Ultrafiltration to characterize PM2.5 water-soluble iron and its sources in an urban environment
Yuhan Yang, Rodney J. Weber
Abstract
Soluble forms of transition metals in airborne particles are linked to adverse health impacts. Iron (Fe) is often the most abundant transition metal in ambient PM2.5. In Atlanta, the water-soluble fraction of Fe had a stronger association with adverse cardiovascular outcomes than PM2.5 mass and total Fe. Water solubility is operationally defined. One determining factor is the pore size in the liquid filtration (0.45 μm in this study) used to isolate water-soluble species, implying colloidal particles smaller than the pore size may be included in the water-soluble group. Here, we characterize the properties of PM2.5 water-soluble iron (WS Fe) utilizing ultrafiltration and investigate sources of the various isolated fractions based on one-year of samples collected in urban Atlanta (N = 355). Annual average Fe solubility (WS Fe divided by total Fe) was 12%, with the highest solubility in the warm season (May–Sept.) when aerosol acidity (pH of 1.8–2.7) and oxalate concentrations were highest. Ultrafiltration showed that 84% of WS Fe (extract that passed through 0.45 μm pore filter) was smaller than nominally 4 nm, with 61% of the soluble iron was smaller than nominally 2 nm. A correlation analysis, experiments involving addition of oxalate to sample extract and a pH cycling experiment with selected Fe species were consistent with the oxalate-Fe ligand being mainly found in the smallest ultrafiltrate size (< nominally 2 nm), whereas pH cycling could lead to the formation of colloidal particles with sizes between nominally 2–4 nm. Magnetite, a solid Fe nanoparticle (smaller than 0.1 μm) detected in human organs in previous studies, was undetectable in the PM2.5 WS Fe in urban Atlanta. These data show that most WS Fe is composed of soluble species or very small colloidal particles (roughly <4 nm), making them highly bioavailable.