Particle water and pH in the southeastern United States

H. Guo, L. Xu, A. Bougiatioti, K. M. Cerully, S. L. Capps, J. R. Hite, A. G. Carlton, S.-H. Lee, M. H. Bergin, N. L. Ng, A. Nenes, and R. J. Weber School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA, USA School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, GA, USA School of Civil & Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA Office of Research and Development, United States Environmental Protection Agency, Research Triangle Park, NC, USA Department of Environmental Sciences, Rutgers University, New Brunswick, NJ, USA College of Public Health, Kent State University, Kent, Ohio, USA Foundation for Research and Technology, Hellas, Greece National Technical University of Athens, Athens, Greece now at: TSI, Inc., Shoreview, MN, USA


Supplementary materials 1. Ion balance vs
An ion balance (or charge balance) of an aerosol is usually calculated as follows, for a NH 4 -SO 4 -NO 3 -Cl-Na-water inorganic aerosol.
(1) where , , , , and are concentrations of these ions in the aerosol in units of g m -3 . An ion balance cannot be used as a proxy for the aerosol concentration of in air (i.e., moles per volume of air, denoted hereafter as ). For example, if sulfate stays in the free form of H 2 SO 4 , it doesn't add protons, but a charge balance simply assumes complete dissociation. Ammonium to sulfate molar ratios are also often used to assess particle acidity. In this study the / mole ratio is 1.4 ± 0.5 (mean ± STD). Orthogonal distance regression (ODR) fits were applied.

Nephelometer RH sensor calibration
The nephelometer RH sensors were calibrated by placing the sensors in a properly sealed container above aqueous saturated salt solutions at known temperatures allowing accurate prediction of equilibrium RH (Greenspan, 1977). More than 3hours were allowed for each salt solution to reach water vapor saturation.

LWC measurement principle by nephelometers
Particle water was indirectly measured by two nephelometers. The difference between ambient and dry scattering coefficients ( ) is assumed to be caused by the loss of water. The ratio between wet or ambient scattering coefficient and dry scattering coefficient is referred to as ( ).
where is scattering efficiency, ( ) is particle number size distribution, and is the total number concentration. Assuming that ̅̅̅̅ is the diameter of average surface, Based on Mie Theory calculations, over fine particle (PM 2.5 ) size range is plotted in the Figure 2. It shows the optimal scattering size range (570-680 nm) with highest scattering efficiencies for accumulation mode particles. The optimal scattering size range is supposed to be covered by both wet and dry particles. Assuming and are of similar magnitude, LWC is equal to the differences between wet particle volume and dry particle volume.

̅̅̅̅̅̅̅̅ ̅̅̅̅̅̅̅̅
Since ̅̅̅̅̅̅̅̅ where is dry PM 2.5 mass concentration and is the density of dry aerosol. Take Equation 5 and 7 into Equation 6. Finally we get, measured by a TEOM was used. was estimated from the particle composition including AMS total organics, ammonium, and sulfate, which accounted for 90% of the measure PM 2.5 dry mass (TEOM) on the SOAS study average. A typical organic density 1.4 g cm -3 is assumed (Turpin and Lim, 2001;King et al., 2007;Engelhart et al., 2008;Kuwata et al., 2012;Cerully et al., 2014), and the density of ammonium sulfate is assumed to be 1.77 g cm -3 (Sloane et al., 1991;Stein et al., 1994). is calculated to be 1.49 ± 0.04 g/cm 3 (n = 4,393) using mass fractions ( ) by Equation 9. The diurnal variation of aerosol density is shown in Figure 3. . S4. Predicted PM 2.5 density diurnal profile. Median hourly density averages and standard error plotted as error bars (too small to be seen) at local hour are plotted.

Particulate organic hygroscopic parameter,
Overall has a study mean (±STD) as 0.126 ± 0.059 (Cerully et al., 2014). However, data is only available after June 20th, and were not measured during the first 20 days of SOAS field study. Diurnal hourly average was calculated and median values are plotted in Figure 4. Because median averages were scattered, three hours running averages were obtained.

Filter based IC analysis at all sites
High-volume PM 2.5 filters, sampled parallel to AMS measurement, were analyzed by a DIONEX IC (UTAC-ULP1 concentrator column, AG11 guard column and AS11 anion column) to provide chemical information of the refractory ions (Na, K, Mg, Ca, PO 4 ). Filter-based nitrate is excluded due to artifacts (Hering and Cass, 1999;Chang et al., 2000). The ion compositions at various sites are similar, all dominated by SO 4 (64-74%) and then NH 4 (22-31%). We found the refractory ions usually accounted for < 5% (RS is the highest at 9%) of the sum of all ion inorganic ion mass for PM 2.5 .