Highly Oxidized Multifunctional Organic Compounds Observed in

Article pubs.acs.org/est

Highly Oxidized Multifunctional Organic Compounds Observed in Tropospheric Particles: A Field and Laboratory Study Anke Mutzel,† Laurent Poulain,† Torsten Berndt,† Yoshiteru Iinuma,† Maria Rodigast,† Olaf Böge,† Stefanie Richters,† Gerald Spindler,† Mikko Sipila,̈ ‡ Tuija Jokinen,‡ Markku Kulmala,‡ and Hartmut Herrmann*,† †

Leibniz Institute for Tropospheric Research (TROPOS), Permoserstraße 15, D-04318 Leipzig, Germany Department of Physics, University of Helsinki, Post Office Box 64, 00014 Helsinki, Finland

‡

S Supporting Information *

ABSTRACT: Very recent studies have reported the existence of highly oxidized multifunctional organic compounds (HOMs) with O/C ratios greater than 0.7. Because of their low vapor pressure, these compounds are often referred as extremely low-volatile organic compounds (ELVOCs), and thus, they are able to contribute significantly to organic mass in tropospheric particles. While HOMs have been successfully detected in the gas phase, their fate after uptake into particles remains unclear to date. Hence, the present study was designed to detect HOMs and related oxidation products in the particle phase and, thus, to shed light on their fate after phase transfer. To this end, aerosol chamber investigations of α-pinene ozonolysis were conducted under near environmental precursor concentrations (2.4 ppb) in a continuous flow reactor. The chemical characterization shows three classes of particle constituents: (1) intact HOMs that contain a carbonyl group, (2) particle-phase decomposition products, and (3) highly oxidized organosulfates (suggested to be addressed as HOOS). Besides chamber studies, HOM formation was also investigated during a measurement campaign conducted in summer 2013 at the TROPOS research station Melpitz. During this field campaign, gas-phase HOM formation was found to be correlated with an increase in the oxidation state of the organic aerosol.

■

their high O/C ratio.2 The fact that organic peroxides are an important class of compounds contributing to secondary organic aerosol (SOA) formation and growth has been demonstrated several times.13−17 On the basis of their high O/C ratio, they are suggested to have vapor pressures significantly lower than those of other known monoterpene oxidation products, such as pinic and pinonic acids. Therefore, it was speculated that these compounds contribute to both particle formation and growth to a large extent.4,6−10 Especially at low aerosol loadings, the condensation of HOMs might be able to explain the entire SOA mass.2 Thus, HOMs contribute significantly to the formation of organic mass during monoterpene oxidation. This finding is consistent with O/C measurement, because a correlation was found between the O/ C ratio determined from the formed organic mass and those of HOMs.2 Despite these observations, HOMs formed in the gas phase via autoxidation processes have not been detected in the particle phase beyond their contribution to organics identified

INTRODUCTION The existence of highly oxidized multifunctional organic compounds (HOMs) in the gas phase has been successfully demonstrated in both lab and field studies.1−5 Because of their high number of functional groups, it can be expected that these compounds are extremely low-volatile; thus, they are often referred as extremely low-volatile organic compounds (ELVOCs). HOMs and their importance in the early growth of freshly formed atmospheric aerosols were theoretically predicted in 1998 and have been investigated in recent times.6−10 Recent studies have revealed that the oxidation of biogenic volatile organic compounds (BVOCs) leads to the formation of HOMs with a molar yield of up to 7% (α-pinene ozonolysis).1,2 Consequently, HOM formation represents an important reaction channel of monoterpene degradation.1 Nevertheless, their formation and subsequent fate in the atmosphere are still under discussion. Very recently, the formation of HOMs in the gas phase was described as an autoxidation process of an alkylperoxy radical (RO2) via multiple intramolecular H atom shifts.2,5,11,12 Such RO2 radicals might be formed from both ozonolysis and OH radical reactions. Accordingly, it was hypothesized that HOMs contain multiple hydroperoxide groups, which also explains © XXXX American Chemical Society

Received: February 18, 2015 Revised: May 19, 2015 Accepted: May 26, 2015

A

DOI: 10.1021/acs.est.5b00885 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology Table 1. Experimental Conditions Used for the Investigation of HOM Formation seed particles 1 2 3 4 5 a

0.06 M 0.08 M 0.08 M organic

Na2SO4 NH4HSO4 NH4HSO4 seeda

O3 mean (ppb)

α-pineneini (ppb)

reaction time (h)

T (K)

RH (%)

19 20 21 18 21

2.5 2.4 2.5 2.3 2.5

48 48 48 48 2.5

292 293 292 292 293

≤1 ≤1 ≤1 ≤1 ≤1

remarks

acid-coated denuder experiments with HR-TOF-AMS

A total of 4.84 g of phthalic acid dipotassium salt [C6H4−1,2-(CO2K)2] and 4.08 g of potassium hydrogen phthalate (C8H5KO4).

Figure 1. Mass spectra obtained from CI-APi-TOF measurements at the Melpitz research station (red) and during the ozonolysis of α-pinene in the aerosol chamber LEAK (black). Compounds were detected as NO3− adducts. Because no OH radical scavenger was used, some signals also originate from OH radical reactions (see ref 5 for a detailed discussion).

directly analyzed and after derivatization with 2,4-dinitrophenylhydrazine (DNPH). For details, see SI 1 of the Supporting Information. Gas-Phase HOMs. Gas-phase HOMs were detected and quantified in LEAK and at the Melpitz research station as NO3− adducts (M + NO3−) using a NO3−−chemical ionization atmospheric pressure interface time-of-flight (CI-APi-TOF) mass spectrometer (Airmodus, Ltd., Tofwerk AG).2,18 Field Measurements. Field measurements were performed at the TROPOS research station at Melpitz [51.54° N, 12.93° E, 86 m above sea level (a.s.l.)], which is located approximately 50 km east of Leipzig, Germany. The site is mainly surrounded by agricultural pasture and forest. The atmospheric aerosol observed at Melpitz can be regarded as representative for central European regional conditions. Online instrumentation includes a HR-TOF-AMS (Aerodyne, Inc.) and a NO3−−CIAPi-TOF mass spectrometer, while off-line measurements were performed using DIGITEL PM10 high volume filter samples. For details, see SI 3 and SI 4 of the Supporting Information.

by high-resolution time-of-flight aerosol mass spectrometer (HR-TOF-AMS) measurements, and thus, their fate after phase transfer and their possible chemical consequences remain unclear. Consequently, the present study addresses this gap by examining both the phase transfer of HOMs into particles and their further reactions within the particle phase. To this end, a series of α-pinene ozonolysis experiments was conducted in the Leipziger Aerosolkammer (LEAK) aerosol chamber. The results of these laboratory observations were also compared to field measurements conducted at the Melpitz research station during summer 2013.

■

MATERIALS AND METHODS Aerosol Chamber LEAK. The aerosol chamber LEAK was used to examine HOM formation from α-pinene ozonolysis. The chamber was operated as a continuous flow reactor (CFR). This was necessary because it was found that, under low mass loadings, HOMs contribute significantly to the SOA mass.2 Experimental details can be found in Table 1 and SI 1 of the Supporting Information. Sample Analysis. Sample analysis of chamber-generated and ambient aerosol particles collected on filters was conducted by high-performance liquid chromatography electrospray ionization coupled to time-of-flight mass spectrometry [HPLC/(−)ESI−TOFMS] and ultra-performance liquid chromatography electrospray ionization coupled to time-of-flight mass spectrometry [UPLC/(−)IMS−QTOFMS]. Filters were

■

RESULTS Detection of HOMs in the Gas Phase. Gas-phase HOMs were detected both in aerosol chamber investigations of αpinene ozonolysis and during the field campaign conducted at the Melpitz research station. It can be seen from Figure 1 that similar m/z values were detected in both sets of experiments. This observation can be attributed to a high contribution of αB


Article

Environmental Science & Technology

Figure 2. Decomposition of a keto-hydroperoxide via the Korcek mechanism.

1) are extremely low-volatile and, thus, are expected to partition into the particle phase. Consequently, condensed HOMs or corresponding reaction products should be detectable in filter extracts. For this reason, a large set of filter samples was collected at the aerosol chamber LEAK, which was subsequently analyzed by HPLC/(−)ESI−TOFMS. Among numerous compounds identified from the analysis, short-chain carbonyl compounds, highly oxidized multifunctional carbonyl compounds, and organosulfates were found in significant amounts. These three groups of compounds will be discussed in the following sections separately. For details regarding sample preparation and chromatographic separation, see SI 1 and SI 2 of the Supporting Information. Highly Oxidized Multifunctional Carbonyl Compounds. After derivatization with DNPH, several compounds were detected in ambient and laboratory samples (Figure 3), and

pinene ozonolysis to the signals of the HOMs observed in the gas phase at the Melpitz research station. This is confirmed by BVOC measurements that identified α-pinene as the most abundant BVOC, with a maximum mixing ratio of up to 2.6 ppbv (see Figure S5 of the Supporting Information). Nevertheless, it should be pointed out that the experiments at the aerosol chamber LEAK were conducted under continuous flow conditions (CFR) to attain near environmental precursor concentrations (see SI 1 of the Supporting Information). The α-pinene mixing ratio in the aerosol chamber was 2.4 ppb, which is close to the mixing ratio measured in Melpitz. In addition to α-pinene, several other BVOCs (e.g., β-pinene and carene) and first-generation oxidation products (e.g., nopinone and campholene aldehyde) were positively identified at the Melpitz research station (see Figure S5 of the Supporting Information). It is likely that the reaction of these additionally volatile organic compounds (VOCs) also led to the formation of HOMs, which might explain the additional signals detected from the gas-phase measurements in Melpitz (e.g., m/z 302, 312, 316, 344, 354, 356, 370 [M + NO3]−). Minor differences in the spectra (Figure 1) can also be observed for the two most important signals detected at m/z 325 and 357 [M + NO3]−. These compounds belong most likely to RO2 radicals, which were detected in significant quantities in the α-pinene ozonolysis experiments conducted in the aerosol chamber but only in small amounts at the Melpitz research station.5 This difference can be attributed to the presence of NO and NO2 at the measurement station (see Figure S4 of the Supporting Information), because RO2 radicals react rapidly with NO and NO2 to form organic nitrates and peroxynitrates (see Figure S7 of the Supporting Information). This reaction might also explain the signals detected at m/z 307, 323, and 339 [M + NO3]−, which were identified as nitrogen-containing compounds.5,19 Most HOMs detected at the Melpitz research station and in chamber experiments were also reported by Ehn and coworkers.1 In particular, HOMs detected at m/z 298, 308, 340, 342, and 372 [M + NO3]− were also reported in the literature and were assigned to C8 H 12O 8 , C 10 H14 O 7, C 10 H 14O 9 , C10H16O9, and C10H14O11, respectively, based on the highresolution mass spectrometric data.1,2,4 Besides these compounds, three additional signals were found at m/z 294, 310, and 324 [M + NO3]−, which correspond to C10H16O6, C10H16O7, and C10H14O8, respectively. Because of the fact that these mentioned compounds were (i) detected in both ambient and laboratory measurements and (ii) reported in the literature, their detection in the particle phase was one main focus of the present study. Detection of HOMs in the Particle Phase. The HOMs detected by CI-APi-TOF measurements in the gas phase at the Melpitz research station and in chamber experiments (Figure

Figure 3. First detection of HOMs in the particle phase obtained from DNPH derivatization of chamber-generated SOA [base peak chromatogram (BPC) from m/z 430 to 660].

their chemical composition (Table 2) was compared to those reported for HOMs in the literature.1,2,4 From this comparison, it was found that HOMs reported in the literature were also found from the off-line analysis after DNPH derivatization (Figure 3). Furthermore, the majority of the HOMs detected from the ozonolysis of α-pinene in the LEAK experiments was also detected in ambient filters (Table 2). This is in good agreement with the results obtained from CI-APi-TOF measurements because comparable m/z values were detected from both LEAK and Melpitz gas-phase measurements (Figure 1). Only m/z 308, 338, 340, 356, and 358 [M + NO3]−, corresponding to C10H14O7, C10H12O9, C10H14O9, C10H14O10, and C10H16O10, respectively, were not detected in filter samples from Melpitz. It is most likely that their concentrations were too low to be detected by the applied off-line analysis. The fact that corresponding compounds were identified by CI-APi-TOF measurements (conducted in the gas phase) and the off-line analysis of particle-phase products supports the picture that C


Article


Table 2. Chemical Composition and Number of Carbonyl Groups Determined for HOMs Detected from Chamber-Generated SOA (LEAK) and Filter Collected at Melpitza detected in the particle phase

a

compound number

chemical formula

1 2 3 4 5 6 7 8 9 10 11 12 13

C8H12O8 C9H14O9 C9H16O9 C9H12O10 C10H14O7 C10H12O9 C10H14O9 C10H16O9 C10H14O10 C10H16O10 C10H16O10 C10H14O11 C10H16O11

Mw of carbonyl compound (+NO3) 236 266 268 280 246 276 278 280 294 296 296 310 312

detected by CI-APi-TOF

number of carbonyl groups

LEAK

Melpitz

LEAK

Melpitz

1 1 1 1 2 1 1 1 2 1 2 1 1

× × × × × × × × × × × × ×

× × × ×

× ×

× ×

× ×

× ×

× × × × × × ×

× × × × × × ×

(298) (328) (330) (342) (308) (338) (340) (342) (356) (358) (358) (372) (374)

× × × ×

All positively identified compounds were also reported by Ehn and co-workers.1,2

Decomposition of the HOM Molecules and the Formation of Short-Chain Carbonyl Compounds. Besides HOMs, a large amount of C1−C4 carbonyl compounds were detected in the particle phase (see a chromatogram given in Figure S2 of the Supporting Information). From the chemical analysis of chamber-generated filters and filters collected at Melpitz, formaldehyde, acetaldehyde, acetone, propionaldehyde, methylglyoxal, methylethylketone, and butyraldehyde were positively identified with authentic standard compounds as their corresponding DNPH derivatives (see SI 2.2 of the Supporting Information). Because of the high number of hydroperoxide groups in the HOM molecules, it is expected that these compounds decompose to yield further oxidation products once they condense into the particle phase.2 On the basis of this, we hypothesize that a certain amount of condensed HOMs decompose within the particle phase forming short-chain carbonyl compounds (C1−C4). This decomposition might proceed via the Korcek mechanism (Figure 2).25−28 Assuming that HOMs are carbonyl hydroperoxides, they might decompose via formation of a five-membered cyclic peroxide with a subsequent fragmentation of O−O and C−C bonds accompanied by a 1,2 H atom shift. The reaction via the Korcek mechanism will lead to the formation of a short-chain carbonyl compound and a carboxylic acid. Because the Korcek mechanism involves the formation of carboxylic acids, a connection might exist to C7−C9 carboxylic acids that were also detected from filter measurements. The detected C7−C9 carboxylic acids include important SOA marker compounds, such as terebic acid, terpenylic acid, norpinic acid, 3-methyl1,2,3-butanetricarboxylic acid, and pinic acid, which were identified with authentic standard compounds (for details, see SI 2.4 of the Supporting Information).29−33 The formation of SOA marker compounds has often been discussed, and several tentative mechanisms exist.30−35 Consequently, the proposed pathway via the Korcek mechanism connects the decomposition of HOMs with the formation of carbonyl compounds and carboxylic acids, both of which have been analytically identified within the present study. Highly Oxidized Organosulfates (HOOS). The analysis of underivatized filter extracts revealed the existence a number of organosulfates, which can be classified by their O/C ratio (see

HOMs are formed in the gas phase and afterward partition into the particle phase. A total of 13 compounds attributed to HOMs were detected after derivatization with DNPH (Table 2). The number of hydrazone groups added to the target compound reveals the number of carbonyl groups present in the molecule.20−23 Especially for the most abundant HOMs described by Zhao et al., at m/z 308, 340, 342, and 310 [M + NO3]− important structural information was obtained by DNPH derivatization (signals 5, 7, 8, and 12 in Figure 3).4 The compounds detected at m/z 340, 342, and 372 [M + NO3]− have been described to be involved in particle formation, whereas m/z 308 is suggested to be correlated to particle growth.4 For the first three compounds, corresponding DNPH derivatives were detected by LC/(−)ESI−TOFMS at m/z 457.0857 ([M − H]−, 17.0 min, hydrazone form of C16H17N4O12−), 459.1022 ([M − H]−, 16.0 min, hydrazone form of C16H19N4O12−), and 489.0747 ([M − H]−, 14.4 min, hydrazone form of C16H17N4O14−), which indicates monocarbonyl compounds. In contrast, for the m/z 308 (Mw + NO3−) compound, a corresponding signal was found at m/z 605.1234 ([M − H]−, 11.5 min, hydrazone form of C22H21N8O13), which indicates that it is a dicarbonyl compound. Up to now, the existence of carbonyl groups in HOMs was only the subject of speculation. The present study provides the first experimental evidence for the presence of carbonyl groups in the HOM skeleton. Three possible explanations can be given for the formation of carbonyl-group-containing compounds in the gas phase. First, the recombination of RO2 radicals can lead to the formation of carbonyl products through the RO2 + RO2 recombination reaction. Additionally, RO2 radicals can form alkoxy radicals, which also form carbonyl compounds via reaction with oxygen. Second, if NOx is present, RO2 radicals are also able to form carbonyl compounds, again via RO formation and subsequent reaction with O2. It is worth noting that this pathway might be off minor relevance for the aerosol chamber experiments because no NOx was introduced. Third, HOMs are proposed to be formed by autoxidation of VOCs; their propagation terminates by a H atom shift accompanied by a loss of an OH radical, which also leads to the formation of a carbonyl group.2,11,12,24 Therefore, it is suggested that HOMs are carbonyl-group-containing compounds.5 D


Article


the described mechanisms ii and iii, the formation of one HOOS can be explained (C10H14O11S). It is most likely that precursor compounds for additionally detected HOOS are not identified thus far and/or other mechanisms exist for HOOS formation. It should be noted that only those HOMs were considered as potential HOOS precursors, which were successfully detected in the particle phase by the off-line analyses applied (compounds summarized in Table 3). It cannot be excluded that more HOMs exist in the particle phase, which were not detected thus far. These unidentified HOMs might also act as HOOS precursor compounds. It should be noted that the most abundant HOOS was detected at m/z 239.0231, which corresponds to C7H11O7S− ([M − H]−). This HOOS has been described as a product of diketone hydrolysis with a subsequent esterification reaction with H2SO4.36 This mechanism seems to be unlikely, however, because none of the previously detected HOOS were detected from experiments conducted in the presence of H2SO4 seed particles. Furthermore, it can be seen from Table 3 that, in addition to C10 HOOS, several compounds with fewer carbon atoms (C7−C9) were detected. The existence of these compounds cannot be explained by the attack of HSO4− only. Instead, it requires a decomposition step, e.g., by the Korcek mechanism, followed by a nucleophilic attack of HSO4− with a subsequent loss of O2/HO2/OH, which could provide a feasible explanation for the formation of C7−C9 HOOS. Nevertheless, on the basis of the obtained data set, the exact reaction sequence cannot be discriminated. It is also possible that HOOS formation occurs first, which is then followed by decomposition. Furthermore, the existence of additional HOOS formation pathways cannot be ruled out. To prove the hypothesis that HOMs act as a precursor for HOOS, experiments were conducted using an acid-coated denuder (see SI 1 of the Supporting Information for more details). Several HOOS were detected from these acid-coated denuders, whereas no HOOS were found from the non-acidic denuder. Especially the most abundant HOOS (m/z 239.0231) was detected at a significant level from acid denuder extracts (see Figure S4 of the Supporting Information). Thus, it is very likely that HOOS are formed from the reaction of gas-phase compounds with acidic particle components. This is also supported by the finding that HOOS were not detected during experiments conducted in the presence of Na2SO4 and organic seed particles (see Figure S4 of the Supporting Information).

Table S3 of the Supporting Information). Compounds of class A have an O/C ratio smaller than 1.0, whereas class B is characterized by an O/C ratio greater than 1.0. The latter is addressed as HOOS (Table 3). The high average carbon oxidation state and the high O/C ratio of class B organosulfates indicate that the formation of these compounds is related to the presence of HOMs. Table 3. HOOS Detected from Chamber-Generated SOA, Filter Extracts Collected at Melpitz and an Acid-Coated Denudera detected in the particle phase

chemical formula

measured m/z [M − H]−

LEAK α-pinene

C7H11O7S C7H7O8S C7H9O8S C7H11O9S C8H9O8S C8H11O8S C8H17O8S C8H13O9S C9H13O8S C9H13O9S C10H15O10S C10H13O11S

239.0231 250.9889 253.0051 271.0150 265.0031 267.0189 273.0644 285.0280 281.0351 297.0280 327.0405 341.0209

× × × × × × × × × × × ×

Melpitz × × × × × × × ×

detected from an acid-coated denuder

potential precusor

× × × × × ×

C8H12O8

C9H14O9 C10H16O10 C10H14O11

a

A complete overview about all organosulfates (OS) is given in Table S3 of the Supporting Information.

When the detected HOMs act as HOOS precursors, the formation of the latter could proceed via a nucleophilic attack of HSO4− at a carbon atom. Subsequently, the formed intermediate can further react via loss of (i) O2 and HO2, (ii) O2, and/or (iii) O2 and OH. With respect to the hydroperoxide moieties suggested in the HOM molecules, an elimination of OH or O2 is feasible; e.g., the elimination of OH has been described to occur during the autoxidation process. With respect to the described mechanism i, in total, four HOOS (C8H12O8S, C9H14O9S, C10H16O10S, and C10H14O11S) can be explained by follow-up reactions of detected HOMs. By

Figure 4. (A) Total concentration of HOMs (black; see SI 1.3 of the Supporting Information) and particulate carbon oxidation state (OSc, purple) during selected days of the measurement campaign in Melpitz. (B) Scatter plot highlights the linear relationship that can be observed between the gas-phase HOM concentration and particulate OSc. E


Article


Figure 5. Time profile of α-pinene (gray), ozone (blue), OM (green), and HOM concentration (red). The highlighted region corresponds to the time with seed introduction (2 min). After introduction of seed particles, dilution is taking place, and thus, the concentration of highly oxidized organic compounds starts to increase again because of the missing condensational sink (gray shaded region).

a significant increase of organic mass in the particle phase. This indicates that formed HOMs (red) condense directly into the provided particle phase, resulting in a significant increase in organic mass (OM) (green). Fate of HOMs in the Particle Phase. We propose three reaction pathways of condensed HOMs based on the detection of highly oxidized carbonyl compounds, short-chain carbonyl compounds, and HOOS. HOMs can undergo (i) condensation into the particle phase without a structural change, (ii) fragmentation reactions leading to short-chain carbonyl compounds, and (iii) formation of HOOS. Several hints were found for all suggested pathways, indicating that these processes occur simultaneously. It should be noted that none of the detected HOOS, except one, have been reported in the literature up to now. Therefore, it can be stated that the condensation of HOMs leads to the formation of an up to now unknown class of compounds, i.e., HOOS in sulfate-containing acidic particles. In terms of the HOM structures, the analysis of the derivatized samples demonstrated that the vast majority of the HOMs contain one or more carbonyl groups. Thus, the hypothesis that HOMs contain carbonyl groups can be supported, at least for the compounds that were positively detected as DNPH derivatives. Thus, this study provides the first evidence for the presence of HOMs in laboratorygenerated and ambient SOA as well as the first experimentally obtained structural information on particle-phase HOMs. Atmospheric Implications. The present study clarified the fate of condensed HOMs, which are also often referred as ELVOCs. The off-line analysis of SOA in field samples and formed by CFR chamber experiments demonstrated the existence of intact HOMs, short-chain carbonyl compounds, and HOOS. HOOS and carbonyl-containing HOMs were also successfully detected in the ambient samples, which clearly shows the atmospheric relevance of the processes described by this study. It is expected that HOOS might play an important

However, even if HOOS formation pathways cannot be completely proven, the suggested mechanism is in accordance with the obtained data set. Furthermore, the majority of the HOOS detected during chamber experiments were also identified in ambient filter extracts. Thus, the HOOS detected in this study are an important particle constituent in both laboratory and field samples. Influence of HOMs in Organic Aerosol. Because HOMs represent low-volatile and highly oxidized compounds, they should directly influence the formation of particulate organic material (OM). To examine this issue, an aerosol mass spectrometer (AMS) was operated at the Melpitz research station and the average carbon oxidation state (OSc = 2O/C − H/C) was determined according to the method described by Kroll and co-workers using the oxygen/carbon ratio (O/C) and the hydrogen/carbon ratio (H/C).37 The OSc will necessarily increase upon oxidation, and thus, its determination enables a description of the evolution of organic species in the organic aerosol. As seen in Figure 4, a good agreement between particulate OSc and the total concentration of gas-phase HOMs was found, which suggests that these compounds might directly contribute to the change in particle-phase OSc (Figure 4A). This is particularly clearly seen for the first hours of the morning when the formation of HOMs is the strongest. During this period, a linear relationship between the total HOM concentration in the gas phase and particulate OSc was observed (Figure 4B). On the basis of this, it is expected that HOMs condense into the particle phase and, thus, make up a large fraction of the organic mass. To further investigate this issue, an AMS was connected to the aerosol chamber and the reaction was run under continuous flow conditions. In these sets of experiments, seed particles were introduced for 2 min after a stable concentration profile of HOMs, α-pinene, and ozone was reached. As seen from Figure 5, the introduction of seed particles led to a dramatic decrease in the concentration of gas-phase HOMs and, at the same time, F


Article


(6) Kulmala, M.; Toivonen, A.; Makela, J. M.; Laaksonen, A. Analysis of the growth of nucleation mode particles observed in Boreal forest. Tellus, Ser. B 1998, 50 (5), 449−462 DOI: 10.1034/j.16000889.1998.t01-4-00004.x. (7) Donahue, N. M.; Ortega, I. K.; Chuang, W.; Riipinen, I.; Riccobono, F.; Schobesberger, S.; Dommen, J.; Baltensperger, U.; Kulmala, M.; Worsnop, D. R.; Vehkamaki, H. How do organic vapors contribute to new-particle formation? Faraday Discuss. 2013, 165 (0), 91−104 DOI: 10.1039/C3FD00046J. (8) Riipinen, I.; Pierce, J. R.; Yli-Juuti, T.; Nieminen, T.; Hakkinen, S.; Ehn, M.; Junninen, H.; Lehtipalo, K.; Petaja, T.; Slowik, J.; Chang, R.; Shantz, N. C.; Abbatt, J.; Leaitch, W. R.; Kerminen, V. M.; Worsnop, D. R.; Pandis, S. N.; Donahue, N. M.; Kulmala, M. Organic condensation: A vital link connecting aerosol formation to cloud condensation nuclei (CCN) concentrations. Atmos. Chem. Phys. 2011, 11 (8), 3865−3878 DOI: 10.5194/acp-11-3865-2011. (9) Riipinen, I.; Yli-Juuti, T.; Pierce, J. R.; Petaja, T.; Worsnop, D. R.; Kulmala, M.; Donahue, N. M. The contribution of organics to atmospheric nanoparticle growth. Nat. Geosci. 2012, 5 (7), 453−458 DOI: 10.1038/ngeo1499. (10) Donahue, N. M.; Kroll, J. H.; Pandis, S. N.; Robinson, A. L. A two-dimensional volatility basis setPart 2: Diagnostics of organicaerosol evolution. Atmos. Chem. Phys. 2012, 12 (2), 615−634 DOI: 10.5194/acp-12-615-2012. (11) Rissanen, M. P.; Kurtén, T.; Sipilä, M.; Thornton, J. A.; Kangasluoma, J.; Sarnela, N.; Junninen, H.; Jørgensen, S.; Schallhart, S.; Kajos, M. K.; Taipale, R.; Springer, M.; Mentel, T. F.; Ruuskanen, T.; Petäjä, T.; Worsnop, D. R.; Kjaergaard, H. G.; Ehn, M. The formation of highly oxidized multifunctional products in the ozonolysis of cyclohexene. J. Am. Chem. Soc. 2014, 136 (44), 15596−15606 DOI: 10.1021/ja507146s. (12) Crounse, J. D.; Nielsen, L. B.; Jorgensen, S.; Kjaergaard, H. G.; Wennberg, P. O. Autoxidation of organic compounds in the atmosphere. J. Phys. Chem. Lett. 2013, 4 (20), 3513−3520 DOI: 10.1021/jz4019207. (13) Mutzel, A.; Rodigast, M.; Iinuma, Y.; Böge, O.; Herrmann, H. An improved method for the quantification of SOA bound peroxides. Atmos. Environ. 2013, 67, 365−369 DOI: 10.1016/j.atmosenv.2012.11.012. (14) Docherty, K. S.; Wu, W.; Lim, Y. B.; Ziemann, P. J. Contributions of organic peroxides to secondary aerosol formed from reactions of monoterpenes with O3. Environ. Sci. Technol. 2005, 39 (11), 4049−4059 DOI: 10.1021/es050228s. (15) Mertes, P.; Pfaffenberger, L.; Dommen, J.; Kalberer, M.; Baltensperger, U. Development of a sensitive long pathlength absorbance photometer to quantify peroxides in aerosol particles (Peroxide-LOPAP). Atmos. Meas. Technol. Discuss. 2012, 5 (1), 1431− 1457 DOI: 10.5194/amtd-5-1431-2012. (16) Chen, Q.; Liu, Y.; Donahue, N. M.; Shilling, J. E.; Martin, S. T. Particle-phase chemistry of secondary organic material: Modeled compared to measured O:C and H:C elemental ratios provide constraints. Environ. Sci. Technol. 2011, 45 (11), 4763−4770 DOI: 10.1021/es104398s. (17) Surratt, J. D.; Murphy, S. M.; Kroll, J. H.; Ng, N. L.; Hildebrandt, L.; Sorooshian, A.; Szmigielski, R.; Vermeylen, R.; Maenhaut, W.; Claeys, M.; Flagan, R. C.; Seinfeld, J. H. Chemical composition of secondary organic aerosol formed from the photooxidation of isoprene. J. Phys. Chem. A 2006, 110 (31), 9665−9690 DOI: 10.1021/jp061734m. (18) Jokinen, T.; Sipila, M.; Junninen, H.; Ehn, M.; Lonn, G.; Hakala, J.; Petaja, T.; Mauldin, R. L.; Kulmala, M.; Worsnop, D. R. Atmospheric sulphuric acid and neutral cluster measurements using CI-APi-TOF. Atmos. Chem. Phys. 2012, 12 (9), 4117−4125 DOI: 10.5194/acp-12-4117-2012. (19) Kulmala, M.; Kontkanen, J.; Junninen, H.; Lehtipalo, K.; Manninen, H. E.; Nieminen, T.; Petaja, T.; Sipila, M.; Schobesberger, S.; Rantala, P.; Franchin, A.; Jokinen, T.; Jarvinen, E.; Aijala, M.; Kangasluoma, J.; Hakala, J.; Aalto, P. P.; Paasonen, P.; Mikkila, J.; Vanhanen, J.; Aalto, J.; Hakola, H.; Makkonen, U.; Ruuskanen, T.;

role in particle growth, as described for less oxidized organosulfates.17 A comparison of the Melpitz field data set to the LEAK data set reveals that a number of identical HOMs were observed. The field study provides evidence that the gas−particle transfer of HOMs leads to a significantly higher oxidation state of ambient organic aerosol. This is also demonstrated by the chamber experiments, where the introduction of seed particles led to a decrease in the gas-phase HOM concentration and, at the same time, an increase in the particle-phase OM.

■

ASSOCIATED CONTENT

* Supporting Information S

Detailed description of all experiments conducted in the aerosol chamber, chromatographic separation, and CI-APi-TOFMS (SI 1), details about identified compounds (chromatograms, retention time, calculated chemical composition, and error of calculation) (SI 2), Melpitz research station and instruments used for the field campaign (SI3), and details about the meteorology, trace gas profiles (NOx, ozone, and SO2), BVOCs present at the station, and results from the AMS measurements are presented in (SI 4). The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b00885.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +49-341-2717-7024. Fax: +49-341-2717-7012. Email: [email protected]. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Ehn, M.; Kleist, E.; Junninen, H.; Petaja, T.; Lonn, G.; Schobesberger, S.; Dal Maso, M.; Trimborn, A.; Kulmala, M.; Worsnop, D. R.; Wahner, A.; Wildt, J.; Mentel, T. F. Gas phase formation of extremely oxidized pinene reaction products in chamber and ambient air. Atmos. Chem. Phys. 2012, 12 (11), 5113−5127 DOI: 10.5194/acp-12-5113-2012. (2) Ehn, M.; Thornton, J. A.; Kleist, E.; Sipila, M.; Junninen, H.; Pullinen, I.; Springer, M.; Rubach, F.; Tillmann, R.; Lee, B.; LopezHilfiker, F.; Andres, S.; Acir, I. H.; Rissanen, M.; Jokinen, T.; Schobesberger, S.; Kangasluoma, J.; Kontkanen, J.; Nieminen, T.; Kurten, T.; Nielsen, L. B.; Jorgensen, S.; Kjaergaard, H. G.; Canagaratna, M.; Dal Maso, M.; Berndt, T.; Petaja, T.; Wahner, A.; Kerminen, V. M.; Kulmala, M.; Worsnop, D. R.; Wildt, J.; Mentel, T. F. A large source of low-volatility secondary organic aerosol. Nature 2014, 506 (7489), 476−479 DOI: 10.1038/nature13032. (3) Mentel, T. F.; Springer, M.; Ehn, M.; Kleist, E.; Pullinen, I.; Kurtén, T.; Rissanen, M.; Wahner, A.; Wildt, J. Formation of highly oxidized multifunctional compounds: Autoxidation of peroxy radicals formed in the ozonolysis of alkenesDeduced from structure− product relationships. Atmos. Chem. Phys. Discuss. 2015, 15 (2), 2791− 2851 DOI: 10.5194/acpd-15-2791-2015. (4) Zhao, J.; Ortega, J.; Chen, M.; McMurry, P. H.; Smith, J. N. Dependence of particle nucleation and growth on high-molecularweight gas-phase products during ozonolysis of α-pinene. Atmos. Chem. Phys. 2013, 13 (15), 7631−7644 DOI: 10.5194/acp-13-76312013. (5) Jokinen, T.; Sipilä, M.; Richters, S.; Kerminen, V. M.; Paasonen, P.; Stratmann, F.; Worsnop, D. R.; Kulmala, M.; Ehn, M.; Herrmann, H.; Berndt, T. Rapid autoxidation forming highly oxidized RO2 radicals in the atmosphere. Angew. Chem., Int. Ed. 2014, 125 (25), 14825−14829. G


Article


particulate products. J. Atmos. Chem. 2001, 38 (3), 231−276 DOI: 10.1023/A:1006487530903. (33) Christoffersen, T. S.; Hjorth, J.; Horie, O.; Jensen, N. R.; Kotzias, D.; Molander, L. L.; Neeb, P.; Ruppert, L.; Winterhalter, R.; Virkkula, A.; Wirtz, K.; Larsen, B. R. cis-Pinic acid, a possible precursor for organic aerosol formation from ozonolysis of α-pinene. Atmos. Environ. 1998, 32 (10), 1657−1661 DOI: 10.1016/s1352-2310(97) 00448-2. (34) Müller, L.; Reinnig, M. C.; Naumann, K. H.; Saathoff, H.; Mentel, T. F.; Donahue, N. M.; Hoffmann, T. Formation of 3-methyl1,2,3-butanetricarboxylic acid via gas phase oxidation of pinonic acid A mass spectrometric study of SOA aging. Atmos. Chem. Phys. 2012, 12 (3), 1483−1496 DOI: 10.5194/acp-12-1483-2012. (35) Winterhalter, R.; Neeb, P.; Grossmann, D.; Kolloff, A.; Horie, O.; Moortgat, G. Products and mechanism of the gas phase reaction of ozone with β-pinene. J. Atmos. Chem. 2000, 35 (2), 165−197 DOI: 10.1023/a:1006257800929. (36) Surratt, J. D.; Gomez-Gonzalez, Y.; Chan, A. W. H.; Vermeylen, R.; Shahgholi, M.; Kleindienst, T. E.; Edney, E. O.; Offenberg, J. H.; Lewandowski, M.; Jaoui, M.; Maenhaut, W.; Claeys, M.; Flagan, R. C.; Seinfeld, J. H. Organosulfate formation in biogenic secondary organic aerosol. J. Phys. Chem. A 2008, 112 (36), 8345−8378 DOI: 10.1021/ jp802310p. (37) Kroll, J. H.; Donahue, N. M.; Jimenez, J. L.; Kessler, S. H.; Canagaratna, M. R.; Wilson, K. R.; Altieri, K. E.; Mazzoleni, L. R.; Wozniak, A. S.; Bluhm, H.; Mysak, E. R.; Smith, J. D.; Kolb, C. E.; Worsnop, D. R. Carbon oxidation state as a metric for describing the chemistry of atmospheric organic aerosol. Nat. Chem. 2011, 3 (2), 133−139 DOI: 10.1038/nchem.948.

Mauldin, R. L.; Duplissy, J.; Vehkamaki, H.; Back, J.; Kortelainen, A.; Riipinen, I.; Kurten, T.; Johnston, M. V.; Smith, J. N.; Ehn, M.; Mentel, T. F.; Lehtinen, K. E. J.; Laaksonen, A.; Kerminen, V. M.; Worsnop, D. R. Direct observations of atmospheric aerosol nucleation. Science 2013, 339 (6122), 943−946 DOI: 10.1126/science.1227385. (20) Kahnt, A.; Iinuma, Y.; Böge, O.; Mutzel, A.; Herrmann, H. Denuder sampling techniques for the determination of gas-phase carbonyl compounds: A comparison and characterisation of in situ and ex situ derivatisation methods. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2011, 879 (17−18), 1402−1411 DOI: 10.1016/ j.jchromb.2011.02.028. (21) Chi, Y. G.; Feng, Y. L.; Wen, S.; Lu, H. X.; Yu, Z. Q.; Zhang, W. B.; Sheng, G. Y.; Fu, J. M. Determination of carbonyl compounds in the atmosphere by DNPH derivatization and LC−ESI−MS/MS detection. Talanta 2007, 72 (2), 539−545 DOI: 10.1016/j.talanta.2006.11.018. (22) Vairavamurthy, A.; Roberts, J. M.; Newman, L. Methods for the deteremination of low-molecular weight carbonyl compounds in the atmosphere. Atmos. Environ., Part A 1992, 26 (11), 1965−1993 DOI: 10.1016/0960-1686(92)90083-w. (23) Vogel, M.; Buldt, A.; Karst, U. Hydrazine reagents as derivatizing agents in environmental analysisA critical review. Fresenius’ J. Anal. Chem. 2000, 366 (8), 781−791 DOI: 10.1007/ s002160051572. (24) Vereecken, L.; Müller, J. F.; Peeters, J. Low-volatility polyoxygenates in the OH-initiated atmospheric oxidation of α-pinene: Impact of non-traditional peroxyl radical chemistry. Phys. Chem. Chem. Phys. 2007, 9 (38), 5241−5248 DOI: 10.1039/b708023a. (25) Hamilton, E. J.; Korcek, S.; Mahoney, L. R.; Zinbo, M. Kinetics and mechanism of the autoxidation of pentaerythrityl tetraheptanoate at 180−220 °C. Int. J. Chem. Kinet. 1980, 12 (9), 577−603 DOI: 10.1002/kin.550120902. (26) Jensen, R. K.; Korcek, S.; Mahoney, L. R.; Zinbo, M. Liquidphase autoxidation of organic compounds at elevated temperatures. 1. The stirred flow reactor technique and analysis of primary products from n-hexadecane autoxidation at 120−180 °C. J. Am. Chem. Soc. 1979, 101 (25), 7574−7584 DOI: 10.1021/ja00519a018. (27) Jensen, R. K.; Korcek, S.; Mahoney, L. R.; Zinbo, M. Liquidphase autoxidation of organic compounds at elevated temperatures. 2. Kinetics and mechanisms of the formation of cleavage products in nhexadecane autoxidation. J. Am. Chem. Soc. 1981, 103 (7), 1742−1749 DOI: 10.1021/ja00397a026. (28) Jalan, A.; Alecu, I. M.; Meana-Paneda, R.; Aguilera-Iparraguirre, J.; Yang, K. R.; Merchant, S. S.; Truhlar, D. G.; Green, W. H. New pathways for formation of acids and carbonyl products in lowtemperature oxidation: The Korcek decomposition of γ-ketohydroperoxides. J. Am. Chem. Soc. 2013, 135 (30), 11100−11114 DOI: 10.1021/ja4034439. (29) Yasmeen, F.; Vermeylen, R.; Szmigielski, R.; Iinuma, Y.; Böge, O.; Herrmann, H.; Maenhaut, W.; Claeys, M. Terpenylic acid and related compounds: Precursors for dimers in secondary organic aerosol from the ozonolysis of α- and β-pinene. Atmos. Chem. Phys. 2010, 10 (19), 9383−9392 DOI: 10.5194/acp-10-9383-2010. (30) Claeys, M.; Iinuma, Y.; Szmigielski, R.; Surratt, J. D.; Blockhuys, F.; Van Alsenoy, C.; Böge, O.; Sierau, B.; Gomez-Gonzalez, Y.; Vermeylen, R.; Van der Veken, P.; Shahgholi, M.; Chan, A. W. H.; Herrmann, H.; Seinfeld, J. H.; Maenhaut, W. Terpenylic acid and related compounds from the oxidation of α-pinene: Implications for new particle formation and growth above forests. Environ. Sci. Technol. 2009, 43 (18), 6976−6982 DOI: 10.1021/es9007596. (31) Szmigielski, R.; Surratt, J. D.; Gomez-Gonzalez, Y.; Van der Veken, P.; Kourtchev, I.; Vermeylen, R.; Blockhuys, F.; Jaoui, M.; Kleindienst, T. E.; Lewandowski, M.; Offenberg, J. H.; Edney, E. O.; Seinfeld, J. H.; Maenhaut, W.; Claeys, M. 3-Methyl-1,2,3-butanetricarboxylic acid: An atmospheric tracer for terpene secondary organic aerosol. Geophys. Res. Lett. 2007, 34 (24), L24811 DOI: 10.1029/ 2007gl031338. (32) Larsen, B. R.; Di Bella, D.; Glasius, M.; Winterhalter, R.; Jensen, N.; Hjorth, J. Gas-phase OH oxidation of monoterpenes: Gaseous and H


Highly Oxidized Multifunctional Organic Compounds Observed in

Recommend Documents