Introduction

Many studies have found that ambient particulate matter (PM) concentrations are positively correlated with respiratory and cardiovascular disease (Vedal, 1997; Samet et al., 2000; Peters et al., 2001). Because people spend most of their time indoors (Robinson et al., 1991; Smith, 1993), researchers have been investigating the relationships between personal, indoor, and outdoor PM to better understand the effect of ambient air on personal exposure and to identify indoor sources of PM exposure. Human exposure studies have shown that personal exposure to PM is highly correlated with indoor air but poorly correlated with outdoor air (Dockery and Spengler, 1981; Özkaynak et al., 1996). However, several recent studies have shown that the personal to outdoor correlation improves with regressions for individual participants over periods of several days (Lioy et al., 1990; Janssen et al., 1997,1998,1999,2000). Investigating indoor sources of PM will improve our understanding of what humans are actually exposed to and how to reduce this exposure.

Long et al. (2000) found that indoor sources of PM are typically short-term, high-concentration events that emit primarily ultrafine (particle diameter D_p0.1 m) and coarse mode (2.5D_p10 m) particles. Although the health effects of short-term PM exposures are not well understood, several recent studies have correlated disease symptoms with increased ambient short-term PM. Asthma symptoms have been associated with 1- and 8-h maximum PM₁₀ concentrations (Delfino et al., 1998), increased chronic obstructive pulmonary disease hospital admissions with 1-h fine particle concentrations (Morgan et al., 1998), and decreased heart rate variability in elderly subjects with 4-h mean PM_2.5 concentrations (Gold et al., 2000).

A summary of major indoor/outdoor concentration studies has shown that cigarette smoking and cooking are the largest identified sources of indoor air particles, while cleaning activities were identified as a contributor to indoor PM of unknown significance (Wallace, 1996). However, normal human activities that resuspend house dust, such as walking around and sitting on furniture, may contribute to the 25% of indoor concentrations for which sources could not be identified in the EPA Particle Total Exposure Assessment Methodology (PTEAM) study (Wallace, 1996). Human activities that resuspend house dust have also been identified as a potential cause for the "personal cloud" effect, an observation that personal exposures are often higher than indoor concentrations (Özkaynak et al., 1996). Although the health effects from breathing house dust are not well understood, resuspended house dust is a potentially important indoor source of human PM exposure. House dust can contain toxins, carcinogens, and allergens such as lead, pesticides, polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), dioxins, molds, pet allergens, and dust mites (Roberts and Dickey, 1995).

A few studies have directly examined the effect of specific human activities on indoor PM concentrations (Thatcher and Layton, 1995; Long et al., 2000). These studies found that walking, dusting, vacuuming, and general cleaning dramatically increased indoor concentrations of PM greater than 1 m. Brauer et al. (1999) measured personal exposure directly for a subject walking about a small dormitory room and found that the personal cloud was more pronounced when the subject was walking about than when the subject was sedentary. Ferro et al. (1999) found that pushing a vacuum cleaner, with the vacuum cleaner on or off, increased human exposure to coarse PM by approximately 20 times as compared with background concentrations.

The purpose of this study is to characterize a larger set of human resuspension activities than has been examined in the past and determine the relative importance of these activities in contributing to human exposure and the personal cloud. In addition to walking, sitting on furniture, vacuuming, and dusting activities, which have been included in previous studies, this study examines other common human activities, such as making a bed, folding clothes, and folding blankets. This study also investigates whether vigor of activity and flooring type impact human exposure to resuspended PM. To conduct the study, a series of prescribed human activities were performed and concurrent measurements from personal exposure monitor (PEM), stationary indoor monitor (SIM), and stationary outdoor monitor (SAM) locations were collected. The use of size-segregated, real-time instruments, in addition to integrated filter samples, provided the ability to examine short-term impacts of the activities on human exposure.

Methods

Collocated real-time particle counters and integrated filter samplers were used to measure personal, indoor, and outdoor PM concentrations for five consecutive days during April 2000 in a home with one occupant in Redwood City, CA, USA. During the first 3 days, three to four prescribed indoor human activities were performed to resuspend house dust. During the remaining 2 days, no prescribed activities took place and the occupant was away from home during the workday. The experiments were conducted without mechanical ventilation and with all external doors and windows closed. The mean daytime air exchange rate while the prescribed activities were taking place was 0.5 air change per hour (ACH) on the first floor and 1.0 ACH in the basement of the house.

A total of 10 prescribed activities were selected to survey common human activities that resuspend house dust. Each prescribed activity period lasted 15 min and was followed by a 45-min period with no activity. This experimental design allowed the activities to be analyzed separately from one another using the concentration time series from the real-time particle counters. All activities took place either in the same room as the SIM or in a room adjacent to the SIM location. On the first day of the experimental period, the SIM was located in the basement and the prescribed activities were performed in the basement. Activities included dancing on a rug, walking on a rug, and dancing on a wood floor. On the second and third days, the SIM was located in the first-floor living room and the prescribed activities took place on the first floor. Prescribed activities on the second day included walking and sitting on furniture, making a bed, folding clothes, and folding blankets. Prescribed activities on the third day included two persons walking and sitting on furniture, dusting, vacuuming the first-floor living room and dining room, and vacuuming the first-floor bedrooms. All activities were performed by one person except where the two persons were walking and sitting on furniture activity.

While performing the activities, the investigator wore personal monitoring instruments attached to a backpack frame. During the nonactivity periods, the investigator sat down several meters from the SIM and removed the backpack frame, placing it so the intakes were in the near-vicinity of the breathing zone. The investigator wore clean, summer clothing and restrained hair to minimize personal shedding. The SIM location consisted of a 6-foot stepladder supporting the instruments with the intakes at breathing height. The SAM location was under a two-sided, roofed shed in the backyard of the home with the filter samplers supported by a metal stand and the real-time particle counters sitting on a table.

The particle counters used for the study were Met-One Model 237B laser particle counters (Grants Pass, OR, USA). The PEM unit had size ranges of 0.3–0.5, 0.5–1.0, 1.0–2.5, 2.5–5.0, 5.0–10, and >10 m and the SIM and SAM units had size ranges of 0.3–0.5, 0.5–0.7, 0.7–1.0, 1.0–2.5, 2.5–5.0, and >5.0 m. The particle counters were calibrated by Pacific Scientific (Grants Pass, OR, USA) by count matching each instrument with a standard particle counter using polystyrene latex (PSL) spheres as the source of particles.

Several instrument collocation periods were included in the experimental protocol, allowing field matching of the particle counters. During the collocation periods, all instruments were placed together indoors with joined intakes. The readings from the indoor particle counter were plotted versus the readings from the personal and outdoor particle counters for all collocation periods. The personal and outdoor particle counter measurements were adjusted to match the indoor readings using the slope and the intercept from the linear regression of the collocation readings. An example of this method is shown in Figure 1, where the collocation readings from the indoor particle counter are plotted against those from the outdoor particle counter for the 1.0–2.5 m size range. While the slopes for some of the size ranges differed from 1.0, the data scatter was minimal for most size ranges, and the adjusted measurements could be compared with confidence.

Figure 1.

Particle counts per 3 min for the indoor particle counter plotted against the outdoor particle counter for four collocation periods (1.0–2.5 m size range).

Full figure and legend (31K)

The filter samples were collected downstream of custom cyclone samplers (John and Reischl, 1980), with PM_2.5 and PM₅ sampled at flow rates of 20 and 10 L/min, respectively. Although PM₅ is not a regulated pollutant in the United States, the PM₅ samples were collected to be consistent with the particle counters, two of which did not have a 10 m size cut. The samples were collected on Pall Gelman TF-1000 PTFE Membrane 47 mm filters. All TF-1000 filter samples were collected in duplicate with the exception of the three personal samples.

To estimate the volume concentration (V) from the number concentration (N) given by the particle counters, we summed the product of N and the volume per particle for each size class k=1–n, which corresponded with the size range desired for V. This method is given in Eq. (1), where D_p is the geometric mean particle diameter for the size range k:

Equation (1) assumes that the particles are spherical with light scattering properties similar to the PSL spheres used to calibrate the particle counters. Multiplying V by a particle density of 2.5 g/cm³, the approximate density of soil dust, the mass concentrations estimated from the particle counts were compared with the mass concentrations measured by the filters. Table 1 shows that for indoor PM_2.5, the measured concentration was almost twice as high as the estimated concentration. The measured/estimated concentration ratios for the other locations and size fractions were closer to one, with the indoor and personal PM₅ measured/estimated mass concentration ratios slightly less than one and the outdoor PM_2.5 and PM₅ mass concentrations slightly more than one. Discrepancies between the mass concentration estimated from the particle counts and the mass concentration measured by the filter samplers may be attributed to the particle density being less than or greater than 2.5 g/cm³ as well as the particles of ambient and indoor origin being nonideal and less efficient at scattering light than the PSL spheres (e.g., Whitby and Willeke, 1979).

Table 1 - Mean and standard deviation of the measured/estimated mass concentration ratios for five consecutive days.

Full table

Although PM composition was not fully determined for this study, filter samples were analyzed for 13 elemental ions using inductively coupled plasma-mass spectrometry (ICP-MS). The ICP-MS results support the hypothesis that the different measured/estimated mass concentration ratios are due to the specific composition of the particles in each location and size range. The elemental ion profiles were similar for the personal and indoor PM₅ samples, suggesting that the overall composition of these samples was similar. Conversely, the elemental ion profiles for the indoor and outdoor samples were very different. To more accurately represent the measured particle mass, the estimated mass concentrations were multiplied by the measured/estimated mass concentration ratios for the corresponding day, monitoring location, and size range.

Results

The mean PM_2.5, PM₅, and PM₁₀ personal exposures (i.e., concentration in the breathing zone of the investigator) for each of the prescribed activities are presented in Figure 2. The personal exposures are corrected for background concentrations, including the increase in particle levels from previous activities. Each mean represents approximately 15 1-min observations. Because PM₁₀ filter samples were not collected, the PM₁₀ mass concentration results are theoretical estimates calculated using Eq. (1) for all size ranges less than 10 m and a particle density of 1 g/cm³. The PM₁₀ mass concentrations should be viewed as lower-bound estimates because the particle density is likely higher than 1 g/cm³. The PM₁₀ data are less reliable than the PM_2.5 and PM₅ data due to the difficulty of calibrating light scattering particle counters in the >5 m size range.

Figure 2.

Mean personal exposure from 15-min human activities. Notes: BR=first-floor bedroom, FF=first floor, BM=basement; the "walk" activity consisted of continuous walking and sitting on furniture; the vacuum cleaner used for the vacuum activities was a Kenmore (Sears) canister model with a paper filtration bag.

Full figure and legend (34K)

Figure 2 shows that most of the mass from the prescribed activities was in the coarse mode. In general, activities that disturbed dust reservoirs on furniture and textiles, such as dry dusting, folding clothes and blankets, and making a bed, produced the highest exposures of PM_2.5, PM₅, and PM₁₀. More vigorous activities produced higher exposures of PM than less vigorous activities, and activity on rugs resuspended more PM than activity on wood floors. For example, dancing on a rug produced PM_2.5, PM₅, and PM₁₀ exposures approximately 2.5 times as high as walking on the same rug and 6–10 times as high as dancing on a wood floor in the same location. Surprisingly, continuous walking around and sitting on furniture increased PM exposures approximately the same amount as vacuuming the same area. Because the walking activity also included sitting on furniture, the effect of walking on the floor could not be separated from that of sitting on furniture.

Figure 3 plots the personal, indoor and outdoor PM₅ estimated mass concentration time series for a representative day of the experimental period. Each observation represents a 3-min average concentration. The time series includes the setup period, four activity periods (each followed by a nonactivity period), and the collocation period. The activity periods are characterized by a sharp increase in indoor and personal concentration, while the nonactivity periods are characterized by decays in the indoor and personal concentrations. The mean outdoor PM₅ concentration for the period from 12:00 to 17:00 is approximately one-fifth of the indoor and personal levels. The personal exposure is higher than the indoor concentration during both the activity periods and the nonactivity decay periods. The observed personal cloud appears to be caused by noninstantaneously mixing in the room and the spatial separation of the PEM and SIM during the activity and nonactivity periods. Movements of the seated investigator during the nonactivity period may also have contributed to the increased particle concentration as measured by the PEM. The plot demonstrates that the mass concentrations estimated from the adjusted particle counts are well matched during the collocation period.

Figure 3.

Personal, indoor, and outdoor PM₅ estimated mass concentration time series.

Full figure and legend (45K)

Human activities contributed to the personal and indoor PM concentrations for all 3 days when prescribed activities took place. The mean daytime 5-h PM_2.5 and PM₅ personal exposures for the three activity days were 348 and 8717 g/m³, respectively. These exposures were 6 and 17 times, respectively, as high as the background concentration, where the background concentration was a 2-h average indoor concentration collected before the activities began each day. The mean 5-h PM_2.5 and PM₅ indoor concentrations for the same periods were 269 and 6229 g/m³. As a comparison, PM_2.5 and PM₅ indoor concentrations for the 2 days when the investigators were away from the home were 7 and 11 g/m³. Mean outdoor PM_2.5 and PM₅ concentrations increased slightly from 7 and 12 g/m³ during the human activity periods to 9 and 15 g/m³ for the days when the investigators were away from the home.

The 3-min average personal exposures during the activity periods were often much higher than the 5-h average personal exposure for the entire activity and nonactivity period. As shown in Figure 3, a peak 3-min PM₅ personal exposure of 305 g/m³ occurred when dry dusting a bookshelf that was not regularly dusted. Two persons walking around resulted in a peak 3-min PM₅ concentration of 131 g/m^-3. In comparison to the 3-min peaks, the mean PM₅ personal exposure for the period from noon to 17:00 was much lower at 67 g/m³. The significance of these short-duration, high-peak exposure events is uncertain because health impacts of exposure at this time scale are currently unknown.

Personal Cloud Observations

The observation that personal exposures were consistently higher than indoor concentrations supports the hypothesis that the personal cloud observed in previous studies is at least partially a result of resuspended particles from human activities. This phenomenon is especially evident from the dry dusting activity in Figure 3, where the personal exposure shows a high peak in concentration that is not measured by the indoor monitor. The mean 5-h PM_2.5 and PM₅ PEM/SIM ratios for all 3 days when prescribed activities took place were 1.4 and 1.6, respectively. The 5-h averaging period includes the experimental setup, prescribed activity periods, and nonactivity periods.

The personal cloud resulting from human activity was found to be a function of both particle size and the distance of the activity from the SIM. Figure 3 plots the mean PEM/SIM ratio by particle size range for all 10 activity periods performed for this study. For comparison, the results of a previous study for which six vacuuming experiments were performed in a residence are also plotted in Figure 4 (Ferro et al., 1999). Both studies collected particle counts using the same Met-One Model 237B laser particle counters, and the PEM/SIM ratios are ratios of volume concentration estimates calculated using Eq. (1). Figure 4 indicates that the PEM/SIM ratio increases with increasing particle size during periods of human activity, ranging from near 1 for the 0.3–0.5 m size fraction to more than 2 for the >5 m size fraction. The results agree with findings by Brauer et al. (1999), who measured personal and area counts using light scattering particle counters for a subject walking about a small dormitory room. They found PEM/SIM ratios of approximately 1.1 for the 0.3–1.0 m size range, 1.3 for the 1.0–5 m size range, and 1.8 for the greater than 5 m size range.

Figure 4.

PEM/SIM ratio by particle size during human activity periods for two independent studies. Ferro et al. (1999) collected personal and indoor concentrations using Met-One Model 237B laser particle counters for six vacuuming experiments in a separate Redwood City, CA, USA, residence.

Full figure and legend (27K)

Because human activity resuspends primarily coarse PM and the PEM/SIM ratio increases with increasing particle size, the personal cloud from resuspending human activities is dominated by particles in the coarse fraction. These results demonstrate that larger particles, which have a greater settling velocity, are able to deposit out of the air between the source and the SIM more efficiently than smaller particles. However, a measurable personal cloud for fine particles still exists. These findings have potential health implications since the particle size affects deposition in the respiratory system.

To determine the effect that the proximity of an activity to the SIM location has on the personal cloud estimates, activities that took place on the first floor of the home were divided into activities that took place near (0–5 m) the SIM versus further away (approximately 10 m) from the SIM. For the near-SIM activities, the investigator performing the activity was primarily in the same room as the SIM, while for the farther-away activities, the investigator was primarily in a room adjacent to the SIM with the door between the rooms fully open. The mean PM_2.5 and PM₅ PEM/SIM ratios for these 15-min activities, given in Table 2, show that the personal cloud is more than twice as large for the farther-away activities as it is for the near-SIM activities. These results indicate that the PEM/SIM ratio increases with increasing distance between the activity and the SIM. This finding is consistent with a previous study that documented a pronounced proximity effect for gas and particle concentrations from a point source in a residence (McBride et al., 1999).

Table 2 - PM_2.5 and PM₅ PEM/SIM ratios for first-floor human activities by distance from SIM location.

Full table

Discussion

The amount of increase in indoor concentration and human exposure from a given indoor human activity is dependent upon a number of factors. As shown in Eq. (2), a general mass balance for a well-mixed indoor volume, the indoor air concentration C_i(t) (g/m³) is a function of the air exchange rate a (h^-1), the penetration efficiency p (dimensionless), the outdoor concentration C_i(t) (g/m³), the indoor decay rate k (h^-1), the emission rate (source strength) of the indoor sources S(t) (g/h), and the indoor mixing volume v (m³):

Each of these factors can vary from residence to residence as well as spatially and temporally within one residence. For example, the air exchange rate is dependent upon factors such as the construction of the residence, outdoor wind velocity, indoor/outdoor temperature difference, forced ventilation conditions, and window and door openings. Similarly, the source strength for a person walking across the floor is dependent upon factors such as the dust loading on the floor, type of flooring, vigor of the activity, and relative humidity. As shown in Eq. (2), a lower source strength does not necessarily result in a lower personal exposure. For example, the source strength for the "one person walk" activity in the basement was actually lower than the same activity on the first floor (Ferro et al., 2004), while the air exchange rate in the basement was twice as high as that on the first floor. However, because the mixing volume of the basement is less than one-third as large as the mixing volume of the first floor, the activity resulted in a higher human exposure and indoor concentration in the basement.

Despite the number of factors affecting indoor concentration and human exposure from indoor sources, the results of indoor concentrations from this study appear to be within the range of previously reported values. This study found that the dry dusting and one person walking around contributed 32 and 15 g/m³, respectively, to indoor PM_2.5 concentrations. In a study of nine homes in the Boston area, Long et al. (2000) found that dusting and vigorous walking contributed 23 (SD=23) and 12 (SD=9.1) g/m³, respectively, to indoor PM_2.5 concentrations. Also, this study found that dancing on a rug produced personal PM_2.5 concentrations 6 times as high as dancing on a wood floor. By comparison, dancing on a rug produced stationary indoor concentrations only 2.4 times as high as similar activities on wood flooring. This result agrees with Long et al. (2000), who found that activities on a carpet produced indoor PM_2.5 concentrations 2 times as high as similar activities on wood or linoleum. For this study, peak 3-min PM₅ concentrations during 15-min activities of vacuuming the first floor, vacuuming the bedrooms, walking on the first floor, and walking in the basement were 81, 69, 98, and 165 g/m³, respectively. Ferro et al. (1999) reported peak 2-min PM₅ concentrations measured by a piezobalance of 90 g/m³ during 30-min vacuuming activity.

The PEM/SIM ratios for human activity presented in this paper are also in the range of those reported by previous studies. For this study, the mean 5-h PM_2.5 and PM₅ PEM/SIM daytime ratios were 1.4 and 1.6, respectively. Rodes et al. (1991) reported median PEM/SIM ratios ranging from 1.6 to 13.4 for five studies of various types of particles and microenvironments. Of these studies, only the EPA PTEAM study compared PEM concentrations with indoor residential SIM concentrations, resulting in a 12-h median PEM/SIM ratio for PM₁₀ of 1.98 (N=283). Unlike the present study, the PTEAM subjects performed normal nonscripted human activities and were not necessarily at home or in the same room as the PEM during the monitoring period. As discussed previously, Brauer et al. (1999) reported PEM/SIM ratios for a subject walking for four 15-min activity periods that ranged from approximately 1.1 for the <1 m size range to 1.8 for the >5 m size range.

This study examines human exposures to PM from 10 prescribed indoor human activities, several of which have not been reported before in scientific literature. The results of this study indicate that house dust resuspended from a range of human activities increases personal PM concentrations over indoor PM concentrations and this resuspension effect significantly contributes to the personal cloud. The results of this study also suggest that normal human activities that resuspend house dust may contribute significantly to the strong correlations found between personal exposure and indoor PM concentrations in previous studies. Further characterization of indoor human activities is necessary to develop the relationships between human exposures and factors such as dust loading, type of flooring and furnishings, and particle size. Further research is also required to determine whether these short-duration, high-peak exposure events associated with indoor sources are unhealthful.

References

1. Brauer M., Hirtle R.D., Hall A.C., and Yip T.R. Monitoring personal fine particle exposure with a particle counter. J Expos Anal Environ Epidemiol 1999: 9: 228–236. | Article |
2. Delfino R.J., Zeiger R.S., Seitzer J.M., and Street D.H. Symptoms in pediatric asthmatics and air pollution: Differences in effects by symptom severity, anti-inflammatory medication use and particulate averaging time. Environ Health Perspect 1998: 106: 751–761. | PubMed |
3. Dockery D.W., and Spengler J.D. Personal exposure to respirable particulates and sulfates. J Air Pollut Control Assoc 1981: 31: 153–159. | PubMed |
4. Ferro A.R., Hildemann L.M., McBride S.J., Ott W., and Switzer P. Human exposure to particles due to indoor cleaning activities. In: C.A. Brebbia et al. (Eds.), Air Pollution VII. WIT Press: Southampton, UK, 1999: pp. 487–496.
5. Ferro A.R., Kopperud R.J., and Hildemann L.M. Source strengths for indoor human activities that resuspend particles. Environ Sci Technol 2004: in press.
6. Gold D.R., Litonjua A., Schwartz J., Lovett E., Larson A., Nearing B., Allen G., Verrier M., Cherry R., and Verrier R. Ambient pollution and heart rate variability. Circulation 2000: 101: 1267–1273. | PubMed | ISI | ChemPort |
7. Janssen N.A.H., Hoek G., Harssema H., and Brunekreef B. Childhood exposure to PM₁₀: Relation between personal, classroom and outdoor concentrations. Occup Environ Med 1997: 54: 888–894. | PubMed | ISI | ChemPort |
8. Janssen N.A.H., Hoek G., Brunekreef B., Harssema H., Mensink I., and Zuidhof A. Personal sampling of PM₁₀ among adults: relation between personal, indoor and outdoor concentrations. Am J Epidemiol 1998: 147: 537–547. | PubMed | ISI | ChemPort |
9. Janssen N.A.H., Hoek G., Harssema H., and Brunekreef B. Personal exposure to fine particles in children correlates closely with ambient fine particles. Arch Environ Health 1999: 54: 95–101. | PubMed | ISI | ChemPort |
10. Janssen N.A.H., Hartog J.J., Hoek G., Brunekreef B., Lanki T., Timonen K.L., and Pekkanen J. Personal exposure to fine particulate matter in elderly subjects: relation between personal, indoor, and outdoor concentrations. Air Waste 2000: 50: 1133–1143.
11. John W., and Reischl G. A cyclone for size-selective sampling of air. J Air Pollut Control Assoc 1980: 30: 872–876.
12. Lioy P.J., Waldman J.M., Buckley T., Butler J., and Pietarinen C. The personal, indoor and outdoor concentrations of PM-10 measured in an industrial community during the winter. Atmos Environ 1990: 24B: 57–66.
13. Long C.M., Suh H.H., and Koutrakis P. Characterization of indoor particle sources using continuous mass and size monitors. Air Waste 2000: 50: 1236–1250.
14. McBride S.J., Ferro A., Ott W., Switzer P., and Hildemann L.M. Investigations of the proximity effect for pollutants in the indoor environment. J Expos Anal Environ Epidemiol 1999: 9: 602–621. | Article |
15. Morgan G., Corbett S., and Wlodarczyk J. Air pollution and hospital admissions in Sydney, Australia, 1990 to 1994. Am J Public Health 1998: 88: 1761–1766. | PubMed | ISI | ChemPort |
16. Özkaynak H., Xue J., Weker R., Butler D., Koutrakis K., and Spengler J. The Particle Team (PTEAM) study: Analysis of the data. EPA Project Summary, EPA/600/SR-95/098, U.S. EPA: Research Triangle Park, NC, 1996.
17. Peters A., Dockery D.W., Muller J.E., and Mittleman M.A. Increased particulate air pollution and the triggering of myocardial infarction. Circulation 2001: 103: 2810–2815. | PubMed | ISI | ChemPort |
18. Roberts J.W., and Dickey P. Exposure of children to pollutants in house dust and indoor air. Rev Environ Contam Toxicol 1995: 143: 59–78. | PubMed |
19. Rodes C.E., Kamens R.M., and Wiener R.W. The significance and characteristics of the personal activity cloud on exposure assessment measurements for indoor contaminants. Indoor Air 1991: 2: 123–145. | Article |
20. Robinson J.P., Thomas J., and Behar J.V. Time spent in activities, locations, and microenvironments: A California-national comparison. EPA Report EPA/600/4-91/006, U.S. EPA: Las Vegas, NV, 1991.
21. Samet J.M., Dominici F., Curriero F.C., Coursac I., and Zerger S.L. Fine particulate air pollution and mortality in 20 U.S. Cities, 1987–1994. N Engl J Med 2000: 343(24): 1742–1749. | Article | PubMed | ISI | ChemPort |
22. Smith K.R. Fuel combustion, air pollution exposure, and health: the situation in developing countries. Annu Rev Energy Environ 1993: 18: 529–566. | Article |
23. Thatcher T.L., and Layton D.W. Deposition, resuspension, and penetration of particles within a residence. Atmos Environ 1995: 29(13): 1487–1497. | Article | ISI | ChemPort |
24. Vedal S. Ambient particles and health: Lines that divide. Air Waste 1997: 47: 551–581.
25. Wallace L. Indoor particles: A review. Air Waste 1996: 46: 98–126.
26. Whitby K.T., and Willeke K. Single particle optical counters: principles and field use. In: Lundgren D.A. (Ed.), Aerosol Measurement. University Presses of Florida: Gainesville, FL, 1979.

Acknowledgements

This work was supported in part by funding from the Center for Indoor Air Research and the Stanford University Shah Family Fellowship. The authors are grateful for the assistance of Wayne Ott and Victor W.C. Chang.