Influence of NH3 on secondary organic aerosols from the ozonolysis and photooxidation of α-pinene in a flow reactor

Atmospheric Environment 164 (2017) 71e84

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Influence of NH3 on secondary organic aerosols from the ozonolysis and photooxidation of a-pinene in a flow reactor Zaeem Bin Babar a, Jun-Hyun Park a, b, Ho-Jin Lim a, * a b

Department of Environmental Engineering, Kyungpook National University, Daegu 702-701, Republic of Korea Mass Spectrometry Research Center, Korea Basic Science Institute, Ochang 363-883, Republic of Korea

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


A new flow reactor (ID 15 cm x L 70 cm) was developed for studying SOA.
The reactor can simulate several days of photochemical aging in the atmosphere.
NH3 promoted the SOA yields and formed N-containing SOA species.
N-containing species enhanced the UV-vis absorbance and fluorescence of SOA.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 February 2017 Received in revised form 17 May 2017 Accepted 19 May 2017 Available online 25 May 2017

This study presents detailed characterizations of a newly-developed flow reactor including (1) residence time distribution measurements, (2) relative humidity (RH) and temperature control, and (3) OH radical exposure range (i.e., atmospheric aging time). Hydroxyl (OH) radical exposures ranged from 8.20  1010 to 7.22  1011 molecules cm3 s (0.5e4.9 d of atmospheric aging). In this study, the effects of NH3 gas on the secondary organic aerosol (SOA) formation of a-pinene by dark ozonolysis and photooxidation were investigated using the newly-developed flow reactor. For both dark a-pinene ozonolysis and photooxidation, higher SOA yields were observed in the presence of NH3 than in the absence of NH3. At RH of ~50%, the SOA yield for ozonolysis and photooxidation in the presence of NH3 increased by 23% and 15% relative to those in the absence of NH3. Similar effects were observed at lower and higher RH conditions. Fourier transform infrared spectroscopy analysis confirmed the presence of nitrogen-containing functional groups in SOA formed in the presence of NH3. The a-pinene SOA formed in the presence of NH3 showed higher absorption and fluorescence for UV-visible radiation than those formed in the absence of NH3. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Secondary organic aerosol a-Pinene NH3 Flow reactor Optical properties Brown carbon

1. Introduction Atmospheric aerosols play a key role in air quality, climate change, and human health (Belalcazar et al., 2009; Davidson et al., 2005; Harrison and Yin, 2000; IPCC, 2013; US EPA, 2012). These fine

* Corresponding author. E-mail address: [email protected] (H.-J. Lim). http://dx.doi.org/10.1016/j.atmosenv.2017.05.034 1352-2310/© 2017 Elsevier Ltd. All rights reserved.

particles are associated with mortality and serious health issues in humans such as cardiopulmonary and bronchial infections (Beelen et al., 2007; Brauer et al., 2012; Lall et al., 2004; WHO, 2013). They also influence the Earth's radiation budget by absorbing and scattering terrestrial and solar radiation; their physicochemical properties affect cloud formation processes (Khalizov et al., 2006). Therefore, understanding the formation mechanisms and key properties of aerosols is essential. Organic aerosols comprise a large

72

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

fraction of atmospheric aerosols (IPCC, 2001; Kanakidou et al., 2005; Putaud et al., 2010). Primary organic aerosols are directly emitted from sources such as vehicular exhaust, fossil fuel combustion, cooking, and biomass burning (Guzman-Morales et al., 2014; Ortega et al., 2013; Suarez-Bertoa et al., 2015). Gaseous organics emitted from biogenic and anthropogenic sources undergo atmospheric oxidation to form low-volatility compounds that partition into the aerosol-phase and form secondary organic aerosols (SOA) (de Gouw and Jimenez, 2009; Hallquist et al., 2009). A substantial fraction of fine organic particles consists of SOA (Hallquist et al., 2009). Recently, brown carbon has received significant attention because it is more abundant than black carbon and has a strong
r, effect on climate via direct light absorption (Andreae and Gelencse 2006; IPCC, 2013; Laskin et al., 2014). However, it is difficult to completely understand the origin of brown carbon and limited knowledge is available on the physicochemical properties of brown carbon. Reduced nitrogen species such as amino acids, amines, and ammonia react with carbonyl compounds to produce brown carbon re et al., 2009; (Galloway et al., 2008; Kampf et al., 2012; Nozie Shapiro et al., 2009). For instance, NHþ 4 ion- or NH3 gas-aging of SOA results in the formation of brown carbon through carbonyl-toimine conversion (Bones et al., 2010; Lee et al., 2013; Nguyen et al., 2012; Updyke et al., 2012). Over the last few decades, simplified laboratory techniques have substantially improved our understanding of the properties and atmospheric formation processes of SOA. Smog chambers have been successfully used as prime tools for physicochemical studies on SOA (Cocker III et al., 2001b; Hu et al., 2014; Paulsen et al., 2005; Platt et al., 2013; Presto et al., 2011; Wang et al., 2014). Smog chambers are batch reactors with sizes ranging from a few liters to 270 m3 (Rohrer et al., 2004). They are typically operated at low oxidant levels comparable to the atmospheric levels, which means long reaction times to simulate atmospheric processes (Kang et al., 2007). Fast particle wall-loss prevents the extension of the reaction time for smog chambers to follow the lifetime aging processes of particles up to several days (Matsunaga and Ziemann, 2010; Ng et al., 2010). Recently, flow reactors with compact sizes of 0.1 Le~1000 L and extremely high oxidant levels comparable to the lifetime of ambient aerosols have been developed as an alternative tool to smog chambers (Ezell et al., 2010; George et al., 2007; Kang et al., 2007; Lambe et al., 2011; Smith et al., 2009). The compact flow reactors have short residence times of several seconds to a few

minutes. In the flow reactor, substantially high levels of OH radicals are produced by the UV photolysis of ozone at high concentrations and successive reactions in humid conditions. To produce high levels of OH radicals, UV source different from the solar spectrum is used. Despite using UV radiation with wavelengths shorter than the solar radiation for O3 photolysis in OH radical formation, SOA yield and composition show similarities in smog chambers and flow tube reactors (Bruns et al., 2015; Kang et al., 2007; Lambe et al., 2015). This approach has been applied for simulating atmospheric photooxidation processes within a time scale of minutes and also for the SOA formation from various precursors, aging of primary aerosols, and inhalation toxicology (Hall IV et al., 2013; Kang et al., 2007; Keller and Burtscher, 2012; Lambe et al., 2011; Slowik et al., 2012). In this study, a new Kyungpook National University (KNU) flow reactor has been developed and applied to the SOA formation. The reactor was used to examine the effect of NH3 on the yield and composition of SOA. The formation of SOA from the dark ozonolysis and photooxidation of a-pinene in the presence and absence of NH3 was investigated. The formation of nitrogen-containing species was examined by the presence of nitrogen-containing functional groups (e.g., amine, amino acid, and nitrate). The effect of NH3 on the optical properties of SOA was also studied. 2. Materials and methods 2.1. Development and characterization of the KNU flow reactor A schematic of the Kyungpook National University (KNU) flow reactor is shown in Fig. 1. It consists of a reaction tube, reactor enclosure, irradiation source, purified air supply, reactant injection system, O3 and OH radical generation system, and analytical instruments. The frame of the reactor enclosure has the dimensions of 80 cm  30 cm  30 cm (L  W  H) and is constructed with aluminum profiles (2 cm  2 cm). The frame is covered with 2-mmthick stainless steel plates. The inner surfaces of the plates are covered with aluminum foil to enhance radiation reflection. The enclosure was sealed with 2-mm thick silicon foam attached between stainless steel plates and aluminum profiles. The inner enclosure space was purged with argon gas to maintain an inert atmosphere. Inside the enclosure, a 0.25 mm cylindrical FEP polymer tube (FHS113, Adtech Polymer Engineering, Stroud, Glos, England) with dimensions of 15.5 cm (I.D.)  70 cm (L) was fixed to the cone-shaped stainless steel ends. The FEP tube was sealed by

Fig. 1. A schematic of KNU flow reactor.

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

heating its surface on the stainless steel ends and tightened with hose clamps after wrapping it with adhesive Teflon tape (ASF110FR, Chukoh-flo, Japan). In order to cool the atmosphere of the enclosure, two 480 W Peltier cooler assemblies were installed on the top cover. These devices were controlled by a temperature controller (TZ4ST, Autonics, Korea) using solid state relays (SDD0-240N, Union Elecom, Korea) and a thermocouple (TJ36-CAXL-116U-18, Omega Engineering, CT). This electric cooling system increases the field applicability of the KNU flow reactor. The irradiation source contained four 32 W UV lamps (l ¼ 254 nm), coupled to a separate control module (HL10054010C, Bestec, Korea), placed in quartz tubings (I.D. 2.54 cm), each positioned at the corner of the enclosure side panels. Openings between the quartz tubings and the enclosure were sealed by Teflon plugs with O-rings. Argon gas was continuously purged through the quartz tubings to maintain an inert atmosphere and cool the lamps. An oil-free scroll compressor (SLPE-07ED, Anestiwata, Torino, Italy) was used to compress the ambient air dried with a condensing air dryer (NSE-10B, Orion Korea) and an adsorption dryer (PD015SFDR07, Walker, WA). Volatile organic compounds (VOCs), NOx, and SO2 in dry air were removed by passing them through two successive beds loaded with a mixture of activated carbon and activated carbon impregnated with phosphoric acid and KMnO4-impregnated alumina (Incul-Aire, Quebec, Canada). Finally, the air was passed through two parallel HEPA filters (PN 12144, Pall, WA) to remove particles. The purified air in the reactor had negligible amounts of VOCs, NOx, and ozone (all less than 1 ppb). The background particle number concentration was always less than 5 counts cm3. The quality of the purified air is described in detail in a previous publication (Babar et al., 2016). Reactive organic gas (ROG) was injected into the flow reactor by using a pressurized cylinder with a specific ROG concentration. For the preparation of the cylinder, the empty cylinder was first purged four times with pressurized ultra-high pure N2 gas (purity > 99.999%) at 3 bar and vacuumed to 160 mmHg. The process of purging and vacuuming was repeated twice to ensure the proper cleaning of the cylinder. Then, the clean cylinder in the vacuumed state was connected to an ROG evaporator. The ROG evaporator is a U-shaped 1/400 stainless steel tubing heated by a 530 W heating tape (HT2520, MTOPS, Korea) maintained at 120  C by a temperature controller (TZ4ST, Autonics, Korea). A known amount of liquid ROG was injected into the ROG evaporator using a microliter syringe (701N, Hamilton, Nevada, USA). Evaporated ROG was introduced to the vacuumed cylinder at 30 mL min1 N2 flow for 40 min. The cylinder was then pressurized by N2 to 6 bar. Finally, the cylinder was mixed thoroughly for 1 h by rolling it on the floor. NH3 gas was injected into the reactor using 100 ppm NH3 gas cylinder (PS CHEM Co. Ltd). O3 was generated by passing high purity O2 through a stainless steel housing with a UV lamp (l ¼ 185 nm). The O3 level was adjusted by controlling the O2 flow to the O3 generator. The flow of pure air, ROG, O3, and NH3 was controlled by mass flow controllers coupled to a computer with flow managing software (TSC-120, KC-mass, Korea). At the inlet of the flow reactor, pure airline split into two lines. One line directly entered the flow reactor. The other airflow line was controlled by a rotameter (RMA-21-SSV, Dwyer, USA), heated by a 530 W cartridge heater and controlled by a temperature controller (TZ4ST, Autonics, Korea) to evaporate H2O in a water evaporator. Pure water (Merck KGaA, Darmstadt, Germany) was injected into the water evaporator through 1/1600 PEEK tubings using a microliter syringe (1025TLL, Hamilton, Nevada, USA) fixed on a syringe pump (Fusion 100T, Chemyx Inc., USA). OH radical generation was instantly carried out in the flow tube

73

reactor by O3 photolysis in the presence of water vapor under UV irradiation (l ¼ 254 nm) in the enclosure. Oxygen radicals were generated by the photodissociation of O3 and then reacted with H2O to form OH radicals as follows:

O3 þ hn/O2 þ O$

(1)

O$þH2 O/2OH$

(2)

No degradation of a-pinene was reported under the conditions used to generate OH radicals (Raff, 2007). The instrumentation for gas and aerosol phase is presented as follows: Sequential Mobility Particle Sizer (SMPSþC, Grimm, Germany) consisting of a differential mobility analyzer (DMA) and a condensation particle counter (CPC) were used to measure aerosol size distribution. The DMA has a size measurement range of 11.0e1083.3 nm. Aerosol and sheath flow rates were 0.3 and 3.0 L min1. The NOx analyzer (42C, Thermo Electron, USA), the O3 analyzer (49C, Thermo Environmental Instruments, USA), and the CO2 sensor (GC0010, COZIR, UK) were connected to a sampling manifold at the outlet of the flow reactor for continuous measurement of NO, NO2, O3, and CO2. The calibration of NOx and O3 analyzers was carried out once per week. The detection limit and accuracy of NOx and O3 analyzers were 0.5 ppb and ±1%, respectively. RH and temperature at the inlet of the flow reactor were measured by a hygro-thermometer (KN-2000W, Konics, Korea) with a measurement range of 0e100% and -40e80  C and an accuracy of RH ± 1% and ±0.2  C, respectively. The Syntech Spectras VOC analyzer (GC-PID GC955 series 600) was used for the online measurement of VOC concentrations. The span time was set to 18 min and the sample flow rate was 1.5 mL min1. The VOC analyzer had a minimum detection limit of 0.15 ppb. Sub-micron aerosol particles were generated by atomizing the aqueous solution of 0.03 M sucrose using a home-made collision atomizer similar to TSI Model 3076. A diffusion dryer containing silica gel was used for drying the droplets from the atomizer. Aerosol charge neutralizer (4530 HCT, Korea) was used to neutralize dried aerosols to avoid electrostatic deposition on the wall of the reaction tube. Residence time distributions (RTDs) of the KNU flow reactor for the gas and particle phase species were measured by introducing pulses of CO2, NO, and sucrose particles for 15 s into the reactor. The pulses of gases (i.e., CO2 or NO) and sucrose particles were generated by manually switching the mass flow controllers (TSC-210, KCmass, Korea) and a three-way valve, respectively. For the RTD measurement using sucrose particles, CPC was used to measure the total particle concentrations at the exit of the flow reactor. For evaluating the wall losses of particles and a-pinene, particle size distributions and concentration of a-pinene were measured at the inlet, center, and outlet of the cylindrical section of the reaction tube, respectively. Details of the experimental methods and results of these evaluations are presented in sections S2 and S3 of the Supplementary Material, respectively. The list of experiments for dark ozonolysis and photooxidation of a-pinene is summarized in Table 1. For all dark ozonolysis experiments, initial a-pinene (ROG) and O3 concentrations were approximately 200 ppb and 1000 ppb, respectively. The effect of NH3 (~400 ppb) on SOA formation and yield was studied at low (RH < 2%), medium (RH ~50%), and high (RH ~90%) humidity conditions. The effect of initial NH3 concentration on the SOA yield for dark a-pinene ozonolysis was examined by changing the NH3/ROG ratio from zero to four at a fixed a-pinene concentration of ~200 ppb and RH ~50%. SOA yields were compared at ROG concentrations of 50, 100, 200, and 400 ppb with and without NH3. aPinene photooxidation experiments were carried out at various atmospheric aging times (0.5, 1.0, and 3 d) at RH 50% for 200 ppb a-

74

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

Table 1 Experimental results for SOA formation from a-pinene. Exp. ID

ROGi (ppb)

ROGf (ppb)

DROG (mg/m3)

NH3,i (ppb)

RH (%)

O3,f (ppb)

OHa (d)

SOAsmps (mg/m3)

YSOA (%)

Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp Exp

194 204 193 197 190 193 187 196 195 54 51 97 96 378 382 200 193 189 181 185 187 56 52 26 23 90 90

12 13 17 18 21 18 14 14 15 3 2 8 7 92 40 0 0 0 0 0 0 0 0 0 0 0 0

1017 1061 986 999 931 972 695 1011 1004 283 273 493 491 1591 1903 1112 1075 1145 1007 1028 1040 313 291 142 125 501 502

e 400 e 400 e 400 100 200 800 e 100 e 200 800 800 e 400 e 400 e 400 e 100 e 50 e 200

1.6 1.3 46.9 48.3 87.0 87.2 46.8 47.3 47.6 46.3 46.9 46.7 47.1 47.5 47.7 47.9 48.3 46.9 48.5 48.2 47.8 47.3 48.1 48.3 47.7 47.1 47.7

972 944 892 911 829 906 943 948 960 966 985 956 938 931 966 444 440 813 812 1277 1318 478 455 505 465 459 460

e e e e e e e e e e e e e e e 0.5 0.5 1.0 1.0 3.0 3.0 0.5 0.5 0.5 0.5 0.5 0.5

99.7 118.6 92.3 115.8 133.2 156.4 112.5 118.5 118.2 12.5 13.6 35.2 37.7 246.2 341 629.5 698.9 509.6 553.0 431.5 499.5 102.8 142.9 25.5 29.4 226.6 253.4

9.8 11.2 9.4 11.6 14.3 16.1 11.7 11.7 11.8 4.4 5.0 7.2 7.7 15.5 17.9 56.7 65.1 44.5 56.2 42.0 48.1 35.4 43.3 17.6 23.4 45.1 50.4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

a OH radical exposure in days assuming the atmospheric OH level of 1.5  106 molecules cm3.

pinene. SOA yields from photooxidation were also compared for samples with and without NH3. For any experiment, the flow reactor was run at a set a-pinene concentration and RH in series without and with O3 introduction for 2.5 and 3 h, respectively. The experiment was repeated to duplicate the experiments. The apinene concentration was the average of the last three consecutive readings at each regime. O3 and SOAsmps levels were the last 1 h average of the readings. The coefficient of variation for O3 and SOAsmps were <2.0% and <5.0%, respectively. NH3 was not measured because an NH3 analyzer was not available. The a-pinene SOA density of 1.32 g cm3 was reported in recent studies (Ng et al., 2007). In this study, the SOA mass concentration of SOAsmps was obtained for particles smaller than 700 nm in diameter by assuming that the a-pinene SOA density is 1.3 g cm3. All experiments were conducted at 25 ± 1  C without OH scavengers or seed particles. The effects of NH3 on optical properties (i.e., UV-Vis absorbance and fluorescence) and bulk composition of a-pinene SOA were also examined. For these examinations, dark ozonolysis at ~2500 ppb O3 and photooxidation at atmospheric aging times of 0.5 and 3 d at ~2500 ppb a-pinene were conducted in the absence and presence of NH3. SOA samples were collected using 47-mm Teflon filters. The Teflon filters were weighed before and after sampling using a microbalance (CPA2P-F, Sartorius, Germany). The experimental results are summarized in Table S1 of Supplementary Material. Ultrasonication of SOA samples was carried out in 4e8 mL of methanol at 17  C for 30 min. The amount of methanol was specified to maintain dissolved SOA concentrations at 1000 ppm by each SOA sample mass. For the analysis of UV-Vis absorption and fluorescence, 3 mL of the SOA extract was used. For the bulk composition analysis, 1.0 mL of the SOA extract was concentrated to 50 mL by blowing a gentle stream of N2 gas. An aliquot of the SOA concentrate (10 mL) was dropped in an aluminum-foil well with a flat bottom (4 mm diameter x 2 mm depth) using a micro syringe. It was dried for 10 min at room temperature, put in a Petri dish, and

kept in the freezer until the following analysis. The analysis of UVVis absorption was performed using a double beam UV-Visible spectrometer (UV-1650PC, SHIMADZU, Japan). The analysis of fluorescence was conducted using a spectrofluorometer (FP-6500, JASCO, Japan) with 1 nm intervals for emissions from 220 to 600 nm and 10 nm step change for excitation from 220 to 500 nm. SOA samples on the aluminum well were analyzed using Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) spectroscopy (Perkin Elmer, Frontier, Waltham, MA, USA). For each sample, three consecutive FTIR analyses were performed with 20 scans at 2 cm1 resolution between 600 and 4000 cm1 and the mean spectra of the multiple analyses are presented in section 3. 3. Results and discussion 3.1. Characterization of KNU flow reactor The results of temperature and RH controls of the KNU flow reactor are provided in Supplementary Material (Fig. S1). Initially, all UV lamps were turned on to attain a constant maximum temperature of 43  C without running Peltier coolers. The temperature was decreased in four stages from 43  C to the set temperatures of 35  C, 30  C, 25  C, and 20  C with Peltier coolers powered on. Corresponding to the set temperatures, actual temperatures of 34.8 ± 0.4  C, 29.6 ± 0.7  C, 25.3 ± 0.7  C, and 21.5 ± 0  C were measured. A minimum temperature of 21.5  C can be achieved with all UV lamps turned on (Fig. S1(a)). With one UV lamp in service, the minimum temperature went down to 11  C. Proper temperature control of the flow reactor would allow SOA formation in a relatively wide temperature range of 21.5e40  C. In contrast to the KNU flow reactor, cooling gas (i.e., N2 and air) was used to dissipate heat generated by the UV lamps in other flow reactors (George et al., 2007; Kang et al., 2007; Smith et al., 2009). Smith et al. (2009) observed a flow tube temperature of 35  C with a cooling system that used pressurized air. SOA yield corrections in the range of 10e20% were applied for the temperature increase by UV lamps with a correction factor of 0.02 per K (Lambe et al., 2011). Measured RH readings of 11.6 ± 1.01%, 24.7 ± 0.2%, 38.1 ± 0.6%, 54.3 ± 0.7%, and 69.8 ± 0.4% were slightly lower than the water vapor injection rates for RH values of 13.2%, 27.9%, 42.6%, 57.3%, and 72%, respectively (Fig. S1(b)). The relative difference between the water vapor injection rates for the set and measured RH decreased with the increase in RH; the differences were 13% and 4% at the set RH values of 13% and 72%, respectively. These differences night be due to the adsorption of water on the surface of the reaction tube. Fig. 2 shows the measured and estimated RTD for ideal plug flow and ideal laminar flow reactors. None of the reactors had an axial

Fig. 2. Residence time distribution flow in KNU flow reactor and corresponding ideal plug flow and ideal laminar RTDs.

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

mixing but the laminar flow reactor had resistance at the wall interface. The ideal laminar flow reactor shows a wider RTD with a long tail than the plug flow reactor. Measured RTDs for both CO2 and NO were non-ideal due to axial mixing. However, their shapes were almost similar to the RTD for the ideal laminar flow reactor except for the early arrival and tailing. NO showed much longer tailing caused by its stronger adherence to the wall of the flow tube than that of CO2. Peak CO2 and NO concentrations were monitored at approximately 0.6 of the plug flow residence time. Measured RTD for sucrose particles was slightly wider than that of CO2. This might be due to the potential interaction of sucrose particles with the wall of the reaction tube. The measured RTD for this flow reactor is comparable to the RTD for the Toronto Photooxidation Tube (TPOT) and Potential Aerosol Mass (PAM) (Lambe et al., 2011). TPOT has dimensions of 34 cm (L)  7.3 cm (I.D.) with a gas velocity of 0.37 cm s1 at 0.9 L min1, while PAM has dimensions of 46 cm (L)  22 cm (I.D.) with a gas velocity of 0.35 cm s1 at 8.5 L min1. The KNU flow reactor has a gas velocity of 0.35 cm s1 at 4 L min1 comparable to that of TPOT and PAM. Particle penetrations for particle sizes of 11.1 nm, 50.3 nm, and 101.1 nm were 0.61, 0.89, and 0.93, respectively. Loss of small particles less than 50 nm in diameter was higher than that of particles greater than 50 nm. This suggests the minimal effect of particle wall loss on SOA yields in the SOA experiments addressed below. Particle penetrations as a function of particle size are shown in Fig. S2 of the Supplementary Material. Loss of reactive organic gases (i.e., a-pinene) was due to the adsorption on the wall of the reaction tube during initial injection (Fig. S3 of Supplementary Material). This initial loss of a-pinene had a negligible effect on the evaluation of SOA yields since all SOA experiments were performed after stabilization for 2.5 h. OH concentrations are determined by UV radiation intensity, O3 concentration, and RH. Atmospheric aging time is affected by OH concentrations and residence time in a flow reactor. In this study, OH radical concentrations were estimated from the photochemical decay of toluene. Toluene is an aromatic compound with a wellknown OH reaction rate constant (kOH) of 5.48  1012 molecules cm3 s1 at 298 K and a negligible rate of reaction with O3 (Atkinson and Aschmann, 1989). For the range of O3 from 2000 to 8000 ppb and RH from 10 to 60%, OH concentrations were calculated as 3.4  108e3.4  109 molecules cm3 by assuming the firstorder decay of toluene by the reaction with OH radical. In comparison with the atmospheric OH concentration of 1.5  106 molecules cm3, this exposure range corresponds to the atmospheric aging time of 0.49 de4.9 d (8.2  1010 to 7.2  1011 molecules cm3

75

s) (Fig. 3) (Mao et al., 2009). Atmospheric aging time tatm (d) was estimated using following equation:

tatm ¼

OHFR tFR OHatm

(3)

where OHFR (molecules cm3) is the concentration of OH radical, tFR (d) is the residence time of flow reactor, and OHatm (molecules cm3) is the concentration of OH radical in the atmosphere. Atmospheric aging time in this study covered the atmospheric life time of particles in less than 5 d (Graf et al., 1998). The OH exposure was lower than that in previous studies (1.8  1011e2.4  1012 molecules cm3 s) (Kang et al., 2011; Lambe et al., 2011; Slowik et al., 2012). The difference is largely due to lower O3 levels compared to the previous studies. PAM and TPOT were operated under OH exposures of up to 2.4  1012 molecules cm3 s and 1.8  1012 molecules cm3 s with maximum O3 concentrations of 9 ppm and 5 ppm, respectively. The maximum OH level in this study was still approximately an order of magnitude greater than the OH exposure of 7.5  1010e1.2  1011 molecules cm3 s in a typical environmental chamber (Lambe et al., 2011). Fig. 3 shows a contour plot of residual O3 concentrations in the reactor as a function of RH and O3. The residual O3 concentration was in the range of 500e3500 ppb with higher levels at higher initial O3 concentrations and lower RH values. The O3-to-OH ratio is higher at high RH as observed in the atmosphere. 3.2. Effects of NH3 on SOA formation In the case of ozonolysis, under the conditions of low, medium, and high RH with and without NH3, temporal profiles of SOA, ROG, and O3 are shown in Fig. 4. Under low RH (dry) conditions in the presence of NH3, SOA yield was significantly higher than that in the absence of NH3. SOA concentrations of 119 ± 4 mg m3 and 100 ± 2 mg m3 resulted in an SOA yield of 11.2% and 9.8% with and without NH3, respectively. The SOA yield increased by 13% as a response to NH3 addition. Na et al. (2007) observed a 23% increase in the SOA yield for dark a-pinene ozonolysis in dry conditions. At medium and high RH, the SOA yields in the presence of NH3 were 11.6% and 16.1%; in the absence of NH3, they were 9.4% and 14.3%, respectively. At medium RH (~50%), in the presence of NH3, the apinene SOA yield increased by approximately 24% relative to that in the absence of NH3. This is comparable to the 31% increase in the SOA yield for a-pinene ozonolysis in the presence of NH3 at RH ~50% (Na et al., 2007). At low, medium, and high RH, the increase in particle number concentrations in the presence of NH3 was evident

Fig. 3. Contour plots of (a) estimated OH exposure range, (b) residual O3 concentration as a function of O3 (ppb) and RH (%). OH exposure corresponding to the atmospheric aging time of one day is 1.64  1011 molecules cm3 s.

76

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

Fig. 4. Aerosol and gas-phase profiles in dry conditions for a-pinene ozonolysis in the (a) absence and (b) presence of NH3, medium RH in the (c) absence and (d) presence of NH3, and high RH in the (e) absence and (f) presence of NH3.

(Fig. 5 (a)). The increase in the SOA yield might be the result of new particle formation in the presence of NH3. The new particles might be ammonium salts formed by the reaction of ammonia and gasphase organic acids such as pinic and pinonic acids (Na et al., 2007). Liu et al. (2015) also observed enhanced particle formation in the presence of NH3 in gasoline vehicle exhaust. NH3 can lower the volatility of organic acids by several orders of magnitude (Paciga et al., 2014), enhancing the partitioning of low-volatility organic acids between the particles and the gas phase to further increase the SOA yield. Consequently, SOA formation is accelerated in the presence of NH3 as shown in Fig. 5 (c). Liu et al. (2015) observed 8.9% of nitrogen-containing organics in a-pinene SOA as a result of the subsequent NH3 uptake after 6 h of ozonolysis at 50% RH. 3.3. Humidity effects In the case of a-pinene ozonolysis, particle, surface, and SOA size distributions as a function of particle diameter Dp at low, medium,

and high RH with and without NH3 are shown in Fig. 5. At medium RH (~50%) without NH3, the SOA yield was slightly lower than in dry conditions. At medium RH, the number concentration of particles with Dp in the range of 100e300 nm was lower than at low RH. Particles with Dp in the range of 100e300 nm contribute more towards SOA formation than particles with Dp less than 100 nm (Fig. 5 (c)). As a result, the SOA yield was lower at medium RH than at low RH, likely because of the suppression of organic acids in the particle-phase at medium RH compared to the dry conditions. The SOA yield data for a-pinene ozonolysis in Hu et al. (2014) was also lower in humid conditions. Na et al. (2007) observed higher SOA yields under dry conditions than under humid conditions whereas in other studies (Fick et al., 2003; Rohr et al., 2003) on a-pinene ozonolysis, at RH of 13e45%, there was almost a negligible effect of RH on the SOA yield. Bonn et al. (2002) suggested that the effect of water on SOA formation can vary at very high and low ROG concentrations. The main difference between the previous studies (Fick et al., 2003; Rohr et al., 2003) and the current study is the concentration of a-pinene used in the experiments, which can be the

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

77

Fig. 5. Size distributions as a function of particle diameter Dp at low, medium, and high RH with and without NH3 for a-pinene ozonolysis (a) number concentration, (b) surface concentration, and (c) mass concentration.

reason of the different effects of medium RH on SOA formation. In this study, the intermediate a-pinene concentration was ~200 ppb. Fick et al. (2003) and Rohr et al. (2003) used a low a-pinene concentration of 20 ppb and very high a-pinene concentrations of 56e266 ppm, respectively. At high RH (90%) without NH3, the SOA yield was higher than it was at medium RH and dry conditions because of the higher number concentration of particles with Dp of 171e692 nm at RH 90% than at medium and low RH. At high RH, a significant number of large particles are formed due to water uptake (Cocker III et al., 2001a; Jonsson et al., 2006; Virkkula et al., 1999). The geometric mean diameter of particles at high RH was 48 nm, significantly higher than 39 nm at dry conditions. The water uptake by particles is characterized by the “growth factor” which is the ratio of the diameter in wet conditions to the diameter in dry conditions (Dwet/ Ddry). For a-pinene ozonolysis at RH 85%, growth factors of 1.07 and 1.09 have been reported as a result of water uptake (Cocker III et al., 2001a; Virkkula et al., 1999). In this study, the measured growth factor of 1.23 was significantly higher than those of previously reported growth factors of 1.07 and 1.09. At high RH (90%), particles grow as a result of condensation of low volatility compounds in addition to the predominant physical uptake of water by the particles. The measured growth factor might also be slightly higher because the presence of newly formed particles can affect the condensational dynamics (Varutbangkul et al., 2006). The aerosol particles primarily had a bimodal distribution (Fig. 5(a)). In the smaller mode, particles were in the range of 11e80 nm in diameter. This range represents the formation of new particles as result of a-pinene oxidation. In the larger mode, the particles were in the range of 85e300 nm in diameter. The large

particles are formed by the condensation of low volatility compounds on the existing particles and the coagulation of smaller particles as a result of the relative motion between them. 3.4. Effects of NH3/a-pinene mixing ratio Fig. 6(a) shows the comparison of SOA yields at the NH3/apinene ratios of 0, 0.5, 1.0, 2.0, and 4.0 with a fixed a-pinene concentration of 200 ppb and RH of 50%. The SOA yield increased from 7.2% to 9.0% as the NH3/a-pinene ratio increased from 0 to 0.5. However, the SOA yield remained nearly constant even with a further increase in the NH3/a-pinene ratio from 0.5 to 4.0. This indicates that organic acids responsible for the increase in SOA formation are utilized by reacting with NH3 at the NH3/a-pinene ratio of 0.5. Na et al. (2007) reported that the increase in the aerosol formation from a-pinene ozonolysis ceased as the NH3 concentration exceeded 200 ppb. 3.5. Effects of a-pinene concentration For ozonolysis, the comparison of SOA yield with and without NH3 at a-pinene concentrations of 50, 100, 200, and 400 ppb and at RH ~50% is shown in Fig. 6(b). An NH3/a-pinene ratio of two was maintained in the experiments carried out in the presence of NH3. The SOA yield increased with the increase in a-pinene concentration regardless of the NH3 presence. However, the SOA yield was higher in the presence of NH3 than in the absence of NH3. In the presence of NH3, the increase in the SOA yield was higher at high apinene concentrations than at low a-pinene concentrations (Fig. 6(b)) because a-pinene at high concentrations is more likely to

78

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

Fig. 6. SOA yield as a function of (a) NH3/a-pinene at 200 ppb a-pinene concentration and (b) a-pinene concentrations at mid RH.

react with NH3 to produce high-concentration organic acids and enhance SOA formation than a-pinene at lower concentrations. 3.6. Estimation of partitioning coefficients The SOA mass yield (Y) as a function of aerosol mass concentration for dark a-pinene ozonolysis and photooxidation in the presence and absence of NH3 is shown in Fig. S4 (section S4 of Supplementary Material). The SOA yield is the ratio of SOA mass concentration (Mo, in mg m3) to the amount of ROG reacted (DROG, in mg m3):

Y¼

Mo

(4)

DROG

The curve fitting for both data sets (with and without NH3) excluding data points at high RH was carried out using the twoproduct absorption model given as:

 Y ¼ Mo

a1 K1 1 þ K1 Mo

þ

a2 K2

 (5)

1 þ K2 Mo

where K1 and K2 are partitioning coefficients (m3 mg1) and a1 and a2 are dimensionless stoichiometric factors of high-volatility and low-volatility aerosol products, respectively. In order to determine the SOA yield for dark a-pinene ozonolysis without NH3, K1 ; K2 ; a1 ;

and a2 were estimated as 0.2080, 0.0017, 0.0453, and 0.3000 (R2 ¼ 0.9986). The partitioning coefficients for dark a-pinene ozonolysis and photooxidation with and without NH3 are compared in Table 2. In previous studies, the partitioning coefficients were estimated using the two-product model for the SOA yield data at various temperatures (Odum Jay et al., 1996; Hoffmann et al., 1997; Griffin et al., 1999; Cocker III et al., 2001b; Na et al., 2007; Saathoff et al., 2008; Wang et al., 2011, 2014). For a better comparison, the partitioning coefficients in the previous studies were corrected to the temperature of this study (i.e., 25  C) using the CalusiusClapeyron equation. The temperature dependence of the partitioning coefficient Ki is given as (Chung and Seinfeld, 2002):

   Ki ðT2 Þ DH 1 1 ¼ exp   Ki ðT1 Þ R T2 T1

(6)

where Ki(T1) and Ki(T2) are the partitioning coefficients at temperatures T1 and T2, respectively, and DH is the enthalpy of vaporization. For the temperature correction, DH of 38 kJ mol1 at 25  C was used (Offenberg et al., 2006). The partitioning coefficients in this study lie within the range of other smog chamber studies for dark a-pinene ozonolysis (Cocker III et al., 2001b; Griffin et al., 1999; Hoffmann et al., 1997; Na et al., 2007; Saathoff et al., 2008; Wang et al., 2011, 2014). The SOA yield data with an assumed SOA density of 1.3 g cm3, without OH scavengers or seed

Table 2 Summary of partitioning coefficients (K1 and K2) and stoichiometric coefficients (a1 and a2) estimated using two-product absorption model for the dark ozonolysis and photooxidation of a-pinene. Reaction

K1 (m3 mg1)

K2 (m3 mg1)

a1

a2

Reactor

References

O3 w /NH3b O3 w/ NH3b O3 w/o NH3a O3 w/o NH3b O3 w/o NH3a O3 w/o NH3c O3 w/o NH3a O3 w/o NH3a O3 w/ NH3c OH w/o NH3b OH w/ NH3b OH w/o NH3b OH w/o NH3b OH w/o NH3d

0.208 0.231 0.069 e 2.173 e 0.155 e 0.0003 e 0.340 e 0.570 e 0.011 e 0.028 0.038 0.644 e 0.510 e 0.017

0.002 0.003 0.002 0.027 0.139 0.015 0.003 0.014 0.010 0.002 0.002 0.019 0.012 0.026

0.045 0.045 0.239 0.168 0.125 0.366 0.11 0.12 0.333 0.344 0.329 0.019 0.038 0.200

0.300 0.205 0.169 0.222 0.102 0.489 0.29 0.19 0.329 0.245 0.400 0.23 0.326 0.396

Flow reactor Flow reactor Smog chamber Smog chamber Smog chamber Smog chamber Smog chamber Smog chamber Smog chamber Flow reactor Flow reactor Smog chamber Smog chamber PAM

This study This study Cocker III et al. (2001b) Saathoff et al. (2008) Griffin et al. (1999) Na et al. (2007) Wang et al. (2011) Hoffmann et al. (1997) Na et al. (2007) This study This study Odum Jay et al. (1996) Hoffmann et al. (1997) Kang et al. (2007)

a b c d e

e e e e e e e

e e

Dry condition without seed and OH scavenger. RH 50% without seed and OH scavenger. RH 50% without seed but with OH scavenger. RH 20% without seed and OH scavenger, K1 and K2 were estimated using data in Kang et al. (2007). For reference studies, K1 and K2 were corrected to temperature of 25  C using equation (6).

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

Fig. 7. SOA yield as a function of organic mass concentration predicted by two-product absorption model using various reference data in the absence of NH3.

particles, have been evaluated by all those studies with the exception of Na et al. (2007). The SOA yield data in Na et al. (2007) were with OH scavengers and without seed particles. However, in this study of the flow reactor, the SOA yield curve of a-pinene ozonolysis was below the curves of the other smog chamber studies (Fig. 7), because the short reaction time in the flow reactor restricts further evolution (decreasing volatility) of the products of a-pinene ozonolysis. Consequently, short reaction time in the flow tube reactor is the main reason of lower SOA yield curve compared to those of the smog chamber studies. Flow reactors have residence times ranging from a few seconds to minutes (George et al., 2007; Kang et al., 2007; Lambe et al., 2015, 2011). In this study, the SOA yield of 9.8% was obtained for ~200 ppb apinene and ~1 ppm O3 in the absence of NH3 under dry conditions. This is comparable with the a-pinene SOA yield of 3% at 1 ppm O3 without NH3 under dry conditions for 100 ppb a-pinene (Kang et al., 2007). In the same study, 12 ppm O3 was required for an SOA yield of 20% in the absence of NH3 under dry conditions using PAM. This indicates that high O3 concentrations are required in flow reactors to make comparisons with yields generated by smog chambers. In Lambe et al. (2011), for 262e263 ppb a-pinene at RH 25e30%, an SOA yield of 10% was obtained without NH3 in TPOT. This is comparable with the SOA yield of 9.4% for ~200 ppb a-pinene without NH3 at RH ~50% in the current study. High a-pinene SOA yield in TPOT might be caused by high ROG concentrations used in the experiment (Kang et al., 2007; Odum Jay et al., 1996). In Chhabra et al. (2015), the a-pinene SOA yield was estimated by developing an algorithm using the results from acetate chemical ionization mass spectrometry (acetate-CIMS) employed in apinene ozonolysis experiments. The SOA yield of a-pinene ozonolysis was estimated to be in the approximate range of 11e24% for 50e100 ppb of a-pinene at 5 ppm O3 and RH 30% for PAM experiments without NH3. The estimated yield of 11% for 50 ppb apinene in Chhabra et al. (2015) is comparable with the SOA yield of 9.4% at RH 50% and 1 ppm O3 for 200 ppb a-pinene in the absence of NH3 in the present study. The higher yield in Chhabra et al. (2015) might be caused by higher O3 concentrations (5 ppm) used in the experiments. In another study of a flow tube reactor, a very low SOA yield of 0.12% was obtained for 43 ppm a-pinene at 550 ppb O3 without NH3 (Gao et al., 2010). The main reason for the very low yield was the reaction conditions of extremely high a-pinene concentrations and very low O3 concentrations. In Gao et al. (2010), high concentrations of a-pinene were used to minimize the

79

fluctuations in the mass loadings because a-pinene concentrations could not be measured in real-time. In the presence of NH3, K1 ; K2 ; a1 ; anda2 were estimated as 0.2307, 0.0033, 0.045, and 0.2046. The comparison of SOA yields at different atmospheric aging times (0.5 d, 1 d, 3 d) and RH 50% with and without NH3 for 200 ppb a-pinene is presented in Table 1 (see Exp 16e21). The SOA mass concentration and yield decreased with increasing atmospheric aging time regardless of NH3. Without NH3, the SOA mass and yield decreased from 629 mg m3 and 56.7% to 431 mg m3 and 42% for apinene, respectively, with an increase in OH levels from 0.5 to 3 d. At low and high atmospheric aging times (0.5 and 3 d), the SOA yield in the presence of NH3 was 15% higher than in the absence of NH3. In Lambe et al. (2011), similar observations for the decrease in SOA yields with increasing OH levels were noted for a-pinene and m-xylene. Kang et al. (2011) also showed that SOA concentrations for a-pinene decreased with increasing OH exposure. In Kang et al. (2011), the SOA yield decreased from 54% to 45% as the OH exposure increased from 2 d to 10 d for 40 ppb a-pinene. The potential oxidation and fragmentation of SOA generate higher-volatility gasphase products which cause the loss of aerosol particles and thus suppress aerosol growth (Heald et al., 2010; Smith et al., 2009). In the PAM reactor study, Kang et al. (2007) reported a-pinene SOA yield of 45% at RH 50% and 5 ppm O3 for 100 ppb a-pinene at 15  C. This is comparable with the SOA yield of 42% at RH 50% and 6 ppm O3 for 200 ppb a-pinene at 25  C in the present study. The higher yield in Kang et al. (2007) might be caused by the temperature difference (15  C versus 25  C in the current study). The a-pinene SOA yield of 54% was obtained at the atmospheric aging time of 2 d for 40 ppb a-pinene (Kang et al., 2011), which was higher than the SOA yield of 44.5% at the atmospheric aging time of 1 d in the present study. The discrepancy is a likely result of the difference in the approach for achieving the atmospheric aging time. For atmospheric aging times ranging from 2 to 20 d, Kang et al. (2011) changed RH from 3 to 45% at a fixed O3 concentration of 9 ppm whereas, in this study, O3 was changed from 2 to 8 ppm to achieve the atmospheric aging times of 0.5e5.0 d. In another PAM reactor study, Lambe et al. (2011) observed a-pinene SOA yield of 19% at an atmospheric aging time of 0.2 d for 78e88 ppb a-pinene. However, in the current study, at the atmospheric aging time of 0.5 d, the apinene SOA yield was 56.4% higher than that of the PAM reactor in Lambe et al. (2011) because of the high a-pinene concentration of ~200 ppb compared to 78e88 ppb used in Lambe et al. (2011). The SOA yield curves of Odum Jay et al. (1996) and Hoffmann et al. (1997) for a-pinene photooxidation were below the curve of this study because of the NOx experimental conditions used in those two studies. High NOx levels limit particle formation by suppressing the particle surface where low volatility compounds condense (Sarrafzadeh et al., 2016). The comparison of a-pinene SOA yield for ozonolysis and photooxidation is shown in Fig. S4 (section S4 of Supplementary Material). It is evident that the apinene SOA yield for photooxidation was approximately six times higher than that for dark ozonolysis at the same RH and temperature. Kang et al. (2007) observed approximately five times higher SOA yields under UV irradiation than dark ozonolysis at similar temperature conditions. As explained earlier, lower SOA yield of apinene for ozonolysis relative to photooxidation might be caused by short reaction time in the flow reactor (George et al., 2007; Kang et al., 2007; Lambe et al., 2015, 2011). 3.7. Bulk composition analysis of SOA The strong peaks of C-H and C¼O stretchings were commonly detected for organic aerosols (Bruns et al., 2010; Maria et al., 2002; Mohiuddin et al., 2016; Roberts et al., 2016). C-H stretching

80

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

Fig. 8. FTIR spectra of a-pinene SOA formed from ozonolysis and photooxidation at the atmospheric aging times of 0.5 and 3 d with and without NH3.

(2854e2925 cm1) and C¼O stretching peaks (1710 cm1) were detected in a kinetic study of the heterogeneous and gas phase reaction of ozone-oleic acid and ozone-phenothrin using a flow

reactor coupled with ATR-FTIR spectrometer (Roberts et al., 2016). C-H stretching (2850e2920 cm1) and C¼O stretching peaks (1720 cm1) were detected for atmospheric aerosols over the Caribbean (Maria et al., 2002) and in the vicinity of steel industries in Australia (Mohiuddin et al., 2016). In addition, specific peaks for nitrogen-containing organic groups are also well documented: N-H stretching of secondary amine (3310e3360 cm1), C-N-H bending of secondary amine (1550e1650 cm1), C-N stretching of secondary amine (1130e1190 cm1), eNO2 symmetric stretching of aliphatic nitro compounds (1350e1380 cm1), and organonitrates (1270e1285 cm1) (Coates, 2000; Socrates, 2004). The N-H stretching (3300e3500 cm1) and C-N-H bending (1600 cm1) of secondary amine were observed in the FTIR spectra for NH3 aged apinene SOA, ambient aerosols, and amines for CO2 capture (Bacsik and Hedin, 2016; Laskin et al., 2014; Mohiuddin et al., 2016). For biogenic SOA and ambient aerosols, the peak of eNO2 symmetric stretching for organonitrates occurred at 1280 cm1 (Bruns et al., 2010; Day et al., 2010; Liu et al., 2012). Fig. 8 shows the ATR-FTIR spectra of the a-pinene SOA samples from the dark ozonolysis and photooxidation with and without NH3. FTIR analysis provides an evidence of nitrogen-containing

Fig. 9. (a) UV-vis absorption spectra, (b) emission spectra for excitation at 350 nm, (c) EEM for ozonolysis with NH3, (d) EEM for 0.5 d photooxidation with NH3, (e) EEM for 3 d photooxidation with NH3, (f) EEM for ozonolysis without NH3, (g) EEM for 0.5 d photooxidation without NH3, (h) EEM for 3 d photooxidation without NH3. Strong diagonal signals in the EEM are Rayleigh scattering lines.

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

functional groups in SOA formed in the presence of NH3. For the apinene SOA from the 3 d photooxidation without NH3, strong peaks of O-H stretching (3404 cm1), C-H stretching (2921e2883 cm1), C¼O stretching (1715 cm1), and C-O-H in-plane bending (1374 cm1) were detected with weak peaks for C-O stretching of tertiary alcohol (1204 cm1 and 1170 cm1). For a-pinene SOA from the 3 d photooxidation with NH3, strong N-H stretching of secondary amine (3258 cm1), C-N-H bending of secondary amine (1590 cm1), C-N stretching of secondary amine (1100 cm1), and eNO2 symmetric stretching of aliphatic nitro compounds and organonitrates (1320 cm1 and 1280 cm1) were detected in addition to the strong peaks for O-H stretching (3410 cm1), C-H stretching (2958e2880 cm1), C¼O stretching (1708 cm1), and CO-H in-plane bending of primary or secondary alcohol (1320 cm1) and weak peaks for C-O stretch of tertiary alcohol (1180 cm1 and 1230 cm1). As presented in Fig. 8 the specific peaks of nitrogencontaining functional groups were commonly found in a-pinene SOA formed under 0.5 d photooxidation and ozonolysis in the presence of NH3. Absorbance ratios of N-H stretching to O-H stretching were compared to examine the effect of NH3 on the composition of apinene SOA. The ratios for ozonolysis and 0.5 d and 3 d photooxidation in the presence of NH3 were 2.06, 1.25, and 1.68, which were greater than 1.05, 0.81, and 1.15 in the absence of NH3, respectively. This indicates the common and substantial formation of N-containing a-pinene SOA species in the presence of NH3. 3.8. Optical properties of SOA There was no change in the color appearance of a-pinene SOAloaded filter samples with and without NH3 regardless of ozonolysis or photooxidation. This qualitative observation is in agreement with previous studies (Bones et al., 2010; Lee et al., 2013; Updyke et al., 2012). However, the degree of browning was quantified in cm2 g1 by the mass absorption coefficient (MAC) and DMAC at specific wavelengths (l) (Bones et al., 2010; Chen and Bond, 2009; Nguyen et al., 2012). The UV-vis absorption spectra of a-pinene SOAs for dark ozonolysis and photooxidations are shown in Fig. 9(a). Based on the spectra, MAC (cm2 g1) at a specific wavelength was determined as:

MAC ¼

81

A CL

(7)

where A is the absorbance, C is the mass concentration (g cm3), and L is the path length (i.e., 1 cm). In the present study, DMAC is the difference in MAC values with and without NH3. Table 3 shows the MAC and DMAC values at 365 nm and 500 nm for a-pinene SOA for photooxidation and ozonolysis. MAC values of organic aerosols measured at or near 365 nm and 500 nm during various field studies are also summarized in Table 3. The a-pinene SOA samples in the presence of NH3 showed higher MAC365 and MAC500 values than those in the absence of NH3 for both ozonolysis and photooxidation. In presence of NH3, DMAC365 and DMAC500 for 3 d photooxidation were higher than those for 0.5 d photooxidation. In the case of ozonolysis, DMAC365 of 134 cm2 g1 in the presence of NH3 was higher than DMAC500 of 12 cm2 g1 in the absence of NH3. DMAC500 of 56 cm2 g1 for a-pinene SOA from the 3 d photooxidation was comparable to 50 cm2 g1 for a-pinene SOA formed by OH-initiated oxidation (Updyke et al., 2012). DMAC365 of 209 cm2 g1 and 352 cm2 g1 for a-pinene SOA from photooxidations in the present study were greater than 200 cm2 g1 for apinene SOA formed under high NOx condition (Liu et al., 2016). MAC500 values for NH3 aged SOA in this study were in the lower range of MAC500 values (i.e., 10e1000 cm2 g1) of various biogenic and anthropogenic SOA aged with NH3 (Updyke et al., 2012). Ambient organic aerosols showed MAC values at or near 365 nm and 500 nm up to two orders higher relative to the MAC of a-pinene SOA in this study. It might be largely due to nitrogen containing brown carbon such as organonitrates and amines (Bones et al., 2010; Chen et al., 2016; Updyke et al., 2012; Zhang et al., 2013). The emission spectra with normalized intensity to the mass of a-pinene SOAs from single wavelength excitation at 360 nm are shown in Fig. 9(b). The figure shows the maximum intensities of all fluorescence spectra in the following order: 3 d photooxidation w/ NH3 > 0.5 d photooxidation w/ NH3 > ozonolysis w/ NH3 > 0.5 d photooxidation w/o NH3 > 3 d photooxidation w/o NH3 > ozonolysis w/o NH3. The normalized three-dimensional excitation-emission matrix (EEM) for ozonolysis and photooxidation is shown in Fig. 9(c)-(h). Negligible fluorescence was detected

Table 3 Summary of the mass absorption coefficients (MAC365 and MAC500) of a-pinene SOA at the wavelengths of 365 and 500 nm and the change in MACs (DMAC365 and DMAC500) by NH3 aging and ambient organic aerosols in field studies. Reaction/field study O3 w/o NH3 O3 w/ NH3 OH 0.5 d w/o NH3 OH 0.5 d w/ NH3 OH 3 d w/o NH3 OH 3 d w/ NH3 O3 w/ NH3 OH w/ NH3 NOx w/o NH3 NOx w/ NH3 CalNex 2010 GoPoEx Beijing 2014 ACE Asia SAFARI 2010 EAST AIRE INTEX/ICARTT a b c d

MAC365 (cm2 g1)

DMAC365

MAC500 (cm2 g1)

DMAC500

References

(cm2 g1)

46 180 332 541 228 580 e e 0 200 3550a 4750a,b 2550 a,c e e e e

e 134 e 209 e 352 e e e 200 e e e e e e e

11 23 30 35 30 42 e e e e e e e 400 (550)d 9000 (500)d 6000 (520)d 6000 (470)d

e 12 e 5 e 12 50 40 e e e e e e e e e

This study This study This study This study This study This study Updyke et al. (2012) Updyke et al. (2012) Liu et al. (2016) Liu et al. (2016) Zhang et al. (2013) Kirillova et al. (2014) Du et al. (2014) Alexander et al. (2008) Kirchstetter et al. (2004) Yang et al. (2008) Clarke et al. (2007)

MAC of WSOC normalized to the mass of organic matter assuming OM/OC ratio of 2. Average MAC value of Northern China. For winter season. Numbers in the parentheses represent wavelength.

(cm2 g1)

82

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

in a-pinene SOA in the absence of NH3 regardless of ozonolysis or photooxidation. For a-pinene SOA from photooxidation, strong fluorescence was detected at the excitation wavelength of 300e475 nm as a result of the aging effect of gas-phase NH3. Additionally, weak fluorescence was observed at the excitation of 300 nm for a-pinene SOA from photooxidation. In the case of apinene SOA from ozonolysis, strong fluorescence was detected at the excitation range of 375e450 nm in the presence of NH3. 4. Conclusions The detailed characterization of a newly-developed flow reactor reveals its applicability for studying SOA formation and chemistry. The flow reactor yielded narrow RTD for inert gas (i.e., CO2). In particular, RH and temperature could be controlled effectively between 0 and 70% and 22e43  C, respectively. The OH radical exposure range was from 8.2  1010 to 7.2  1011 molecules cm3 s, corresponding to 0.5e4.9 d of atmospheric aging. The application of the flow reactor for the study of a-pinene SOA formation by dark ozonolysis and photooxidation demonstrated that the SOA yield by photooxidation was approximately six times higher than by ozonolysis at the same temperature, RH, and ROG concentrations. The reasons of the low SOA yield might be the short residence time and low O3 concentration in a-pinene ozonolysis in the flow reactor. Under low, medium, and high RH conditions, the presence of NH3 in the a-pinene ozonolysis and photooxidation revealed higher SOA yields than in the absence of NH3. At medium RH (~50%), the SOA yield for ozonolysis and photooxidation increased by 23% and 15% in the presence of NH3 compared to the values in the absence of NH3. FTIR analysis confirmed the presence of secondary amines in the SOA samples of ozonolysis and photooxidation in the presence of NH3. SOA from a-pinene ozonolysis and photooxidation in the presence of NH3 showed enhanced absorption and fluorescence in UV-vis radiation. SOA samples showed higher values of MAC365 and MAC500 in the presence of NH3 than in absence of NH3. SOA from ozonolysis and photooxidation in the presence of NH3 showed strong fluorescence in the excitation range of 375e450 nm and 300e475 nm respectively, compared to the negligible fluorescence in the absence of NH3. Acknowledgement This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP), which granted financial resources from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20142010201810) through POSCO E & C. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atmosenv.2017.05.034. References Alexander, D.T.L., Crozier, P.A., Anderson, J.R., 2008. Brown carbon spheres in East Asian outflow and their optical properties. Science 321, 833e836. http:// dx.doi.org/10.1126/science.1155296.
r, A., 2006. Black carbon or brown carbon? The nature of Andreae, M.O., Gelencse light-absorbing carbonaceous aerosols. Atmos. Chem. Phys. Discuss. 6, 3419e3463. http://dx.doi.org/10.5194/acpd-6-3419-2006. Atkinson, R., Aschmann, S.M., 1989. Rate constants for the gas-phase reactions of the OH radical with series of aromatic hydrocarbons at 296±2 K. I Int. J. Chem. Kinet. 21, 355e365. http://dx.doi.org/10.1002/kin.550210506. Babar, Z. Bin, Park, J.-H., Kang, J., Lim, H.-J., 2016. Characterization of a smog chamber for studying formation and physicochemical properties of secondary organic aerosol. Aerosol Air Qual. Res. 16, 1e12. http://dx.doi.org/10.4209/ aaqr.2015.10.0580. Bacsik, Z., Hedin, N., 2016. Effects of carbon dioxide captured from ambient air on

the infrared spectra of supported amines. Vib. Spectrosc. 87, 215e221. http:// dx.doi.org/10.1016/j.vibspec.2016.10.006. Beelen, R., Hoek, G., van den Brandt, P.A., Goldbohm, R.A., Fischer, P., Schouten, L.J., Jerrett, M., Hughes, E., Armstrong, B., Brunekreef, B., 2007. Long-Term effects of traffic-related air pollution on mortality in a Dutch Cohort (NLCS-AIR study). Environ. Health Perspect. 116, 196e202. http://dx.doi.org/10.1289/ehp.10767. Belalcazar, L.C., Fuhrer, O., Ho, M.D., Zarate, E., Clappier, A., 2009. Estimation of road traffic emission factors from a long term tracer study. Atmos. Environ. 43, 5830e5837. http://dx.doi.org/10.1016/j.atmosenv.2009.07.059. Bones, D.L., Henricksen, D.K., Mang, S.A., Gonsior, M., Bateman, A.P., Nguyen, T.B., Cooper, W.J., Nizkorodov, S.a., 2010. Appearance of strong absorbers and fluorophores in limonene-O3 secondary organic aerosol due to NHþ 4 -mediated chemical aging over long time scales. J. Geophys. Res. Atmos. 115, D05203. http://dx.doi.org/10.1029/2009jd012864. Bonn, B., Schuster, G., Moortgat, G.K., 2002. Influence of water va- por on the process of new particle formation during monoterpene ozonolysis. J. Phys. Chem. A 106, 2869e2881. Brauer, M., Amann, M., Burnett, R.T., Cohen, A., Dentener, F., Ezzati, M., Henderson, S.B., Krzyzanowski, M., Martin, R.V., Van Dingenen, R., van Donkelaar, A., Thurston, G.D., 2012. Exposure assessment for estimation of the global burden of disease attributable to outdoor air pollution. Environ. Sci. Technol. 46, 652e660. http://dx.doi.org/10.1021/es2025752. Bruns, E.A., Perraud, V., Zelenyuk, A., Ezell, M.J., Johnson, S.N., Yu, Y., Imre, D., Finlayson-Pitts, B.J., Alexander, M.L., 2010. Comparison of FTIR and particle mass spectrometry for the measurement of particulate organic nitrates. Environ. Sci. Technol. 44, 1056e1061. http://dx.doi.org/10.1021/es9029864. Bruns, E.a., El Haddad, I., Keller, A., Klein, F., Kumar, N.K., Pieber, S.M., Corbin, J.C.,
vo ^t, A.S.H., 2015. Inter-comparSlowik, J.G., Brune, W.H., Baltensperger, U., Pre ison of laboratory smog chamber and flow reactor systems on organic aerosol yield and composition. Atmos. Meas. Tech. 8, 2315e2332. http://dx.doi.org/ 10.5194/amt-8-2315-2015. Chen, Y., Bond, T.C., 2009. Light absorption by organic carbon from wood combustion. Atmos. Chem. Phys. Discuss. 9, 20471e20513. http://dx.doi.org/ 10.5194/acpd-9-20471-2009. Chen, Q., Ikemori, F., Mochida, M., 2016. Light absorption and excitation-emission fluorescence of urban organic aerosol components and their relationship to chemical structure. Environ. Sci. Technol. 50, 10859e10868. http://dx.doi.org/ 10.1021/acs.est.6b02541. Chhabra, P.S., Lambe, A.T., Canagaratna, M.R., Stark, H., Jayne, J.T., Onasch, T.B., Davidovits, P., Kimmel, J.R., Worsnop, D.R., 2015. Application of high-resolution time-of-flight chemical ionization mass spectrometry measurements to estimate volatility distributions of a-pinene and naphthalene oxidation products. Atmos. Meas. Tech. 8, 1e18. http://dx.doi.org/10.5194/amt-8-1-2015. Chung, S.H., Seinfeld, J.H., 2002. Global distribution and climate forcing of carbonaceous aerosols. J. Geophys. Res. 107 http://dx.doi.org/10.1029/2001JD001397, 14e1-14-33. Clarke, A., McNaughton, C., Kapustin, V., Shinozuka, Y., Howell, S., Dibb, J., Zhou, J., Anderson, B.E., Brekhovskikh, V., Turner, H., Pinkerton, M., 2007. Biomass burning and pollution aerosol over North America: organic components and their influence on spectral optical properties and humidification response. J. Geophys. Res. Atmos. 112, 1e13. http://dx.doi.org/10.1029/2006JD007777. Coates, J., 2000. Interpretation of infrared spectra, a practical approach interpretation of infrared spectra. A Pract. Approach. Encycl. Anal. Chem. 10815e10837. http://dx.doi.org/10.1002/9780470027318. Cocker III, D.R., Clegg, S.L., Flagan, R.C., Seinfeld, J.H., 2001a. The effect of water on gas-particle partitioning of secondary organic aerosol. Part I: a-pinene/ozone system. Atmos. Environ. 35, 6049e6072. http://dx.doi.org/10.1016/S13522310(01)00404-6. Cocker III, D.R., Flagan, R.C., Seinfeld, J.H., 2001b. State-of-the-art chamber facility for studying atmospheric aerosol chemistry. Environ. Sci. Technol. 35, 2594e2601. http://dx.doi.org/10.1021/es0019169. Davidson, C.I., Phalen, R.F., Solomon, P.A., 2005. Airborne particulate matter and human health: a review. Aerosol Sci. Technol. 39, 737e749. http://dx.doi.org/ 10.1080/02786820500191348. Day, D.A., Liu, S., Russell, L.M., Ziemann, P.J., 2010. Organonitrate group concentrations in submicron particles with high nitrate and organic fractions in coastal southern California. Atmos. Environ. 44, 1970e1979. http://dx.doi.org/10.1016/ j.atmosenv.2010.02.045. de Gouw, J., Jimenez, J.L., 2009. Organic aerosols in the Earth's atmosphere. Environ. Sci. Technol. 43, 7614e7618. http://dx.doi.org/10.1021/es9006004. Du, Z., He, K., Cheng, Y., Duan, F., Ma, Y., Liu, J., Zhang, X., Zheng, M., Weber, R., 2014. A yearlong study of water-soluble organic carbon in Beijing II: light absorption properties. Atmos. Environ. 89, 235e241. http://dx.doi.org/10.1016/ j.atmosenv.2014.02.022. EPA, 2012. Our Nation's Air - Status and Trends through 2010. US Environmental Protection Aency, Office of Air Quality Planning and Standards, Research Triangle Park, Noth Carolina. Ezell, M.J., Johnson, S.N., Yu, Y., Perraud, V., Bruns, E.A., Alexander, M.L., Zelenyuk, A., Dabdub, D., Finlayson-Pitts, B.J., 2010. A new aerosol flow system for photochemical and thermal studies of Tropospheric aerosols. Aerosol Sci. Technol. 44, 329e338. http://dx.doi.org/10.1080/02786821003639700. Fick, J., Pommer, L., Nilsson, C., Andersson, B., 2003. Effect of OH radicals, relative humidity, and time on the composition of the products formed in the ozonolysis of a-pinene. Atmos. Environ. 37, 4087e4096. http://dx.doi.org/10.1016/ S1352-2310(03)00522-3.

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84 Galloway, M.M., Chhabra, P.S., Chan, A.W.H., Surratt, J.D., Flagan, R.C., Seinfeld, J.H., Keutsch, F.N., 2008. Glyoxal uptake on ammonium sulphate seed aerosol: reaction products and reversibility of uptake under dark and irradiated conditions. Atmos. Chem. Phys. Discuss. 8, 20799e20838. http://dx.doi.org/10.5194/ acpd-8-20799-2008. Gao, Y., Hall, W.A., Johnston, M.V., 2010. Molecular composition of monoterpene secondary organic aerosol at low mass loading. Environ. Sci. Technol. 44, 7897e7902. http://dx.doi.org/10.1021/es101861k. George, I.J., Vlasenko, A., Slowik, J.G., Abbatt, J.P.D., 2007. Heterogeneous oxidation of saturated organic aerosols by hydroxyl radicals: uptake kinetics and condensed-phase products. Atmos. Chem. Phys. Discuss. 7, 6803e6842. http:// dx.doi.org/10.5194/acpd-7-6803-2007. Graf, H.F., Langmann, B., Feichter, J., 1998. The contribution of Earth degassing to the atmospheric sulfur budget. Chem. Geol. 147, 131e145. http://dx.doi.org/10.1016/ S0009-2541(97)00177-0. Griffin, R.J., Cocker III, D.R., Flagan, R.C., Seinfeld, J.H., 1999. Organic aerosol formation from the oxidation of biogenic hydrocarbons. J. Geophys. Res. http:// dx.doi.org/10.1029/1998jd100049. Guzman-Morales, J., Frossard, A.A., Corrigan, A.L., Russell, L.M., Liu, S., Takahama, S., Taylor, J.W., Allan, J., Coe, H., Zhao, Y., Goldstein, A.H., 2014. Estimated contributions of primary and secondary organic aerosol from fossil fuel combustion during the CalNex and Cal-Mex campaigns. Atmos. Environ. 88, 330e340. http://dx.doi.org/10.1016/j.atmosenv.2013.08.047. Hall IV, W.A., Pennington, M.R., Johnston, M.V., 2013. Molecular transformations accompanying the aging of laboratory secondary organic aerosol. Environ. Sci. Technol. 47, 2230e2237. http://dx.doi.org/10.1021/es303891q. Hallquist, M., Wenger, J.C., Baltensperger, U., Rudich, Y., Simpson, D., Claeys, M., Dommen, J., Donahue, N.M., George, C., Goldstein, A.H., Hamilton, J.F., Herrmann, H., Hoffmann, T., Iinuma, Y., Jang, M., Jenkin, M.E., Jimenez, J.L., Kiendler-Scharr, A., Maenhaut, W., McFiggans, G., Mentel, T.F., Monod, A.,
vo ^t, A.S.H., Seinfeld, J.H., Surratt, J.D., Szmigielski, R., Wildt, J., 2009. The Pre formation, properties and impact of secondary organic aerosol: current and emerging issues. Atmos. Chem. Phys. 9, 5155e5236. http://dx.doi.org/10.5194/ acp-9-5155-2009. Harrison, R.M., Yin, J., 2000. Particulate matter in the atmosphere: which particle properties are important for its effects on health? Sci. Total Environ. 249, 85e101. http://dx.doi.org/10.1016/S0048-9697(99)00513-6. Heald, C.L., Kroll, J.H., Jimenez, J.L., Docherty, K.S., DeCarlo, P.F., Aiken, A.C., Chen, Q., Martin, S.T., Farmer, D.K., Artaxo, P., 2010. A simplified description of the evolution of organic aerosol composition in the atmosphere. Geophys. Res. Lett. 37, L08803. http://dx.doi.org/10.1029/2010gl042737. Hoffmann, T., Odum, J.R., Bowman, F., Collins, D., Klockow, D., Flagan, R.C., Seinfeld, J.H., 1997. Formation of organic aerosols from the oxidation of biogenic hydrocarbons. J. Atmos. Chem. 26, 189e222. http://dx.doi.org/10.1023/A: 1005734301837. Hu, C., Cheng, Y., Pan, G., Gai, Y., Gu, X., Zhao, W., Wang, Z., Zhang, W., Chen, J., Liu, F., Shan, X., Sheng, L., 2014. A smog chamber facility for qualitative and quantitative study on atmospheric chemistry and secondary organic aerosol. Chin. J. Chem. Phys. 27, 631e639. http://dx.doi.org/10.1063/1674-0068/27/06/631-639. Intergovernmental Panel on Climate Change (IPCC), 2001. Working Group I, Climate Change: the Scientific Basis. Cambridge Univ. Press, New York. Intergovernmental Panel on Climate Change (IPCC), 2013. Climate Change 2013: the Physical Science Basis. Cambridge Univ. Press, New York. €m, E., 2006. Impact of humidity on the ozone Jonsson, Å.M., Hallquist, M., Ljungstro initiated oxidation of limonene, D3-carene, and a-pinene. Environ. Sci. Technol. 40, 188e194. http://dx.doi.org/10.1021/es051163w. Kampf, C.J., Jakob, R., Hoffmann, T., 2012. Identification and characterization of aging products in the glyoxal/ammonium sulfate system- implications for lightabsorbing material in atmospheric aerosols. Atmos. Chem. Phys. 12, 6323e6333. http://dx.doi.org/10.5194/acp-12-6323-2012. Kanakidou, M., Seinfeld, J.H., Pandis, S.N., Barnes, I., Dentener, F.J., Facchini, M.C., Van Dingenen, R., Ervens, B., Nenes, A., Nielsen, C.J., Swietlicki, E., Putaud, J.P., Balkanski, Y., Fuzzi, S., Horth, J., Moortgat, G.K., Winterhalter, R., Myhre, C.E.L., Tsigaridis, K., Vignati, E., Stephanou, E.G., Wilson, J., 2005. Organic aerosol and global climate modelling: a review. Atmos. Chem. Phys. 5, 1053e1123. http:// dx.doi.org/10.5194/acp-5-1053-2005. Kang, E., Root, M.J., Brune, W.H., 2007. Introducing the concept of potential aerosol mass (PAM). Atmos. Chem. Phys. Discuss. 7, 9925e9972. http://dx.doi.org/ 10.5194/acpd-7-9925-2007. Kang, E., Toohey, D.W., Brune, W.H., 2011. Dependence of SOA oxidation on organic aerosol mass concentration and OH exposure: experimental PAM chamber studies. Atmos. Chem. Phys. 11, 1837e1852. http://dx.doi.org/10.5194/acp-111837-2011. Keller, A., Burtscher, H., 2012. A continuous photo-oxidation flow reactor for a defined measurement of the SOA formation potential of wood burning emissions. J. Aerosol Sci. 49, 9e20. http://dx.doi.org/10.1016/j.jaerosci.2012.02.007. Khalizov, A.F., Earle, M.E., Johnson, W.J.W., Stubley, G.D., Sloan, J.J., 2006. Development and characterization of a laminar aerosol flow tube. Rev. Sci. Instrum. 77, 1e9. http://dx.doi.org/10.1063/1.2175958. Kirchstetter, T.W., Novakov, T., Hobbs, P.V., 2004. Evidence that the spectral dependence of light absorption by aerosols is affected by organic carbon. J. Geophys. Res. D. Atmos. 109, 1e12. http://dx.doi.org/10.1029/2004JD004999. € 2014. Sources and light Kirillova, E.N., Andersson, A., Han, J., Lee, M., Gustafsson, O., absorption of water-soluble organic carbon aerosols in the outflow from northern China. Atmos. Chem. Phys. 14, 1413e1422. http://dx.doi.org/10.5194/

83

acp-14-1413-2014. Lall, R., Kendall, M., Ito, K., Thurston, G.D., 2004. Estimation of historical annual PM 2.5 exposures for health effects assessment. Atmos. Environ. 38, 5217e5226. http://dx.doi.org/10.1016/j.atmosenv.2004.01.053. Lambe, A.T., Ahern, A.T., Williams, L.R., Slowik, J.G., Wong, J.P.S., Abbatt, J.P.D., Brune, W.H., Ng, N.L., Wright, J.P., Croasdale, D.R., Worsnop, D.R., Davidovits, P., Onasch, T.B., 2011. Characterization of aerosol photooxidation flow reactors: heterogeneous oxidation, secondary organic aerosol formation and cloud condensation nuclei activity measurements. Atmos. Meas. Tech. 4, 445e461. http://dx.doi.org/10.5194/amt-4-445-2011. Lambe, A.T., Chhabra, P.S., Onasch, T.B., Brune, W.H., Hunter, J.F., Kroll, J.H., Cummings, M.J., Brogan, J.F., Parmar, Y., Worsnop, D.R., Kolb, C.E., Davidovits, P., 2015. Effect of oxidant concentration, exposure time, and seed particles on secondary organic aerosol chemical composition and yield. Atmos. Chem. Phys. 15, 3063e3075. http://dx.doi.org/10.5194/acp-15-3063-2015. Laskin, J., Laskin, A., Nizkorodov, S.A., Roach, P., Eckert, P., Gilles, M.K., Wang, B., Lee, H.J., Hu, Q., 2014. Molecular selectivity of brown carbon chromophores. Environ. Sci. Technol. 48, 12047e12055. http://dx.doi.org/10.1021/es503432r. Lee, H.J., Laskin, A., Laskin, J., Nizkorodov, S.A., 2013. Excitation-emission spectra and fluorescence quantum yields for fresh and aged biogenic secondary organic aerosols. Environ. Sci. Technol. 47, 5763e5770. http://dx.doi.org/10.1021/ es400644c. Liu, S., Shilling, J.E., Song, C., Hiranuma, N., Zaveri, R.A., Russell, L.M., 2012. Hydrolysis of organonitrate functional groups in aerosol particles. Aerosol Sci. Technol. 46, 1359e1369. http://dx.doi.org/10.1080/02786826.2012.716175. Liu, Y., Liggio, J., Staebler, R., Li, S.M., 2015. Reactive uptake of ammonia to secondary organic aerosols: kinetics of organonitrogen formation. Atmos. Chem. Phys. Discuss. 15, 17449e17490. http://dx.doi.org/10.5194/acpd-15-17449-2015. Liu, J., Lin, P., Laskin, A., Laskin, J., Kathmann, S.M., Wise, M., Caylor, R., Imholt, F., Selimovic, V., Shilling, J.E., 2016. Optical properties and aging of light-absorbing secondary organic aerosol. Atmos. Chem. Phys. 16, 12815e12827. http:// dx.doi.org/10.5194/acp-16-12815-2016. Mao, J., Ren, X., Brune, W.H., Olson, J.R., Crawford, J.H., Fried, A., Huey, L.G., Cohen, R.C., Heikes, B., Singh, H.B., Blake, D.R., Sachse, G.W., Diskin, G.S., Hall, S.R., Shetter, R.E., 2009. Airborne measurement of OH reactivity during INTEX-B. Atmos. Chem. Phys. Discuss. 9, 163e173. http://dx.doi.org/10.5194/ acpd-8-14217-2008. Maria, S.F., Russell, L.M., Turpin, B.J., Porcja, R.J., 2002. FTIR measurements of functional groups and organic mass in aerosol samples over the Caribbean. Atmos. Environ. 36, 5185e5196. http://dx.doi.org/10.1016/S1352-2310(02) 00654-4. Matsunaga, A., Ziemann, P.J., 2010. Gas-wall partitioning of organic compounds in a Teflon film chamber and potential effects on reaction product and aerosol yield measurements. Aerosol Sci. Technol. 44, 881e892. http://dx.doi.org/10.1080/ 02786826.2010.501044. Mohiuddin, K., Strezov, V., Nelson, P.F., Evans, T., 2016. Bonding structure and mineral analysis of size resolved atmospheric particles nearby steelmaking industrial sites in Australia. Aerosol Air Qual. Res. 16, 1638e1650. http:// dx.doi.org/10.4209/aaqr.2015.02.0076. Na, K., Song, C., Switzer, C., Cocker III, D.R., 2007. Effect of Ammonia on Secondary Organic Aerosol formation from a-Pinene ozonolysis in dry and humid conditions. Environ. Sci. Technol. 41, 6096e6102. http://dx.doi.org/10.1002/ star.19900420204. Ng, N.L., Chhabra, P.S., Chan, A.W.H., Surratt, J.D., Kroll, J.H., Kwan, A.J., McCabe, D.C., Wennberg, P.O., Sorooshian, A., Murphy, S.M., Dalleska, N.F., Flagan, R.C., Seinfeld, J.H., 2007. Effect of NOx level on secondary organic aerosol (SOA) formation from the photooxidation of terpenes. Atmos. Chem. Phys. 7, 5159e5174. http://dx.doi.org/10.5194/acp-7-5159-2007, 2007a. Ng, N.L., Canagaratna, M.R., Zhang, Q., Jimenez, J.L., Tian, J., Ulbrich, I.M., Kroll, J.H., Docherty, K.S., Chhabra, P.S., Bahreini, R., Murphy, S.M., Seinfeld, J.H.,
v o ^ t, A.S.H., Dinar, E., Hildebrandt, L., Donahue, N.M., DeCarlo, P.F., Lanz, V.a., Pre Rudich, Y., Worsnop, D.R., Prevot, A.S.H., Dinar, E., Rudich, Y., Worsnop, D.R., 2010. Organic aerosol components observed in northern hemispheric datasets from aerosol mass spectrometry. Atmos. Chem. Phys. 10, 4625e4641. http:// dx.doi.org/10.5194/acp-10-4625-2010. Nguyen, T.B., Lee, P.B., Updyke, K.M., Bones, D.L., Laskin, J., Laskin, A., Nizkorodov, S.A., 2012. Formation of nitrogen- and sulfur-containing lightabsorbing compounds accelerated by evaporation of water from secondary organic aerosols. J. Geophys. Res. 117, D01207. http://dx.doi.org/10.1029/ 2011JD016944. re, B., Dziedzic, P., Co
rdova, A., 2009. Products and kinetics of the liquid-phase Nozie reaction of glyoxal catalyzed by ammonium ions (NH4(þ)). J. Phys. Chem. A 113, 231e237. http://dx.doi.org/10.1021/jp8078293. Odum Jay, R., Hoffmann, T., Bowman, F., Collins, D., Flagan Richard, C., Seinfeld John, H., 1996. Gas particle partitioning and secondary organic aerosol yields. Environ. Sci. Technol. 30, 2580e2585. http://dx.doi.org/10.1021/es950943. Offenberg, J.H., Kleindienst, T.E., Jaoui, M., Lewandowski, M., Edney, E.O., 2006. Thermal properties of secondary organic aerosols. Geophys. Res. Lett. 33, L03816. http://dx.doi.org/10.1029/2005GL024623, 2006. Ortega, A.M., Day, D.A., Cubison, M.J., Brune, W.H., Bon, D., de Gouw, J.A., Jimenez, J.L., 2013. Secondary organic aerosol formation and primary organic aerosol oxidation from biomass-burning smoke in a flow reactor during FLAME3. Atmos. Chem. Phys. 13, 11551e11571. http://dx.doi.org/10.5194/acp-13-115512013. Paciga, A., Riipinen, I., Pandis, S., 2014. Effect of inorganic salts on the volatility of

84

Z.B. Babar et al. / Atmospheric Environment 164 (2017) 71e84

organic acids. Environ. Sci. Technol. 48, 13769e13775. http://dx.doi.org/10.1021/ es5037805.
vo ^t, A.S.H., Richter, R., Sax, M., Paulsen, D., Dommen, J., Kalberer, M., Pre Steinbacher, M., Weingartner, E., Baltensperger, U., 2005. Secondary organic aerosol formation by irradiation of 1,3,5-trimethylbenzene-NOx-H2O in a new reaction chamber for atmospheric chemistry and physics. Environ. Sci. Technol. 39, 2668e2678. http://dx.doi.org/10.1021/es0489137. Platt, S.M., El Haddad, I., Zardini, A.A., Clairotte, M., Astorga, C., Wolf, R., Slowik, J.G., €hler, O., Temime-Roussel, B., Marchand, N., Je zek, I., Drinovec, L., Mo cnik, G., Mo
vo ^t, A.S.H., 2013. SecRichter, R., Barmet, P., Bianchi, F., Baltensperger, U., Pre ondary organic aerosol formation from gasoline vehicle emissions in a new mobile environmental reaction chamber. Atmos. Chem. Phys. 13, 9141e9158. http://dx.doi.org/10.5194/acp-13-9141-2013. Presto, A.A., Nguyen, N.T., Ranjan, M., Reeder, A.J., Lipsky, E.M., Hennigan, C.J., Miracolo, M.A., Riemer, D.D., Robinson, A.L., 2011. Fine particle and organic vapor emissions from staged tests of an in-use aircraft engine. Atmos. Environ. 45, 3603e3612. http://dx.doi.org/10.1016/j.atmosenv.2011.03.061. Putaud, J.P., Van Dingenen, R., Alastuey, A., Bauer, H., Birmili, W., Cyrys, J., Flentje, H., Fuzzi, S., Gehrig, R., Hansson, H.C., Harrison, R.M., Herrmann, H., Hitzenberger, R., Hüglin, C., Jones, A.M., Kasper-Giebl, A., Kiss, G., Kousa, A., €schau, G., Maenhaut, W., Molnar, A., Moreno, T., Kuhlbusch, T.A.J., Lo Pekkanen, J., Perrino, C., Pitz, M., Puxbaum, H., Querol, X., Rodriguez, S., Salma, I., Schwarz, J., Smolik, J., Schneider, J., Spindler, G., ten Brink, H., Tursic, J., Viana, M., Wiedensohler, A., Raes, F., 2010. A European aerosol phenomenology - 3: physical and chemical characteristics of particulate matter from 60 rural, urban, and kerbside sites across Europe. Atmos. Environ. 44, 1308e1320. http:// dx.doi.org/10.1016/j.atmosenv.2009.12.011. Raff, J.D., 2007. Transport of Atmospheric Pollutants and Their Atmospheric Fates. Indiana University, Bloomington, IN, USA. Ph.D. Dissertation. Roberts, J.E., Zeng, G., Maron, M.K., Mach, M., Dwebi, I., Liu, Y., 2016. Measuring heterogeneous reaction rates with ATR-FTIR spectroscopy to evaluate chemical fates in an atmospheric environment: a physical chemistry and environmental chemistry laboratory experiment. J. Chem. Educ. 93, 733e737. http://dx.doi.org/ 10.1021/acs.jchemed.5b00290. Rohr, A.C., Weschler, C.J., Koutrakis, P., Spengler, J.D., 2003. Generation and quantification of ultrafine particles through Terpene/ozone reaction in a chamber setting. Aerosol Sci. Technol. 37, 65e78. http://dx.doi.org/10.1080/ 02786820390112597. Rohrer, F., Bohn, B., Brauers, T., Brüning, D., Johnen, F.-J., Wahner, A., Kleffmann, J., 2004. Characterisation of the photolytic HONO-source in the atmosphere simulation chamber SAPHIR. Atmos. Chem. Phys. Discuss. 4, 7881e7915. http:// dx.doi.org/10.5194/acpd-4-7881-2004. €hler, O., Jonsson, Å.M., Hallquist, M., KiendlerSaathoff, H., Naumann, K.-H., Mo Scharr, A., Mentel, T.F., Tillmann, R., Schurath, U., 2008. Temperature dependence of yields of secondary organic aerosols from the ozonolysis of a-pinene and limonene. Atmos. Chem. Phys. Discuss. 8, 15595e15664. http://dx.doi.org/ 10.5194/acpd-8-15595-2008. Sarrafzadeh, M., Wildt, J., Pullinen, I., Springer, M., Kleist, E., Tillmann, R., Schmitt, S.H., Wu, C., Mentel1, T.F., Zhao, D., Hastie, D.R., Kiendler-Scharr, A., 2016. Impact of NOx and OH on secondary organic aerosol formation from bpinene photooxidation. Atmos. Chem. Phys. 16, 11237e11248. http://dx.doi.org/

10.5194/acp-16-11237-2016. Shapiro, E.L., Szprengiel, J., Sareen, N., Jen, C.N., Giordano, M.R., McNeill, V.F., 2009. Light-absorbing secondary organic material formed by glyoxal in aqueous aerosol mimics. Atmos. Chem. Phys. 2009 (9), 2289e2300. Slowik, J.G., Wong, J.P.S., Abbatt, J.P.D., 2012. Real-time, controlled OH-initiated oxidation of biogenic secondary organic aerosol. Atmos. Chem. Phys. 12, 9775e9790. http://dx.doi.org/10.5194/acp-12-9775-2012. Smith, J.D., Kroll, J.H., Cappa, C.D., Che, D.L., Liu, C.L., Ahmed, M., Leone, S.R., Worsnop, D.R., Wilson, K.R., 2009. The heterogeneous reaction of hydroxyl radicals with sub-micron squalane particles: a model system for understanding the oxidative aging of ambient aerosols. Atmos. Chem. Phys. 9, 3209e3222. http://dx.doi.org/10.5194/acpd-9-3945-2009. Socrates, G., 2004. Infrared and Raman Characteristic Group Frequencies: Tables and Charts, third ed. Wiley, Chichester. accessed 10.01.17. http://eu.wiley. com/WileyCDA/WileyTitle/productCd-0470093072.html. Suarez-Bertoa, R., Zardini, A.A., Platt, S.M., Hellebust, S., Pieber, S.M., El Haddad, I.,
vo ^t, A.S.H., Astorga, C., Temime-Roussel, B., Baltensperger, U., Marchand, N., Pre 2015. Primary emissions and secondary organic aerosol formation from the exhaust of a flex-fuel (ethanol) vehicle. Atmos. Environ. 117, 200e211. http:// dx.doi.org/10.1016/j.atmosenv.2015.07.006. Updyke, K.M., Nguyen, T.B., Nizkorodov, S.A., 2012. Formation of brown carbon via reactions of ammonia with secondary organic aerosols from biogenic and anthropogenic precursors. Atmos. Environ. 63, 22e31. http://dx.doi.org/ 10.1016/j.atmosenv.2012.09.012. Varutbangkul, V., Brechtel, F.J., Bahreini, R., Ng, N.L., Keywood, M.D., Kroll, J.H., Flagan, R.C., Seinfeld, J.H., Lee, A., Goldstein, A.H., 2006. Hygroscopicity of secondary organic aerosols formed by oxidation of cycloalkenes, monoterpenes, sesquiterpenes, and related compounds. Atmos. Chem. Phys. 6, 2367e2388. http://dx.doi.org/10.5194/acp-6-2367-2006. Virkkula, A., Van Dingenen, R., Raes, F., Hjorth, J., 1999. Hygroscopic properties of aerosol formed by oxidation of limonene, a-pinene, and b-pinene. Sect. Title Air Pollut. Ind. Hyg. 104, 3569e3579. http://dx.doi.org/10.1029/1998JD100017. Wang, J., Doussin, J.F., Perrier, S., Perraudin, E., Katrib, Y., Pangui, E., PicquetVarrault, B., 2011. Design of a new multi-phase experimental simulation chamber for atmospheric photosmog, aerosol and cloud chemistry research. Atmos. Meas. Tech. 4, 2465e2494. http://dx.doi.org/10.5194/amt-4-2465-2011. Wang, X., Liu, T., Bernard, F., Ding, X., Wen, S., Zhang, Y., Zhang, Z., He, Q., Lü, S., Chen, J., Saunders, S., Yu, J., 2014. Design and characterization of a smog chamber for studying gas-phase chemical mechanisms and aerosol formation. Atmos. Meas. Tech. 7, 301e313. http://dx.doi.org/10.5194/amt-7-301-2014. World Health Organization (WHO), 2013. Health Effects of Particulate Matter: Policy Implications For Countries in Eastern Europe, caucasus and Central Asia. Yang, M., Howell, S.G., Zhuang, J., Huebert, B.J., 2008. Attribution of aerosol light absorption to black carbon, brown carbon, and dust in China e interpretations of atmospheric measurements during EAST-AIRE. Atmos. Chem. Phys. Discuss. 8, 10913e10954. http://dx.doi.org/10.5194/acpd-8-10913-2008. Zhang, X., Lin, Y.H., Surratt, J.D., Weber, R.J., 2013. Sources, composition and absorption Angstrom exponent of light-absorbing organic components in aerosol extracts from the los angeles basin. Environ. Sci. Technol. 47, 3685e3693. http:// dx.doi.org/10.1021/es305047b.