Heterogeneous Chemistry of Reactive Species on Aerosol Surfaces

A photo-fragmentation laser induced fluorescence instrument coupled to an aerosol flow-tube was validated and calibrated for the measurement of HONO from sub-micron aerosol surfaces. The production of HONO from illuminated TiO2 aerosol surfaces in the presence of NO2 was investigated and the reactive uptake coefficient of NO2 to form HONO for NO2 mixing ratios of 34 – 400 ppb was found to be in the range of γ(NO2→HONO)= (9.97 ± 3.51) × 10-6 at 286 ppb to a maximum of γ(NO2→HONO) = (1.26 ± 0.17) × 10-4 at 51 ppb NO2 for a lamp photon flux of (1.65 ± 0.02) × 1016 photons cm-2 s-1. The reactive uptake of NO2 to form HONO on TiO2 was found to increase with NO2 mixing ratio to ~ 51 ppb NO2, followed by a decrease with further increase in NO2. Box modelling studies supported a mechanism involving two NO2 molecules per HONO molecule formed, suggesting the formation of an NO2 dimer intermediate on the surface of the aerosol. Using the maximum experimental value of γ(NO2→HONO), a production rate of HONO for Beijing Summertime was estimated to be 1.70 × 105 molecules cm-3 s-1 (~24.8 ppt hr-1). From the calculated net gas phase production of HONO for the same conditions (~3.8 ppb hr-1) it was concluded that production of HONO from TiO2 aerosol surfaces in the presence of NO2 would have little effect on the overall HONO budget for Summertime in Beijing. In the absence of NO2, the production of HONO from ammonium nitrate aerosols was investigated due to recent studies proposing particulate nitrate photolysis as an important source of HONO and hence NOx following photolysis (renoxification) in the marine boundary layer. Significant production of HONO from pure ammonium nitrate aerosols was not seen. However, with the addition of TiO2 in a 1:1 mixture with ammonium nitrate, the production of HONO observed was significant. Using experimental results for mixed TiO2/ammonium nitrate aerosol experiment, a rate of HONO production from nitrate photolysis for ambient marine conditions was calculated to be 68 ppt hr-1, taking Cape Verde in the tropical Atlantic Ocean as an example of such environments. This is a significant rate of production of HONO, with a magnitude similar to the missing HONO production rate calculated in the RHaMBLe campaign which took place at Cape Verde in 2007. A Potential Aerosol Mass chamber was built and characterised for the production of secondary organic aerosols (SOA) from oxidation of volatile organic compounds (VOC). The effect of changing the concentrations of limonene and O3 on the size distributions of SOA was investigated, with size distributions for SOAs derived from α-pinene and limonene with both O3 and OH as the oxidants compared. The PAM chamber was coupled to an aerosol flow tube system and was validated against previous observations for the measurement of HO2 uptake onto aerosol surfaces. The uptake of HO2 onto limonene-derived SOA was measured with an interference observed, postulated to be due to RO2 species coming off the SOA surface after the charcoal denuders. Trace amounts of limonene after the denuders and OH from the sliding injector (formed as a bi-product of water photolysis for the generation of HO2) were also found to be present within the flow tube leading to a production of peroxy radicals as the injector was pulled back due to VOC + OH reactions. With the addition of a Gas Chromatography (GC) trap after the denuders which significantly reduced any interference seen, though to an unknown extent, a decay of HO2 in the presence of limonene-derived SOA was observed and uptake coefficients of γ(HO2)= 0.10 ± 0.03 and γ(HO2 )= 0.15 ± 0.03 were measured for 22 % and 58 % RH respectively at atmospheric pressure and 298 K. Using a novel parameterisation by Song et al., 2020, the uptake coefficient of HO2, γ(HO2), was calculated using measured values of temperature, RH, aerosol pH, aerosol liquid water content (ALWC), particulate matter concentration ([PM]) and [Cu2+]eff for Beijing Summer AIRPRO campaign in 2017. For the entire campaign, the average HO2 uptake coefficient calculated was 0.070 ± 0.035 (1σ) with values ranging from as low as 0.002 to as high as 0.15, though still significantly lower than the value of γ(HO2)= 0.2, commonly used in modelling studies. Using the calculated values of γ(HO2) for the Summer AIRPRO campaign, the OH, HO2 and RO2 radical concentrations were calculated using a box-model incorporating the Master Chemical Mechanism (v. 3.3.1), with and without the addition of aerosol uptake, and compared to the measured concentrations. The effect of HO2 uptake onto aerosols was investigated using 2 models: MCM_αpinene_SA with γ(HO2) fixed at 0.2, and MCM_αpinene_gamma with the values of γ(HO2) calculated by the Song parametrisation. A rate of destruction analysis showed the dominant loss pathway for HO2 to be HO2 + NO for all NO concentrations across the campaign with HO2 uptake contributing < 0.3 % to the total loss of HO2, as expected for a polluted urban location. However, at low NO r, i.e. < 0.1 ppb NO, up to 78 % of HO2 loss was due to HO2 uptake within MCM_αpinene_SA (γ(HO2)= 0.2) and up to 28 % of HO2 loss was due to HO2 uptake within MCM_αpinene_gamma, despite the much lower γ(HO2) values compared to γ(HO2) of 0.2. From this, it can be concluded that in cleaner environments away from urban centres with high concentrations of NO but where aerosol surface area is high still, values of γ(HO2) of less than 0.2 could still have a significant effect on the overall HO2 concentration.