Factors Influencing the Formation of Nitrous Acid from Photolysis of Particulate Nitrate

Enhanced photolysis of particulate nitrate (pNO3) to form photolabile species, such as gas-phase nitrous acid (HONO), has been proposed as a potential mechanism to recycle nitrogen oxides (NOx) in the remote boundary layer (“renoxification”). This article presents a series of laboratory experiments aimed at investigating the parameters that control the photolysis of pNO3 and the efficiency of HONO production. Filters on which artificial or ambient particles had been sampled were exposed to the light of a solar simulator, and the formation of HONO was monitored under controlled laboratory conditions. The results indicate that the photolysis of pNO3 is enhanced, compared to the photolysis of gas-phase HNO3, at low pNO3 levels, with the enhancement factor reducing at higher pNO3 levels. The presence of cations (Na+) and halides (Cl–) and photosensitive organic compounds (imidazole) also enhance pNO3 photolysis, but other organic compounds such as oxalate and succinic acid have the opposite effect. The precise role of humidity in pNO3 photolysis remains unclear. While the efficiency of photolysis is enhanced in deliquescent particles compared to dry particles, some of the experimental results suggest that this may not be the case for supersaturated particles. These experiments suggest that both the composition and the humidity of particles control the enhancement of particulate nitrate photolysis, potentially explaining the variability in results among previous laboratory and field studies. HONO observations in the remote marine boundary layer can be explained by a simple box-model that includes the photolysis of pNO3, in line with the results presented here, although more experimental work is needed in order to derive a comprehensive parametrization of this process.

First, some of the filters on which artificial particles had been sampled were analyzed by Ion Chromatography (IC, see Methods: Instrumentation), instead of being used for the illumination experiments.Second, the concentration of particulate nitrate, as measured by IC, was found to be linearly correlated to the Molarity of ammonium nitrate or sodium nitrate in the atomizer solution (Figure S2).Third, the correlation between these two parameters was used to estimate pNO 3 from the Molarity of nitrate in the atomizer solution for all the filters that were not analyzed by ion chromatography.

S1.2 Ambient particles
The illumination experiments with ambient particles were conducted with PM2.5 teflon filters collected at Cape Verde and Delhi (India).Since the experimental procedure is destructive, the concentrations of particulate nitrate on the filters that were analyzed in the photocell had to be estimated.In order to do so, contemporaneous and colocated particle composition measurements were used, as described below.
In the case of Cape Verde, filter samples were taken at the Cape Verde Atmospheric Observatory (CVAO, Carpenter et al. (2010)).PM10 samples were collected on quartz filters from the 30 meters tower at CVAO, using a Digitel DHA-80 high volume sampler.To extract the analytes, a quarter of the quartz filter was extracted in deionized water using a sonication technique for 2 hours.The resulting solution was then filtered through a 0.45 µm syringe filter to remove any insoluble particles, and the filtrate was analyzed for standard water soluble ions, including nitrate, using a Dionex ICS3000 ion chromatography system.To account for S3 background contamination, blank field filters were also analyzed using similar procedures and their concentrations were subtracted from the sample concentrations.
For more details on the sampling and analytical procedure, see Fomba et al. (2014); Deabji et al. (2021).PM2.5 samples were collected on teflon filters from the 7.5 meters tower of CVAO: the sampling and analysis of the filters is described in the Methods section.An average PM2.5 (7.5m) /PM10 (30m) ratio of 0.4 for pNO 3 was derived from these measurements and used to estimate the nitrate concentration on the PM2.5 samples that were used in the illumination experiments.
In the case of Delhi, PM2.5 samples were collected using an Envirosense Digitel high volume air sampler (50 slpm) on pre-baked quartz filters (150 mm diameter, Whatman QM-A).The instrument was located on the campus of the Indian Institute of Technology Delhi (IITD), next to the Partisol sampler that collected the PM2.5 samples used in the illumination experiments.The quartz filters were analyzed by ion chromatography, as described in Srivastava et al..It was assumed that the concentrations of pNO 3 on the quartz and on the teflon filters was the same.

S2 Photolysis rates
The photolysis rate of NO 2 was determined by flowing the photocell with a constant amount of NO 2 and monitoring the mixing ratios of NO and NO 2 when the shutter of the solar simulator was opened (see Methods: Experimental setup).A Thermo Scientific 42i-TL NO-NO 2 -NO x monitor was used for this purpose.The instrument has a stated detection limit of 50 ppt (at 2 minutes averaging time).
Repeat experiments at different levels of NO 2 were made to ensure that the results were reproducible.Two sets of 3 experiments each are shown in Figure S3.
The photo-stationary state (PSS) equation was then used to calculate j(NO 2 ), by fitting Equation S1 to the experimental results:

S4
The average from nine NO 2 photolysis experiments yielded a value of 1.33 × 10 −2 s −1 for j(NO 2 ), which is comparable to the ambient observations made at Cape Verde (Andersen et al., 2022(Andersen et al., , 2023)).The photolysis rates of gas phase HNO 3 and HONO are needed for the calculations presented in this paper.These were obtained by calculating the average j(NO 2 )/j(HNO 3 ) and j(NO 2 )/j(HONO) ratios, using ambient measurements taken at Cape Verde with a spectral radiometer, as described in the Supplementary Information of Andersen et al. (2022).
Artificial particles were generated from solutions of various composition using a TSI 3076 Constant Output Atomizer (see Methods: Artificial and ambient particles).The number and surface area distributions of the particles were measured S1 using a TSI 3936 Scanning Mobility Particle Sizer (SMPS).Examples for ammonium sulfate + ammonium nitrate particles are shown in Figure S1.

Figure S1 :
Figure S1: Number and surface area distributions for ammonium sulfate + ammonium nitrate artificial particles.

Figure S2 :
Figure S2: Nitrate concentration on artificial particle filters (measured by IC) vs nitrate concentration in the atomizer solution.

Figure S3 :
Figure S3: NO 2 photolysis experiments.The red line is the fit to the photostationary state equation (PSS, Equation S1).