The influence of air cleaners on indoor particulate matter components and oxidative potential in residential households in Beijing.

Abstract

In many developing regions with poor air quality, the use of air filtration devices to clean indoor air is growing rapidly. In this study, we collected indoor, outdoor and personal exposure filter-based samples of fine particulate matter (PM2.5) with both properly operating, and sham air cleaners in six Beijing residences from July 24th to August 17th, 2016. Mass concentrations of PM2.5 and several health relevant components of PM2.5 including organic carbon, elemental carbon, sulfate, nitrate, ammonium, and 21 selected metals, were analyzed to evaluate the effectiveness of air cleaners. The effect of air purification on PM2.5 reactive oxygen species (ROS) activity, a metric of the oxidative potential of the aerosol, was also evaluated. The average indoor PM2.5 concentration during true filtration was 8.47μg/m3, compared to 49.0μg/m3 during sham filtration; thus, air cleaners can significantly reduce the indoor PM2.5 concentration to well below WHO guideline levels and significantly lower all major components of PM2.5. However, the utility of air cleaners in reducing overall personal exposure to PM2.5 and its components was marginal in this study: the average personal exposure PM2.5 concentration was 67.8 and 51.1μg/m3 during true and sham filtration respectively, and it is likely due to the activity patterns of the subjects. Short-term exposure contributions from environments with high PM2.5 concentrations, including exposure to traffic related emissions as well as uncharacterized indoor microenvironments, likely add substantially to the total PM2.5 exposure burden. The toxicity assay indicates that the air cleaners can also significantly reduce ROS activity in the indoor environment; however, this decrease did not translate to a reduction in personal exposure. Elemental carbon, lead, and arsenic were well-correlated with the ROS activity, thus adding to the knowledge base of drivers for ROS activity.

DOI
10.1016/j.scitotenv.2018.01.024
Year