Effects of aerosol collection and extraction procedures on the optical properties of water-soluble organic compounds

Water-soluble organic compounds (WSOC) are routinely collected using active and passive aerosol samplers and, after extraction in water, analyzed using UV-vis absorbance and fluorescence techniques. These analyses provide important information regarding the chemical character and sources of aerosols worldwide. To evaluate the effects of various aerosol collection and processing methods on the optical properties of WSOC, two-dimensional absorption spectra from 200 to 900 nm and three-dimensional fluorescence excitation-emission spectra (EEMs) from 240 to 450 nm excitation and 300 to 560 nm emission were analyzed in samples obtained simultaneously with different procedures. Samples included: milli-Q purified water passed through 140 mm diameter glass fiber and quartz fiber filters used in high volume PM10 aerosol samplers, 47 mm glass fiber filters used for organic matter analyses, and mixed cellulose 0.2 micron and 0.015 micron filters used for bacterial and viral filtration, respectively; milli-Q purified water rinsed in plastic buckets used for passive wet and dry deposition collection; and WSOC samples extracted from filters by soaking, sonication, and agitation. Parallel factor analysis (PARAFAC) modeling of WSOC was performed to quantify the influence of various collection and extraction procedures on fluorescence signatures. All filters examined were found to leach some amount of fluorescent compounds (Figure 1). Mixed cellulose filters, especially those with small pore size, leached substantially more amino acid-like and humic-like material than other filters, whereas leaching from quartz fiber filters used for high volume aerosol collection was minimal (Figure 1). Fluorescence intensities of filter leachates decreased with increased rinsing of filters, indicating that rinsing with purified water prior to filtration is advisable, even for pre-combusted filters. Dissolved organic carbon concentrations of WSOC extracted from filters by sonication, agitation, and soaking techniques were not statistically different. The WSOC of aerosol samples collected passively (as deposition) and actively (on filters) had similar fluorescence spectra. Our results suggest that WSOC studies may be highly comparable irrespective of collection procedures, but that methodological differences related to filter use can be significant and should be reported in detail in the literature. Figure 1. Excitation-emission matrices (EEMs) of a WSOC sample extracted from a pre-combusted quartz fiber filter (left) by agitation in 60 mL milli-Q purified water and a 100 mL sample of milli-Q purified water filtered through a 0.015 μm pore size mixed cellulose ester filter (right).