The underappreciated role of nonvolatile cations in aerosol ammonium-sulfate molar ratios

Abstract

Abstract. Overprediction of fine-particle ammonium-sulfate molar ratios (R) by thermodynamic models is suggested as evidence for interactions with organic constituents that inhibit the equilibration of gas-phase ammonia with aerosol sulfate and questions the equilibrium assumption long thought to apply for submicron aerosol. This hypothesis is tested through thermodynamic analysis of ambient observations. We find that the deviation between R from a molar ratio of 2 is strongly correlated with the concentration of sodium (Na+), a nonvolatile cation (NVC), but exhibits no correlation to organic aerosol (OA) mass concentration or mass fraction. Thermodynamic predictions of both R and ammonia gas–particle partitioning can accurately reproduce observations when small amounts of NVCs are included in the calculations, whereas exclusion of NVCs results in a predicted R consistently near 2. The sensitivity of R to small amounts of NVCs arises because, when the latter are present but not included in the thermodynamic calculations, the missing cations are replaced with ammonium in the model (NH3–NH4+ equilibrium shifts to the particle), resulting in an R that is biased high. Results and conclusions based on bulk aerosol considerations that assume all species are internally mixed are not changed even if NVCs and sulfate are largely externally mixed; fine-particle pH is found to be much less sensitive to mixing state assumptions than molar ratios. We also show that the data used to support the “organic inhibition” of NH3 from equilibrium, when compared against other network and field campaign datasets, display a systematically and significantly lower NH4+ (thought to be from an evaporation bias), that is of the order of the effect postulated to be caused by organics. Altogether, these results question the postulated ability of organic compounds to considerably perturb aerosol acidity and prevent ammonia from achieving gas–particle equilibrium, at least for the locations considered. Furthermore, the results demonstrate the limitations of using molar ratios to infer aerosol properties or processes that depend on particle pH.