3.2.1. Low NaNO<sup>2</sup> Concentrations

As has recently been reported in our previous study on the nighttime aqueous-phase formation of MNC, the two identified products, i.e., 3M5NC and 3M4NC, strongly absorb in the near UV range, whereas the reaction mixture also exhibited notable absorption in the visible range, which are all characteristic of BrC [27]. Moreover, in the dark, the enhanced absorption in the visible range can be attributed to MNC degradation by hydroxylation and oxidative cleavage by water addition [28].

Figure 3a–f shows the time evolution of absorption spectra during the aging of 3MC in mildly acidic NaNO<sup>2</sup> solutions under illumination. The dependence of absorbance on the initial concentration ratio of 3MC to NaNO<sup>2</sup> was investigated. Compared to the photolysis of 3MC only (without the addition of NaNO2), where we did not observe any change in the absorption spectrum over the reaction time scale (Figure 3a), the spectrum noticeably evolves already at low NaNO<sup>2</sup> concentration (i.e., 0.1 mM, Figure 3b). The absorption around 350 nm increases with time and corresponds to the formation of first-generation products, 3M5NC and 3M4NC [27]. Nevertheless, only a slight increase in the absorption is in agreement with the ascertainment that at low NOx, BrC formation by nitration is not preferred [40,41]. Note also the low product yield at the lowest NO<sup>2</sup> − concentration (Table 1).

The time evolution of absorbance at 350, 400, and 410 nm, measured experimentally, and modeled using Equation (7), is presented in Figure 4. The predicted absorption describes the experimental data fairly well; a somewhat weaker correlation can be explained by extensive scattering of data due to unreliable measurements at low absorbance (below 0.2). The developed absorption above 400 nm at long reaction times may, among others, correspond to hydroxylated and ring cleavage products formed from MNC, the same as in the dark [28]. See further discussion for justification.

### 3.2.2. High NaNO<sup>2</sup> Concentrations

At higher NaNO2/3MC concentration ratios, the absorption at 350 nm is substantially enhanced in comparison to the experiment with equimolar amounts of both reactants, resulting in the appearance of a distinctive peak with the maximum at this wavelength (Figure 3c–e). In addition, the absorption above 400 nm becomes more pronounced. The enhancement in absorption at examined wavelengths is the fastest at the highest NaNO<sup>2</sup> concentration (Figure 3f).

As already mentioned before, the two products, 3M5NC and 3M4NC with a pronounced absorption peak at 350 nm can react further, forming second-generation products, which can be responsible for the development of absorption in the visible region. According to our recent study [28], the secondary products can be 3M5NC-OH and the oxidative cleavage products of 3M4NC. Their spectroscopic characteristics (both exhibit the absorption maxima at cca. 400 nm and the latter another one at around 500 nm [28]) are in agreement with the presented results and were also confirmed in one of the investigated reaction mixtures by comparison with the recently published data. The chromatogram of the reaction mixture is shown in Figure 5 and mass spectra corresponding to the characteristic peaks were additionally analyzed by an LC-MS/MS experiment. The peak at 2.5 min showed the *m*/*z* ratio of 183 corresponding to 3M5NC-OH, while the peak at 4.5 min with the *m*/*z* of 184 corresponds to the oxidative cleavage products of 3M4NC, which additionally confirmed the identity of investigated compounds.

**Figure 3.** Absorption spectra of the reaction mixture during the photolysis of 3-methylcatechol (3MC) in aqueous NaNO2 solutions (pH = 4–5, T = 25 °C) at different initial concentrations of NaNO2. The concentration of 3MC was always 0.1 mM, while the concentrations of NaNO2 were 0 (**a**), 0.1 (**b**), 0.5 (**c**), 1 (**d**), 2 (**e**), and 5 mM (**f**). The discontinuation in the absorption spectra (d,f) is due to the issues with the instrument and does not affect the results in any way. **Figure 3.** Absorption spectra of the reaction mixture during the photolysis of 3-methylcatechol (3MC) in aqueous NaNO<sup>2</sup> solutions (pH = 4–5, T = 25 ◦C) at different initial concentrations of NaNO<sup>2</sup> . The concentration of 3MC was always 0.1 mM, while the concentrations of NaNO<sup>2</sup> were 0 (**a**), 0.1 (**b**), 0.5 (**c**), 1 (**d**), 2 (**e**), and 5 mM (**f**). The discontinuation in the absorption spectra (d,f) is due to the issues with the instrument and does not affect the results in any way. *Atmosphere* **2020**, *11*, 131 10 of 17

**Figure 4.** Brown carbon (BrC) formation in 0.1 mM 3-methylcatechol and 0.1 mM NaNO2 under sunlight conditions. Experimentally measured (symbols) and modeled absorbance (solid lines) at **Figure 4.** Brown carbon (BrC) formation in 0.1 mM 3-methylcatechol and 0.1 mM NaNO<sup>2</sup> under sunlight conditions. Experimentally measured (symbols) and modeled absorbance (solid lines) at different wavelengths representative of BrC.

different wavelengths representative of BrC.

3.2.2. High NaNO2 Concentrations

identity of investigated compounds.

wavelengths is the fastest at the highest NaNO2 concentration (Figure 3f).

in comparison to the experiment with equimolar amounts of both reactants, resulting in the appearance of a distinctive peak with the maximum at this wavelength (Figure 3c–e). In addition, the absorption above 400 nm becomes more pronounced. The enhancement in absorption at examined

As already mentioned before, the two products, 3M5NC and 3M4NC with a pronounced absorption peak at 350 nm can react further, forming second-generation products, which can be responsible for the development of absorption in the visible region. According to our recent study [28], the secondary products can be 3M5NC-OH and the oxidative cleavage products of 3M4NC. Their spectroscopic characteristics (both exhibit the absorption maxima at cca. 400 nm and the latter another one at around 500 nm [28]) are in agreement with the presented results and were also confirmed in one of the investigated reaction mixtures by comparison with the recently published data. The chromatogram of the reaction mixture is shown in Figure 5 and mass spectra corresponding to the characteristic peaks were additionally analyzed by an LC-MS/MS experiment. The peak at 2.5 min showed the *m*/*z* ratio of 183 corresponding to 3M5NC-OH, while the peak at 4.5 min with the *m*/*z* of 184 corresponds to the oxidative cleavage products of 3M4NC, which additionally confirmed the

**Figure 5.** Chromatogram recorded at 388 nm for the reaction mixture of 0.1 mM 3MC and 1 mM NaNO2 under sunlight conditions is comparable to the one from our previous study [28]. The peak at 2.5 min with the *m*/*z* ratio of 183 corresponds to 3MC5NC-OH, while the peak at 4.5 min with the *m*/*z* of 184 corresponds to oxidative cleavage products of 3M4NC. **Figure 5.** Chromatogram recorded at 388 nm for the reaction mixture of 0.1 mM 3MC and 1 mM NaNO<sup>2</sup> under sunlight conditions is comparable to the one from our previous study [28]. The peak at 2.5 min with the *m*/*z* ratio of 183 corresponds to 3MC5NC-OH, while the peak at 4.5 min with the *m*/*z* of 184 corresponds to oxidative cleavage products of 3M4NC.

The time evolution of absorbance, experimentally measured and modeled by using Equation (7), is presented in Figure 6 for all remaining four experimental conditions. At the initial concentrations of 0.5 and 1 mM NaNO2, the experimental data are very well fitted with the applied absorption model (Figure 6a,b). However, at higher NaNO2 concentrations, the exponential shape of absorbance changes into the sigmoidal one, which is most noticeable at 5 mM NaNO2 in the visible range (Figure 6c,d). This behavior could be attributed to the formation of the second-generation products, which are particularly responsible for the enhanced absorption above 400 nm (e.g., 3M5NC-OH and the oxidative cleavage products of 3M4NC). As the proposed absorption model only considers the one-step transformation of 3MC into BrC, it fails to precisely reproduce such complex behavior of the experimental system. Furthermore, at the highest NO2 <sup>−</sup> concentration, 3MC is completely consumed during the experiment; therefore, primary reaction products, 3M5NC and 3M4NC, stopped forming already before the end of the experiment. Consequently, the decay of absorbance is observed in the second part of the experiment, and the predicting ability of the absorption model is even worse in this case because it does not account for any of the explained phenomena. The time evolution of absorbance, experimentally measured and modeled by using Equation (7), is presented in Figure 6 for all remaining four experimental conditions. At the initial concentrations of 0.5 and 1 mM NaNO2, the experimental data are very well fitted with the applied absorption model (Figure 6a,b). However, at higher NaNO<sup>2</sup> concentrations, the exponential shape of absorbance changes into the sigmoidal one, which is most noticeable at 5 mM NaNO<sup>2</sup> in the visible range (Figure 6c,d). This behavior could be attributed to the formation of the second-generation products, which are particularly responsible for the enhanced absorption above 400 nm (e.g., 3M5NC-OH and the oxidative cleavage products of 3M4NC). As the proposed absorption model only considers the one-step transformation of 3MC into BrC, it fails to precisely reproduce such complex behavior of the experimental system. Furthermore, at the highest NO<sup>2</sup> − concentration, 3MC is completely consumed during the experiment; therefore, primary reaction products, 3M5NC and 3M4NC, stopped forming already before the end of the experiment. Consequently, the decay of absorbance is observed in the second part of the experiment, and the predicting ability of the absorption model is even worse in this case because it does not account for any of the explained phenomena. *Atmosphere* **2020**, *11*, 131 12 of 17

**Figure 6.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and (**a**) 0.5, (**b**) 1, (**c**) 2, and (**d**) 5 mM NaNO2 under sunlight conditions. Experimentally measured (symbols) and modelled **Figure 6.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and (**a**) 0.5, (**b**) 1, (**c**) 2, and (**d**) 5 mM NaNO<sup>2</sup> under sunlight conditions. Experimentally measured (symbols) and modelled absorbance (solid lines) at different wavelengths, representative of BrC are shown.

absorbance (solid lines) at different wavelengths, representative of BrC are shown.

λ= 350 **nm** λ= 400 **nm** λ= 410 **nm**

> R2 = 0.99

undergo secondary reactions substantially.

0.00

wavelengths, representative of BrC are shown.

0.15

0.30

Absorbance (mAU)

0.45

0.60

02468

Time(h)

From characteristic MAC values at 350, 400, and 410 nm listed in Table 3 one can see that with increasing NaNO2 concentration, higher MAC values are obtained corresponding to more BrC production; i.e., the highest was in the order of 4 m2 g−1 for 5 mM NaNO2 at 350 nm. It should be

**Figure 7.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and 2 mM NaNO2 in the dark. Experimentally measured (symbols) and modelled absorbance (solid lines) at different

R2 = 0.96

> R2 = 0.91

In contrast to the illuminated experiments, the model describes the experimental data of dark

0.0

Absorbance (mAU)

0.00

0.13

0.26

0.39

0.1

0.2

0.3

0.4

a

c

R2 = 0.98

R2 = 0.94

> R2 = 0.84

In contrast to the illuminated experiments, the model describes the experimental data of dark reactions very well, also at high NaNO<sup>2</sup> concentrations (0.1 mM 3MC and 2 mM NaNO2; Figure 7). These results additionally support our discussion that under the sunlight, the formed BrC species undergo secondary reactions substantially. reactions very well, also at high NaNO2 concentrations (0.1 mM 3MC and 2 mM NaNO2; Figure 7). These results additionally support our discussion that under the sunlight, the formed BrC species undergo secondary reactions substantially.

absorbance (solid lines) at different wavelengths, representative of BrC are shown.

**Figure 6.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and (**a**) 0.5, (**b**) 1, (**c**) 2, and (**d**) 5 mM NaNO2 under sunlight conditions. Experimentally measured (symbols) and modelled

In contrast to the illuminated experiments, the model describes the experimental data of dark

02468

Time (h)

R2 = 0.79

R2 = 0.99

R2 = 0.99

λ = 350 **nm**  λ = 400 **nm**  λ = 410 **nm**

*Atmosphere* **2020**, *11*, 131 12 of 17

b

R2 = 0.96

0.0

Absorbance (mAu)

0.48

0.00

0.16

0.32

02468

Time (h)

R2 = 0.47

R2 = 0.99

R2 = 0.99

R2 = 0.52

R2 d = 0.67

0.1

0.2

0.3

0.4

**Figure 7.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and 2 mM NaNO2 in the dark. Experimentally measured (symbols) and modelled absorbance (solid lines) at different wavelengths, representative of BrC are shown. **Figure 7.** BrC formation in the reaction mixture of 0.1 mM 3-methylcatechol and 2 mM NaNO<sup>2</sup> in the dark. Experimentally measured (symbols) and modelled absorbance (solid lines) at different wavelengths, representative of BrC are shown.

From characteristic MAC values at 350, 400, and 410 nm listed in Table 3 one can see that with increasing NaNO2 concentration, higher MAC values are obtained corresponding to more BrC production; i.e., the highest was in the order of 4 m2 g−1 for 5 mM NaNO2 at 350 nm. It should be From characteristic MAC values at 350, 400, and 410 nm listed in Table 3 one can see that with increasing NaNO<sup>2</sup> concentration, higher MAC values are obtained corresponding to more BrC production; i.e., the highest was in the order of 4 m<sup>2</sup> g −1 for 5 mM NaNO<sup>2</sup> at 350 nm. It should be emphasized that in the dark (experiments 6 and 7), roughly 50% more BrC is produced at comparable solution conditions. This is consistent with the finding that under the sunlight, BrC degradation, and so the solution bleaching, may proceed via the reactions with OH radicals leading to products with poor absorption in the visible region [33].


4 2 YES 3.77 1.54 1.17 5 5 YES 4.09 1.74 1.42 6 1 NO 5.41 1.73 1.75 7 2 NO 5.63 1.90 1.38

**Table 3.** Wavelength-dependent mass absorption coefficients (MAC350, MAC400, and MAC410) for BrC formation during the photolysis of 3-methylcatechol in aqueous NaNO<sup>2</sup> solutions (pH 4–5, T = 25 ◦C)

MAC values for the photolysis of 3MC in aqueous NaNO<sup>2</sup> solutions (from ca. 0.4 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 410 nm for 0.1 mM NO<sup>2</sup> <sup>−</sup> to ca. 4 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 350 nm for 5 mM NO<sup>2</sup> −, see Table 3) are of the same order of magnitude as reported in the literature [42]. Liu et al. [43] found out that MAC of SOA, especially those produced from aromatic precursors, are much higher for high-NO<sup>x</sup> conditions than for low NOx. In high-NOx, SOA from toluene was largely composed of nitroaromatic compounds (e.g., NC, dinitrocatechol, and NP), the total absorbance of which accounted for 60% and 41% of the overall

absorbance in the ranges of 300–400 and 400–500 nm, respectively [44]. Jiang et al. [45] has recently reported on MAC values of SOA from unsaturated heterocyclic organic compounds; for example, for pyrrole SOA from nighttime NO<sup>3</sup> oxidation, the value of (MAC)290–700nm was 0.34 <sup>±</sup> 0.07 m<sup>2</sup> ·g −1 . In addition, MAC values of water-soluble organic aerosols from different environments have been reported [42]; MAC of 0.3 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 532 nm was found for water-soluble Humic-like-substances(HULIS) fraction of BB aerosol (BBA) in Amason basin [46] and 0.5–1.5 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 404 nm for ambient BB organic aerosol (BBOA) in Boulder, Colorado [47]. Based on our results, we can conclude that during the reaction of 3MC and NaNO2, both in the sunlight and in the dark, light-absorbing components are formed, largely MNC, which efficiently absorb light in the range characteristic of BrC. For comparison, in the aqueous reaction of 3MC at low NaNO<sup>2</sup> concentration (experiment 1), for a factor of more than 2 more BrC is formed (MAC is ca. 1.8 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 350 nm) than from toluene at high-NO<sup>x</sup> conditions at moderate RH (MAC, 0.8 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 365 nm) [43].

### **4. Conclusions**

This study demonstrates that aqueous-phase transformations of 3MC in the presence of HNO2/NO<sup>2</sup> − under atmospheric sunlight conditions importantly contribute to the secondary BrC formation. The two primary products, 3M5NC and 3M4NC, which have been identified as the main products of nighttime processing and shown to strongly absorb in the near UV range [27], are also formed under the sunlight. However, their cumulative yield is only up to about 45% under illumination and was additionally found to be dependent on the initial concentration of NO<sup>2</sup> <sup>−</sup>. When NO<sup>2</sup> − is in excess with respect to 3MC, the rate constant of 3MC degradation (*k*2nd(global) = 0.075 M−<sup>1</sup> ·s −1 ) is twice as high as in the dark (*k*2nd = 0.032 M−<sup>1</sup> ·s −1 ) under the same initial conditions. Several different degradation mechanisms are possible in the sunlight. Besides the non-radical nitration reaction, which has already been studied in the dark, different reactive species can be formed during the photolysis of HNO<sup>2</sup> (e.g., OH, NO, NO2), which are also possible sinks of 3MC.

The dependence of absorbance evolution on the initial NO<sup>2</sup> −/3MC concentration ratio during the photolysis showed that when the ratio is 1, the absorption around 350 nm increases with time, which corresponds to the formation of identified first-generation products (3M5NC and 3M4NC). At higher concentration ratios, the absorption at this wavelength substantially increases, and the absorption above 400 nm becomes pronounced as well. The second-generation products, i.e., 3M5NC-OH and the oxidative cleavage products of 3M4NC, were confirmed to be responsible for the absorption in the visible region.

With higher NO<sup>2</sup> <sup>−</sup> concentrations, the characteristic MAC values (up to more than 4 m<sup>2</sup> ·g <sup>−</sup><sup>1</sup> at 350 nm for 5 mM NO<sup>2</sup> −) increase due to more BrC formation. We believe that the aqueous-phase reaction of 3MC in the presence of HNO2/NO<sup>2</sup> −, either under the sunlight or in the dark, may significantly contribute to SOA light absorption. This is supported by the fact that MNC and secondary hydroxylated/ring cleavage products produced from 3MC significantly absorb light in the range characteristic of BrC. Our conclusions are also supported by recent field measurements of aerosol particles, where MNC have been identified as significant contributors to atmospheric BrC [15–19,22].

**Author Contributions:** K.V. performed the experiments, conducted the research, and helped to write the manuscript; M.Š. helped in product identification; A.K. contributed to the data analysis and writing process; and I.G. wrote the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We acknowledge the financial support from the Slovenian Research Agency (research core funding No. P1-0034). We thank Luka Nunar for his help with experimental work.

**Conflicts of Interest:** The authors declare no competing financial interests.
