Comparison of light scattering devices and impactors for particulate measurements in indoor, outdoor, and personal environments

Short-term monitoring of individual particulate matter (PM) exposures on subjects and inside residences in health effect studies have been sparse due to the lack of adequate monitoring devices. The recent development of small and portable light scattering devices, including the Radiance nephelometer(neph) and the personal DataRAM (pDR) has made this monitoring possible. This paper evaluates the performance of both the passive pDR and neph (without any size fractionation inlet) against measurements from both Harvard impactors (Hl(2.5)) and Harvard personal environmental monitors (HPEM2.5)for PM2.5 in indoor, outdoor, and personal settings. These measurements were taken at the residences and on the person of nonsmoking elderly subjects across the metropolitan Seattle area and represent a wide range of light scattering measurements directly related to exposures and health effects. At low PM levels, nephs provided finer resolution and more precise measurements (precision = 3-8% and uncertainty = 2.8 x 10(-7) m(-1) or <1 mug/m(3)) than the pDRs. The unbiased precision of pDRs above 10 mug/m(3) is around 5% (with an unbiased uncertainty of 4.4 mug/m(3)). The 24-h average responses of the pDR and neph, as compared to 24-h integrated gravimetric measurements, are not affected by indoor sources of PM. When regressed against 24-h gravimetric measurements, nephs showed higher coefficients of determination (R-2 = 0.81-0.93) than pDRs (R-2 = 0.77-0.84). The default mass calibration on the pDRs generally overestimated indoor Hl(2.5) measurements by 56%. When carried by subjects, the pDR overestimated the HPEM2.5 measurements by approximately 27%. Collocated real-time indoor nephs and pDRs at diverse residential sites had varied coefficients of determination across homes (R-2 = 0.75-0.96), and the difference between pDR and neph responses increased during cooking hours. This difference was larger during baking or frying episodes than during other cooking or cleaning activities. Relative humidity, ranging between 25% and 64% indoors in our study, was not a significant factor affecting the differences in neph or pDR response. In summary, for nonsmoking residences, the mass scattering efficiency (m(2)/g) of a stationary indoor neph on a 24-h basis does not vary by residence, including residences with and without cooking activities. This is also true for the pDR. These same stationary indoor pDRs and nephs correlate well with each other, even on a 10-min basis, in the absence of indoor source activities. The fact that these activities comprised a relatively small percentage (cooking + cleaning = 2.3%) of the overall sampling time meant that the overall correlation between these two instruments for all time periods was good. However, when examining the cooking and cleaning periods separately, the correlation was not very good. Thus, during these short-term PM episodes,the 24-h average calibrations versus gravimetric mass should be used with caution. Both devices should be potentially useful in future exposure assessment and health effects studies.