Evaporation-Induced Transformations in Volatile Chemical Product-Derived Secondary Organic Aerosols: Browning Effects and Alterations in Oxidative Reactivity

Volatile chemical products (VCPs) are increasingly recognized as significant sources of volatile organic compounds (VOCs) in urban atmospheres, potentially serving as key precursors for secondary organic aerosol (SOA) formation. This study investigates the formation and physicochemical transformations of VCP-derived SOA, produced through ozonolysis of VOCs evaporated from a representative room deodorant air freshener, focusing on the effects of aerosol evaporation on its molecular composition, light absorption properties, and reactive oxygen species (ROS) generation. Following aerosol evaporation, solutes become concentrated, accelerating reactions within the aerosol matrix that lead to a 42% reduction in peroxide content and noticeable browning of the SOA. This process occurs most effectively at moderate relative humidity (∼40%), reaching a maximum solute concentration before aerosol solidification. Molecular characterization reveals that evaporating VCP-derived SOA produces highly conjugated nitrogen-containing products from interactions between existing or transformed carbonyl compounds and reduced nitrogen species, likely acting as chromophores responsible for the observed brownish coloration. Additionally, the reactivity of VCP-derived SOA was elucidated through heterogeneous oxidation of sulfur dioxide (SO2), which revealed enhanced photosensitized sulfate production upon drying. Direct measurements of ROS, including singlet oxygen (1O2), superoxide (O2•–), and hydroxyl radicals (•OH), showed higher abundances in dried versus undried SOA samples under light exposure. Our findings underscore that drying significantly alters the physicochemical properties of VCP-derived SOA, impacting their roles in atmospheric chemistry and radiative balance.

Figure S7, UV-Vis absorption spectra of undried and dried VCP-derived SOA and those dried at higher RH levels.
Figure S8, Relationship between the difference in sulfate production and the difference in EPR signal peak height under UV and dark conditions.In this study, 2-nitrobenzaldehyde (2NB), a recognized chemical actinometer, was utilized to determine the photon flux in the flow cell, in accordance with the method described by Liang et al. 1 This procedure involved the introduction of 50 μM 2NB aqueous solution into the flow cell using syringe aliquots.Subsequently, aliquots were periodically taken for photolysis monitoring.Every 5 minutes, an aliquot was collected and transferred to a sample vial, continuing for a total duration of 25 minutes, to determine the concentration of 2NB.The quantification of 2NB was performed using Ultra-High-Performance Liquid Chromatography with Photodiode Array detection (UHPLC-PDA; UHPLC, Waters Acquity H-Class, Waters, Milford, USA), with the detailed settings referenced in prior literature 2 .The UV absorption channel at 254 nm was specifically employed for the measurement of 2NB.The concentration of 2nitrobenzaldehyde (2NB) demonstrated an exponential decay during photolysis, and the corresponding decay rate constant was determined using the following equation: where [2NB] t and [2NB] 0 are the 2NB concentrations at time t and 0, respectively.The following equation can also be used to calculate (2NB): where N A is Avogadro's number,  λ is the actinic flux (photons cm −2 s −1 nm −1 ), Δλ is the wavelength interval between actinic flux data points (nm), and  2NB,λ and  2NB are the base-10 molar absorptivity (M −1 cm −1 ) and quantum yield (molecule photon −1 ) for 2NB, respectively.Values of  2NB,λ at each wavelength under 298 K and a wavelengthindependent  2NB were adapted from Galbavy et al. 3 The spectral shape of the photon output of our illumination system (i.e., the relative flux at each wavelength) was measured using a high-sensitivity spectrophotometer (Brolight Technology Co. Ltd, Hangzhou, China).Using a scaling factor (SF), this measured relative photon output,  λ, relative , is related to  λ as follows: S3 λ was obtained by combining (1), (2), and (3), as shown in Figure S1.The actinic flux during typical haze over Beijing (40ºN, 116ºE) on January 26, 2015 at 12:00 pm (GMT+8) was estimated using the National Center for Atmospheric Research Tropospheric Ultraviolet Visible (TUV) Radiation Model (Figure S1) 4,5 .The input environmental parameters were set to be as follows: clouds optical depth = 0, base = 4, top = 5; aerosol optical depth = 2.3, single scattering albedo = 0.9, Angstrom exponent = 0.9; direct beam = diffuse down = diffuse up = 1.For clear days, the actinic flux was estimated over Beijing (at the same date and time) using the default parameters.  , Lambe et al. 7 , and Hu et al. 8 The triangles help guide the eye to the regions where ambient oxygenated organic aerosol components typically fall.The legends "CV" and "SV" represent 'capture vaporizer' and 'standard vaporizer,' respectively, denoting the vaporizer used in the instruments.

Figure S1 ,
Figure S1, Photon flux in the flow cell and during typical haze days or clear days in Beijing, China.

Figure S2 ,
Figure S2, GC×GC chromatogram of VOC emissions from the VCP.

Figure S3 ,
FigureS3, Particle number size distribution of the VCP-derived SOA and total aerosol mass from ToF-ACSM measurements vs. mass estimated from SMPS measurements.

Figure S4 ,
Figure S4, Average mass spectra of the VCP-derived SOA measured by ToF-ACSM.

Figure S5 ,
Figure S5, Absorption spectra of dried VCP-derived SOA filter samples extracted immediately and 2 days after the collection.

Figure S6 ,
Figure S6, Van Krevelen plot of H:C vs. O:C obtained from assigned mass spectra of undried and dried VCP-derived SOA.

Figure S9 ,
FigureS9, The normalized sulfate production ratio of UV to dark conditions for dried sample and partially dried sample.TableS1, Chemical profile identified in VOC emissions from the VCP.

Figure S1 .
Figure S1.The photon flux in the flow cell and during typical haze days or clear days in Beijing, China.

Figure S2 .
Figure S2.GC×GC chromatogram of VOC emissions from the VCP.The molecular structures of the top ten emitted species are shown.

Figure
Figure S3.(a) The particle number size distribution of the VCP-derived SOA measured at the exit of the OFR and (b) scatter pot of total aerosol mass from ToF-ACSM measurements vs. mass estimated from SMPS measurements (assuming the aerosol density of 1 g cm -3 ).

Figure
Figure S4 (a) The average mass spectra of VCP-derived SOA and (b) The fractions of the total organic signal at m/z 43 (f 43 ) vs. m/z 44 (f 44 ) from SOA data in this work.It also features triangle plots adapted from Ng et al.6 , Lambe et al.7 , and Hu et al.8The triangles help guide the eye to the regions where ambient oxygenated organic aerosol components typically fall.The legends "CV" and "SV" represent 'capture vaporizer' and 'standard vaporizer,' respectively, denoting the vaporizer used in the instruments.

Figure S5 .
Figure S5.The absorption spectra of dried VCP-derived SOA filter samples extracted immediately (green) and 2 days after the collection (purple).

Figure S9 .
Figure S9.The normalized sulfate production ratio of UV to dark conditions for the dried sample and partially dried sample (~37-42% RH), after 10 h of reaction at 2 ppm SO 2 and 80% RH.

Table S1 .
Chemical profile identified in VOC emissions from the VCP a .Compounds with a relative signal intensity greater than 0.03% are shown.