INVESTIGATION OF THE INFLUENCES OF ANTHROPOGENIC EMISSIONS ON ISOPRENE-DERIVED SECONDARY ORGANIC AEROSOL FORMATION DURING THE

2013 SOUTHERN OXIDANT & AEROSOL STUDY AT THE BIRMINGHAM, AL GROUND SITE

Kevin Chu

A thesis submitted to the faculty at the University of North Carolina at Chapel Hill in partial

fulfillment of the requirements for the degree of Master of Science in the Department of Environmental Sciences and Engineering.

Chapel Hill 2015

Approved by: Jason Surratt Avram Gold William G. Vizuete

  ii    

© 2015 Kevin Chu

ALL RIGHTS RESERVED

  iii    

ABSTRACT

Kevin Chu: Investigation of the influences of anthropogenic emissions on isoprene-derived secondary organic aerosol formation during the 2013 Southern Oxidant & Aerosol Study at the

Birmingham, AL ground site (Under the direction of Jason Surratt)

Fine particulate matter (PM2.5) impacts public health, air quality and global climate. In the

southeastern U.S., emissions of isoprene from deciduous trees undergo atmospheric oxidation to

form secondary organic aerosol (SOA) that contributes to PM2.5. To examine how anthropogenic

pollutants influence isoprene-derived SOA formation, PM2.5 filter samples taken at the

Birmingham, AL ground site during the 2013 Southern Oxidant and Aerosol Study were

analyzed by gas chromatography/electron impact-mass spectrometry for known isoprene SOA

tracers. Tracers were compared with collocated measurements of anthropogenic and

meteorological variables. On average, isoprene-derived SOA contributed 7.1% (up to ~22%) of

total fine organic aerosol mass. Isoprene-SOA correlated with sulfate (r2 = 0.34) and acidity (r2 =

0.24), but not with oxides of nitrogen (NOx). Ozone (O3) correlated with MAE/HMML-derived

tracers (r2 = 0.31) and with 2-methyltetrols (r2 = 0.16), phenomena not previously observed in

field studies. These results demonstrate that anthropogenic pollutants enhance isoprene-derived

SOA formation.

  iv    

ACKNOWLEDGEMENTS

This work was funded by the US Environmental Protection Agency (EPA) through grant

number 835404. The contents of this publication are solely the responsibility of the authors and

do not necessarily represent the official views of the US EPA. Further, the US EPA does not

endorse the purchase of any commercial products or services mentioned in the publication. The

authors would also like to thank the Electric Power Research Institute (EPRI) for their support.

This study was supported in part by the National Oceanic and Atmospheric Administration

(NOAA) Climate Program Office’s AC4 program, award # NA13OAR4310064. The authors

thank Louisa Emmons for her assistance with forecasts made available during the SOAS

campaign. The authors thank John Offenberg for providing access to a Sunset Laboratory OC/EC

analysis instrument. We would like to thank Annmarie Carlton, Joost deGouw, Jose Jimenez,

and Allen Goldstein for helping to organize the SOAS campaign and coordinating

communication between ground sites.

  v    

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... vi

LIST OF FIGURES ...................................................................................................................... vii

LIST OF ABBREVIATIONS ...................................................................................................... viii

CHAPTER 1: INTRODUCTION ....................................................................................................1

CHAPTER 2: EXPERIMENTAL ....................................................................................................6

2.1 Site Description and Measurements Made During 2013 SOAS at BHM ............................. 6

2.2 Gas Chromatography/Electron Impact-Mass Spectrometry (GC/EI-MS) Analysis ........... 10

2.3. Estimations of aerosol pH ................................................................................................. 12

CHAPTER 3: RESULTS AND DISCUSSION .............................................................................13

3.1 Overview of Study .............................................................................................................. 13

3.2 Characterization of Isoprene SOA ...................................................................................... 16

3.3 Effect of Sulfate and Acidity .............................................................................................. 19

3.4 Effect of Anthropogenic Emissions .................................................................................... 21

3.5 Nighttime Nitrate Radical Production ................................................................................ 24

CHAPTER 4: CONCLUSION ......................................................................................................26

REFERENCES ..............................................................................................................................29

  vi    

LIST OF TABLES

Table 1. Sampling schedule during SOAS at the BHM ground site. .............................................. 7

Table 2. Instrumentation and operating parameters of collocated measurements. ......................... 9

Table 3. Summary of collocated measurements. .......................................................................... 14

Table 4. Summary of Isoprene-derived SOA Tracer Concentrations ........................................... 18

Table 5. Summary of isoprene-SOA tracers from the three SOAS ground sites .......................... 28

  vii    

LIST OF FIGURES Figure 1. Isoprene oxidation pathways under high-NOx (upper) and low-NOx (lower) conditions. ..................................................................................... 3 Figure 2. Map of SEARCH network locations. ........................................................................ 6 Figure 3. Total ion chromatogram (TIC) illustrating the isoprene SOA tracers ..................... 11 Figure 4. Wind rose illustrating wind direction during the campaign at the BHM site.. ........ 14 Figure 5. Time series of meteorological conditions during the campaign at BHM. ............... 15 Figure 6. Time series concentrations of trace gas species. ..................................................... 15 Figure 7. Time series mass concentrations of PM2.5 constituents ........................................... 16 Figure 8. Sulfate aerosol correlated with all three GC/EI-MS quantified tracers. .................. 20 Figure 9. NOx/NOy concentrations exhibit ~0 correlation with major SOA Tracers .............. 21 Figure 10. Ozone correlates with MAE/HMML-derived SOA tracers. .................................. 23 Figure 11. Overall nighttime NO3 radical production is mildly correlated with 2-MG abundance. ................................................................................... 25

  viii  

LIST OF ABBREVIATIONS BHM Birmingham, AL

CTR Centerville, AL

CMAQ Community Multiscale Air Quality Model

FLEX-PART Flexible Particle dispersion model

GC/EI-MS Gas chromatography/electron impact-mass spectrometry

HMML Hydroxymethyl-methyl-a-lactone

IEPOX Isoprene-derived Epoxydiol

LRK Look Rock, TN

MACR Methacrolein

MAE Methacrylic acid epoxide

MeTHF Methyl tetrahydrofuran

MEGAN Model of Emissions of Gas and Aerosols from Nature

MOZART Model for Ozone and Related Chemical Tracers

MPAN Methacryloyl peroxynitrate

OA Organic Aerosol

OM Organic Matter

OS Organosulfate

POA Primary Organic Aerosol

PM2.5 Fine Particulate Matter

SEAC4RS Studies of Emissions and Atmospheric Composition, Clouds, and Climate Coupling by Regional Surveys

SEARCH Southeastern Aerosol Research and Characterization Study

  ix  

SOA Secondary Organic Aerosol

SOAS Southern Oxidant and Aerosol Study

TIC Total ion chromatogram

UPLC/ESI-HR-QTOFMS Ultra performance liquid chromatography/electrospray ionization high-resolution quadrupole time-of-flight mass spectrometry

VOCs Volatile Organic Compounds

  1    

CHAPTER 1: INTRODUCTION  

Fine aerosol (PM2.5), suspensions of liquid or solid matter in a gaseous medium that

are ≤ 2.5 μm in diameter, play a key role in physical and chemical atmospheric processes.

They influence climate patterns both directly, through the absorption and scattering of solar

and terrestrial radiation, and indirectly, through cloud formation.1,2 In addition to climatic

effects, PM2.5 have been demonstrated to pose a potential human health risk through

inhalation exposure.3,4 Despite the strong association of PM2.5 with climate change and

environmental health, there remains a need to more fully resolve its composition, sources,

and chemical formation processes, resulting in the development of effective control strategies

to address potential hazards in a cost-effective manner.4–6

Atmospheric PM2.5 are comprised, in a large part (upwards of 90% by mass in some

locations) by organic matter (OM).4,7 OM can be derived from many sources. Primary

organic aerosol (POA) are initial emissions from natural (e.g., pollen, fungal spores,

vegetative detritis) and anthropogenic sources (e.g., fossil fuel and biomass burning) prior to

atmospheric processing. As a result of large anthropogenic sources, POA tends to be

abundant in urban areas. Processes such as biomass burning and combustion also yield

volatile organic compounds (VOCs), which exhibit high vapor pressures and volatility and

undergo atmospheric oxidation to form secondary organic aerosol (SOA) through reactive

uptake onto seed aerosol or gas-to-particle phase partitioning with subsequent particle-phase

  2    

chemical reactions. SOA, as recently demonstrated, can comprise up to 90%, by mass in

some areas, of total atmospheric PM2.5.7

At around 600 Tg emitted per year into the atmosphere, isoprene (2-methyl-1,3-

butadien, C5H8) is the most abundant volatile no n-methane hydrocarbon.8 The abundance

of isoprene is particularly high in the southeastern U.S. due to emissions from broad leaf

deciduous tree species.8 Though it was long known to play a central role in the

photochemical formation of ozone (O3),9 isoprene was traditionally thought not to form SOA

due to the high volatility of its known oxidation products.10,11 However, work over the last

decade has revealed that isoprene, via hydroxyl radical (OH)-initiated oxidation, is indeed a

major source of SOA.12–18 In addition, it is now known that SOA formation is enhanced by

anthropogenic emissions, namely oxides of nitrogen (NOx) and sulfur dioxide (SO2), that are

a source of acidic seed aerosol onto which photochemical oxidation products of isoprene are

reactively taken up to yield a variety of SOA products.14,19,20

Recent work has begun to elucidate some of the critical intermediates of isoprene

oxidation that lead to SOA formation through acid-catalyzed heterogeneous (multiphase)

chemistry.13,18 Under low-NOx conditions, such as in a pristine environment, isomeric

epoxydiols (IEPOX) have been demonstrated to be the critical to the formation of isoprene

SOA, especially in the presence of acidic sulfate aerosol.13,17,19 This pathway has been shown

to yield 2-methyltetrols as one of the major SOA constituents observed in ambient PM2.5.17,21

In addition, further work has revealed a slew of IEPOX-derived tracers, including C5-alkene

triols,17,22 cis- and trans-3-methyltetrahydrofuran-3,4-diols (3-MeTHF-3,4-diols),17,23 IEPOX-

derived organosulfates (OS),17 and IEPOX-derived oligomers.24 Some of the IEPOX-derived

oligomers have been shown to contribute to aerosol components known as brown carbon that

  3    

absorb light at the near UV and visible ranges of the electromagnetic spectrum.24 Under high-

NOx conditions, such as an urban environment, isoprene SOA formation occurs via the

oxidation of methacrolein (MACR).13,15 The OH-initiated oxidation of MACR under these

conditions yields methacryloyl peroxynitrate (MPAN), especially under high NO2/NO

conditions.21,25,26 It has been recently shown that when MPAN is oxidized by OH it yields at

least two SOA precursors, including methacrylic acid epoxide (MAE) and hydroxymethyl-

methyl-α-lactone (HMML).13,16,21,25 Under both high- and low-NOx conditions, aerosol acidity

is a critical parameter that enhances the reaction kinetics through acid-catalyzed reactive

uptake and multiphase chemistry of either IEPOX or MAE/HMML.19,21,20 The current

understanding of the chemical pathways leading to SOA formation from isoprene is

represented in Figure 1.

Figure 1. Isoprene oxidation pathways under high-NOx (upper) and low-NOx (lower) conditions.16,21 23,27

OH

O2 OHOO

NO NO2

O

isoprene

OH

O2/NO2O

OH

O

methacrylic acid

epoxide (MAE)

MACR

OHO

OH

HO

OHO

OH

HO3SO

OHO

OH

O

O

HOOH

Organosulfate

2-methylglyceric acid Dimeric ester

HO2

OHOOH

ISOPOOH

OHO2

O

OH

HO

isoprene epoxyidols (IEPOX)

HOOH

OHOH

HO

OHOH

HOOH

OHO

OH

OHOH

HOOSO3H

OHO

OH

OHOH

HO

O

OH

HOOSO3H

OHOH

2-methyltetrols

C5-alkene triols

Dimer

Organosulfate of dimerOrganosulfate

3-MeTHF-3,4-diols

H+

H+

Gas Phase Aerosol Phase

H2O

H+

SO42-

MAE

IEPOX

IEPOX

H+

H+

H+SO42-

H2O

H+

Brown Carbon

NOOH

ONO2

OH

O

O

OH

+

hydroxymethyl-methyl-α-lactone

(HMML)

  4    

Though once considered not to form SOA, it is becoming more apparent that isoprene

may be a major precursor to organic aerosol (OA). Due to the considerable emissions of

isoprene, even a meager 1% SOA yield would translate into significant SOA contributions to

the global atmospheric environment.7,28 This suggestion is supported by estimates of up to a

third of total fine OA mass attributed to IEPOX-derived SOA tracers in field studies in

Atlanta, GA.29,30 A recent study in Yorkville, GA, found similar results where IEPOX-

derived SOA comprised 20% of the fine organic aerosol mass.19 Another SOAS site at

Centerville (CTR), Alabama (AL) revealed IEPOX-SOA contributed 17% of total OA

mass.31 The individual field sites corroborate measurements made in the Studies of Emissions

and Atmospheric Composition, Clouds, and Climate Coupling by Regional Surveys

(SEAC4RS) aircraft campaign, which estimates IEPOX-SOA contribution of 32% to OA

mass in the southeastern U.S.31

It is clear now that particle-phase chemistry of isoprene-derived oxidation products

plays a large role in atmospheric SOA formation. Yet much remains unknown regarding the

exact nature of its formation, limiting the ability of models to accurately account for isoprene

SOA.32,33 Currently, traditional air quality models in the southeastern U.S. that do not

incorporate detailed particle-phase chemistry of isoprene oxidation products (IEPOX or

MAE/HMML) under-predict isoprene SOA formation.34 Recent work demonstrates that

incorporating the specific chemistry of isoprene epoxide precursors into models increases the

accuracy of predicted isoprene SOA,35,36 suggesting that understanding the formation

mechanisms of biogenic SOA, especially with regard to how it is affected by anthropogenic

emissions, such as NOx, sulfate, and aerosol acidity, will be key to yielding more accurate

modeling methods. More accurate models are needed in order to make cost-effective control

  5    

strategies for reducing PM2.5 levels. Since isoprene is primarily biogenic in origin, and

therefore not controllable, the key to understanding the public health and environmental

implications of isoprene SOA lies in resolving the effects of anthropogenic pollutants.

This study presents results from the 2013 Southeastern Oxidant and Aerosol Study

(SOAS), where several well-instrumented ground sites dispersed throughout the southeastern

U.S. made intensive gas- and particle-phase measurements from June 1 – July 16, 2013. The

primary purpose of this campaign was to examine, in greater detail, the formation

mechanisms, composition, and properties of biogenic SOA, including how it is affected by

anthropogenic emissions. This study pertains specifically to the results from the Birmingham

(BHM), AL ground site, where the city’s ample urban emissions mix with biogenic

emissions from the surrounding rural areas, making it an ideal location to investigate such

interactions. The results presented here focus on gas chromatography interfaced to electron

impact-mass spectrometry (GC/EI-MS) analysis of PM2.5 collected on filters during the

campaign. More specifically, GC/EI-MS analysis of PM2.5 was conducted in order to measure

quantities of known isoprene SOA tracers (2-methyltetrols, 3-MeTHF-3,4-diol isomers, C5-

alkene triols, 2-methylglyceric acid, and IEPOX-dimers) and using collocated air quality and

meteorological measurements to investigate how anthropogenic pollutants (NOx, SO2, aerosol

acidity, PM2.5, sulfate) affect isoprene SOA formation. These results, along with the results

presented in similar studies from the 2013 SOAS campaign, seek to elucidate the chemical

relationships between anthropogenic emissions and SOA formation in order to provide better

parameterizations needed to improve the accuracy of air quality models in this region of the

U.S.

  6    

     

CHAPTER 2: EXPERIMENTAL  

2.1 Site Description and Measurements Made During 2013 SOAS at BHM

Filter samples were collected in the summer of 2013 as part of the SOAS field

campaign at the BHM ground site, located in downtown Birmingham. In addition to the

SOAS campaign, the site is also part of the Southeastern Aerosol Research and

Characterization (SEARCH) Study (Figure 2), an observation and monitoring program

initiated in mid-1998. A detailed description of SEARCH and this site is provided

elsewhere.37,38 From June 1 – July 14, 2013, PM2.5 samples were collected onto TissuquartzTM

Filters (Pall Life Sciences) using high-volume PM2.5 samplers (Tisch Environmental)

operated at 1 m3 min-1 (adjusted to ambient temperature). All quartz filters were pre-baked at

550 °C for 18 hours prior to sampling before being cooled down to 25 °C over 12 hours.

Figure 2. Map of SEARCH network locations. Figure adapted from Hansen et al. 200338

  7    

Samples were obtained according to the schedule listed in Table 1. Either two or four

samples were collected per day. The regular schedule consisted of two samples per day, one

during the day, the second at night, each collected for 11 hours. On days when four samples

were collected, the day sampling consisted of three separate samples. This intensive sampling

schedule was performed on days when high levels of isoprene and anthropogenic emissions,

which specifically included sulfate (SO42-) and NOx, where forecast by the National Center

for Atmospheric Research (NCAR) using the Flexible Particle (FLEXPART) dispersion

model39 in conjunction with the Model for Ozone and Related Chemical Tracers (MOZART)

model40 to predict periods of high anthropogenic emissions as well as the Model of

Emissions of Gases and Aerosols from Nature (MEGAN)8 to predict biogenic gas and

particle abundance. The higher collection frequency allowed enhanced time resolution for

offline analysis methods to examine the effect of anthropogenic emissions on the evolution

of isoprene SOA throughout the day. In total, 120 samples were collected throughout the

field campaign. All filters were stored at -20 °C under dark conditions until extraction and

analysis.

Table 1. Sampling schedule during SOAS at the BHM ground site.

No. of samples/day Sampling schedule Dates

8am-7pm, Jun 1 - Jun 9,8pm-7am next day Jun 13,

Jun 17 – Jun 28, July 2- July 9 15-Jul

8am-12pm, Jun 10 - Jun 12, 1pm-3pm, Jun 14- Jun 16, 4pm-7pm, Jun 29 - Jun 30, 8pm-7am next day Jul 9 - Jul 14

2 (regular)

4 (intensive)

  8    

In addition to filter sampling of PM2.5, a suite of additional instruments at the site

provided collected measurements of a variety of variables, including meteorology, gas

concentrations, and continuous PM monitoring. These variables and their respective

instrumentation are summarized in Table 2.

  9    

Table 2. Instrumentation and operating parameters of collocated measurements at BHM. Time resolution given in minutes with averaging interval, i.e. “5, 60” denotes measurements collected every 5 minutes and averaged every 60 minutes.

Category Variable Analyzer/Sensor

Time Resolution (Interval, average)

(minutes)Meteorology Wind Speed/Direction RMYoung 81000 sonic 5, 60

T/RH/BP Paroscientific Met4A 5, 60T/RH Vaisala 5, 60PAR Licor 5, 60Precipitation ETI-NOAH IV 5, 60Aerosol/cloud layers JenOptik CHM 15k ceilometer 5, 60Surface wetness Vaisala (SWS2) 5, 60

Trace Gases O3 Thermo 49i 5, 60CO Thermo 48i 5, 60SO2 Thermo 43i 5, 60NO Thermo 42i 5, 60NO2 photolysis/Thermo 49i 5, 60HNO3 continuous denuder diff/Thermo 42i 5, 60NOy cat. reduction/Thermo 42i 5, 60NH3 continuous denuder diff/Thermo 42i 5, 60

Continuous PM PM2.5 Mass TEOM 60PMcoarse Mass dichotomous TEOM 60PM2.5 SO4 cat. reduction/Thermo 43i 60PM2.5 NO3 cat. reduction/Thermo 42i 60PM2.5 NH4 cat. oxidation/Thermo 42i 60PM2.5 TC/EC Sunset 60dry Babs (880 nm) Radiance Research M903 5, 60dry Bsp (530 nm) Magee 2ch. Aeth 5, 60ambient Bsp (530 nm) Optec NGN-2a 5, 60

Filter-Based PM PM2.5 Mass gravimetry 1440, dailyPM2.5 ions IC 1440, 1 in 3 daysPM2.5 major/minor elements XRF 1440, dailyPM2.5 water-soluble metals ICPMS 1440, 1 in 3 daysPM2.5 OC/EC TOR 1440, 1 in 3 daysPMcoarse Mass gravimetry 1440, 1 in 3 daysPMcoarse ions IC 1440, 1 in 3 daysPMcoarse major/minor elements XRF 1440, 1 in 3 daysPMcoarse water-soluble metals ICPMS 1440, 1 in 3 days

Hi-Vol Based PM PM2.5 OC/EC TOR 23-hr, dailyPM2.5 ions IC 23-hr, dailyPM2.5 (other) various 11-hr, daily

  10  

2.2 Gas Chromatography/Electron Impact-Mass Spectrometry (GC/EI-MS) Analysis

Quartz filters collected in the field were extracted for isoprene SOA tracers that were

quantified by GC/EI-MS with prior trimethylsilylation. 37-mm diameter circular punches of

each quartz filter were extracted in pre-cleaned scintillation vials with 20 mL high-purity

methanol (LCMS CHROMASOLV-grade, Sigma-Aldrich) by sonication for 45 minutes. The

extracts were filtered through PTFE syringe filters (Pall Life Science, Acrodisc®, 0.2-μm

pore size) in order to remove suspended particles and residual quartz fibers and blown dry

under a gentle stream of N2 at room temperature. Residues were immediately

trimethylsilylated by reaction with 100 μL BSTFA + TMCS (99:1 v/v, Supelco) and 50 μL

pyridine (anhydrous, 99.8 %, Sigma-Aldrich) at 70 °C for 1 hour. Derivatized samples were

analyzed within 24 hours using a Hewlett-Packard (HP) 5890 Series II Gas Chromatograph

equipped with an Econo-Cap®-EC®-5 Capillary Column (30 m x 0.25 mm i.d.; 0.25-μm

film thickness) coupled to a HP 5971A Mass Selective Detector. 1 μL aliquots were injected.

Operating conditions and procedures of the GC/EI-MS technique have been described

elsewhere.21

Filter extraction efficiency was taken into account for the quantification of all SOA

tracers. Extraction efficiency was determined by analyzing 4 pre-baked filters spiked with 50

ppb of 2-methyltetrols, 2-methylglyceric acid, levoglucosan, and cis- and trans-meTHF-3,4-

diols. The chromatographic peaks illustrated in the total ion chromatogram (TIC) shown in

Figure 3 were used to identify and quantify isoprene-derived SOA tracers. Extracted ion

chromatograms (EICs) of m/z 262, 219, 231, 335 were used to quantify the cis-/trans-3-

MeTHF-3,4-diols, 2-methyltetrols and 2-methylglyceric acid, C5-alkene triols, and IEPOX-

dimers, respectively.

  11  

Figure 3. Total ion chromatogram (TIC) illustrating the isoprene SOA tracers measured by the GC/EI-MS (1) cis- and trans-3-MeTHF-3,4-diols, (2) 2-methylglyceric acid, (3) C5-alkene triols, and (4) 2-methyltetrols. IEPOX-derived dimers were not detected in this sample. TIC illustrated was taken from the June 17, 2013 Day sample.

2-Methyltetrols were quantified using an authentic 1:1 erythro/threo diastereomer reference

standard. Similarly, 3-MeTHF-3,4-diol isomers were quantified using authentic standards;

however, the 3-MeTHF-3,4-diol isomers were detected in only a few field samples. C5-

alkene triols and IEPOX-derived dimers were quantified using the average response factor of

the 2-methyltetrol mixture. 2-Methylglyceric acid was also quantified using an authentic

standard. Synthesis procedures for the 2-methyltetrol diastereomers, 3-MeTHF-3,4-diol

isomers, and 2-methylglyceric acid have been described elsewhere.23,30 Filters were also

analyzed by ultra performance liquid chromatography/electrospray ionization high-resolution

quadrupole time-of-flight mass spectrometry (UPLC/ESI-HR-QTOFMS) for OS derived

from isoprene oxidation products (MAE/HMML and IEPOX). Details of this method have

been described by Lin et al.19 and Budisulistiorini e al.30 Although the UPLC/ESI-HR-

  12  

QTOFMS analysis is beyond the scope of this report, they have been incorporated into the

discussion in order to further support the source of isoprene SOA tracers measured by

GC/EI-MS. Authentic standards were available for quantifying the MAE- and IEPOX-

derived OS. Details of the synthesis for the two authentic standards have also been reported

elsewhere.30

2.3. Estimations of aerosol pH

Aerosol pH was estimated using the ISORROPIA-II thermodynamic model.41 Sulfate (SO42-),

nitrate (NO3-), and ammonium (NH4

+) ion concentrations measured in PM2.5 collected from

BHM, as well as relative humidity (RH), temperature and gas-phase ammonia (NH3) were

used as inputs into the model. Due to the limited number of NH4+ measurements, an average

NH4+ concentration was used from these measurements to calculate aerosol pH for all other

points. All variables needed for the model estimates were obtained from the SEARCH

network at BHM, which collected these measurements during the same time as the SOAS

campaign. The ISORROPIA-II model estimated particle hydronium ion concentration per

unit volume of air (H+, μg m−3), aerosol liquid water content (LWC, μg m−3), and aqueous

aerosol mass concentration (μg m−3). The following formula used these model-estimated

parameters to calculate the aerosol pH:

Aerosol  pH = −log!"𝑎!! =  −log!"(  𝐻!"#!

𝐿𝑀𝐴𝑆𝑆  ×  𝜌!"#  ×  1000  )

Where  𝑎!! is the H+ activity in the aqueous phase (molL-1), LMASS is the total liquid-phase

aerosol mass (μg m−3) and  𝜌!"# is the aerosol density. Details of the ISORROPIA-II model

and its ability to predict pH, LWC, and gas-to-particle partitioning are discussed elsewhere.42

  13  

     

CHAPTER 3: RESULTS AND DISCUSSION  

3.1 Overview of Study

The campaign spanned June through mid-July, 2013. Temperature during this period

averaged 26.4 °C with a high and low temperature of 32.6 and 20.5 °C. RH varied from 37-

96% throughout the campaign, with an average of 71.5%. Rainfall occurred intermittently,

occurring in 2-3 day periods. Wind analysis shows air masses approached largely from the

south-south east at an average wind speed of 2 m s-1. Summaries of meteorological

conditions during the course of the campaign are listed in Table 3 and illustrated in Figures 4

and 5.

O3 levels for the majority of the campaign averaged 31.1 ppb. O3 levels were slightly

higher on intensive sampling days (average 37.0 ppb). Carbon monoxide (CO), a combustion

byproduct, had an average mixing ratio of 208.7 ppb. SO2, NOx, and NH3 levels were present

in lower concentrations with averages of 0.9, 7.8, and 1.9 ppb, respectively. The largest

inorganic component of PM2.5 was SO42-, which varied between 0.4 and 4.9 μg m-3 during the

campaign and NH4+ and NO3

= averaged 1.5 and 0.8 μg m-3, respectively, during the campaign.

Time series summaries of gas and PM2.5 concentrations are illustrated in Figures 6 and 7,

respectively.

  14  

Table 3. Summary of collocated measurements of meteorological factors, gaseous species, and PM2.5 constituents.

Figure 4. Wind rose illustrating wind direction during the campaign at the BHM site. Bars indicate direction of incoming wind, with 0 degrees set to geographic north. Length of bar size indicates frequency with color segments indicating the wind speed in m s-1.

Category Variable Average SD Min MaxMeteorology Rainfall (in) 0.08 0.22 0.00 1.37

Temperature °C 26.38 2.96 20.49 32.65RH (%) 71.53 14.95 36.93 96.06Aerosol pH 3.21 0.37 1.44 4.11

Trace Gas (ppb) O3 31.12 14.78 8.30 62.24CO 208.72 71.97 99.56 422.89SO2 0.89 0.78 0.10 3.72NOx 7.84 5.99 1.34 29.68NH3 1.86 0.81 0.67 3.98

PM2.5 (μg m-3) SO4-2 1.99 0.90 0.36 4.91

NO3- 0.14 0.10 0.00 0.83

NH4+ 0.66 0.30 0.15 1.54

Total OC 7.21 3.20 1.39 14.91

  15  

Figure 5. Time series of meteorological conditions during the campaign at BHM.

Figure 6. Time series concentrations of trace gas species during the 2013 SOAS campaign at the BHM site.

  16  

Figure 7. Time series mass concentrations of PM2.5 constituents (SO4

2-, NO3-, NH4

+, and OC) measured during the 2013 SOAS campaign at the BHM site.

3.2 Characterization of Isoprene SOA

GC/EI-MS analysis of the filter samples revealed the presence of IEPOX-derived

SOA tracers, including 2-methyltetrols and C5-alkene triols, as well as the MAE/HMML-

derived SOA tracer, 2-methylglyceric acid. Concentrations of 3-MeTHF-3,4-diols and

IEPOX-derived dimers were often near or below detection limits, and were therefore only

detected in a limited number of samples. Levoglucosan was also quantified as a marker for

biomass burning. Table 3 summarizes the measurements of isoprene SOA tracers and

levoglucosan in PM2.5 samples from BHM. Isoprene SOA contribution to total OM was

estimated by assuming an OM/OC ratio of 1.4, as traditionally used for urban regions.43

Overall, isoprene-derived SOA (sum of both IEPOX- and MAE/HMML-derived SOA)

contributed, on average, 7.11% (up to 21.70%) of total particulate OM. This fraction is lower

than measured at other sites in the southeastern US including both rural Look Rock, TN,30,31

  17  

and urban Atlanta, GA,29 though the measured absolute concentrations of isoprene-SOA

tracers were comparable and in some cases, higher.

2-Methyltetrols were the most abundant of the isoprene-derived SOA tracers

measured during this study, with 2-methylthrietol contributing 20.9% (107.26 ng m-3) and 2-

methylerithritol contributing 40.9% (266.73 ng m-3) on average to the total mass of tracers

detected by GC/EI-MS. On average, 2-methyltetrols comprised 61.8% (373.99 ng m-3) of

detected tracer mass. The ratio of average 2-methylthrietol to 2-methylerythritol in BHM was

0.40, slightly higher than at rural sites located at CTR, AL (0.36) and Look Rock (LRK), TN

(0.35). Sum of the C5-alkene-triol isomers accounted for 24.9% on average (169.7 ng m-3),

with the most abundant isomer, 2-methylbut-3-ene-1,2,4-triol, contributing 16.2% (108.96 ng

m-3) of the average isoprene SOA tracer mass identified by GC/EI-MS. MAE/HMML-

derived SOA tracers exhibited lower concentrations than IEPOX-derived tracers. On average,

MAE/HMML-derived 2-methylglyceric acid contributed 2.3% (10.39 ng m-3) of the total

isoprene-derived SOA tracer mass. The levels of isoprene-derived SOA tracers measured at

BHM were higher than those concurrently measured at the SOAS ground site located at

LRK,44 and higher than those reported for a previous study at Yorkville, GA.19 SOA

abundance was also higher than reported in previous field studies conducted in central

Europe,45 as well as at rural and urban sites in China.46,47 However, 2-methyltetrols and C5-

alkene-triols were quantitated in these prior studies using surrogate standards structurally

unrelated to the target analytes to quantify, rather than the authentic 2-methyltetrol standards

used in this and the other studies in the southeastern U.S.45–47

  18  

Table 4. Summary of Isoprene-derived SOA Tracer Concentrations Measured by GC/EI-MS

Interestingly, 2-methyltetrols were more abundant, but C5-alkene-triols less abundant

at BHM than in concurrent sampling at the SOAS site at CTR. At CTR, 2-methylerythritol

contributed 26.5% (204.8 ng m-3) of detected tracer mass while C5-alkene-triols accounted for

27.8% (214.5 ng m-3) of the isoprene-derived SOA tracer mass at CTR. The ratio of average

total 2-methyltetrols to average total C5-alkene-triols in BHM was 2.2, nearly double that of

CTR (1.3) and LRK (1.1). Although 2-methyltetrols and C5-alkene-triols are considered to

form readily from the acid-catalyzed reactive uptake and multiphase chemistry of IEPOX,21,22

this observation suggests an additional source of 2-methyltetrols. The disparity may be

explained by ozonolysis of isoprene by urban O3 in the presence of acidic aerosol which has

been shown to yield the 2-methyltetrols.48 The role of ozonolysis is discussed in section 3.4.

The high levels of IEPOX-derived SOA products also provide support for recent work

detailed in Jacobs et al.,49 which reported significant IEPOX yield from isoprene-derived

# of Samples

Detected*

Avg % mass of GC/EI-MS

detected tracers

Avg % of total

particulate organic mass

Maximum MeanLevoglucosan 39.5 922.57 98.69 120 - -2-methylglyceric acid 23.4 43.82 10.39 112 2.27% 1.07%2-methylthreitol 32.9 388.85 107.26 120 20.92% 2.72%2-methylerythritol 33.7 1048.86 266.73 119 49.90% 0.10%cis-3-meTHF-3,4-diol 20.0 98.93 6.84 27 1.02% 0.35%trans-3-meTHF-3,4-diol 21.0 137.62 8.57 12 1.01% 0.21%IEPOX-derived dimer 54.0 2.22 0.04 12 0.003% 1.017%

(Z)-2-methylbut-3-ene-1,2,4-triol 25.6 287.79 37.28 115 5.36% 0.07%

2-methylbut-3-ene-1,2,3-triol 26.6 260.58 23.43 113 3.29% 0.09%

(E)-2-methylbut-3-ene-1,2,4-triol 26.9 878.45 108.96 116 16.23% 0.00%

Tracer Retention Time (min)

Concentration (ng m-3)

*total samples: 120

  19  

hydroxy nitrates suggesting that the current route of IEPOX formation under low-NOx

conditions might be revised.

3.3 Effect of Sulfate and Acidity

Sulfate was moderately correlated with IEPOX-derived SOA (r2 = 0.34),

MAE/HMML-derived SOA (r2 = 0.31), and total isoprene SOA (r2 = 0.34). With regard to

specific SOA tracers, sulfate was weakly correlated with 2-methylglyceric acid (r2 = 0.11)

and slightly better correlated with 2-methyltetrols (r2 = 0.26) and the C5-alkene-triols (r2 =

0.33). These results agree with recent field work by Xu et al.50 that suggests that sulfate

aerosol concentration is paramount to the formation of biogenic SOA. Aerosol pH averaged

3.21, and ranged between 1.44 and 4.11, as modeled by the ISORROPIA thermodynamic

model. Acidity correlated with total isoprene SOA (r2= 0.24), and with IEPOX- and

MAE/HMML-derived SOA (r2= 0.18 and 0.32, respectively). Acidity was consistently

correlated between major IEPOX-SOA components, 2-methyltetrols (r2 = 0.19) and C5-

alkene-triols (r2 = 0.20). MAE/HMML-derived OS exhibited higher correlation with acidity

(r2 = 0.32) than 2-MG (r2 = 0.20). The correlation between aerosol acidity and total isoprene

SOA formation demonstrates that aerosol acidity may have a local role for aerosol formation.

  20  

Figure 8. Sulfate aerosol correlated with all three GC/EI-MS quantified tracers, but more strongly with IEPOX-derived tracers, especially C5-alkene triols.

  21  

3.4 Effect of Anthropogenic Emissions

Figure 9. NOx/NOy concentrations exhibit ~0 correlation with major SOA Tracers

Total IEPOX-, MAE- and isoprene-derived SOA mass were not correlated with NOx

or NOy. None of the SOA tracers, including 2-MG and MAE-derived OS, were correlated

with NOx (r2 = 0). This is inconsistent with the current understanding of SOA formation from

  22  

isoprene oxidation pathways under high-NOx conditions, which proceeds through MAE16,

and, as recently suggested, HMML25, to yield 2-MG and its OS derivative. Several

explanations may account for the disparity between laboratory studies and field results from

BHM. First, tracers may form distant from BHM and be transported to the site. Wind

direction measurements during the campaign casts doubts on the first explanation, as wind

came largely from the south-southeast of BHM, a rural area that would exhibit little NOx

abundance. The age of the air plume may also play a role in 2-MG and its related OS.

Nguyen et al.25 demonstrated that 2-MG is highly volatile, especially under acidic conditions,

whereas its OS counterpart is relatively stable. The NOx:NOy ratio was calculated as a metric

for plume age.51 Plume age was mildly inversely correlated with 2-MG abundance (r2= 0.22)

but not correlated with MAE/HMML-derived OS (r2= 0.10). Another hypothesis is that other

as yet unknown NOx-independent pathways may account for a significant proportion of

isoprene SOA formation. This would explain the ambient 2-MG observed at the remote

SOAS site at LRK, TN,30 far from the influences of primary anthropogenic emissions,

including NOx. Surprisingly, O3 modestly correlated with both 2-MG (r2= 0.26), and total

MAE/HMML-SOA (r2= 0.31) (which includes 2-MG and MAE/HMML-OS). Further

examination reveals that O3 was highly correlated with plume age (r2= 0.79). This is

unsurprising as tropospheric O3 results from photolysis of NO2, which is abundant due to

traffic at the urban ground site. The correlation between MAE/HMML-derived SOA tracers

with O3 and plume age but not NOx or NOy suggests that ozone may have a more prominent

role in the local production of 2-MG and its related OS. This may occur via enhanced

oxidation of isoprene by ozone to form MACR, a precursor to MAE and HMML.16,25

  23  

Figure 10. Ozone correlates with MAE/HMML-derived SOA tracers.

As discussed in section 3.2, O3 was weakly correlated with 2-methyltetrols (r2 = 0.16)

although not found to correlate with 2-methyltetrols (r2 < 0.1) in previous field studies.19,44

Recent laboratory work reported by Riva et al.48 have demonstrated that 2-methyltetrols can

be formed via isoprene ozonolysis in the presence of acidified sulfate aerosol; however, C5-

alkene-triols are not formed via this oxidation pathway. Lower abundance of C5-alkene-triols

and higher abundance of 2-methyltetrols at the urban BHM site relative to the rural CTR site

is consistent with 2-methyltetrols formed by ozonolysis adding to the level formed by

reactive uptake of IEPOX shown in Figure 1.

  24  

3.5 Nighttime Nitrate Radical Production

Nitrate (NO3) radical production (PNO3) was calculated using the following equation:

𝑃!"! = 𝑁𝑂! 𝑂! 𝑘

Where [NO2] is measured ambient NO2 concentration in mol cm-3, [O3] is measured O3 in

mol cm-3, and k is the temperature-dependent rate constant. When considering all data points,

including day, night, and intensive samples, 2-MG was not correlated with PNO3 (r2 = 0.07).

However, during the nighttime samples (sampling between 20:00 and 7:00) 2-MG was

moderately correlated with PNO3 (r2 = 0.24). This correlation is closer (r2 = 0.56) on days

when 2-MG abundance is relatively low (<10 ng m-3). This observation suggests that when

the planetary boundary layer (PBL) heights decrease at night, the remaining isoprene, NO2,

and O3 continue to react. In conjunction with the previously noted correlation between

MAE/HMML-derived SOA tracers, this result may indicate that O3 plays a larger role in

SOA formation at night, since PNO3 is a direct function of O3 abundance. Stronger correlation

may occur because 2-MG formation is NO2 dependent,13,26 and the absence of sunlight at

night prevents photolysis of NO2 to NO. Alternatively, there may be pathways in addition to

the photolysis of MACR high-NOx conditions, for example, an unknown NO3 oxidation

pathway. O3 may also explain this phenomenon. Future studies might consider examining

MACR oxidation under dark conditions using NO3 radical, since work by Ng et al.52

demonstrated that dark oxidation reactions of isoprene by NO3 radicals yields substantial

SOA formation.

  25  

Figure 11. Overall nighttime NO3 radical production (purple and red) is mildly correlated with 2-MG abundance, but more strongly correlated when high 2-MG abundance outliers (red) are not considered. High correlation between low abundance of 2-MG (purple) and nighttime NO3 radical production suggests a dual regime for 2-MG formation’s dependence on NO3 radical.

   

  26  

CHAPTER 4: CONCLUSION  

This study examined isoprene SOA tracers in PM2.5 samples collected at the BHM

ground site during the 2013 SOAS campaign. Our analysis reveals the complexity of and

potential multitude of chemical pathways that lead to isoprene SOA formation. Analysis, to

date, of SOAS campaign sites including CTR, BHM, and LRK offer insights into the factors

that influence SOA formation in rural and urban environments in the southeastern US. A

summary of SOA tracers from these sites is shown in Table 5. A comparison of tracer

abundance reveals a stark contrast between trends in both high- and low-NOx isoprene SOA

products that may be explained by anthropogenic influences. Out of the three 2013 SOAS

ground sites that collected and chemically analyzed PM2.5 samples for isoprene-derived SOA

tracers, BHM exhibited the highest abundance of 2-methyltetrols, as well as the highest

contribution of 2-methyltetrols to total quantified tracers. The unique correlation between O3

and isoprene-derived SOA tracers observed at the BHM site suggests a greater role for O3 in

SOA formation. O3 may explain the divergence in BHM’s ratio of 2-methyltetrols to C5-

alkene-triols relative to measurements from CTR and LRK. This result may provide evidence

to support the recent work by Riva et al.48 demonstrating the potential role of O3 in

generating isoprene-derived SOA besides only through reactive uptake of IEPOX onto acidic

aerosol. Future chamber experiments examining the relative potential for SOA formation

from isoprene via both ozonolysis and IEPOX uptake may elucidate which chemical

processes are responsible for SOA abundance in urban areas with high atmospheric O3 and

  27  

NOx. SOA products in the high-NOx regime exhibit a similar contrast from rural field

measurements, with 2-MG being the dominant high-NOx tracer over its related OS. A high

correlation between O3 and MAE/HMML-derived SOA tracers may suggest a greater role for

O3 in the formation of high-NOx SOA products, especially at the local level in urban areas,

where O3 is abundant. In addition, a correlation between nighttime nitrate production and 2-

MG at night suggests ozonolysis may be a source of SOA at night. These results also provide

new insights into the previously proposed high-NOx pathway of isoprene-SOA formation.

The lack of association of MAE/HMML-derived SOA tracers with anthropogenic NOx

further suggests a gap in our current understanding regarding the formation of 2-MG and

related OS. Future laboratory studies should elaborate on the phenomena observed here by

testing the potential for MAE/HMML-derived SOA under dark and light conditions, as well

as exploring additional pathways for SOA formation outside of the current NOx-dependent

understanding, especially via ozonolysis. Recent work has demonstrated that pathways

leading to isoprene-SOA products may be more complex than previously thought.25 Together,

the results presented here demonstrate an enhancement of isoprene SOA loadings in the

presence of anthropogenic pollutants (e.g. O3 and NOx). Additional laboratory studies such as

those suggested above may do much to improve our understanding of SOA formation

mechanisms and how anthropogenic influences might enhance it; this can help to inform how

existing air quality models, such as the EPA CMAQ model, can be improved. Prior work has

already started to add new information on the chemistry of isoprene SOA formation into the

EPA CMAQ model, as recently demonstrated in Pye et al.35 Understanding of these chemical

processes will be crucial to crafting appropriate air quality policies that mitigate the

  28  

enhancement of isoprene SOA formation by anthropogenic pollutants (i.e., acidified sulfate

aerosol, NOx, and O3).

Table 5. Summary of isoprene-SOA tracers from the three SOAS ground sites at Centerville, AL, Birmingham, AL, and Look Rock, TN.

Mean (ng m-3)

Max (ng m-3)

Average amount detected tracers

(%)

Mean (ng m-3)

Max (ng m-3)

Average amount detected tracers

(%)

Mean (ng m-3)

Max (ng m-3)

Average amount detected tracers

(%)Isoprene high-NOx SOA tracers

(MAE/HMML-SOA)

MAE/HMML-derived organosulfate 10.2 43.1 1.30% 7.2 35.7 1.10% 8.2 57.3 1.80%

2-methylglyceric acid 5.1 34.7 0.70% 10.4 43.8 1.66% 7.5 36.7 1.60%

Isoprene Low-NOx SOA tracers (IEPOX-SOA)

IEPOX-derived organosulfates 207.1 755.3 26.80% 164.5 865.1 24.31% 139.2 835.3 30.30%

IEPOX-derived dimer organosulfate 0.7 3.2 0.10% 0.04 2.2 0.003% 1.1 10.3 0.20%

trans-3-MeTHF-3,4-diol 0 0 0.00% 8.6 137.6 1.01% 2.7 18.8 0.60%

cis-3-MeTHF-3,4-diol 0.2 4.8 0.00% 6.8 98.9 1.02% 1.7 5.7 0.40%

2-methylthreitol 73.7 261.2 9.50% 107.3 388.8 15.83% 42.4 329.8 9.20%

2-methylerythritol 204.8 676.4 26.50% 266.7 1048.9 37.91% 120.7 1269.7 26.30%

(Z)-2-methylbut-3-ene-1,2,4-triol 50.7 350.2 6.60% 37.3 287.8 4.10% 29.1 260 6.10%

2-methylbut-3-ene-1,2,3-triol 26.1 149.5 3.40% 23.4 260.6 2.46% 16.5 162.5 3.60%

(E)-2-methylbut-3-ene-1,2,4-triol 137.3 734.3 17.80% 109.0 878.4 12.34% 98.8 1127 21.50%

SOA tracers

CTR BHM LRKSOAS Site tracer concentrations

  29  

REFERENCES

1. Ipcc. Summary for Policymakers. In: (2013). doi:10.1017/CBO9781107415324

2. Kanakidou, M. et al. Organic aerosol and global climate modelling: a review. Atmos. Chem. Phys. 5, 1053–1123 (2005).

3. Pope, C. A. & Dockery, D. W. Health effects of fine particulate air pollution: lines that connect. J. Air Waste Manag. Assoc. 56, 709–42 (2006).

4. Hallquist, M. et al. The formation, properties and impact of secondary organic aerosol: current and emerging issues. Atmos. Chem. Phys. 9, 5155–5236 (2009).

5. Nozière, B. et al. The Molecular Identification of Organic Compounds in the Atmosphere: State of the Art and Challenges. Chem. Rev. 150203102816008 (2015). doi:10.1021/cr5003485

6. Boucher, O. et al. 7. Clouds and Aerosols. Clim. Chang. 2013 Phys. Sci. Basis. Contrib. Work. Gr. I to Fifth Assess. Rep. Intergov. Panel Clim. Chang. 571–657 (2013). doi:10.1017/CBO9781107415324.016

7. Carlton, a. G., Wiedinmyer, C. & Kroll, J. H. A review of Secondary Organic Aerosol (SOA) formation from isoprene. Atmos. Chem. Phys. Discuss. 9, 8261–8305 (2009).

8. Guenther, A. et al. Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos. Chem. Phys. 6, 3181–3210 (2006).

9. Chameides, W., Lindsay, R., Richardson, J. & Kiang, C. The role of biogenic hydrocarbons in urban photochemical smog: Atlanta as a case study. Science (80-. ). 241, 1473–1475 (1988).

10. 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. Gen. Top. 25, 997–1008 (1991).

11. Kamens, R. M., Gery, M. W., Jeffries, H. E., Jackson, M. & Cole, E. I. Ozone-isoprene reactions: Product formation and aerosol potential. Int. J. Chem. Kinet. 14, 955–975 (1982).

12. Claeys, M. Formation of Secondary Organic Aerosols Through Photooxidation of Isoprene. Science (80-. ). 303, 1173–1176 (2004).

  30  

13. Surratt, J. D. et al. Chemical composition of secondary organic aerosol formed from the photooxidation of isoprene. J. Phys. Chem. A 110, 9665–9690 (2006).

14. Edney, E. O. et al. Formation of 2-methyl tetrols and 2-methylglyceric acid in secondary organic aerosol from laboratory irradiated isoprene/ NOX/SO2/air mixtures and their detection in ambient PM2.5 samples collected in the eastern United States. Atmospheric Environment 5281–5289 (2005). at <http://www.unc.edu/courses/2008fall/envr/890/001/lecture5/05 Atmso Envir Edney Cleays.pdf>

15. Kroll, J. H., Ng, N. L., Murphy, S. M., Flagan, R. C. & Seinfeld, J. H. Secondary organic aerosol formation from isoprene photooxidation. Environ. Sci. Technol. 40, 1869–1877 (2006).

16. Lin, Y.-H. 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. 110, 6718–23 (2013).

17. Lin, Y. H. et al. Isoprene epoxydiols as precursors to secondary organic aerosol formation: Acid-catalyzed reactive uptake studies with authentic compounds. Environ. Sci. Technol. 46, 250–258 (2012).

18. Kroll, J. H., Ng, N. L., Murphy, S. M., Flagan, R. C. & Seinfeld, J. H. Secondary organic aerosol formation from isoprene photooxidation under high-NO x conditions. Geophys. Res. Lett. 32, n/a–n/a (2005).

19. Lin, Y. H., Knipping, E. M., Edgerton, E. S., Shaw, S. L. & Surratt, J. D. Investigating the influences of SO2 and NH3 levels on isoprene-derived secondary organic aerosol formation using conditional sampling approaches. Atmos. Chem. Phys. 13, 8457–8470 (2013).

20. Surratt, J. D. et al. Effect of acidity on secondary organic aerosol formation from isoprene. Environ. Sci. Technol. 41, 5363–5369 (2007).

21. Surratt, J. D. et al. Reactive intermediates revealed in secondary organic aerosol formation from isoprene. Proc. Natl. Acad. Sci. U. S. A. 107, 6640–6645 (2010).

22. Wang, W. et al. Characterization of oxygenated derivatives of isoprene related to 2-methyltetrols in Amazonian aerosols using trimethylsilylation and gas chromatography/ion trap mass spectrometry. Rapid Commun. Mass Spectrom. 19, 1343–51 (2005).

23. Zhang, Z. et al. Technical Note: Synthesis of isoprene atmospheric oxidation products: isomeric epoxydiols and the rearrangement products cis- and trans-3-methyl-3,4-dihydroxytetrahydrofuran. Atmos. Chem. Phys. 12, 8529–8535 (2012).

  31  

24. Lin, Y. et al. Light-Absorbing Oligomer Formation in Secondary Organic Aerosol from Reactive Uptake of Isoprene Epoxydiols. Environ. Sci. Technol. 48, 12012–12021 (2014).

25. Nguyen, T. B. et al. Mechanism of the hydroxyl radical oxidation of methacryloyl peroxynitrate (MPAN) and its pathway toward secondary organic aerosol formation in the atmosphere. Phys. Chem. Chem. Phys. (2015). doi:10.1039/C5CP02001H

26. Chan, a. W. H. et al. Role of aldehyde chemistry and NOx concentrations in secondary organic aerosol formation. Atmos. Chem. Phys. 10, 7169–7188 (2010).

27. Paulot, F. et al. Unexpected epoxide formation in the gas-phase photooxidation of isoprene. -. Science 325, 730–3 (2009).

28. Henze, D. K., Seinfeld, J. H. & Shindell, D. T. Inverse modeling and mapping US air quality influences of inorganic PM2.5 precursor emissions using the adjoint of GEOS-Chem. Atmos. Chem. Phys. Discuss. 8, 15031–15099 (2008).

29. Budisulistiorini, S. H. et al. Real-time continuous characterization of secondary organic aerosol derived from isoprene epoxydiols in downtown Atlanta, Georgia, using the aerodyne aerosol chemical speciation monitor. Environ. Sci. Technol. 47, 5686–5694 (2013).

30. Budisulistiorini, S. H. et al. Examining the effects of anthropogenic emissions on isoprene-derived secondary organic aerosol formation during the 2013 Southern Oxidant and Aerosol Study (SOAS) at the Look Rock, Tennessee, ground site. Atmos. Chem. Phys. 15, 8871–8888 (2015).

31. Hu, W. W. et al. Characterization of a real-time tracer for Isoprene Epoxydiols-derived Secondary Organic Aerosol (IEPOX-SOA) from aerosol mass spectrometer measurements. Atmos. Chem. Phys. Discuss. 15, 11223–11276 (2015).

32. Carlton, A. G., Pinder, R. W., Bhave, P. V. & Pouliot, G. a. To what extent can biogenic SOA be controlled? Environ. Sci. Technol. 44, 3376–3380 (2010).

33. Foley, K. M. et al. Incremental testing of the Community Multiscale Air Quality (CMAQ) modeling system version 4.7. Geosci. Model Dev. 3, 205–226 (2010).

34. Carlton, A. G. et al. Model representation of secondary organic aerosol in CMAQv4.7. Environ. Sci. Technol. 44, 8553–60 (2010).

35. Pye, H. O. T. et al. Epoxide pathways improve model predictions of isoprene markers and reveal key role of acidity in aerosol formation. Environ. Sci. Technol. 47, 11056–64 (2013).

  32  

36. Karambelas, A., Pye, H. O. T., Budisulistiorini, S. H., Surratt, J. D. & Pinder, R. W. Contribution of Isoprene Epoxydiol to Urban Organic Aerosol: Evidence from Modeling and Measurements. Environ. Sci. Technol. Lett. 1, 278–283 (2014).

37. Edgerton, E. S. et al. The Southeastern Aerosol Research and Characterization Study, part 3: continuous measurements of fine particulate matter mass and composition. J. Air Waste Manag. Assoc. 56, 1325–1341 (2006).

38. Hansen, D. A. et al. The Southeastern Aerosol Research and Characterization Study: part 1--Overview. J. Air Waste Manag. Assoc. 53, 1460–1471 (2003).

39. Stohl, a, Forster, C., Frank, a, Seibert, P. & Wotawa, G. Technical note  : The Lagrangian particle dispersion model FLEXPART v6.2. Atmos. Chem. Phys. 2005, 2461–2474 (2005).

40. Emmons, L. K. et al. Description and evaluation of the Model for Ozone and Related chemical Tracers, version 4 (MOZART-4). Geosci. Model Dev. 43–67 (2010). doi:10.5194/gmdd-2-1157-2009

41. Nenes, A., Pandis, S. N. & Pilinis, C. ISORROPIA: A new thermodynamic equilibrium model for multiphase multicomponent inorganic aerosols. Aquat. Geochemistry 4, 123–152 (1998).

42. Fountoukis, C. et al. Thermodynamic characterization of Mexico City aerosol during MILAGRO 2006. Atmos. Chem. Phys. 9, 2141–2156 (2009).

43. White, W. H. & Roberts, P. T. On the nature and origins of visibility-reducing aerosols in the los angeles air basin. Atmos. Environ. 11, 803–812 (1977).

44. Budisulistiorini, S. H. et al. Examining the effects of anthropogenic emissions on isoprene-derived secondary organic aerosol formation during the 2013 Southern Oxidant and Aerosol Study (SOAS) at the Look Rock, Tennessee, ground site. Atmos. Chem. Phys. Discuss. 15, 7365–7417 (2015).

45. Kourtchev, I., Copolovici, L., Claeys, M. & Maenhaut, W. Characterization of Atmospheric Aerosols at a Forested Site in Central Europe. Environ. Sci. Technol. 43, 4665–4671 (2009).

46. Hu, D., Bian, Q., Li, T. W. Y., Lau, A. K. H. & Yu, J. Z. Contributions of isoprene, monoterpenes, β-caryophyllene, and toluene to secondary organic aerosols in Hong Kong during the summer of 2006. J. Geophys. Res. Atmos. 113, 1–14 (2008).

47. Ding, X. et al. Tracer-based estimation of secondary organic carbon in the Pearl River Delta, south China. J. Geophys. Res. 117, D05313 (2012).

  33  

48. Riva, M., Budisulistiorini, S. H., Zhang, Z., Gold, A. & Surratt, J. D. Chemical Characterization of Secondary Organic Aerosol Constituents from Isoprene Ozonolysis in the Presence of Acidic Aerosol. Atmos. Environ. (2015). doi:10.1016/j.atmosenv.2015.06.027

49. Jacobs, M. I., Burke, W. J. & Elrod, M. J. Kinetics of the reactions of isoprene-derived hydroxynitrates: gas phase epoxide formation and solution phase hydrolysis. Atmos. Chem. Phys. Discuss. 14, 12121–12165 (2014).

50. Xu, L. et al. Effects of anthropogenic emissions on aerosol formation from isoprene and monoterpenes in the southeastern United States. Proc. Natl. Acad. Sci. 112, 37–42 (2015).

51. Dommen, J. et al. Photochemical production and aging of an urban air mass. J. Geophys. Res. 104, 5493 (1999).

52. Ng, N. L. et al. Secondary organic aerosol (SOA) formation from reaction of isoprene with nitrate radicals (NO3). Atmos. Chem. Phys. Discuss. 8, 3163–3226 (2008).