Understanding radical chemistry in Beijing through observations and modelling

In Beijing, poor urban air quality has a demonstrable effect on human health; with high aerosol loadings during winter and high ozone episodes during summer. The hydroxyl radical (OH) mediates virtually all the oxidative chemistry in the atmosphere, being responsible for the transformation of primary emissions into secondary pollutants such as NO2, O3 and SOA (secondary organic aerosol). Comparison of measured radicals with results from detailed chemical box models serves as a vital tool to assess our understanding of the underlying chemical mechanisms involved in tropospheric oxidation. A recent comparison of radical measurements with those predicted by box models highlight missing understanding at both high and low NO mixing ratios.
The Leeds FAGE (fluorescence assay by gas expansion) instrument was deployed in Beijing for two field campaigns, one in the winter in November/December 2016 and the other in summer in May/ June 2017. The chemical conditions varied vastly between the two campaigns, with NO concentrations exceeding 250 ppb in the winter, while O3 levels over 100 ppbv were frequently observed during the summer. The average OH concentration during the winter campaign was high (~ 2.5 x 106 cm-3) even during haze events, and during the summer elevated levels of OH were observed, reaching up to 2.5 x 107 cm-3.
The OH measurements were compared with the OH calculated using a photostationary steady-state (PSS) calculation and showed that in winter the experimental budget could be closed, but in the summer the PSS calculation underpredicts OH by a factor of ~2 and highlights a missing source of OH. The measured radicals have been compared to results from a box model incorporating the Master Chemical Mechanism (MCM, v3.3.1). The model underpredicts OH, HO2 and RO2 in winter, and the underprediction increases with increasing NO. By contrast, in summer the model can replicate OH very well, but overpredicts HO2 and underpredicts RO2. Several model scenarios have been performed to investigate the differences between the measured radical values and the model results. In winter, the model can reproduce the measured OH reactivity. When the model is constrained to measured HO2, the model can reproduce OH, which suggests the missing understanding is linked to RO2 radical chemistry. In summer the model can reproduce the OH, HO2 and RO2 radical concentration when the RO2 + NO rate constant is reduced by a factor of ~10, and may suggest that the RO2 species are undergoing autoxidation forming HOMs (highly oxidised molecules) which is competing with RO2 + NO reaction. HOMs species were also observed during the summer campaign in both the gas phase and aerosol phase by a CIMS (chemical ionisation mass spectroscopy) instrument. The impact of HO2 uptake onto aerosols on both modelled HO2 and calculated O3 production was assessed for both the winter and summer campaign. It shows that HO2 uptake decreases O3 by up to ~6 ppbv hr-1 in summer; while there is a much smaller impact in winter as the reaction of HO2 + NO outcompetes HO2 uptake.