- Email: [email protected]
outdoor inhalable and respirable particles in domestic environments
the Science of the Total Environment ELSEVIER
The Science of the Total Environment 151 (1994) 205-211
Relationships of indoor/outdoor inhalable and respirable particles in domestic environments C h i h - S h a n Li* DiLision of Ent~ironmental Health Science, Institute of Public Health, College of Medicine, National Taiwan Unitersity, No.1 Jen Al Road, 1st Sect., Taipei, Taiwan, ROC Received 4 May 1993; accepted 6 July 1993
Abstract
Simultaneous air monitoring inside and outside of three domestic environments in residential and industrial communities were conducted to evaluate infiltrations of outdoor aerosols into indoor environments, and indoor aerosol characteristics. The relationships of total (TSP), inhalable (< 10 ~m, PM10) and respirable ( < 2.5 /xm, PM2.5) suspended particles were compared. It was found that the fractions of the indoor (20%) and outdoor (40%) PM10 levels exceeding 150 /xg/m 3 were substantial. In addition, it was indicated that indoor TSP and PM10 concentrations (R z = 0.99) were well correlated with their corresponding outdoor levels. Moreover, there was a close relationship between the paired PM10 and PM2.5 levels indoors (R 2= 0.75) and outdoors (R z = 0.83). The average indoor/outdoor ( I / O ) and median I / O ratios observed were 0.60. Therefore, steps to mitigate the situations with particulate levels above ambient particulate standards requires further study.
Keywords: Indoor air quality; Residental environment; pH 10; pH 2.5
1. Introduction Indoor air quality has become an important issue in recent years, since activity investigations indicate that people spend 80-90% of their time indoors [1-3]. In addition, indoor air quality could deteriorate through contributions of outdoor particles and the presence of indoor particle sources [4]. Therefore, it is necessary to measure indoor particle levels for the purpose of characterizing total human exposure to airborne particulate matter [5]. There have been a number of studies that
* Corresponding author.
evaluate correlations between indoor and outdoor particles in different domestic environments. In 24 residences in Vermont, USA, the mean mass levels of indoor and outdoor respirable particles during 14 sampling days in the winter were found to be 25 / z g / m 3 and 1 7 / x g / m 3, respectively [6]. Additionally, the average indoor levels exceeded outdoor levels on 12 out of the 14 sampling days. The indoor fine particle concentrations varied from 6 to 40 / z g / m 3, with mean indoor/outdoor ( I / O ) ratios ranging from 0.7 to 2.1 in 11 Houston (USA) homes [7]. Regarding fine and coarse particles in urban and suburban Helsinki, Finland, the geometric mean mass concentrations were found to be 16 /xg/m 3 and 13 /xg/m 3, respectively [8]. The I / O ratios were < 1 in most
0048-9697/94/$07.00 © 1994 Elsevier Science BV. All rights reserved.
SSDI 0048-9697(94)04014-Z
206
C-S. Li / Sci. Total Entiron. 151 (1994) 205-211
situations, however, they were > 1 for elements of fine particles originating from indoor sources [9]. Lioy et al. [10] investigated the indoor and outdoor relationships of PM10 in an industrial community during the winter in Phillipsburg, N J, USA. Here, indoor concentrations in the measured homes were related more to outdoor particulate sources. Another similar investigation was performed in 10 homes in southern California, USA [11], where the observed I / O median ratio was ~ 0.70. Furthermore, the percentages of the indoor and outdoor PM10 observations greater than the California standard (50 /zg/m 3) were 7% and 29%, respectively. The indoor particles were influenced by not only the infiltration of outdoor air, but also indoor particle generation [12]. It was estimated that the mean infiltration of outdoor fine particles was ~ 70% [9]. In addition, the most important single determinant of indoor particle sources was found to be cigarette smoking [7,9]. In some cases, it was also reported that indoor particle levels were highly affected by outdoor traffic [13]. From a field survey in 98 homes in Tucson, Arizona, USA, the mean mass concentrations of PM2.5 and PM10 were observed to be 20 and 4 0 / z g / m 3 for non-smoking, 35 and 55 /~g/m 3 for 1-20 cigarettes/day, and 85 and 100 /zg/m 3 for > 20 cigarettes/day, respectively [14]. Moreover, the I / O ratio of PM2.5 and PM10 was found to be 0.63 for non-smoking homes and 1.1 for those with smoking. In three woodstove homes in Raleigh, NC, USA, the highest I / O fine particle ratio was observed in the home with a leaking woodstove [15]. Another related comparison of indoor and outdoor respirable suspended particles (RSP) was conducted in seven houses during woodburning periods. The I / O ratios of RSP were generally > 1, because of woodburning and other unidentified sources [16]. In brief, the correlations between indoor and outdoor particles are quite different from area to area, because of different characteristics of I / O sources as well as particle infiltration into indoors. Taiwan is a subtropical country with a warm climate throughout the year. The influences of ambient atmosphere on indoor air quality should be different from those observed in colder areas,
because of direct transport into indoors through open doors and windows. To date there have been few investigations regarding specific compounds and compound classes indoors, as well as the relationship of outdoor to indoor concentrations of PM2.5 and PM10 in Taiwan. The purpose of these field measurements was to characterize the particulate matter in the domestic environments of residential and industrial communities in Taiwan (a subtropical country with warm weather all year round) and to compare simultaneously the indoor results with the corresponding outdoor measurements. 2. Materials and methods
2.1. Equipment Open sampling cassettes, without particle-size fractionations, were equipped with personal SKC sampling pumps for collecting TSP at a 3.5-1/min flowrate. PM10 and PM2.5 environmental monitor samplers (model 200, MSP Co., Minneapolis, MN) were used for collection of inhalable and respirable particles at a flowrate of 10 l/min [17]. Both PM10 and PM2.5 samplers had twin impaction plates, providing sharp 10 and 2.5 p,m particle cut sizes, respectively. Plastic filter holders were fitted with a Teflon membrane filter (Teflon; Gelman Filter Co., Ann Arbor, MI) and the flowrates were calibrated in the field with a bulb meter (Gilian Instrument Corp., West Caldwell, N J). Simultaneous field measurements were performed indoors and outdoors with a portable high volume air sampler (Gilian, AirCon-2 with timer and controller to permit sequential collection by PM10 and PM2.5 samplers, as well as by PM10 and a cassette using the same pump). Teflon filters (2/zm pore size, 37 mm diameter) for TSP, PM10 and PM2.5 samplers were placed into a 5-cm petric dish for handling. Particulate filters were pre- and post-weighed following the standardized operating procedure for balance calibration, filter conditioning and filter weighing. Quality control checks included limit checks for flow rates and filter weights as well as range checks of concentrations in relation to sample type and location.
C.-S. Li /Sci. Total En tiron. 151 (1994) 205-211
2.2. Sampling In Taiwan, the homes are located in either residential or industrial communities. To evaluate the concentration differences in TSP, PM10 and PM2.5 particulate matter in these two different areas, the simultaneous indoor and outdoor field samplings of three residences were conducted during a 10-week period, from November 1992 to February 1993, with multiple measurements. Home 1 is a lst-floor apartment located in the residential area. Home 2 (lst floor) and Home 3 (5th floor) are located near several iron pipe manufacturing foundries in the industrial community. Moreover, the occupants of the three homes are non-smokers. The average sampling time was 24 h and the field measurements were carried out at 1-1.5 m above ground level to simulate the breathing zone. Field samples were taken during the first 3 weeks in Home 1, the 4-6th week in Home 2, and the final 4 weeks in Home 3. Therefore, the field characterization of the particulate matter in the domestic environments, and a simultaneous comparison of the results with the outdoor measurements could be made in both residential and industrial communities. 3. R e s u l t s a n d d i s c u s s i o n
207
HOME I 120 I00
so g
60
•
40
INDOOR OUTDOOR
20 0 2
4
6
Observation
HOME 2 50O
t~
g
400 300 • INDOOR ------o--- O U T D O O R
200 100
0 0
2
4
6
Observation
HOME 3 400
300
i
• INDOOR ------o----- O U T D O O R
200 I00
The indoor and outdoor TSP, PM10 and PM2.5 mass levels, measured in the three domestic environments over the course of the monitoring period, are summarized graphically in Figs. 1, 2 and 3. The particulate levels in the different size ranges varied widely from day to day indoors and outdoors of the three monitored homes. The indoor TSP levels ranged from 20 to 70 /xg/m 3 for Home 1, from 60 to 350 /xg/m 3 for Home 2 and from 30 to 200 /xg/m 3 for Home 3, respectively. Compared with the indoor TSP levels, the corresponding outdoor levels observed were much higher and could reach as high as 450 /zg/m 3 in the outdoor air of Home 2. The particulate mass concentrations of Homes 2 and 3 in the industrial community were generally higher, in most cases, than those measured in Home 1 of the residential community. These observed higher particle mass concentrations are probably related to their sur-
0
---~ 0
I 2
t 4
t 6
i 8
Observation
Fig. 1. Mass concentration variations of indoor and outdoor TSP observed in the three homes.
rounding environments, e.g. some manufacturing foundries in the industrial community. For PM10, the indoor and outdoor levels were in the range of 20-300 /xg/m 3 and 40-350 /xg/m 3, respectively. In addition, the PM2.5 concentrations ranged from 20 to 150 txg/m 3 indoors and 10 to 200 /xg/m 3 outdoors, respectively. In brief, the indoor and outdoor particulate concentrations observed were higher, to some degree, than those measured in other investigations [7,8,1031]. Therefore, indoor and outdoor particles are a problem of major concerned, and steps to miti-
208
C.-S. Li / Sci. Total Environ. 151 (1994) 205-211 HOMEI
HOMEI
200
J
~--
~. IS0
150
o *
100
J |
200
INDOOR OUTDOOR
100
50
~ .
0 0
i
i
i
i
i
i
2
4
6
8
10
12
• iNDOOR ------o--- O U T D O O R
5o 0
,
0
i
2
,
i
,
4
i
6
i
i
8
Observation Observation
HOME2
HOME2
400
i
200
300
I$0
200
J
|
• -----o---
INDOOR 100
OUTDOOR
i
100
~ •
0 0
t
2
.
i
4
.
i
i
t
i
6
8
l0
12
• ---o---
INDOOR OUTDOOR
so
l 0 0
Observation
t
i
i
i
2
4
6
8
Observation
HOME3 ~..
~
HOME 3
400
~--
120
~
so
loo
300
~
200
E J
100
• ------o--
INDOOR OUTDOOR
60 i
0 0
t
i
I
i
t
i
2
4
6
8
10
12
Observation
Fig. 2. Mass concentration variations of indoor and outdoor PM10 observed in the three homes.
|
* ----o---
40
INDOOR OUTDOOR
2o 0 0
i 2
i 4
i 6
i 8
OBSERVATION
Fig. 3. Mass concentration variations of indoor and outdoor
PM2.5 observed in the three homes. gate situations with particulate levels above 150 p g / m 3 for 24 h need further investigations. The summary statistics for indoor and outdoor TSP, PM10 and PM2.5 mass levels are presented in Table 1. These data show that the highest indoor and outdoor mean (arithmetic) levels were observed in H o m e 2, but the lowest was found in Home 1. The nearby iron pipe foundries probably play an important role in these substantial concentration differences. Clearly, there were substantial fractions of the indoor (20%) and outdoor (40%) PM10 levels exceeding the U S E P A (United
States Environmental Protection Agency) 24 h standard for PM10 (150 / z g / m 3) in the three monitored homes. Additionally, a considerable number of the indoor (70%) and outdoor (85%) PM10 measurements over the 24 h California State standard of 5 0 / x g / m 3 were observed. These findings indicate that the percentages of the indoor and outdoor particulate levels over the particulate standard in Taipei are much higher compared with those found in other places [10,11]. The Wilcoxon Signed Rank test was used for paired data to determine whether indoor or out-
C.-S. Li / Sci. Total En~ron. 151 (1994) 205-211
door concentrations were found to be higher. In most samples, the outdoor levels were significantly higher than the corresponding indoor concentrations, especially for TSP and PM10. From the results of our field measurements, the indoor concentrations consistently followed but were generally lower than the corresponding outdoor ones. This indicates that the variation of indoor concentrations is driven by variation in ambient concentrations. Furthermore, an indicator of the average difference between indoor and outdoor TSP, PM10 and PM2.5 is the computation, as the indoor- to outdoor ratios, of the average or median concentrations, including I / O (mean) and I50,~/O50 % (median) ratios. The results indicated that I / O ratios were < 1, ranging from 0.52 to 0.60 for TSP, from 0.58 to 0.71 for PM10 and from 0.54 to 0.91 for PM2.5. Results with regard to the I50%/O50 % ratio were 0.51 for TSP, 0.61 for PM10 and 0.68 for PM2.5. These values indicate that ~ 40% decrease occurs in suspended particle mass concentration when going from outdoors to indoors. In addition, this decrease might represent the overall filtration efficiency of the homes investigated, and the modification of outdoor concentrations by the building envelope. The
209
Indoor = -9.05 + 0.62 (Outdoor) for TSP, R ~ = 93.1% 400 [
Indoor = 1.87 + 0.65 (Outdoor) for PM 10, R 2 = 78.6%
F
Indoor = 29.66 + 0.31 (Outdoor) for PM25, R 2 = 30.7%
• TOTAL
Z
o
300 t E
.i
/
o PMIO
•
/
////
PM~.s
o
•
•
o:// > ;.o:°:
200
100
o
o
100
200
.
•
.
.
.
300
.
.
.
400
Outdoor Concentrations (ug/m3)
Fig. 4. S c a t t e r p l o t of p a i r e d i n d o o r vs. o u t d o o r TSP, PM10 a n d PM2.5 mass c o n c e n t r a t i o n s .
Table 1 C o m p a r i s o n of i n d o o r to o u t d o o r m a s s c o n c e n t r a t i o n s ( t x g / m 3) of a i r b o r n e particles Indoor na Home 1 TSP PM10 PM2.5 Home 2 TSP P M 10 PM2.5 Home 3 TSP PM10 PM2.5
7 11 7
Outdoor Mean
37.11 46.88 36.65
S.D. b
n
Mean
19.36 20.82 12.90
7 11 7
71.52 81.20 51.23
T e s t value c
I / O ratio
- 6.85 d - 3.24 a - 0.68
0.52 0.58 0.72
S.D.
29.99 37.74 52.51
6 9 6
163.1 172.1 83.7
100.60 57.47 39.2
6 9 6
271.8 240.9 155.0
144.4 78.4 36.1
-4.13 ~ - 3.78 d - 4.28 ~
0.60 0.71 0.54
5 9 6
90.2 99.6 53.0
70.9 67.1 33.4
5 9 6
171.1 156.7 58.0
124.2 99.4 21.7
- 3.33 c - 3.31 e - 0.48
0.53 0.64 0.91
an = number. bS.D. = s t a n d a r d deviation. c W i l c o x o n signed r a n k test. d p < 0.05. e p < 0.01.
500
C.-S. Li /Sci. Total Ent~ron. 151 (1994) 205-211
210
TSP = -3.10 + 1.17 (PMI0) f o r I n d o o r , R 2 = 99.7%
400
P M I 0 = 16.83+ i.26 (PM2.$) f o r I n d o o r , R 2 = 75.3%
TSP = -14.27+ 1.28 (PM 10) f o r O u t d o o r , R 2 = 99.5%
P M I 0 = 44.67+ 1.05 (PM2.5) for O u t d o o r , R 2 = 83.4%
500 •
INDOOR
o//~ •
300 400
INDOOR
o OUTDOOR
/S
o
300
.=
200
r.)
E ¢
o
o •
•
o o • e
200
100
eL
o
100 i
i
0
0
~ _
0
i
100
PMI0 Mass
i
i
i
200
300
400
Concentrations
(ug/m3)
Fig. 5. S c a t t e r p l o t o f p a i r e d T S P vs. P M 1 0 m a s s c o n c e n t r a tions indoors and outdoors.
value also demonstrates the effects of indoor sources on indoor air quality. It was demonstrated that indoor particle sources were not significantly apparent in our samples, compared with outdoor particulate sources, such as heavy traffic and manufacturing emissions. Our results are consistent with other findings that suggest an I / O ratio < 1 is a reasonable hypothesis. These I / O ratio results also support the lack of coarse particle penetration indoors, as there was < 70% penetration of the outdoor particles' to the indoor environments. A number of field investigations indicate that ~ 70% of the respirable fraction will penetrate indoors [9,18]. The individual indoor and outdoor pairs of TSP, PM10 and PM2.5 are presented in a scatterplot as shown in Fig. 4. From the regression, the results indicate that indoor TSP and PM10 con-
50
PM2.5
Mass
i
i
i
100
150
200
Concentrations
(ng/m3)
Fig. 6. S c a t t e r p l o t o f p a i r e d P M 1 0 vs. P M 2 . 5 m a s s c o n c e n trations indoors and outdoors.
centrations are well correlated with their corresponding outdoor levels; the outdoor concentrations accounting for between 78 and 93% of the variation of indoor concentrations. Regarding indoor and outdoor PM2.5, the indoor levels were less correlated with the corresponding outdoor concentrations. The outdoor concentrations explain only 30% of the variation of the indoor ones which may be related to indoor particulate sources, and the modification of outdoor concentrations by the apartment envelope. The paired TSP and PM10 concentrations indoors and outdoors were highly correlated as shown in Fig. 5, with R 2 = 0.99. Regarding the paired PM10 and PM2.5 levels (Fig. 6), good correlations were observed with R 2 = 0.75 for indoors and 0.83 for outdoors. These results indicate that PM2.5 mass concentrations account for between 75 and 83% of the variation of PM10 concentrations. This high proportional mass explains why there is a good correlation between
C.-S. Li / Sci. Total Endron. 151 (1994) 205-211
PM10 and PM2.5 mass. Although these observations only imply for the sampling periods and locations of our investigation, PM2.5 is a good predicator of PM10 mass. In summary, the approach undertaken in this field investigation should be extended to a large sample of residences for the purpose of constructing a total exposure model and performing a risk assessment for PM10 and PM2.5.
Acknowledgements
8
9
10
11
This work is supported by Taiwan National Science Council under Grant NSC 81-0421-F-002541-Z. 12
References 1 A. Szalai, The use of time: daily activities of urban and suburban populations in twelve countries, Moughton, The Hague, Paris, 1972. 2 F.S. Chapin, Human activity patterns in the city, Wiley, New York, 1974. 3 J.A. Wiley, P.J. Robinson, T. Piazza, K. Garrett, K. Cirksena, Y. Cheng and G. Martin, Activity patterns of California residents, Final rep. for the Research Division, Califronia Air Resources Board, contract no. A 6-177-33, Sacramento, CA, 1991. 4 National Research Council, Indoor Pollutants, Committee on Indoor Pollutants, National Academy Press, Washington, DC, 1981. 5 National Research Council, Human Exposure Assessment for Airborne Pollutants: Advances and Opportunities, National Academy Press, Washington, DC, 1991. 6 K. Sexton, J.D. Spengler and R.D. Treitman, Personal exposure to respirable particles: a case study in Waterbury, Vermont, Atmos. Environ., 18 (1984) 1385-1398. 7 Y.S. Kim and T.H. Stock, House-specific characterization
13
14
15
16
17
18
211
of indoor and outdoor aerosols, Environ. Int., 12 (1986) 75 -92. T. Raunemaa, M. Kulmala, H. Saari, M. Olin and M.H. Kulmala, Indoor air aerosol model: transport indoors and deposition of fine and coarse particles, Aerosol Sci. Technol., 11 (1989) 11-25. D.W. Dockery and J.D. Spengler, Indoor-outdoor relationships of respirable sulfates and particles, Atmos. Environ., 15 (1981) 335-343. P.J. Lioy, J.M. Waldman, T. Buckley, J. Butler and C. Pietarinen, The personal, indoor and outdoor concentrations of PM-10 measured in an industrial community during the winter, Atmos. Environ., 24B (1990) 57-66. S.D. Colome, N.Y. Kado, P. Jaques and M. Kleinman, Indoor-outdoor air pollution relations: particulate matter less than 10 /zm in aerodynamic diameter (PM10) in homes of asthmatics, Atmos. Environ., 26A (1992) 2173-2178. D.J. Moschandreas, J.W. Winchster, J.W. Nelson and R.M. Buton, Fine particle residential indoor air pollution, Atmos. Environ., 13 (1979) 1413-1418. C.R. Thompson, E.G. Hensel and G. Kats, Outdoor-indoor levels of six air pollutants. J. Air Pollut. Control Assoc., 23 (1973) 881-886. J.J. Quackenboss, M.D. Lebowitz and C.D. Crutchfield, Indoor-outdoor relationships for particulate matter: exposure classifications and health effects, Environ. Int., 15 (1989) 353-360. V.R. Highsmith, R.B. Zweidinger and R.G. Merrill, Characterization of indoor and outdoor air associated with residences using woodstoves: a pilot study, Environ. Int., 14 (1988) 213-219. J.M. Daisey, J.D. Spengler and P. Kaarakka, A comparison of the organic chemical composition of indoor aerosols during woodburning and non-woodburning periods, Environ. Int., 15 (1989) 435-442. V.A. Marple, K.L. Rubow, W. Turner and J.D. Spengler, Low flow rate sharp cut impactors for indoor air sampling: design and calibration, J. Air Pollut. Control Assoc., 37 (1987) 1303-1307. J.E. Yocom, Indoor-outdoor air quality relationship, J. Air Pollut. Control Assoc., 32 (1982) 500-52[).
The Science of the Total Environment 151 (1994) 205-211
Relationships of indoor/outdoor inhalable and respirable particles in domestic environments C h i h - S h a n Li* DiLision of Ent~ironmental Health Science, Institute of Public Health, College of Medicine, National Taiwan Unitersity, No.1 Jen Al Road, 1st Sect., Taipei, Taiwan, ROC Received 4 May 1993; accepted 6 July 1993
Abstract
Simultaneous air monitoring inside and outside of three domestic environments in residential and industrial communities were conducted to evaluate infiltrations of outdoor aerosols into indoor environments, and indoor aerosol characteristics. The relationships of total (TSP), inhalable (< 10 ~m, PM10) and respirable ( < 2.5 /xm, PM2.5) suspended particles were compared. It was found that the fractions of the indoor (20%) and outdoor (40%) PM10 levels exceeding 150 /xg/m 3 were substantial. In addition, it was indicated that indoor TSP and PM10 concentrations (R z = 0.99) were well correlated with their corresponding outdoor levels. Moreover, there was a close relationship between the paired PM10 and PM2.5 levels indoors (R 2= 0.75) and outdoors (R z = 0.83). The average indoor/outdoor ( I / O ) and median I / O ratios observed were 0.60. Therefore, steps to mitigate the situations with particulate levels above ambient particulate standards requires further study.
Keywords: Indoor air quality; Residental environment; pH 10; pH 2.5
1. Introduction Indoor air quality has become an important issue in recent years, since activity investigations indicate that people spend 80-90% of their time indoors [1-3]. In addition, indoor air quality could deteriorate through contributions of outdoor particles and the presence of indoor particle sources [4]. Therefore, it is necessary to measure indoor particle levels for the purpose of characterizing total human exposure to airborne particulate matter [5]. There have been a number of studies that
* Corresponding author.
evaluate correlations between indoor and outdoor particles in different domestic environments. In 24 residences in Vermont, USA, the mean mass levels of indoor and outdoor respirable particles during 14 sampling days in the winter were found to be 25 / z g / m 3 and 1 7 / x g / m 3, respectively [6]. Additionally, the average indoor levels exceeded outdoor levels on 12 out of the 14 sampling days. The indoor fine particle concentrations varied from 6 to 40 / z g / m 3, with mean indoor/outdoor ( I / O ) ratios ranging from 0.7 to 2.1 in 11 Houston (USA) homes [7]. Regarding fine and coarse particles in urban and suburban Helsinki, Finland, the geometric mean mass concentrations were found to be 16 /xg/m 3 and 13 /xg/m 3, respectively [8]. The I / O ratios were < 1 in most
0048-9697/94/$07.00 © 1994 Elsevier Science BV. All rights reserved.
SSDI 0048-9697(94)04014-Z
206
C-S. Li / Sci. Total Entiron. 151 (1994) 205-211
situations, however, they were > 1 for elements of fine particles originating from indoor sources [9]. Lioy et al. [10] investigated the indoor and outdoor relationships of PM10 in an industrial community during the winter in Phillipsburg, N J, USA. Here, indoor concentrations in the measured homes were related more to outdoor particulate sources. Another similar investigation was performed in 10 homes in southern California, USA [11], where the observed I / O median ratio was ~ 0.70. Furthermore, the percentages of the indoor and outdoor PM10 observations greater than the California standard (50 /zg/m 3) were 7% and 29%, respectively. The indoor particles were influenced by not only the infiltration of outdoor air, but also indoor particle generation [12]. It was estimated that the mean infiltration of outdoor fine particles was ~ 70% [9]. In addition, the most important single determinant of indoor particle sources was found to be cigarette smoking [7,9]. In some cases, it was also reported that indoor particle levels were highly affected by outdoor traffic [13]. From a field survey in 98 homes in Tucson, Arizona, USA, the mean mass concentrations of PM2.5 and PM10 were observed to be 20 and 4 0 / z g / m 3 for non-smoking, 35 and 55 /~g/m 3 for 1-20 cigarettes/day, and 85 and 100 /zg/m 3 for > 20 cigarettes/day, respectively [14]. Moreover, the I / O ratio of PM2.5 and PM10 was found to be 0.63 for non-smoking homes and 1.1 for those with smoking. In three woodstove homes in Raleigh, NC, USA, the highest I / O fine particle ratio was observed in the home with a leaking woodstove [15]. Another related comparison of indoor and outdoor respirable suspended particles (RSP) was conducted in seven houses during woodburning periods. The I / O ratios of RSP were generally > 1, because of woodburning and other unidentified sources [16]. In brief, the correlations between indoor and outdoor particles are quite different from area to area, because of different characteristics of I / O sources as well as particle infiltration into indoors. Taiwan is a subtropical country with a warm climate throughout the year. The influences of ambient atmosphere on indoor air quality should be different from those observed in colder areas,
because of direct transport into indoors through open doors and windows. To date there have been few investigations regarding specific compounds and compound classes indoors, as well as the relationship of outdoor to indoor concentrations of PM2.5 and PM10 in Taiwan. The purpose of these field measurements was to characterize the particulate matter in the domestic environments of residential and industrial communities in Taiwan (a subtropical country with warm weather all year round) and to compare simultaneously the indoor results with the corresponding outdoor measurements. 2. Materials and methods
2.1. Equipment Open sampling cassettes, without particle-size fractionations, were equipped with personal SKC sampling pumps for collecting TSP at a 3.5-1/min flowrate. PM10 and PM2.5 environmental monitor samplers (model 200, MSP Co., Minneapolis, MN) were used for collection of inhalable and respirable particles at a flowrate of 10 l/min [17]. Both PM10 and PM2.5 samplers had twin impaction plates, providing sharp 10 and 2.5 p,m particle cut sizes, respectively. Plastic filter holders were fitted with a Teflon membrane filter (Teflon; Gelman Filter Co., Ann Arbor, MI) and the flowrates were calibrated in the field with a bulb meter (Gilian Instrument Corp., West Caldwell, N J). Simultaneous field measurements were performed indoors and outdoors with a portable high volume air sampler (Gilian, AirCon-2 with timer and controller to permit sequential collection by PM10 and PM2.5 samplers, as well as by PM10 and a cassette using the same pump). Teflon filters (2/zm pore size, 37 mm diameter) for TSP, PM10 and PM2.5 samplers were placed into a 5-cm petric dish for handling. Particulate filters were pre- and post-weighed following the standardized operating procedure for balance calibration, filter conditioning and filter weighing. Quality control checks included limit checks for flow rates and filter weights as well as range checks of concentrations in relation to sample type and location.
C.-S. Li /Sci. Total En tiron. 151 (1994) 205-211
2.2. Sampling In Taiwan, the homes are located in either residential or industrial communities. To evaluate the concentration differences in TSP, PM10 and PM2.5 particulate matter in these two different areas, the simultaneous indoor and outdoor field samplings of three residences were conducted during a 10-week period, from November 1992 to February 1993, with multiple measurements. Home 1 is a lst-floor apartment located in the residential area. Home 2 (lst floor) and Home 3 (5th floor) are located near several iron pipe manufacturing foundries in the industrial community. Moreover, the occupants of the three homes are non-smokers. The average sampling time was 24 h and the field measurements were carried out at 1-1.5 m above ground level to simulate the breathing zone. Field samples were taken during the first 3 weeks in Home 1, the 4-6th week in Home 2, and the final 4 weeks in Home 3. Therefore, the field characterization of the particulate matter in the domestic environments, and a simultaneous comparison of the results with the outdoor measurements could be made in both residential and industrial communities. 3. R e s u l t s a n d d i s c u s s i o n
207
HOME I 120 I00
so g
60
•
40
INDOOR OUTDOOR
20 0 2
4
6
Observation
HOME 2 50O
t~
g
400 300 • INDOOR ------o--- O U T D O O R
200 100
0 0
2
4
6
Observation
HOME 3 400
300
i
• INDOOR ------o----- O U T D O O R
200 I00
The indoor and outdoor TSP, PM10 and PM2.5 mass levels, measured in the three domestic environments over the course of the monitoring period, are summarized graphically in Figs. 1, 2 and 3. The particulate levels in the different size ranges varied widely from day to day indoors and outdoors of the three monitored homes. The indoor TSP levels ranged from 20 to 70 /xg/m 3 for Home 1, from 60 to 350 /xg/m 3 for Home 2 and from 30 to 200 /xg/m 3 for Home 3, respectively. Compared with the indoor TSP levels, the corresponding outdoor levels observed were much higher and could reach as high as 450 /zg/m 3 in the outdoor air of Home 2. The particulate mass concentrations of Homes 2 and 3 in the industrial community were generally higher, in most cases, than those measured in Home 1 of the residential community. These observed higher particle mass concentrations are probably related to their sur-
0
---~ 0
I 2
t 4
t 6
i 8
Observation
Fig. 1. Mass concentration variations of indoor and outdoor TSP observed in the three homes.
rounding environments, e.g. some manufacturing foundries in the industrial community. For PM10, the indoor and outdoor levels were in the range of 20-300 /xg/m 3 and 40-350 /xg/m 3, respectively. In addition, the PM2.5 concentrations ranged from 20 to 150 txg/m 3 indoors and 10 to 200 /xg/m 3 outdoors, respectively. In brief, the indoor and outdoor particulate concentrations observed were higher, to some degree, than those measured in other investigations [7,8,1031]. Therefore, indoor and outdoor particles are a problem of major concerned, and steps to miti-
208
C.-S. Li / Sci. Total Environ. 151 (1994) 205-211 HOMEI
HOMEI
200
J
~--
~. IS0
150
o *
100
J |
200
INDOOR OUTDOOR
100
50
~ .
0 0
i
i
i
i
i
i
2
4
6
8
10
12
• iNDOOR ------o--- O U T D O O R
5o 0
,
0
i
2
,
i
,
4
i
6
i
i
8
Observation Observation
HOME2
HOME2
400
i
200
300
I$0
200
J
|
• -----o---
INDOOR 100
OUTDOOR
i
100
~ •
0 0
t
2
.
i
4
.
i
i
t
i
6
8
l0
12
• ---o---
INDOOR OUTDOOR
so
l 0 0
Observation
t
i
i
i
2
4
6
8
Observation
HOME3 ~..
~
HOME 3
400
~--
120
~
so
loo
300
~
200
E J
100
• ------o--
INDOOR OUTDOOR
60 i
0 0
t
i
I
i
t
i
2
4
6
8
10
12
Observation
Fig. 2. Mass concentration variations of indoor and outdoor PM10 observed in the three homes.
|
* ----o---
40
INDOOR OUTDOOR
2o 0 0
i 2
i 4
i 6
i 8
OBSERVATION
Fig. 3. Mass concentration variations of indoor and outdoor
PM2.5 observed in the three homes. gate situations with particulate levels above 150 p g / m 3 for 24 h need further investigations. The summary statistics for indoor and outdoor TSP, PM10 and PM2.5 mass levels are presented in Table 1. These data show that the highest indoor and outdoor mean (arithmetic) levels were observed in H o m e 2, but the lowest was found in Home 1. The nearby iron pipe foundries probably play an important role in these substantial concentration differences. Clearly, there were substantial fractions of the indoor (20%) and outdoor (40%) PM10 levels exceeding the U S E P A (United
States Environmental Protection Agency) 24 h standard for PM10 (150 / z g / m 3) in the three monitored homes. Additionally, a considerable number of the indoor (70%) and outdoor (85%) PM10 measurements over the 24 h California State standard of 5 0 / x g / m 3 were observed. These findings indicate that the percentages of the indoor and outdoor particulate levels over the particulate standard in Taipei are much higher compared with those found in other places [10,11]. The Wilcoxon Signed Rank test was used for paired data to determine whether indoor or out-
C.-S. Li / Sci. Total En~ron. 151 (1994) 205-211
door concentrations were found to be higher. In most samples, the outdoor levels were significantly higher than the corresponding indoor concentrations, especially for TSP and PM10. From the results of our field measurements, the indoor concentrations consistently followed but were generally lower than the corresponding outdoor ones. This indicates that the variation of indoor concentrations is driven by variation in ambient concentrations. Furthermore, an indicator of the average difference between indoor and outdoor TSP, PM10 and PM2.5 is the computation, as the indoor- to outdoor ratios, of the average or median concentrations, including I / O (mean) and I50,~/O50 % (median) ratios. The results indicated that I / O ratios were < 1, ranging from 0.52 to 0.60 for TSP, from 0.58 to 0.71 for PM10 and from 0.54 to 0.91 for PM2.5. Results with regard to the I50%/O50 % ratio were 0.51 for TSP, 0.61 for PM10 and 0.68 for PM2.5. These values indicate that ~ 40% decrease occurs in suspended particle mass concentration when going from outdoors to indoors. In addition, this decrease might represent the overall filtration efficiency of the homes investigated, and the modification of outdoor concentrations by the building envelope. The
209
Indoor = -9.05 + 0.62 (Outdoor) for TSP, R ~ = 93.1% 400 [
Indoor = 1.87 + 0.65 (Outdoor) for PM 10, R 2 = 78.6%
F
Indoor = 29.66 + 0.31 (Outdoor) for PM25, R 2 = 30.7%
• TOTAL
Z
o
300 t E
.i
/
o PMIO
•
/
////
PM~.s
o
•
•
o:// > ;.o:°:
200
100
o
o
100
200
.
•
.
.
.
300
.
.
.
400
Outdoor Concentrations (ug/m3)
Fig. 4. S c a t t e r p l o t of p a i r e d i n d o o r vs. o u t d o o r TSP, PM10 a n d PM2.5 mass c o n c e n t r a t i o n s .
Table 1 C o m p a r i s o n of i n d o o r to o u t d o o r m a s s c o n c e n t r a t i o n s ( t x g / m 3) of a i r b o r n e particles Indoor na Home 1 TSP PM10 PM2.5 Home 2 TSP P M 10 PM2.5 Home 3 TSP PM10 PM2.5
7 11 7
Outdoor Mean
37.11 46.88 36.65
S.D. b
n
Mean
19.36 20.82 12.90
7 11 7
71.52 81.20 51.23
T e s t value c
I / O ratio
- 6.85 d - 3.24 a - 0.68
0.52 0.58 0.72
S.D.
29.99 37.74 52.51
6 9 6
163.1 172.1 83.7
100.60 57.47 39.2
6 9 6
271.8 240.9 155.0
144.4 78.4 36.1
-4.13 ~ - 3.78 d - 4.28 ~
0.60 0.71 0.54
5 9 6
90.2 99.6 53.0
70.9 67.1 33.4
5 9 6
171.1 156.7 58.0
124.2 99.4 21.7
- 3.33 c - 3.31 e - 0.48
0.53 0.64 0.91
an = number. bS.D. = s t a n d a r d deviation. c W i l c o x o n signed r a n k test. d p < 0.05. e p < 0.01.
500
C.-S. Li /Sci. Total Ent~ron. 151 (1994) 205-211
210
TSP = -3.10 + 1.17 (PMI0) f o r I n d o o r , R 2 = 99.7%
400
P M I 0 = 16.83+ i.26 (PM2.$) f o r I n d o o r , R 2 = 75.3%
TSP = -14.27+ 1.28 (PM 10) f o r O u t d o o r , R 2 = 99.5%
P M I 0 = 44.67+ 1.05 (PM2.5) for O u t d o o r , R 2 = 83.4%
500 •
INDOOR
o//~ •
300 400
INDOOR
o OUTDOOR
/S
o
300
.=
200
r.)
E ¢
o
o •
•
o o • e
200
100
eL
o
100 i
i
0
0
~ _
0
i
100
PMI0 Mass
i
i
i
200
300
400
Concentrations
(ug/m3)
Fig. 5. S c a t t e r p l o t o f p a i r e d T S P vs. P M 1 0 m a s s c o n c e n t r a tions indoors and outdoors.
value also demonstrates the effects of indoor sources on indoor air quality. It was demonstrated that indoor particle sources were not significantly apparent in our samples, compared with outdoor particulate sources, such as heavy traffic and manufacturing emissions. Our results are consistent with other findings that suggest an I / O ratio < 1 is a reasonable hypothesis. These I / O ratio results also support the lack of coarse particle penetration indoors, as there was < 70% penetration of the outdoor particles' to the indoor environments. A number of field investigations indicate that ~ 70% of the respirable fraction will penetrate indoors [9,18]. The individual indoor and outdoor pairs of TSP, PM10 and PM2.5 are presented in a scatterplot as shown in Fig. 4. From the regression, the results indicate that indoor TSP and PM10 con-
50
PM2.5
Mass
i
i
i
100
150
200
Concentrations
(ng/m3)
Fig. 6. S c a t t e r p l o t o f p a i r e d P M 1 0 vs. P M 2 . 5 m a s s c o n c e n trations indoors and outdoors.
centrations are well correlated with their corresponding outdoor levels; the outdoor concentrations accounting for between 78 and 93% of the variation of indoor concentrations. Regarding indoor and outdoor PM2.5, the indoor levels were less correlated with the corresponding outdoor concentrations. The outdoor concentrations explain only 30% of the variation of the indoor ones which may be related to indoor particulate sources, and the modification of outdoor concentrations by the apartment envelope. The paired TSP and PM10 concentrations indoors and outdoors were highly correlated as shown in Fig. 5, with R 2 = 0.99. Regarding the paired PM10 and PM2.5 levels (Fig. 6), good correlations were observed with R 2 = 0.75 for indoors and 0.83 for outdoors. These results indicate that PM2.5 mass concentrations account for between 75 and 83% of the variation of PM10 concentrations. This high proportional mass explains why there is a good correlation between
C.-S. Li / Sci. Total Endron. 151 (1994) 205-211
PM10 and PM2.5 mass. Although these observations only imply for the sampling periods and locations of our investigation, PM2.5 is a good predicator of PM10 mass. In summary, the approach undertaken in this field investigation should be extended to a large sample of residences for the purpose of constructing a total exposure model and performing a risk assessment for PM10 and PM2.5.
Acknowledgements
8
9
10
11
This work is supported by Taiwan National Science Council under Grant NSC 81-0421-F-002541-Z. 12
References 1 A. Szalai, The use of time: daily activities of urban and suburban populations in twelve countries, Moughton, The Hague, Paris, 1972. 2 F.S. Chapin, Human activity patterns in the city, Wiley, New York, 1974. 3 J.A. Wiley, P.J. Robinson, T. Piazza, K. Garrett, K. Cirksena, Y. Cheng and G. Martin, Activity patterns of California residents, Final rep. for the Research Division, Califronia Air Resources Board, contract no. A 6-177-33, Sacramento, CA, 1991. 4 National Research Council, Indoor Pollutants, Committee on Indoor Pollutants, National Academy Press, Washington, DC, 1981. 5 National Research Council, Human Exposure Assessment for Airborne Pollutants: Advances and Opportunities, National Academy Press, Washington, DC, 1991. 6 K. Sexton, J.D. Spengler and R.D. Treitman, Personal exposure to respirable particles: a case study in Waterbury, Vermont, Atmos. Environ., 18 (1984) 1385-1398. 7 Y.S. Kim and T.H. Stock, House-specific characterization
13
14
15
16
17
18
211
of indoor and outdoor aerosols, Environ. Int., 12 (1986) 75 -92. T. Raunemaa, M. Kulmala, H. Saari, M. Olin and M.H. Kulmala, Indoor air aerosol model: transport indoors and deposition of fine and coarse particles, Aerosol Sci. Technol., 11 (1989) 11-25. D.W. Dockery and J.D. Spengler, Indoor-outdoor relationships of respirable sulfates and particles, Atmos. Environ., 15 (1981) 335-343. P.J. Lioy, J.M. Waldman, T. Buckley, J. Butler and C. Pietarinen, The personal, indoor and outdoor concentrations of PM-10 measured in an industrial community during the winter, Atmos. Environ., 24B (1990) 57-66. S.D. Colome, N.Y. Kado, P. Jaques and M. Kleinman, Indoor-outdoor air pollution relations: particulate matter less than 10 /zm in aerodynamic diameter (PM10) in homes of asthmatics, Atmos. Environ., 26A (1992) 2173-2178. D.J. Moschandreas, J.W. Winchster, J.W. Nelson and R.M. Buton, Fine particle residential indoor air pollution, Atmos. Environ., 13 (1979) 1413-1418. C.R. Thompson, E.G. Hensel and G. Kats, Outdoor-indoor levels of six air pollutants. J. Air Pollut. Control Assoc., 23 (1973) 881-886. J.J. Quackenboss, M.D. Lebowitz and C.D. Crutchfield, Indoor-outdoor relationships for particulate matter: exposure classifications and health effects, Environ. Int., 15 (1989) 353-360. V.R. Highsmith, R.B. Zweidinger and R.G. Merrill, Characterization of indoor and outdoor air associated with residences using woodstoves: a pilot study, Environ. Int., 14 (1988) 213-219. J.M. Daisey, J.D. Spengler and P. Kaarakka, A comparison of the organic chemical composition of indoor aerosols during woodburning and non-woodburning periods, Environ. Int., 15 (1989) 435-442. V.A. Marple, K.L. Rubow, W. Turner and J.D. Spengler, Low flow rate sharp cut impactors for indoor air sampling: design and calibration, J. Air Pollut. Control Assoc., 37 (1987) 1303-1307. J.E. Yocom, Indoor-outdoor air quality relationship, J. Air Pollut. Control Assoc., 32 (1982) 500-52[).