Environ. Sci. Technol. 2005, 39, 1707-1714
Use of Personal-Indoor-Outdoor Sulfur Concentrations to Estimate the Infiltration Factor and Outdoor Exposure Factor for Individual Homes and Persons LANCE WALLACE* AND RON WILLIAMS National Exposure Research Laboratory, U.S. EPA, Research Triangle Park, North Carolina
A study of personal, indoor, and outdoor exposure to PM2.5 and associated elements has been carried out for 37 residents of the Research Triangle Park area in North Carolina. Participants were selected from persons expected to be at elevated risk from exposure to particles, and included 29 persons with hypertension and 8 cardiac patients with implanted defibrillators. Participants were monitored for 7 consecutive days in each of four seasons. One goal of the study was to estimate the contribution of outdoor PM2.5 to indoor concentrations. This depends on the infiltration factor Finf, the fraction of outdoor PM2.5 remaining airborne after penetrating indoors. After confirming with our measurements the findings of previous studies that sulfur has few indoor sources, we estimated an average Finf for each house based on indoor/outdoor sulfur ratios. These estimates ranged from 0.26 to 0.87, with a median value of 0.55. Since these estimates apply only to particles of size similar to that of sulfur particles (0.06-0.5 µm diameter), and since larger particles (0.5-2.5 µm) have lower penetration rates and higher deposition rates, these estimates are likely to be higher than the true infiltration factors for PM2.5 as a whole. In summer when air conditioners were in use, the sulfur-based infiltration factor was at its lowest (averaging 0.50) for most homes, whereas the average Finf for the other three seasons was 0.62-0.63. Using the daily estimated infiltration factor for each house, we calculated the contribution of outdoor PM2.5 to indoor air concentrations. The indoor-generated contributions to indoor PM2.5 had a wider range (0-33 µg/m3) than the outdoor contributions (5-22 µg/m3). However, outdoor contributions exceeded the indoor-generated contributions in 27 of 36 homes. A second goal of the study was to determine the contribution of outdoor particles to personal exposure. This is determined by the “outdoor exposure factor” Fpex, the fraction of outdoor PM2.5 contributing to personal exposure. As with Finf, we estimated Fpex by the personal/outdoor sulfur ratios. The estimates ranged from 0.33 to 0.77 with a median value of 0.53. Outdoor air particles were less important for personal exposures than for indoor concentrations, with the median outdoor contribution to personal exposure just 49%. We regressed the outdoor contributions to personal exposures on measured outdoor PM2.5 at the central site. The regressions had R2 values ranging from 0.19 to * Corresponding author phone: (703)620-4543; fax: (703)860-0678; e-mail:
[email protected]. 10.1021/es049547u Not subject to U.S. Copyright. Publ. 2005 Am. Chem. Soc. Published on Web 01/26/2005
0.88 (median ) 0.73). These values provide an indication of the extent of misclassification error in epidemiological estimates of the effect of outdoor particles on health.
Introduction Many studies worldwide in the past decade have documented an association between health effects and particle concentrations measured at central monitoring sites (1). Since the health effects are presumably related to personal exposures, an important research need (identified by the National Academy of Sciences in 1995), is to determine how personal exposures correlate with these outdoor concentrations (NRC-NAS 1998). A number of studies (2-31) have been undertaken to measure personal exposure directly using personal monitors, and the correlations of personal exposure with outdoor concentrations are straightforward to determine (32). However, the correlation that interests many epidemiologists is not that between total personal exposure and outdoor concentrations, but the correlation between that component of personal exposure due to outdoor particles and the outdoor concentrations. This requires the ability to estimate the contribution to personal exposures from particles originating outdoors. Only a few studies have reported making this attempt (33-35). The goal of this report is to estimate the contribution of outdoor particles to personal exposure for a group of 37 persons monitored one week per season over four seasons in 2000-2001. The correlation between this outdoor-generated personal exposure and PM2.5 outdoor concentrations will then be calculated for each person. Since persons spend on average 89% of their time indoors, the contribution of outdoor particles to indoor concentrations will also be explored. For many people, the indoor-outdoor relationship may be the major determinant of the personaloutdoor relationship.
Study Methods and Database In this study, 37 persons in 36 homes in the Raleigh, NC area were monitored for four seasons, 7 days per season. Previous publications have described the study design, methods and PM2.5 results (30, 31). Briefly, two cohorts were selected: 29 African-American persons with hypertension and 8 persons with implanted cardiac defibrillators. Since the selection was made in a nonprobabilistic manner, the findings in this study may not be extrapolated to any larger population. The study was focused on the exposures of high-risk subpopulations, whose activities may not be similar to those of the population at large. Two gravimetric PM2.5 monitors were employed at the homes: the Harvard impactor (HI), operating at 20 Lpm, and a personal exposure monitor (PEM), operating at 2 Lpm for a nominal 24-h period. The PEM was used for personal samples and the HI was used for indoor-outdoor samples. All of the PEM and HI filters were analyzed for a suite of elements using X-ray fluorescence. Two PM2.5 monitors were also employed at the central monitoring site: an HI and a federal reference method (FRM) monitor. Estimating the Contribution of Outdoor Air Particles to Indoor Concentrations. The contribution of outdoor particles to indoor concentrations is described by the mass balance equation. The full mass balance equation includes such phenomena as coagulation, condensation, and gasto-particle conversion, etc. (36). We will consider here a simplified version involving only infiltration, exfiltration, deposition, and indoor sources. The differential form of this VOL. 39, NO. 6, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1707
simplified mass balance equation is
dCin/dt ) PaCout - (a + k) Cin + S/V
(1)
where Cin ) indoor number or mass concentration (cm-3 or (µg/m)3), Cout ) outdoor number or mass concentration, P ) penetration coefficient across building envelope, a ) air exchange rate (h-1), V ) volume of building (cm3 or m3), S ) source strength (h-1 or µg h-1), and k ) total decay rate of particles (h-1). The equation is considered to be applicable to all particle sizes, with all terms except air exchange rate and building volume considered to be functions of particle size. The assumption is made that the entire house is a single wellmixed zone, with instantaneous mixing of particles throughout the house, and that the measured air exchange rate in one room applies to the entire house. We assume that the averaging time (24 h) over which the equation is to be evaluated is sufficiently long that transient terms due to shortterm changes in the outdoor concentration are negligible compared to the long-term average concentrations. Under these assumptions, the time-averaged solution to the mass balance equation is
Cin ) [Pa/(a + k)] Cout + S/[V(a + k)]
(2)
The coefficient of the outdoor concentration is sometimes called the infiltration factor Finf:
Finf ) Pa/(a + k)
(3)
The infiltration factor is a major variable determining the indoor-outdoor, and to a large degree the personal-outdoor, relationship. This is expected to vary by household and resident characteristics. For example, a tightly built house may have a lower penetration coefficient than a drafty house. A house with a large surface/volume ratio (e.g., many carpets, rugs, or fibrous wall hangings) may have higher deposition rates (37). Use of fans or filters may also increase particle deposition rates (38-40). Open windows will increase Finf by increasing the air exchange rate (41, 42) and possibly by redirecting infiltrating particles through the open window (P ) 1) rather than through the rest of the building envelope (P < 1) (43-45). Use of air conditioning has been shown to lower the infiltration factor, either because of reducing air exchange rates by shutting windows or increasing deposition rates by recirculation of indoor air through ductwork (21, 38, 46-48). Despite these clear indications that exposure to outdoor air particles depends heavily on household and behavioral characteristics, studies capable of estimating the infiltration factor reliably for individual homes are rather few (34). In this study we attempt to estimate Finf for individual homes, together with an estimate of the uncertainties involved. Several investigators have noted that sulfur has few indoor sources (23, 35). If that is the case, the source term in the solution above (eq 2) may be ignored, and the equation takes the very simple form
Sin/Sout ) Finf
(4)
where Sin and Sout are the sulfur concentrations indoors and outdoors. Indoor-outdoor comparisons of sulfur concentrations thus provide a direct way to estimate Finf for each individual home. Strictly speaking, the sulfur data provides only information for particles with behavior similar to that of sulfur with respect to penetration and deposition. Sulfur particles are somewhat smaller than most other fine particles (0.060.5 µm) (35). Since the particles in the 0.5-2.5 µm range may 1708
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 6, 2005
have higher deposition velocities than sulfur (36, 37) and may also have lower penetration rates through the building envelope (43), the infiltration factors calculated using sulfur ratios may be somewhat higher than the infiltration factors for PM2.5 as a whole. Note that this method does not depend on the actual fraction of PM2.5 accounted for by sulfur, nor does it depend on a constant value of that fraction or any correlation at all with PM2.5. This is because the method is not using sulfur as a tracer of PM2.5 but rather the indoor/outdoor sulfur ratio as a tracer of PM2.5 infiltration. In fact, the fraction of PM2.5 accounted for by sulfur varied widely, from 2% to 20% (mean 9.9%, SD 3.1%, N = 775). A second way to calculate Finf is to regress indoor sulfur concentrations on outdoor values. Although this calculation is obviously related to the simple calculation of the mean indoor/outdoor ratio, since the slope of the regression is an estimate of Finf, it adds one element to our understanding: that is, a nonzero intercept might indicate possible indoor sources. In fact, the intercept for home 18 was very high (about 1000 ng/m3) and investigation revealed that this subject was using a humidifier in his home for three of the four seasons. It is suspected that he was using tap water in the humidifier and that dissolved solids in the tap water included sulfates or sulfides. Therefore, the personal and indoor sulfur values for that participant during the three seasons not including summer (21 person-days) were deleted from the data set. Using the calculated values of Finf, the contribution of outdoor particles to indoor PM2.5 concentrations can be calculated by multiplying the concentration just outside the house by Finf. Finally, this daily outdoor contribution can be regressed longitudinally for each home against the outdoor concentrations measured at the central site. Estimating the Contribution of Outdoor Air Particles to Personal Exposure. Personal exposure E is the sum of outdoor and nonoutdoor sources
E ) Eo + Eno
(5)
where Eo is the exposure due to outdoor sources and Eno is the exposure due to nonoutdoor sources. Nonoutdoor sources include indoor sources in the home, indoor sources in other locations, sources encountered during transit, and sources associated with the “personal cloud” such as resuspension of particles from floors, furniture, or clothes. We now define an “outdoor personal exposure factor” Fpex such that
E ) FpexCout + Eno
(6)
The factor Fpex plays a role with respect to personal exposure that Finf played with respect to indoor concentrations. Some investigators have labeled Fpex as an “attenuation factor,” marking the attenuation of the outdoor contribution to personal exposure. However, this terminology would imply that attenuation increases as Fpex increases, whereas the reverse is true. We shall call Fpex the “outdoor exposure factor”, indexing the contribution of outdoor particles to personal exposure. Assuming no nonoutdoor sources for sulfur, eq 6 provides a simple equation for Fpex
Fpex ) Spers/Sout
(7)
where Spers is the personal exposure to sulfur and Sout is the outdoor concentration. The value for Fpex obtained from the sulfur measurements may be inserted into eq 6 to determine the outdoor and nonoutdoor contributions to personal exposure to PM2.5. The
TABLE 1. PM2.5 (µg/m3) and Sulfur Concentrations (ng/m3) and Ratios in Matched Personal-Indoor-Outdoor Samples: Means and Selected Percentiles variable
N
mean
SD
min
10th
25th
median
75th
90th
max
PM2.5 pers PM2.5 in PM2.5 out PM2.5in/PM2.5out PM2.5pers/PM2.5out FRM2.5a S pers S in S out Spers/Sout Sin/Sout
727 727 727 727 727 727 727 727 727 727 727
23.0 19.4 19.5 1.08 1.31 18.1 1046 1098 1951 0.55 0.59
16.4 16.5 8.6 1.05 0.99 8.1 633 652 1137 0.14 0.16
3.6 2.3 5.0 0.24 0.21 5.0 137 123 407 0.16 0.17
9.7 6.7 8.8 0.47 0.57 8.1 389 423 762 0.38 0.38
12.8 9.8 12.6 0.58 0.73 11.3 544 603 1053 0.45 0.47
18 14 19 0.75 1.00 17 880 947 1685 0.54 0.58
27 22 25 1.13 1.47 23 1371 1402 2596 0.63 0.68
44 36 32 1.88 2.34 30 1970 2013 3648 0.73 0.79
142 119 52 9.48 10.08 46 3254 3852 5406 1.08 1.06
a
Federal reference method PM2.5 monitor at the central site.
estimated outdoor contributions to personal PM2.5 exposure can then be regressed longitudinally for each person against the measured outdoor concentrations at the central site. The nonoutdoor contribution to personal exposure Eno is made up of two partssindoor-generated concentrations (Cig) while at home and “other” contributions
Eno ) Cig fhome + other
(8)
where fhome is the fraction of time the participant was at home (determined from the activity logs). The “other” term is made up of a combination of indoor-generated particles while indoors away from home and “personal cloud” particles generated at all locations whether indoors or outdoors. Relationship Between Fpex and Finf. Finally, if the calculated value for Finf can be applied to all indoor locations visited by the participant during the monitoring period, then the relationship between Fpex and Finf would be
Fpex ) finFinf + fout
(9)
where fin and fout are the fractions of time the person spent indoors and outdoors (i.e., outdoors or in vehicles) during the monitoring period. Since fin + fout ) 1 and Finf < 1, eq 9 predicts that Fpex is always larger than Finf. Relating Outdoor Measurements at a Central Site to Personal Exposures to Particles of Outdoor Origin. When epidemiologists relate increased mortality to an increase in outdoor particle concentrations, they are assuming a proportional increase in personal exposure to particles of outdoor origin. The constant of proportionality is in fact Fpex. However, if Fpex changes from day to day, misclassification of exposures will occur. One way to estimate the magnitude of the misclassification is to regress the outdoor central-site concentration on the exposures attributed to particles of outdoor origin. The higher the R2 value, the less error due to misclassification.
Results A total of 876 person-days had at least one personal, indoor, or outdoor PM2.5 measurement. About 868 days had at least one filter analyzed by XRF. Samples were flagged if they failed any of a number of quality control criteria. For example, flow rates were required to be within 10% of the target values, and filters were discarded if torn or pierced. PM2.5 samples were flagged if the concentrations exceeded those for co-located PM10 samples (N ) 29). All outdoor PM2.5 filters collected between April 7 and April 11 were flagged due to excessive contamination with pollen (N ) 30). Since the PEM flow rate was only 1/10 of the HI flow rate, it was important to test the PEM and HI side by side with respect to the sulfur measurements. During the summer season, 166 co-located indoor and outdoor samples were
collected. The agreement was excellent, with the slopes ranging between 0.98 (0.03 SE) and 1.03 (0.03 SE), both not significantly different from unity, intercepts that were not significantly different from zero, and R2 values of 0.97 for both the indoor and outdoor sets of paired samples. The HI mass measurements were considered to have a precision of about 5%; the PEM measurements had a precision of about 8%. The precision of the sulfur measurements was calculated to be about 8%. Table 1 lists the observed distributions of all triplets of personal-indoor-outdoor samples with validated PM2.5 and sulfur measurements. Also included in Table 1 are the indoor/ outdoor and personal/outdoor sulfur ratios (Finf and Fpex, respectively), as well as the indoor/outdoor and personal/ outdoor PM2.5 ratios. The lack of indoor sources of sulfur can be seen in the small number of indoor/outdoor ratios exceeding unity, whereas the prevalence of indoor sources of PM2.5 can be seen in the fact that even the mean indoor/ outdoor and personal/outdoor ratios exceed unity. The mean outdoor value at the central site measured by the HI agreed to within 0.1% of the mean residential outdoor value; the FRM was about 6% lower than the HI. The Spearman correlation between the residential and central-site HI monitors was 0.98; between the central-site HI and FRM was 0.94; and between the residential HI monitors and the FRM was 0.92 (N ) 778). The mean sulfur concentrations indoors and outdoors are listed for each home, along with their standard deviations (Table 2). The ratio of these means is the estimated infiltration factor Finf. The interquartile range of 0.49 to 0.66 stays within 20% of the median ratio of 0.56. The complete range of Finf values spans a greater than 3-fold variation, from 0.26 to 0.87. We expect the indoor/outdoor sulfur ratios to vary as a function of air exchange rate, which varies by season. The highest mean air exchange rates were in the winter (1.01 h-1, SD ) 0.73 h-1) and the lowest were in the summer (0.49 h-1, SD ) 0.57 h-1). The indoor/outdoor sulfur ratios averaged across all 37 homes were lowest in the summer (mean 0.50, 0.06 SD), and were virtually identical (0.62-0.63) for the other three seasons. The regressions of indoor on outdoor sulfur resulted in generally high R2 values for each individual home: 14 values of 90% or higher and only one value below 50%. The slope of the regression (an estimate of Finf) was significantly different from zero (p < 0.05) for every home. Of 36 homes, 22 had intercepts not significantly different from zero, indicating no apparent source of sulfur in the home. The intercepts that were different from zero may be more likely due to scatter than to possible indoor sources of sulfur or sulfates. The scatter is due both to variations in Finf with changing seasons and air exchange rates and to measurement error. These VOL. 39, NO. 6, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1709
FIGURE 1. Personal vs outdoor sulfur.
TABLE 2. Indoor and Outdoor 24-Hour Sulfur Measurements (ng/m3) Averaged Over All Visits to Each Home house
N
Sin
SD
Sout
SD
Sin/Sout
SDcalc
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 31 32 33 34 36 37 38
27 26 29 23 27 28 28 6 24 27 8 28 7 27 27 28 26 6 27 27 27 13 13 27 8 24 28 25 34 14 28 23 12 6 20 17
955 1069 1516 882 1199 1293 1476 1270 1087 1049 1661 944 553 701 928 1040 923 1328 1568 1139 1240 1231 1175 909 1827 1829 1335 1059 991 1678 607 522 1746 658 1079 534
372 477 534 359 723 651 901 408 688 558 825 449 224 342 449 554 467 436 769 641 704 607 601 498 679 1009 704 502 446 722 287 322 808 300 327 137
1532 1877 2078 2106 1892 2116 1947 3664 1968 1945 2446 1945 1534 1579 1322 1639 1889 2625 2198 2272 2152 2338 2558 1795 2712 2620 2113 1938 2001 1920 1730 1493 2811 2525 1836 974
612 1030 756 1056 1034 1008 1261 1376 1310 1202 1266 1149 661 774 588 952 1140 996 1218 1229 1322 1175 1508 1153 1292 1457 1276 1147 1269 839 769 978 1139 1130 698 299
0.62 0.57 0.73 0.42 0.63 0.61 0.76 0.35 0.55 0.54 0.68 0.49 0.36 0.44 0.70 0.63 0.49 0.51 0.71 0.50 0.58 0.53 0.46 0.51 0.67 0.70 0.63 0.55 0.50 0.87 0.35 0.35 0.62 0.26 0.59 0.55
0.35 0.40 0.37 0.27 0.52 0.42 0.67 0.17 0.51 0.44 0.49 0.37 0.21 0.31 0.46 0.50 0.38 0.25 0.53 0.39 0.48 0.37 0.36 0.43 0.41 0.55 0.51 0.41 0.38 0.54 0.23 0.31 0.38 0.17 0.29 0.22
all
775
1139
541
2058
1058
0.56
0.39
sources of variability affect regressions by causing lower slopes and higher intercepts than the true values. Therefore the calculated slopes are likely to be underestimates of the true infiltration factor. Only 4 of the 36 homes had slopes higher than the indoor/outdoor sulfur ratio. The overall average slope was only 0.49 (Figure 1), compared to the overall average indoor/outdoor ratio of 0.59. Because of the likelihood that the regressions underestimated the value of the infiltration factor, we chose to use the indoor/outdoor sulfur ratio as our best estimate of Finf in the following calculations. Contribution of Outdoor Particles to Indoor PM2.5. We multiplied the daily Finf for each home by the outdoor air 1710
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 6, 2005
TABLE 3. Estimated Contributions of Outdoor and Indoor-Generated Particles to Total Indoor PM2.5 Concentrations by House (µg/m3) outdoor contribution indoorto indoor generated particles SDcalc SEcalc particles SDcalc SEcalc
house
N
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 31 32 33 34 36 37 38
27 26 29 23 27 28 28 6 24 27 8 28 7 27 27 28 26 6 27 27 27 13 13 27 8 24 28 25 34 14 28 23 12 6 20 17
11.0 13.8 19.4 9.4 12.2 11.7 18.0 9.6 11.3 10.9 15.5 9.1 4.8 7.4 10.8 10.7 8.7 10.7 17.2 10.8 13.0 10.6 9.9 9.4 21.4 17.6 12.7 10.4 9.7 21.7 5.8 6.4 15.4 6.2 10.4 7.4
12.9 18.8 16.6 16.8 17.4 15.0 22.6 15.8 20.0 18.9 17.8 16.1 9.6 13.3 11.4 15.1 15.4 11.9 20.4 19.8 20.6 15.6 19.0 16.6 19.1 22.3 18.1 15.8 16.8 20.7 12.3 18.0 16.7 17.4 10.4 8.3
2.5 3.7 3.1 3.5 3.4 2.8 4.3 6.5 4.1 3.6 6.3 3.1 3.6 2.6 2.2 2.9 3.0 4.8 3.9 3.8 4.0 4.3 5.3 3.2 6.7 4.5 3.4 3.2 2.9 5.5 2.3 3.7 4.8 7.1 2.3 2.0
5.0 2.3 1.6 31.0 5.9 5.8 1.6 8.1 4.8 0.2 11.8 3.7 3.8 4.1 7.2 -0.5 4.1 1.4 3.3 -0.3 25.4 14.0 15.1 27.6 32.6 2.2 22.7 1.0 16.3 -0.2 7.3 5.1 1.3 3.0 1.3 0.3
15.9 21.3 20.0 37.0 20.9 16.5 25.5 16.5 22.0 19.3 22.2 16.9 10.4 14.8 20.3 15.6 18.6 12.2 21.5 20.2 29.6 23.5 33.7 29.1 29.0 24.2 32.1 16.6 22.7 23.2 13.6 20.0 17.3 17.5 10.9 8.8
3.1 4.2 3.7 7.7 4.0 3.1 4.8 6.7 4.5 3.7 7.8 3.2 3.9 2.8 3.9 3.0 3.6 5.0 4.1 3.9 5.7 6.5 9.4 5.6 10.3 4.9 6.1 3.3 3.9 6.2 2.6 4.2 5.0 7.2 2.4 2.1
all
775
11.7
16.5
3.9
7.8
20.5
4.8
concentration to estimate the contribution of outdoor particles to the total indoor particle level (Table 3). The indoor-generated particle concentrations are then the difference between the measured indoor PM2.5 and the calculated outdoor contribution. Since these calculated concen-
FIGURE 2. Comparison of infiltration factor Finf and outdoor exposure factor Fpex by participant. trations are now the result of arithmetical operations involving four measurements, each with error, the relative errors of the estimates of the indoor-generated particles are quite large. Many of the estimates are in fact not significantly different from zero. The relative importance of the indoor and outdoor sources varied by household, with some homes having essentially no indoor contribution while others (N ) 9) had more than 50% indoor-generated contributions. We performed longitudinal regressions of the estimated outdoor contributions to indoor PM2.5 vs the PM2.5 measured just outside the homes. The individual regressions by home had R2 values between 0.39 and 0.95 (median ) 0.77). The overall Pearson correlation coefficient was 0.82, with an R2 (adjusted) value of 0.67. A Spearman rank correlation was also 0.82, indicating that the Pearson coefficient was not affected by outliers. When the regression was run against the HI at the central site, the overall R2 value was reduced somewhat to 0.58 (N ) 736). Against the Federal reference method (FRM) monitor at the central site, it was reduced further to 0.49 (N ) 775). Contribution of Outdoor PM2.5 to Personal Exposure. Personal sulfur exposures averaged over all seasons are compared to residential outdoor values for each home, together with the ratios of personal to outdoor sulfur (Fpex) (Table 4). The 24-h personal exposures to sulfur are very similar to the indoor concentrations in the home. Fpex ranged from 0.33 to 0.77 with a mean value of 0.54. The calculated Fpex values were very similar to the Finf values (Figure 2). We used the estimated Fpex for each participant to calculate the contribution to personal exposure made by particles of outdoor origin (Table 5). The contribution of both outdoor and nonoutdoor sources to personal exposure is shown for each subject in Figure 3. The nonoutdoor contribution was divided into the contribution of sources inside the home and that of indoor or personal sources elsewhere using eq 8. Overall, the contribution of outdoor particles to total exposure was somewhat less than half (47%). The nonoutdoor contributions were divided almost evenly between indoorgenerated particles encountered while the participant was at home (mean 6.6 ( 7.0 µg/m3 SD), and particles from other nonoutdoor sources, including indoor-generated particles away from home and particles from the “personal cloud” at all locations whether indoors or outdoors (mean 5.9 ( 4.8 µg/m3 SD).
TABLE 4. Personal and Outdoor Sulfur Concentrations (ng/m3) by Subject subject
N
Spers
SD
Sout
SD
Spers/Sout
SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 31 32 33 34 35 36 37 38
25 23 21 23 26 27 27 6 25 21 8 27 7 26 27 27 25 6 22 22 28 13 14 23 14 21 28 25 27 14 28 22 13 27 6 21 20
878 920 1209 934 1254 1082 1210 1293 1097 792 1785 892 643 707 841 951 867 1315 1364 1088 1369 1258 1148 1044 1682 1571 1101 879 977 1469 811 489 1769 894 1143 937 571
370 426 496 550 719 610 804 467 755 504 899 441 297 352 417 560 479 492 783 743 787 684 771 611 694 908 560 457 577 604 341 267 793 336 610 357 256
1532 1871 1901 2009 1871 2119 1958 3664 1939 1730 2446 1995 1534 1619 1331 1625 1886 2625 2103 2209 2087 2338 2418 1890 2865 2601 2113 1938 1852 1920 1730 1468 2812 1668 2525 1791 1068
633 916 762 1085 1048 1027 1283 1376 1291 1263 1266 1139 661 761 582 967 1164 996 1327 1347 1343 1175 1540 1187 1292 1550 1276 1147 1214 839 769 994 1090 974 1130 711 452
0.57 0.49 0.64 0.46 0.67 0.51 0.62 0.35 0.57 0.46 0.73 0.45 0.42 0.44 0.63 0.59 0.46 0.50 0.65 0.49 0.66 0.54 0.47 0.55 0.59 0.60 0.52 0.45 0.53 0.77 0.47 0.33 0.63 0.60 0.45 0.52 0.53
0.34 0.33 0.37 0.37 0.54 0.38 0.58 0.18 0.54 0.44 0.53 0.34 0.26 0.30 0.42 0.49 0.38 0.27 0.55 0.45 0.57 0.40 0.44 0.47 0.36 0.50 0.41 0.36 0.47 0.46 0.29 0.29 0.37 0.42 0.32 0.29 0.33
all
765
1093
568
2038
1072
0.54
0.40
We then regressed our estimate of the outdoor contribution to personal exposure vs the residential outdoor PM2.5 concentrations for each person. The R2 values ranged from 0.42 to 0.93 (median 0.81). When the regression was run against the Federal reference method PM2.5 values at the central site, the range of R2 values was from 0.19 to 0.88 VOL. 39, NO. 6, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1711
FIGURE 3. Outdoor and nonoutdoor contributions to personal PM2.5 exposure. The nonoutdoor contribution is divided into indoor-generated particles encountered while at home and particles from other sources.
TABLE 5. Estimated Contribution of Outdoor Particles to Personal Exposure (µg/m3)
SD
SE
nonoutdoor contribution to personal exposure
TABLE 6. Regressions of Personal Exposure to Outdoor-Generated Particles vs Outdoor PM2.5 Measured by the FRM at the Central Site
subject
N
outdoor contribution to personal exposure
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 31 32 33 34 35 36 37 38
25 23 21 23 26 27 27 6 25 21 8 27 7 26 27 27 25 6 22 22 28 13 14 23 14 21 28 25 27 14 28 22 13 27 6 21 20
9.9 11.6 15.5 9.8 12.8 9.8 14.5 9.7 11.2 7.9 16.4 8.4 5.6 7.3 10.3 9.7 8.2 10.5 13.3 9.0 14.2 10.6 9.9 9.8 17.8 14.8 10.5 8.5 9.6 18.8 7.8 5.9 15.5 9.6 10.6 9.0 7.1
13.3 17.9 17.4 19.4 17.2 15.9 23.8 16.5 20.5 18.9 17.9 16.3 10.0 13.3 12.6 15.8 16.1 12.4 18.7 19.1 20.8 16.3 21.2 16.8 20.1 22.6 17.9 16.3 17.6 20.5 11.8 17.7 16.0 13.7 18.6 11.2 10.1
2.7 3.7 3.8 4.0 3.4 3.1 4.6 6.7 4.1 4.1 6.3 3.1 3.8 2.6 2.4 3.0 3.2 5.1 4.0 4.1 3.9 4.5 5.7 3.5 5.4 4.9 3.4 3.3 3.4 5.5 2.2 3.8 4.4 2.6 7.6 2.5 2.3
10.9 7.4 19.0 29.6 16.6 11.7 10.4 13.4 9.0 5.1 13.5 5.5 7.9 5.2 10.0 7.8 12.8 5.9 6.6 6.7 31.9 18.4 15.7 27.1 33.0 19.5 13.7 9.1 9.0 5.9 8.7 9.9 5.4 13.9 7.3 10.3 6.6
16.2 3.2 20.3 4.2 33.1 7.2 36.1 7.5 29.8 5.8 25.6 4.9 29.7 5.7 17.8 7.3 22.8 4.6 19.7 4.3 21.9 7.8 16.9 3.2 13.7 5.2 14.6 2.9 16.7 3.2 17.8 3.4 23.5 4.7 13.2 5.4 20.1 4.3 20.4 4.3 24.9 4.7 23.5 6.5 26.9 7.2 26.2 5.5 52.7 14.1 27.1 5.9 22.0 4.1 18.7 3.7 19.0 3.7 24.5 6.5 14.3 2.7 21.4 4.6 17.9 5.0 10.4 3.4 19.4 7.9 16.5 3.6 11.6 2.6
all
765
10.9
16.9 4.0
12.5
21.8
SD
SE
subject
N
slope
SE
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 31 32 33 34 35 36 37 38
25 24 21 23 27 27 27 6 25 21 9 27 7 27 27 27 25 6 22 22 28 13 14 23 14 21 28 25 27 14 28 22 13 27 6 21 21
0.58 0.45 0.48 0.40 0.76 0.50 0.57 0.36 0.55 0.46 0.62 0.35 0.54 0.48 0.71 0.58 0.35 0.52 0.47 0.58 0.51 0.43 0.45 0.41 0.68 0.60 0.41 0.34 0.37 0.60 0.31 0.33 0.57 0.23 0.56 0.46 0.53
0.06 0.06 0.06 0.05 0.10 0.04 0.07 0.06 0.09 0.06 0.18 0.04 0.15 0.05 0.05 0.07 0.10 0.22 0.06 0.07 0.14 0.12 0.06 0.09 0.10 0.07 0.04 0.04 0.06 0.09 0.06 0.06 0.11 0.08 0.09 0.06 0.08
all
770 0.49 0.08
p (slope) intercept 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.004
0.95 2.71 5.81 1.49 0.24 0.39 2.07 0.49 1.11 1.05 3.08 2.58 -0.92 -0.11 0.25 0.85 2.09 0.08 5.18 -1.47 4.08 1.64 1.39 2.48 -1.56 1.56 3.01 2.23 3.61 5.77 2.66 0.85 0.93 6.06 -2.39 1.99 0.38 1.69
SE
p (Int)
R2
1.08 1.16 1.35 1.19 1.75 0.87 1.77 1.54 1.82 0.97 4.32 0.67 1.88 0.77 0.83 1.11 1.77 4.54 1.22 1.32 2.93 2.64 1.20 1.68 3.12 1.70 0.73 0.83 1.14 2.30 1.09 1.00 2.95 1.40 2.20 1.02 1.10
0.39 0.03 0.00 0.22 0.89 0.65 0.25 0.77 0.55 0.29 0.50 0.00 0.64 0.88 0.76 0.45 0.25 0.99 0.00 0.28 0.17 0.55 0.27 0.15 0.63 0.37 0.00 0.01 0.00 0.03 0.02 0.41 0.76 0.00 0.34 0.07 0.73
0.77 0.74 0.76 0.73 0.69 0.84 0.69 0.88 0.59 0.74 0.57 0.77 0.67 0.8 0.87 0.74 0.34 0.47 0.71 0.77 0.33 0.5 0.83 0.49 0.76 0.78 0.82 0.73 0.56 0.76 0.47 0.59 0.68 0.19 0.88 0.73 0.67
1.65 0.36 0.67
5.2
Discussion (median ) 0.73). This last set of regressions (Table 6) is the one that interests epidemiologists, who use values at the central site to estimate personal exposure. 1712
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 6, 2005
This study had an unusually large number of measurements of personal exposure and indoor concentrations of PM2.5 and sulfur across all four seasons. The high precision of the
measurements allowed the infiltration factor Finf for each home, and the personal exposure factor Fpex for each participant, to be estimated with reasonable confidence. However, sulfates are smaller than many fine particles. Since particles near the upper end of the PM2.5 size range may have lower penetration efficiencies into homes, and also have higher deposition rates, the infiltration factors estimated using sulfur may be overestimates of the average infiltration factor for PM2.5 as a whole. For example, the average indooroutdoor ratio for iron in this study was 0.38, compared to the value of 0.59 for sulfur. To the extent that iron and other crustal elements contribute to the PM2.5 total mass, the sulfurbased estimates of the outdoor contribution to PM2.5 exposure will be overestimates. However, for the Eastern region of the United States where our study took place, sulfur in the form of sulfates contributes much more to fine particle mass (about 40% in this case) than crustal particles (probably less than 10%). For each home and for each person, the 24-h values of Finf and Fpex showed a fair amount of day-to-day variation. Air conditioning in the summer had a major effect on the values of these parameters, but day-to-day variations occurred within any season. The greater these variations, the greater the amount of error in assuming that a given increase in outdoor fine particle concentrations will produce a proportional increase in personal exposure. The magnitude of this error is suggested by the range in R2 values of the regressions of personal exposure to particles of outdoor origin against the central-site measurements. To the extent that these R2 values fall short of unity, the estimated effect of central-site fine particles on health will be weakened. For our population of persons at high risk of such health effects, the R2 values ranged from 0.19 to 0.88 (median 0.73). However, we caution that these must be upper bounds, because the assumptions of our model, which force the infiltrated particles of outdoor origin to be proportional to outdoor concentrations, virtually ensure that R2 values will be high. Epidemiologists and statisticians may be able to use the observed variation in Fpex and these R2 values to refine their estimates of the health effects of fine particles. Finf and Fpex had similar ranges and means, suggesting that the infiltration factor measured for a home is a good estimator for the personal exposure factor. This in turn indicates that indoor and outdoor measurements of sulfur concentrations could be used even in the absence of personal exposure measurements to estimate the contribution of outdoor fine particles to personal exposures.
Acknowledgments We thank Linda Sheldon for many stimulating discussions. Robert Kellogg of METI oversaw the X-ray fluorescence analyses that resulted in the excellent sulfur data. Charles Rodes of Research Triangle Institute International led the RTII team in collecting the field data. Jack Suggs, Carry Croghan, and Anne Rea of EPA put together the databases used in this report. The Exposure Measurements and Analysis Branch of EPA’s National Exposure Research Laboratory was responsible for designing and overseeing the study. We especially thank the participants who carried the burden of responsibility for up to 28 days over a year’s time.
Disclaimer The U.S. Environmental Protection Agency through its Office of Research and Development funded and conducted the research described here through Contract 68-D-99-012 to the Research Triangle Institute International. It has been subjected to Agency review and approved for publica-
tion. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
Literature Cited (1) Schwartz, J.; Dockery, D. W.; Neas, L. M. Is daily mortality associated specifically with fine particles? J. Air Waste Manage. Assoc. 1996, 46, 927-939. (2) Abt, E.; Suh, H. H.; Allen, G.; Koutrakis, P. Characterization of indoor particle sources: a study conducted in the metropolitan Boston area. Environ. Health Persp. 2000, 108 (1), 35-44. (3) Abt, E.; Suh, H. H.; Catalano, P.; Koutrakis, P. Relative contribution of outdoor and indoor particle sources to indoor concentrations. Environ. Sci. Technol. 2000, 34, 3579-3587. (4) Clayton, C. A.; Perritt, R. L.; Pellizzari, E. D.; Thomas, K. W.; Whitmore, R. W.; Wallace, L. A.; O ¨ zkaynak, H.; Spengler, J. D. Particle Total Exposure Assessment Methodology (PTEAM) study: distributions of aerosol and elemental concentrations in personal, indoor, and outdoor air samples in a southern California community. J. Exposure Analysis Environ. Epidemiol. 1993, 3, 227-250. (5) Evans, G. F.; Highsmith, R. V.; Sheldon, L. S.; Suggs, J. C.; Williams, R. W.; Zweidinger, R. B.; Creason, J. P.; Walsh, D.; Rodes, C. E.; Lawless, P. A. The 1999 Fresno particulate matter exposure studies: comparison of community, outdoor, and residential PM mass measurements. J. Air Waste Manage. Assoc. 2000, 50 (11), 1887-1896. (6) Hopke P. K.; Ramadan, Z.; Paatero, P.; Norris, G. A.; Landis, M. L.; Williams, R. W.; Lewis, C. W. Receptor modeling of ambient and personal exposure samples: 1998 Baltimore Particulate Matter Epidemiology-Exposure Study. Atmos. Environ. 2003, 37 (23), 3289-3302. (7) Howard-Reed, C.; Rea, A. W.; Zufall, M. J.; Burke, J. M.; Williams, R. W.; Suggs, J. C.; Sheldon, L. S.; Walsh, D.; Kwock, R. Use of a continuous nephelometer to measure personal exposure to particles during the U. S. Environmental Protection Agency Baltimore and Fresno panel studies. J. Air Waste Manage. Assoc. 2000, 50, 1125-1132. (8) Janssen, N. A. H.; Hoek, G.; Brunekreef, B.; Harssema, H.; Mensink, I.; Zuldhof, A. Personal sampling in adults: relation among personal, indoor, and outdoor air concentrations. Am. J. Epidemiol. 1998, 147, 537-547. (9) Janssen, N. A. H.; Hoek, G.; Harssema, H.; Brunekreef, B. Personal exposure to fine particles in children correlates closely with ambient fine particles. Arch. Environ. Health 1999, 54, 95-101. (10) Janssen, N. A. H.; De Hartog, J. J.; Hoek, G.; Brunekreef, B.; Lanki, T.; Timonen, K. L.; Pekkanen, J. Personal exposure to fine particulate matter in elderly subjects: relation between personal, indoor, and outdoor concentrations. J. Air Waste Manage. Assoc. 2000, 50, 1133-1143. (11) Keeler, G. J.; Dvonch, J. T.; Yip, F. Y.; Parker, E. A.; Israel, B. A.; Marsik, F. J.; Morishita, M.; Barres, J. A.; Robins, T. G.; BrakefieldCaldwell, W.; Sam, M. Assessment of personal and communitylevel exposures to particulate matter among children with asthma in Detroit, Michigan, as part of Community Action Against Asthma (CAAA). Environ. Health Persp. 2002, 110(supplement 2), 173-181. (12) Landis, M. S.; Norris, G. A.; Williams, R. W.; Weinstein, J. P. Personal exposures to PM2.5 mass and trace elements in Baltimore, Maryland. Atmos. Environ. 2001, 35, 6511-6524. (13) Liu, L.-J. S.; Box, M.; Kalman, D.; Kaufman, J.; Koenig, J.; Larson, T.; Sheppard, L.; Slaughter, C.; Lewtas, J.; Wallace, L. A. Exposure assessment of particulate matter for susceptible populations in Seattle, WA. Environ. Health Persp. 2003, 111 (7), 909-918. (14) Long, C. M.; Suh, H. H.; Catalano, P.; Koutrakis, P. Using timeand size-resolved particulate data to quantify indoor penetration and deposition behavior. Environ. Sci. Technol. 2001, 35, 20892099. (15) Long, C. M.; Suh, H. H.; Koutrakis, P. Characterization of indoor particle sources using continuous mass and size monitors. J. Air Waste Manage. Assoc. 2000, 50 (7), 1236-1250. (16) O ¨ zkaynak, H.; Xue, J.; Spengler, J. D.; Wallace, L. A.; Pellizzari, E. D.; Jenkins, P. Personal exposure to airborne particles and metals: results from the Particle TEAM Study in Riverside, CA. J. Exposure Anal. Environ. Epidemiol. 1996, 6, 57-78. (17) O ¨ zkaynak, H.; Xue, J.; Weker, R.; Butler, D.; Koutrakis, P.; Spengler, J. The Particle TEAM (PTEAM) Study: Analysis of the Data. Volume 3. EPA/600/R-95/098; U.S. Environmental Protection Agency: Research Triangle Park, NC, 1996. (18) Pellizzari, E. D.; Thomas, K. W.; Clayton, C. A.; Whitmore, R. W.; Shores, R. C.; Zelon, H. S.; Perritt, R. L. Particle Total Exposure VOL. 39, NO. 6, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1713
(19)
(20)
(21)
(22) (23) (24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32) (33)
(34)
Assessment Methodology (PTEAM): Riverside, California pilot study. Volume I, Final Report. EPA Contract 68-02-4544; U.S. Environmental Protection Agency: Research Triangle Park, NC, 1992. Rea, A.; Zufall, M.; Williams, R.; Reed, C.; Sheldon, L. The influence of human activity patterns on personal PM exposure: a comparative analysis of filter-based and continuous particle measurements. J. Air Waste Manage. Assoc. 2001, 51, 12711279. Rojas-Bracho, L.; Suh, H. H.; Koutrakis, P. Relationships among personal, indoor, and outdoor fine and coarse particle concentrations for individuals with COPD. J. Exposure Anal. Environ. Epidemiol. 2000, 10, 294-306. Sarnat, J. A.; Koutrakis, P.; Suh, H. H. Assessing the relationship between personal particulate and gaseous exposures of senior citizens living in Baltimore, MD. J. Air Waste Manage. Assoc. 2000, 50 (7), 1184-1198. Sarnat, J. A.; Schwartz, J.; Catalano, P. J.; Suh, H. H. Gaseous pollutants in particulate matter epidemiology: confounders or surrogates? Environ. Health Persp. 2001 109, 1053-1061. Sarnat, J. A.; Long, C. M.; Koutrakis, P.; Coull, B. A.; Schwartz, J.; Suh, H. H. Using sulfur as a tracer of outdoor fine particulate matter. Environ. Sci. Technol. 2002, 36, 5305-5314. Thomas, K. W.; Pellizzari, E. D.; Clayton, C. A.; Whitaker, D. A.; Shores, R. C.; Spengler, J. D.; O ¨ zkaynak, H.; Wallace, L. A. Particle Total Exposure Assessment Methodology (PTEAM) 1990 study: method performance and data quality for personal, indoor, and outdoor monitoring. J. Exposure Anal. Environ. Epidemiol. 1993, 3, 203-226. U.S. EPA. Preliminary Particulate Matter Mass Concentrations Associated with Longitudinal Panel Studies: Assessing Human Exposures of High-Risk Subpopulations to Particulate Matter. National Exposure Research Laboratory, Office of Research and Development, EPA/600/R-01/086; U.S. Environmental Protection Agency: Washington, DC, 2002. Vette, A. F.; Rea, A. W.; Lawless, P. A.; Rodes, C. E.; Evans, G.; Highsmith, V. R.; Sheldon, L. Characterization of indoor-outdoor aerosol concentration relationships during the Fresno PM exposure studies. Aerosol Sci. Technol. 2001, 34, 118-126. Wallace, L. A.; Mitchell, H.; O’Connor, G. T.; Liu, L.-J. S.; Neas, L.; Lippmann, M.; Kattan, M.; Koenig, J.; Stout, J. W.; Vaughn, B. J.; Wallace, D.; Walter, M.; Adams, K. Particle concentrations in inner-city homes of children with asthma: the effect of smoking, cooking, and outdoor pollution. Environ. Health Persp. 2003, 111, 1265-127. Williams, R.; Creason, J.; Zweidinger, R.; Watts, R.; Sheldon, L.; Shy, C. Indoor, outdoor and personal exposure monitoring of particulate air pollution: the Baltimore elderly epidemiologyexposure pilot study. Atmos. Environ. 2000, 34, 4193-4204. Williams, R.; Suggs, J.; Zweidinger, R.; Evans, G.; Creason, J.; Kwok, R.; Rodes C.; Lawless, P.; Sheldon, L. Comparison of PM2.5 and PM10 monitors. J. Exposure Anal. Environ. Epidemiol. 2000, 10, 497-505. Williams, R. W.; Suggs, J.; Rea, A.; Leovic, K.; Vette, A.; Croghan, C.; Sheldon, L.; Rodes, C.; Thornburg, J.; Ejire, A. et al. The Research Triangle Park particulate matter panel study: PM mass concentration relationships. Atmos. Environ. 2003, 37 (38), 5349-5363. Williams, R.; Suggs, J.; Rea, A.; Sheldon, L.; Rodes, C.; Thornburg, J. The Research Triangle Park particulate matter panel study: modeling ambient source contribution to personal and residential PM mass concentrations. Atmos. Environ. 2003, 37 (38), 5365-5378. Wallace, L. A. Correlations of personal exposure to particles with outdoor air measurements: a review of recent studies. Aerosol Sci. Technol. 2000, 32, 15-25. Allen R.; Larson, T.; Sheppard, L.; Wallace, L.; Liu, L.-J S. Use of real-time light scattering data to estimate the contribution of infiltrated and indoor-generated particles to indoor air. Environ. Sci. Technol. 2003, 37, 3484-3492. Allen R.; Larson, T.; Sheppard, L.; Wallace, L.; Liu, L.-J S. Estimated hourly personal exposures to ambient and nonambient particulate matter among sensitive populations in Seattle,
1714
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 6, 2005
(35)
(36) (37)
(38)
(39)
(40)
(41)
(42)
(43) (44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
Washington. J. Air Waste Manage. Assoc. 2004, 38 (11), 15671577. Ebelt, S. T.; Petkau, A. J.; Vedal, S.; Fisher, T. V.; Brauer, M. Exposure of chronic obstructive pulmonary disease patients to particulate matter: relationship between personal and ambient air concentrations. J. Air Waste Manage. Assoc. 2000, 50, 10811094. Nazaroff, W. W.; Cass, G. R. Mathematical modeling of indoor aerosol dynamics. Environ. Sci. Technol. 1989, 23, 157-166. Lai, A. C. K.; Nazaroff, W. W. Modeling indoor particle deposition from turbulent flow onto smooth surfaces. J. Aerosol Sci. 2000, 31, 463-476. Howard-Reed, C.; Wallace, L.; Emmerich, S. J. Effect of ventilation systems and air filters on decay rates of particles produced by indoor sources in an occupied townhouse Atmos. Environ. 2003, 37 (38), 5295-5306. Thatcher, T. L.; Lai, A. C. K.; Moreno-Jackson, R.; Sextro, R. G.; Nazaroff, W. W. Effects of room furnishings and air speed on particle deposition rates indoors. Atmos. Environ. 2002, 36 (11), 1811-1819. Wallace, L. A.; Emmerich, S, J.; Howard-Reed, C. Effect of central fans and in-duct filters on deposition rates of ultrafine and fine particles in an occupied townhouse. Atmos. Environ. 2004, 38 (4), 405-413. Howard-Reed, C. H.; Wallace, L. A.; Ott, W. R. The effect of opening windows on air change rates in two homes. J. Air Waste Manage. Assoc. 2002, 52, 147-159. Wallace, L. A.; Emmerich, S. J.; Howard-Reed, C. Continuous measurements of air change rates in an occupied house for 1 year: The effect of temperature, wind, fans, and windows. J. Exposure Anal. Environ. Epidemiol. 2002, 12, 296-306. Liu, D. L.; Nazaroff, W. W. Modeling pollutant penetration across building envelopes. Atmos. Environ. 2001, 35, 4451-4462. Mosley, R. B.; Greenwell, D. J.; Sparks, L. E.; Guo, Z.; Tucker, W. G.; Fortmann, R.; Whitfield, C. Penetration of ambient fine particles into the indoor environment. Aerosol Sci. Technol. 2001, 34 (1), 127-136. Thornburg, J.; Ensor, D. S.; Rodes, C. E.; Lawless, P. A.; Sparks, L. E.; Mosley, R. B. Penetration of particles into buildings and associated physical factors. Part 1: model development and computer simulations. Aerosol Sci. Technol. 2001, 34, 284-296. Lai, A. C. K.; Burne, M. A.; Goddard, A. J. H. Measured deposition of aerosol particles on a two-dimensional ribbed surface in a turbulent duct flow. J. Aerosol Sci. 1999, 30 (9), 1201-1214. Thornburg, J. W.; Rodes, C. E.; Lawless, P. A.; Stevens, C. D.; Williams, R. W. A pilot study of the influence of residential HAC duty cycle on indoor air quality. Atmos. Environ. Wallace, L. A.; Howard-Reed, C. H. Continuous monitoring of ultrafine, fine, and coarse particles in a residence for 18 months in 1999-2000. J Air Waste Manage. Assoc. 2002, 52 (7), 828-844. U.S. EPA. Air quality criteria for particulate matter (fourth external review draft) June 2003. U. S. Environmental Protection Agency, Office of Research and Development, National Center For Environmental Assessment: Research Triangle Park, NC, 2003. Ott, W.; Wallace, L.; Mage, D. Predicting Particulate (PM10) Personal exposure distributions using a random component superposition statistical model. J. Air Waste Manage. Assoc. 2000, 50, 1390-1406. Wilson, W. E.; Mage, D. T.; Grant, L. D. Estimating separately personal exposure to ambient and nonambient particulate matter for epidemiology and risk assessment: why and how. J Air Waste Manage. Assoc. 2000, 50, 1167-1183. Wilson, W.; Suh, H. Fine and coarse particles: concentration relationships relevant to epidemiological studies. J. Air Waste Manage. Assoc. 1997, 47, 1238-1249.
Received for review March 24, 2004. Revised manuscript received October 14, 2004. Accepted December 3, 2004. ES049547U
