Formation mechanisms and atmospheric implications of summertime nitrous acid (HONO) during clean, ozone pollution and double high-level PM2.5 and O3 pollution periods in Beijing

Nitrous acid (HONO) is a key precursor of the hydroxyl radicals (OH) and has a significant impact on air quality. Nowadays, the source of HONO is still controversial due to its complex formation mechanisms, which is widely explored in extensive field and laboratory studies. In this study, the pollution characteristics and source contribution of HONO under different air quality conditions in summer in Beijing were analyzed. The observation periods were classified as three typical periods: clean, ozone pollution, and double high pollution (co-occurrence of high PM2.5 and O3 concentrations). The average concentrations of observed HONO were 0.38 ± 0.35 ppb, 0.21 ± 0.18 ppb, 0.26 ± 0.20 ppb and 0.54 ± 0.45 ppb during the whole, clean, ozone and double high periods, respectively. The elevated HONO levels at night were attributed to vehicle emissions and the RH-dependent heterogeneous conversion of NO2 to HONO. The average emission ratio (HONO/NOx) was 0.85 % ± 0.38 %, and the mean value of calculated nocturnal NO2 to HONO conversion frequency was 0.0076 ± 0.0031 h-1. Based on daytime HONO budget analysis, the largest potential source of HONO was the homogeneous reaction of NO and OH (0.33 and 0.34 ppb h-1), followed by the unknown source (0.11 and 0.21 ppb h-1) during clean and ozone periods, while the unknown source (0.49 ppb h-1) played the predominant role during double high period. The unknown sources of HONO could be attributed to the photo-enhanced heterogeneous conversion of NO2 and the photolysis of particulate nitrate. Furthermore, the photolysis of ozone (0.17, 0.34 and 0.44 ppb h-1) was the major contributor to primary OH during three typical periods. HONO photolysis contributed considerable amounts of primary OH (0.32 ppb h-1) during double high period. These results are helpful to further understand the linkage between HONO and air quality variation.