases did not equal those created by house cleaning, and they occurred concurrent with other activities. McNall (1975) studied cigarette smoke in domestic premises experimentally. In this test, a machine located in the basement of a three bedroom house consumed 35 cigarettes/hour in one case, and 12 cigarettes/hour in another instance. Particu- lates reached -2700 and - 1100 µg/m3 in the two cases, respectively, while outside levels were 60 f.cg/m3. Despite the extreme numbers of cigarettes smoked, particu- lates immediately dropped to outdoor levels when an electronic filter was activated. A similar experiment was conducted by DeRouane et al. (1974). In a closed 50 m3 room of a house, three cigarettes were smoked by a smoking machine over a total of 24 minutes. Peak concentrations were 1000 µg/m' aerosol and 7.5 ppm CO. As information is as yet sparse, an assessment of the contribution of tobacco smoke to domestic pollution must be inferred from relevant studies in public buildings. POLLUTANTS IN PUBLIC BUILDINGS Offices, libraries, schools, and public halls of assembly such as theaters, res- taurants, and areas where individuals gather in large numbers expose people to a HASSTER- SPP 003088

HK®21'70090 EXPOSURE TO POLLUTANTS 7 variety of pollutants and toxic substances which are generated from many sources inside the structure or which penetrate from the outside. All pollutants, generated inside the building or penetrating from the outside, become part of the internal environment. The escape of all pollutants from the building depends on the type of existing ventilation. There are some major differences between household dwellings and public buildings. Public buildings very often are situated in industrialized, more polluted areas. Larger buildings are also better sealed. This is true especially of the new, modern, air-conditioned and completely enclosed office buildings. However, even in older structures, the very size of the building will decrease the amount of ventilation per unit of space. As a consequence, not only must air be brought into the building but active filtration and ventilation are much more important in public buildings than in homes. Various pollution elimination devices or built-in filtration plants must be provided. Filters may be effective in reducing particulate concen- trations but do little in regard to gases such as CO, CO2, NO, NO2, and others. SOZ may be a lesser problem since it is usually absorbed by building materials regard- less of filtration. Finally, many pollutants are generated by the activities of man and machinery inside public buildings just as in submarines. (For a discussion of current work on building pollution-reducing systems see Holcombe et at. (1971).) The contribution of indoor- and outdoor-generated pollution is much more dif- ficult to determine in large buildings than in households. However, both sources have been clearly identified. A great deal of attention has always been paid to adequate ventilation in public schools. Much of the concern in the past was with odor problems. Korenevskaya et al. (1965) observe that upper floors in school buildings get pollution from kitchens, gyms, boiler rooms, and other structures that are located below them. They noticed a very definite increase in smell, dust, and CO levels. Grusha et al. (1964) measured changes in relative humidity and CO2 as an index of metabolic by-products of school children. They found that relative humidities in schools rose to one and a half times the accepted level by the end of the first class period (values given were 78 to 80%). Temperatures also increased rapidly. The inves- tigators observed that while CO2 levels were normal at the beginning of a class period, they were double by its end. Although the function of ventilation and air-conditioning units is to renew and purify internal building atmospheres, they may do just the reverse. Banaszak et a!. (1970, 1974); and Fink et al. (1971) report air-conditioning and heating units con- taminated with thermophilic fungi. The systems, in turn, pollute indoor spaces with the fungi. Another source of pollution arises from the current building prac- tice of designing ceiling spaces as return air plenums. Air is allowed to circulate through these areas which have been sprayed with asbestos. The asbestos is gradually eroded and circulated throughout the building (Castleman and Fritsch, 1973). For most public and office buildings the relation of outdoor to indoor pollution is exceedingly important. Studies have compared outdoor to indoor dust, SO2, CO, and hydrocarbons (for buildings with and without filtration). DeRouane (197t) found that indoor (total particulate) concentrations varied to t . ~~STER-5PP 003863

HK121'7U09 1 8 STERLING AND KOBAYASHI some extent, depending on whether the building was old or new, but were gener- ally around 80% of outside values. However, SO2 was much reduced due to surface absorption (approximately 25% of outdoor levels). The Japan Air Cleaning Association (in 1968) examined indoor and outdoor sources of various pollutants: SOZ, CO, hydrocarbons, and dusts, in rooms with and without filtration. As in the DeRouane study, SOz was found to be one-fifth of the outdoor concentration. Carbon filters were found to be effective in reducing dust. However, there was no appreciable decrease in CO and hydrocarbon con- centrations with the carbon filters in operation. Yocom et af. (1971a,b) also found little relation between indoor and outdoor concentrations of SOZ (again because of the absorption in internal structures). Particulate concentrations were highest in buildings near roadways but generally were found to be lower indoors than outdoors. However, the organic fraction of particulate matter was consistently higher inside than outside in all public build- ings. At times more than twice the organic contamination was found inside than outside. (In contrast, lead enrichment of particulate matter was about the same indoors as outdoors.) The soiling index in all buildings was about 80 to 90% of outdoor levels (except in a library where, during winter, it was only 50%). CO levels indoors showed a direct correlation with the structures' proximity to road- ways, but CO was spread fairly evenly throughout the buildings. Filtration was not effective in reducing CO levels indoors and mean indoor CO levels were higher than mean outdoor CO levels. At the same time, the semi sealed buildings prevented the escape of CO. CO levels rose sharply, beginning around 7:00 AM in response to the buildup of outdoor CO due to traffic. But, after traffic reached its peak, indoor CO levels remained extremely high for long periods, while outside levels decreased. Yocom's findings are replicated in part in a study by Godin et a!. (1972). In a building in which indoor CO values averaged 2.2 ± 1.3 ppm on the first floor and 2.8 ± 1.5 ppm on the second floor, fluctuations were similar to those found by Yocom et al. With the windows and doors shut, indoor concentrations fell less rapidly than outdoor concentrations. A number of studies have been conducted on particulate matter. Jacobs et al. (1962) found that indoor dust contained more small particulate matter. Jacobs et at. (1962) found that indoor dust contained more small particles than outdoor dust (I µm or less). Jacobs also supplied a number of measures for amounts of particu- lates found inside buildings, ranging between 4.0 and 53.4 mp/ft3 of particulate matter. Lead as a component of indoor dust, in addition to other components, was reported by Hunt and Cadoff (1971) and McNesby et al. (1972). Both studies found lead to be a consistent trace element, along with ammonium sulfate, in both indoor and outdoor dust. Few studies exist that measure air pollutants in public places of assembly as opposed to public-office-type buildings. One important study was conducted by Matsumoto and Kitamura (1971), who measured CO2 and dust concentrations in the underground market streets of Osaka-in tea rooms, bowling alleys, movie theaters, and basements of department stores. CO2 was found to be higher in all areas than in the outside air with the exception of the street itself. Dust under- ground was found to be double that above ground, with peaks of ten times the . MR~TER-~~'~ ~~~~~0

N KE21'70002 EXPOSURE TO POLLUTANTS 9 outside concentration. In department stores, dust was found to be "severe" in the underground floor. Unfortunately, there was no analysis on the organic content of this dust. One other study (Johnson et al., 1975a) measured CO in a public place of assembly. Ice resurfacing machines operating in indoor rinks were found to be a source of CO levels of up to 304 ppm on the average. TOBACCO-INDUCED POLLUTION IN PUBLIC BUILDINGS Measurements have been made of smoke constituents of room air under natural and experimental conditions. Also measured have been blood levels of COHb and nicotine contents of urines. Bridge and Corn (1972) measured CO during two experimental "parties." In one 5120 ft3 room containing 50 people, 25 people consumed 50 cigarettes and seven cigars in I~i hours. With a room air exchange rate of seven times per hour, CO averaged 7 ppm during the course of the party. During the second experiment in a 3750 ft3 room containing 73 people, 36 smokers consumed 63 cigarettes and 10 cigars in 11h hours and the average CO content was 9 ppm. These values actually coincided with values predicted using Turk's (1963) equation (6.5 and 8 ppm, respectively). In order to determine mainstream to sidestream smoke ratios produced by cigarettes, Hoegg (1972) measured CO and total particulate matter from varying numbers of cigarettes. In a sealed 25 m3 chamber, CO levels increased with the number of cigarettes smoked. Concentrations ranged from - 10 ppm for 4 ciga- rettes to 69.8 ppm for 24 cigarettes. For total particulate matter, initial or peak values ranged from -2.5 mg/m3 for 4 cigarettes to 16.65 mg/m3 for 24 cigarettes. Utilizing these experimentally obtained values, Hoegg modified Turk's equation with the addition of a decay function for cigarette-produced particulate matter. A study by Anderson and Dalhamn (1973) determined, in addition to CO, nicotine and smoke density produced by cigarettes in a medium-sized meeting room (80m3). Fifty cigarettes were smoked in 120 minutes. With six air changesJ hour, initial levels were 2 ppm and average peaks during smoking were around 6 ppm. Smoke density prior to testing was 0.02 mg/m3. Highest concentrations were found at the beginning of the experiment but they rapidly dissipated. Nicotine content of the air increased from zero to 0.377 mg/m3 during the course of the experiment, but it rapidly decreased also. The seven smokers and five non-smok- ers in the experiment were tested for their COHb levels. Changes in non-smoker COHb were not significant. Harke conducted several experiments with cigarettes in enclosed office rooms under conditions of "severe" and "realistic" smoking. Twenty-one persons smoking two cigarettes each within 16 to 18 minutes in a room 57 m3 produced 0.5 mg/m3 nicotine and 49 ppm CO. Ventilating the room decreased these concentra- tions by 80%. In the case of one person smoking i 1 cigarettes in 5 hours in a room 30 m3, nicotine reached 0.04 mg/m3 and CO was still under 10 ppm. With the window closed, nicotine was 0.06 to 0.09 mglm3 and CO was still under 10 ppm. (Background pollutants, however, were not mentioned (Harke, 1970).) In a number of experiments in 1972, Harke measured pollutants in large and small 11 M~"t~! t h.R"3PP k~'~.,t~~ ..~ ~

HKE2170093 10 STERLING AND KOBAI'ASHI rooms under extreme conditions. In the first experiment, a smoking machine consumed 30, 15, 10, and 5 cigarettes in 13 minutes in a small room (38.2m;). After 30 cigarettes had been smoked, 0.52 mg/m3 nicotine was found in the room air. In 21 minutes, 0.46 mg acrolein/m' was reached. AcetaldehydeJm' attained a level of 6.5 mg in the same time period. The highest concentration of CO, 64 ppm, was reached immediately after smoking. With 5 cigarettes smoked in 13 minutes and without ventilation, CO was 11.5 ppm at the end of smoking; nicotine was 0.06 mglm3; acrolein reached 0.07 mg/m3; and acetaldehyde was 1.3 mg/m'. In a considerably larger (170m3) second test room, after the machine had smoked 150 cigarettes in 60 minutes, CO was 53 ppm; nicotine reached 0.69 mg/m3; acrolein 0.38 mg/m3; and acetaldehyde 4.2 mg/m3. Ventilation reduced all levels by a factor of 2 to 5. In every test situation, with or without ventilation, ail constituents fell rapidly with time after smoking, nicotine being the most rapid. Harke (1974d) measured particulate matter produced by 30 cigarettes in a 38 m3 office room. In eight determinations, average concentrations ranged from 20.8 mgJm3 in 11 to 31 minutes to 16.2 mg/m3 in 41 to 61 minutes. Particulate concentra- tions rapidly diminished with time at the end of the smoking phase. Russell et al. (1973) studied room contamination and subject COHb levels of 21 volunteers who spent I hour in a 15 x 12 x 8 ft unventilated room. Before the test, 30 cigarettes were left to burn in ashtrays. An additional 32 cigarettes and two cigars were smoked, and 18 cigarettes were left smoldering. After 18 minutes, CO reached 37 and 32.5 ppm (two samples). After 53 minutes, CO reached 41.8 and 41.3 ppm. the mean level CO was 38.2 for the entire experiment. The non- smokers' mean COHb levels were 1.6% before and 2.6% after the experiments. Some very preliminary results of cigarette-produced CO pollution were re- ported by Lawther and Commins (1970). In a 15 m3 exposure chamber, CO rose to 20 ppm after seven cigarettes were smoked in 1 hour. Particulate matter reached 3 mg/m3. The ventilation rate was one room change per hour. Further details, how- ever, were not specified. Harmsen and Effenberger (1957) reported results from an experiment con- ducted in an unventilated 98 m3 room where a number (unspecified) of persons smoked a large (62) number of "nicotine-rich" cigarettes in 30 minutes. CO reached 0.008% by volume or 80 ppm, and nicotine was 5.2 mghn3. (These high values, however, have never been replicated by any other investigator.) Dublin (1972) burned two standard-brand unfiltered cigarettes. The room was medium-sized, 18 x 30 x 9 ft. Compared to background levels of 1 ppm of CO, a transient peak immediately after lighting a cigarette and in the immediate vicinity of the smoker was between 20.5 and 32.5 ppm (simultaneous samples). Further away, the levels were 13 and 17 ppm. Five minutes after smoking, the room reached equilibrium at 2 ppm of CO. The high initial concentration was the direct result of lighting up. A number of investigations report on cigarette-produced pollution under natural conditions. CO was monitored for 18 days by Harke (1974a) in two office build- ings, one air-conditioned, the other not. No significant overall increase in CO was found after employees started to smoke. The CO curve instead correlated well 0 KASTER-SPP 003692