10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Accuracy of Sensors in Assessing Exposure to Traffic-Related Aerosols
JENNIE COX, Seung-Hyun Cho, Sergey A. Grinshpun, James Ross, Steven Chillrud, Zheng Zhu, Roman Jandarov, Tiina Reponen, University of Cincinnati
Abstract Number: 825 Working Group: Indoor Aerosols
Abstract Traffic-related airborne particles (TRAP) penetrate into buildings affecting the indoor air quality. Improved methods are needed for rapid and accurate assessment of TRAP and the efficiency of control methods. In this study, we evaluated two novel sensors: RTI’s Micro Personal Exposure Monitor (MicroPEM) and the MicroAeth real-time black carbon (AE51) sensor. It is recognized in the field that in order to obtain credible real-time measurement results, the data from sensors should be corrected utilizing the information from filter analysis which requires additional time and money. The objective of this study was to create universal correction factors for the MicroPEM and MicroAeth sensors to provide accurate real-time data on PM2.5 and black carbon (BC) in Cincinnati metropolitan area homes, as an alternative to correcting each dataset with the respective filter weights. The two sensors and a PM2.5 impactor (Personal Modular impactor; SKC, Inc.) were collocated in 27 indoor sampling events for 2 days in residences <500 meters from a major roadway and/or had a TRAP score (based on a land-use regression model) of at least 0.33. The PM2.5 impactor samples were analyzed for BC by multi-wavelength integrative sphere method which was gravimetrically calibrated. The real-time PM2.5 concentration from the MicroPEM (RTPM) was averaged and corrected using the gravimetric concentration from the MicroPEM filter. The BC results from the MicroAeth real-time sensor were corrected utilizing BC results from the PM2.5 impactor. Subsets of independent variables in an analysis procedure utilizing multiple linear regression were selected based on lowest Akaike information criterion (AIC). Variables in the model included age and type of building, type of flooring, type of heat (gas or electric), number of smokers, TRAP score, the average daily truck counts within 400 meters, distance to state highways, and distance to a major road (state highways or federal interstates with an average daily truck count of more than 1000). Additional variables of conditions during sampling included season, month, frying food, wood or candle burning, and window opening. The two-day PM2.5 concentrations obtained with the MicroPEM filter ranged from 0.4 to 47.1 µg/m3 with a median of 9.5 µg/m3. The average correction factor (CF) (MicroPEM filter/RTPM two-day mean) was 1.51 with a coefficient of variation (CV) of 149% (n=27). For the MicroPEM, the final regression model included RTPM values, the average daily truck counts, distance to state highways and distance to federal interstates. By including these variables into the model, the CV was reduced to 40%. The BC values from the PM2.5 impactor ranged from 0.03 to 3.1 µg/m3 with a median of 0.4 µg/m3. The 48-hour average values from the MicroAeth BC sensor ranged from 0.1 to 2.6 µg/m3 with a median of 0.3 µg/m3. The average CF (SKC filter/MicroAeth) was 1.3 with a CV of 46% (n=25). Linear regression models considered the variety of variables, but the best model (CV = 49%) did not reduce the CV compared to the correction factor, indicating that these variables are not predictive of indoor concentrations for BC. The poor performance of the models could be due to small sample size and not including other potentially meaningful variables. Still, the PM2.5 model identified three significant factors that indicate the importance of home characteristics for predicting the correction factor. The results from this study will help ensure that the real-time exposure monitors are capable of accurately detecting PM2.5 and BC, which creates the foundation for the further development of user-friendly, field-compatible aerosol instruments for indoor air quality monitoring.