Heterogeneous uptake of HO2 radicals onto atmospheric aerosols

HO2 uptake coefficients were measured onto a variety of sub-micron aerosols and over a range of
experimental conditions using an aerosol flow tube coupled with a sensitive Fluorescence Assay
by Gas Expansion (FAGE) cell. Experiments showed that the deliquesced salt aerosols had larger
HO2 uptake coefficients (γ = 0.003 – 0.016) than the effloresced salt aerosols (γ < 0.004).
Similarly, solid organics had smaller uptake coefficients (γ < 0.004) than aqueous organics which
were small (γ < 0.004 – γ = 0.008) unless metal ions were present. No observable dependence
upon aerosol size or aerosol pH was measured for aqueous salt aerosols. The mass
accommodation was measured as 0.5 ± 0.3 by doping the aerosols with copper. Measurements
also showed that the HO2 uptake coefficient was highly dependent upon the copper and iron concentrations and increased between copper concentrations of 10-4 – 10-2 M within the aerosol.
However, the addition of organics such as EDTA and oxalic acid to copper doped aerosols
decreased the HO2 uptake coefficient by a factor of ~ 50 – 100. The HO2 uptake coefficient onto
copper doped sucrose aerosols increased with increasing relative humidity. Secondary organic
aerosols were generated in situ in a smog chamber and small uptake coefficients were measured
onto α-pinene derived aerosols (γ < 0.001) and 1,3,5 trimethylbenzene derived aerosols (γ =
0.004 ± 0.002).
Measurements onto Arizona Test Dust (ATD) aerosols showed much higher HO2 uptake
coefficients (γ = 0.018 ± 0.006) than salt and organic aerosols at an initial HO2 concentration of 1 × 109 molecule cm-3 and increased with increasing humidity. Experiments were also performed
over a temperature range of 263 -313 K onto effloresced sodium chloride and ammonium
sulphate aerosols and onto deliquesced ammonium nitrate and copper doped ammonium nitrate.
For the deliquesced ammonium nitrate aerosols the HO2 uptake coefficient increased with
decreasing temperature. Finally, a time and apparent HO2 concentration dependence was
observed for aqueous salt aerosols, copper doped aqueous aerosols and for ATD with larger HO2
uptake coefficients at shorter times and at lower HO2 concentrations. Modelling was performed
using the kinetic multi-layer model of aerosol surface and bulk chemistry (KM-SUB) and showed
that for undoped aqueous aerosols the time dependence could be explained by a decrease in the
HO2 concentrations along the flow tube. The apparent HO2 concentration had the potential to be
explained by a Fenton-like reaction whereby hydrogen peroxide exiting the injector was
converted to HO2 within the aerosols, due to the presence of trace amounts of transition metal
ions, which then partitioned back to the gas phase.