The positive effect of formaldehyde on the photocatalytic renoxification of nitrate on TiO2 particles

Abstract. Renoxification is the process of recycling NO3- / HNO3 into NOx under illumination and is mostly ascribed to the photolysis of nitrate. TiO2, a typical mineral dust component, is able to play a photocatalytic role in the renoxification process due to the formation of NO3 radicals; we define this process as “photocatalytic renoxification”. Formaldehyde (HCHO), the most abundant carbonyl compound in the atmosphere, may participate in the renoxification of nitrate-doped TiO2 particles. In this study, we established a 400 L environmental chamber reaction system capable of controlling 0.8 %–70 % relative humidity at 293 K with the presence of 1 or 9 ppm HCHO and 4 wt % nitrate-doped TiO2. The direct photolyses of both nitrate and NO3 radicals were excluded by adjusting the illumination wavelength so as to explore the effect of HCHO on the “photocatalytic renoxification”. It was found that NOx concentrations can reach up to more than 100 ppb for nitrate-doped TiO2 particles, while almost no NOx was generated in the absence of HCHO. Nitrate type, relative humidity and HCHO concentration were found to influence NOx release. It was suggested that substantial amounts of NOx were produced via the NO3-–NO3⚫–HNO3–NOx pathway, where TiO2 worked for converting “NO3-” to “NO3⚫ ”, that HCHO participated in the transformation of “NO3⚫ ” to “HNO3” through hydrogen abstraction, and that “HNO3” photolysis answered for mass NOx release. So, HCHO played a significant role in this “photocatalytic renoxification” process. These results were found based on simplified mimics for atmospheric mineral dust under specific experimental conditions, which might deviate from the real situation but illustrated the potential of HCHO to influence nitrate renoxification in the atmosphere. Our proposed reaction mechanism by which HCHO promotes photocatalytic renoxification is helpful for deeply understanding atmospheric photochemical processes and nitrogen cycling and could be considered for better fitting atmospheric model simulations with field observations in some specific scenarios.