Modeling the Formation, Degradation, and Spatiotemporal Distribution of 2-Nitrofluoranthene and 2-Nitropyrene in the Global Atmosphere

Polycyclic aromatic hydrocarbons (PAHs) are common atmospheric pollutants and known to cause adverse health effects. Nitrated PAHs (NPAHs) are formed in combustion activities and by nitration of PAHs in the atmosphere and may be equally or more toxic, but their spatial and temporal distribution in the atmosphere is not well characterized. Using the global EMAC model with atmospheric chemistry and surface compartments coupled, we investigate the formation, abundance, and fate of two secondarily formed NPAHs, 2-nitrofluoranthene (2-NFLT) and 2-nitropyrene (2-NPYR). The default reactivity scenario, the model with the simplest interpretation of parameters from the literature, tends to overestimate both absolute concentrations and NPAH/PAH ratios at observational sites. Sensitivity scenarios indicate that NO2-dependent NPAH formation leads to better agreement between measured and predicted NPAH concentrations and that photodegradation is the most important loss process of 2-NFLT and 2-NPYR. The highest concentrations of 2-NFLT and 2-NPYR are found in regions with strong PAH emissions, but because of continued secondary formation from the PAH precursors, these two NPAHs are predicted to be spread across the globe.


Contents of Supporting Information
Additional information S1. Formation chemistry of 2-NFLT and 2-NPYR S2. Partitioning between gas and particle phase S3. Particle-phase loss of 2-NFLT and 2-NPYR S4. Comparing simulated with measured concentrations Tables   Table S1. Physicochemical properties and degradation rate coefficients of 2-NFLT and 2-NPYR used. Table S2. Full collection of NPAH observations used for comparison with the model.

S1. Formation Chemistry of 2-NFLT and 2-NPYR
Y 2-NFLT,OH , Y 2-NPYR,OH and Y 2-NFLT,NO3 is the total yield of either 2-nitrofluoranthene  or 2-nitroyprene (2-NPYR) from the amount consumed of fluoranthene (FLT) or pyrene (PYR), by their respective reactions. Experimental values can be seen in Table S1 and ranged between 0.5-24%. 1 One key factor contributing to this value of total yield is the % conversion of the PAH-radical adduct into NPAH (reaction with NO 2 ) instead of other oxygenated products (reaction with O 2 ) (Fig. S1). Several other factors conflate to reduce the empirical Y NPAH/PAH of 2-NFLT and 2-NPYR. These likely include: (1) further reaction and loss of NPAH and (2) formation of other NPAH isomers. The contribution of (2) may be estimated from the theoretical calculations of Zhang et al. (2014), by accounting for the fact that only certain adducts may lead to the formation of 2-NFLT or 2-NPYR (≈ 20-60%). 2 Initial radical addition is described by experimentally determined rate coefficients (Table S1). 1 The kinetics of reaction of FLT with OH has been measured by Brubaker and Hites. 3 These rate coefficients are in reasonable agreement with theoretical calculations for the reaction of OH and NO 3 at specific positions on FLT and PYR. 2 Notably, the rate of reaction between FLT and NO 3 is dependent on NO 2 .
With respect to fate of the PAH-radical adduct, Ghigo et al. (2006), calculated theoretically the ratio k NO2 /k O2 of rate coefficients for benzene and naphthalene (9×10 3 and 8×10 4 respectively). 4 This is in good agreement with the experimentally determined value, 3.6×10 4 , for k NO2naphthalene /k O2-benzene (i.e. rate coefficient for reaction with NO 2 and O 2 is for the naphthalene-and benzene-OH adducts, respectively). [5][6][7] The ratio k NO2 /k O2 for the pyrene-OH and pyrene-NO 3 adducts were calculated as 5×10 9 and 2×10 9 , respectively. 4 By using the concentration of NO 2 and O 2 , an analytical expression for the yield scaling factor (Ω) may be obtained from the ratio of rate coefficients k NO2 /k O2 (equation 1). In the NO 2 -dependent scheme, the yield scaling factor is multiplied by the empirically determined yield to ensure that (a) the model yield is equivalent to the empirically determined one at high NO 2 concentrations and (b) decreases towards lower NO 2 concentrations.
Results from a sensitivity study using a formation scheme in which Y 2-NFLT,OH , Y 2-NPYR,OH and Y 2-NFLT,NO3 were dependent on NO 2 are shown in Fig. S2. In this scheme the value of k NO2 /k O2 was set as 1 × 10 7 (see Table S1). The dependence of Y 2-NFLT,OH , Y 2-NPYR,OH and Y 2-NFLT,NO3 on NO 2 in this scheme is shown (Fig. S5) as well as the spatial distribution of Y 2-NFLT,OH for an illustrative month (Fig. S6).

S2. Partitioning Between Gas and Particle Phase
Semi-volatile compounds are distributed to significant mass fractions between the particle and gas phases of aerosols. Polyparameter linear free energy relationships (ppLFERs) are suitable to predict the mass distribution of NPAHs. 8 Experimentally determined ppLFER solute specific descriptors for 1-nitropyrene are used in the model. 9 These parameters are expected to be reasonably well representative for both 2-NFLT and 2-NPYR. Phase equilibrium is reestablished at each model time step (30 min).

S3. Particle-phase Loss of 2-NFLT and 2-NPYR
In the gas-phase, NPAHs are generally less reactive than their precursors. 10 11 The same reaction was never reported for 2-NPYR. Particle phase 1-nitropyrene was shown to be less reactive than PYR with O 3 or NO 2. 12,13

S4. Comparing Simulated and Measured Concentrations
In order to minimize incommensurability issues, model simulated concentration values were compared with observed values in the following ways:  Bilinear interpolation within the model grid cell is used to obtain simulated concentrations that better represents the location of each observational site.  Comparison between model and observations is now solely presented at rural rather than urban sites. Rural sites are less affected by local sources, and make for more representable comparison.  Observation sites where data was available over multiple seasons, were split into separate data points in order to prevent seasonal information being lost by averaging over time.  Figure S1. a) Mechanism of 2-nitrofluoranthene formation by the addition of a radical (X = OH or NO 3 ). Following initial radical addition (k add ), three processes compete for the fluoranthene-radical adduct: unimolecular decomposition (k rev ), reaction with O 2 to form oxygenated products (k O2 ) or reaction with NO 2 to form 2-nitrofluoranthene. b) The same scheme is applicable for 2-nitropyrene with addition of an OH radical. a) b) S14 Figure S2. Dependence of yield on NO 2 mixing ratio in the NO 2 -dependent reactivity schemes for the reaction of FLT or PYR with a) OH or b) NO 3 . S15 Figure S3. Comparison between simulated and measured near-surface concentrations [pg m -3 ] of 2-NFLT (above) and 2-NPYR (below) using the default reactivity scenario and sensitivity test with homogeneous reaction of NPAH with the OH radical. S16 Figure S4. Comparison between simulated and measured near-surface concentrations [pg m -3 ] at rural and urban sites of 2-NFLT comparing alternative scenarios with α = 0.05 (above) and α = 0.005 (below). S17 Figure S5. Comparison between simulated and measured near-surface concentrations [pg m -3 ] at rural and urban sites of 2-NPYR comparing alternative scenarios with α = 0.05 (above) and α = 0.005 (below).