Highly Oxidized Second-Generation Products from the Gas-Phase

Article pubs.acs.org/JPCA

Highly Oxidized Second-Generation Products from the Gas-Phase Reaction of OH Radicals with Isoprene Torsten Berndt,*,† Hartmut Herrmann,† Mikko Sipila,̈ ‡ and Markku Kulmala‡ †

Leibniz Institute for Tropospheric Research, TROPOS, 04318 Leipzig, Germany Department of Physics, University of Helsinki, P.O. Box 64, Helsinki 00014, Finland

‡

S Supporting Information *

ABSTRACT: The gas-phase reaction of OH radicals with isoprene has been investigated in an atmospheric pressure flow tube at 293 ± 0.5 K with special attention to the second-generation products. Reaction conditions were optimized to achieve a predominant reaction of RO2 radicals with HO2 radicals. Chemical ionization−atmospheric pressure interface−time-of-flight mass spectrometry served as the analytical technique in order to follow the formation of RO2 radicals and closed-shell products containing at least four O atoms in the molecule. The reaction products were detected as adducts with the reagent ions using acetate, lactate, or nitrate in the ionization process. Observed signals were attributed to a series of C5-products with multiple hydroxy, hydroperoxy, and probably carbonyl groups. H/D exchange experiments supported the product identification. The generation of the detected second-generation products can be mechanistically explained starting from the OH radical reaction of hydroxy hydroperoxide isomers, HO−C5H8−OOH. These isomers represent the dominant products of the initial OH radical attack on isoprene. Dihydroxy dihydroperoxides, (HO)2−C5H8−(OOH)2, were analyzed as the main second-generation products beside the dihydroxy epoxides. A simple kinetic analysis revealed that the observed second-generation products in total (other than dihydroxy epoxides) were formed with an estimated molar yield of 10.0+2.1 −1.5 % with respect to converted hydroxy hydroperoxides. A formation yield of 5.8+1.3 % has been deduced for the main product (HO)2−C5H8−(OOH)2. The detected, highly oxidized −0.9 isoprene products represent potential secondary organic aerosol precursors. An annual, global (HO)2−C5H8−(OOH)2 formation strength of (16−35) × 106 metric tons is estimated based on product measurements of this study and literature data regarding the formation of the dihydroxy epoxide isomers for an annual isoprene emission of 454 × 106 metric tons of carbon. OH + C5H8 ( +O2 ) → HOC5H8O2

1. INTRODUCTION Isoprene (C5H8) represents the most important nonmethane emission to the Earth’s atmosphere with a global rate of about 500 million metric tons per year.1−3 The role of isoprene’s oxidation products for the process of secondary organic aerosol (SOA) formation is still subject of investigations, despite of the progress in mechanistic understanding of the atmospheric oxidation pathways in the last years.4,5 Formerly, it was assumed that isoprene does not contribute significantly to the SOA formation.6 The experimental evidence of tetrols bearing an isoprene carbon-skeleton, however, pointed to the importance of isoprene for the atmospheric SOA burden.7 Presently, second-generation products from the OH radical initiated isoprene oxidation under low-NO conditions, such as dihydroxy epoxides (often called IEPOX)5,8−10 or other low volatility products, 11 are considered as potential SOA precursors. In the atmosphere, isoprene reacts readily with the OH radical forming an adduct that rapidly adds O2 resulting in a series of RO2 isomers:12 © XXXX American Chemical Society

(1)

The RO2 radicals react in bimolecular steps with NO, HO2, or with other RO2 radicals depending on the availability of the second reactant. Moreover, intramolecular H atom transfer reactions are possibly leading to the formation of hydroperoxy aldehydes (HPALD) and other products.4 The latter reaction pathway does not dominate the RO2 fate under atmospheric conditions.13,14 In remote areas with very low NO concentrations,15 the preferred reaction represents the reaction with HO2 radicals forming the corresponding hydroxy hydroperoxide isomers (ISOPOOH): HOC5H8O2 + HO2 → HOC5H8OOH + O2 (2)

1,2-ISOPOOH, 2-hydroperoxy-2-methylbut-3-en-1-ol, and 4,3ISOPOOH, 2-hydroperoxy-3-methylbut-3-en-1-ol, are the two Received: November 1, 2016 Revised: November 24, 2016

A

DOI: 10.1021/acs.jpca.6b10987 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

± 0.5 K and a pressure of 1 bar of purified air. OH radicals were generated via UV photolysis of ozone in the presence of water vapor. The flow tube (i.d. 8 cm; length 425 cm) consists of a first section (56 cm) containing the gas inlet system, a second section (344 cm) surrounded by eight UV lamps (Hg-lamps made of PN235 quartz-glass with a cutoff wavelength of 210 nm), and a nonirradiated end section (25 cm) incorporating the sampling devices. Ozone was supplied from an ozone generator UVP OG-2, and it was injected through a nozzle to the gas mixture of isoprene/hydrogen in humidified air just before entering the irradiated middle section. The humidified air was supplied by flushing a part of the air stream through three water saturators filled with water taken from an ultrapure water system (Barnstead, resistivity: 17.4 MΩ cm). The relative humidity of the mixed reaction gas was set at 25%, which was continuously controlled at the outlet of the TROPOS flow tube by a humidity sensor (Vaisala). Ozone was measured by means of a gas monitor (Thermo Environmental Instruments 49C). The isoprene concentration was continuously monitored by a proton transfer reaction−mass spectrometer (Ionicon, high sensitivity PTR−MS).19 The total gas flow rate in the experiments was 30 L min−1 (STP) resulting in a bulk residence time of 32 s in the irradiated middle section. Gas flows were set by calibrated mass flow controllers (MKS 1259/ 1179). Isoprene (>99%, Aldrich), H2 (99.9999%, Air Liquide), D2O (99.9 D atom%, Aldrich), acetic acid (99.99%, Aldrich), and lactic acid (∼90%, Merck) were used without further purification. The air was taken from a pressure swing adsorption unit with further purification by activated 4 Å molecular sieve, charcoal, and subsequently by a GateKeeper CE-500 KF-O-4R unit (AERONEX). The initial isoprene concentration was 1.0 × 1011 molecules cm−3, the concentration of hydrogen, if used, was 4.7 or 7.8 × 1015 molecules cm−3, and the initial ozone concentrations ranged from 1.5 × 1011 to 5.9 × 1012 molecules cm−3. Converted isoprene was determined based on the PTR−MS measurements. A comparison of the calculated isoprene conversion via ozonolysis20 with the measured total amount of converted isoprene from the photolysis experiments revealed that more than 99.5% of the reacted isoprene was due to the photolytic OH radical reaction. Under dark conditions, no significant isoprene conversion from O3 + isoprene was detectable. The degradation of formed hydroperoxides via photolysis in the flow system was estimated to account for less than 1% applying the experimentally observed H2O2 photolysis frequency.14 2.1. Product Detection. Detection of the highly oxidized products (also termed highly oxidized multifunctional organic molecules; HOMs) was conducted by means of a CI-APi-TOF (chemical ionization−atmospheric pressure interface−time-offlight) mass spectrometer (Airmodus, Tofwerk, resolving power > 3000 Th/Th) using acetate,21−26 lactate,26 or nitrate27−29 as the reagent ion. The gas sample was taken from the center flow of the TROPOS flow tube through a sampling inlet (i.d. 1.6 cm; length 28 cm) including a dilution unit. The total flow rate through the inlet was 10 L min−1 (STP) and a dilution factor of 7 (air as dilution gas) was applied in all experiments. A flow of 1−2 mL min−1 air (ionization by acetate and nitrate) or 20 mL min−1 air (ionization by lactate) over a concentrated acid sample (HX: acetic, lactic, or nitric acid) was added to a flow of 35 L min−1 (STP) of purified air. This setup produced the acid containing sheath gas that forms the reagent ions, X−, (HX)X−, and (HX)2X−, after ionization with a 241Am

atmospherically important hydroxy hydroperoxide isomers.11,16 Applying mass spectrometric detection techniques, the occurrence of these hydroxy hydroperoxides was experimentally confirmed from laboratory experiments5 and tentatively from field measurements as well.17 The OH radical can attack the remaining double bond of the hydroxy hydroperoxides in a subsequent reaction step. The resulting adduct undergoes an intramolecular rearrangement to form a dihydroxy epoxide, HO−C5H8(O)−OH, connected with OH radical recycling, pathway 3a.5 In a parallel way (probably depending on the energy distribution of the OH-adduct), O2 addition can take place resulting in the corresponding RO2 radicals, pathway 3b: OH + HOC5H8OOH → HOC5H8(O)OH + OH

(3a)

OH + HOC5H8OOH ( +O2 ) → (HO)2 C5H8(OOH)O2 OH + HOC5H8OOH → H 2O + products

(3b) (3c)

Pathway 3a accounts for 70−80%5,16 and pathway 3b for ∼13%16 of the OH radical initiated conversion of the HO− C5 H8 −OOH isomers. Additionally, significant H atom abstraction by OH radicals was experimentally observed, helping to fulfill the reaction balance, pathway 3c.16 Total OH rate coefficients k3 of (7.5 ± 1.2) × 10−11 and (1.18 ± 0.19) × 10−10 cm3 molecule−1 s−1 have been recently measured at 297 K for 1,2- and 4,3-ISOPOOH, respectively.16 As a result of a former study, average Arrhenius parameters for all isomers have been determined for pathways 3a and 3b.5 According to that, k3a = 7.5 × 10−12 and k3b = 7.2 × 10−11 cm3 molecule−1 s−1 follows for a temperature of 293 K as used in this study. The resulting OH rate coefficient k3 = 7.9 × 10−11 cm3 molecule−1 s−1 was applied for modeling of the reaction system within this study, as discussed later. The knowledge regarding the closed-shell product formation of the RO2 isomers, (HO)2−C5H8(OOH)−O2, under HO2 dominated conditions is very sparse at the moment. No products were detected in the previously mentioned chamber study,16 probably caused by effective wall losses. However, as a result of a second study, performed under similar reaction conditions, but equipped with a different detection technique,11 a series of low volatile products were identified. These products arose likely from reactions of the (HO)2−C5H8(OOH)−O2 radicals. In this work, the formation of second-generation products from the reaction of OH radicals with isoprene, other than dihydroxy epoxides, is investigated for low-NO conditions. In order to ensure a predominate reaction of RO2 radicals with HO2 radicals, the HO2 radical concentration was enhanced in the reaction gas by adding hydrogen. The product formation is followed by means of chemical ionization−atmospheric pressure interface−time-of-flight mass spectrometry (CI-APiTOF) using acetate, lactate, or nitrate as the reagent ion. The total yield of second-generation products, other than dihydroxy epoxides, is calculated based on estimated lower end concentrations of the individual products.

2. EXPERIMENTAL SECTION The gas-phase reaction of OH radicals with isoprene has been conducted in the TROPOS flow tube18 at a temperature of 293 B


Article

The Journal of Physical Chemistry A source. Figure S1a−c in Supporting Information shows the reagent ion spectra as recorded for commonly used reaction conditions. In the CI-inlet (Airmodus), the ions from the sheath gas flow are electrostatically guided into the diluted sample flow without mixing of both gas streams.30 In this process, the sheath flow surrounds the diluted sample flow that streams in the middle of the CI-inlet. This approach minimizes sample losses to the walls. The CI-inlet design using coaxial flows was developed by Eisele and Tanner.27 More detailed information on the fundamentals of the measurement technique is given by Jokinen et al.28 and Junninen et al.31 Concentrations of highly oxidized multifunctional organic compounds (HOMs) were determined for all ionization schemes according to eq I: [HOM] = fHOM

[(HOM)X−] [X−] + [(HX)X−] + [(HX)2 X −]

(I) Figure 1. Mass spectrum from the reaction of OH radicals with isoprene using acetate for ionization, reacted isoprene: 7.3 × 1010 molecules cm−3. The products appear as adducts with acetate. The red spectrum shows the background recorded from ozone photolysis in the absence of isoprene. For more clarity, the background spectrum was shifted horizontally by 2 s−1. Initial isoprene concentration was 1.0 × 1011 molecules cm−3, the relative humidity 25%, and the reaction time 32 s.

The values in the brackets are the measured ion signals. The lack of needed reference substances does not allow an absolute signal calibration. A lower limit value of f HOM can be calculated considering (HOM)X− adduct formation in the CI-inlet via reaction 4: HOM + (HX)n X− → (HOM)X− + (HX)n , n = 0, 1, 2

(4) −9

The rate coefficient k = k4 can be set to (2−3) × 10 cm3 molecule−1 s−1, typical for ion−molecule reactions being close to the collision limit.32−34 Diffusion controlled wall loss in the inlet tube (assuming a diffusion coefficient D = 0.08 cm2 s−1) leads to a HOM loss of 12% and accordingly to an inlet factor f inlet = 0.88. With a reaction time of the ion−molecule reaction t = 0.2−0.3 s, a lower end value f HOM = f HOM, calc = (1.3−2.8) × 109 molecules cm−3 follows according to eq II.35,36

fHOM,calc =

dihydroxy epoxides formed via pathways 2 and 3a, respectively. No attempt was undertaken to distinguish between these products being present with different isomers each. Three groups of formed substances retaining the C5-skeleton were detected, which contain at least four O atoms, i.e., C5HxOy with x = 10, 11, and 12 and with y = 4, 5, and 6. Products with an odd number of H atoms (C5H11O4, C5H11O5, and C5H11O6) were assigned to RO2 radicals as confirmed by the formation of the corresponding organic nitrates in the presence of NO according to RO2 + NO → RONO2, see Figure S4. Even a relatively small NO addition of 5 × 109 molecules cm−3 yielded strong additional signals in the mass spectrum, which were attributed to the corresponding organic nitrates of the different RO2 radicals, i.e., C5H11O5N, C5H11O6N, and C5H11O7N. The other products with 10 or 12 H atoms, x = 10 or 12 in C5HxOy, represent closed-shell products. Estimated concentrations of closed-shell products and RO2 radicals as a function of converted isoprene are depicted in Figure S5. It is supposed that the closed-shell product formation proceeds either via bimolecular reactions of RO2 radials with HO2 or other RO2 radicals or via unimolecular steps. In the absence of NO additions, a significant contribution of the RO2 + NO reaction to RO2 removal can be ruled out due to the absence of measurable RONO2 signals. That justifies the conclusion that the background NO concentration in the reaction gas was clearly smaller than 109 molecules cm−3. For comparison, Krechmer et al.11 observed products with a molecular formula C5HxOy with x = 10 and 12 and with y = 5 and 6 as a result of their study on the reaction of OH radicals with 2-hydroperoxy-3-methylbut-3-en-1-ol (4,3-ISOPOOH). These results are in good agreement with our study. 4,3ISOPOOH is one of the two important hydroxy hydroperoxide isomers representing the first-generation products of the OH + isoprene reaction for low-NO conditions, see pathway 2.11,16 Krechmer et al.11 used CI-APi-TOF mass spectrometry with nitrate as the reagent ion for product identification. Detection of the corresponding RO2 radicals was not reported.

1 finlet kt

(II)

The calculated value of f HOM is in very good agreement with the experimentally obtained H2SO4 calibration factor f H2SO4,exp = 1.85 × 109 molecules cm−3 in this system37 based on the ion−molecule reaction H2SO4 + (HNO3)nNO3−, n = 0,1,2.38,39 By practical reasons and due to the similarity of the ion− molecule reactions, f HOM in eq I was set equal to f H2SO4,exp in the calculations. The total uncertainty of the lower end HOM concentrations determined via eq I is estimated with a factor of 2 including changing ion transmission in the considered mass range.31 The mass spectrometer was running in the “weak field” mode for the three reagent ions applied, i.e., the APi voltages were set in such a way that collisional dissociation of neutralion adducts was minimized.21,23,40

3. RESULTS AND DISCUSSION 3.1. Product Spectrum and Assignment. Figure 1 shows an example of a mass spectrum from the reaction of OH radicals with isoprene applying acetate as the reagent ion. The background spectrum, recorded under photolysis conditions in the absence of isoprene (in red), contains almost exclusively the signal of the acetate adduct with two acetic acid molecules, (CH3COOH)2CH3COO−, see also Figure S1. In the product spectrum (in black), the strongest signal is attributed to substances with the molecular formula C5H10O3, standing most likely for the hydroxy hydroperoxides and the isobaric C


Article

The Journal of Physical Chemistry A 3.1.1. H/D Exchange Experiments. Rapid H/D exchange reactions in the gas phase allow the determination of the number of acidic H atoms in the molecules, i.e., the H atoms bound to electronegative elements such as O atoms.41−43 Thus, it is excepted that the H atoms from −OH and −OOH groups of the isoprene products will be exchanged by D atoms in the presence of heavy water (D2O). The resulting signal shift in the mass spectrum indicates the total number of these groups in the molecule. In the experiments, the dilution air needed for the sampling inlet was passed through a D2O saturator generating a D2O vapor enriched air flow. This dilution gas was mixed in the sampling inlet with the sample flow from the flow tube, which contained the “normal” water vapor (H2O) used for OH radical formation via O(1D) + H2O in the flow tube. In order to maximize the D2O content in the inlet after mixing of sample and dilution flow, the relative humidity (H2O) in the flow tube was reduced to ∼8%. For these conditions and using acetate as the reagent ion, the signal ratio of the reagent ion dimers in the mass spectrum (CH 3 COOD)CH 3 COO − /(CH 3 COOH)CH3COO− accounted for 87/13. This value served as a measure of the H/D exchange efficiency of the D2O/H2O containing air. In Figure 2, sample mass spectra recorded in the absence and presence of D2O are depicted in the mass-to-charge range of

reagent ion dimer, the mass spectrum recorded in the presence of D2O can be reasonably reproduced starting from the mass spectrum in the absence of D2O. For instance, a product distribution C5H8D4O6/C5H9D3O6/C5H10D2O6/C5H11D1O6/ C5H12O6 = 1/0.82/0.25/0.034/0.0017 follows for the four exchangeable H atoms of the compound C5H12O6. That means that the signal ratio of the acetate adducts at nominal 231 and 230 Th should be 1/0.82, which is in good agreement with the measurement, see the upper spectrum in Figure 2. The other part of the H/D exchange spectrum between 226 and 229 Th is characterized by overlapped signals from different compounds. Nevertheless, the signal at nominal 229 Th can be mainly assigned to the acetate adduct of C5H8D3O6 (from C5H11O6) and that at 228 Th to the acetate adduct of C5H7D3O6 (from C5H10O6) justifying the deduced number of −OH and −OOH groups in these compounds. The mass spectrum of the products with five O atoms in the range up to 215 Th can be explained in a similar way. Results of the H/D exchange experiments, i.e., the deduced number of −OH and −OOH groups in the respective products, were used to develop a proposed reaction scheme for the second-generation products, see section 3.4. 3.2. Product Formation for Conditions of a Preferred RO2 + HO2 Reaction. Investigations with a preferred RO2 + HO2 reaction were conducted in order to determine the product formation for conditions close to the atmospheric lowNO scenario where the RO2 fate is dominated by the HO2 radical reaction.5,15 Addition of hydrogen to the feed gas led to an increase of the HO2 radical concentration in the reaction system according to the HO2 formation pathway 5: OH + H 2( +O2 ) → H 2O + HO2

(5)

The effect of hydrogen addition is demonstrated in Figure 3 showing exemplarily the concentrations of C5H10O3 (shared signal of hydroxy hydroperoxides and the isobaric dihydroxy epoxides) and C5H12O6 (dihydroxy dihydroperoxides as explained later) as a function of converted isoprene for different hydrogen additions. Both plotted product concen-

Figure 2. Mass spectra of the reaction products with five and six O atoms recorded in the absence (lower spectrum in black) and presence of heavy water (upper spectrum in blue) using acetate in the ionization process. The red spectra show the background recorded each from ozone photolysis in the absence of isoprene. Initial reactant concentrations: [O3] = 5.9 × 1012 and [isoprene] = 1 × 1011 molecules cm−3, relative humidity ∼8%, and reaction time 32 s.

the product groups with five and six O atoms. Measurements of the product signals with four O atoms were interfered by a relatively high background level in the presence of D2O preventing a reliable data analysis. The comparison of the spectra with five and six O atoms indicates that the signals of products with 12 H atoms (C5H12O6 and C5H12O5) shifted by four mass units in the presence of D2O. This finding suggests the presence of four −OH and −OOH groups in these molecules. For the other product signals (C5H10O5, C5H11O5, C5H10O6, and C5H11O6), a shift by three mass units can be inferred pointing to three −OH and −OOH groups. Assuming a H/D exchange efficiency of 0.87, as determined for the

Figure 3. Estimated concentrations of C5H10O3 (hydroxy hydroperoxides and the isobaric dihydroxy epoxides, in black) and C5H12O6 (dihydroxy dihydroperoxides, in red) as a function of the converted isoprene for different hydrogen additions. The dilution factor of 7 is considered in the plotted concentrations. Acetate served as the reagent ion. Initial reactant concentrations: [O3] = (1.5−54) × 1011 and [isoprene] = 1 × 1011 molecules cm−3, relative humidity 25%, and reaction time 32 s. D


Article


× 1015 molecules cm−3, the closed-shell products in part A and the RO2 radicals in part B. A corresponding mass spectrum, recorded for an isoprene conversion of 4.75 × 1010 molecules cm−3, is given in Figure S6 in Supporting Information. The main closed-shell product under conditions of the preferred RO2 + HO2 reaction is the compound with the chemical formula C5H12O6. The estimated C5H12O6 concentration accounts for more than 50% of the total observed product concentration. It is also noteworthy, that the relative RO2 radical abundance is changed comparing the measurements without and with hydrogen addition, i.e., for relatively low and high HO2 radical concentrations. For reacted isoprene of ∼4.8 × 1010 molecules cm−3, the RO2 concentration ratio [C5H11O6]/[C5H11O5] increases from 0.6 to 2.7, see part B each in Figures S5 and 4. This finding suggests that a RO2 + R′O2 reaction is likely responsible for the C5H11O5 formation, which is repressed in the presence of elevated HO2 radical concentrations. 3.3. Variation of the Reagent Ion. In order to test the reagent ion specific sensitivity of the product detection, the measurement series performed under conditions of the preferred RO2 + HO2 reaction was repeated using lactate or nitrate as the reagent ion instead of acetate. In the case of lactate ionization, the same products were detectable with an almost identical sensitivity as described before using acetate in the ionization process. Exceptions are the products containing four O atoms. Their signals were smaller by a factor of 2−3. The RO2 radical signal C5H11O6 was disturbed by any interfering signal that hampered the data analysis. The estimated closed-shell products and RO2 radical concentrations as a function of reacted isoprene are depicted in Figures S7 and S8, respectively. In the case of nitrate ionization, only four closed-shell products were detected and all with a definitely lower sensitivity compared with the results using acetate ionization. The estimated C5H12O6 concentration was lower by a factor of ∼3 and those of C5H10O6, C5H10O5, and C5H12O5 by a factor of 10−20. Significant RO2 radical signals were only detectable for C5H11O6 with estimated concentrations not exceeding 2 × 106 molecules cm−3. Figure S9 shows the obtained data for the closed-shell products as a function of reacted isoprene. A comparison of C5H12O6 concentrations from the three ionization schemes is depicted in Figure 5. Acetate and lactate ionization yielded almost identical results for this main product, while the data obtained by nitrate ionization were smaller by a factor of ∼3. Sensitivity differences can be due to different adduct stabilities for the different reagent ions. A significantly higher detection sensitivity for products from cyclohexene ozonolysis with four or five O atoms24 was already observed using acetate instead of nitrate in the ionization process. Furthermore, acetate ionization was much more sensitive than nitrate ionization in the case of highly oxidized RO2 radicals from the OH radical reaction of α- and β-pinene.26 Theoretical calculations on the stability of adducts of model RO2 radicals with these reagent ions supported the experimental findings.26 Moreover, for the ozone-initiated oxidation of β-caryophyllene, the signal height of RO2 radicals containing only one hydroperoxy group increased by an order of magnitude when changing from nitrate to acetate ionization.46 Taking all these facts together, acetate ionization seems to be a good choice at the moment for an efficient detection of highly oxidized products, especially for those with a relatively low oxygen content. The difference of the detection sensitivity for OH

trations exhibited a strong rise with increasing HO 2 concentration caused by the hydrogen addition. The other products were also sensitive to the HO2 concentration change. Measured product distributions became almost independent of the hydrogen concentration if it exceeded ∼5 × 1015 molecules cm−3. Thus, it can be concluded that at the highest hydrogen addition of 7.8 × 1015 molecules cm−3 used in the experiments, the RO2 radical fate was dominated by its reaction with HO2 radicals. A reliable estimate of the concentrations of HO2 radicals and the first-generation RO2 radicals, formed via pathway 1, is difficult. It can be merely stated that the ratio of the formation rate of HO2 radicals and the first-generation RO2 radicals, k5 × [H2]/(k1 × [C5H8]), is larger than five for a hydrogen concentration of 7.8 × 1015 molecules cm−3, k1 = 1.0 × 10−1044 and k5 = 6.7 × 10−15 cm3 molecule−1 s−1.45 The firstgeneration RO2 radicals dominate the total RO2 radical budget in this system because of the relatively small importance of pathway 3b forming the second-generation RO2 radicals and the limited isoprene conversion. Therefore, in the case of the high hydrogen additions, this estimation suggests also that the fate of all RO2 radicals is governed by the reaction with HO2 radicals rather than by RO2 self-reactions. Figure 4 shows the estimated concentrations from the measurement series with the highest hydrogen addition of 7.8

Figure 4. Estimated product concentrations detected in the presence of 7.8 × 1015 molecules cm−3 hydrogen as a function of the converted isoprene, closed-shell products in part A and RO2 radicals in part B. The measurements have been performed using acetate as the reagent ion. The dilution factor of 7 is considered in the plotted concentrations. Initial reactant concentrations: [O3] = (5.6−53) × 1011 and [isoprene] = 1 × 1011 molecules cm−3, relative humidity 25%, and reaction time 32 s. E


Article


radical (3) or add O2 forming the RO2 radical (4). The isomerization step (2) → (3) was adopted from a proposed reaction mechanisms given by St. Clair et al.16 No information on the number of acidic H atoms in (3) was achieved from the H/D exchange experiments caused by experimental difficulties. For the RO2 radical (4), C5H11O6, three acidic H atoms were detected in line with the proposed structure bearing one hydroperoxy and two hydroxy groups. The formation of the RO2 radical (6), C5H11O5, can be explained by a reduction step (4) → (5) followed by the isomerization (5) → (6).47 The presence of three hydroxy groups in (6) is consistent with the result of the H/D exchange experiment, which indicated three acidic H atoms for C5H11O5. The reduction proceeds likely via RO2 + R′O2 → RO + R′O + O2.48 Support for this assumption comes from the observed increase of the RO2 radical ratio [C5H11O6]/[C5H11O5] for elevated HO2 radical concentrations where the importance of RO2 + R′O2 reactions is repressed. Nevertheless, also a contribution of a pathway according to RO2 + HO2 → RO + OH + O2 for the formation of (5) cannot be ruled out.49,50 It is to be noted that Scheme 1 represents an example of a proposed reaction scheme for the formation of the RO2 radicals. The external OH attack as well as all OH radical reaction of the other hydroxy hydroperoxide isomers can lead to the same RO2 radicals or isomers of those. 3.4.2. Closed-Shell Products. Scheme 2 shows possible pathways for the formation of the closed-shell products (7), C5H12O6, and (9), C5H10O6, starting from the RO2 radical (4), C5H11O6. The formation of the main product C5H12O6 for HO2 radical dominated conditions can be explained by a RO2 + HO2 reaction forming the corresponding hydroperoxide (7). Four acidic H atoms were analyzed for C5H12O6 in the H/D exchange experiment being in line with the expected two hydroperoxy and two hydroxy groups in this molecule. Thus, it is very likely that C5H12O6 is a dihydroxy dihydroperoxide according to (7) or isomers of that. The minor closed-shell product with six O atoms, C5H10O6, could be produced via isomerization of (4) and subsequent carbonyl formation from the alkoxy radical (8) via H atom abstraction by O2, (8) → (9).

Figure 5. Estimated C5H12O6 concentrations as a function of the converted isoprene using different reagent ions, i.e., acetate (in black), lactate (in red), or nitrate (in blue). The dilution factor of 7 is considered in the plotted concentrations. Initial reactant concentrations: [O3] = (5.6−53) × 1011, [isoprene] = 1 × 1011, and [H2] = 7.8 × 1015 molecules cm−3, relative humidity 25%, and reaction time 32 s.

radical derived RO2 radicals from α- and β-pinene was less pronounced comparing acetate with lactate ionization.26 Also in the present study, almost identical results have been received for products with multiple hydroxy and hydroperoxy groups using acetate and lactate ionization, as shown here for C5H12O6 in Figure 5. 3.4. Proposed Reaction Pathways. 3.4.1. RO2 Radicals. The detected signals assigned to C5H11O4, C5H11O5, and C5H11O6 were attributed to RO2 radicals. Experiments with NO additions, and subsequent organic nitrate formation according to RO2 + NO → RONO2, supported the assignment, see Figure S4. In Scheme 1, starting from the most abundant hydroxy hydroperoxide isomer, 2-hydroperoxy-2-methylbut-3en-1-ol (1), the internal OH radical attack16 yields the alkyl radical (2). Species (2) can either isomerize forming the RO2

Scheme 1. Proposed Reaction Scheme for the Formation of Detected RO2 Radicals Starting from the Internal OH Radical Reaction of 2-Hydroperoxy-2-methylbut-3-en-1-ol

F


Article


Scheme 2. Proposed Pathways for the Formation of Closed-Shell Products Originated from RO2 Radical (4), C5H11O6

kM4 = 0.018 s−1 estimated from kwall = 3.65 × D/r2 with a diffusion coefficient D = 0.08 cm2 s−1. The system of differential equations resulting from M1−M4 was integrated numerically in the time interval [0; 32 s] for each measurement point by means of a semi-implicit method.51 The integration procedure was connected with a Newtontechnique51 to determine the free parameter “y” in a leastsquares analysis that minimizes the difference between measured and calculated “products” formed via pathway M2. The needed OH radical concentration was calculated using the measured isoprene concentrations with and without isoprene conversion, termed [C5H8] and [C5H8]0, respectively, [OH] = ln([C5H8]0/[C5H8])/(k1t). As a result of this analysis, a molar formation yield of 10.0 ± 0.4% was determined for the total product concentration measured by means of acetate ionization, see the red measurement points and the corresponding modeling curve in Figure 6. Summation has been done for all closed-shell products and RO2 radicals. Considering that the RO2 radicals C5H11O6 will be converted to the dihydroxy dihydroperoxide, C5H12O6, in the further course of the reaction, the measured concentrations of C5H11O6 were added to C5H12O6 leading to the “final” C5H12O6 concentrations. The analysis resulted in a molar formation yield of 5.8 ± 0.2% for the dihydroxy

The presence of one hydroxy and two hydroperoxy groups in C5H10O6 is consistent with the detected three acidic H atoms from the H/D exchange experiment. Similar pathways can be proposed for the formation of C 5 H 12 O 5 and C 5 H 10 O 5 originating from the RO2 radical (6), C5H11O5, and for the formation of C5H12O4 and C5H10O4 from the RO2 radical (3), C5H11O4. According to that, it can be expected that the product C5H12O5 stands for a trihydroxy hydroperoxide and C5H10O5 for a product with two hydroxy, one hydroperoxy, and one carbonyl group. This assignment is in line with the obtained number of four acidic H atoms in C5H12O5 and three acidic H atoms in C5H10O5. 3.5. Modeling of the Reaction System and Formation Yields. A direct determination of second-generation product yields is impossible by following these products as a function of the reacted primary reactant isoprene. Thus, a kinetic analysis has been performed for conditions of the preferred RO2 + HO2 reaction applying a simple model. This modeling approach assumes (i) that the HO−C5H8−O2 radical reacts immediately with HO2 radicals via pathway 2 forming the hydroxy hydroperoxides, HO−C5H8−OOH, and (ii) that the OH radical concentration is constant within the course of an individual experiment. The latter assumption seems to be justified taking into account the constant OH production term, based on constant values for the ozone and water vapor concentrations and for the photolysis frequency, and an almost constant OH loss term. The OH loss term, k1 × [C5H8] + k3 × [HO−C5H8−OOH] + k5 × [H2], altered less than 10% in the course of a reaction even for the runs with the highest isoprene conversion of ∼50%. The following reaction steps were considered in the simple model: OH + C5H8(+ O2 ) → 0.88 × HOC5H8OOH + ... (M1)

OH + HOC5H8OOH(+ O2 ) → y × products (M2)

HOC5H8OOH → wall

(M3)

products → wall

(M4) Figure 6. Comparison of experimental results with the outcome of the modeling. The experimental data were taken from the runs in the presence of 7.8 × 1015 molecules cm−3 hydrogen using acetate ionization, see Figure 4. The red circles show the sum of all closedshell products and RO2 radicals. Black circles represent the “final” dihydroxy dihydroperoxide concentrations, C5H12O6, incorporating the measured precursor RO2 radical C5H11O6. The corresponding curves show the modeling results.

The HO−C5H8−OOH formation yield in pathway M1 is set at 0.885 and kM1 = k1 = 1.0 × 10−10 cm3 molecule−1 s−1.44 The average rate coefficient for the OH radical reaction of the hydroxy hydroperoxide isomers at 293 K kM2 = 7.9 × 10−11 cm3 molecule−1 s−1 was taken from the literature.5 Diffusion controlled wall loss in the flow tube was assumed for the reaction products HO−C5H8−OOH and “products” with kM3 = G


Article


of magnitude lower than those for the hydroxy hydroperoxides, HO−C5H8−OOH, 7 × 10−6 atm, and for the dihydroxy epoxides, 3.5 × 10−6 atm. The low vapor pressure along with the expected good water solubility and adhesiveness qualify especially the dihydroxy dihydroperoxides as potential SOA precursors. The other highly oxidized products can act in a similar way. Very recently, from two experimental studies it is conformably reported that highly oxidized reaction products of isoprene other than dihydroxy epoxides substantially contribute to SOA formation.53,54 Riva et al.53 detected as the main SOA components C5H12O6 and C5H12O5 formed from the OH radical reaction of one hydroxy hydroperoxide isomer (1,2ISOPOOH). Liu et al.54 analyzed also C5H12O6 as a major particle-phase product from the OH radical reaction of isoprene as well as from the OH radical reaction of the other important hydroxy hydroperoxide isomer (4,3-ISOPOOH). A SOA mass yield of up to 15% is stated.54 Both studies53,54 conclusively demonstrate the importance especially of C5H12O6 (dihydroxy dihydroperoxides) as a SOA constituent in line with the result of the present gas-phase study identifying the dihydroxy dihydroperoxides as a second-generation main product from OH + isoprene.

dihydroperoxides. In Figure 6, the outcome from the modeling (black curve) is compared with the measurements (black circle). Despite the simplicity of the applied model, both modeling curves are able to describe the experimental findings well. The most critical input parameter for the accuracy of the yield determination in the reaction scheme M1−M4 is the HO−C5H8−OOH formation yield of 0.88 taken from the literature.5 Experimental limitations, i.e., the occurrence of the isobaric dihydroxy epoxide and probably insufficient detection sensitivity, prevented a separate determination of this yield under the conditions of our experiment. In order to account for a possible uncertainty of this value, the data analysis was repeated using presumed HO−C5H8−OOH formation yields of 1.0 or 0.75 in the calculations. Table 1 summarizes the Table 1. Formation Yields of the Total Products and of the Dihydroxy Dihydroperoxides for Different HO−C5H8− OOH Formation Yield Utilized in the Data Analysisa

a

HO−C5H8−OOH yield in pathway M1

total products (%)

dihydroxy dihydroperoxides (%)

0.88 1.0 0.75

10.0 ± 0.4 8.8 ± 0.3 11.7 ± 0.4

5.8 ± 0.2 5.1 ± 0.2 6.8 ± 0.3

4. CONCLUSIONS Second-generation products from the gas-phase reaction of OH radicals with isoprene under conditions of the preferred RO2 + HO2 reaction are, beside the known dihydroxy epoxide,5,16 highly oxidized products with up to six O atoms bearing multiple hydroxy, hydroperoxy, and probably carbonyl groups. Their estimated, molar formation yield is 10.0+2.1 −1.5 % with respect to the reacted hydroxy hydroperoxides, i.e., about 1/7− 1/8 of the dihydroxy epoxide formation yield.5,16 These highly oxidized products can be regarded as potential precursors for secondary organic aerosol formation. Dihydroxy dihydroperoxides, (HO)2−C5H8−(OOH)2, represent the main fraction of the highly oxidized products with an estimated molar yield of 5.8+1.3 −0.9 %. An annual (HO)2−C5H8− (OOH)2 formation strength of (16−35) × 106 metric tons total mass is estimated. The findings emphasize the role of the highly oxidized, second-generation products from OH + isoprene for the process of SOA formation.

Error limits represent 2σ values of the statistical error.

modeling results. An increased HO−C5H8−OOH formation yield leads to a lowering of the calculated formation yield of second-generation products, and vice versa, for a given set of experimental data. The error limits were enlarged based on these findings resulting in a molar formation yield of 10.0+2.1 −1.5 % +1.3 for the total products and 5.8−0.9 % for the dihydroxy dihydroperoxides. It is to be noted here that these yields have been determined based on estimated concentrations, which represent lower end limits. Hence, resulting yields can be only treated as lower end limits as well. Nevertheless, the highly oxidized reaction products from the OH radical reaction of hydroxy hydroperoxides, detected in this study, represent a substantial product fraction beside the dihydroxy epoxides reported with a formation yield of 70−80%.5,16 3.6. Implication for the Atmosphere. An annual, global dihydroxy epoxide formation of (94−122) × 106 metric tons of carbon is stated in the literature resulting from global simulation runs with an annual isoprene emission of 454 × 106 metric tons of carbon.5,16 Taking into account a dihydroxy epoxide formation yield of 70−80% from OH + HO−C5H8− OOH,5,16 the total products obtained in this study with a yield 6 of 10.0+2.1 −1.5 % account for (10−21) × 10 metric tons of carbon 16 per year. St. Clair et al. reported a similar value as a result of their product studies for OH + HO−C5H8−OOH under highNO conditions. For the dihydroxy dihydroperoxides, (HO)2− C5H8−(OOH)2, obtained with a formation yield of 5.8+1.3 −0.9 %, an annual formation of (5.8−12.4) × 106 metric tons of carbon follows. These estimates require that the second-generation product formation proceeds under HO2 dominated conditions for the RO2 radical fate in the same way as needed for the formation of the hydroxy hydroperoxides. Regarding the absolute mass, an annual global (HO)2−C5H8−(OOH)2 formation strength of (16−35) × 106 metric tons can be stated. Applying the increment method SIMPOL.1,52 a vapor pressure of 2 × 10−10 atm was estimated for the dihydroxy dihydroperoxides, (HO)2−C5H8−(OOH)2, more than 4 orders

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.6b10987. Acetate reagent ion spectrum; lactate reagent ion spectrum; nitrate reagent ion spectrum; organic nitrate formation; estimated concentrations of second-generation products; mass spectrum recorded for high hydrogen concentrations; results from lactate ionization; results from nitrate ionization (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Torsten Berndt: 0000-0003-2014-6018 Notes

The authors declare no competing financial interest. H


Article


■

generation isoprene hydroxy hydroperoxide (ISOPOOH) with OH. J. Phys. Chem. A 2016, 120, 1441−1451. (17) Crutzen, P. J.; Williams, J.; Pöschl, U.; Hoor, P.; Fischer, H.; Warneke, C.; Holzinger, R.; Hansel, A.; Lindinger, W.; Scheeren, B.; et al. High spatial and temporal resolution measurements of primary organics and their oxidation products over the tropical forest of Surinum. Atmos. Environ. 2000, 34, 1161−1165. (18) Berndt, T.; Böge, O.; Stratmann, F.; Heintzenberg, J.; Kulmala, M. Rapid formation of sulfuric acid particles at near-atmospheric conditions. Science 2005, 307, 689−700. (19) Lindinger, W.; Hansel, A.; Jordan, A. On-line monitoring of volatile organic compounds at ppt levels by means of Proton-Transfer Reaction Mass Spectrometry (PTR-MS) Medical application, food control and environmental research. Int. J. Mass Spectrom. Ion Processes 1998, 173, 191−241. (20) Grosjean, E.; Grosjean, D. Rate constants for the gas-phase reaction of ozone with 1,1-disubstituted alkenes. Int. J. Chem. Kinet. 1996, 28, 911−918. (21) Bertram, T. H.; Kimmel, J. R.; Crisp, T. A.; Ryder, O. S.; Yatavelli, R. L. N.; Thronton, J. A.; Cubison, M. J.; Gonin, M.; Worsnop, D. R. A Field-deployable, chemical ionization time-of-flight mass spectrometer. Atmos. Meas. Tech. 2011, 4, 1471−1479. (22) Lopez-Hilfiker, F. D.; Mohr, C.; Ehn, M.; Rubach, F.; Kleist, E.; Wildt, J.; Mentel, Th. F.; Carrasquillo, A.; Daumit, K.; Hunter, J.; et al. Phase partitioning and volatility of secondary organic aerosol components formed from α-pinene ozonolysis and OH oxidation: The importance of accretion products and other low volatility compounds. Atmos. Chem. Phys. 2015, 15, 7765−7776. (23) Aljawhary, D.; Lee, A. K. Y.; Abbatt, J. P. D. High-resolution chemical ionization mass spectrometry (ToF-CIMS): application to study SOA composition and processing. Atmos. Meas. Tech. 2013, 6, 3211−3224. (24) Berndt, T.; Richters, S.; Kaethner, R.; Voigtländer, J.; Stratmann, F.; Sipilä, M.; Kulmala, M.; Herrmann, H. Gas-phase ozonolysis of cycloalkenes: Formation of highly oxidized RO2 radicals and their reactions with NO, NO2, SO2, and other RO2 radicals. J. Phys. Chem. A 2015, 119, 10336−10348. (25) Brophy, P.; Farmer, D. K. A switchable reagent ion high resolution time-of-flight chemical ionization mass spectrometer for real-time measurements of gas phase oxidized species: characterization from the 2013 southern oxidant and aerosol study. Atmos. Meas. Tech. 2015, 8, 2945−2959. (26) Berndt, T.; Richters, S.; Jokinen, T.; Hyttinen, N.; Kurtén, T.; Otkjær, R. V.; Kjaergaard, H. G.; Stratmann, F.; Herrmann, H.; Sipilä, M.; et al. Hydroxyl radical induced formation of highly oxidized organic compounds. Nat. Commun. 2016, 7, 13677. (27) Eisele, F. L.; Tanner, D. J. Ion-assisted tropospheric OH measurements. J. Geophys. Res. 1991, 96, 9295−9308. (28) Jokinen, T.; Sipilä, M.; Junninen, H.; Ehn, M.; Lönn, G.; Hakala, J.; Petäjä, T.; Mauldin, R. L., III; Kulmala, M.; Worsnop, D. R. Atmospheric sulphuric acid and neutral cluster measurements using CI-APi-TOF. Atmos. Chem. Phys. 2012, 12, 4117−4125. (29) Ehn, M.; Kleist, E.; Junninen, H.; Petäjä, T.; Lönn, G.; Schobesberger, S.; Dal Maso, M.; Trimborn, A.; Kulmala, M.; Worsnop, D. R.; et al. Gas phase formation of extremely oxidized pinene reaction products in a chamber and ambient air. Atmos. Chem. Phys. 2012, 12, 5113−5127. (30) Sipilä, M.; Sarnela, N.; Jokinen, T.; Junninen, H.; Hakala, J.; Rissanen, M. P.; Praplan, A.; Simon, M.; Kürten, A.; Bianchi, F.; et al. Bisulfate − cluster based atmospheric pressure chemical ionization mass spectrometer for high-sensitivity (< 100 ppqV) detection of atmospheric dimethyl amine: proof-of-concept and first ambient data from boreal forest. Atmos. Meas. Tech. 2015, 8, 4001−4011. (31) Junninen, H.; Ehn, M.; Petäjä, T.; Luosujärvi, L.; Kotiaho, T.; Kostiainen, R.; Rohrer, U.; Gonin, M.; Fuhrer, K.; Kulmala, M.; et al. A high-resolution mass spectrometer to measure atmospheric ion composition. Atmos. Meas. Tech. 2010, 3, 1039−1053. (32) Viggiano, A. A.; Seeley, J. V.; Mundis, P. L.; Williamson, J. S.; Morris, A. M. Rate constants for the reactions of XO3−(H2O)n (X = C,

ACKNOWLEDGMENTS The authors thank K. Pielok, S. Richters, and A. Rohmer for technical assistance.

■

REFERENCES

(1) Guenther, A.; Hewitt, C. N.; Erickson, D.; Fall, R.; Geron, C.; Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.; McKay, W. A.; et al. A global model of natural volatile organic compound emissions. J. Geophys. Res. 1995, 100, 8873−8892. (2) Guenther, A. B.; Jiang, X.; Heald, C. L.; Sakulyanontvittaya, T.; Duhl, T.; Emmons, L. K.; Wang, X. The model of emissions of gases and aerosols from nature version 2.1 (MEGAN2.1): an extended updated framework for modeling biogenic emissions. Geosci. Model Dev. 2012, 5, 1471−1492. (3) Goldstein, A. H.; Galbally, I. E. Known and unexplored organic constituents in the Earthś atmosphere. Environ. Sci. Technol. 2007, 41, 1514−1521. (4) Peeters, J.; Nguyen, T. L.; Vereecken, L. HOx radical regeneration in the oxidation of isoprene. Phys. Chem. Chem. Phys. 2009, 11, 5935− 5939. (5) Paulot, F.; Crounse, J. D.; Kjaergaard, H. G.; Kurten, A.; St. Clair, J. M.; Seinfeld, J. H.; Wennberg, P. O. Unexpected epoxide formation in the gas-phase photoxidation of isoprene. Science 2009, 325, 730− 733. (6) Pandis, S. N.; Paulson, S. E.; Seinfeld, J. H.; Flagan, R. C. Aerosol formation in the photooxidation of isoprene and β-pinene. Atmos. Environ., Part A 1991, 25A, 997−1008. (7) Claeys, M.; Graham, B.; Vas, G.; Wang, W.; Vermeylen, R.; Pashynska, V.; Cafmeyer, J.; Guyon, P.; Andreae, M. O.; Artaxo, P.; et al. Formation of secondary organic aerosols through photooxidation of isoprene. Science 2004, 303, 1173−1176. (8) Lin, Y.-H.; Zhang, H.; Pye, H. O. T.; Zhang, Z.; Marth, W. J.; Park, S.; Arashiro, M.; Cui, T.; Budisulistiorini, S. H.; Sexton, K. G.; et al. Epoxide as a precursor to secondary organic aerosol formation from isoprene photooxidation in the presence of nitrogen oxides. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 6718−6723. (9) Nguyen, T. B.; Coggon, M. M.; Bates, K. H.; Zhang, X.; Schwantes, R. H.; Schilling, K. A.; Loza, C. L.; Flagan, R. C.; Wennberg, P. O.; Seinfeld, J. H. Organic aerosol formation from the reactive uptake of isoprene epoxydiols (IEPOX) onto non-acidified inorganic seeds. Atmos. Chem. Phys. 2014, 14, 3497−3510. (10) Liu, Y.; Kuwata, M.; Strick, B. F.; Geiger, F. M.; Thomson, R. J.; McKinney, K. A.; Martin, S. T. Uptake of epoxydiol isomers accounts for half of the particle-phase material produced from isoprene photooxidation via the HO2 pathway. Environ. Sci. Technol. 2015, 49, 250−258. (11) Krechmer, J. E.; Coggon, M. M.; Massoli, P.; Nguyen, T. B.; Crounse, J. D.; Hu, W.; Day, D. A.; Tyndall, G. S.; Henze, D. K.; Rivera-Rios, J. C.; et al. Formation of low volatility organic compounds and secondary organic aerosol from isoprene hydroxyhydroperoxide low-NO oxidation. Environ. Sci. Technol. 2015, 49, 10330−10339. (12) Jenkin, M. E.; Boyd, A. A.; Lesclaux, R. Peroxy radical kinetics from the OH-initiated oxidation of 1,3-butadiene, 2,3-dimethyl-1,3butadiene and isoprene. J. Atmos. Chem. 1998, 29, 267−298. (13) Crounse, J. D.; Paulot, F.; Kjaergaard, H. G.; Wennberg, P. O. Peroxy radical isomerisation in the oxidation of isoprene. Phys. Chem. Chem. Phys. 2011, 13, 13607−13613. (14) Berndt, T. Formation of carbonyls and hydroperoxyenals (HPALDs) from the OH radical reaction of isoprene for low-NOx conditions: Influence of temperature and water vapour content. J. Atmos. Chem. 2012, 69, 253−272. (15) Lelieveld, J.; Butler, T. M.; Crowley, J. N.; Dillon, T. J.; Fischer, H.; Ganzeveld, H.; Harder, H.; Lawrence, M. G.; Martinez, M.; Taraborrelli, D.; et al. Atmospheric oxidation capacity sustained by a tropical forest. Nature 2008, 452, 737−740. (16) St. Clair, J. M.; Rivera-Rios, J. C.; Crounse, J. D.; Knap, H. C.; Bates, K. H.; Teng, A. P.; Jorgensen, S.; Kjaergaard, H. G.; Keutsch, F. N.; Wennberg, P. O. Kinetics and products of the reaction of the firstI


Article

The Journal of Physical Chemistry A HC, and N) and NO3−(HNO3)n with H2SO4: Implications for atmospheric detection of H2SO4. J. Phys. Chem. A 1997, 101, 8275− 8278. (33) Mackay, G. I.; Bohme, D. K. Proton-transfer reactions in nitromethane at 297 K. Intern. Int. J. Mass Spectrom. Ion Phys. 1978, 26, 327−343. (34) Mackay, G. I.; Tanner, S. D.; Hopkinson, A. C.; Bohme, D. K. Gas-phase proton-transfer reactions of the hydronium ion at 298 K. Can. J. Chem. 1979, 57, 1518−1523. (35) Berresheim, H.; Elste, T.; Plass-Dülmer, C.; Eisele, F. L.; Tanner, D. J. Chemical ionization mass spectrometer for long-term measurements of atmospheric OH and H2SO4. Int. J. Mass Spectrom. 2000, 202, 91−109. (36) Ehn, M.; Thornton, J. A.; Kleist, E.; Sipilä, M.; Junninen, H.; Pullinen, I.; Springer, M.; Rubach, F.; Tillmann, R.; Lee, B.; et al. A large source of low-volatility secondary organic aerosol. Nature 2014, 506, 476−479. (37) Berndt, T.; Jokinen, T.; Sipilä, M.; Mauldin, R. L., III; Herrmann, H.; Stratmann, F.; Junninen, H.; Kulmala, M. H2SO4 Formation from the gas-phase reaction of stabilized Criegee intermediates with SO2: Influence of water vapour content and temperature. Atmos. Environ. 2014, 89, 603−612. (38) Eisele, F.; Tanner, D. Measurement of the gas phase concentration of H2SO4 and methane sulfonic acid and estimates of H2SO4 production and loss in the atmosphere. J. Geophys. Res. 1993, 98, 9001−9010. (39) Mauldin, R. L., III; Frost, G. J.; Chen, G.; Tanner, D. J.; Prevot, A. S. H.; Davis, D. D.; Eisele, F. L. OH measurements during the first aerosol characterization experiment (ACE 1): Observations and model comparisons. J. Geophys. Res. 1998, 103, 16713−16729. (40) Brophy, P.; Farmer, D. K. Clustering, methodology, and mechanistic insights into acetate chemical ionization using highresolution time-of-flight mass spectrometry. Atmos. Meas. Technol. Discuss. 2016, 1. (41) Wine, P. H.; Astalos, R. J.; Mauldin, R. L., III Kinetic and mechanistic study of the hydroxyl + formic acid reaction. J. Phys. Chem. 1985, 89, 2620−2624. (42) Vaghjiani, G. L.; Ravishankara, A. R. Kinetics and mechanism of OH reaction with CH3OOH. J. Phys. Chem. 1989, 93, 1948−1959. (43) Rissanen, M. P.; Kurten, T.; Sipilä, M.; Thornton, J. A.; Kangasluoma, J.; Sarnela, N.; Junninen, H.; Jorgensen, S.; Schallhart, S.; Kajos, M. K.; et al. The formation of highly oxidized multifunctional products in the ozonolysis of cyclohexene. J. Am. Chem. Soc. 2014, 136 (44), 15596−15606. (44) Atkinson, R. Kinetics and mechanisms of the gas-phase reactions of the hydroxyl radical with organic compounds under atmospheric conditions. Chem. Rev. 1986, 86, 69−201. (45) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.; Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J. Evaluated kinetic and photochemical data for atmospheric chemistry: Volume I - gas phase reactions of Ox, HOx, NOx and SOx species. Atmos. Chem. Phys. 2004, 4, 1461−1738. (46) Richters, S.; Herrmann, H.; Berndt, T. Highly oxidized RO2 radicals and consecutive products from the ozonolysis of three sesquiterpenes. Environ. Sci. Technol. 2016, 50, 2354−2362. (47) Atkinson, R.; Aschmann, S. M. Alkoxy radical isomerization products from the gas-phase OH radical-initiated reactions of 2,4dimethyl-2-pentanol and 3,5-dimethyl-3-hexanol. Environ. Sci. Technol. 1995, 29, 528−536. (48) Lightfoot, P. D.; Cox, R. A.; Crowley, M.; Hayman, G. D.; Zabel, F. Organic peroxy radicals: Kinetics, spectroscopy and tropospheric chemistry. Atmos. Environ., Part A 1992, 26, 1805−1961. (49) Schwantes, R. H.; Teng, A. P.; Nguyen, T. B.; Coggon, M. M.; Crounse, J. D.; St. Clair, J. M.; Zhang, X.; Schilling, K. A.; Seinfeld, J. H.; Wennberg, P. O. Isoprene NO3 oxidation products from the RO2 + HO2 pathway. J. Phys. Chem. A 2015, 119, 10158−10171. (50) Praske, E.; Crounse, J. D.; Bates, K. H.; Kurtén, T.; Kjaergaard, H. G.; Wennberg, P. O. Atmospheric fate of methyl vinyl ketone:

Peroxy radical reactions with NO and HO2. J. Phys. Chem. A 2015, 119, 4562−4572. (51) Stoer, J.; Bulirsch, R. Einführung in die Numerische Mathematik II; Springer Verlag: Berlin, 1978. (52) Pankow, J. F.; Asher, W. E. SIMPOL.1: a simple group contribution method for predicting vapor pressures and enthalpies of vaporization of multifunctional organic compounds. Atmos. Chem. Phys. 2008, 8, 2773−2796. (53) Riva, M.; Budisulistiorini, S. H. H.; Chen, Y.; Zhang, Z.; D’Ambro, E. L.; Zhang, X.; Gold, A.; Turpin, B. J.; Thornton, J. A.; Canagaratna, M. R.; et al. Chemical characterization of secondary organic aerosol from oxidation of isoprene hydroxyhydroperoxides. Environ. Sci. Technol. 2016, 50, 9889−9899. (54) Liu, J.; D’Ambro, E. L.; Lee, B. H.; Lopez-Hilfiker, F. D.; Zaveri, R. A.; Rivera-Rios, J. C.; Keutsch, F. N.; Iyer, S.; Kurten, Th.; Zhang, Z.; et al. Efficient isoprene secondary organic aerosol formation from a non-IEPOX pathway. Environ. Sci. Technol. 2016, 50, 9872−9880.

J


Highly Oxidized Second-Generation Products from the Gas-Phase

Recommend Documents