Development of a wet-chemical method for the speciation of iron in atmospheric aerosols.

The ability to quantify the chemical and physical forms of transition metals in atmospheric particulate matter (PM) is essential in determining potential human health and ecological effects. A method for the speciation of iron in atmospheric PM has been adapted which involves extraction in a well-defined solution followed by oxidation state specific detection. The method was applied to a suite of environmental aerosols. Ambient atmospheric aerosols in an urban area of St. Louis (the St. Louis-Midwest Supersite) were collected on Teflon substrates and were leached in one of four different solutions: (1) >18.0 Momega water; (2) 140 microM NaCl solution; (3) pH = 7.4 NaHCO3 solution; and (4) pH = 4.3 acetate buffering system. Fe(ll) was determined directly using the Ferrozine method as adapted to liquid waveguide spectrophotometry using a 1 m path-length cell. Fe(lll) was determined similarly after reduction to Fe(ll). It was found that, at low ionic strength, pH exerted a major influence on Fe(ll) solubility with the greatest Fe(ll) concentration consistently found in the pH = 4.3 acetate buffer. Soluble Fe(lll) (as defined by a 0.2 microm filter) varied little with extractant, which implies that most of the Fe(lll) detected was colloidal. To characterize well-defined materials for future reference, NIST standard reference materials were also analyzed for soluble Fe(ll) and Fe(lll). For all SRMs tested, a maximum of 2.4% of the total iron (Urban Dust 1649a) was soluble in pH = 4.3 acetate buffer. For calibration curves covering the ranges of 0.5-20 microg/L Fe(ll), excellent linearity was observed in all leaching solutions with R2 values of > 0.999. Co-located filters were used to test the effect of storage time on iron oxidation state in the ambient particles as a function of time. On two samples, an average Fe(ll) decay rate of 0.89 and 0.57 ng Fe(ll) g(-1) PM day(-1) was determined from the slope of the regression, however this decrease was determined not to be significant over 3 months (95% confidence). As an application of this method to mobile source emissions, size-resolved PM10 samples were collected at the inlet and outlet of the Caldecott Motor Vehicle Tunnel in northern California. These samples indicate that the coarse fraction (PM10-PM2.5) contains almost 50% of the total soluble Fe(ll) in the aerosol.