The Science of the Total Environment, 11 (1979) 53--58 © Elsevier Scientific Publishing Company, Amsterdam -- Printed in England
AN APPRAISAL OF RELATIVE AIRBORNE SUB-URBAN CONCENTRATIONS OF POLYCYCLIC AROMATIC HYDROCARBONS MONITORED INDOORS AND OUTDOORS
J. D. BUTLER AND P. CROSSLEY
Department of Chemistry, University of Aston, Birmingham, B4 7ET (England) (Received July l l t h , 1978; in final form August 15th, 1978)
ABSTRACT
Particle,size distribution studies of polycyclic aromatic hydrocarbons in city and sub-urban atmospheres indicate that these compounds are associated with particles having MMD values of a b o u t 0.5 pro. Ambient airborne concentrations of pyrene, chrysene, b e n z o [ a ] p y r e n e , b e n z o [ e ] p y r e n e and coronene at suburban sites between 11 and 17 km from a city centre are shown to be approximately similar inside and outside detached residential houses. At all sites, monitored chrysene appears in highest concentration in the range 4 - - 7 n g m -3, followed by b e n z o [ a ] p y r e n e and b e n z o [ e ] p y r e n e between 2--4 ng m - 3 , with coronene generally less than 1 ng m-3 By taking b e n z o [ a ] p y r e n e as a typical example of this class of compounds, these results indicate that the lungs of sub-urban populations will be exposed to about 34 ng of benzo[a ]pyrene per day. This background amount is roughly equivalent to the exposure to this chemical experienced b y the lung when one cigarette is smoked per day.
INTRODUCTION
It has been established that indoor airborne lead concentrations approach those existing out-of-doors, especially at night [1]. The principal reason for this arises from the fact that lead emissions from m o t o r vehicles are predominantly in the sub-micron particle-size range [2]. Such particles will survive in the atmosphere in suspended form until either growth through coagulation eventually causes gravitational deposition [3] or impaction and capture at a surface results in their removal from the atmosphere [4]. Besides lead particles, polycyclic aromatic hydrocarbons exist in urban aerosols and evidence from Canada [5] and the U.S.A. [6] indicates that these compounds are likewise associated with sub-micron particles. This paper confirms that this is also true in a large city in the U.K. The fact that many polycyclic aromatic hydrocarbons are known carcinogens has p r o m p t e d this investigation, so that some knowledge of the extent to which urban and sub-urban populations are exposed to these compounds may be made available.
54 MONITORING PROCEDURE
(i) R o u t i n e collection and analysis Atmospheric particulate matter was collected on 20 x 25 cm 2 GFA filters m o u n t e d on Staplex 'Hi-Vol' air samplers operating at 0.57 m 3 min -1 . Sampling times of 3 to 4 days, sufficient for the filtration of a b o u t 3,000 m 3 of air were employed. After collection the filters were extracted with dimethyl sulphoxide at boiling-water-bath temperatures for 12 h. The extract was filtered and the filtrate diluted with an equal volume of water. This aqueous extract was shaken with n-pentane (x3) to concentrate the polycyclic aromatic hydrocarbons into the n-pentane layer. These layers were separated off, then combined, before the n-pentane was evaporated at room temperature. The residue was dissolved in cyclohexane. The cyclohexane solution of polycyclic aromatic hydrocarbons was chromatographed on an activated alumina column using cyclohexane as eluting solvent. The fractions collected from the column were transferred to 1-cm silica cells and analyzed by scanning the UV spectrum between 200 and 450 nm using a Unicam SP800.
(ii) Particle size evaluation An Andersen 5-stage cascade impactor, Model 67-000, operating at 0.57 m 3 rain -1 was used for the particle-size distribution studies. This instrument fractionates airborne particulate matter into the following size ranges: > 7 . 0 pm; 3.3--7.0 #m; 3.3--2.0 pm; 1.1--2.0 p m and < 1 . 1 pm. After sampling, the impactor was dismantled and the weight collected on each stage was determined. The glass-fibre collection surfaces for each stage were extracted b y immersion in dimethyl sulphoxide and analyzed for polycyclic aromatic hydrocarbons as described in the previous section.
(iii) Location o f sites (a) Outdoor sites. Three o u t d o o r sites designated S-C, C-C and S-U corresponding to Salford Circus at the Midlands m o t o r w a y interchange M6-A38 (M) Birmingham, U.K., a city centre site on the University of Aston campus and at a sub-urban house approximately 15 km due north from the city centre have been used. These sites represent contrasting environments, near heavy vehicular traffic, typical city centre although not in close proximity to traffic and residential, respectively. (b) Indoor sites. Three detached residential houses situated away from main roads have been employed, measured from the city centre, t w o were due north at distances of 11 and 15 km and the other was 17 km due west. None of the occupants of these houses smoked cigarettes, cigars or pipes. Staplex 'Hi-Vol' pumps were used for the collection of indoor airborne particulate matter, operating under similar conditions to those employed outside.
55 TABLE 1 AEROSOL CHARACTERISTICS F O R O U T D O O R SITES
OF TOTAL
AIRBORNE
Site
MMD (pro)
og
~1/~m
~2 pm
~3 pln
S--C C--C S--U
1.15 1.25 1.15
7.0 6.2 13.0
48 45 48
62
69 67
59 59
P A R T I C U L A T E MA'FrER
65
RESULTS
(i) Particle-size evaluation Table 1 summarizes the aerosol characteristics of total particulate m a t t e r found at the three o u t d o o r sites. The data for this table have been compiled by weighing the a m o u n t of particulate collected on each stage of the Andersen impactor and then plotting the cumulative percent mass < particle diameter vs. particle diameter on log-probability paper in the usual manner [7]. The mass median equivalent diameters (MMD's) have been read from the plot at the 50% probability p o i n t and the standard geometric deviation (og) has been estimated from the ratio of the distribution at the 84 and 50 percent values. Table 2 gives the results of the chemical analysis of the impactor stages at the three sites and indicates the particle size with which the polynuclear hydrocarbons are associated.
(ii) Atmospheric concentrations Table 3 shows the concentrations of airborne polycyclic aromatic hydrocarbons monitored indoors and compares it with that f o u n d outside at TABLE 2 DISTRIBUTION CHARACTERISTICS OF THE POLYCYCLIC AROMATIC HYDROCARBONS IN OUTDOOR AIR Polynuclear aromatic hydrocarbon
Site
MMD (/~m)
Pyrene
S--C C--C S--U
1.85 0.55 0.60
Benzo[a ]pyrene
Coronene
ag
Percentage ~ 90% on backup filter
86 83
90 89
S--C C--C S--U
0.56 0.56 0.40
83 70 79
89 75 89
2.6 23.6 14.2
3.8 11.4 7.3
67 60 67
56 TABLE 3 COMPARISON
OF EXTERNAL
A N D I N T E R N A L C O N C E N T R A T I O N S O F POLY-
CYCLIC AROMATIC HYDROCARBON AT SUB-URBAN SITES Limits q u o t e d are at the 10% significance level.
TPM (pgrn -3) Pyrene Chrysene Benzo[a]pyrene Benzo[e]pyrene Coronene Pyrene Chrysene Benzo[a]pyrene Benzo[e]pyrene Coronene
Outside
Inside
47.5 -+ 13.2 ng m - 3 2.62 + 1.42 4.56 -+ 2.19 2.88 -+ 0.54 2.19 + 0.95 0.92 + 0.38 /ag g - I 54.3 + 21.08 91.2 + 30.71 72.0 -+ 42.67 46.5 + 14.10 21.8 + 7.72
56 + 41.3 ng m - 3 1.32 + 1.41 3.98 + 3.31 2.10 + 1.88 2.28 -+ 2.16 0.43 + 0.24 p.g g - I 24.97 + 27.2 72.55 -+ 63.7 41.00 + 37:4 40.80 -+ 43.07 7.32_+ 1.46
the sub-urban site. The results are expressed b o t h in terms of weight per cubic metre of air sampled and as weight per unit weight of airborne particulate collected on the filters.
DISCUSSION
The notable feature of these results is that despite the environmentally contrasting nature of the sites, aerosol characteristics are remarkably similar with MMD's of total particulate all around 1 pm. Such a distribution implies that these particles are respirable and can proceed into the human b o d y b e y o n d the trachea. This statement assumes that the MMD's determined b y the impactor are truly representative of the aerosol sampled and that aggregated s o o t particles, for instance, do not disintegrate into smaller fragments during collision with the impactor surface. The other principal finding reported in Table 3 is that the concentrations of the polycyclic hydrocarbons measured are of the same order of magnitude inside and outside. In both cases, chrysene is present in highest concentrations of b e t w e e n 4 and 7 n g m - 3 , followed b y b e n z o [ a ] p y r e n e and benzo[e] pyrene b e t w e e n 2 and 3 ng m -3 with coronene generally less than 1 n g m -3 These monitored concentrations can be used to estimate the dally background intake of these c o m p o u n d s by sub-urban populations. Taking b e n z o [ a ] p y r e n e as an example, we can see that a person breathing an average 15 m 3 of air per day will inspire during this period 15 x 2.5 ng. As the distributions in Table 2 show, 90 percent of the b e n z o [ a ] p y r e n e is associated with particles that are less than 3 pm in size and which consequently are capable of penetrating the lung as far as the bronchi. Thus,
57 a background exposure to the lung of about 15 x 2.5 x 90/100 ~ 34 ng per day of benzo[a] pyrene is to be expected. For comparative purposes, it is relevant to enquire how cigarette smoking, the smoke from which also contains benzo[a]pyrene can add to this background intake. The data collected by Wynder and Hoffmann [8] can be used as the basis of simple calculation. According to the information summarized by these authors, one cigarette will deliver about 39 ng of benzo[a]pyrene to the body by way of mainstream smoke, i.e., smoke drawn through the cigarette, when smoked at a rate of one puff of two seconds duration every minute. Cigarette smoke held in the lungs for 5 and 30 sec will result in 82 and 97 percent retention, respectively, so assuming about 85 percent retention as average, something like 33 ng of benzo[a]pyrene will be captured by the lungs. Hence, it is apparent that the exposure to benzo [a ] pyrene in one day due to breathing ambient air is approximately equivalent to smoking one cigarette per day. Although carcinogens may enter the lungs, it is important to appreciate that the carcinogen is present in an adsorbed state, usually on carbon particles and may not, therefore, come into contact with lung tissue. Animal experiments have revealed that consideration should be given to particle size and to the rate at which polycyclic aromatic hydrocarbons are leached from the carbon carriers present in ambient aerosols. Creasia et al. [9] have recently shown that 50 percent of benzo[a] pyrene adsorbed on 0.5--1.0 gm sized carbon particles intratracheally instilled into mice was cleared from lung tissue after 1.5 days, but that benzo[a]pyrene adsorbed on 15--30 pm sized carbon particles requires 4--5 days for 50 percent elimination. Furthermore, in the latter case lung clearance of carbon particles and benzo[a] pyrene was at the same rate. This implies that benzo[a] pyrene was not eluted from the carrier and did not contact lung tissue. Thus, one broad conclusion that emerges is that carcinogenic material associated with particles around 1 pm or less as reported in Table 2 and which is typical of ambient aerosols, is more likely to contact lung tissue whilst resident in the lung, than carcinogens adsorbed on particles an order of magnitude larger.
ACKNOWLEDGEMENTS
We thank the Department of the Environment through the Transport and Road Research Laboratory, the Department of Health and Social Security for a research studentship to P.C. and the Health and Safety Directorate of the Commission of the European Communities, Luxemburg for supporting this work. REFERENCES 1 2
J.D. Butler and S. D. MacMurdo, Int. J. Environ. Stud., {3 (1974) 181. R.E. Lee, Jr., R.K. Patterson, W.L. Crider and J. Wagman, Atmos. Environ., 5
(1971) 225.
58 3 4 5 6 7 8 9
D.A. Gillette, Atmos. Environ., 6 (1972) 451. A.C. Chamberlain, Contemp. Phys., 8 (1967) 561. R.C. Pierce and M. Katz, Environ. Sci. Technol., 9 (1975) 347. L. Demaio and M. Corn, J. Air Pollut. Control Assoc., 16 (1966) 67. R.E. Lee, Jr., and S. Goransen, Environ. Sci. Technol., 6 (1972) 1019. E.L. Wynder and D. Hoffmann, in A. Haddow and S. Weinhouse (Eds.), Advances in Cancer Research, Vol. 8, Academic Press, New York and London, 1964, p. 249. D.A. Creasia, J. K. Poggenburg, Jr., and P. Nettesheim, J. Toxicol. Environ. Health, 1 (1976) 967.
AN APPRAISAL OF RELATIVE AIRBORNE SUB-URBAN CONCENTRATIONS OF POLYCYCLIC AROMATIC HYDROCARBONS MONITORED INDOORS AND OUTDOORS
J. D. BUTLER AND P. CROSSLEY
Department of Chemistry, University of Aston, Birmingham, B4 7ET (England) (Received July l l t h , 1978; in final form August 15th, 1978)
ABSTRACT
Particle,size distribution studies of polycyclic aromatic hydrocarbons in city and sub-urban atmospheres indicate that these compounds are associated with particles having MMD values of a b o u t 0.5 pro. Ambient airborne concentrations of pyrene, chrysene, b e n z o [ a ] p y r e n e , b e n z o [ e ] p y r e n e and coronene at suburban sites between 11 and 17 km from a city centre are shown to be approximately similar inside and outside detached residential houses. At all sites, monitored chrysene appears in highest concentration in the range 4 - - 7 n g m -3, followed by b e n z o [ a ] p y r e n e and b e n z o [ e ] p y r e n e between 2--4 ng m - 3 , with coronene generally less than 1 ng m-3 By taking b e n z o [ a ] p y r e n e as a typical example of this class of compounds, these results indicate that the lungs of sub-urban populations will be exposed to about 34 ng of benzo[a ]pyrene per day. This background amount is roughly equivalent to the exposure to this chemical experienced b y the lung when one cigarette is smoked per day.
INTRODUCTION
It has been established that indoor airborne lead concentrations approach those existing out-of-doors, especially at night [1]. The principal reason for this arises from the fact that lead emissions from m o t o r vehicles are predominantly in the sub-micron particle-size range [2]. Such particles will survive in the atmosphere in suspended form until either growth through coagulation eventually causes gravitational deposition [3] or impaction and capture at a surface results in their removal from the atmosphere [4]. Besides lead particles, polycyclic aromatic hydrocarbons exist in urban aerosols and evidence from Canada [5] and the U.S.A. [6] indicates that these compounds are likewise associated with sub-micron particles. This paper confirms that this is also true in a large city in the U.K. The fact that many polycyclic aromatic hydrocarbons are known carcinogens has p r o m p t e d this investigation, so that some knowledge of the extent to which urban and sub-urban populations are exposed to these compounds may be made available.
54 MONITORING PROCEDURE
(i) R o u t i n e collection and analysis Atmospheric particulate matter was collected on 20 x 25 cm 2 GFA filters m o u n t e d on Staplex 'Hi-Vol' air samplers operating at 0.57 m 3 min -1 . Sampling times of 3 to 4 days, sufficient for the filtration of a b o u t 3,000 m 3 of air were employed. After collection the filters were extracted with dimethyl sulphoxide at boiling-water-bath temperatures for 12 h. The extract was filtered and the filtrate diluted with an equal volume of water. This aqueous extract was shaken with n-pentane (x3) to concentrate the polycyclic aromatic hydrocarbons into the n-pentane layer. These layers were separated off, then combined, before the n-pentane was evaporated at room temperature. The residue was dissolved in cyclohexane. The cyclohexane solution of polycyclic aromatic hydrocarbons was chromatographed on an activated alumina column using cyclohexane as eluting solvent. The fractions collected from the column were transferred to 1-cm silica cells and analyzed by scanning the UV spectrum between 200 and 450 nm using a Unicam SP800.
(ii) Particle size evaluation An Andersen 5-stage cascade impactor, Model 67-000, operating at 0.57 m 3 rain -1 was used for the particle-size distribution studies. This instrument fractionates airborne particulate matter into the following size ranges: > 7 . 0 pm; 3.3--7.0 #m; 3.3--2.0 pm; 1.1--2.0 p m and < 1 . 1 pm. After sampling, the impactor was dismantled and the weight collected on each stage was determined. The glass-fibre collection surfaces for each stage were extracted b y immersion in dimethyl sulphoxide and analyzed for polycyclic aromatic hydrocarbons as described in the previous section.
(iii) Location o f sites (a) Outdoor sites. Three o u t d o o r sites designated S-C, C-C and S-U corresponding to Salford Circus at the Midlands m o t o r w a y interchange M6-A38 (M) Birmingham, U.K., a city centre site on the University of Aston campus and at a sub-urban house approximately 15 km due north from the city centre have been used. These sites represent contrasting environments, near heavy vehicular traffic, typical city centre although not in close proximity to traffic and residential, respectively. (b) Indoor sites. Three detached residential houses situated away from main roads have been employed, measured from the city centre, t w o were due north at distances of 11 and 15 km and the other was 17 km due west. None of the occupants of these houses smoked cigarettes, cigars or pipes. Staplex 'Hi-Vol' pumps were used for the collection of indoor airborne particulate matter, operating under similar conditions to those employed outside.
55 TABLE 1 AEROSOL CHARACTERISTICS F O R O U T D O O R SITES
OF TOTAL
AIRBORNE
Site
MMD (pro)
og
~1/~m
~2 pm
~3 pln
S--C C--C S--U
1.15 1.25 1.15
7.0 6.2 13.0
48 45 48
62
69 67
59 59
P A R T I C U L A T E MA'FrER
65
RESULTS
(i) Particle-size evaluation Table 1 summarizes the aerosol characteristics of total particulate m a t t e r found at the three o u t d o o r sites. The data for this table have been compiled by weighing the a m o u n t of particulate collected on each stage of the Andersen impactor and then plotting the cumulative percent mass < particle diameter vs. particle diameter on log-probability paper in the usual manner [7]. The mass median equivalent diameters (MMD's) have been read from the plot at the 50% probability p o i n t and the standard geometric deviation (og) has been estimated from the ratio of the distribution at the 84 and 50 percent values. Table 2 gives the results of the chemical analysis of the impactor stages at the three sites and indicates the particle size with which the polynuclear hydrocarbons are associated.
(ii) Atmospheric concentrations Table 3 shows the concentrations of airborne polycyclic aromatic hydrocarbons monitored indoors and compares it with that f o u n d outside at TABLE 2 DISTRIBUTION CHARACTERISTICS OF THE POLYCYCLIC AROMATIC HYDROCARBONS IN OUTDOOR AIR Polynuclear aromatic hydrocarbon
Site
MMD (/~m)
Pyrene
S--C C--C S--U
1.85 0.55 0.60
Benzo[a ]pyrene
Coronene
ag
Percentage ~ 90% on backup filter
86 83
90 89
S--C C--C S--U
0.56 0.56 0.40
83 70 79
89 75 89
2.6 23.6 14.2
3.8 11.4 7.3
67 60 67
56 TABLE 3 COMPARISON
OF EXTERNAL
A N D I N T E R N A L C O N C E N T R A T I O N S O F POLY-
CYCLIC AROMATIC HYDROCARBON AT SUB-URBAN SITES Limits q u o t e d are at the 10% significance level.
TPM (pgrn -3) Pyrene Chrysene Benzo[a]pyrene Benzo[e]pyrene Coronene Pyrene Chrysene Benzo[a]pyrene Benzo[e]pyrene Coronene
Outside
Inside
47.5 -+ 13.2 ng m - 3 2.62 + 1.42 4.56 -+ 2.19 2.88 -+ 0.54 2.19 + 0.95 0.92 + 0.38 /ag g - I 54.3 + 21.08 91.2 + 30.71 72.0 -+ 42.67 46.5 + 14.10 21.8 + 7.72
56 + 41.3 ng m - 3 1.32 + 1.41 3.98 + 3.31 2.10 + 1.88 2.28 -+ 2.16 0.43 + 0.24 p.g g - I 24.97 + 27.2 72.55 -+ 63.7 41.00 + 37:4 40.80 -+ 43.07 7.32_+ 1.46
the sub-urban site. The results are expressed b o t h in terms of weight per cubic metre of air sampled and as weight per unit weight of airborne particulate collected on the filters.
DISCUSSION
The notable feature of these results is that despite the environmentally contrasting nature of the sites, aerosol characteristics are remarkably similar with MMD's of total particulate all around 1 pm. Such a distribution implies that these particles are respirable and can proceed into the human b o d y b e y o n d the trachea. This statement assumes that the MMD's determined b y the impactor are truly representative of the aerosol sampled and that aggregated s o o t particles, for instance, do not disintegrate into smaller fragments during collision with the impactor surface. The other principal finding reported in Table 3 is that the concentrations of the polycyclic hydrocarbons measured are of the same order of magnitude inside and outside. In both cases, chrysene is present in highest concentrations of b e t w e e n 4 and 7 n g m - 3 , followed b y b e n z o [ a ] p y r e n e and benzo[e] pyrene b e t w e e n 2 and 3 ng m -3 with coronene generally less than 1 n g m -3 These monitored concentrations can be used to estimate the dally background intake of these c o m p o u n d s by sub-urban populations. Taking b e n z o [ a ] p y r e n e as an example, we can see that a person breathing an average 15 m 3 of air per day will inspire during this period 15 x 2.5 ng. As the distributions in Table 2 show, 90 percent of the b e n z o [ a ] p y r e n e is associated with particles that are less than 3 pm in size and which consequently are capable of penetrating the lung as far as the bronchi. Thus,
57 a background exposure to the lung of about 15 x 2.5 x 90/100 ~ 34 ng per day of benzo[a] pyrene is to be expected. For comparative purposes, it is relevant to enquire how cigarette smoking, the smoke from which also contains benzo[a]pyrene can add to this background intake. The data collected by Wynder and Hoffmann [8] can be used as the basis of simple calculation. According to the information summarized by these authors, one cigarette will deliver about 39 ng of benzo[a]pyrene to the body by way of mainstream smoke, i.e., smoke drawn through the cigarette, when smoked at a rate of one puff of two seconds duration every minute. Cigarette smoke held in the lungs for 5 and 30 sec will result in 82 and 97 percent retention, respectively, so assuming about 85 percent retention as average, something like 33 ng of benzo[a]pyrene will be captured by the lungs. Hence, it is apparent that the exposure to benzo [a ] pyrene in one day due to breathing ambient air is approximately equivalent to smoking one cigarette per day. Although carcinogens may enter the lungs, it is important to appreciate that the carcinogen is present in an adsorbed state, usually on carbon particles and may not, therefore, come into contact with lung tissue. Animal experiments have revealed that consideration should be given to particle size and to the rate at which polycyclic aromatic hydrocarbons are leached from the carbon carriers present in ambient aerosols. Creasia et al. [9] have recently shown that 50 percent of benzo[a] pyrene adsorbed on 0.5--1.0 gm sized carbon particles intratracheally instilled into mice was cleared from lung tissue after 1.5 days, but that benzo[a]pyrene adsorbed on 15--30 pm sized carbon particles requires 4--5 days for 50 percent elimination. Furthermore, in the latter case lung clearance of carbon particles and benzo[a] pyrene was at the same rate. This implies that benzo[a] pyrene was not eluted from the carrier and did not contact lung tissue. Thus, one broad conclusion that emerges is that carcinogenic material associated with particles around 1 pm or less as reported in Table 2 and which is typical of ambient aerosols, is more likely to contact lung tissue whilst resident in the lung, than carcinogens adsorbed on particles an order of magnitude larger.
ACKNOWLEDGEMENTS
We thank the Department of the Environment through the Transport and Road Research Laboratory, the Department of Health and Social Security for a research studentship to P.C. and the Health and Safety Directorate of the Commission of the European Communities, Luxemburg for supporting this work. REFERENCES 1 2
J.D. Butler and S. D. MacMurdo, Int. J. Environ. Stud., {3 (1974) 181. R.E. Lee, Jr., R.K. Patterson, W.L. Crider and J. Wagman, Atmos. Environ., 5
(1971) 225.
58 3 4 5 6 7 8 9
D.A. Gillette, Atmos. Environ., 6 (1972) 451. A.C. Chamberlain, Contemp. Phys., 8 (1967) 561. R.C. Pierce and M. Katz, Environ. Sci. Technol., 9 (1975) 347. L. Demaio and M. Corn, J. Air Pollut. Control Assoc., 16 (1966) 67. R.E. Lee, Jr., and S. Goransen, Environ. Sci. Technol., 6 (1972) 1019. E.L. Wynder and D. Hoffmann, in A. Haddow and S. Weinhouse (Eds.), Advances in Cancer Research, Vol. 8, Academic Press, New York and London, 1964, p. 249. D.A. Creasia, J. K. Poggenburg, Jr., and P. Nettesheim, J. Toxicol. Environ. Health, 1 (1976) 967.
