Structural and Light-Absorption Characteristics of Complex Water-Insoluble Organic Mixtures in Urban Submicrometer AerosolsQingcai Chen,†,‡ Fumikazu Ikemori,†,§ Yuua Nakamura,† Petr Vodicka,∥ Kimitaka Kawamura,∥

and Michihiro Mochida*,†

†Graduate School of Environmental Studies, Nagoya University, Nagoya 464-8601, Japan‡School of Environmental Science and Engineering, Shaanxi University of Science and Technology, Xi’an 710021, China§Nagoya City Institute for Environmental Sciences, Nagoya 457-0841, Japan∥Chubu Institute for Advanced Studies, Chubu University, Kasugai 487-8501, Japan

*S Supporting Information

ABSTRACT: Submicrometer aerosols in the urban atmosphere ofNagoya, Japan, were collected in late winter and early spring, and thewater-insoluble organic matter (WISOM) in the samples werefractionated into six subfractions based on their polarities by usingsolvent and normal-phase solid-phase extractions: nonpolar (F1), low-polar (F2 and F3), and medium-polar (F4, F5, and F6) fractions. Theoverall structural characteristics of these subfractions were then analyzedusing Fourier transform infrared spectroscopy, nuclear magneticresonance spectroscopy, and high-resolution aerosol mass spectrometry.Quantitative information related to the overall chemical characteristics ofthe WISOM in the different polarity fractions, including their elementalcompositions, the relative abundances of different functional groups andtheir fragments from electron impact ionization, was obtained. Thesewater-insoluble fractions accounted for half of the total light absorption by the extracted aerosol matter at 400 nm. Thecontributions of the medium-polar fractions to both the total organic carbon and light absorption by the extracts were dominantamong the contributions from the six subfractions. Large molecules with aromatic and heteroatomic (O and N) groups, includingcharge transfer complexes, might have greatly contributed to the light absorption by the fraction F4, which is the largest fractionof the extracted water-insoluble organic matter.

■ INTRODUCTION

Submicrometer aerosol particles in the atmosphere containabundant organic compounds (20−90%).1,2 Their presenceinfluences the properties of these aerosols and thereby affects theroles of the aerosols in the climate and in human health.3 Thechemical characterization of these organic compounds isimportant to understand and predict these roles. However, theorganic aerosol components remain poorly characterizedbecause of the highly complex nature of their composition.4−6

The brown carbon (BrC) in organic aerosols can absorb light atvisible and near-ultraviolet (UV) wavelengths. Hence, BrCpotentially plays important roles in the photoinitiated reactionsof organic aerosols and thus in the radiative balance of the Earth.The characterization of BrC is important to gain insight into theirsource and formation processes and to evaluate the effects of BrCon the climate.7,8

Over the past decade, much of the work on the character-ization of organic aerosol components has focused on water-soluble organic matter (WSOM).9−13 However, the overallcharacteristics of the remaining organic matter, that is, the water-insoluble organic matter (WISOM), have not received muchattention14 even though the remaining organic matter constitutes

a large fraction of the organic carbon in urban and marineaerosols (26−94% in OC; 19−92% in OM).14−17 AerosolWISOM is a complex organic matter that is composed of notonly hydrocarbons but also N- and O-containing com-pounds.16,18,19 WISOM also contains high-molecular-weightoligomers.20−22 The analysis of the chemical composition ofaerosol WISOM has generally been performed using gaschromatography (GC) or two-dimensional GC (2D-GC) incombination with mass spectrometry.19,23,24 However, thecomposition of a large mass fraction of the organic matter,including unresolved complex mixtures (branched and cycliccompounds) and nonvolatile or thermally unstable compounds(oxygenated compounds with high molecular weights), has notbeen well characterized.6,19,23,24 BrC is in both the WSOM andWISOM fractions, and recent studies have shown that WISOMhas a significant light absorptivity that is stronger than that ofWSOM.9,25 The light absorption by the methanol-extracted

Received: April 13, 2017Revised: June 20, 2017Accepted: June 22, 2017Published: June 22, 2017

Article

pubs.acs.org/est

© 2017 American Chemical Society 8293 DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

matter from the ambient aerosols in Los Angeles was ∼3 (365nm) and 21 (532 nm) times higher than that extracted by water.9

Among the organic compounds in WISOM, unsaturatedcompounds, such as polycyclic aromatic hydrocarbons (PAHs)and oxygenated and nitrated PAHs, have electron-conjugatedsystems and absorb light at near-ultraviolet and -visiblewavelengths.26−28 It is important to clarify which classes orstructures of the organic compounds contribute largely to thelight absorption of WISOM.In this study, we employed Fourier transform infrared (FT-IR)

spectroscopy, nuclear magnetic resonance (NMR) spectroscopyand high-resolution aerosol mass spectrometry (HR-AMS) tocharacterize the structures of a WISOM comprised of a widevariety of organics. These analytical techniques are suitable toobtain the overall structural information on complex organicmatter.6 A normal-phase solid-phase extraction (SPE) methodwas employed to separate the WISOM into six subfractions, thatis, nonpolar (F1), low-polar (F2 and F3), andmedium-polar (F4,F5, and F6) fractions. The fractions were quantified, and thecharacteristic compound classes were identified using the aboveanalytical techniques. Furthermore, the light-absorption charac-teristics of the WISOM and its subfractions were investigatedusing the UV−visible spectra. The major organic compoundgroups contributing to light absorption in the WISOM wereidentified, and their relationship to the chemical structures isdiscussed.

■ EXPERIMENTAL SECTIONCollection, Extraction and Separation of Aerosol

Components. Submicrometer urban atmospheric particles(aerodynamic diameter: <0.95 μm) were collected on therooftop of a building at Nagoya University in Nagoya, Japan.Nagoya is in Chukyo, which is a large metropolitan area(population of the area: 9 million); the studied aerosols should

have been strongly influenced by local traffic/industrial/domestic sources. Aerosol sampling was performed using ahigh-volume sampler (model-120B, Kimoto Electric Co. Ltd.)equipped with a cascade impactor (TE-230, Tisch Environ-mental, Inc.). The <0.95 μm diameter particles in the air werecollected on 8 × 10-in. quartz-fiber filters (Tissuquartz 2500QAT-UP, Pall Life Science) after the removal of >0.95 μmparticles in the impactor. A total of 24 samples, including fourweekly samples (for all the chemical analysis and light-absorptionmeasurements) and 20 two-day samples (for analysis using FT-IR and UV−visible spectrometers), were collected from 25August to 28 September 2013 and from 4March to 7 April 2014.The filters for the determination of the field blank were preparedby operating the sampler for 10 s. The collected aerosol sampleswere stored at ∼−20 °C or lower until further analysis.Three filter punches (diameter: 34 mm) for each sample were

used for the extraction and fractionation of the organiccomponents (Figure 1). The water-soluble matter (WSM) wasextracted by ultrasonicating the punches with 3 g of Fluka water(Sigma-Aldrich) three times (10 min each) and filtering through0.2 μm polytetrafluoroethylene (PTFE) membrane filters on astainless-steel syringe filter holder (Millex). After the extraction,the WISOM was further extracted from the same filter punchesby ultrasonicating continuously with 3 g of methanol (MeOH),dichloromethane (DCM) and n-hexane for 10 min, respectively.The PTFE membrane filter used to filter the WSM was usedagain to filter the extract containing the WISOM to recover theremaining organics on the PTFE filters.The extracted WISOM fractions were dried using a rotary

evaporator and were redissolved in different solvents in the orderof 6 mL (2 mL × 3 times) of n-hexane, n-hexane/DCM (2/1, v/v), DCM,DCM/MeOH (2/1, v/v), andMeOH and 4mL (2mL× 2 times) ml of MeOH containing 2 wt % of ammonia for thesame samples. These solutions were placed on the same silica gel

Figure 1. Procedure of the solvent extraction, fractionation and analysis of the organic mixtures in aerosols.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8294

columns (500 mg, glass hardware, Sigma-Aldrich) one by one inthe order of the redissolution, and the six effluents after theaddition of the extracts with the respective six solvents werecollected separately in 15 -ml precleaned glass bottles. Theextracts obtained with the solvents of n-hexane, n-hexane/DCM(2/1, v/v), DCM, DCM/MeOH (2/1, v/v), and MeOH and 4mL of MeOH containing 2 wt % of ammonia are hereafterreferred to as F1, F2, F3, F4, F5, and F6, respectively. The silicagel columns were prerinsed with 6 mL of MeOH, DCM and n-hexane prior to the sample loading. Note that the Snyder’spolarity indices of the organics used for solvents, n-hexane, DCMand MeOH, are 0.0, 3.4, and 6.6, respectively.29

Analysis of Carbonaceous Components. The extractsfrom the weekly samples were dried under a nitrogen flow andredissolved in 100 μL of the corresponding solvents to analyzethe carbonaceous components. Then, 20-μL aliquots of theextracts were dropped onto 1.0 × 1.5 cm prebaked quartz-fiberfilters, and the solvents were allowed to volatilize for ∼25 min.The prepared filter samples were analyzed using a laboratory-based thermal-optical carbon analyzer (Sunset Laboratory Inc.)with the IMPROVE temperature protocol.30 The quantifiedcarbon fractions were OC1−4 (OC1, OC2, OC3, and OC4detected at 120 °C, 250 °C, 450 °C, and 550 °C, respectively, in ahelium atmosphere) and EC1−3 (EC1, EC2, and EC3 detectedat 550 °C, 700 °C, and 800 °C, respectively, in a 2% oxygen/98%helium atmosphere). The reflectance and transmittance of lasersignals with the light wavelength of 658 nm were used formonitoring the pyrolyzed organic carbon. The relative standarddeviation of the quantification of the carbon masses in theextracts by the thermal-optical analysis was 3.9% (n = 5). In theanalysis, three blank samples for each fraction were also analyzedto correct for blank levels. The blank levels corresponded to 1−2%, 3.5%, and 7.6% of the lowest concentrations of the extractsF1−F4, F5, and F6, respectively.The concentrations of the extracted organic carbon (EOC =

WSOC + extracted water-insoluble organic carbon (E-WISOC))were 1.96 ± 0.16 μg m−3 (mean ± standard deviation (SD)),which corresponds to 87% ± 5% and 106%± 10% (mean± SD)of the filtered total organic carbon (TOC) derived via thelaboratory-based thermal-optical carbon analyzer with thethermal-optical transmittance (TOT) and reflectance (TOR)methods, respectively. The mean ± SD of the ratios of EOC toTOC were 74% ± 5% if the TOC concentrations from asemicontinuous carbon analyzer (Sunset Laboratory Inc.) wereused (Supporting Information (SI) section S2). The highpercentages indicate that most of the TOC on the filters wasextracted, although the percentage smaller than 100% may, inpart, be caused by the presence of organics insoluble to thesolvents used, and the loss of some organics during the extractionprocedure, for instance, by the volatilization.Diffuse Reflectance FT-IR Analysis. The liquid extracts

from the weekly samples and those from the combination of fivetwo-day samples were concentrated to∼0.5 mL under a nitrogenflow. Then, 0.1 g of KBr (FT-IR grade, Sigma-Aldrich) was addedand concentrated to be fully dried. The mixture of the extractsand KBr was ground in an agate mortar for FT-IR spectroscopyanalysis.An FT-IR spectrometer (6100, JASCO) with a diffuse

reflectance accessory (DR PR04 10-M, JASCO) was employedat ambient pressure to quantify the chemical functional groups inthe extracts. FT-IR spectra in the range of 600−4000 cm−1 wererecorded 64 times, with a resolution of 4 cm−1. The spectrumbaseline was determined by analyzing the pure KBr before the

analysis of the aerosol-extract samples. Six types of functionalgroups, that is, aliphatic C−H (saturated and unsaturated),alcoholic C−OH, nonacidic carbonyl CO, carboxylic COOH,amine C−NH2, and organic nitrate C−ONO2, were quantified.The details of the processing for the quantification are describedin Chen et al.,31 which is based on works by Takahama et al.,32

among others. No comparable signals were detected in the blanksamples for each fraction (SI Figure S1).

1H NMR Analysis. Extracts of the fractions from the fourweekly filters were combined, dried under a nitrogen flow, andredissolved in 0.6 mL of chloroform-d (CDCl3) and 0.75 mL ofmethanol-d4 (99.8 atom % D, Sigma-Aldrich) for the fractionsF1−F4 and F5, respectively; 0.03% (v/v) of the TMS inchloroform-d and methanol-d4 served as the internal standardsfor the 1H NMR measurements. The 1H NMR spectra of thedifferent extracts in 5 mmNMR tubes were recorded 1024 timeson a JEOL JNM ECA-600 spectrometer equipped with a 5 mmFG/HCN probe.The baseline and phase corrections and the data integration

were performed using matching software (JEOL). The peaks inthe 1H NMR spectra of the blank samples were subtracted fromthe 1HNMR spectra of the samples. The intensities of the signalsin the four chemical shift ranges, that is, the saturated aliphaticregion (H−C−C, δ0.6−δ1.9 ppm), the unsaturated aliphaticregion (H−C−C, δ1.9−δ3.2 ppm), the oxygenated saturatedaliphatic region (H−C−O, δ3.2−δ4.5 ppm) and the aromaticregion (H−Ar, δ6−δ9 ppm), were integrated.33−35 For theidentification of specific molecular structures that correspond tothe peaks in the 1H NMR spectra and the 1H−1H correlationspectra (COSY), the online tools available at nmrdb.org wereemployed.36

HR-AMS Measurements. The extracts from the weeklysamples were atomized with argon (purity: 99.9999%) and thenpassed through two diffusion scrubbers (one with silica gel andthe other with activated carbon mixed with silica gel (12 wt %))to remove the water and organic solvent vapors. The HR-AMSspectra were acquired in both the V- and W-modes. The datawere analyzed using the Squirrel v.1.56A and Pika v1.15Asoftware (http://cires.colorado.edu/jimenez-group/ToFAMSResources/ToFSoftware/). In the analysis using Pika,all the CO+ ion signals were considered to be from organicsbecause pure argon was used as the carrier gas, and all organicions were classified into Cx, CH, CHO1, CHOgt1, CHN, CHON,NO, HO, CO, and CO2 groups based on their elementalcompositions. The elemental analyses to determine the molarelemental ratios and the mass ratio of organic matter (OM) toorganic carbon (OC) were performed as described in Aiken et al.(for N/C)37 and Canagaratna et al. (for O/C, H/C, and OM/OC).38 The mass concentrations of the organics in the WSMextracts were quantified by using the internal standard methodand phthalic acid.17,39 When the solvents for the extraction wereatomized, the signal intensities of the organics in the HR-AMSspectra were only 0.8% or less of those of the organic aerosolcomponents extracted using the appropriate solvents for the HR-AMS analysis. Hence, the interference of the signals from theorganic vapors from the solvents was not evident.

UV−visible Absorption Spectra. The extract solutionsfrom the weekly and two-day samples were placed in 1 cm path-length quartz cells and were subjected to analysis using a UV−visible spectrophotometer (V-570, JASCO). The UV−visibleabsorption spectra in the range of 250−800 nm were recorded atintervals of 0.5 nm 3 times. The UV−visible absorption spectra ofthe blank samples were also recorded to subtract their

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8295

contribution from the spectra of the extracts. Based on the light-absorption spectra, themass absorption efficiency (MAEλ, m

2 g−1

C) at a wavelength of λ nm of the organics in the extracts werecalculated:11

= ·λ λ CMAE Abs ln(10)/ OC (1)

where Absλ is the light-absorption coefficients (m−1) of theextract solutions at λ, and COC (g m−3) is the organic carbonconcentrations in the extracts. For the fraction F4, thecontribution of structures with charge transfer (CT) to theobserved absorption was assessed by the addition of NaBH4,according to the protocol described by Ma et al.40

Analyses of Molecular Weight Distribution. Analyses ofthe molecular weight distributions of the fractionated organicsfor F4 and WSOM were performed using a mass spectrometer(micrQTOF-QII, Bruker) equipped with electrospray ionizationsources (ESI). Details are provided in the SI.

■ RESULTS AND DISCUSSIONCarbon Concentrations and Carbon Fractions. The

concentrations of carbon in the aerosols and the six subfractionsof WISOM therein were determined using the thermal-opticalmethod (SI Tables S1 and S2). The thermograms from thethermal-optical analysis of the extracted organic carbon areshown in SI Figure S2. The concentrations of E-WISOC in eachaerosol sample were calculated as the sum of the concentrationsof the organic carbon in the six subfractions, where theconcentrations of the organic carbon in each fraction, F1−F6,were the total concentrations of the thermally derived carbon(OC1−4 + EC1−3) detected in the thermal-optical analysis.Figure 2a presents the mean of the relative abundances of the

organic carbon masses in the different fractions. The ratio of E-WISOC toWSOCwas 0.41± 0.02 (mean± SD), indicating thata substantial part of the organic compounds in PM0.95 wasinsoluble. On average, half of the E-WISOC (50%) was found inF4, a low-polar fraction of WISOM. By contrast, relatively smallfractions of carbon were detected in the nonpolar fraction F1

(mean: 9%), low-polar fractions F2 (15%) and F3 (18%), andmedium-polar fractions F5 (6%) and F6 (2%).The carbon-mass percentages of the thermally derived carbon

fractions (OC1−4 and EC1−3) in the six subfractions ofWISOM were obtained (SI Table S1). The relative abundancesof OC2, which is associated with high-volatility compounds(≤250 °C), in the nonpolar fraction F1 and low-polar fraction F2were higher than those of the other fractions. This resultindicates that the organic compounds in the fractions F1 and F2have relatively high volatilities. Furthermore, the thermaldecomposition of the organic compounds during the carbonanalysis was evident from the decrease in the reflectance andtransmittance laser signals during the steps to analyze OC3 andOC4, and a large fraction of the organic carbon was found to bepyrolyzed organic carbon (SI Figure S2). There was a positivecorrelation between the total mass fractions of the carbon inEC1−3 and the mean of the O/C ratios of the organics in each ofthe fractions F1−F6 that were derived from the HR-AMS spectra(r = 0.88, p < 0.05). This result suggests that the oxygenatedorganic compounds in E-WISOC were prone to charring.41

Analysis of Functional Groups Using FT-IR Spectros-copy. The FT-IR spectra of the organic fractions studied, whichprovide quantitative information about the different organicfunctional groups in the aerosol extracts, are presented in Figure3. The main peaks and their assignments are shown as the colorbars in the top part of the figure. The relative contents of theorganic functional groups are also presented. Nearly all of thequantified organic functional groups in F1 were aliphatic C−Hgroups (mean: 98%). The result suggests that F1 were aliphatichydrocarbons, which is supported by the NMR analysis asexaplained in the next section. A weak absorption in the range of1000−1100 cm−1 was observed, typically corresponding to theC−O stretching vibration of alcohols, ethers and esters.42

However, absorption from esters (C−O−CO) was unlikelybecause the peaks of carbonyl (CO) groups were absent in theFT-IR spectra. Hence, a small mass fraction of alcohols (mean:<2%) and/or ethers was present in F1.The FT-IR spectra of F2 exhibited strong absorption peaks at

1728, 1280, 1125, and 1073 cm−1. These originated from thestretching vibrations of CO and C−O. Peaks at 1600 and 1580cm−1, which are generally assigned to aromatic CC ringvibrations, were also observed. The peak at 744 cm−1

corresponded to the H−Ph plane bending vibration, indicatingthat the phenyl rings in this fraction were ortho-substituted. Theabove peak assignments and comparison of the spectra to thoseof the IR spectra of phthalate esters (PAEs) in the NIST libraryand previous studies suggest that F2 contained abundantPAEs.43−45

An abundant C−ONO2 in F3 was indicated by the peaks at1630, 1280, and 863 cm−1.46,47 The mean of the masspercentages of C−ONO2 in F3 were estimated to be 15%,which was higher than the percentages in the other fractions.This result suggests that F3 contained abundant organic nitrates.In addition, the peaks at 1720 cm−1 (CO) and in the range of1000−1200 cm−1 (C−O−C) suggest that F3 may also havecontained abundant aliphatic esters.Strong absorptions of the C−OH (3100−3600 cm−1), CO

(1720 cm−1), and C−O−C (1000−1200 cm−1) groups wereobserved for F4, indicating that the organics were primarilyesters, alcohols, ketones and/or ethers. The peaks atapproximately 1635 cm−1 corresponded to nitrogen-containinggroups, which were presumably organic nitrates and/or amines.A peak at 1580 cm−1 was also observed, suggesting that this

Figure 2. (a) Mean of the relative abundances of the organic carbonmass in the different fractions of the aerosol components and (b) meanof the relative contributions of the aerosol extracts to the lightabsorption at 400 nm.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8296

fraction contained aromatics.48 The peaks at 1384 and 1462 cm−1

were attributed to the C−H deformation of methyl (CH3) andmethylene (CH2) groups;

42 the high intensity ratio of CH3/CH2

for F4 (mean: 2.9) was higher than that in fractions F1 (mean:0.2), F2 (mean: 0.5), and F3 (mean: 0.8), indicating abundantbranched structures.The FT-IR spectra of F5 were characterized by the high

relative abundances of alcohols, which were indicated by strongabsorptions in the wavenumber ranges of 3100−3600 and 1100−1200 cm−1. Significant absorptions at approximately 1715, 1675,and 1580 cm−1, which are generally assigned to saturated andunsaturated carbonyl (CO) groups and aromatics (CC),were also observed.49 Substantial parts of the C−OH (mean:35%) and CO (mean: 5%) functional groups were not in thecarboxylic acids because the carboxylic COOH in F5 was notabundant (mean: 3%). The fraction F5 may have containedabundant alcohols and quinones, resulting in the observedspectra.

1H NMR Spectroscopy. The 1H NMR spectra furtherprovided quantitative information about the structural character-istics of the complex water-insoluble organic matter with respectto the organic functional groups with protons. Although thefunctional groups detectable by 1H NMR and FT-IR spectraoverlap, 1H NMR analysis has an advantage in the identificationof some of the detailed chemical structures, such as aromatics,esters and ethers. As seen in Figure 4f, aromatic protons (δH 6−9ppm) accounted for, on average, 5% of the total protons in theorganics of F2 and F5, which was significantly higher than thepercentages in the other fractions (≤2%). This result isconsistent with the assignment of the compound groups from

the FT-IR spectra. Multiple signals at approximately δH 7.71,7.53, 4.22, and 1.69 ppm were identified in the 1H NMR spectraof F2 (marked as a−d in panel (B) of Figure 4), which areassigned to protons from PAE molecules. Long-chain ester orketone structures were identified in F4 based on the 1H−1Hcorrelation spectra (COSY) (SI Figure S3).35 The oxygenatedsaturated aliphatic region of δH 3.2−4.5 ppm corresponds to H−C−Oproton resonances from the aliphatic forms of esters, ethersand alcohols. The signals in this range accounted for 3%, 7%, 8%,8%, and 20% of the total proton signals for the fractions F1, F2,F3, F4, and F5, respectively.

HR-AMS Characteristics of Various Organic Fractions.The HR-AMS analysis provides the bulk chemical structuralcharacteristics of the complex organic matter in aerosols,although their molecular structures are lost in the electronionization. Further, more molecular-level information can beinferred from the fragmentation patterns. Figure 5(a) shows theHR-AMS spectra of the six subfractions ofWISOM. Based on thehigh-resolution analysis (m/z < 130) of the HR-AMS spectra forthe fractions F1, F2, F3, F4, F5, and F6, the mean of their O/Cratios were determined to be 0.01, 0.06, 0.13, 0.20, 0.69, and 0.79,respectively. As expected, this result suggests that the O/C ratiosof the six subfractions of WISOM increased as their polaritiesincreased. Among them, the O/C ratios of F5 and F6 in thisstudy were significantly higher than the O/C ratios of the totalWISOM reported by Chen et al. for 12 total suspendedparticulate samples.31

The ions in the mass spectra of the non- and low-polarfractions F1−F4 were primarily CxHy

+, which are presumablyattributed to aliphatic hydrocarbon structures. The CxHy

+ ions

Figure 3. (a−e) FT-IR spectra of the five fractions of the organic matter. Because the signals of the organics in the FT-IR spectra of the fraction F6 werelow, the result is not presented. (f−j) Enlarged views of the spectra. (k) Pie charts of the mean of the mass fractions of the quantified functional groups inthe organic matter. The mass fractions of the functional groups are also summarized in SI Table S4.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8297

accounted for an average of 97%, 82%, 63%, and 55% of the totalmass of the fractions F1, F2, F3, and F4, respectively. Conversely,

the ions of CxHyOz+, CO2

+, CO+ and HyOz+ in the mass spectra

of the medium-polar fractions F5 and F6 accounted for more

Figure 4. (a−e) 1H NMR spectra of the five fractions of organic matter. Because the signals of the organics in the 1H NMR spectra of the fraction F6were low, the result is not presented. (f) The pie charts represent the mean of the mass fractions of the quantified functional groups in their respectivefractions. The mass fractions of the quantified functional groups are also summarized in SI Table S5.

Figure 5. (a) Mean of the normalized HR-AMS spectra collected in the W-mode for fractionated organic matter groups, F1−F6. (b) Mean of the massfractions of the ion groups in the HR-AMS spectra. The mass fractions of the ion groups are also summarized in SI Table S6.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8298

than 70% of the total signals, indicating an abundance ofoxygenated functional groups. The CO2

+ ion generally originatesfrom acids and esters in the HR-AMS spectra.38 The high relativeintensity of CO2

+ in the HR-AMS spectra of the fractions F5 andF6 (on average 20% and 25%, respectively) suggests thatcarboxylic acids and/or carboxylate esters are abundant in thesefractions. Because the carboxylic acid group was not found inabundance in F5 in the FT-IR analysis, a large fraction of theCO2

+ signals from F5 may have originated from carboxylateesters and other oxygenated compounds, such as organichydroperoxides.50

Nitrogen-containing ions CxHyOzNm+, CxHyNm

+ and NmOz+

were observed in the HR-AMS spectra of the extracts. TheCxHyNm

+ ions, attributed to amines, accounted for an average of<0.1%, 0.1%, 0.2%, 0.7%, 2.2%, and 3.2% of the ions from theorganics in the mass spectra of the fractions F1, F2, F3, F4, F5,and F6, respectively, showing an increase with the increase in theO/C of the extracted organics. Furthermore, the ratios of NO+/NO2

+ in the HR-AMS spectra of the fractionated organic matter(5.1 ± 0.4) were higher than those of NH4NO3 (2.4 ± 0.2),confirming the presence of organic nitrates.51,52 The percentageof ions from the organic nitrates was highest in the case of F3(mean: 4%); this result agrees with that from the FT-IR spectra.The specific PAEs and PAHs molecules in the extract were

investigated using the HR-AMS spectra. The high-resolutionanalysis of the mass spectra for F2 showed the high signalintensities of the C7H4O

+ (m/z 104), C8H5O3+ (m/z 149),

C8H6O3+ (m/z 150) and C8H7O4

+ (m/z 167) ions (SI FigureS5). These ions in the EI mass spectra are generally attributed toPAEs,53 suggesting that F2 contained abundant PAEs. This isalso suggested by the FT-IR and 1H NMR spectra of F2. BecauseBis(2-ethylhexyl) phthalate (DEHP) has previously beenidentified as the predominant PAE in the atmospheric environ-ment,54 it may have been abundant in F2. The prominentfragment ion of C16H23O4

+ in the HR-AMS spectra of F2 is acharacteristic ion of DEHP, supporting the likelihood that DEHPwas abundant in F2. Furthermore, the possible fragment ions ofPAHs in the fractionated organic matter were investigated byprocessing the unit resolution AMS spectra in the m/z range of198−479, as per Dzepina et al.55 The PAH marker signals,including those at m/z 202, 215, 239, 252, and 276, were furtherseparated by processing the high-resolution spectra of thefractions F1−F4 (SI Figure S6). The ion-formula analysisindicated that at least a part of these signals was attributed to theC16H10

+, C17H11+, C19H12

+, C20H12+ and C22H12

+ ions (SI FigureS7). The analysis showed that PAHs accounted for an average of0.26%, 0.13%, 0.14%, and 0.11% of themasses of the fractions F1,F2, F3, and F4, respectively. The PAH ions in the mass spectrafor the fractions F3 and F4 likely originated from oxygenated-PAHs or nitro-PAHs because the organics in the fractions musthave a certain degree of polarity.56

Light-Absorption Characteristics of Extracts and TheirRelationship to the Chemical Structures. The fractionationmethod in this study determined the important light-absorptionorganic groups in WISOM. Figure 2b presents the relativecontributions of the aerosol extracts to the light absorption at 400nm. The average UV−visible spectra of the different aerosolextracts and the relative contributions of the extracted organicsfrom the weekly samples to the total light absorption in thewavelength range of 280−600 nm are also presented in Figure 6.The result indicated that the relative contributions to the lightabsorption were markedly different between different extractsand wavelengths. In the wavelength range of 250−400 nm (UV

region), the light absorption by the organic matter wasdominated by that of the water-soluble fraction, which accountedfor an average of 66% of the total light absorption of the extractsat 300 nm. By contrast, the light absorption in the wavelengthrange of 400−600 nm (visible region) was primarily due to theextracted water-insoluble fractions F1−F6, which accounted foran average of 64% of the light absorption at 600 nm. Hence,WISOM accounted for a large part of the total light absorption byorganic matter, and the contribution was higher than that ofwater-soluble organics in the visible region. This finding is in linewith a previous report by Zhang et al.; their result for the ambientaerosols in Los Angeles also suggested that water-insolubleorganics were more light-absorbing than water-soluble organicsin the visible frequencies.11 In Nagoya, although the lightabsorption by water-insoluble BrC was relatively more importantthan that by water-soluble BrC in the visible region, the total lightabsorption coefficients by the extracted aerosol organics in theatmosphere was low (0.43 and 0.08 Mm−1 at 400 and 500 nm,respectively). The absorption corresponds to an average of 0.099and 0.023 times the light absorption by EC at 400 and 500 nm,respectively, which were estimated by the method in Chen et al.(SI Figure S8).25

The light absorption of WISOM was dominated by themedium-polar fractions F4, F5, and F6. The sum of the lightabsorption by the three components in the visible region (≥400nm) accounted for more than 90% of the light absorption by thefractions F1−F6. In particular, an average of 62% of the lightabsorption by WISOM was attributed to F4. Furthermore, theMAEs of the fractions F4, F5, and F6 at 365 nm (on average 0.77,2.3, and 1.3 m2 g−1, respectively) were much higher than those of

Figure 6. (a) Mean of the light absorption of aerosol extracts and (b)mean of the relative contributions of the extracts to the total lightabsorption. (c) Mass absorption efficiencies (MAEs) of the aerosolextracts at the wavelengths of 280, 400, and 550 nm. The solid circlesand whiskers indicate the mean and standard deviation, respectively.The MAE values of the organics in the different extracts are alsosummarized in SI Table S7.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8299

the fractions F1−F3 or of the water-soluble extracts (<0.4 m2

g−1). The results show that the medium-polar fractions F4, F5,and F6 contain light-absorbing substances, whose massabsorption efficiencies were higher than those of the water-soluble BrC in Los Angeles, Atlanta and even Beijing (0.1−0.7m2 g−1 at 365 nm).11,57 Aromatic and polar functional groups,containing O or both O and N atoms, in the organic matter maycontribute to the light absorption of those low- and medium-polar fractions because these functional groups were moreabundant in the fractions F4, F5, and F6 than in the otherextracts, as shown by the structural characteristics identified bythe FT-IR, 1H NMR and HR-AMS analysis (SI Table S3). Thisresult is in line with the general knowledge that heteroatomic (Oand N) and aromatic groups in molecules can form conjugatedsystems such as quinone, nitrophenol, nitro-aromatic and N-heterocyclic structures, and lower excitation energies of electronsin the molecules and lead to light absorption in the visiblerange.27,28,58,59 In particular, F5 had the highest MAE value. Thisresult was likely due to the large contribution by quinones, thepresence of which is suggested by the FT-IR spectra, given thatthey are considered an important source of BrC in atmosphericparticulates.60−62 Note that the large contributions of WISOMand the subfractions F4 and F5 to the light-absorption in thevisible region were further confirmed by the analysis of a set oftwo-day samples (SI Figure S9). This result suggests that theircontributions are generally large, at least in the studied urbanarea.The light absorption in the studied organics may also be

caused by CT interactions. In the case of the WSM in Athens,GA, the CT complexes that formed from interactions betweenalcohol and carbonyl moieties accounted for approximately 50%of the absorption (300−600 nm) based on reduction experi-ments using NaBH4.

63,64 Because F4 greatly contributed to boththe light absorption and organic masses of WISOM, similarexperiments were performed for the F4 of two samples, whichresulted in 20−60% decreases in the light absorption in thewavelength range of 300−600 nm (SI Figure S10). This resultsuggests 20−60% contributions of the CT complexes to the lightabsorption of F4 because the decrease of the absorption waslikely caused by the loss of CT complexes through the reductionof the electron acceptor carbonyl groups. As seen in Figure 3k,the fraction F4 contained abundant aromatic and oxygenatedgroups (primarily CO and C−OH); the CT complexes mayhave formed between carbonyl groups as electron acceptors andas electron donors in other groups, for example, phenolic andaliphatic hydroxyl groups. The CT complexes in F4 may havebeen associated with the high molecular weights and branchedstructures of organics. The abundant branched structures in F4were identified from the FT-IR spectra, and the ESI mass spectraof the fraction F4 showed that the number-average molecularweight of the organic compounds with an m/z in the 50−1000range was higher (464−516 Da) than that in WSOM (316−372Da) (SI Figures S11and S12). The branched and large molecularstructures of the organics in F4 with aromatic and oxygenatedgroups may have led to the abundant presence of CT complexesand may have contributed to the light absorption in the visiblerange. Note that organometallic complexes may have alsocontributed to part of the light absorption by aerosol matter.Novelty of Results and Environmental Implications.

This study obtained the elemental composition, the relativeabundances of different functional groups, and the characteristicsof the fragments from the electron impact ionization of thesubfractions of WISOM, which provided novel quantitative

information related to the overall chemical structure character-istics of WISOM. The clarified characteristics, such as the largeabundance of high-molecular-weight and oxygenated com-pounds, provide basic information to elucidate the possibleeffects of WISOM on the physicochemical properties ofatmospheric particulates and therefore its potential environ-mental impacts. Furthermore, these characteristics might beuseful for the development and validation of a representation ofthe relation between the chemical structures of organic matterand their physicochemical properties, that is, the development ofQuantitative Structure Activity Relationships (QSAR). Fororganic aerosols, such representations have been proposed fordensity, viscosity, hygroscopicity parameter k, reaction rates withhydroxyl radicals, and phase distribution parameters.65−70

As expected, the O/C ratios and the relative abundances of thepolar functional groups (e.g., C−OH and CO) of the sixsubfractions of WISOM increased as their polarity increased.However, the relationship between the light absorptions and thepolarities of the organic matter is not as simple. The result warnsthat the light absorption of the brown carbon in the modelsmight not be represented accurately by the degree of theoxidation alone. The results showed that WISOM accounts for alarge part of the total light absorption by organics. Thecontribution in the visible region was noticeably greater thanthat of the water-soluble organics. The obtained values of lightabsorption and MAEs of the organic aerosol components mightserve as inputs for atmospheric radiative models and might beuseful to evaluate estimates of the direct radiative effect of BrC inmodel simulations. Although the total light absorption by theorganic aerosol components in the visible region probably doesnot greatly contribute to the total light absorption by aerosols inurban areas, the light absorption in the ultraviolet region isexpected to be important because the calculated MAEs increasedsharply toward shorter wavelengths. Consequently, WISOMwould affect photochemistry by accelerating the photolysis rateof aerosol organics71 and decreasing the concentrations of O3

and NO2.72 It is noteworthy that applying the obtained values of

light absorption and MAEs of the studied components toatmospheric aerosol particles requires an approximation; TheMAEs in the matrices of the aerosol components might not beidentical to those in solutions because of the differentintermolecular interactions and the morphology of the ambientaerosol particles.The light-absorption properties of WISOM have been

characterized: the medium-polar fractions (F4, F5, and F6)had high MAEs and showed strong light absorption. Theseresults underscore the potential importance of the WISOMfractions for organic-phase photoreactions in particles,73 whichcan alter the chemical composition and therefore various otherproperties of organic aerosols. The light absorption by themedium-polar fractions might be associated with aromatic andheteroatomic (O and N) groups. Further studies must beconducted for these compound groups, including investigationsof the abundances and sources in different environments. Thesewater-insoluble organic fractions might be important reactiveoxygen species, given that the oxidative potential of aerosolextracts has been reported to be positively correlated with thelight absorbance of the extracts.74 The analytical approachpresented herein, using solvent extraction and normal-phasesolid-phase extraction, is important for the study of WISOMtoxicity.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8300

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.est.7b01630.

Additional details, including the methods for thedetermination of molecular weight distribution and theassessment of the nonuniformity of aerosol samplescollected on filters, are found here. Figures include theFT-IR spectra of the five fractions of organic matter for theblanks, thermograms of the thermal-optical analysis forextracted organic carbon, 1H−1H correlation spectra ofthe organic matter for F4, Van Krevelen diagram of thefractionated organics, PAEs and PAHmarker signals in theHR-AMS spectra, wavelength-resolved light spectra by theextracts and EC, light absorption of the fractionatedorganic matter for a set of two-day samples, absorptionspectra of the reduced F4, ESI-QTOF mass spectrum ofF4 and WSOM and variations of the light-absorption perextract. Tables showing the concentrations of carbon inthe aerosols and six subfractions of WISOM, the massfractions of the major functional groups and ion groups inthe extracted organic fractions from the FT-IR, 1H NMRand HR-AMS spectra, the mass absorption efficiency ofthe extracted organic fractions, and the distribution of thetotal carbon on quartz-fiber filters are also included (PDF)

■ AUTHOR INFORMATIONCorresponding Author*Phone/fax: 052-788-6157; e-mail: mochida.michihiro@g.mbox.nagoya-u.ac.jp.ORCIDMichihiro Mochida: 0000-0001-9557-5138NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank Lynn M. Russell and Satoshi Takahama for providingFT-IR peak pattern data for carboxylic COH. We also thank theResearch Center for Materials Science, Nagoya University, forthe use of the FT-IR, NMR and ESI-MS, and Kinichi Oyama andYuwu Maeda for the technical support in the use of theinstruments. The distribution of the aerosol samples on the filterswas clarified thanks to a related analysis by Tomoki Nakayamaand Hiroaki Fujinari. This work was supported in part by JSPSKAKENHI (grant numbers: JP25281002, JP24221001, andJP16F16760).

■ REFERENCES(1) Zhang, Q.; Jimenez, J. L.; Canagaratna, M. R.; Allan, J. D.; Coe, H.;Ulbrich, I. M.; Alfarra, M. R.; Takami, A.; Middlebrook, A.M.; Sun, Y. L.;Dzepina, K.; Dunlea, E.; Docherty, K.; DeCarlo, P. F.; Salcedo, D.;Onasch, T. B.; Jayne, J. T.; Miyoshi, T.; Shimono, A.; Hatakeyama, S.;Takegawa, N.; Kondo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S. L.;Weimer, S.; Demerjian, K.; Williams, P.; Bower, K. N.; Behreini, R.;Cottrell, L.; Griffin, R. J.; Rautiainen, J.; Sun, J. Y.; Zhang, Y. M.;Worsnop, D. R. Ubiquity and dominance of oxygenated species inorganic aerosols in anthropogenically-influenced northern hemispheremidlatitudes. Geophys. Res. Lett. 2007, 34 (13), 1−12.(2) Jimenez, J. L.; Canagaratna, M. R.; Donahue, N. M.; Prevot, A. S.H.; Zhang, Q.; Kroll, J. H.; DeCarlo, P. F.; Allan, J. D.; Coe, H.; Ng, N. L.Evolution of organic aerosols in the atmosphere. Science 2009, 326,1525−1529.

(3) Poschl, U. Atmospheric aerosols: composition, transformation,climate and health effects. Angew. Chem., Int. Ed. 2005, 44 (46), 7520−7540.(4) Hamilton, J. F.; Webb, P. J.; Lewis, A. C.; Hopkins, J. R.; Smith, S.;Davy, P. Partially oxidised organic components in urban aerosol usingGCXGC-TOF/MS. Atmos. Chem. Phys. 2004, 4, 1279−1290.(5) Andreae, M. O. A new look at aging aerosols. Science 2009, 326,1525−1529.(6) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.;Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.;Goldstein, A. H.; Hamilton, J. F.; Herrmann, H.; Hoffman, T.; Linuma,Y.; Jang, M.; Jenkin, M.; Jimenez, J. L.; Kiendler-Scharr, A.; Maenhaut,W.;McFiggans, G.;Mentel, Th. F.; Monod, A.; Prevot, A. S. H.; Seinfeld,J. H.; Surratt, J. D.; Szmigielski, R.; Wildt, J. The formation, propertiesand impact of secondary organic aerosol: Current and emerging issues.Atmos. Chem. Phys. 2009, 9, 5155−5236.(7) Laskin, A.; Laskin, J.; Nizkorodov, S. A. Chemistry of AtmosphericBrown Carbon. Chem. Rev. 2015, 115, 4335−4382.(8) Moise, T.; Flores, J. M.; Rudich, Y. Optical Properties of SecondaryOrganic Aerosols and Their Changes by Chemical Processes.Chem. Rev.2015, 115, 4400−4439.(9) Gysel, M.; Weingartner, E.; Nyeki, S.; Paulsen, D.; Baltensperger,U.; Galambos, I.; Kiss, G. Hygroscopic properties of water-solublematter and humic-like organics in atmospheric fine aerosol. Atmos.Chem. Phys. 2004, 4, 35−50.(10) Verma, V.; Fang, T.; Xu, L.; Peltier, R. E.; Russell, A. G.; Ng, N. L.;Weber, R. J. Organic aerosols associated with the generation of reactiveoxygen species (ROS) by water-soluble PM2.5. Environ. Sci. Technol.2015, 49 (7), 4646−4656.(11) Zhang, X.; Lin, Y. H.; Surratt, J. D.; Weber, R. J. Sources,composition and absorption Ångstrom exponent of light-absorbingorganic components in aerosol extracts from the Los Angeles basin.Environ. Sci. Technol. 2013, 47 (8), 3685−3693.(12) Utry, N.; Ajtai, T.; Filep, A.; Pinter, M. D.; Hoffer, A.; Bozokia, Z.;Szabo, G. Mass specific optical absorption coefficient of HULIS aerosolmeasured by a four-wavelength photoacoustic spectrometer at NIR, VISand UV wavelengths. Atmos. Environ. 2013, 69, 321−324.(13) Chen, Q.; Miyazaki, Y.; Kawamura, K.; Matsumoto, K.; Coburn,C.; Volkamer, R.; Iwamoto, Y.; Kagami, S.; Deng, Y.; Ogawa, S.;Ramasamy, S.; Kato, S.; Ida, A.; Kajii, Y.; Mochida, M. Characterizationof Chromophoric Water-Soluble Organic Matter in Urban, Forest, andMarine Aerosols by HR-ToF-AMS Analysis and Excitation−EmissionMatrix Spectroscopy. Environ. Sci. Technol. 2016, 50 (19), 10351−10360.(14) Willoughby, A. S.; Wozniak, A. S.; Hatcher, P. G. A molecular-level approach for characterizing water-insoluble components ofambient organic aerosol particulates using ultrahigh-resolution massspectrometry. Atmos. Chem. Phys. 2014, 14, 10299−10314.(15) Weber, R. J.; Sullivan, A. P.; Peltier, R. E.; Russell, A.; Yan, B.;Zheng, M.; de Gouw, J.; Warneke, C.; Brock, C.; Holloway, J. S.; Atras,E. L.; Edgerton, E. A study of secondary organic aerosol formation in theanthropogenic-influenced southern United States. J. Geophys. Res. 2007,112 (D13302), DOI: n/a10.1029/2007JD008408.(16) Facchini, M. C.; Rinaldi, M.; Decesari, S.; Carbone, C.; Finessi, E.;Mircea, M.; Fuzzi, S.; Ceburnis, D.; Flanagan, R.; Nilsson, E. D.; deLeeuw, G.; Martino, M.; Woeltjen, J.; O’Dowd, C. D. Primarysubmicron marine aerosol dominated by insoluble organic colloidsand aggregates. Geophys. Res. Lett. 2008, 35 (L17814), DOI: 10.1029/2008GL034210.(17) Mihara, T.; Mochida, M. Characterization of solvent-extractableorganics in urban aerosols based on mass spectrum analysis andhygroscopic growth measurement. Environ. Sci. Technol. 2011, 45 (21),9168−9174.(18) Favez, O.; Sciar, J.; Cachier, H.; Alfaro, S. C.; Abdelwahab, M. M.Significant formation of water-insoluble secondary organic aerosols insemi-arid urban environment. Geophys. Res. Lett. 2008, 35 (L15801),DOI: 10.1029/2008GL034446.(19) Alama, M. S.; Westb, C. E.; Scarlettb, A. G.; Rowlandb, S. J.;Harrison, R. M. Application of 2D-GCMS reveals many industrial

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8301

chemicals in airborne particulate matter. Atmos. Environ. 2013, 65, 101−111.(20) Kalberer, M.; Sax, M.; Samburova, V. Molecular size evolution ofoligomers in organic aerosols collected in urban atmospheres andgenerated in a smog chamber. Environ. Sci. Technol. 2006, 40 (19),5917−5922.(21) Tan, Y.; Lim, Y. B.; Altieri, K. E.; Seitzinger, S. P.; Turpin, B. J.Mechanisms leading to oligomers and SOA through aqueous photo-oxidation: insights from OH radical oxidation of acetic acid andmethylglyoxal. Atmos. Chem. Phys. 2012, 12, 801−813.(22) Kourtchev, I.; Doussin, J. F.; Giorio, C.; Mahon, B.; Wilson, E. M.;Maurin, N.; Pangui, E.; Venables, D. S.; Wenger, J. C.; Kalberer, M.Molecular composition of fresh and aged secondary organic aerosolfrom a mixture of biogenic volatile compounds: a high-resolution massspectrometry study. Atmos. Chem. Phys. 2015, 15, 5683−5695.(23) Simoneit, B. R. T.; Bi, X.; Oros, D. R.; Medeiros, P. M.; Sheng, G.;Fu, J. Phenols and hydroxy-PAHs (Arylphenols) as tracers for coalsmoke particulate matter: source tests and ambient aerosol assessments.Environ. Sci. Technol. 2007, 41 (21), 7294−7302.(24) Chan, A. W. H.; Isaacman, G.; Wilson, K. R.; Worton, D. R.;Ruehl, C. R.; Nah, T.; Gentner, D. R.; Dallman, T. R.; Kirchstetter, T.W.; Harley, R. A.; Gilman, J. B.; de Gouw, J. A.; Offenberg, J. H.;Kleindienst, T. E.; Lin, Y. H.; Rubitschun, C. L.; Surratt, J. D.; Goldstein,A. H. Detailed chemical characterization of unresolved complexmixtures (UCM) in atmospheric organics: Insights into emissionsources and atmospheric processing. J. Geophys. Res. 2013, 118, 2169−8996.(25) Chen, Q.; Ikemori, F.; Mochida, M. Light Absorption andExcitation−Emission Fluorescence of Urban Organic Aerosol Compo-nents and Their Relationship to Chemical Structure. Environ. Sci.Technol. 2016, 50 (20), 10859−10868.(26) Nocun, M. S.; Schantz, M. M. Determination of selectedoxygenated polycyclic aromatic hydrocarbons (oxy-PAHs) in diesel andair particulate matter standard reference materials (SRMs). Anal.Bioanal. Chem. 2013, 405 (16), 5583−5593.(27) Lin, P.; Aiona, P. K.; Li, Y.; Shiraiwa, M.; Laskin, J.; Nizkorodov, S.A.; Laskin, A. Molecular Characterization of Brown Carbon in BiomassBurning Aerosol Particles. Environ. Sci. Technol. 2016, 50 (21), 11815−11824.(28) Bluvshtein, N.; Lin, L.; Flores, J. M.; Segev, L.; Mazar, Y.; Tas, E.;Snider, G.; Weagle, C.; Brown, S. S.; Laskin, A.; Rudich, Y. Broadbandoptical properties of biomass-burning aero soland identification ofbrown carbon chromophores. J. Geophys. Res. 2017, 122, 122.(29) Snyder, L. R. Classification of the solvent properties of commonliquids. J. Chromatogr A 1974, 92 (2), 223−230.(30) Chow, J. C.; Watson, J. G.; Crow, D.; Lowenthal, D. H.;Merrifield, T. Comparison of IMPROVE and NIOSH CarbonMeasurements. Aerosol Sci. Technol. 2001, 34, 23−34.(31) Chen, Q.; Ikemori, F.; Higo, H.; Asakawa, D.; Mochida, M.Insight into chemical characteristics of aerosol HULIS and otherfractionated complex matters using mass spectrometry and FT-IRspectroscopy. Environ. Sci. Technol. 2016, 50 (4), 1721−1730.(32) Takahama, S.; Johnson, A.; Russell, L. M. Quantification ofcarboxylic and carbonyl functional groups in organic aerosol infraredabsorbance spectra. Aerosol Sci. Technol. 2013, 47 (3), 310−325.(33) Duarte, R. M. B. O.; Duarte, A. C. Unraveling the structuralfeatures of organic aerosols by NMR spectroscopy: a review. Magn.Reson. Chem. 2015, 53 (9), 658−666.(34) Chalbot, M. C. G.; Kavouras, I. G. Nuclear magnetic resonancespectroscopy for determining the functional content of organic aerosols:A review. Environ. Pollut. 2014, 191, 232−249.(35) Seaton, P. J.; Kieber, R. J.; Willey, J. D.; Brooks Avery, G., Jr.;Dixon, J. L. Seasonal and temporal characterization of dissolved organicmatter in rainwater by proton nuclear magnetic resonance spectroscopy.Atmos. Environ. 2013, 65, 52−60.(36) Banfi, D.; Patiny, L. www.nmrdb.org: Resurrecting and ProcessingNMR Spectra On-line. Chimia 2008, 62, 280−281.10.2533/chi-mia.2008.280

(37) Aiken, A. C.; DeCarlo, P. F.; Kroll, J. H.; Worsnop, D. R.;Huffman, J. A.; Docherty, K.; Ulbrich, I. M.; Mohr, C.; Kimmel, J. R.;Sueper, D.; et al. O/C and OM/OC ratios of primary, secondary, andambient organic aerosols with a high resolution time-of-flight aerosolmass spectrometer. Environ. Sci. Technol. 2008, 42 (12), 4478−4485.(38) Canagaratna, M. R.; Jimenez, J. L.; Kroll, J. H.; Chen, Q.; Kessler,S. H.; Massoli, P.; Hildebrandt Ruiz, L.; Fortner, E.; Williams, L. R.;Wilson, K. R.; Surratt, J. D.; Donahue, N. M.; Jayne, J. T.; Worsnop, D.R. Elemental ratio measurements of organic compounds using aerosolmass spectrometry: characterization, improved calibration, andimplications. Atmos. Chem. Phys. 2015, 15, 253−272.(39) Han, Y.; Kawamura, K.; Chen, Q.; Mochida, M. Formation ofhigh-molecular-weight compounds via the heterogeneous reactions ofgaseous C8−C10 n-aldehydes in the presence of atmospheric aerosolcomponents. Atmos. Environ. 2016, 126, 290−297.(40)Ma, J.; del Vecchio, R.; Golanoski, K. S.; Boyle, E. S.; Blough, N. V.Optical properties of humic substances and CDOM: Effects ofborohydride reduction. Environ. Sci. Technol. 2010, 44, 5395−5402.(41) Yu, J. Z.; Xu, J.; Yang, H. Charring characteristics of atmosphericorganic particulate matter in thermal analysis. Environ. Sci. Technol.2002, 36, 754−761.(42) Colthup, N. B.; Daly, L. H.; Wiberley, S. E. Introduction to infraredand Raman spectroscopy; Academic press: New York and London, 1964.(43) Das, M. R.; Mahiuddin, S. Kinetics and adsorption behaviour ofbenzoate and phthalate at the alpha-alumina−water interface: Influenceof functionality. Colloids Surf., A 2005, 264, 90−100.(44) Salim, C. J.; Liu, H.; Kennedy, J. F. Comparative study of theadsorption on chitosan beads of phthalate esters and their degradationproducts. Carbohydr. Polym. 2010, 81, 640−644.(45) Toja, F.; Saviello, D.; Nevin, A.; Comelli, D.; Lazzari, M.; Levi, M.;Toniolo, L. The degradation of poly(vinyl acetate) as a material fordesign objects: A multi-analytical study of the effect of dibutyl phthalateplasticizer. Part 1. Polym. Degrad. Stab. 2012, 97, 2441−2448.(46) Bruns, E. A.; Perraud, V.; Zelenyuk, A.; Ezell, M. J.; Johnson, S. N.;Yu, Y.; Imre, D.; Finlayson-Pitts, B. J.; Alexander, M. L. Comparison ofFTIR and particle mass spectrometry for the measurement of particulateorganic nitrates. Environ. Sci. Technol. 2010, 44 (3), 1056−1061.(47) Liu, S.; Shilling, J. E.; Song, C.; Hiranuma, N.; Zaveri, R. A.;Russell, L. M. Hydrolysis of organonitrate functional groups in aerosolparticles. Aerosol Sci. Technol. 2012, 46 (12), 1359−1369.(48) Popovicheva, O.; Kistler, M.; Kireeva, E.; Persiantseva, N.;Timofeev, M.; Kopeikin, V.; Kasper-Giebl, A. Physicochemicalcharacterization of smoke aerosol during large-scale wildfires: Extremeevent of August 2010 in Moscow. Atmos. Environ. 2014, 96, 405−414.(49) Ramesh, P.; Sampath, S. Electrochemical and spectroscopiccharacterization of quinone functionalized exfoliated graphite. Analyst2001, 126, 1872−1877.(50) Lim, Y. B.; Turpin, B. J. Organic peroxide and OH formation inaerosol and cloud water. Atmos. Chem. Phys. Discuss. 2015, 15, 17367−17396.(51) Farmer, D. K.; Matsunaga, A.; Docherty, K. S.; Surratt, J. D.;Seinfeld, J. H.; Ziemann, P. J.; Jimenez, J. L. Response of an aerosol massspectrometer to organonitrates and organosulfates and implications foratmospheric chemistry. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 6670−6675.(52) Hao, L.; Romakkaniemi, S.; Kortelainen, A.; Jaatinen, A.; Portin,H.; Miettinen, P.; Komppula, M.; Leskinen, A.; Virtanen, A.; Smith, J.N.; Sueper, D.; Worsnop, D. R.; Lehtinen, K. E. J.; Laaksonen, A.Aerosol chemical composition in cloud events by high resolution time-of-flight aerosol mass spectrometry. Environ. Sci. Technol. 2013, 47 (6),2645−2653.(53) Wu, P. G.; Yang, D. J.; Zhang, L. Q.; Shen, X. H.; Pan, X. D.;Wang, L. Y.; Zhang, J.; Tan, Y.; Feng, L.; Ying, Y. Simultaneousdetermination of 17 phthalates esters in edible vegetable oils by GC-MSwith silica/PSA-mixed solid-phase extraction. J. Sep. Sci. 2012, 35,2932−2939.(54) Net, S.; Sempere, R.; Delmont, A.; Paluselli, A.; Ouddane, B.Occurrence, fate, behavior and ecotoxicological state of phthalates in

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8302

different environmental matrices. Environ. Sci. Technol. 2015, 49 (7),4019−4035.(55) Dzepina, K.; Arey, J.; Marr, L. C.; Worsnop, D. R.; Salcedo, D.;Zhang, Q.; Onasch, T. B.; Molina, L. T.; Molina, M. J.; Jimenez, J. L.Detection of particle-phase polycyclic aromatic hydrocarbons in MexicoCity using an aerosol mass spectrometer. Int. J. Mass Spectrom. 2007,263, 152−170.(56) Musa Bandowe, B. A.; Meusel, H.; Huang, R.; Ho, K.; Cao, J.;Hoffmann, T.; Wilcke, W. PM2.5-bound oxygenated PAHs, nitro-PAHsand parent-PAHs from the atmosphere of a Chinese megacity: Seasonalvariation, sources and cancer risk assessment. Sci. Total Environ. 2014,473−474, 77−87.(57) Cheng, Y.; He, K. B.; Zheng, M.; Duan, F. K.; Du, Z. Y.; Ma, Y. L.;Tan, J. H.; Yang, F. M.; Liu, J. M.; Zhang, X. L.; Weber, R. J.; Bergin, M.H.; Russell, A. G. Mass absorption efficiency of elemental carbon andwater-soluble organic carbon in Beijing, China.Atmos. Chem. Phys. 2011,11, 11497−11510.(58) Lin, P.; Liu, J.; Shilling, J. E.; Kathmann, S.; Laskin, J.; Laskin, A.Molecular characterization of brown carbon (BrC) chromophores insecondary organic aerosol generated from photo-oxidation of toluene.Phys. Chem. Chem. Phys. 2015, 17, 23312−23325.(59) De Haan, D. O.; Hawkins, L. N.; Kononenko, J. A.; Turley, J. J.;Corrigan, A. L.; Tolbert, M. A.; Jimenez, J. L. Formation of nitrogen-containing oligomers by methylglyoxal and amines in simulatedevaporating cloud droplets. Environ. Sci. Technol. 2011, 45 (3), 984−991.(60) Cory, R. M.; McKnight, D. M. Fluorescence spectroscopy revealsubiquitous presence of oxidized and reduced quinones in dissolvedorganic matter. Environ. Sci. Technol. 2005, 39 (21), 8142−8149.(61) Chang, J. L.; Thompson, J. E. Characterization of coloredproducts formed during irradiation of aqueous solutions containingH2O2 and phenolic compounds. Atmos. Environ. 2010, 44, 541−551.(62) Mladenov, N.; Alados-Arboledas, L.; Olmo, F. J.; Lyamani, H.;Delgado, A.; Molina, A.; Reche, I. Applications of optical spectroscopyand stable isotope analyses to organic aerosol source discrimination inan urban area. Atmos. Environ. 2011, 45, 1960−1969.(63) Phillips, S. M.; Smith, G. D. Light absorption by charge transfercomplexes in brown carbon aerosols. Environ. Sci. Technol. Lett. 2014, 1,382−386.(64) Phillips, S. M.; Smith, G. D. Further evidence for charge transfercomplexes in brown carbon aerosols from excitation−emission matrixfluorescence spectroscopy. J. Phys. Chem. A 2015, 119, 4545−4551.(65) Kuwata, M.; Zorn, S. R.; Martin, S. T. Using Elemental Ratios toPredict the Density of Organic Material Composed of Carbon,Hydrogen, and Oxygen. Environ. Sci. Technol. 2012, 46 (2), 787−794.(66) Wang, C.; Goss, K. U.; Lei, Y. D.; Abbatt, J. P. D.; Wania, F.Calculating Equilibrium Phase Distribution during the Formation ofSecondary Organic Aerosol Using COSMO therm. Environ. Sci. Technol.2015, 49 (14), 8585−8594.(67) Monod, A.; Doussin, J. F. Structure-activity relationship for theestimation of OH-oxidation rate constants of aliphatic organiccompounds in the aqueous phase: alkanes, alcohols, organic acids andbases. Atmos. Environ. 2008, 42, 7611−7622.(68) Raventos-Duran, T.; Camredon, M.; Valorso, R.; Mouchel-Vallon, C.; Aumont, B. Structure-activity relationships to estimate theeffective Henry’s law constants of organics of atmospheric interest.Atmos. Chem. Phys. 2010, 10, 7643−7654.(69) Wania, F.; Lei, Y. D.; Wang, C.; Abbatt, J. P. D.; Goss, K. U. Novelmethods for predicting gas−particle partitioning during the formation ofsecondary organic aerosol. Atmos. Chem. Phys. 2014, 14, 13189−13204.(70) Wei, Y.; Cao, T.; Thompson, J. E. The chemical evolution &physical properties of organic aerosol: A molecular structure basedapproach. Atmos. Environ. 2012, 62, 199−207.(71) Hodzic, A.; Kasibhatla, P. S.; Jo, D. S.; Cappa, C.; Jimenez, J. L.;Madronich, S.; Park, R. J. Rethinking the global secondary organicaerosol (SOA) budget: stronger production, faster removal, shorterlifetime. Atmos. Chem. Phys. 2016, 16, 7917−7941.

(72) Jo, D. S.; Park, R. J.; Lee, S.; Kim, S. W.; Zhang, X. A globalsimulation of brown carbon: implications for photochemistry and directradiative effect. Atmos. Chem. Phys. 2016, 16, 3413−3432.(73) Smith, J. D.; Sio, V.; Yu, L.; Zhang, Q.; Anastasio, C. SecondaryOrganic Aerosol Production from Aqueous Reactions of AtmosphericPhenols with an Organic Triplet Excited State. Environ. Sci. Technol.2014, 48 (2), 1049−1057.(74) Verma, V.; Rico-Martinez, R.; Kotra, N.; King, L.; Liu, J.; Snell, T.W.; Weber, R. J. Contribution of water-soluble and insolublecomponents and their hydrophobic/hydrophilic subfractions to thereactive oxygen species-generating potential of fine ambient aerosols.Environ. Sci. Technol. 2012, 46 (20), 11384−11392.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b01630Environ. Sci. Technol. 2017, 51, 8293−8303

8303