University of Southern California Dissertations and Theses

Development of novel techniques for evaluating physical, chemical and toxicological properties of particulate matter in ambient air

(USC Thesis Other)

Development of novel techniques for evaluating physical, chemical and toxicological properties of particulate matter in ambient air

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

DEVELOPMENT OF NOVEL TECHNIQUES FOR EVALUATING
PHYSICAL, CHEMICAL AND TOXICOLOGICAL PROPERTIES OF
PARTICULATE MATTER IN AMBIENT AIR

by

Dongbin Wang

A dissertation presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(ENVIRONMENTAL ENGINEERING)

August 2016

ii

Dedication

To my parents Gang and Yan,

for their unconditional love, trust and support in my life.

&

To my wife Xiaomeng,

for her persistent love, encouragement and confidence in me.

iii

Acknowledgements

The last five years of my doctoral degree studying has been a precious experience in my
life. Since joining the USC Aerosol Lab in 2011, I have worked with a number of great people
who have contributed significantly to the development of my research expertise and intellectual
maturity. I would like to take this opportunity to convey my sincere gratitude to them with my
humble acknowledgement.

First and foremost, I would like to express my sincere thanks to my graduate advisor,
Prof. Constantinos Sioutas, who gave me an opportunity of working under his guidance and his
whole hearted support, trust and encouragement. His truly scientist intuition has made him as a
lighthouse in the sea, guiding the young voyager of me to the proper direction in scientific
research. His keen scientific vision, creative discussions and enthusiasm have significantly
inspired and enriched my growth from a student to a mature scientist, as quoted from Sir Isaac
Newton: ―If I have seen a little further it is by standing on the shoulders of Giants.‖

I would also like to gratefully acknowledge the members of my guidance committee, Prof.
Caleb Finch and Prof. Ronald Henry, for taking the time to review this dissertation and for their
valuable suggestions on it. My special thanks to Prof. James Schauer and Dr. Martin Shafer at
University of Wisconsin-Madison for their extraordinary contributions in our collaborations in
research for this dissertation.

It is my great pleasure to work and study with a number of former colleagues at the group:
Dr. Payam Pakbin, Dr. Kalam Cheung, Dr. Neelakshi Hudda, Dr. Winnie Kam, Dr. Nancy Daher
and James Liacos. Without their self-giving assistance and help, my research and this
dissertation wouldn’t be a success. I would like to specially acknowledge Dr. Payam Pakbin,
Research Scientist at South Coast Air Quality Management District (SCAQMD), for his
mentorship during the early stages of my study. I also extend my thanks to the current colleagues
at the group: Sina Hasheminassab, Arian Saffari, Farimah Shirmohammadi, Mohammad Sowlat
and Giulia Simonetti for their valuable support during the past years.
iv

Last but not the least, I would like to acknowledge my parents, my wife, and all my other
family members for their unwavering love and support. They are my true motivations being able
to accomplish this work.

v

Table of Contents

Dedication ....................................................................................................................................... II
Acknowledgements ....................................................................................................................... III
List of Tables ................................................................................................................................ IX
List of Figures ................................................................................................................................ X
Abstract ...................................................................................................................................... XIV
Chapter 1 Introduction ................................................................................................................ 1
1.1 Background ...................................................................................................................... 1
1.2 Characteristic of particulate matter .................................................................................. 2
1.2.1. Particle size ............................................................................................................... 2
1.2.2. Particle mass ............................................................................................................. 2
1.2.3. Particle number ......................................................................................................... 3
1.3 Health effects of particulate matter .................................................................................. 3
1.4 Rationale for proposed research ....................................................................................... 4
1.4.1. Motivations ............................................................................................................... 4
1.4.2. Objectives ................................................................................................................. 7
1.5 Dissertation layout............................................................................................................ 8
Chapter 2 Development of a Two-Stage Virtual Impactor System for High
Concentration Enrichment of Ultrafine, PM
2.5
and Coarse Particulate Matter . 10
2.1. Introduction .................................................................................................................... 10
2.2. Methodology .................................................................................................................. 12
2.3. Results and discussion .................................................................................................... 16
2.3.1. Laboratory characterization of second-stage VI ..................................................... 16
2.3.2. System configuration test ........................................................................................ 17
2.3.3. Laboratory evaluation of the two-stage VI system ................................................. 18
2.3.4. Field evaluation of two stage VI system ................................................................. 21
2.4. Summary and conclusions .............................................................................................. 23
2.5. Acknowledgements ........................................................................................................ 24
Chapter 3 Macrophage Reactive Oxygen Species Activity of Water-soluble
and Water-insoluble Fractions of Ambient Coarse, PM
2.5
and Ultrafine
Particulate Matter (PM) in Los Angeles ................................................................. 25
3.1. Introduction .................................................................................................................... 25
vi

3.2. Methodology .................................................................................................................. 28
3.2.1. Sample collection .................................................................................................... 28
3.2.2. Sample analysis ....................................................................................................... 30
3.3. Results and discussion .................................................................................................... 32
3.3.1. Chemical composition of coarse, PM
2.5
and ultrafine PM total samples ................ 32
3.3.2. Macrophage ROS results ........................................................................................ 35
3.3.3. Correlation between ROS activity and chemical components ................................ 40
3.4. Summary and conclusions .............................................................................................. 45
3.5. Acknowledgements ........................................................................................................ 46
Chapter 4 Development and Evaluation of a High-Volume Aerosol-Into-Liquid
Collector for Fine and Ultrafine Particulate Matter ............................................. 47
4.1. Introduction .................................................................................................................... 47
4.2. Methodology .................................................................................................................. 50
4.2.1. Description of the system and its components ........................................................ 50
4.2.2. Laboratory evaluation tests of the impactor’s collection efficiency ....................... 51
4.2.3. Field tests of the aerosol-into-liquid collector and ambient PM sample collection 53
4.3. Results and discussion .................................................................................................... 55
4.3.1. Laboratory characterization of the impactor ........................................................... 55
4.3.2. Laboratory evaluation of the collection efficiency of the aerosol-into-liquid
collector ……………………………………………………………………………………..57
4.3.3. Field evaluation of continuous, unattended operation of the aerosol-into-liquid
collector ……………………………………………………………………………………..59
4.3.4. Chemical results for PM
2.5
samples ........................................................................ 62
4.4. Summary and conclusions .............................................................................................. 70
4.5. Acknowledgement .......................................................................................................... 71
Chapter 5 Development of a Technology for Online Measurement of Total and
Water-soluble Copper (Cu) in PM
2.5
....................................................................... 72
5.1. Introduction .................................................................................................................... 72
5.2. Methodology .................................................................................................................. 74
5.2.1. Ion Selective Electrodes (ISEs) .............................................................................. 74
5.2.2. Description of the system and its components ........................................................ 75
5.2.3. Laboratory evaluation of Cupric ISE and system performance .............................. 77
5.2.4. Field evaluation of Cu measurement system: ......................................................... 79
5.3. Results and discussion .................................................................................................... 80
5.3.1. Calibration of cupric ISE: ....................................................................................... 80
vii

5.3.2. Effect of ionic strength, temperature and pH of sample ......................................... 82
5.3.3. Collection and measurement of lab generated Cu(NO
3
)
2
aerosols ......................... 85
5.3.4. Field evaluations ..................................................................................................... 86
5.4. Summary and Conclusions ............................................................................................. 91
5.5. Acknowledgements ........................................................................................................ 92
Chapter 6 A New Technique for Online Measurement of Total and Water-soluble
Copper (Cu) in coarse particulate matter (PM) .................................................... 93
6.1. Introduction .................................................................................................................... 93
6.2. Methodology .................................................................................................................. 95
6.2.1. Description of the system and its components ........................................................ 95
6.2.2. Copper ISE .............................................................................................................. 98
6.2.3. Laboratory evaluation of modified BioSampler’s collection efficiency ................. 99
6.2.4. Field evaluation of Cu measurement system: ....................................................... 100
6.3. Results and discussion .................................................................................................. 101
6.3.1. Laboratory evaluation of collection efficiency of modified BioSampler ............. 101
6.3.2. Field evaluations ................................................................................................... 105
6.4. Summary and Conclusions ........................................................................................... 110
6.5. Acknowledgements ...................................................................................................... 111
Chapter 7 Development and Evaluation of a Novel Monitor for Online Measurement
of Iron, Manganese, and Chromium in Ambient Particulate Matter (PM) ...... 112
7.1. Introduction .................................................................................................................. 112
7.2. Methodology ................................................................................................................ 114
7.2.1. Reagents and Standards ........................................................................................ 114
7.2.2. Aerosol-Into-Liquid Collector .............................................................................. 115
7.2.3. Micro Volume Flow Cell (MVFC) ....................................................................... 116
7.2.4. System configuration and operation ..................................................................... 116
7.2.5. Laboratory evaluation tests ................................................................................... 119
7.2.6. Site description and field evaluation tests ............................................................. 120
7.3. Results and Discussion ................................................................................................. 121
7.3.1. Calibration............................................................................................................. 121
7.3.2. Laboratory aerosol generation tests ...................................................................... 122
7.3.3. Comparison between online measurements and offline sample results................ 124
7.3.4. Field deployment of the developed metal monitor ............................................... 126
7.4. Summary and conclusions ............................................................................................ 130
viii

7.5. Acknowledgements ...................................................................................................... 130
Chapter 8 Conclusions and recommendations for future research ..................................... 132
8.1. Conclusions .................................................................................................................. 132
8.2. Recommendations for future research.......................................................................... 133
8.2.1. Limitations of current studies ............................................................................... 133
8.2.2. Recommendations for future research .................................................................. 134
Bibliography: .............................................................................................................................. 136

ix

List of Tables

Table 2.1 Upstream and downstream the two-stage concentrator mass concentrations, mass
enrichment factors, and enrichment efficiency based on PSL tests ............................. 21
Table 3.1 Mass concentration for total, water-soluble and water-insoluble portions of
collected samples. ........................................................................................................ 37
Table 3.2 Correlation between chemical components of ambient PM with total, water-
soluble and water-insoluble ROS activity .................................................................... 41
Table 4.1 Collection efficiency of PSL particles and BC ............................................................. 58
Table 4.2 Particle recovery of BioSampler and aerosol-into-liquid collector .............................. 58
Table 4.3 Collection efficiency of polydisperse particles ............................................................. 59
Table 5.1 Cu measurements showing the necessity of adding ISA .............................................. 83
Table 5.2 Cupric ISE measurements under different pH values ................................................... 85
Table 6.1 Collection efficiency of PSL particles ........................................................................ 103
Table 6.2 Particle recovery of modified BioSampler ................................................................. 105
Table 7.1 Summary of the reagents used in different procedure for metal detections.
(Percentage values in parentheses represent volume ratio)........................................ 119
Table 7.2 Calibration results for different metals ....................................................................... 122

x

List of Figures

Figure 2.1 Schematic of the Two-Stage Virtual Impaction System ............................................. 13
Figure 2.2 Collection efficiency curves of the second-stage virtual impactor at different
intake flows ................................................................................................................. 17
Figure 2.3 (a-d) Particle size distributions for laboratory-generated polydisperse particles for
(a) ammonium sulfate, (b) adipic acid, (c) glutaric acid, (d) ammonium nitrate
(with minor flow of 1 L/min) upstream and downstream the 2-stage concentrator.
Physical properties of the size distribution including median, mean, mode, and
total number and mass concentrations are shown. The secondary y-axis is
multiplied by a factor of 100 (120 for ammonium nitrate) for visibility purposes. .... 19
Figure 2.4 Average enrichment efficiency in different size ranges. ............................................. 20
Figure 2.5 Particle size distributions of ambient PM upstream and downstream of the two-
stage virtual impactor system ..................................................................................... 22
Figure 2.6 Average enrichment factor of PAHs. .......................................................................... 23
Figure 3.1 System schematic of PM
2.5
and Ultrafine PM (UFP) collection ................................. 29
Figure 3.2 Chemical composition of PM samples. ....................................................................... 33
Figure 3.3 (a-b) ROS measurement results: (a) volume-based ROS; (b) mass-based ROS ......... 37
Figure 3.4 ROS comparison with previous studies ....................................................................... 40
Figure 3.5 (a-d) Correlation between selected chemical species and ROS: (a) Total ROS; (b),
(c) Water-soluble ROS; (d) Water-insoluble ROS ..................................................... 44
Figure 4.1 Schematic of impactor ................................................................................................. 51
Figure 4.2 System schematic for collection efficiency tests ......................................................... 53
Figure 4.3 Collection efficiency as a function of aerodynamic particle diameter of the
impactor ...................................................................................................................... 57
Figure 4.4 (a-c) Continuity test results and particle size distribution: a. Continuity test results
(Few gaps in downstream measurement is due to temporary malfunction of the
OPS monitor); b. Typical ambient particle size distribution at sampling site
during sampling period; c. Droplet size distribution upstream and downstream of
the impactor. ............................................................................................................... 61
Figure 4.5 (a-b) Inorganic ions (nitrate, ammonium and sulfate) comparison between filter,
BioSampler and impactor: a. between BioSampler and aerosol-into-liquid
collector. The p value from t-test for nitrate, ammonium and sulfate is 0.35, 0.85
xi

and 0.8, respectively; b. between filter and aerosol-into-liquid collector. The p
value from t-test for nitrate, ammonium and sulfate is 0.89, 0.97 and 0.31,
respectively. Error bars represent the standard deviation of multiple samples. The
sulfate (SF-ICPMS) results in Figure 4.5(a) represent the estimated sulfate
concentrations based on the sulfur (S) measurement results by SF-ICPMS
analysis. ...................................................................................................................... 63
Figure 4.6 TOC and WSOC comparison between BioSampler and aerosol-into-liquid
collector. The p value from t-test for TOC and WSOC is 0.66 and 0.73,
respectively. Error bars represent the standard deviation of multiple samples. ......... 65
Figure 4.7 (a-b): Metals and trace elements comparison between filter, BioSampler and
impactor samples: a. between filter and aerosol-into-liquid collector. The p value
from t-test is 0.71; b. between BioSampler and aerosol-into-liquid collector. The
p value from t-test is 0.94. Error bars represent the standard deviation of multiple
samples. Note: inserted plot shows the correlations on log-scale. The line Y=X is
also included for visibility purposes. .......................................................................... 67
Figure 4.8 (a-b) ROS activity comparison between filter, BioSampler and aerosol-into-liquid
collector: a. between BioSampler and aerosol-into-liquid collector. The p value
from t-test for unfiltered and filtered slurry is 0.75 and 0.68, respectively; b.
between filter extractions, filtered and unfiltered aerosol-into-liquid collector
slurries. The p value from t-test between filter extraction and filtered slurry is
0.71. Error bars represent the standard deviation of multiple samples. ...................... 69
Figure 5.1 Schematic of Cu measurement system ........................................................................ 77
Figure 5.2 Calibration curve of cupric ISE. The top x-axis represents the Cu concentration in
standard solution. The bottom x-axis represents the corresponding airborne Cu
concentration based on a sampling flow rate of 200 L/min and a slurry collection
rate of 4 ml per hour. Error bars represent the standard deviations of multiple
calibration tests (at least 5 tests were performed) ....................................................... 82
Figure 5.3 Effect of sample temperature on Cu measurements. Standard solution of 100 ppb
was used as sample solution. Error bar represents the standard deviation of
multiple reading (3 readings for each data point). ...................................................... 84
Figure 5.4 Cu generation test results. Error bar represents the standard deviation of multiple
measurements (3 measurements for each sample)...................................................... 86
Figure 5.5 (a-b): Metals and trace elements comparison between filter and slurry samples: a.
total metal and trace elements; b. water-soluble metal and trace elements. Error
bars represent the standard deviation of multiple samples (7 sets of samples).
Note: inserted plot shows the correlations on log-scale. ............................................ 88
Figure 5.6 (a-b): Metals and trace elements comparison between cupric ISE and ICPMS
results: a. total metal and trace elements; b. water-soluble metal and trace
xii

elements. Error bars represent the standard deviation of multiple measurements
(3 measurements for each sample). ............................................................................ 90
Figure 5.7 Field continuous operation test results (Gap was due to temporary system
maintenance). Error bars represent the standard deviation of multiple
measurements (3 measurements for each sample)...................................................... 91
Figure 6.1 Schematic of the online Cu measurement system for coarse PM ............................... 97
Figure 6.2 Schematic of the modified BioSampler ....................................................................... 98
Figure 6.3 Comparison of the BioSampler collection efficiency tests with a previous study
by Willeke et al (Willeke et al., 1998) ...................................................................... 104
Figure 6.4 (a-b): Coarse PM metals and trace elements comparison between filter and slurry
samples by SF-ICPMS: a. total metal and trace elements; b. water-soluble metal
and trace elements. Data in these figures represent the average of different
samples. Error bars represent the standard deviation of multiple samples. Copper
is indicated by arrow in both figures. (Note: dash line represents Y=X line) .......... 107
Figure 6.5 Coarse PM Cu measurements comparison between copper ISE and SF-ICPMS
results. Error bars represent the standard deviation of multiple measurements.
(Note: Dash line represents the Y=X line) ............................................................... 108
Figure 6.6 Field continuous total Cu measurements VS. dominate wind direction. .................. 109
Figure 6.7 Comparison of the average diurnal concentrations of coarse PM Cu to those
reported previously in the same sampling location by Cheung el al. (2011) in
winter 2010 ............................................................................................................... 110
Figure 7.1 Schematic of the developed on-line metal monitoring system. ................................. 118
Figure 7.2 (a-b) Lab aerosol generation and collection results: a) Fe tests; b) Mn tests. Error
bars correspond to one standard deviation of multiple tests. .................................... 123
Figure 7.3(a-c): Comparison between online metal measurements and off-line parallel
samples analyzed by SF-ICPMS analysis for a) Fe; b) Mn; and c) Cr. Error bars
represent one standard deviation of multiple online measurements during parallel
sample collection period. Data in black color represent slurry samples analyzed
by SF-ICPMS, whereas data in grey represent filter samples. Dashed line
represents Line Y=X. ................................................................................................ 126
Figure 7.4(a-c): Continuous online metal measurements during field deployment: a) Fe
measurements; b) Cr measurements; c) Mn measurements. Few gaps in Cr and
Mn measurements are due to data points below detection limit. .............................. 128
xiii

Figure 7.5(a-c): Near-continuous (2 hour) measurements of total and water-soluble
concentrations of target metals: (a) Fe measurements; (b) Mn measurements; and
(c) Cr measurements. ................................................................................................ 130
Figure 7.6 Field measurements of oxidation states of Fe ........................................................... 130

xiv

Abstract

There has been a large body of epidemiological and toxicological studies indicating the
strong associations between exposures to ambient particulate matter (PM) and adverse health
outcomes. Some specific chemical component in ambient PM, including but not restricted to
airborne transition metals, has been hypothesized to be mostly responsible for generating excess
cellular oxidation stress and eventually result in PM induced health risks. Due to the complex
physical property and chemical composition of ambient PM, assessing which PM constituents
are linked to adverse health outcomes as well as the exact mechanisms leading to these outcomes
remains an active topic of research. Therefore, novel technologies for characterizing ambient PM
with high time resolutions are in great needs, which will assist in investigations of the physical
properties and chemical composition of PM, as well as enabling a better understanding of
toxicological properties of ambient PM.

This dissertation focuses on the development and evaluations on novel techniques in
determining the physical, chemical and toxicological properties of ambient PM. As the first part
of this dissertation, a two-stage particle concentration enrichment system was developed to
provide highly concentrated particles at low flow rate (i.e. 1.5 L/min). This system can enrich
particle concentration by a factor roughly of 100-120 without altering their physical and
chemical properties. Secondly, as a principle investigation of the new PM collection technology,
the relative contributions of water-soluble and water-insoluble portions of ambient PM to cellular
redox activity were investigated. Results from this study indicated that both water-soluble and
water-insoluble portions of PM played important roles in influencing potential cellular toxicity.
Next, a novel Aerosol-Into-Liquid Collector was developed to provide concentrated slurries of
fine and/or ultrafine PM, in which both water-soluble and water-insoluble components were well
preserved in the collected slurry samples. This new aerosol collection system could achieve an
excellent collection efficiency (over 90%), and has the unique ability to be continuously operated
unattended for at least 4 to 5 days without any obvious shortcomings in its operation. Following
the successfully development of Aerosol-Into-Liquid Collector, this new PM sampler was further
developed into a novel monitor for online, in-situ measurement of copper (Cu) in ambient fine
PM. Evaluations of the Cu monitor indicates a very good agreement for total and water-soluble
xv

Cu concentrations obtained online by this monitor, with measurements performed by inductively
coupled plasma mass spectrometry (ICP-MS) as a reference method, suggesting the excellent
performance of this Cu monitor in aspect of collection efficiency and measurement accuracy.
This technology is then extended to coarse PM by utilizing two virtual impactors combined with
a modified liquid impinger (BioSampler) as PM collector. Lastly, a prototype atmospheric
aerosol monitor was developed and evaluated for online measurement of other three
toxicologically relevant redox-active metals (Fe, Mn, and Cr) in ambient PM
2.5
based on the
developed instrument described in the previous parts of this dissertation.

The novel techniques developed in this dissertation will greatly advance the capabilities
of atmospheric pollution monitoring to understand the physical properties, as well as the
chemical and toxicological active components of ambient PM. Moreover, such technologies will
provide significant insights on developing a better understanding of the sources, formation
mechanisms, and transport of PM in the atmosphere. Ultimately, air pollution monitoring goals
can be more directly linked with the protection of public health. More effective and targeted
control strategies to better protect human health can thus be implemented.

1 | P a g e

Chapter 1 Introduction
1.1 Background
Particulate matter (PM), also known as aerosols, is defined as a suspension of liquid
droplet and/or solid particles in the atmosphere. Generally ambient PM are conglomerates of
many pollutant subclasses, potentially comprising of different organic and inorganic species.
Particulate matter is one of the extremely important criteria air pollutants, which is listed under
the 1990 Clean Air Act Amendment, together with ozone (O
3
), sulfur dioxide (SO
2
), nitrogen
oxides (NO
X
), lead (Pb) and carbon monoxide (CO) (USEPA, 1990). PM contributes to
formation of smog and visibility degradation, and also influences the radiative forcing by
changing the amount of heat reaching the ground (Seinfeld and Pandis, 2006). More importantly,
numerous health studies have demonstrated that exposure to atmospheric particulate matter (PM)
has strong associations with significant impacts on human health (Campbell et al., 2005; Davis et
al., 2013; Delfino et al., 2005; Li et al., 2003; Pope et al., 2004; Wilhelm and Ritz, 2005).
Therefore, understanding the physical, chemical and toxicological characteristics of ambient PM
is significantly essential in allowing for more effective regulatory control strategies, more
targeted air quality standards, and ultimately, reductions in population exposure to harmful types
of airborne PM.

Typically, ambient PM are directly emitted from different type of sources (defined as
primary aerosol), or formed by gas-to-particle partition process (defined as secondary aerosol)
during photo-chemical reaction in the atmosphere. Both natural and anthropogenic sources
contribute to primary aerosol in ambient air. Natural sources include wood smoke, volcanic
activities, sea spray and etc. (Seinfeld and Pandis, 2006), while anthropogenic sources refers to
vehicular emission, industrial source, combustion of fossil fuels and etc. which are related to
human activities. On the other hand, once PM and other type of primary pollutants are emitted
into atmosphere, they also participate in gas-to-particle conversion and form the secondary
pollutant, for example the oxides of SO
2
and NO
X
, sulfate and nitrate particles, respectively
(Robinson et al., 2007). For both primary and secondary aerosols, the chemical composition of
them is very complex and highly dependent on local pollution sources and meteorological
conditions, both of which may vary in short period of time (Cheung et al., 2012; Fine et al., 2004;
Pancras et al., 2013; Petit et al., 2015; Salameh et al., 2015). The size and composition of
2 | P a g e

ambient PM can also be changed via different physical and chemical processes such as vapor
condensation, evaporation, coagulation, and chemical reactions (Hinds, 2012).

1.2 Characteristic of particulate matter
1.2.1 Particle size
Particulate matter in ambient air is usually classified by their aerodynamic diameter
rather than their actual shape. The size of particles is one of the most influential characteristic of
particles in context of their behavior, source, and health outcome. Generally there are three major
size modes of PM according to their aerodynamic diameter: (a) the coarse mode, which refers to
particles with a diameter between 2.5 and 10 µ m; (b) the accumulation mode (also known as fine
mode), which refers to particles with a diameter between 0.1 and 2.5 µ m; (c) the ultrafine mode,
which refers to particles with a diameter less than 0.1 µ m. Typically, ultrafine mode dominates
the number distribution while coarse and accumulation modes contribute to majority of total
PM
10
mass concentration in the atmosphere. These three size modes also differ in their sources,
formation mechanism, chemical composition, as well as their lifetime in the ambient air. Coarse
PM are generated primarily by mechanical process such as grinding, abrasion and re-suspension
by wind. Due to the larger size, they are generally removed by gravitational settling, resulting a
relative short lifetime in the atmosphere. Fine mode particles contribute to a major amount of PM
mass concentration and especially the surface area. These particles are formed by primary
emission from different type of sources, and coagulation or condensational growth of ultrafine
particles. Fine particles tend to express longer lifetime in atmosphere because its removal
mechanism is neither dominated by gravitational settling or diffusion processes. Lastly, ultrafine
particles are primarily originated by combustion sources and gas-to-particle formation processes
in ambient air. Giving the fact that they are generally very small in size, ultrafine particles are
mostly removed by physical processes such as diffusion or coagulation.

1.2.2 Particle mass
PM mass a significant parameter in terms of air quality regulation and ambient PM
measurements. US EPA has established PM
10
mass concentration standard since 1987 because
these particles are considered to be the respirable fraction of suspended particles in the
atmosphere. In 1997, a legislation of PM
2.5
mass concentration was also carried out. Ultrafine
3 | P a g e

particles are generally very small in size and they contribute to a relative small fraction of PM
mass concentration in the atmosphere.

1.2.3 Particle number
Particle number is another important parameter especially for ambient particle
measurements. As opposite to particle mass concentration, the majority of ambient particle
number concentration is contributed by ultrafine particles. Given the fact that currently there is
no regulation standard for ultrafine PM in the United States, measurement of particle number
concentration may provide valuable information in evaluating the diurnal and seasonal trends of
ambient ultrafine particles, as well as in studies about indoor air quality.

1.3 Health effects of particulate matter
For the last few decades there has been significant concern over adverse health outcomes
introduced by elevated level of particulate matter. It has become the most prominent derives and
motivations for aerosol researches. A number of health studies have demonstrated that exposure
to high concentrations of PM has been linked to respiratory diseases (Dominici et al., 2006;
Jansen et al., 2005; Li et al., 2003), cardiovascular diseases (Delfino et al., 2005; Dominici et al.,
2006; Pope et al., 2009), neurodegenerative disorders (Davis et al., 2013; Gatto et al., 2014), and
adverse birth outcomes (Ritz and Wilhelm, 2008; Sapkota et al., 2010; Wilhelm and Ritz, 2005).
A broadly accepted hypothesis is that PM-induced toxicity is driven by the interaction of PM
with cells and macrophages to generate reactive oxygen species (ROS), which change the redox
status of the cells (Araujo et al., 2008; Li et al., 2008; Risom et al., 2005). There is increasing
evidence that PM characteristics (e.g. size and chemical composition) are important factors in
mediating this cellular oxidative stress (Cho et al., 2005; Valavanidis et al., 2008; Xia et al.,
2006). Therefore, understanding sources, atmospheric transformations and aging of ambient PM
allows for more effective regulatory control strategies, more targeted air quality standards, and
ultimately, reductions in population exposure to harmful types of airborne PM.

Particle size, governing the site of particle deposition along the respiratory tract, has an
important role in health effects induced by PM exposure. Studies have shown that the ultrafine
mode particles may be more toxic than larger PM size fractions such as coarse mode and
4 | P a g e

accumulation mode (Cho et al., 2005; Kleinman et al., 2008; Li et al., 2003). However, the
composition of ambient PM is very complex and highly dependent on a number of factors
including regional meteorology, topography, and natural (i.e. sea-salt) and anthropogenic (i.e.
vehicular emissions) source strengths of PM, all of which may vary in short periods of time (Fine
et al., 2004; Salameh et al., 2015). For instance, urban ambient PM components usually include
metals, trace elements, black carbon, inorganic salts and a large number of organic species
(Daher et al., 2013; Putaud et al., 2010). Due to the complex composition of ambient PM,
assessing which PM constituents are linked to adverse health outcomes as well as the exact
mechanisms leading to these outcomes remains an active topic of research. Therefore, ambient
PM measurement in high time resolution becomes critical to identify the toxicologically relevant
PM species, as well as the sources of these species.

1.4 Rationale for proposed research
1.4.1 Motivations
Conventional PM sampling methodologies usually involve collections of particles onto
substrates by filtration or inertial impaction. These sampling methodologies have been widely
used in aerosol studies and are regarded as referenced methodologies. After time integrated PM
collection, the quantitative composition measurements are usually performed offline by
extracting the collected particles using different types of solvents (e.g., water, methanol and etc.)
and analyzed by various analytical techniques, such as ion chromatography (IC), inductively
coupled plasma mass spectrometry (ICP-MS) and carbonaceous analysis. Although these
sampling methodologies have been widely adopted to-date, they are associated with a variety of
intrinsic drawbacks, such as long turn-around time for processed results of chemical analysis,
long sampling intervals for collection of sufficient mass for subsequent chemical analysis (hours
to days, depending on sampling flow rate and ambient concentrations), incomplete extraction of
insoluble PM-bound species, potential sampling artifacts (e.g., losses of labile species such as
ammonium nitrate and semi-volatile organic carbon (SVOC)) (Eatough et al., 2003; Schauer et
al., 2003) and/or changes in chemical speciation (e.g. oxidation state shifts in redox active metals)
due to prolonged sampling time (Eatough et al., 2003; Turpin et al., 1994). These drawbacks may
significantly limit the feasibility and accuracy of conventional sampling approaches in
investigating chemical composition and toxicological properties of ambient PM.
5 | P a g e

A number of advanced aerosol sampling technologies have been developed in recent
years to overcome these limitations in conventional methodologies. In these newer technologies,
ambient PM is directly collected as an aqueous suspension; for instance PM collection onto a
filter that is periodically washed by water (Buhr et al., 1995; Phan and McFarland, 2004),
impaction into a flowing liquid (Karlsson et al., 1997), collection by a water cyclone (Orsini et
al., 2008) and the combination of particle growth by water vapor condensation and impaction on
surfaces covered by liquid flow (Weber et al., 2001). These technologies effectively eliminate
the need for elaborate extraction procedures and have been well documented in applications of
ambient PM monitoring. As an alternative methodology, a novel system as Versatile Aerosol
Concentration Enrichment System (VACES) was also developed (Kim et al., 2001a, 2001b). In
VACES, ambient PM is first grown to super-micrometer size droplets via condensational growth,
and subsequently concentrated by virtual impactor (VI). In tandem with a liquid impinger PM
collector (BioSampler, SKC West, Inc., Fullerton, CA), VACES can directly collect both fine
and ultrafine particles in an aqueous suspension. Moreover, the highly concentrated aerosol that
VACES produced can enhance the signal-to-noise ratios of on-line monitors (Khlystov et al.,
2005; Zhao et al., 2005), or used in applications of in molecular and cellular in-vitro toxicity
assays and real-time in-vivo exposures (Kleinman et al., 2008; Morgan et al., 2011). A similar
system was also developed by Kidwell and Ondov (2001). Daher et al. (2011) compared
sampling techniques of filtration, impaction and VACES/BioSampler tandem and results
suggested that in VACES/BioSampler tandem, both water-soluble and water-insoluble portions
of ambient PM were essential for measurements of PM redox properties.

Despite the robust PM collection performance of the VACES/BioSampler technology,
several drawbacks still exists that hinder its use for high volume PM collection. Firstly, the
VACES system can concentrate particles generally by 20-30 times of their ambient PM
concentrations. Such enrichment factor might be insufficient in some cases for the areas of direct
cellular exposures (Savi et al., 2008; Volckens et al., 2009), which are currently limited by the
amount of PM material used in the exposure, and for some applications requiring highly
concentrated particles especially at low flow rates (on the order of 1-2 L/min). Moreover, long
period of operating the VACES associates with other type of drawbacks, including excess water
6 | P a g e

accumulation, clogging of the virtual impactor acceleration and especially collection nozzles,
clogging of BioSampler nozzles and the limitation collector volume (20 mL) of the conventional
BioSampler, which requires periodic removal of the collected suspension samples. As a result,
the operation of that system requires routine field supervisions, which limits its unattended
operation in long periods of time.

Refined health studies require the development of online measurement techniques that
can provide high-time-resolution and high accuracy data. Several online systems involving
automated PM bulk composition measurements have been recently developed. These techniques
have been successfully used to measure PM bulk composition such as organic carbon (Sullivan
et al., 2004) and ionic species, including nitrate, sulfate and ammonium (Orsini et al., 2003;
Takegawa et al., 2005; Weber et al., 2001). These chemical species are quite abundant in
ambient aerosols, allowing their measurement with traditional analytical detection tools (e.g. ion
chromatography, total organic carbon analyzer). However, given typical ambient concentrations
and due to the small analytic mass/concentration, these approaches are generally not adequate to
measure metals and elements in trace levels (such as Cu, Cr, Fe and Mn), all of which have been
implicated in inducing oxidative stress (Shafer et al., 2010). Past efforts to develop field-
deployable, higher time resolution tools for characterization of the elemental composition of
ambient PM have focused on the technologies of aerosol time-of-flight mass spectrometry
(ATOFMS) and X-ray fluorescence (XRF). However, ATOFMS lacks the sensitivity to detect
several important metals in ambient air and is semi-quantitative at best for measurement of total
metals only (Gard et al., 1997). Another commercially available technique for semi-continuous
measurement of metals/elements in ambient PM is based on X-ray fluorescence (XRF) (Park et
al., 2014). Though capable of detecting a number of major and minor elements in filter-collected
PM, the XRF technique provides information on total element levels only - it is insensitive to
oxidation state and clearly is unable to distinguish between potentially soluble species and
insoluble species. Recently, the spectrophotometry approach has been incorporated into a few
online systems in which ambient PM are directly collected as slurry samples, and metal
concentrations in the slurry are sequentially measured. For example, Rastogi et al. (2009)
developed a prototype online system for detecting water-soluble Fe (II) in ambient PM
2.5
, while
Khlystov and Ma (2006) reported measurements of Cr (VI) and Cr (III) in ambient PM using a
7 | P a g e

new online system. There is still great need, however, to measure the speciation, in particular
oxidation state, of important redox active metals (i.e. Fe, Mn and Cr), as the biological pathways
underlying the adverse health effects of ambient PM are speciation-driven. Neither the XRF nor
ATOFMS techniques can provide this speciation information.

1.4.2 Objectives
Improving our understanding of the processes governing PM chemistry and the
association of real-world PM exposure with adverse human health effects, requires novel
approaches to overcome the limitations of conventional sampling methods (both in terms of
detection limit and time resolution) and efficiently integrate PM physical property and chemical
composition measurement, which best represents real-world PM exposure. The overarching
objectives of this dissertation were therefore to:

 To develop an extension form of the VACES system to achieve higher concentration
enrichment under low flow rate;
 To investigate the toxicological contribution of water-soluble and water-insoluble
portions of ambient PM, in order to illustrate the advantages of collecting PM directly as
slurry samples comparing to conventional sampling;
 To develop a novel technique to collect ambient PM directly into liquid suspension with
the capability to be operated unattended for long time period;
 To develop a series of online monitors to measure the important metal species (e.g. Cu,
Fe, Mn and Cr) which potentially related to the toxicological activities of ambient PM.

Findings from this dissertation will provide significant insights on advanced PM
measurement technologies to overcome the potential limitations in current techniques, and
elucidate the physical and chemical properties of ambient PM that lead to PM-induced toxicity.
Moreover, these findings will tend to create the foundation for transformative changes in air
pollution monitoring that will more directly connect monitoring goals with the protection of
public health. The advances in monitoring technology will include the development and
demonstration of technologies to measure the total and water-soluble fraction of metals in PM, as
well as their oxidation states, which is the biologically relevant fraction of PM for many
8 | P a g e

toxicological pathways. The deployment of these tools for online sampling and analysis will
allow the investigation of diurnal/temporal changes in PM components that is expected to change
with sources and atmospheric redox cycling. These advances, in themselves, will provide
significant insights on chemical reactions of aerosol during atmospheric transport, as well as
controls on PM induced toxicity.

1.5 Dissertation layout
This dissertation presents my doctoral research work under the supervision of Professor
Costantinos Sioutas, with the goal to develop a series of novel techniques for ambient PM
collections and characterizations. The thesis includes the following chapters:

Chapter 1 provides an overview of urban particulate matter regarding their properties and
particle related health effects. The rationale of this dissertation is also identified and discussed.

Chapter 2 describes the development and evaluations of a Two-Stage Virtual Impactor
(VIs) system to achieve high concentration enrichment. This Two-Stage VIs system is an
extension of conventional VACES configuration and can achieve high particle concentration
enrichments by a factor of 100-120, without altering their physical and chemical properties.

Chapter 3 investigates of the relative contributions of water-soluble and water-insoluble
portions of ambient PM to cellular redox activity as measurement of macrophage reactive
oxygen species (ROS) activity, which suggests that collection of particles directly into a liquid
suspension for toxicological analysis may be superior to conventional methodologies.

Chapter 4 describes the development and evaluation of a novel Aerosol-Into-Liquid
Collector, developed to provide concentrated slurries of fine and/or ultrafine PM to be used for
unattended, in-situ measurements of PM chemistry and toxicity. This system operates at 200
L/min flow, and is capable for continuous and unattended collection of concentrated suspensions
for at least several days without any obvious shortcomings in its operation.

9 | P a g e

Chapter 5 and 6 introduce novel monitors for online, in-situ measurement of copper (Cu)
in ambient PM. Chapter 5 described the new Cu monitor for ambient PM
2.5
, in which particles

are collected as concentrated slurries in the Aerosol-Into-Liquid Collector, and the Cu
concentration in slurry samples is subsequently determined by a cupric Ion Selective Electrode
(ISE). This technology is then extended to coarse PM as presented in Chapter 6, by utilizing two
virtual impactors combined with a modified BioSampler as PM collector.

In Chapter 7, a prototype atmospheric aerosol monitor was developed for online
measurement of three toxicologically relevant redox-active metals (Fe, Mn, and Cr) in PM
2.5
.
Similar as the PM
2.5
Cu monitor, this new metal monitor for Fe, Mn and Cr measurements
utilizes the Aerosol-Into-Liquid Collector as PM collection module. The concentrations of target
metals in the collected slurries are subsequently measured in a Micro Volume Flow Cell (MVFC)
using spectrophotometry to quantify the light absorption of colored complexes resulting from the
reaction between the target metals and added analytical reagents.

Finally, Chapter 8 provides a conclusion of this dissertation and recommendations for
future developments and investigations.

10 | P a g e

Chapter 2 Development of a Two-Stage Virtual Impactor System
for High Concentration Enrichment of Ultrafine, PM
2.5
and Coarse Particulate Matter

This chapter is based on the following publication:
Wang, D., Kam, W., Cheung, K., Pakbin, P., & Sioutas, C. (2013). Development of a two-stage
virtual impactor system for high concentration enrichment of ultrafine, PM
2.5
, and coarse
particulate matter. Aerosol Science and Technology, 47(3), 231-238.

2.1. Introduction
A number of health studies have demonstrated that exposure to high concentrations of
PM has been linked to respiratory diseases (Li et al., 2003a), cardiovascular diseases (Delfino et
al., 2005b), neurodegenerative disorders (Campbell et al., 2005; Morgan et al., 2011), and
adverse birth outcomes (Ritz and Wilhelm, 2008). Studies have also shown that the ultrafine
fraction (defined as particles having a physical and/or aerodynamic diameter dp < 100-200 nm)
may be more toxic than larger PM size fractions (Cho et al., 2005; Kleinman et al., 2008; Li et al.,
2003b). In addition, the composition of PM is highly complex and variable as it depends on a
number of factors including regional meteorology, topography, and natural (i.e. sea-salt) and
anthropogenic (i.e. vehicular emissions) source strengths of PM. Urban ambient PM components
include metals, trace elements, black carbon, inorganic salts and a large number of organic
species (Cheung et al., 2011). Due to the complex composition of ambient PM, assessing which
PM constituents are linked to adverse health outcomes as well as the exact mechanisms leading
to these outcomes remains an active topic of research. To that end, recent advances in particle
concentration techniques have become effective tools in facilitating health studies (Demokritou
et al., 2003, 2002; Li et al., 2009; Rastogi et al., 2012) and in assisting ambient particle sampling
(Geller et al., 2005).

The concentration enrichment of PM is achieved by means of virtual impactors (VIs), the
operation of which is based on inertial separation of particles above and below a designated
cutpoint. Particles larger than the VI cutpoint will cross the air streamlines and entrain the minor
11 | P a g e

flow, while particles smaller than the cutpoint will follow the deflected streamlines and exit
through the major flow. Hence, particles larger than the cutpoint are enriched by a theoretical
factor equivalent to the intake-to-minor flow ratio. The design characteristics of VIs, such as
separation efficiency vs particle aerodynamic diameter and internal wall losses have been
extensively examined in both theoretical (Marple and Chien, 1980) and experimental studies
(Chen and Yeh, 1987). Because of the adverse health effects associated with fine PM (dp < 2.5
μm), low cutpoint (0.1-0.25 μm) round nozzle or slit nozzle VIs were developed in the mid-90s
(Sioutas et al., 1999). However, ultrafine particles (dp < 0.1-0.2 μm) were not effectively
concentrated by these technologies.

To expand the abilities of VIs to concentrate ultrafine PM, the Versatile Aerosol
Concentrator Enrichment System (VACES) was developed, capable of concentrating both fine
and ultrafine particles (Kim et al., 2001a, 2001b). In the VACES, the particles are grown to
super-micrometer size droplets via condensational growth, and subsequently concentrated by
virtual impaction. Following concentration enrichment, particles pass through diffusion dryers
where excess moisture is removed and they are returned to their original size. Previous studies
have demonstrated that particle enrichment through the VACES does not have significant effects
on the physical properties (i.e. mass and number concentration) and chemical composition of the
particles (Khlystov et al., 2005; Zhao et al., 2005). In addition to time-integrated filter sampling,
applications of this technology have included uses in molecular and cellular in-vitro toxicity
assays (Li et al., 2003; Xia et al., 2006), real-time in-vivo exposures (Kleinman et al., 2008), and
direct PM collection in aqueous solutions for chemical and toxicological analysis (Daher et al.,
2011a). An advantage of using VACES for in-vitro exposure is the reduction in sampling time,
which favors cell viability and exposure characterization (Sillanpä ä et al., 2008). The VACES
can also be coupled with a heater, in which semi-volatile particle-bound PM species are
partitioned to gas phase, to be used for exposure studies separately to non-volatile and semi-
volatile components of PM (Pakbin et al., 2011). In addition to its applications for health studies
and PM sampling efforts, high concentrations of PM can also enhance signal-to-noise ratios of
on-line monitors including the Aerodyne Aerosol Mass Spectrometer (AMS) (Khlystov et al.,
2005) and the rapid single-particle mass spectrometer (RSMS-3) (Zhao et al., 2005).

12 | P a g e

The VACES can concentrate particles at most by 20-30 times their ambient PM
concentrations. The current study represents an extension of the VACES capabilities by placing
two virtual impaction stages in series, yielding theoretical PM enrichment factors of up to 130.
The resulting highly concentrated particles produced by the modified VACES broaden the scope
of PM exposure studies, including recent and exciting advances in the areas of direct cellular
exposures, which are currently limited by the amount of PM material used in the exposure, and
facilitate ambient PM sampling efforts, by improving the signal-to-noise ratios of many on-line
monitors. This manuscript will discuss the laboratory and field validation tests of the two-stage
VACES system.

2.2. Methodology
The two-stage VI system is an extension of the previously developed VACES system
(Kim et al., 2001a; Kim et al., 2001b). A second-stage miniature VI was placed in series after the
first-stage VI to provide particles sequentially enriched through each stage. The system was first
deployed and tested using stable laboratory particles, which were generated by a HOPE nebulizer
(B&B Medical Technologies, Carlsbad, CA). Particle mass and number concentrations as well as
particle size distributions were measured before and after enrichment, using continuous monitors.
Upon completion of the lab evaluation, the system was moved in the field and its concentration
enrichment performance was validated using ambient PM.

The system schematic is presented in Figure 2.1. The air stream was first drawn through
the saturator tank filled with high performance liquid chromatography (HPLC) water, maintained
at 35 ° C, to saturate the air with ultrapure water vapor. The particles leaving the saturator at
about 27-28 ° C entered two condensational tubes that were connected to a circulating chiller to
achieve super-saturation and promote condensational particle growth. The flow rate through each
condensational tube was set at 100 L/min; particles were generally grown to 3-4 µ m aqueous
droplets, as measured by an Aerosol Particle Sizer (APS 3320, TSI Inc., Shoreview, MN) and as
reported in previous studies (Kim et al., 2001a; Kim et al., 2001b). The particle droplets then
passed through the first-stage VI in each line, which had a major flow rate of 100 L/min and a
minor flow rate of 7.5 L/min (the determination of the optimum minor flow rate will be
discussed in the next section). The two minor flows of the first-stage VIs were combined to 15
13 | P a g e

L/min to form the total intake flow of the second-stage VI, which operated with a minor flow of
1.5 L/min. Following the second-stage VI, concentrated particles were dried with a silica gel
diffusion dryer (Model 3062, TSI Inc., Shoreview, MN) and returned to their original size for
characterization and evaluation.

Figure 2.1 Schematic of the Two-Stage Virtual Impaction System

The first part of the laboratory tests was to characterize the second-stage VI. Different
sizes of monodisperse fluorescent polystyrene latex (PSL) particles (Polyscience Inc.,
Sacramento, CA) were used to determine the 50% cutpoint of the VI. PSL particle sizes of 1, 1.5,
3 and 6 µm were generated by a nebulizer and the PM mass concentrations before and after the
two-stage VI system were continuously measured with a DustTrak DRX Aerosol Monitor
(Model 8534, TSI Inc., Shoreview, MN). Different intake flows of 10, 15 and 30 L/min were
tested with a minor flow fixed at 1.5 L/min.

14 | P a g e

The second part of the laboratory tests was to determine the optimal enrichment
performance of the system. Different configuration tests were performed by adjusting the
intermediate flow (see next section), defined as the combined minor flows of the first-stage VIs
or the intake flow of the second-stage VI. In this test, polydisperse glutaric acid particles were
generated and the number concentration was measured both upstream and downstream of the
system with a Condensation Particle Counter (CPC 3022A, TSI Inc., Shoreview, MN). Number
concentration enrichment factors (i.e., the ratio of downstream to upstream values) measured by
CPC were then compared to the theoretical enrichment factors to determine the enrichment
efficiency. Intermediate flow rates of 10, 15 and 30 L/min were tested.

In addition, several other types of particles were generated by atomization and the
particle size distribution before and after concentration enrichment by the system was measured.
These included polydisperse particles of ammonium sulfate, ammonium nitrate, adipic acid and
glutaric acid. Ammonium sulfate and ammonium nitrate were chosen because they represent
predominant inorganic species in PM
2.5
in most areas of the U.S (Malm et al., 2004). In addition,
ammonium nitrate was chosen because it is a semi-volatile species (Chang et al., 2000;
Mozurkewich, 1993). Adipic acid and glutaric acid were selected because they are typical
organic dicarboxylic acids formed by ozone photo-oxidation (Cruz and Pandis, 1999; Saleh et al.,
2012). A Scanning Mobility Particle Sizer (SMPS 3936, TSI Inc., Shoreview, MN) coupled with
a CPC was used to measure the particle size distribution upstream and downstream of the system.
The SPMS/CPC was operated at sheath/particle flow rates of 3/0.3 L/min, covering a particle
diameter range of 13-700 nm. The minor flow of the system was set to 1.5 L/min for ammonium
sulfate, adipic acid and glutaric acid and 1 L/min for ammonium nitrate.

In addition to polydisperse particles, a series of monodisperse polystyrene latex (PSL)
particles were generated to further evaluate the enrichment performance of the two-stage VI
system. This series of tests was conducted to confirm that the concentration enrichment is not
affected by a monodisperse particle size distribution as well as for ascertaining that the two-stage
system is capable of concentrating coarse PM equally effectively. These tests also represent a
theoretically greater challenge for this system, since PSL particles are hydrophobic and thus
more difficult to be activated and grown by super saturation of water vapor. PSL particles of 100,
15 | P a g e

300, 400 nm and 1, 3, 6 µm were generated by atomization, as described earlier, and the mass
concentrations before and after the two stage concentration system were measured by a DustTrak
DRX. The tests for 3 and 6µm were conducted without using the saturator and condenser parts of
VACES since these particles are in the coarse PM range and already greater than the cutpoint of
both Vis.

After completion of the laboratory evaluation, the system was deployed in the Particle
Instrumentation Unit (PIU) of the University of Southern California from April to June 2012.
The site is located in an urban area near downtown Los Angeles, California, within 150 m of a
major freeway (I-110), and thus represents a typical urban mix of particle sources. The local
ambient conditions during the sampling periods were 22-27 ° C and 20-40% relative humidity
(RH), with PM
2.5
concentration of 10-50 µ g/m
3
, based on the data obtained from a local
monitoring site maintained by the South Coast Air Quality Management District (SCAQMD).
Particle size distribution measurements upstream and downstream of the system were conducted
using SMPS/CPC system similar to the lab tests. Additionally, black carbon (BC) concentrations
before and after the two-stage VI system were determined by means of a two-channel
Aethalometer (Model AE-21, Thermo Andersen, Smyrna, GA). In urban areas BC is often
dominated by vehicular emissions, which have been associated with adverse health outcomes
(Invernizzi et al., 2011).

Finally, time-integrated ambient and concentrated PM
2.5
samples were collected for PM-
bound polycyclic aromatic hydrocarbon (PAH) analysis since PAHs are regarded as one of the
most harmful components of ambient PM to human health (Ris, 2007). Ambient PM
2.5
was
collected by a Micro-Orifice Uniform Deposit Impactor (MOUDI model 110, MSP Corp,
Shoreview, MN). The MOUDI was operated at 30 L/min using 37-mm Teflon filters (Teflo, Pall
Corp., Life Sciences, 1-μm pore, Ann Arbor, MI). A 37-mm Teflon filter was connected to the
minor flow of the second-stage VI to collect concentrated PM. Ambient and concentrated PM
samples were collected concurrently for about 7-8 hours a day, and for about 2 days for each
sample. The PAH analysis was conducted using Gas Chromatography-Mass Spectrometry (GC-
MS) (Sheesley et al., 2004).

16 | P a g e

2.3. Results and discussion
2.3.1. Laboratory characterization of second-stage VI:
The characteristics of the first-stage virtual impactor (VI) were described in detail in
previous studies by Kim et al. (2001b). This VI was designed to have a 50% cutpoint of 1.5 µ m
at an intake flow rate of 110 L/min. In order to characterize the second-stage VI, different intake
flow rates were tested, ranging from 10-30 L/min, with a constant minor flow of 1.5 L/min. The
50% cutpoint (d
50
) can be calculated by Stokes equation:
2
50
50
9
PC
j
d UC
Stk
D




where, Cc, η and ρp are the Cunningham correction factor, the kinematic air viscosity and
the particle density, which is 1, 1.8× 10-5 Pa∙S and 1000 kg/m
3
, respectively. Dj is the impactor
nozzle diameter, which is 0.25 cm in our case. U is the velocity at the nozzle. Stk
50
is the Stokes
number corresponding to the 50% collection efficiency cutpoint, corresponding to a value of
0.48 for circular nozzle VIs (Marple and Chien, 1980), and d
50
represents the theoretical 50%
cutpoint. Based on this equation, the theoretical cutpoint for intake flow rates of 10, 15 and 30
L/min are 2.4, 2.0 and 1.4 µ m, respectively. Figure 2.2 shows the collection efficiency curves of
the second-stage VI, defined as the ratio of downstream particle mass concentration to the
upstream particle mass concentration divided by the ratio of the intake to minor flow, plotted as a
function of aerodynamic particle diameter. As indicated in Figure 2.2, the experimentally
determined 50% cutpoints are approximately 2.2, 2.0 and 1.9 µm for 10, 15 and 30 L/min,
respectively. Except for 30 L/min, the experimental cutpoints are quite consistent with their
corresponding theoretical values, and the measured values are not statistically different from the
theoretical values (p=0.76). Marple and Chien (1980) found that particle losses on the tip of the
minor flow probe increase when the minor-to-intake flow decreases below a given ratio, usually
5% of the total flow; these losses are more pronounced near the 50% cutpoint size range. This
might explain the somewhat higher than predicted cutpoint at 30 L/min. At this low minor-to-
intake flow ratio, the concentration enrichment for particles near the cutpoint size range seems to
be reduced by particle losses. Irrespective of the actual cutpoint of each configuration and the
agreement between theoretical values and experimentally determined cutpoints, all sampled
particles following the hydration process, thereby having grown to 3-4 µ m droplets, should be
17 | P a g e

theoretically drawn into the minor flow of each configuration tested, based on the results plotted
in Figure 2.2.

Figure 2.2 Collection efficiency curves of the second-stage virtual impactor at different intake
flows

2.3.2. System configuration test
To test whether the overall concentration enrichment factor of the two-stage VI system is
significantly influenced by the intermediate flow (i.e., the minor flow of the first-stage VIs and
intake flow of the second-stage VI), different intermediate flow rates of 10, 15 and 30 L/min
were tested. The theoretical enrichment factor, which is defined as the ratio of the intake flow of
the first-stage VIs (200 L/min) to the minor flow of the second-stage VI (1.5 L/min), is 133.
Results from the test shows that the two-stage system has relatively higher enrichment efficiency
and less variability when operating at an intermediate flow rate of 15 L/min, with an enrichment
efficiency of 80% (± 6%), in comparison to 10 L/min or 30 L/min, with an enrichment efficiency
of 59% (± 9%) and 61% (± 10%). Using 15 L/min as the intermediate flow also yields a more
even enrichment at each stage, with theoretical concentration enrichment factors of 13.3 in the
first stage (200 to 15 L/min) and 10 in the second stage (15 to 1.5 L/min). Therefore, a higher
overall enrichment efficiency may be achieved in addition to the lower variability in the overall
concentration enrichment. As a result, 15 L/min was chosen as the optimal intermediate flow rate
18 | P a g e

during all laboratory and field evaluation tests, with concentration enrichment factors of 100 to
120, corresponding to approximately 80% to 90% of the ideal theoretical value of 133.

2.3.3. Laboratory evaluation of the two-stage VI system
Figures 2.3(a-d) illustrate the concentration enrichment and size distribution of lab-
generated particles. In each figure, the upstream concentration is presented on the primary y-axis
on the left side and downstream concentration on the secondary y-axis on the right side. The
secondary y-axis is 100 times (120 times in Figure 2.3(d)) greater than the primary y-axis to
make both datasets visible in one plot. Figures 3a-c show the size distribution results of
ammonium sulfate, adipic acid and glutaric acid, performed with a minor flow of 1.5 L/min.
Although we observe some degree of distortion in the size distribution between the upstream and
downstream, the overall mass and number enrichment factors are relatively stable and
consistently over 100 times, corresponding to about 80% of the theoretical enrichment factor of
133. This is in close agreement with the expected value of 81% assuming that each VI stage
performs at about 90% concentration efficiency. In addition to the consistent enrichment factors,
physical properties of the particles, including the median, mode and mean diameters for the
upstream and downstream particles are also presented in the figures. For each species, the
differences of median, mode and mean diameter between upstream and downstream are within
10%-15%, indicating that major physical properties of particle size distribution are not altered
before and after particle concentration. The total number and mass concentration for both
upstream and downstream are also included in these figures. The total number concentration
enrichment factors for ammonium sulfate, adipic acid and glutaric acid are 120± 31, 110± 10 and
121± 14; the total mass concentration enrichment factors are 130± 27, 105± 16 and 109± 10,
respectively. Figure 2.3(d) shows the results of ammonium nitrate with a minor flow of 1 L/min.
The total number and mass enrichment factors are 150± 16 and 195± 22, respectively, compared
to the theoretical value of 200.

19 | P a g e

Figure 2.3(a-d). Particle size distributions for laboratory-generated polydisperse particles for (a)
ammonium sulfate, (b) adipic acid, (c) glutaric acid, (d) ammonium nitrate (with minor flow of 1
L/min) upstream and downstream the 2-stage concentrator. Physical properties of the size
distribution including median, mean, mode, and total number and mass concentrations are shown.
The secondary y-axis is multiplied by a factor of 100 (120 for ammonium nitrate) for visibility
purposes.

The slight discrepancy between mass and number–based enrichment factors can be to a
certain degree attributed to the lower concentration enrichment of sub-20 nm particles, as noted
in our earlier publications (Geller et al., 2005) in which we discuss in greater detail the
relationship between the minimum particle diameter that can be activated and grown by super-
saturation in our VACES for externally-mixed ambient particles (e.g. urban background). The
experimentally determined super-saturation ratio (S) in the VACES is in the range of 1.05-1.15,
sufficient to activate externally-mixed insoluble particles of at least 20 nm and above (Ning et al.,
20 | P a g e

2006). From the lab tests results, the average enrichment factors for particles in 13-20 and 20-30
nm are 53± 9 and 89± 21, respectively.

In Figures 2.3(a-d), a local peak is observed at 50-60 nm for both the upstream and
downstream particle size distribution among all four tested particles. Figure 2.4 presents the
average enrichment efficiency (the average ratio of measured enrichment factor divided by
theoretical value) for different size ranges among the lab and field evaluation tests. It is clear that
in size range of 30~60 nm, the average enrichment efficiency is over 100%, which corresponds
to the local peak in Figure 2.3(a-d). This peak may have been introduced by the sampling
methodology, considering that the SMPS scans were not taken concurrently, and may also be the
result of potential coagulation of smaller particles into larger particles for these size ranges. With
the exception of sub-20 nm and 30~60 nm particles, the results shown in Figures 2.3(a-d) and
Figure 2.4, prove the ability of the two-stage VI system to achieve an overall enrichment factor
of 100-120 for sub-micrometer PM without drastically altering the particle size distribution and
its physical properties (i.e. mode, median).

Figure 2.4 Average enrichment efficiency in different size ranges.

In addition to polydisperse particles, similar concentration enrichment tests were
conducted with monodisperse PSL particles of 100, 300, 400 nm and 1, 3 and 6 µ m using the
21 | P a g e

DustTrak DRX. Table 2.1 shows the results of mass concentration enrichment for different PSL
particle sizes. Generally, the concentration enrichment factors were in the range of 100-120 with
uncertainties of 10-20%. These results corroborate the concentration enrichment performance of
the two-stage VI system for ultrafine PM, PM
2.5
and coarse PM.

Table 2.1 Upstream and downstream the two-stage concentrator mass concentrations, mass
enrichment factors, and enrichment efficiency based on PSL tests
PSL Particle
Upstream
Concentration
Downstream
Concentration
Mass
enrichment
factor
Enrichment efficiency
(measured/theoretical)
100 nm
a
0.039± 0.005 3.932± 0.095 100.82± 13.15 75%± 10%
300 nm
a
2.28± 0.25 249.20± 36.05 109.30± 19.80 82%± 15%
400 nm
a
2.27± 0.31 235.80± 21.15 103.88± 16.91 78%± 12%
1 μm
b
0.016± 0.002 1.62± 0.04 104.18± 13.39 79%± 10%
3 μm
b
0.058± 0.014 6.54± 0.69 112.76± 29.07 85%± 21%
6 μm
b
0.045± 0.011 4.90± 0.27 109.89± 27.18 83%± 20%
a
concentration in µ g/m
3

b
concentration in mg/m
3

2.3.4. Field evaluation of two stage VI system
Field validation measurements were carried out after completion of all laboratory tests.
Particle size distribution tests were also conducted using the SMPS/CPC system and over 20
scans were taken for both ambient air (upstream) and concentrated air (downstream). These
results are shown in Figure 2.5. The median diameter for both upstream and downstream is
approximately 30 nm for the duration of the field tests. Similar to the observations in the
laboratory, the overlap of the two plots validates the consistent enrichment by a factor of around
100 for ambient PM. In addition, the mode, median and mean diameters for both upstream and
downstream differ by approximately 5%. However, we do observe some degree of distortion in
size distribution for sub-20 nm and 30~60 nm particles in the upstream and downstream, which
may be attributed to the dynamic nature of ambient aerosols and the alternate (as opposed to
concurrent) sampling between up-and downstream of our system.

22 | P a g e

Figure 2.5 Particle size distributions of ambient PM upstream and downstream of the two-stage
virtual impactor system

Similar results were observed for black carbon tests conducted by the two-channel
Aethalometer. Ambient black carbon mass concentrations had an average of 500± 63 ng/m
3
while
downstream concentrations after passing through the two-stage VI system had an average of
54,000± 5100 ng/m
3
, corresponding to an enrichment factor of 100-110, consistent with
laboratory results discussed earlier.

The final field evaluation test was based on time-integrated measurements of polycyclic
aromatic hydrocarbon (PAH) concentrations upstream and downstream of the two-stage VI
system. Results are presented in Figure 2.6. Sixteen PAH species were analyzed by GC/MS. In
Figure 2.6, the average enrichment factor among different PAH species are presented. Aside
from four species that are under detection limit (acenaphthene, acenaphthylene and indeno(1,2,3-
cd)pyrene, and naphthalene are not shown), twelve species are presented in the plots. Overall,
with the exception of dibenz(a,h)anthracene, the concentration enrichment factors for each
species are very consistent, with an average value of 109± 26, which is in good agreement with
those observed for other species in lab and field tests. Based on these results, the concentration
enrichment performance of the two-stage VI system also seems to be unaffected by the
molecular weight as well as the vapor pressure and volatility of the measured PAH
23 | P a g e

Figure 2.6 Average enrichment factor of PAHs.

2.4. Summary and conclusions
A two-stage virtual impactor system was developed and evaluated in this study. The
particle size distribution measurements in the lab and in the field illustrate that an enrichment
factor of 100-120 for both number and mass concentration is achieved along with similar
physical properties of the size distribution (i.e. mode, median). Even though some degree of
distortion in PM size distribution was observed, especially for sub-20 nm and 50~60 nm particles,
the results from this study have shown that the two-stage concentration system is very effective
at consistently enriching particles to very high number and mass concentrations. Field test results
also demonstrated the ability of the system to concentrate black carbon and particle-bound PAH
by enrichment factors close to their ideal value. Overall, the results in this study demonstrate that
the two-stage VI system is appropriate for providing a robust and consistent concentration
enrichment for different types of particles, a feature that makes it a potentially valuable tool in
applications requiring highly concentrated particles especially at low flow rates (on the order of
1-2 L/min).
24 | P a g e

2.5. Acknowledgements
This study was funded by the National Institutes of Health through Grant #1566G HB474,
and the Environmental Protection Agency through Grants/Awards #1566G HB474 and #RD-
83374301-0 to the University of Southern California (USC). The research described herein has
not been subjected to the agency’s required peer and policy review and therefore does not
necessarily reflect the views of the agency, and no official endorsement should be inferred. The
authors would like to thank Dr. Junfeng Zhang (USC Keck School of Medicine) for his
assistance in the chemical analysis.

25 | P a g e

Chapter 3 Macrophage Reactive Oxygen Species Activity of
Water-soluble and Water-insoluble Fractions of
Ambient Coarse, PM
2.5
and Ultrafine Particulate
Matter (PM) in Los Angeles

This chapter is based on the following publication:
Wang, D., Pakbin, P., Shafer, M. M., Antkiewicz, D., Schauer, J. J., & Sioutas, C. (2013).
Macrophage reactive oxygen species activity of water-soluble and water-insoluble fractions of
ambient coarse, PM
2.5
and ultrafine particulate matter (PM) in Los Angeles. Atmospheric
Environment, 77, 301-310.

3.1. Introduction
Numerous health studies have demonstrated that exposure to atmospheric particulate
matter (PM) has strong associations with significant impacts on human health, and may induce
respiratory illnesses (Li et al. 2003a), cardiovascular diseases (Pope et al. 2004; Delfino et al.
2005), neurodegenerative disorders (Campbell et al. 2005; Morgan et al. 2011), and adverse birth
outcomes (Wilhelm and Ritz 2005). A broadly accepted hypothesis is that PM-induced toxicity is
driven by the interaction of PM with cells and macrophages to generate reactive oxygen species
(ROS), which change the redox status of the cells (Donaldson et al., 2002; Xia et al., 2006).
There is increasing evidence that PM characteristics (e.g. size and chemical composition) are
important factors in mediating this cellular oxidative stress (Cho et al. 2005; Kleinman et al.
2008). A variety of methods have been developed to evaluate the potential of ambient PM in
generating ROS both in purely chemical as well as cell-based assays. For example, previous
studies have demonstrated that the dithiothreitol (DTT) assay can provide a good measure of the
redox activity of particles by determining superoxide radical formation as the initial step in the
generation of ROS (Cho et al. 2005). Cellular approaches for ROS activity measurement likely
provide a more comprehensive assessment of the oxidative stress potential of PM by further
examining the ROS generated from cellular stimulation in relevant biological models (Fach et al.,
2002; Verma et al., 2009). In a commonly applied alveolar macrophage assay model, ROS
generated intracellularly is quantified using the fluorescent probe, 2, 7-dichlorodihy-
26 | P a g e

drofluorescein diacetate (DCFH-DA) (Landreman et al., 2008). The complexity and variability of
PM chemical composition, which includes water-soluble and water-insoluble metals and trace
elements, black carbon, inorganic salts and a large number of organic species (e.g. (Cheung et al.
2011)), makes it difficult to address the specific role of these species in inducing oxidative stress.
Therefore, assessing the relationships between specific PM constituents and observed adverse
health outcomes still remains an active and important topic of aerosol research.

In conventional PM collection methodologies, ambient PM is first collected on filters and
other substrates and then removed/extracted into different solvents (e.g., water, methanol, etc)
for in-vitro and in-vivo toxicological evaluations. However, several drawbacks may limit the
feasibility and accuracy of this approach in investigating the toxicity of ambient PM, including
poor extraction of insoluble PM-bound species, potential physical and chemical changes in PM
during extraction e.g. due to sonication and vortexing (Dick et al., 2000), and also loss of labile
species such as ammonium nitrate and semi-volatile organic carbons (SVOCs) and/or changes in
chemical speciation (e.g. oxidation state shifts in redox active metals) due to prolonged sampling
time (Turpin et al. 1994; Eatough et al. 2003). To overcome these limitations of conventional
sampling methodologies, a novel technology was developed that combines the versatile aerosol
concentration enrichment system (VACES) (Kim et al. 2001a; Kim et al. 2001b) in tandem with
a liquid impinger PM collector (BioSampler, SKC West, Inc., Fullerton, CA). In this technology,
ambient fine (dp < 2.5 µ m) and ultrafine PM (dp < 100 - 200 nm) are grown to super-micrometer
sizes via condensational growth, and subsequently concentrated by virtual impaction. The
concentration-enriched and enlarged particles are then injected into a swirling flow and are
collected by a combination of impaction and centrifugal forces in the BioSampler. Thus, ambient
PM is directly collected as a highly concentrated suspension in an aqueous ―solvent‖ while
certain uncertainties and limitations related to conventional PM collection and extraction are
effectively eliminated. In a previous study by Daher et al. (Daher et al. 2011) a comparison
between the sampling techniques of filtration, impaction, and the VACES/BioSampler tandem
was performed. Results from this study showed a good overall agreement among the different
sampling methodologies for PM collection efficiency and chemical composition, however the
redox activity of the aqueous extracts of the filter was substantially lower than that of the
BioSampler slurry, and became virtually identical to the BioSampler activity after filtering
27 | P a g e

insoluble PM components from the slurry, thus underscoring the contribution of insoluble
species in total redox activity of ambient PM. Therefore, the Daher et al study demonstrated the
ability of the VACES/Biosampler tandem to capture both water-soluble and water-insoluble
portions of ambient PM for subsequent ROS analysis. Thus, this novel sampling technique is
beneficial comparing to conventional filtration methodology particularly in investigating relative
toxicity of ambient PM.

Efforts have been made to associate specific physical and chemical properties of PM to
oxidative stress and toxicity (Hu et al., 2008; Ntziachristos et al., 2007; Zhang et al., 2008). The
water-solubility of the particles can vary greatly depending on the atmospheric conditions and
local anthropogenic and natural PM sources. Therefore the solubility of chemical components of
ambient PM may be a suitable metric for identifying a specific group of compounds that affect
the redox activity. Various studies have reported toxicity responses of both soluble (Dreher et al.,
1997) and insoluble species (Jalava et al., 2008); however, the relative contribution of water-
soluble and water-insoluble portions of ambient PM to ROS activity is still not entirely clear.
This is mostly due to the limitations of the available methods for separating water-soluble and
water-insoluble portions of PM samples collected on filter substrates that have to be effectively
extracted for subsequent chemical analysis. In addition to conventional water extraction, other
organic solvents and acid leaching have been used to separate the water-soluble and water-
insoluble portions of PM. Verma et al. (2012) illustrated the high correlation between organic
carbon (OC) and DTT activity from water-soluble and water-insoluble portion of ambient PM
2.5

samples by employing water and methanol extraction, respectively. Other organic solvents such
as RTLF and dichloromethane were applied in previous studies (Shinyashiki et al., 2009;
Zielinski et al., 1999). However, a major drawback of organic solvent extraction is that these
solvents must be removed by evaporation and/or similar procedures prior to the ROS
measurements due to their own intrinsic toxicity especially for macrophage assay (Jalava et al.,
2005). As a result, losses of some potentially important toxic PM-bound labile species, such as
semi-volatile organics, are likely to occur during the solvent removal process (Eatough et al.
2003). Alternatively, if acid solvents are involved, some of chemical components may be
converted into soluble forms and/or change redox species, which will probably alter their toxicity.
Moreover, organic compounds might be oxidized into carbon dioxide during acid extraction. The
28 | P a g e

VACES/BioSampler sampling technique eliminates the solvent extraction procedure and could
be advantageous in such studies, especially for the purpose of separating water-soluble and
water-insoluble portions of PM (Khlystov et al. 2005; Zhao et al. 2005).

The goal of the current study was to investigate the toxicological contribution of water-
soluble and water-insoluble portions of ambient PM as measured by the aforementioned ROS
assay. Size-fractionated ambient PM samples (coarse, PM
2.5
and ultrafine PM) were collected
using VACES/BioSampler system, directly into an aqueous suspension. Separation of water-
soluble and water-insoluble fractions was achieved by ultra-filtration of the collected suspension
samples, which allows the separation of soluble and insoluble components of suspension samples,
regardless of their size, since the conventional filtration is not a suitable method for separation of
smaller particles such as ultrafine suspension samples. Oxidative potential of the PM for water-
soluble and total (un-filtered) extracts was measured by the macrophage ROS assay. In addition,
the slurry samples were analyzed for water-soluble organic carbon/total organic carbon
(WSOC/TOC), metals and trace elements, and inorganic ions. Associations between these
detailed chemistry data and ROS activity were explored to examine major drivers of cellular
oxidative stress in both water-soluble and water-insoluble portions of ambient PM.

3.2. Methodology
3.2.1. Sample collection
Sampling was performed at the Particle Instrumentation Unit (PIU) of the University of
Southern California (USC) during 8 weeks in August-September 2012. The site is located in an
urban area near downtown Los Angeles, California, within 150 m of a major freeway (I-110),
and thus represents a typical urban mix of particle sources (Ning et al., 2007). The local ambient
temperature and relative humidity during the sampling periods were 22-27 ° C and 20-40%,
respectively. Ambient coarse, PM
2.5
and ultrafine PM samples were collected using the USC
VACES/BioSampler system. Development and evaluation of the VACES/BioSampler system are
described in detail in previous studies (Kim et al. 2001a; Kim et al. 2001b; Daher et al. 2011).
The system schematic is presented in Figure 3.1. For PM
2.5
sampling, two parallel sampling lines
from the VACES were employed. Ambient PM
2.5
was first drawn through the saturation-
condensation system to promote condensational particle growth by super-saturation process. The
29 | P a g e

flow rate through each sampling line was set at 105 L/min, and particles were generally grown to
3-4 µ m aqueous droplets, as reported in previous studies (Kim et al. 2001a; Kim et al. 2001b).
The individual streams of droplets were then concentrated by a virtual impactor (VI) in each line,
which had a major flow rate of 100 L/min and a minor flow rate of 5 L/min and a 50% cutpoint
of 1.5 µ m in aerodynamic diameter. The concentrated droplets then exited their corresponding
VI stage as part of the minor flow. For one streamline, the minor flow was directly connected to
a pre-cleaned (methanol & nitric acid) BioSampler to accumulate droplets for the suspension
samples. Operation and evaluation of the BioSampler was discussed in details by Kim et al.
(Kim et al. 2001b). A 37-mm Teflon ―after-filter‖ (Teflo, Pall Corp., Life Sciences, 1-μm pore,
Ann Arbor, MI) was placed downstream of the BioSampler to collect any droplets escaping
collection for validation purposes. In the other streamline, concentrated particles were diffusion-
dried with a silica gel diffusion dryer (Model 3062, TSI Inc., Shoreview, MN) following the VI
and returned to their original size. The diffusion-dried stream was then directed to a 37-mm
Teflon filter in order to estimate the mass loading in the sampling trains.

Figure 3.1 System schematic of PM
2.5
and Ultrafine PM (UFP) collection

The sampling of coarse PM was conducted similarly to PM
2.5
sampling but without using
the saturator and condenser parts of the VACES, since these particles are already greater than the
cutpoint of the VI. 20ml of high purity water (HPLC grade) was injected into the BioSampler
30 | P a g e

before sampling in order to capture the coarse PM in liquid suspension samples. For ultrafine PM
collection, a low pressure drop multi-slit nozzle impactor was placed in front of
VACES/BioSampler system in order to remove particles larger than 180nm. The design and
operating performance of the impactor have been described in detail previously (Misra et al.,
2002). Three parallel sampling lines were used for ultrafine PM collection. BioSamplers were
placed in two of the sampling lines and the third line went into a diffusion dryer followed by a
37-mm Teflon filter, similar to that described above for PM
2.5
sampling.

Time-integrated ambient PM samples were collected during the 8-week sampling
campaign for different time periods of each day. For coarse PM, daytime samples (10AM to
4PM) and nighttime samples (10PM to 4AM) were collected; while for PM
2.5
and ultrafine PM,
samples were collected during ―morning time‖ (6AM to 10AM) and ―afternoon time‖ (12PM to
4PM). Suspension samples collected from different days over a two month period were then
composited resulting in 6 sets of samples in total (two for each size fraction) capturing a broad
range of PM chemical composition. All suspension samples were kept frozen at –20
◦
C after each
sampling section and before chemical analysis.

3.2.2. Sample analysis
In the laboratory, each BioSampler slurry was divided into two portions: one portion was
ultrafiltered at 10 kilo-Dalton (kD) using centrifugal ultrafilters (Amicon Ultra-15 Centrifugal
Filter Device, regenerated cellulose membrane in an all polypropylene device), and the other
portion was left un-fractionated. The ultrafiltration devices were pre-cleaned with dilute acid and
high-purity water just prior to use. The chemical species contained in the solution passing
through the 10 kD ultrafilter (permeate) are termed water-soluble portion, as all particles,
colloids and macromolecules are isolated from this fraction, and what remains are truly soluble
species or small molecules/complexes. The un-fractionated (unfiltered) BioSampler slurries
contain all chemical/physical forms (i.e. total) and the difference between total and ultrafiltered
is considered the water-insoluble portion. Aliquots of both the water-soluble and total fractions
were sub-sampled and subsequently analyzed for inorganic ions, organic carbon and elements.
Total sub-samples were processed appropriately to quantitatively recover all of the species of
interest. The inorganic ion composition was determined by ion chromatography (IC). Measured
31 | P a g e

ions included chloride, sodium, nitrate, ammonium, potassium and sulfate. WSOC/TOC analyses
were conducted by injecting water-soluble and total sub-samples into a Sievers 900 Total
Organic Carbon Analyzer (Stone et al., 2009). The TOC results from water-soluble subsamples
were considered as WSOC and as TOC for total samples. For analysis of metals and elements,
water soluble sub-samples (acidified with dilute acid) were directly measured, while total
samples were first digested in acid before analysis. Measurements were conducted using high-
resolution magnetic sector inductively coupled plasma mass spectrometry (SF-ICP-MS, Thermo-
Finnigan Element 2) (Zhang et al. 2008).

A portion of the water-soluble and total sub-samples were analyzed for macrophage ROS
activity. The macrophage ROS assay is a fluorogenic cell-based method to investigate the
production of ROS in rat alveolar macrophages (NR8383, American Type Culture Collection)
using 2, 7-dichlorodihydrofluorescein diacetate (DCFH-DA) as the fluorescent probe. DCFH-DA
is a non-fluorescent, membrane permeable compound. After entering cells, DCFH-DA is rapidly
de-acetylated to DCFH, which can be converted by a broad range of ROS into 2, 7-
dichlorofluorescein (DCF), which is highly fluorescent and can be readily detected. Details of the
assay, extraction protocol, and detection methodology are explained in Landreman et al
(Landreman et al. 2008). Briefly, rat alveolar macrophages (prepared as detailed in Landreman et
al. 2008)) were exposed to the samples (particle extracts and suspensions) for a total time of 2.5
hours (at 37 °C under a 6% CO
2
atmosphere). At the completion of the exposure period the
fluorescence intensity of each well was determined (in kinetic mode) at 504 nm excitation and
529 nm emission (515 cutoff) using a m5e microplate reader (Molecular Devices). Negative
controls included complete method blanks (sample collection/processing/extraction blanks) and
each sample on the 96-well plate was preceded by an ―untreated‖ macrophage control (i.e. plated
macrophages exposed only to SGM). Raw fluorescence data were control-corrected and
normalized to the Zymosan positive control. Sample fluorescence fold-changes (with respect to
the untreated controls) over the 2.5 hours exposure period ranged from 1.5 to over 10 (depending
upon sample activity and serial dilution). A minimum of four dilutions (each of them in triplicate)
of every sample extract/suspension was performed to ensure that a linear dose-response region
could be identified. Overall uncertainty in the ROS method was estimated by propagating the
three primary sources of variation: (a) ROS measurement (standard deviation of the triplicate
32 | P a g e

measurements); (b) applied control/blank (standard deviation of the controls); and (c) positive
control response (standard deviation of the Zymosan controls). Thus a total uncertainty metric
was associated with each individual sample measurement and subsequently used to determine
significant differences between different exposure groups.

3.3. Results and discussion
3.3.1. Chemical composition of coarse, PM
2.5
and ultrafine PM total samples
The chemical composition of the size-fractionated PM slurry samples is presented in
Figure 3.2. Chemical components were grouped in six broad categories, each reflecting a distinct
group of sources, as follows: (1) crustal material (CM); (2) vehicle abrasion (VA); (3) water-
soluble organic matter (WSOM); (4) water-insoluble organic matter (WIOM); (5) sea spray (SS)
and (6) secondary ions (SI; i.e. ammonium, sulfate and nitrate).

Concentration of crustal material (CM) (as oxides) is calculated using the following
equation (Chow et al., 1994a; Hueglin et al., 2005):

Crustal material = 1.89Al + 1.21K + 1.43Fe + 1.4Ca + 1.66Mg + 1.67Ti + 2.14Si [1]

where Si is estimated as Si=3.41Al (Hueglin et al. 2005).

Vehicle abrasion is calculated as the sum of concentrations of several metal species,
including V, Cr, Mn, Ni, Cu, Zn, As, Se, Br, Sr, Ba and Pb (Harrison et al., 2001).

Water-insoluble organic carbon (WIOC) is calculated by subtracting WSOC (measured
from water-soluble sub-samples) from TOC. To convert WSOC and WIOC to WSOM and
WIOM, respectively, there are two main sources of uncertainty: 1) the split between soluble and
insoluble organic carbon, and 2) the ratio of organic mass to organic carbon for WSOC and
WIOC. ―Water-soluble‖ OC is operationally defined, and reflects the sum of colloidal and truly
dissolved carbon. Nonetheless, the conversions of the operationally defined soluble and
insoluble carbon to WSOM and WIOM, respectively, are well-established in the literature, and
are made by measurement techniques that are very reproducible (Turpin and Lim, 2001).
33 | P a g e

Typically, a factor of 1.4 ± 0.2 and 2.0 ± 0.2 is recommended for urban and non-urban aerosols,
respectively, to account for the contributions of non-carbon atoms (e.g. oxygen, hydrogen) to the
mass of organic matter (OM) (Turpin and Lim 2001). In this study, a multiplier of 1.8 was used
for converting WSOC to WSOM, and 1.4 for converting WIOC to WIOM through all the
samples, based on the recommendations of Turpin and Lim.

Secondary ions (SI) include NH
4
+
, NO
3
-
and SO
4
2-
. Sea spray is primarily formed by
bursting of seawater bubbles due to the surface wind during the formation of whitecap (Gong,
2003). The estimation of sea spray components is usually performed using soluble Na
+
and the
sea spray fraction of typical sea water components such as Cl
-
, Mg
2+
, K
+
, Ca
2+
and SO
4
2-
as
follows (Cheung et al. 2011):

Sea spray (SS) = Na
+
+ ssCl
-
+ ssMg
2+
+ssK
+
+ ssCa
2+
+ssSO
4
2-
[2]

where ss (Cl
-
) =1.8, ss (Mg
2+
) = 0.12, ss (K
+
) = 0.036, ss (Ca
2+
) =0.038 and ss (SO
4
2-
) =
0.252 (Seinfeld and Pandis 2006).

Figure 3.2 Chemical composition of PM samples.
34 | P a g e

Overall, the reconstituted mass from the measured chemical species is in good agreement
(65% to 131%, with an average value of 96 ± 24%) with the measured mass concentration
determined gravimetrically on parallel filters (Figure 3.2). PM samples collected in this study
include coarse, fine and ultrafine PM size fractions and were collected during different sampling
time periods to expand and maximize the variability in chemical composition among them,
which is evident in Figure 3.2. OM originates primarily from vehicular emissions and photo-
oxidation of reactive gaseous precursors (Seinfeld and Pandis, 2006). As a result, the
contribution of OM to total PM mass increases for the smaller particles, with the coarse PM
samples consisting of 15-18% OM, and 36-54% and 57-64% for PM
2.5
and ultrafine PM samples,
respectively. Secondary ions, including sulfate, nitrate and ammonium, are the most abundant
chemical components in coarse PM and PM
2.5
samples, representing 30-41% and 40-48% of total
mass, respectively, while somewhat lower in ultrafine PM, representing 17-20% of total mass.
Generally sulfate, nitrate and ammonium are formed by secondary processes of precursors such
as sulfur dioxide, nitric acid and ammonia (Seinfeld and Pandis, 2006). For crustal material, the
mass contribution decreases for smaller size fraction, accounting for 8-23% in coarse PM, 6-10%
and 4-5% for PM
2.5
and ultrafine PM samples, respectively. Sea spray contributes to a large
portion of mass in coarse PM samples, i.e. 12-40% of total mass, and to a lesser extent to PM
2.5
and ultrafine PM samples (about 13-15% and 5-15% of total mass, respectively). The observed
mass fractions of chemical species in this study are within good agreement with previously
reported speciation results from studies conducted at the same sampling location (Ntziachristos
et al. 2007; Verma et al. 2009; Cheung et al. 2011). As noted earlier, the same factors of 1.8 and
1.4 for WSOC and WIOC are applied to all three size fractions samples in order to estimate OM
mass concentrations. Using these multipliers for all three size fractions may introduce some
uncertainties in estimating total mass of OM, however the agreement between re-constituted
mass concentration and measured concentration is generally very good, which indicates that
these multiplication factors are reasonable .

It should be noted that the overall percentage of water soluble organic matter to total
organic matter (WSOM/OM) is relatively high among all samples, with an average value of
73± 8%, compared with previous studies, with reported WSOM/OM ratios varying between 30%
35 | P a g e

to 60% (Hu et al. 2008; Verma et al. 2012). One of the major differences in the current study is
that PM samples were directly collected into an aqueous suspension in contrast to the
conventional filtration sampling used in previous studies. The prolonged sampling times required
for collection of sufficient mass by filtration may be susceptible to sampling artifacts such as
losses of water-soluble semi-volatile organic carbons (SVOC) (Turpin et al. 1994; Eatough et al.
2003) which would result in under-reporting WSOC using these methods. Previous studies have
demonstrated that particle concentrators such as the VACES could significantly reduce the
sampling artifacts induced by volatilization losses of components such as nitrate, ammonium and
volatile organic compounds by reducing the sampling time required for collecting sufficient mass
(Chang et al. 2000; Khlystov et al. 2005). Moreover, SVOC species are captured more
effectively when collected directly into liquid suspension as evidenced by the observation of
higher concentration of WSOC in slurry samples compared to filter substrates as reported in
previous studies conducted at the same sampling site (Verma et al. 2009; Cheung et al. 2011;
Daher et al. 2011). Both of these factors may contribute to the higher WSOC/TOC ratios in this
study. Zhang et al (Zhang et al., 2012) investigated the formation mechanisms of WSOC and its
contribution to ambient concentrations in Los Angeles and central Atlanta, showing that WSOC
in Los Angeles is generally formed by gas-to-particle phase photochemical conversion
of anthropogenic VOC, while WSOC in central Atlanta is mostly generated from the photo-
oxidation of biogenic sources. The PM samples in current study were collected during the warm
summer season when photochemical activity is at its peak. These factors may contribute to
higher WSOC content of ambient PM

and the increased WSOC/OC ratio. Although there is no
TOC data available from the study conducted by Zhang et el., their reported WSOC
concentrations in Pasadena (in Los Angeles basin, roughly 10 km north-east of our sampling site)
is in very good agreement with the current measurements; the average measured WSOC
concentration of PM
2.5
samples in Pasadena was 7.59 ±3.17 μg/m
3
(Zhang et al. 2012) compared
to 8.79 ±0.58 μg/m
3
for PM
2.5
samples in the current study.

3.3.2. Macrophage ROS results
The macrophage ROS measurement results are presented in Figure 3.3(a-b). The results
are expressed on a per volume (of air) based activity metric (Fig. 3.3a) in µ g Zymosan units/m
3

of air as well as on a per mass (of PM) based activity metric (Fig. 3.3b) in µ g Zymosan units/mg
36 | P a g e

of PM mass, respectively. Since it was not possible to obtain an accurate gravimetric mass
concentration for water-soluble and water-insoluble portions of PM samples after separation by
ultra-filtration, the mass concentration of water-soluble and water-insoluble portions of PM are
estimated based on the mass reconstruction for each portion from their respective chemical
composition. The total mass concentration of PM is determined from the mass collected on the
parallel-sampled 37mm Teflon filter. The estimated water-soluble mass concentration is
calculated by deploying the same equations calculating chemical component categories (i.e. CM,
VA, SS, WSOM, SI) detailed in previous section, using only the water-soluble species in the
mass reconstitution calculation. Water-insoluble mass concentration is calculated by subtracting
the water-soluble mass concentration from total mass concentration. The total, water-soluble and
water-insoluble mass concentrations for each sample are presented in Table 3.1. The ROS
activity is measured directly for water-soluble sub-samples (referred to as water-soluble ROS),
while the water-insoluble ROS is calculated by subtracting water-soluble ROS from the total
ROS. The mass-based ROS activity for water-soluble and water-insoluble portions of PM are
calculated based on the aforementioned water-soluble and water-insoluble mass concentration
calculations.

(a)

37 | P a g e

(b)
Figure 3.3 (a-b). ROS measurement results: (a) volume-based ROS; (b) mass-based ROS

Table 3.1 Mass concentration for total, water-soluble and water-insoluble portions of collected
samples.
Mass concentration (μg/m
3
)
Sample set Total Water-soluble Water-insoluble
CPM day 10.62 5.57 5.05
CPM night 9.25 7.64 1.61
PM
2.5
AM 22.20 20.54 1.66
PM
2.5
PM 21.49 18.68 2.81
UFP AM 8.94 7.99 0.95
UFP PM 8.22 6.80 1.42

As shown in Figure 3.3(a-b), PM
2.5
samples have highest volume-based ROS activity
compared to coarse PM and ultrafine PM samples, while on average, the intrinsic ROS activity
of ultrafine PM is slightly higher than coarse PM. This likely reflects the much higher total mass
concentrations in PM
2.5
samples, as evident in Figure 3.2. Similar observations (i.e. higher
volume-based ROS activity in PM
2.5
than coarse and ultrafine PM) were reported previously for
38 | P a g e

samples collected at the same sampling site (Hu et al. 2008). For PM
2.5
and ultrafine PM samples,
it is also noted that samples collected during the afternoon had higher ROS activity than morning
samples collected on the same day. One explanation is that higher ROS activity in the afternoon
samples is induced by species formed by photo-chemical reaction during afternoon time period.
Verma et al. (Verma et al. 2009) demonstrated that secondary aerosols produced by
photochemical processes in the afternoon period contributed to higher redox activity in ultrafine
PM collected at the same sampling site. ROS results from current study are therefore consistent
with the findings presented in Verma et al.

The ROS activity of water-soluble and water-insoluble fractions can be normalized by the
collected PM mass or the sampled air volume, each revealing important insights about the
contribution of soluble and insoluble species to PM toxicity. While the ROS activity on a per
volume of air basis for the water-soluble portion is typically higher than the water-insoluble
portion, an opposite trend is observed for the mass-normalized ROS, showing higher redox
activity per mass of PM for the water-insoluble portion. However, despite their higher intrinsic
ROS activity, the lower mass fraction of the water-insoluble portions make their overall
contribution to PM redox activity on a per m
3
of air volume quite small. The volume-based ROS
metric may be more relevant for population exposure assessment, as it accounts for PM
concentrations and therefore reflects the total amount of potentially toxic air pollutants taken up
by individuals via inhalation. Our results therefore indicate that the water-soluble components
contribute to a significant fraction of the total oxidative activity of ambient PM.

A comparison of the PM mass-normalized ROS activity from the current study with
earlier studies using the exact same assay (Shafer et al. 2010; Verma et al. 2009; Daher et al.
2011; Daher et al., 2012; Verma et al., 2010) is presented in Figure 3.4. The study by Daher et
al (Daher et al. 2011) was conducted at the same sampling site using essentially the same
sampling methodology. Results from this study (regarding PM
2.5
samples particularly) are in
very good agreement with the results of Daher et al. Slightly higher TOC concentration is
observed in current study, with concentration of 6.6± 0.5 µ g/m
3
comparing to 5.5± 1.7 µ g/m
3
from
Daher et al. However, the PM mass-based ROS activity is much higher than the results from
Daher et al, with the average value of 11331 (± 6712) µ g Zymosan unit/mg of PM compared to
39 | P a g e

6126 (± 2838) µ g Zymosan unit/mg of PM, respectively. Considering that these two studies were
conducted at the same sampling site and using a similar sampling methodology, a possible
explanation for the observed differences might be the seasonal variations in PM composition.
Samples in Daher et al were collected during January to March 2010, representing the cooler
winter season in Los Angeles, impacted mostly by primary emissions of PM (Sardar et al. 2005;
Cheung et al. 2011; Daher et al. 2013), while the current study was conducted during August to
September 2012, when photochemical activity is at its peak and contributes to a potentially
higher WSOC concentrations and thus ROS activity (Docherty et al., 2008; Verma et al., 2009).
In addition to the study by Daher et al, ROS results from a few other studies are also compared
(Figure 3.4) to the ROS activity reported in this study. Overall the observed ROS activity in this
study is comparable and in some cases even higher than those reported in previous studies. Apart
from the differences in sampling time periods and locations, a major difference between the
current study and previous ones is the different sampling methodology employed in collecting
PM. Of particular note is for example the comparability in ROS values between the PM samples
of this study collected in a typical urban environment of Los Angeles to those found in the
heavily polluted city of Lahore, Pakistan (Shafer et al. 2010), or in fresh diesel bus exhaust
samples (Verma et al. 2010) which also should be expected to produce highly redox active PM.
These findings suggest that the VACES/BioSampler technology may capture a higher fraction of
the overall ROS activity in ambient PM and more efficiently than conventional filter sampling,
due to the reasons note earlier related to incomplete PM sample extraction as well as losses of
semi-volatile species during sample collection from the latter methodology. This argument is
further supported by our earlier publication (Daher et al. 2011) in which we show that based on
the same macrophage reactive oxygen species (ROS) assay, the oxidative potential of aqueous
extracts of concurrent filter and impactor substrates was similar yet substantially lower than that
of the VACES/BioSampler slurries also collected at the same time.

40 | P a g e

Figure 3.4 ROS comparison with previous studies

3.3.3. Correlation between ROS activity and chemical components
In order to investigate the association between PM oxidative potential (ROS activity) and
different chemical components of the PM, simple linear regression analysis was conducted
between measured chemical species and macrophage ROS activity. Similar to the calculation of
ROS activity, the concentration of all chemical components in water-insoluble portion are
calculated by subtracting the water-soluble portion from their corresponding total measurement.
In order to diminish the effect of co-variability among PM components, in addition to
representing the actual contribution of the species to redox activity, regression analysis was
performed on mass fraction of each species as discussed by Verma et al. (Verma et al. 2012).
Water-soluble and water-insoluble chemical components are normalized to the particle mass
using estimated mass concentration of water-soluble and water-insoluble portions of all samples,
41 | P a g e

respectively. Pearson correlation coefficients (R) and the associated coefficient of significance (p)
of macrophage ROS vs selected PM components are presented in Table 3.2. The species
exhibiting high correlation (defined here as p<0.05 and R>0.70) with the ROS activity are
highlighted in bold in the table. Regression plots between specific species with high R values and
ROS activity are presented in Figure 3.5(a-d).

Table 3.2 Correlation between chemical components of ambient PM with total, water-soluble
and water-insoluble ROS activity
Total Water soluble Water insoluble
R p R P R P
TOC 0.80 0.05 - - - -
WSOC - - 0.84 0.03 - -
WIOC - - - - 0.70 0.04
SO
4
2-
0.90 0.02 0.96 <0.01 - -
NO
3
-
0.51 0.30 0.79 0.06 - -
NH
4
+
0.71 0.11 0.82 0.04 - -
Ti 0.30 0.55 -0.33 0.52 0.10 0.84
V 0.72 0.10 0.80 0.05 0.16 0.75
Cr 0.19 0.70 -0.20 0.70 0.03 0.95
Mn 0.45 0.36 0.80 0.05 0.13 0.80
Fe 0.67 0.13 -0.44 0.38 0.13 0.80
Ni 0.48 0.32 -0.46 0.53 -0.03 0.95
Cu 0.87 0.02 0.78 0.05 0.16 0.76
Zn 0.51 0.29 0.80 0.05 0.86 0.03
Sr 0.53 0.27 0.19 0.68 0.44 0.36
Ba 0.40 0.43 0.20 0.69 0.11 0.76
Pb 0.16 0.75 0.37 0.47 0.22 0.63

A high correlation between TOC and ROS activity as well as the water-soluble and
water-insoluble portions (WSOC and WIOC) is observed. This result indicates that the organic
components of ambient PM, regardless of their solubility, represent one of the major drivers of
redox activity in PM. This observation is generally consistent with a number of previous studies
investigating correlations between ROS activity and individual chemical components (Cho et al.
2005; Hu et al. 2008; Zhang et al. 2008; Verma et al. 2012).
42 | P a g e

Besides organic compounds, sulfate and ammonium are also well correlated with the
ROS activity of the soluble PM fraction. Ammonium and sulfate are generally considered as
nontoxic species since they have no reactive functional groups to result in the formation of ROS.
Several studies reported no correlation between these species and ROS activity (Ntziachristos et
al. 2007; Hu et al. 2008). However, given the fact that ammonium sulfate and WSOC are all
mostly formed through secondary photochemical process (Seinfeld and Pandis, 2006), it is
possible that the concentrations and mass fractions of ammonium and sulfate are in co-linearity
with those of WSOC. Therefore the observed correlations between ROS activity, ammonium and
sulfate are likely attributed to their correlation to WSOC (the correlation coefficients (R) for
WSOC vs ammonium and sulfate are 0.81 and 0.87, respectively) rather than their own
contribution to ROS activity (Cho et al. 2005; Verma et al. 2009).

Several elements such as Cu, Mn, V and Zn also exhibit a high correlation with ROS
activity in the water-soluble portion. These species are primarily generated from various
vehicular sources including fuel combustion, brake wear and tire abrasion depending on PM size
fractions (Hjortenkrans et al., 2007) and their likely contribution to the redox activity of ambient
PM has been documented in previous studies (Hu et al. 2008; Zhang et al. 2008). Moreover, high
internal correlation between these metal species is observed among all samples (e.g. the
correlation coefficients of water-soluble Mn vs water-soluble V, Cu and Zn are 0.84, 0.90 and
0.77, respectively), which indicates that they are probably contributed by similar sources (i.e.
vehicular sources). In the water-insoluble portion, Zn is highly correlated with ROS activity.
Results discussed above illustrate that besides OC, several transition metals such as Cu, Mn, V
and Zn are also contributing to overall redox activity of ambient PM. Although the mechanisms
of how these metal species generate redox activity are still not completely clear, their
correlations with ROS activity may suggest the potential of these chemicals in driving the
toxicity of ambient PM.

43 | P a g e

(a)

(b)
44 | P a g e

(c)

(d)
Figure 3.5(a-d) Correlation between selected chemical species and ROS: (a) Total ROS; (b), (c)
Water-soluble ROS; (d) Water-insoluble ROS

Generally, species with significant correlation with ROS activity in the current study
were also reported in previous studies as redox active species. Cheung et al. and Hu et al. (Hu et
45 | P a g e

al. 2008; Cheung et al. 2012) reported that OC and several water-soluble metal species such as V,
Cu and Ni are highly correlated with ROS activity in Los Angeles area. Zhang et al. (Zhang et al.
2008) reported WSOC and Zn in Denver area which are also correlated with ROS activity, while
Shafer et al. (Shafer et al. 2010) reported Mn, Co, Cd and Ce in Lahore, Pakistan as redox active
species. Given the fact that these studies were conducted in different sampling periods and
locations, the consistency of our observation in current study significantly validates the high
ROS correlations observed in this study.

3.4. Summary and conclusions
In this study, size-fractionated ambient PM samples (coarse, PM
2.5
and ultrafine PM)
were collected directly into liquid suspensions, in order to investigate the macrophage ROS
activity of their water-soluble and water-insoluble fractions and their correlation with different
PM chemical components. Collecting particles directly into a liquid suspension is beneficial
compared to the conventional filtration process, as the procedure to extract the collected PM
from the filter substrates into suspension is effectively eliminated. Moreover, we have shown
earlier that collecting particles in suspension samples is less prone to the losses of semi-volatile
species.

Results from this study indicate that both water-soluble and water-insoluble portions play
a very important role in the potential toxicity of ambient PM. While the ROS activity per volume
of air for soluble species is generally higher, on a per PM mass basis a higher specific ROS
activity was found in the insoluble fraction, indicating that some of the insoluble species may be
intrinsically more ROS active. However, during this study their overall contribution to total ROS
activity on a per m
3
of air volume was small due to their low mass fractions. The ROS activity
exhibited a high correlation with organic compounds in both water-soluble and water-insoluble
portions of ambient PM, as well as with several transition metal species. It must be noted that by
the nature of our study design, limited number of data points (N=6) were used in the regression
analysis. However, samples in this study represent three distinct PM size fractions, and were
intentionally collected in different time periods during the day and over a two-month sampling
period to capture a broader range of chemical composition (verified in the results of our chemical
analysis). We thus believe that despite these limitations noted above, the high ROS correlations
46 | P a g e

reported between selected PM species should be considered quite robust, particularly given the
high variability and wide concentration range of these species observed in our field
measurements.

3.5. Acknowledgements
This study has been supported by South Coast Air Quality Management District (AQMD)
through award number #11527, and the National Institutes of Health (NIH) through Grant
#7R01AI065617-12 to the University of Southern California (USC). The research described
herein has not been subjected to the agency’s required peer and policy review and therefore does
not necessarily reflect the views of the agency, and no official endorsement should be
inferred. Mention of trade names or commercial products does not constitute an endorsement or
recommendation for use.

47 | P a g e

Chapter 4 Development and Evaluation of a High-Volume
Aerosol-Into-Liquid Collector for Fine and Ultrafine
Particulate Matter

This chapter is based on the following publication:
Wang, D., Pakbin, P., Saffari, A., Shafer, M. M., Schauer, J. J., & Sioutas, C. (2013).
Development and evaluation of a high-volume aerosol-into-liquid collector for Fine and ultrafine
particulate matter. Aerosol Science and Technology, 47(11), 1226-1238.

4.1. Introduction
Numerous epidemiological and toxicological studies have documented robust
associations between ambient particulate matter (PM), particularly fine and ultrafine PM, and
adverse health outcomes (Pope and Dockery, 2006; Ritz et al., 2002). Recent studies have shown
that elevated concentration of PM may be linked to cardiovascular diseases (Pope et al. 2004;
Delfino et al. 2005), pulmonary injury (Li et al. 2009) and lung cancer (Castranova et al., 2001;
Garshick et al., 2004). Understanding sources, atmospheric transformations and aging of ambient
PM allows for more effective regulatory control strategies, more targeted air quality standards,
and ultimately, reductions in population exposure to harmful types of airborne PM. However, the
composition of ambient PM is very complex and highly dependent on local pollution sources
and meteorological conditions, both of which may vary in short periods of time (Cheung et al.,
2011; Daher et al., 2013; Sardar et al., 2005). Thus, ambient PM measurement in high time
resolution becomes critical to identify the toxicologically relevant PM species, as well as the
sources of these species, and remains an active topic of aerosol research.

Traditional PM sampling methodologies usually involve collection of particles onto
substrates by filtration or inertial impaction. Since the quantitative composition measurements
are usually performed offline, the collected particles are extracted using different types of
solvents (e.g., water, methanol, etc) and analyzed by various analytical techniques, such as ion
chromatography (IC), inductively coupled plasma mass spectrometry (ICPMS) and total organic
carbon (TOC) analysis. Although these sampling methodologies have been widely adopted to-
48 | P a g e

date, they are associated with a variety of intrinsic drawbacks, such as long turn-around time for
processed results of chemical analysis, long sampling intervals for collection of sufficient mass
for subsequent chemical analysis (hours to days, depending on sampling flow rate and ambient
concentrations), incomplete extraction of insoluble PM-bound species, potential sampling
artifacts (e.g., losses of labile species such as ammonium nitrate and semi-volatile organic carbon
(SVOC)) (Eatough et al. 2003; Schauer et al. 2003) and/or changes in chemical speciation (e.g.
oxidation state shifts in redox active metals) due to prolonged sampling time (Turpin et al. 1994;
Eatough et al. 2003). These drawbacks may significantly limit the feasibility and accuracy of
conventional sampling approaches in investigating chemical composition and toxicological
properties of ambient PM.

A number of advanced aerosol sampling technologies have been developed in recent
years to overcome these limitations in conventional methodologies. In these newer technologies,
ambient PM is directly collected as an aqueous suspension; for instance PM collection onto a
filter that is periodically washed by water (Buhr et al. 1995; Phan and McFarland 2004),
impaction into a flowing liquid (Karlsson et al. 1997), collection by a water cyclone (Orsini et al.
2008) and the combination of particle growth by water vapor condensation and impaction on
surfaces covered by liquid flow (Weber et al. 2001). These technologies effectively eliminate the
need for elaborate extraction procedures and have been well documented in achieving more
reliable measurements of chemical species such as organic carbon (Zhang et al. 2012) and
secondary ions, including nitrate, sulfate and ammonium (Takegawa et al. 2005). However, most
of these technologies are likely inadequate for the measurement of trace level species such as
metals and trace elements due to their low signal-to-noise ratio. As an alternative methodology, a
novel system was developed that combines the versatile aerosol concentration enrichment system
(VACES) (Kim et al. 2001a; Kim et al. 2001b) in tandem with a liquid impinger PM collector
(BioSampler, SKC West, Inc., Fullerton, CA) to effectively concentrate and collect both fine and
ultrafine particles in an aqueous suspension. In VACES/BioSampler tandem, ambient PM is first
grown to super-micrometer size droplets via condensational growth, and subsequently
concentrated by virtual impactor (VI), and finally collected as a highly concentrated liquid
suspension in the BioSampler, in a technique described in details by Kim et al (Kim et al. 2001a).
A similar system was also developed by Kidwell and Ondov (Kidwell and Ondov 2001). Daher
49 | P a g e

et al. (Daher et al. 2011) compared sampling techniques of filtration, impaction and
VACES/BioSampler tandem and an overall very good agreement in PM collection efficiency and
chemical composition was observed among the different samplers. In contrast, dissimilarities in
PM oxidative properties, measured by an ROS assay (Landreman et al. 2008), were apparent
between the BioSampler slurry and aqueous extracts of the filter and impactor samples, with the
former technology measuring considerably higher ROS levels, whereas filtering the
VACES/BioSampler slurries brought their ROS content to virtually identical levels to those of
the impactor and filter. The higher ROS values of the VACES/Biosampler were thus attributed to
potentially toxic insoluble PM species, which cannot be effectively extracted from filter or
impactor substrates with water. The study by Daher et al. demonstrated that both water-soluble
and water-insoluble portions of ambient PM were effectively captured in VACE/BioSampler
tandem, which is essential for toxicological measurements of PM redox properties, including
their reactive oxygen species (ROS) content. Despite the robust PM collection performance of
the VACES/BioSampler technology, several drawbacks may still exist that hinder its use for high
volume PM collection; these including excess water accumulation and clogging of the virtual
impactor acceleration, and especially collection nozzles, clogging of BioSampler nozzles and the
limited collector volume (20 mL) of the conventional BioSampler, which requires periodic
removal of the collected suspension samples. As a result, the operation of that system requires
some field supervision, which limits its unattended operation in long periods of time.

The current study presents a novel sampling technology for collecting aerosol directly
into liquid suspension using the saturation-condensation part of the VACES system. The new
methodology replaces the virtual impactor in VACES with a low pressure drop, high flow rate
impactor that can be operated for a prolonged periods of time, with no clogging issues observed
due to its relatively large nozzle size (0.3 cm wide and 1.35 cm long). This system can be used
for collection of fine or ultrafine PM directly into an aqueous suspension for subsequent
chemical analysis, as well as for collection of PM for in-vitro health studies, hence effectively
eliminating elaborate procedures of extracting collected particles from a filter substrate.
Moreover, as it will be discussed in subsequent sections, collecting particles directly into an
aqueous solution allows more efficient capturing of both water-soluble and water-insoluble
50 | P a g e

components of ambient PM and offers better collection efficiency for semi-volatile species, such
as semi-volatile organic compounds (SVOC) and ammonium nitrate.

4.2. Methodology
4.2.1. Description of the system and its components
The aerosol-into-liquid collector utilizes the saturation-condensation, particle-to-droplet
growth system developed for the VACES (Kim et al. 2001a; Kim et al. 2001b) that grows
sampled airborne particles to 3-4 µ m droplets, which are subsequently collected in a low-
pressure drop, high flow rate (200 L/min) impactor. The sampled air is first drawn through a
PM
2.5
or ultrafine PM impactor (already developed for the VACES) (Kim et al, 2001a; Misra et
al, 2002) before entering a saturator tank filled with ultrapure water produced by lab water
purification system (Millipore A-10, EMD Millipore, Billerica, MA) and maintained at 30° C, to
saturate the air. The particle-vapor mixture then enters two condensation tubes that are connected
to a circulating chiller maintained at about -3° C to promote condensational particle growth. The
grown aerosol leaves the condensational tubes at roughly 21-22° C. The flow rate through each
condensational tube is set at 100 L/min (Kim et al. 2001a; Kim et al. 2001b). The two air streams
are then combined before entering the impactor to provide the 200 L/min operational flow rate.
The schematic of the impactor is presented in Figure 4.1. The impactor is made of chemically
inert Teflon and consists of three stages: impaction stage, middle stage and suspension collection
stage. In the impaction stage, the droplet aerosol is accelerated through two slit impaction
nozzles (0.3 cm wide and 1.35 cm long) and is impacted on the lateral cylindrical surfaces of the
impactor, as shown in Figure 1. The impacting droplets gradually drain into the collection stage,
shaped like an inverse conical cavity, and form slurry samples at an approximate accumulation
rate of 4-6 mL/hour. The designed 50% cutpoint and pressure drop of the impactor are 1.5 µ m
and 0.02 atm (7 inches of water), respectively. In the middle stage, a micro-liquid level detector
(ML 101, Cosense Inc., Hauppauge, New York) is deployed for automatic measurement of the
liquid sample volume, a metric which is essential for converting aqueous PM concentrations to
airborne PM concentrations. The liquid level detector is an ultrasonic measurement system that
performs high speed, non-contact ultrasonic measurements with an accuracy of ± 0.1mm, and
introduces no disturbance of the sample because there is no physical contact with the liquid. The
volume of slurry collected is determined by the dimension of reservoir and the liquid level
51 | P a g e

distance to the sensor. A suspension outlet port is installed at the bottom of the collection stage,
thus automatic sample transport and removal can be achieved by using peristaltic pumps for
continuous and unattended operation.

Figure 4.1 Schematic of impactor

4.2.2. Laboratory evaluation tests of the impactor’s collection efficiency
The first part of the laboratory tests was to characterize the new impactor. Different sizes
of monodisperse fluorescent polystyrene latex (PSL) particles (Polyscience Inc., Sacramento, CA)
with sizes of 1, 1.5, 3 and 6 µm in diameter were generated by a HOPE nebulizer (B&B Medical
Technologies, Carlsbad, CA) to determine the experimental 50% cutpoint of the impactor. A 37-
mm Teflon filter (Teflo, Pall Corp., Life Sciences, 1-μm pore, Ann Arbor, MI) was connected
before and after the impactor to collect the generated PSL particles. At the end of each run, the
Teflon filters from both upstream and downstream of the impactor were extracted with 10 mL of
ethyl acetate to dissolve the hydrophobic fluorescent dye from the collected particles. The
quantities of the fluorescent dye in the extraction solutions were quantified by a Fluorescence
Detector (FD-500, GTI, Concord, MA) to determine particle concentration.
52 | P a g e

The second part of the laboratory tests was to evaluate the collection performance of the
aerosol-into-liquid collector using aerosols initially much smaller than the impactor’s cutpoint.
Fluorescent PSL particles in sub-micrometer size range (i.e., 42.5, 100, 300, 750 and 1000 nm)
were generated as test aerosols by the nebulizer and were collected directly as suspensions in the
impactor. Since these particles were smaller than the cutpoint of the impactor, they were first
grown in the saturation-condensation module. A silica gel diffusion dryer (Model 3062, TSI Inc.,
Shoreview, MN) was placed before the upstream and downstream Teflon filters to remove the
excess water. In a similar manner to that outlined previously, the Teflon filters from both
upstream and downstream of the impactor were extracted with 10 mL of ethyl acetate for
sequential fluorescence measurement. Moreover, the residuals on the impactor inlet and
impaction surfaces were carefully washed with ethyl acetate and extracted in order to estimate
the inlet and wall losses.

During the tests with 300 nm and 750 nm PSL particles, a third condensation tube
coupled with a virtual impactor followed by a BioSampler (i.e. the traditional VACES
/BioSampler tandem configuration) was added to the system and operated in parallel. The major
and minor flow rates of the VI were 100 L/min and 5 L/min, respectively. Since PSL particles
are mostly hydrophobic, suspension samples collected by BioSampler and the aerosol-into-liquid
collector were dried first by evaporating the collected water using laboratory HEPA-filtered,
purified compressed air. The remaining residuals were re-extracted with 10 mL of ethyl acetate
and the extraction solutions were defined as ―recovery‖ samples, representing the amount of
incoming PM that is actually collected by either the aerosol-into-liquid collector or the
VACES/BioSampler tandem. Recoveries from both BioSampler collection and impactor
collection were then measured by means of fluorescence analysis as described above. The system
configuration is presented in Figure 4.2.

53 | P a g e

Figure 4.2 System schematic for collection efficiency tests

Lastly, polydisperse glutaric acid and ammonium nitrate particles were generated to
simulate the ―real-world‖ particles. Ammonium nitrate was chosen because it represents the
predominant inorganic species in PM
2.5
in Los Angeles (Malm et al. 2004), which, importantly,
is also semi-volatile in nature and thus poses significant challenges to traditional aerosol
samplers (Mozurkewich 1993; Chang et al. 2000). Glutaric acid was selected because it is a
typical organic dicarboxylic acid formed by ozone photo-oxidation (Cruz and Pandis 1999; Saleh
et al. 2012). A small fraction (1 L/min) of the aerosols before and after the impactor were
diverted first into a diffusion dryer (Model 3062, TSI Inc., Shoreview, MN) and then passed
through a DustTrak (Model 8520, TSI Inc., Shoreview, MN) to measure mass concentrations and
thus characterize the impactor's collection efficiency.

4.2.3. Field tests of the aerosol-into-liquid collector and ambient PM sample collection
After completion of the laboratory evaluation, the system described in Figure 4.2 was
deployed in the Particle Instrumentation Unit (PIU) of the University of Southern California for
field evaluation tests during January and February 2013. The site is located in an urban area near
54 | P a g e

downtown Los Angeles, California, within 150 m of a major freeway (I-110), and thus represents
a typical urban mix of particles emitted by mostly traffic sources (Ning et al. 2007). The first part
of the field evaluation was the continuity (or unattended sampling) tests. Ambient PM was
continuously collected by the system for periods ranging between 48-72 consecutive hours. PM
number concentrations were measured concurrently upstream of the impactor with an Aerosol
Particle Sizer (APS 3320, TSI Inc., Shoreview, MN), and with an Optical Particle Size (OPS
3330, TSI Inc., Shoreview, MN) after the impactor. Suspensions collected in the impactor were
transferred to collection vials (LDPE bottles, Thermal Scientific, Rockwood, TN) by a peristaltic
pump (Mityflex mode 907, Anko product Inc., Bradenton, FL, using Tygon R-3603 PVC
laboratory tubing) automatically every four hours.

In order to assess the collection efficiency of the system for soot particles, ambient PM
2.5
were collected by aerosol-into-liquid collector and black carbon (BC) concentrations before and
after the impactor as well as in the ambient air were determined by means of a two-channel
Aethalometer (Model AE-22, Magee Scientific, Berkeley, CA). A silica gel diffusion dryer was
placed before the Aethalometer to remove excess water and return the aerosols to ambient
relative humidity levels.

Finally, ambient PM
2.5
samples were collected concurrently using either the conventional
filter or the VACES/BioSampler tandem as a reference, in parallel with the aerosol-into-liquid
collector system, in order to evaluate comparability in terms of PM
2.5
chemical composition. The
system configuration for these experiments was similar to the previous description in laboratory
evaluations, shown in Figure 4.2. A diffusion dryer followed by a Teflon filter were placed
immediately after the impactor and in parallel with the BioSampler (see Figure 4.2) to estimate
the PM mass loading in the BioSampler sample. The flow rate passing through the parallel
Teflon filter was 1 L/min. For filter collection, the 5 L/min minor flow of the virtual impactor
(VI) was directly connected to a diffusion dryer, followed by a 37-mm Teflon filter. All samples
were kept frozen at –20 ° C after the sampling and prior to chemical analysis.

The PM
2.5
samples (both slurries and filters) were subsequently analyzed for inorganic
ions, total major and trace elements as well as total organic carbon (TOC), water-soluble organic
55 | P a g e

carbon (WSOC), and macrophage reactive oxygen species activity (ROS). TOC/WSOC analysis
was restricted to the BioSampler and aerosol-into-liquid collector, since TOC analysis is not
possible on Teflon filters, and WSOC analysis was not performed on filter samples due to
insufficient collected PM mass loading. The slurry samples were divided into two portions: one
portion was ultra-filtered at 10 kilo-Dalton (kD) using centrifugal ultrafilters (Amicon Ultra-15
Centrifugal Filter Device, regenerated cellulose membrane in an all polypropylene device), and
was used for the analysis of water-soluble PM components and the other portion was left un-
fractionated and represented the total mass of chemical species in the PM samples collected by
the BioSampler or the aerosol-into-liquid collector. Sub-samples of the un-fractionated (total
elements, TOC and ROS) and ultra-filtered (inorganic ions, WSOC and ROS) slurry samples
were taken and distributed for chemical analysis. The slurry sub-sample for total elements was
acid digested prior to elemental chemical analysis. Filter samples were sectioned in quarters to
provide sub-samples for both total and water soluble analyses that paralleled the slurry
processing. One section was digested (microwave-aided acid digestion in Teflon bombs) for total
elemental analysis. The other filter sections were extracted with water to provide water-soluble
samples (ROS, inorganic ions). The inorganic ion composition was determined by ion
chromatography (IC). Measured ions included ammonium, nitrate and sulfate. TOC/WSOC
analyses were carried out using Sievers 900 Total Organic Carbon Analyzer (Stone et al. 2009).
Measurements for major and trace elements were conducted using high-resolution magnetic
sector inductively coupled plasma mass spectrometry (SF-ICPMS, Thermo-Finnigan Element 2)
(Zhang et al. 2008). The macrophage ROS assay employed a fluorogenic cell-based method to
investigate the production of ROS in rat alveolar macrophages (NR8383, American Type Culture
Collection) using 2, 7-dichlorodihydrofluorescein diacetate (DCFH-DA) as the fluorescent probe.
Details of the assay, extraction protocol, and detection methodology are explained in previous
study (Landreman et al. 2008). All the analysis mentioned above was also performed on lab
blanks, and all the chemical/ROS speciation data presented in this manuscript are blank corrected.

4.3. Results and discussion
4.3.1. Laboratory characterization of the impactor
The impactor is designed with an operating flow rate of 200 L/min with the measured
total pressure drop through the impactor of 0.02 atm (i.e., 7 inches of water). The 50% cutpoint
56 | P a g e

(d
50
) of impactor is determined by the Stokes number of a particle having a 50% probability of
capturing by impaction, Stk
50
, defined as follows for slit geometry impactors:

2
50
50
9
PC
d UC
Stk
W




[1]
where ρ
p
is the particle density, C
c
is the Cunningham Slip correction factor and µ is the
dynamic viscosity of the air, the values of which are 1000 kg/m
3
, 1 and 1.8× 10
-5
kg/m∙s,
respectively. d
50
represents the geometrical (or physical) diameter of a particle having a 50%
probability of impaction; U is the jet velocity at the nozzle (104 m/s in our case ) and W is the
rectangular impactor slit nozzle width equal to 0.3 cm (the length of the slit nozzle is 1.35 cm),
respectively. A typical Stk
50
value for rectangular geometry impactors is about 0.5 (Marple and
Willeke, 1976; Misra et al., 2002) and is used in our case. The impactor was designed with a
target 50% cutpoint of 1.5 μm in aerodynamic diameter to ensure near 100% collection of the
grown droplets, 3-4 μm in diameter (a typical size distribution of the grown aerosol is shown in
Figure 4.4(c) in a following section, having a geometric mean diameter of 3.8±0.6 μm). The
experimental collection efficiency curve, which is defined as the ratio of downstream particle
mass concentration to the upstream particle mass concentration of the impactor, is presented in
Figure 4.3. It can be seen that the 50% cutpoint is approximately 1.4 μm (corresponding to a
Stk
50
value of 0.44), agreeing very well with the theoretical prediction.

57 | P a g e

Figure 4.3 Collection efficiency as a function of aerodynamic particle diameter of the impactor

4.3.2. Laboratory evaluation of the collection efficiency of the aerosol-into-liquid collector
In order to evaluate the collection performance of the aerosol-into-liquid collector, PSL
particles of 42.5, 100, 300, 750 nm and 1 µ m, all smaller than the impactor's cutpoint, were used
as the test aerosols. The total collection efficiency is defined as the ratio of the downstream mass
concentration to upstream concentration of the impactor, which was determined by measuring
the fluorescent signal from corresponding filter extractions, following the methodology
described earlier in this manuscript. Moreover, the inlet and wall losses were estimated by
extracting the PSL residuals on inlet and impaction wall. The fraction collected in the suspension
was determined by subtracting estimated inlet and impaction wall losses from total collection
efficiency. Results from these tests are presented in Table 4.1, showing very good overall
collection for both fine and ultrafine size particle ranges, with average total collection efficiency
over 90%. The average inlet loss and wall loss of the impactor were quite low across the particle
size range tested (i.e. 2.5% for inlet loss and 8.1% for wall loss).

58 | P a g e

Table 4.1 Collection efficiency of PSL particles and BC
Particle
Total collection
efficiency
Inlet loss Wall loss
Fraction collected
in suspension
PSL 42.5nm 96.4% 4.5% 9.4% 82.4%
PSL 100nm 90.3% 2.1% 7.5% 80.7%
PSL 300nm 98.9% 1.1% 6.2% 91.4%
PSL 750nm 95.1% 2.8% 8.1% 84.1%
PSL 1 μm 99.1% 1.7% 9.3% 88.0%
Ambient BC 94.3% - - -

During the tests with 300 nm and 750 nm PSL particles, a comparison in the collection
efficiencies between the aerosol-into-liquid collector and VACES/BioSampler tandem was also
performed to demonstrate equivalence between these two samplers. Results of these tests are
presented in Table 4.2. Overall a very good agreement in the recoveries of the BioSampler and
the aerosol-into-liquid collector was observed (the average recovery was 86.6% for BioSampler
and 82.2% for the aerosol-into-liquid collector). Despite the uncertainties due to the evaporation-
extraction procedure discussed earlier, the recovery collection efficiencies of BioSampler and the
aerosol-into-liquid collector show an excellent agreement with the estimated fraction collected in
suspension considering inlet and wall losses from total collection efficiency (agreeing within 8%
with the values reported in Table 4.1).

Table 4.2 Particle recovery of BioSampler and aerosol-into-liquid collector

BioSampler
recovery
aerosol-into-liquid collector
recovery
PSL 300 nm 85.6± 2.1% 83.7± 3.5%
PSL 750 nm 87.6± 1.9% 80.6± 1.9%

In addition to monodisperse PSL particles, similar collection efficiency tests were
conducted with polydisperse glutaric acid and ammonium nitrate particles using a DustTrak for
mass concentration measurements upstream and downstream of the impactor. Table 4.3 shows
the results of mass concentration measurements. Generally, the mass collection efficiencies of
59 | P a g e

the impactor were above 95%, corroborating the collection efficiency results based on
monodisperse sub micrometer fluorescent aerosols discussed earlier.

Table 4.3 Collection efficiency of polydisperse particles
Glutaric acid Ammonium nitrate
Upstream concentration (mg/m
3
) 4.97± 0.07 1.62± 0.03
Downstream concentration (mg/m
3
) 0.14± 0.01 0.06± 0.01
Collection efficiency 97.1% 96.0%

The collection efficiency of black carbon (BC) is also included in Table 4.1. Overall the
collection efficiency of the BC collection in the aerosol-into-liquid collector is 94.3 ± 4.0%,
which is in excellent agreement with the collection efficiency reported for the PSL particles.
Both the collection efficiency results for PSL particles and BC validate the ability of the aerosol-
into-liquid collector to collect water-insoluble particles with near–ideal collection efficiencies.

4.3.3. Field evaluation of continuous, unattended operation of the aerosol-into-liquid
collector
Field evaluation of the system was conducted after completion of lab evaluation tests.
The system was deployed at PIU and ambient PM was continuously sampled for about 72 hours
(shown in Figure 4.4(a)) to validate the aerosol collector’s ability to operate unattended for
prolonged periods. The accumulation rate of the water from the aerosol droplets inside the
impactor’s collection cavity during the field tests varied between 4 to 8 mL per hour, depending
on ambient temperature and relative humidity, as well as the ambient particle number
concentrations. As shown in Figure 4.4(a), the overall collection efficiency of the system was
consistently above 90% (i.e. the average upstream number concentration was about on average
3600 ± 1168 particles/cm
3
while the downstream number concentration was 96 ± 52
particles/cm
3
, indicating an overall collection efficiency of 96.8%), which confirms the ability of
the system to operate unattended and for at least 2-3 days. Typical ambient particle size
distribution (Figure 4.4 (b)) during the sampling period, as well as the droplet size distribution
upstream and downstream of the impactor (Figure 4.4 (c)), demonstrate that over 95% of the
ambient particles (which are mostly smaller than the cutpoint of the impactor) are effectively
60 | P a g e

grown through saturation-condensation (note the very close agreement in number concentration
between ambient PM and grown droplets upstream of the impactor) and subsequently collected
in the impactor.

(a)
61 | P a g e

(b)

(c)
Figure 4.4 (a-c) Continuity test results and particle size distribution: a. Continuity test results
(Few gaps in downstream measurement is due to temporary malfunction of the OPS monitor); b.
62 | P a g e

Typical ambient particle size distribution at sampling site during sampling period; c. Droplet size
distribution upstream and downstream of the impactor.

4.3.4. Chemical results for PM
2.5
samples
4.3.4.1. Inorganic ions
Figure 4.5(a-b) presents the comparison between the filter, BioSampler and the aerosol-
into-liquid collector samples for the three major inorganic ions in PM
2.5
: nitrate, ammonium and
sulfate. Error bars represent the standard deviation of multiple samples. Overall, agreement
between filter, BioSampler and the aerosol-into-liquid collector was obtained, considering the
experimental uncertainties, with the exception of sulfate and nitrate levels measured on the filters,
which were lower than those of the aerosol-into-liquid collector by about 35-45%. It should be
noted that the sulfate concentrations in the samples were generally very low and close to the IC
detection limit which increases the measurement uncertainty. The samples were also analyzed by
SF-ICPMS which has a much lower detection limit compared to IC, and considering that sulfate
is the dominant species of soluble sulfur in ambient aerosols, the sulfate comparison after
conversion of measured sulfur to sulfate yields excellent agreement (see Figure 4.5a).

63 | P a g e

(a)

(b)
Figure 4.5 (a-b) Inorganic ions (nitrate, ammonium and sulfate) comparison between filter,
BioSampler and impactor: a. between BioSampler and aerosol-into-liquid collector. The p value
from t-test for nitrate, ammonium and sulfate is 0.35, 0.85 and 0.8, respectively; b. between filter
and aerosol-into-liquid collector. The p value from t-test for nitrate, ammonium and sulfate is
0.89, 0.97 and 0.31, respectively. Error bars represent the standard deviation of multiple samples.
The sulfate (SF-ICPMS) results in Figure 4.5(a) represent the estimated sulfate concentrations
based on the sulfur (S) measurement results by SF-ICPMS analysis.

Previous studies have demonstrated sampling artifacts for labile semi-volatile species,
such as ammonium nitrate and semi-volatile organic carbon (SVOC), in traditional filter
sampling methodologies (Turpin et al. 1994; Eatough et al. 2003). These artifacts are primarily
introduced by volatilization losses of these labile species on filters during prolonged sampling
periods. It has also been documented that pre-concentration of PM and/or collection of PM
directly into liquid (such as our case) may significantly reduce such artifacts for particulate
nitrate (Chang et al. 2000; Khlystov et al. 2005). Our observations are generally consistent with
64 | P a g e

these findings, given that volatilization loss of nitrate is probably reduced, if not eliminated, by
collecting PM directly into a suspension, manifested by the near-excellent agreement between
the nitrate levels measured in the aerosol-into-liquid collector and the BioSampler suspensions.
We also acknowledge that condensation of gas-phase nitric acid and/or ammonia onto the
sampled particles might occur during the saturation/condensation process. During the evaluation
of the VACES, which uses the same components to the sampler described in this paper to
achieve super-saturation, Khlystov et al (2005) reported a positive artifact of 2.5-7.5% from gas-
phase condensation, which was more pronounced in ammonia-limited conditions, whereas in
ammonia-rich environment these artifacts were negligible. Given that the samples in this study
were collected in Los Angeles under ammonia-rich conditions (Misra et al., 2004), the formation
of extra material in the system is expected to be insignificant.

4.3.4.2. Total (TOC) and water-soluble organic carbon (WSOC)
To investigate the equivalence in their organic carbon concentrations, suspensions
collected by BioSampler and aerosol-into-liquid collector were analyzed for both TOC and
WSOC. As mentioned in previous section, TOC analysis cannot be done on Teflon filter samples
and WSOC analysis was not performed on filter samples due to insufficient mass loadings.
Results for this analysis are presented in Figure 4.6. Error bars represent the standard deviation
of multiple samples. Both TOC and WSOC in samples collected by VACES/BioSampler tandem
were very close to those collected by the aerosol-into-liquid collector, considering the
experimental uncertainties. The average TOC concentration is 5.6 ± 0.8 µ g/m
3
for BioSampler
and 7.1 ± 1.7 µ g/m
3
for aerosol-into-liquid collector, while 5.1 ± 0.9 µ g/m
3
for BioSampler and
6.2 ± 1.8 µ g/m
3
for aerosol-into-liquid collector for WSOC measurements. The slightly higher
(roughly 15%) TOC and WSOC levels measured by the aerosol-into-liquid collector could be
due to its slightly smaller 50% cutpoint (1.4 μm in aerodynamic diameter) of the impactor
compared to that of the BioSampler (2 μm aerodynamic diameter at 5 L/min (Kim et al. 2001a)).
The droplet size distribution of the grown ambient aerosols, shown in Figure 4c, reveals that
there is approximately 10-15% of the total droplet mass between 1.4 - 2 μm, which is consistent
with the slightly higher concentrations of the aerosol-into-liquid collector. This may also explain
the slightly higher sulfate and nitrate levels of the aerosol-into-liquid collector compared to the
Biosampler, discussed in the previous section. Nonetheless these differences are not statistically
65 | P a g e

significant as indicated by the p-values from the t-tests; with a p-value of 0.66 and 0.73 for TOC
and WSOC, respectively.

Figure 4.6 TOC and WSOC comparison between BioSampler and aerosol-into-liquid collector.
The p value from t-test for TOC and WSOC is 0.66 and 0.73, respectively. Error bars represent
the standard deviation of multiple samples.

As mentioned in a previous section, we also acknowledge the formation of certain
secondary organic products by reactions after water condenses to form cloud droplets, as well as
the positive artifacts due to condensation of gas-phase semi-volatile organic species may occur
during sample collection. Based on our own field experiments, condensation of gas phase
organics may be at most in the 1.5 - 2 μg/m
3
range, so the reported particle–phase levels may be
overestimated by about 20%.

4.3.4.3. Metals and trace elements
Figure 4.7(a-b) illustrate the comparison for total metals and trace elements between filter,
BioSampler and the aerosol-into-liquid collector samples. Error bars represent the standard
66 | P a g e

deviation of multiple samples. Overall agreement for most metal species is observed between
samples collected by the BioSampler and the aerosol-into-liquid collector, with correlation slope
close to 1 and with high R
2
value (R
2
> 0.9). The average concentration ratio of the aerosol-into-
liquid collector vs BioSampler across all measured trace elements and metals is 1.03 ± 0.10. This
agreement is obtained over a large number of elements (a total of 24) covering a concentration
range that span 5 orders of magnitude. Regarding the filters vs aerosol-into-liquid collector
comparison, very good overall agreement is obtained for most total metals, with the average
aerosol-into-liquid collector vs filter concentration ratio of 1.04 ± 0.14. Species are also well
correlated with R
2
= 0.95. A few species were excluded from these correlations due to the fact
that the measured concentrations were very close to the method detection limit (i.e. analytical,
sampling, and handling) and the associated relative uncertainties were very high.

(a)

67 | P a g e

(b)
Figure 4.7 (a-b) Metals and trace elements comparison between filter, BioSampler and impactor
samples: a. between filter and aerosol-into-liquid collector. The p value from t-test is 0.71; b.
between BioSampler and aerosol-into-liquid collector. The p value from t-test is 0.94. Error bars
represent the standard deviation of multiple samples. Note: inserted plot shows the correlations
on log-scale. The line Y=X is also included for visibility purposes.

4.3.4.4. Macrophage ROS activity
In addition to the chemical analysis discussed previously, a portion of PM
2.5
samples
collected by the filter, BioSampler and the aerosol-into-liquid collector were analyzed for
macrophage ROS activity. This cellular macrophage ROS assay provides an estimate of the
overall oxidative potential of PM by measuring their reactive oxygen species (ROS) content
(Verma et al. 2009; Shafer et al. 2010; Daher et al. 2012). Figure 4.8(a-b) represents the results
from these comparisons. Error bars represent the standard deviation of multiple samples. The
ROS levels are expressed on a per volume (of air) based activity metric in µ g Zymosan units/m
3

of air. As shown in Figure 4.8(a), the ROS activity of suspension samples collected by aerosol-
into-liquid collector is generally very close to that from BioSampler samples, for both unfiltered
68 | P a g e

and filtered suspension samples (the latter representing the ROS activity presumably associated
with water-soluble species), and within the experimental uncertainty.

(a)

69 | P a g e

(b)
Figure 4.8 (a-b) ROS activity comparison between filter, BioSampler and aerosol-into-liquid
collector: a. between BioSampler and aerosol-into-liquid collector. The p value from t-test for
unfiltered and filtered slurry is 0.75 and 0.68, respectively; b. between filter extractions, filtered
and unfiltered aerosol-into-liquid collector slurries. The p value from t-test between filter
extraction and filtered slurry is 0.71. Error bars represent the standard deviation of multiple
samples.

When comparing the ROS levels between filter and aerosol-into-liquid collector
suspensions, however, we see that the ROS activity of the latter is higher on average by about 40%
(i.e., average ROS activity on filter samples is 254± 110 µ g Zymosan units/m
3
of air while
365± 116 µ g Zymosan units/m
3
of air for unfiltered aerosol-into-liquid collector slurries). In a
previous study, Daher et al. (Daher et al. 2011) demonstrated that the ROS activity of PM
2.5
samples collected by the VACES/BioSampler was higher than that of parallel-sampling filters,
but filtering the BioSampler slurry brought the ROS levels to almost identical values to the filter-
based ROS. Similarly to the results presented in that study, filtering the aerosol-into-liquid
collector slurries to remove insoluble PM species decreases the ROS activity to about the same
70 | P a g e

levels as those of the filter extracts. These findings underscore the contribution of insoluble
species to the overall PM
2.5
redox activity. Given the rather small number of our samples, the
observed difference between the ROS levels on slurries vs filters does not reach statistical
significance, as we noted earlier. Nonetheless the reported trends are consistent with the more
rigorous data of Daher et al (2011) and illustrate the advantages of collecting PM directly into
liquid suspension compared to conventional filter sampling, particularly for evaluating
toxicological properties of PM such as redox activity. By eliminating extraction of PM from a
filter substrate, sampling methodologies such as the aerosol-into-liquid collector, capture
effectively both water-soluble and water-insoluble redox active PM components in suspension
samples. Thus this approach is likely to be more appropriate for collecting PM for in-vitro and
in-vivo assays in the future.

4.4. Summary and conclusions
This study describes the development, laboratory and field evaluation of a new aerosol
sampler operated in conjunction with a saturation-condensation system for growing the particles
to super-micrometer sizes, capable of high volume collection of fine and ultrafine particles
directly into aqueous suspensions. This sampling methodology was developed as an alternative
to the VACES/BioSampler tandem for collection of PM directly into liquid suspensions. The
performance of the VACES/BioSampler tandem as a tool for collecting highly concentrated fine
and ultrafine PM into liquid suspensions was demonstrated previously (Kim et al. 2001b;
Khlystov et al. 2005). However, this system is not suitable for long-term sampling periods as it
requires attendance by trained personnel. The aerosol-into-liquid collector improves this system
by replacing the virtual impactor with a newly designed low pressure drop impactor. This
modification eliminates clogging of the BioSampler and virtual impactor nozzles as a result of
excessive particle accumulation that comes with prolonged sampling periods, and therefore it
makes it possible to achieve continuous and unattended collection of concentrated suspensions
for at least 2 to 3 days. The collection performance of the system was evaluated in laboratory
experiments using different types of particles, demonstrating excellent collection efficiency for
hygroscopic and hydrophobic test aerosols.

71 | P a g e

The chemical composition of PM collected in the suspension with this sampler was quite
similar to those obtained using the traditional filtration as well as the VACES/BioSampler
system, for PM species including inorganic ions (except of nitrate), selected elements and metal
species, and total and water soluble organic carbon. Moreover, the field experiments indicated
that volatile species such as nitrate are better preserved when the particles are collected directly
into suspension compared to conventional filtration. Similar results were observed based on the
ROS assays, demonstrating that this system can provide concentrated suspensions for use in in-
vitro studies that contain important water-soluble and water-insoluble redox active species, a
unique advantage comparing to using filters to determine the oxidant properties of PM. Although
there are some limitations, such as potential positive artifacts resulting from condensation of gas
phase semi-volatile species onto the sampled particles, the new aerosol-into-liquid collector is
highly adaptable and versatile for potential applications in long-term, in-situ measurement of PM
chemistry by overcoming the drawbacks of traditional sampling methods, both in terms of
detection limit and time resolution.

4.5. Acknowledgement
This study has been supported by South Coast Air Quality Management District (AQMD)
through award number #11527, and the National Institutes of Health (NIH) through
Grants R01AI065617-13, and R21AG040753-02 to the University of Southern California (USC).
The research described herein has not been subjected to the agency’s required peer and policy
review and therefore does not necessarily reflect the views of the agency, and no official
endorsement should be inferred. Mention of trade names or commercial products does not
constitute an endorsement or recommendation for use. The authors would like to thank Dr.
Massoud Pirbarzari and Woonhoe Kim (USC) for their assistance in the chemical analysis.

72 | P a g e

Chapter 5 Development of a Technology for Online
Measurement of Total and Water-soluble Copper (Cu)
in PM
2.5

This chapter is based on the following publication:
Wang, D., Shafer, M. M., Schauer, J. J., & Sioutas, C. (2014). Development of a technology for
online measurement of total and water-soluble copper (Cu) in PM
2.5
. Aerosol Science and
Technology, 48(8), 864-874.

5.1. Introduction
Epidemiological and toxicological studies have demonstrated robust associations between
ambient particulate matter (PM), especially fine and ultrafine PM, and adverse health outcomes
(Delfino et al., 2011; Li et al., 2013; Pope and Dockery, 2006). A broadly accepted hypothesis is
that PM-induced toxicity is driven by the interaction of PM with cells and macrophages to
generate reactive oxygen species (ROS), which change the redox status of the cells (Donaldson
et al., 2002b; Xia et al., 2004). There is increasing interest in exploring the emission and
concentrations of redox-active metal species, such as Cu and Fe, due to the associations between
these species and ROS activities of size-fractioned PM as described in previous studies (Daher et
al., 2012; Nishida et al., 2002; Shafer et al., 2010; Wang et al., 2013b; Zheng et al., 2006).
Therefore, measurement of atmospheric metal and elements in high time resolution remains an
active topic of aerosol research.

Conventional PM sampling methodologies typically involve collection of particles onto
substrates by filtration or inertial impaction. Although widely used and adopt to date, these
approaches have many drawbacks, including long turn-around time for processed results of
chemical analysis, incomplete extraction of insoluble PM-bound species, as well as potential
sampling artifacts for labile species such as semi-volatile organic compounds (SVOC). In order
to overcome these drawbacks, advanced PM sampling technologies have been developed since
1990s, in which PM is directly collected in liquid phase as suspension samples (Buhr et al., 1995;
Karlsson et al., 1997; Kidwell and Ondov, 2001; Weber et al., 2001). These technologies have
73 | P a g e

been well documented in achieving reliable measurements of bulk chemical species such as
organic carbon (Sullivan et al., 2004) and secondary ions (Orsini et al., 2003). However, most of
these technologies may be seemingly inadequate for the measurement of metals and trace
elements due to their low signal-to-noise ratios. As an alternative technique, a high flow rate
Aerosol-Into-Liquid Collector was developed to provide concentrated slurries of fine and/or
ultrafine PM (Wang et al., 2013a). This system operates at a flow rate of 200 liters per minute
(L/min) and utilizes the saturation-condensation, particle-to-droplet growth component of the
Versatile Aerosol Concentration Enrichment System (VACES) (Kim et al., 2001a, 2001b),
growing fine or ultrafine PM to 3-4 µ m droplets, in conjunction with a newly designed impactor,
in which grown particles are collected gradually forming highly concentrated slurries. In addition,
the new Aerosol-Into-Liquid Collector is highly adaptable and versatile for potential applications
in long-term, in-situ measurement of PM chemistry, especially for trace level metal and elements
measurement since slurry samples collected by this system are highly concentrated.

Significant progress has been made towards advancing these sampling techniques into
fully automated chemical composition measurement instruments (Kidwell and Ondov, 2004;
Orsini et al., 2003; Sullivan et al., 2004). Multiple techniques have been developed for metal and
element analysis, such as X-ray fluorescence (Creatchman, 1999), instrumental neutron
activation (Ondov et al, 1992), and inductively coupled plasma mass spectrometry (ICP-MS)
(Montaser, 1998). Some of these techniques have been commercialized for rapid detection of the
ambient aerosol particle’s elemental composition. However, these techniques are not without
drawbacks. Their application and flexibility are limited by complexities and high costs
associated with installation and operation. There is a need for the development of a modular,
relative inexpensive, easily operated and maintained platform for online metal and elements
measurements. Ion Selective Electrodes (ISEs) are widely used for determining the concentration
of selected ions in aqueous solution (Gupta and D’Arc, 2000; Schwarz et al., 2000). ISEs are
generally inexpensive, flexible, and have a high measurement accuracy (errors of less than 3%
(Rundle, 2000)). The useable concentration range over which most commercially available ISEs
function can exceed several orders-of-magnitude. Thus, ISEs are potentially valuable instruments
for continuous monitoring of selective metal species, especially in providing improved temporal
resolution of elemental species concentrations in the atmosphere.
74 | P a g e

Cu is commonly observed as an abundant metal species in atmospheric PM (Cheung et
al., 2012; Hueglin et al., 2005; Pastuszka et al., 2010). It is primarily generated from various
vehicular sources including fuel combustion, brake wear and tire abrasion depending on PM size
fractions (Hjortenkrans et al., 2007; Lin et al., 2005; Manoli et al., 2002), as well as copper
mining/smelting sources (Prabhakar et al., 2014). More importantly, numerous recent studies
have demonstrate strong associations between Cu and ROS activities of PM (Charrier and
Anastasio, 2012; Daher et al., 2012; Nishida et al., 2002; Shinyashiki et al., 2009; Wang et al.,
2013b), suggesting that it may potentially contribute to overall PM-induced toxicity. Therefore,
Cu was selected as the primary target metal species in the current study.

This manuscript describes the development of a novel technique for field online
measurement of Cu in ambient PM
2.5
. The novel technique employs the Aerosol-Into-Liquid
Collector (Wang et al., 2013a) as a PM sampling module, in which ambient PM
2.5
are directly
collected as concentrated slurry samples. Slurries collected in the sampling module are
subsequently transferred into the Cu measurement module, in which the water-soluble/total Cu
concentration is determined by a cupric ISE. This system can provide accurate and robust
measurements of Cu concentration in ambient PM as demonstrated in this paper that effectively
equivalent data are obtained by the cupric ISE measurements and off-line inductively coupled
plasma mass spectrometry (ICP-MS) analysis. Moreover, the new system can be operated
continuously for at least several days with minimal maintenance or supervision. Both laboratory
and field evaluations of this new Cu measurement system suggest that it is an effective and
valuable technology for PM collection and characterization of Cu in ambient aerosols.

5.2. Methodology
5.2.1. Ion Selective Electrodes (ISEs)
Ion selective electrodes measure the activity of ions in aqueous solution. Typically an ISE
incorporates a membrane, which selectively permits target ions to migrate through it. The
potential difference between internal phase of the membrane and a stable reference system is
then measured by a high impedance millivolt meter. The relationship between ionic
concentration (activity) and the electrode potential is given by the Nernst equation:
75 | P a g e

0
(2.303 / ) ( ) E E RT nF Log A    [1]

where E is the total potential difference (in mV) between the sensing and reference electrodes. E
0

is a constant which is characteristic of the particular ISE and n is the charge on the ion. T is the
absolute temperature (K). R and F are gas constant and Faraday constant, the values of which are
8.314 J/degree/mole and 96,500 Coulombs, respectively. Log (A) is the logarithm of the activity
of the measured ion.

The performance and accuracy of any ISE measurement can be affected by several
factors, including ionic strength, temperature and pH of samples. As discussed previously, the
ISE measures activity of ions under an equilibrium condition at the membrane surface. Ion
activity is a function of ionic strength. In order to minimize biases between samples (and
standards), a buffer is added to both samples and standards to effectively ―swamp-out‖ these
biases. The potential difference measured by the ISE is also a function of temperature (refer to
the Nernst equation above). Typically, a 1 degree change in temperature will result in 4% change
in response for divalent ions such as Cu ions (Rundle, 2000). Meanwhile, the formation of
insoluble Cu(OH)
2
limits the pH range over which cupric ISE measurements can be made (when
pH is greater than 7), but this is not normally a factor for typical environmental samples,
especially if an Ionic Strength Adjuster (ISA) buffer is employed. Other factors such as
interfering ions (i.e. Bromide and Silver ions for cupric ISE) may also affect the measurement
accuracy of ISEs. However, the levels of these ions in ambient aerosols are well below the
threshold at which they may bias the cupric ISE measurements.

5.2.2. Description of the system and its components
The online Cu measurement system utilizes the Aerosol-Into-Liquid Collector described
previously (Wang et al., 2013a) as particle collection module and a cupric ISE as the Cu
measurement module. Briefly, in Aerosol-Into-Liquid Collector, the sampled aerosol is first
drawn into a saturator tank filled with ultrapure water produced by lab water purification system
(Millipore A-10, EMD Millipore, Billerica, MA) and maintained at 30 ° C, to saturate the air. The
particle-vapor mixture then enters two condensation tubes maintained at about -3° C to promote
76 | P a g e

condensational particle growth. Sampled aerosol is cooled to about 21-22 ° C, and the resulting
super saturation grows all particles to droplets in the 3-4 μm range. The grown droplets are then
collected in a high flow rate impactor and form concentrated slurry sample. In the impactor, a
micro liquid level detector (ML 101, Cosense Inc., Hauppauge, New York) is deployed for
automatic measurement of the liquid sample volume, a metric which is essential for converting
aqueous PM concentrations to airborne PM concentrations. The volume of slurry sample was
determined by converting level distance to the sensor to volume based on the dimension of slurry
reservoir of Aerosol-Into-Liquid Collector. The collected samples were then transferred to
measuring vials (polypropylene, Thermal Scientific, Rockwood, TN) by a computer controlled
syringe pump (Model C 3000, 3K step, TriContinent Inc., Grass Valley, CA, using FEP
laboratory tubing) with a 4-way distribution valve and a 12.5 mL PTFE syringe automatically
after each sample collection. Before Cu measurements, cupric ionic strength adjuster (ISA) (0.5
M NaNO
3
solution) is transferred into the sample vial and mixed with the sample by a volume
ratio of 2%, in order to provide a constant background ionic strength for sample and standards.
Sample acidification can be added by mixing the slurry sample with 10% v/v nitric acid
(resulting a pH value of 2-3), in order to distinguish water-soluble Cu measurement (without
acidification) and total Cu measurement (with acidification). The Cu concentration is determined
by a cupric combination ISE (Model 9629BNWP, Thermo Fisher Scientific Inc., Waltham, MA)
which is connected to a high impedance millivolt meter (Orion Star A214, Thermo Fisher
Scientific, Waltham, MA). The cupric combination ISE has the sensing and reference half-cells
built into one electrode. In parallel of the cupric ISE, a stirrer probe (Orion stirrer probe, Thermo
Fisher Scientific Inc., Waltham, MA) is deployed to stir the samples thoroughly during Cu
measurement. The Cu concentration was measured by the cupric ISE over a period of 5 minutes.
After each measurement, the Aerosol-Into-Liquid Collector, the cupric ISE and system sample
lines were purged with ultrapure water to remove any carryover from the previous sample. The
cupric ISE was maintained in ultrapure water between measurements. The schematic of the
system is presented in Figure 5.1.

77 | P a g e

Figure 5.1 Schematic of Cu measurement system

5.2.3. Laboratory evaluation of Cupric ISE and system performance
The first part of laboratory test was the calibration of cupric ISE. The calibration
standards (from 10 ppb to 1000 ppb) were prepared by series dilution of 1000 mg/L stock
solution that was made in the laboratory using analytical grade salts of cupper (Cu(NO
3
)
2
, Fisher
Scientific). Before Cu measurement, cupric ionic strength adjuster (ISA) was added to standard
solutions by a volume ratio of 2%. Calibrations were performed multiple times to evaluate the
consistency of cupric ISE response.

In the second part of laboratory tests, the impact of different factors on cupric ISE
measurements, including ionic strength, temperature and pH of samples were evaluated, as
described in previous section of this manuscript. Typically it is suggested that cupric ionic
strength adjuster (ISA) should be added into samples before measurements. In order to evaluate
the impact of ionic strength, comparison of measurements using and without using ISA were
performed by measuring same standard samples with and without adding ISA solution (the
volume ratio of adding ISA was 2%). Secondly, to evaluate the measurement error introduced by
78 | P a g e

changing sample temperature, Cu standard solution of 100 ppb was injected into the sample vial
and temperature changes of sample was achieved by heating it in a water bath. The potential
difference and standard solution temperature were measured in parallel by cupric ISE and a
pH/temperature probe (Orion 9107BNMD, Thermo Fisher Scientific Inc., Waltham, MA),
respectively. Standard solutions (i.e. 10 ppb, 100 ppb, 1000 ppb) were acidified to different pH
values ranging from 1.7 to 6.8 by adding 1% v/v nitric acid (trace metal grade, VWR) into
standard solutions, and measurements were taken by cupric ISE in order to evaluate effect of pH
on measurement accuracy. The pH of samples was determined by a pH/temperature probe (Orion
9107BNMD, Thermo Fisher Scientific Inc., Waltham, MA) in parallel with Cu measurements.

Finally the online Cu monitor’s performance was evaluated by collecting and measuring
samples of lab generated Cu(NO
3
)
2
aerosols with known mass concentration. Cu(NO
3
)
2
particles
were generated as test aerosols by atomizing Cu(NO
3
)
2
solutions of known and variable
concentrations using a HOPE nebulizer (B&B Medical Technologies, Carlsbad, CA) and were
collected directly as slurry samples in the Aerosol-Into-Liquid Collector. A small fraction (1.5
L/min) of the aerosols before entering the Aerosol-Into-Liquid Collector were diverted first into
a diffusion dryer (Model 3062, TSI Inc., Shoreview, MN) and then passed through a Scanning
Mobility Particle Sizer (SMPS 3936, TSI Inc., Shoreview, MN) coupled with a Condensation
Particle Counter (CPC 3022A, TSI Inc., Shoreview, MN) to measure particle size distributions.
The mass concentration of generated aerosols was estimated by multiplying the density of
Cu(NO
3
)
2
(i.e. 3.05 g/cm
3
) (Haynes, 2013) with volume concentration obtained from the
number-based particle size distributions, with the assumption of spherical particles. The Cu
concentration in the slurry samples was measured by the cupric ISE after every 3-4 hours
collection (after sufficient volume of slurry was collected for the measurement). The measured
Cu concentration was then compared with theoretical Cu concentration, which was calculated by
mass concentration of generated aerosol, sampled air volume and volume of slurry samples, as
shown below:

Cu conc. (theoretical) = Mass conc. (generated) × Volume of Air / Volume of slurry [2]

79 | P a g e

5.2.4. Field evaluation of Cu measurement system:
After completion of the laboratory evaluation, the system described in Figure 1 was
deployed at the Particle Instrumentation Unit (PIU) of the University of Southern California for
field evaluation tests from December 2013 to March 2014. The ambient atmospheric monitoring
site is located in an urban area near downtown Los Angeles, California, within 150 m of a major
freeway (I-110), and thus represents a typical urban mix of particles emitted by mostly traffic
sources (Ning et al., 2007). Ambient PM
2.5
samples were collected concurrently using the filter
collection, in parallel with the Aerosol-Into-Liquid Collector system, in order to evaluate
comparability of the novel and traditional sampling methods in terms of PM
2.5
elemental
composition. To implement the traditional filter sampling comparison, a third condensation tube
coupled with a virtual impactor was added to the system and operated in parallel. For filter
collection, the 5 L/min minor flow of the virtual impactor (VI) was directly connected to a
diffusion dryer (Model 3062, TSI Inc., Shoreview, MN), followed by a 37-mm Teflon filter
(Teflo, Pall Corp., Life Sciences, 1-μm pore, Ann Arbor, MI). Detailed information of system
configuration was reported in a previous study (Wang et al., 2013a). For each set of samples the
sampler was operated for 4 hours to ensure adequate mass loading (primarily for the off-line
sample for subsequent comparative analysis). The slurry samples collected by Aerosol-Into-
Liquid Collector were divided into two portions: in one portion, Cu was measured directly online
using the cupric ISE. Measurements were taken before and after acidification of samples in order
to acquire both water-soluble and total Cu data. The other portion (together with the parallel
filter off-line samples) were stabilized and subsequently analyzed for major and trace elements
by magnetic sector ICP-MS. All samples were kept frozen at –20 ° C after the sampling and prior
to chemical analysis.

The slurry samples preserved for chemical analysis were processed as follows: A sub-
sample was separated and ultra-filtered at 10 kilo-Dalton (kD) using centrifugal ultra-filters
(Amicon Ultra-15 Centrifugal Filter, regenerated cellulose membrane in an all polypropylene
device). The elements passing a 10 kD filter can be considered nearly truly dissolved and as such
is an ideal fraction to compare with the ISE (which is sensitive only to ions and labile
complexes). Another sub-sample of the slurry was left un-fractionated and acid-digested to
80 | P a g e

determine the total mass of chemical species in the PM samples collected by the Aerosol-Into-
Liquid Collector. This fraction is directly comparable to the ISE data acquired post-acidification.
Filter samples were sectioned in quarters to provide sub-samples for both total and water soluble
analyses that paralleled the slurry processing. One section was digested (microwave-aided acid
digestion in Teflon bombs) for total elemental analysis. The other filter sections were extracted
with water to provide water-soluble samples.

Measurements of major and trace elements were conducted using high-resolution
magnetic sector inductively coupled plasma mass spectrometry (SF-ICPMS, Thermo-Finnigan
Element 2). The Thermo-Finnigan Element 2 (with fast scanning magnet and Pt guard electrode)
was interfaced with a quartz cyclonic spray-chamber fitted with an ESI low-flow (80 µ L/min)
Teflon micro-concentric nebulizer. The complete analytical system is located within a trace
metal clean room. Instrumental detection limits (3-sigma) were in the range of 0.01 to 20 ng/L
(0.01-20 pg/m
3
of air) for most trace elements and in the range of 5 to 50 ng/L (5-50 pg/m
3
of air)
for major elements. Details of the ICP-MS analysis are presented in a previous study (Zhang et
al., 2008).

Lastly, the Cu monitor was operated continuously and total Cu concentration was
measured for periods ranging between 4-7 consecutive days in March 2014. PM
2.5
slurry samples
were collected every 4 hours, with an approximate slurry accumulation rate of 4-5 mL/hour. A
one-point standard check (using 100 ppb Cu standard solution) was performed daily to confirm
the cupric ISE performance. The sensing surface of the ISE was polished every two days and ISE
was re-calibrated after each polishing. The trends in ISE-determined Cu concentrations were also
referenced to PM
2.5
mass concentration data reported by the South Coast Air Quality
Management District (AQMD), measured at a monitoring site located approximately 5 km north
of our sampling site in downtown Los Angeles during the sampling period.

5.3. Results and discussion
5.3.1. Calibration of cupric ISE:
The calibration of ISE is normally carried out by immersing cupric ISE in a series of
standard solutions with known concentration. Based on the Nernst equation (described in
81 | P a g e

previous section), the potential difference (mV reading) obtained by cupric ISE is linearly
correlated with the logarithmic of sample concentration. However, the response of ISE to
varying concentrations becomes progressively lower as the concentration decreases below a
certain level, resulting in a non-linear correlation in the lower concentration range (Rundle,
2000). Calibration results of cupric ISE are presented in Figure 5.2. Generally the cupric ISE
responds linearly between 50 ppb to 1000 ppb (corresponding to an airborne Cu concentration of
17 to 333 ng/m
3
), with a R
2
value of 0.92, while response becomes non-linear in the
concentration range of 10-60 ppb, thus a more detailed calibration curve needs to be prepared
separately at this lower range. Previous studies in the Los Angeles Basin have reported that the
average Cu concentration in PM
2.5
varies between 18.5-43.1 ng/m
3
(Daher et al., 2011; Hu et al.,
2008; Wang et al., 2013a, 2013b), corresponding to a slurry concentration of 60-150 ppb (For
lower concentration levels, a separate calibration curve in non-linear ranges can be applied to
achieve accurate measurements), respectively. This Cu concentration level is within the linear
response range of the proposed cupric ISE, which ensures the accuracy of Cu measurement.
Moreover, the slope of the calibration curve in linear response region is about 25.3± 0.1
mV/decade, which agrees with the recommended value of 25-28 mV/decade in the manual of
cupric ISE. Multiple calibration tests were performed and the calibration curve was very
reproducible, illustrating that the obtained calibration curve is reliable in determining the Cu
concentration in real-world PM samples.

82 | P a g e

Figure 5.2 Calibration curve of cupric ISE. The top x-axis represents the Cu concentration in
standard solution. The bottom x-axis represents the corresponding airborne Cu concentration
based on a sampling flow rate of 200 L/min and a slurry collection rate of 4 ml per hour. Error
bars represent the standard deviations of multiple calibration tests (at least 5 tests were
performed)

5.3.2. Effect of ionic strength, temperature and pH of sample
As discussed earlier, the ionic strength of measured samples may significantly affect the
cupric ISE measurement, especially in high concentration ranges. This is because the inter-ionic
interactions (both positive and negative) tend to reduce the mobility of ions thus fewer ions will
diffuse through the membrane of ISE, resulting under-estimation of real concentration in bulk
solution (Rundle, 2000; Schwarz et al., 2000). In most of applications, specific ISA solutions are
added before measurements to provide constant ionic strength background. Table 5.1 presents
the comparison of Cu measurement results between adding/without adding ISA. As shown in
Table 5.1, potential differences measure by cupric ISE are consistently lower when ISA are not
added, comparing to those obtained after adding ISA. These results indicate that the error of
83 | P a g e

under-estimation may occur when ISA is not added to the samples. Moreover, the error
introduced by ionic strength is more substantial in higher concentration range (i.e. 6.3% error in
10 ppb standard solution while 34.6% for 1000 ppb standard solution). Thus, adding ISA
solution prior to Cu measurement is critical in obtaining accurate measurements, and was
followed in all other tests described in this manuscript.

Table 5.1 Cu measurements showing the necessity of adding ISA
Conc. (ppb) Without ISA (mV) With ISA (mV) Error
10 -43.3± 0.2 -46.2± 0.1 6.3%
100 -31.9± 0.1 -36.8± 0.2 13.3%
1000 -8.4± 0.1 -12.8± 0.1 34.6%

According to the Nernst equation, the potential difference measured by ISE can vary due
to changes in temperature. The effect of sample temperature to cupric measurement is presented
in Figure 5.3. Generally the potential difference obtained by cupric ISE increases linearly
(R
2
=0.98) with temperature which has an average slope of 0.53 mV/° C, illustrating that
maintaining constant sample temperature is essential in Cu measurements. The temperature of
PM
2.5
slurry samples collected by Aerosol-Into-Liquid Collector are normally about 22.2± 0.2 ° C,
resulting in about ± 0.1~± 0.2 mV uncertainties. Therefore, the effect of sample temperature on
cupric ISE measurements was neglected in all laboratory tests, as well as in field sample
collection and measurements described in the following sections.

84 | P a g e

Figure 5.3 Effect of sample temperature on Cu measurements. Standard solution of 100 ppb was
used as sample solution. Error bar represents the standard deviation of multiple reading (3
readings for each data point).

Moreover, the pH of measured sample also has substantial impact on cupric ISE
measurements due to the formation of the formation of insoluble Cu(OH)
2
, which reduces the
level of free cupric ion concentration. Results of cupric ISE measurements under different pH
values are summarized in Table 5.2. Generally the cupric ISE measurements remained very
consistent when sample pH varied between 1.7 and 6.8, indicating the impact of pH on Cu
measurement is relatively low in this pH range, especially after sample acidification. The pH of
PM
2.5
slurry samples collected by Aerosol-Into-Liquid Collector is normally about 6.7-7.2.
Therefore, the proposed system is capable of measuring both water-soluble Cu (if samples are
not acidified), as well as total Cu (if samples are acidified) in ambient PM, an additional
advantage compared to other online element monitors.

85 | P a g e

Table 2.2 Cupric ISE measurements under different pH values

Cupric ISE measurements (mV)
pH=1.7 pH=2.5 pH=3.9 pH=5.1 pH=6.8
10 ppb -46.6± 0.1 -45.9± 0.2 -45.8± 0.2 -46.9± 0.1 -47.0± 0.2
100 ppb -36.4± 0.3 -36.8± 0.2 -37.0± 0.1 -36.2± 0.2 -36.6± 0.1
1000 ppb -13.1± 0.1 -12.5± 0.2 -12.8± 0.1 -13.2± 0.2 -12.5± 0.1

It is noted that the performance of cupric ISE is also influenced by interference metal
species, such as bromide and silver ions. Considering that concentrations of these metal species
are usually below detection limit in ambient PM
2.5
samples (Hu et al., 2008; Saffari et al., 2013;
Wang et al., 2013b), the measurement error introduced by interference species are negligible in
the experiments described in this manuscript.

5.3.3. Collection and measurement of lab generated Cu(NO
3
)
2
aerosols
In order to demonstrate system collection and measurement performance, Cu(NO
3
)
2

aerosols were generated with known mass concentration (calculated based on particle size
distribution and density, as described in previous sections) and subsequently collected as slurry
samples in Aerosol-Into-Liquid Collector. The Cu concentration in collected slurry samples were
determined by the cupric ISE. Comparison between theoretical and measured Cu concentration is
presented in Figure 5.4, showing very good agreement between generated and measured Cu
concentration, with correlation slope close to 1 and with high R
2
value (R
2
> 0.9). This
agreement is obtained over a concentration range from 100 ppb to over 3000 ppb, indicating an
excellent collection efficiency and measurement accuracy of the Cu monitor. It must be noted
that by the nature of this study design, the lowest Cu concentration generated in this test is still
slightly higher than typical Cu concentration in ambient PM
2.5
samples reported previously (i.e.
63 to 147 ppb). However, samples in this test covered a very wide concentration range while the
agreement between generated and measured Cu remains consistent, corroborating the
effectiveness of this technology for measuring Cu in ambient PM.

86 | P a g e

Figure 7.4 Cu generation test results. Error bar represents the standard deviation of multiple
measurements (3 measurements for each sample)

5.3.4. Field evaluations
5.3.4.1. Chemical results for PM
2.5
samples
Figures 5.5(a-b) present a comparison for total and water-soluble elements between the
filter-based collections and the Aerosol-Into-Liquid Collector samples. Overall, excellent
agreement for most total elements is observed, with a correlation slope close to 1 and with high
R
2
value (R
2
= 0.85). The median concentration ratio of the Aerosol-Into-Liquid Collector vs.
filter across all measured total elements was 1.02 ± 0.12. Similarly for the water-soluble element
comparison, very good overall agreement is observed with an average Aerosol-Into-Liquid
Collector vs. filter concentration ratio of 1.08 ± 0.14. Species are also well correlated with R
2
=
0.71. A few species (i.e. Br, Pb and etc.) were excluded from these correlations because the
measured concentrations were very close to the method detection limit and the associated
relative uncertainties were very high. These outcomes are consistent with observations reported
in a previous study (Wang et al., 2013a) and confirm that the Aerosol-Into-Liquid Collector is a
87 | P a g e

robust and versatile tool for aerosol collection with great potential for applications in long-term,
in-situ measurement of PM chemistry. Moreover, the overall agreement obtained suggests that
the Aerosol-Into-Liquid Collector is a platform that could be modified to monitor many other
metal species in airborne PM.

(a)

88 | P a g e

(b)

Figure 5.5 (a-b): Metals and trace elements comparison between filter and slurry samples: a. total
metal and trace elements; b. water-soluble metal and trace elements. Error bars represent the
standard deviation of multiple samples (7 sets of samples). Note: inserted plot shows the
correlations on log-scale.

To validate the equivalence between the cupric ISE measurements and the other
elemental analysis techniques, a comparison of Cu measurement obtained by the cupric ISE
(total after acidification and water-soluble without acidification) and ICP-MS for total and water-
soluble Cu was performed (presented in Figure 5.6(a-b)). Generally very good agreement
between cupric ISE and ICP-MS measurements were observed (R
2
> 0.87 in both comparisons).
Given that ICP-MS measurements can detect some Cu-bound complexes as well, it is generally
reasonable that slopes are slightly less than 1. The very good agreement also indicates that there
is no evidence of interference from other species on the cupric ISE measurements. In summary,
these results indicate that this system is capable of providing Cu measurements that are virtually
equivalent to those obtained by ICP-MS for both total and water-soluble portions of ambient PM.
89 | P a g e

(a)

(b)
90 | P a g e

Figure 5.6 (a-b): Metals and trace elements comparison between cupric ISE and ICPMS results:
a. total metal and trace elements; b. water-soluble metal and trace elements. Error bars represent
the standard deviation of multiple measurements (3 measurements for each sample).

5.3.4.2. Field operation of the online Cu monitor
The online monitor was deployed at PIU and ambient PM
2.5
was continuously sampled
and Cu concentration was measured for 4-7 consecutive days (shown in Figure 5.7) to evaluate
the system’s ability to operate for prolonged periods with minimal supervision. The
accumulation rate of the slurry from the aerosol droplets inside the Aerosol-Into-Liquid Collector
varied from 4 to 6 mL per hour, depending on ambient temperature and relative humidity, as well
as the ambient particle number concentrations. As shown in Figure 5.7, the measured total Cu
concentration during the test period varied between 15.3 and 65.5 ng/m
3
, a range which is
generally consistent with PM
2.5
Cu concentrations reported in previous studies in the same
location (Daher et al., 2011; Hu et al., 2008; Wang et al., 2013a, 2013b). Overall the ISE-
determined Cu concentrations have similar diurnal trends as the PM
2.5
mass concentration data
monitored by AQMD at a nearby site. It is noted that Cu concentrations in PM
2.5
may be affected
by the local meteorological conditions. During the sampling period, wind speed and ambient
temperature (obtained from the same AQMD site) were highest in the midday (12pm-4pm) and
afternoon (4pm-8pm) periods. This phenomenon may to some degree explain the relative lower
Cu concentrations during these time periods comparing to morning (8am-12pm) period, as
presented in Figure 5.7. In summary, the field measurements data indicate that the online Cu
monitor is an effective technology for characterization of Cu in ambient aerosols, capable of
prolonged field operation and measurements (4 to 7 days) with minimal supervision.

91 | P a g e

Figure 5.7 Field continuous operation test results (Gap was due to temporary system
maintenance). Error bars represent the standard deviation of multiple measurements (3
measurements for each sample).

5.4. Summary and Conclusions
This study describes the development, laboratory and field evaluation of a new monitor
for online, in-situ measurement of total and water-soluble Cu in ambient PM
2.5
. This novel
technique employs the Aerosol-Into-Liquid Collector (Wang et al., 2013a) to provide
concentrated slurry samples of ambient PM
2.5
. Slurries collected in the sampling module are
subsequently measured for water-soluble/total Cu concentration by a cupric ISE. Laboratory
Cu(NO
3
)
2
aerosol collection tests demonstrated an excellent system measurement performance.
The field experiments demonstrated that proposed system provides Cu concentration
measurements that are equivalent to those obtained by ICP-MS for both total and water-soluble
components of ambient PM. In addition, the very good agreement obtained for Cu indicates that
interferences from other elements in the slurry are minimal. The online Cu monitor could
achieve near-continuous measurements (at 2-4 hour intervals) for at least 4 to 7 days without any
obvious limitations in its operation and with minimal supervision. Overall, this monitor is
versatile with potential applications in long-term, in-situ measurement of many other chemical
species in ambient PM. Although this study reports only Cu measurements, applications of this
92 | P a g e

system could be extended to other important metal species by deploying different types of ISEs,
and eventually provide more comprehensive measurement and evaluation of important redox
active metal species in ambient PM.

5.5. Acknowledgements
This study has been supported by South Coast Air Quality Management District (AQMD)
through award number #11527, and the National Institutes of Health (NIH) through Grants
R01AI065617-13, and R21AG040753-02 to the University of Southern California (USC). The
research described herein has not been subjected to the agency’s required peer and policy review
and therefore does not necessarily reflect the views of the agency, and no official endorsement
should be inferred. Mention of trade names or commercial products does not constitute an
endorsement or recommendation for use.

93 | P a g e

Chapter 6 A New Technique for Online Measurement of Total
and Water-soluble Copper (Cu) in Coarse Particulate
Matter (PM)

This chapter is based on the following publication:
Wang, D., Shafer, M. M., Schauer, J. J., & Sioutas, C. (2015). A new technique for online
measurement of total and water-soluble copper (Cu) in coarse particulate matter (PM).
Environmental Pollution, 199, 227-234.

6.1. Introduction
Numerous studies have linked particulate matter (PM) exposure to a wide range of
adverse health outcomes, including cardiovascular diseases (Delfino et al., 2005), pulmonary
injury (Li et al., 2003), as well as neurodegenerative disorders (Davis et al., 2013). It is
recognized that PM-induced adverse health effects are linked to the chemical composition of PM
rather that the total PM mass (Claiborn et al., 2002; Verma et al., 2009), though the influence of
specific species is to-date poorly characterized and likely size dependent. Improved PM
collection tools, coupled with speciated chemical and toxicological methods are needed to
advance our current understanding of PM drivers of human disease. Transition metals such as
copper and iron, together with other chemical elements are important contributors to the overall
toxicity of PM due to their ability to influence the redox status of cells (Gasser et al., 2009;
Nishida et al., 2002). The solubility of these specific metals may greatly influence their toxicity.
Although it is generally accepted the water-soluble fraction of metals may better represent the
bioavailable fraction of total metals in PM (Heal et al., 2005; Shi et al., 2003), determination of
both total and water-soluble trace metal fraction in PM samples is still important for
understanding their potential effects on human health and atmospheric environment (Karthikeyan
et al., 2006). Copper (Cu) is commonly released from various vehicular sources, especially in the
form of brake wear and tire abrasions (Hays et al., 2011; Hjortenkrans et al., 2007; Thorpe and
Harrison, 2008). Strong and consistent associations between Cu and the Reactive Oxygen
Species (ROS) activity of PM has been observed in PM collections from across the globe
(Charrier and Anastasio, 2012; Daher et al., 2012; Dellinger et al., 2001; Midander et al., 2009;
94 | P a g e

Wang et al., 2013b; Zhang et al., 2008). Novel tools for speciated and time-resolved copper
characterization of PM will improve our understanding of the likely significant role of Cu in
overall PM-induced toxicity.

Advanced PM collection and measurements technologies in high time resolution are
essential in evaluating the characteristics and the potential sources of ambient PM. Such
techniques have been developed to achieve near continuous characterization of bulk chemical
components including secondary ions (Orsini et al., 2003) and certain organics (DeCarlo et al.,
2006; Sullivan et al., 2004). More recently online tools for assessment of the relative toxicity of
PM, using oxidative potential as measured by the dithiothreitol (DTT) assay (Fang et al., 2014)
have been developed. However, novel technologies with the ability to measure in the field
concentrations of trace metals/elements with a high time resolution are still needed. Some
techniques have been developed for metal and element analysis, such as X-ray fluorescence
(Creatchman, 1999a; Park et al., 2014). Recently, an online Cu monitor was described to achieve
near-continuous measurements of total and water-soluble Cu in fine particles (PM
2.5
) (Wang et
al., 2014). This system utilizes a newly developed Aerosol-Into-Liquid Collector (Wang et al.,
2013a) as the PM collection module and a copper ion selective electrode (ISE) as the Cu
measurement module. In the Aerosol-Into-Liquid Collector, ambient fine PM are first grown to
3-4 µ m droplets through a saturation-condensation process, and are then sequentially collected in
a high flow rate (i.e. 200 L/min) impactor that results in highly concentrated PM slurries. The Cu
concentration in the slurry samples is subsequently determined by a copper Ion Selective
Electrode (ISE) at collection intervals of 2-3 hours. This new PM
2.5
Cu monitor was shown to be
highly adaptable and versatile for long-term, field measurements of an important redox active
metal species in ambient PM (Daher et al., 2012; Wang et al., 2013b).

While the technology for measuring PM
2.5
Cu has been described and evaluated as
discussed above, technologies which extend these trace element measurements to the coarse PM
range (defined as particles with a physical and/or aerodynamic diameter d
p
between 2.5 µ m to 10
µ m) are still needed and are currently unavailable. Recent studies suggest that the role of coarse
PM on asthma and other diseases has been significantly underestimated (Becker et al., 2005;
Becker and Soukup, 2003). Some other studies also demonstrated that Cu in coarse PM was
95 | P a g e

strongly associated with the redox activity of these particles (Cheung et al., 2012; Shafer et al.,
2010). This manuscript describes the development of a technique for online measurement of Cu
in ambient coarse PM. The approach described here employs two virtual impactors integrated
with a modified liquid impinger (BioSampler) (Willeke et al., 1998) as a coarse PM sampling
module, in which coarse PM

are directly collected as concentrated slurry samples. Copper
concentrations in the slurries collected in the modified BioSampler are subsequently measured
by a copper ISE. The new system also has the ability to be operated continuously, unattended,
for at least 5-6 days, making it a robust technology for characterization of the chemical
composition of ambient PM. We demonstrate that this system can provide measurements of Cu
concentrations in ambient coarse PM comparable to those obtained off-line by inductively
coupled plasma mass spectrometry (ICP-MS) analysis.

6.2. Methodology
6.2.1. Description of the system and its components
The online coarse PM Cu measurement system utilizes two virtual impactors (VIs)
developed in previous studies (Kim et al., 2001a; Wang et al., 2012) combined with a modified
liquid impinger (BioSampler, SKC West, Inc., Fullerton, CA) as the particle collection module
and a copper ISE (Model 9629BNWP, Thermo Fisher Scientific Inc., Waltham, MA) as the Cu
measurement module. A schematic of the complete system is presented in Figure 6.1 and a
schematic of the modified BioSampler is presented in Figure 6.2. Briefly, the sampled aerosol is
first drawn into two round nozzle VIs, which each have a major flow rate of 100 L/min and a
minor flow rate of 5 L/min (corresponding to a theoretical 50% cutpoint of 1.5 µ m in
aerodynamic diameter), to concentrate ambient coarse PM into the minor flow of VIs. It is noted
that the 50% cutpoint of both VIs is somewhat different than the traditionally defined coarse PM
cut size of 2.5 µ m. However, the slightly lower cutpoint of VIs was elected to ensure an ideal
concentrating/collection performance for the entire coarse PM size range. The two minor flows
of the VIs are then combined (10 L/min) to form the total intake flow of the modified
BioSampler. Just before sampling is initiated, 20 mL of ultrapure water (from a Millipore high
purity water system; A-10, EMD Millipore, Billerica, MA) is injected into the modified
BioSampler in order to capture the coarse PM as liquid suspension samples. In the current study,
the commercially available SKC BioSampler was modified by adding a glass tube on the liquid
96 | P a g e

reservoir, approximately 1 cm above the impaction point of the aerosol from the nozzles (shown
in Figure 2), to serve as a rinsing/replenished water inlet. Another glass tube was added to the
bottom of the reservoir as the sample outlet (Figure 6.2), and the bottom of the reservoir is
modified into a conical cavity for improved sample transfer efficiency. During prolonged
sampling (i.e. 2-4 hours), 2.5 mL of ultrapure water (based on the average water evaporation rate,
as discussed in the following sections) is added into modified BioSampler each hour by a
peristaltic pump (Mityflex mode 907, Anko product Inc., Bradenton, FL, using Tygon R-3603
PVC laboratory tubing) which is controlled by programmable time switches (Model FM1D20E,
Intermatic, Inc., Spring Grove, Illinois) to replenish the water evaporation and maintain the
volume of slurry at about 20 mL through the entire sampling period. After each sample
collection period, slurry samples are transferred by another peristaltic pump into a 100 mL
capacity vial (polypropylene, Thermo Scientific, Rockwood, TN) coupled with a micro liquid
level detector (ML 101, Cosense Inc., Hauppauge, New York) for automatic measurement of the
actual liquid sample volume (liquid height is proportional to volume) in order to convert aqueous
PM concentrations to airborne PM concentrations. The collected samples are then transferred to
the Cu measuring vial (polypropylene, Thermal Scientific, Rockwood, TN) by a computer
controlled syringe pump (Model C 3000, 3K step, TriContinent Inc., Grass Valley, CA, using
FEP laboratory tubing) with a 4-way distribution valve and a 12.5 mL PTFE syringe
automatically after each sample collection. Cu ionic strength adjuster (ISA) (0.5 M NaNO
3

solution) is mixed with the sample at a volume ratio of 2%, in order to provide a constant
background ionic strength for sample and standards. If measurement of total copper is needed,
the samples can be acidified before ISE measurement to a pH value of 2-3 using 10% v/v nitric
acid, in order to distinguish water-soluble Cu measurement (without acidification) to total Cu
measurement (with acidification). The Cu concentration is determined by the copper ISE
mentioned above, the signal of which is monitored by a millivolt meter (Orion Star A214,
Thermo Fisher Scientific, Waltham, MA). A stirrer probe (Orion stirrer probe, Thermo Fisher
Scientific Inc., Waltham, MA) is used in the measurement vial to rapidly bring the electrode to
equilibrium and to maintain stability during the Cu measurements. The Cu concentration is
measured by the copper ISE over a period of 5 minutes. After each measurement, the entire
system is rinsed with ultrapure water to remove any material carried over from the previous
sample. The copper ISE is maintained in ultrapure water between measurements.
97 | P a g e

Figure 6.1 Schematic of the online Cu measurement system for coarse PM

98 | P a g e

Figure 6.2 Schematic of the modified BioSampler

6.2.2. Copper ISE
The copper ISE used in this study was a combination electrode with the sensing and
reference half-cells built into one electrode body. The copper ISE incorporates a membrane
which selectively permits Cu ions to migrate through it and ion-exchange with active sites on the
sensing element. A small electrical potential is generated from this ion-exchange: a signal that is
measured by a sensitive voltmeter. The electric potential is compared to the constant potential
reference half-cell system to determine the ion concentration. The sensing element is sensitive to
only the cupric (Cu
2+
) oxidation state of copper (cuprous (Cu
1+
) will interfere, but Cu
1+
are a very
small fraction of Cu). It is also important to note that the electrode is responsive to only the truly
dissolved (or very labile/exchanging) species/form of Cu. Particulate, colloidal, and strongly
complexed forms of Cu are not measured. Thus, the ISE determined Cu concentration may in
principle be slightly lower than the operationally defined by filtration ―water-soluble‖ fraction.
The calibration of the copper ISE was periodically checked over the course of the study using
standards (from 10 ppb to 1000 ppb) prepared by serial dilution of a 1000 mg/L stock Cu
99 | P a g e

solution (made using analytical grade salts of cupper (Cu(NO
3
)
2
, Fisher Scientific). A typical
calibration curve of the copper ISE is presented in Figure S1, which is generally consistent
through the entire study. The performance and accuracy of copper ISE measurement can be
affected by several factors, including ionic strength, temperature and pH of samples. Details of
laboratory evaluations of the influence of these factors on copper ISE performance are described
in a previous study (Wang et al., 2014).

6.2.3. Laboratory evaluation of modified BioSampler’s collection efficiency
A series of experiments were performed to identify a sampling flow rate that optimally
balanced high particle collection efficiency with low pressure drop. A low pressure drop is
desired in order to keep the rate of water evaporation from the BioSampler reservoir to a
minimum. As supplied, the design sampling flow rate of the SKC BioSampler is 12.5 L/min,
resulting in a pressure drop of approximately 0.5 atm (Willeke et al., 1998). We explored particle
collection efficiency at the lower flow rates of 5, 9 and 10 L/min where pressure drops would be
significantly less. Different sizes of monodisperse fluorescent polystyrene latex (PSL) particles
(Polyscience Inc., Sacramento, CA) varying from 0.75, 1, 1.5, 3 and 6 µ m in diameter were
generated by a HOPE nebulizer (B&B Medical Technologies, Carlsbad, CA) to determine the
collection efficiency of the modified BioSampler. 20 mL of ultrapure water was injected into the
modified BioSampler before each test. A 37-mm Teflon filter (Teflo, Pall Corp., Life Sciences,
1-μm pore, Ann Arbor, MI) was connected before and after the modified BioSampler to
determine the concentrations of PSL arriving at and exiting the modified BioSampler,
respectively. At the end of each run, the Teflon filters from both upstream and downstream of
the modified BioSampler were extracted with 10 mL of ethyl acetate to dissolve the fluorescent
dye from the collected particles. The amount of fluorescent dye in the extraction solutions was
quantified with a Fluorescence Detector (FD-500, GTI, Concord, MA) to determine particle
concentrations. Moreover, the residuals on the modified BioSampler impaction surfaces were
carefully washed with ethyl acetate and extracted in order to estimate the impaction wall losses.

To provide a complete ―mass‖ balance of PSL and check on the ―in-minus-out‖ estimates
of recovery, the concentrations of PSL particles in the modified Biosampler slurries from the 3
and 6 µm PSL particle trials were determined. Aliquots of the slurries were first dried by
100 | P a g e

evaporating the collected water using laboratory HEPA-filtered, purified compressed air. The
remaining residuals were re-extracted with 10 mL of ethyl acetate and the fluorescence of the
extraction solutions determined as described above. The fluorescence measures were then
converted to PSL concentration to estimate the amount of incoming PM (PSL) that was actually
collected by the modified BioSampler.

6.2.4. Field evaluation of Cu measurement system:
Following completion of the laboratory evaluation, the system described in Figure 6.1
was deployed at the Particle Instrumentation Unit (PIU) of the University of Southern California
for field evaluation tests from July 9th to August 3rd 2014. This site is located in an urban area
near downtown Los Angeles, California, within 150 m of a major freeway (I-110), and thus
represents a typical urban mix of particles emitted by mostly traffic sources (Ning et al., 2007).
Ambient coarse PM samples were collected concurrently using a filter sampler and the coarse
PM Cu measurement system, in order to directly compare the performance of this novel system
to the traditional and well-documented filter-based sampling method in terms of coarse PM
elemental composition and mass. A third virtual impactor was added to the system and the 5
L/min minor flow of it was directly connected to a 37-mm Teflon filter (Teflo, Pall Corp., Life
Sciences, 1-μm pore, Ann Arbor, MI) held in a 37mm air sampling cassette (Zefon International
Inc., Ocala, FL). For each experiment (set of samples), the sampler was operated for 3 hours to
ensure adequate mass loading (primarily for the filter samples for subsequent comparative
analysis). The slurry samples collected by the modified BioSampler were divided equally into
two portions by the syringe pump: in one portion, Cu was measured directly online using the
copper ISE. Measurements were taken before and after acidification of the samples in order to
distinguish water-soluble and total Cu data. The other portion (together with the parallel filter
samples) were reserved and subsequently analyzed for total and water-soluble major and trace
elements by high-resolution magnetic sector inductively coupled plasma mass spectrometry (SF-
ICPMS). All samples were kept frozen at –20 ° C immediately after each collection and prior to
chemical analysis.

In the laboratory (Wisconsin State Laboratory of Hygiene, WSLH), the slurry samples
reserved for offline chemical analysis were divided into two sub-samples. One sub-sample was
101 | P a g e

ultra-filtered at 10 kilo-Dalton (kD) using pre-cleaned centrifugal ultra-filters (Amicon Ultra-15
Centrifugal Filter; regenerated cellulose membrane in an all polypropylene device). The fraction
of elements passing the 10 kD ultra-filter can be considered the water-soluble component of
these species. The other sub-sample of the slurry was left un-fractionated and acid-digested to
determine the total concentration of chemical elements in the PM slurry samples. Parallel
collected filter samples were sectioned into two equal half’s to provide sub-samples for both total
and water-soluble analyses that were comparable to the slurry processing. One section was
digested (microwave-aided acid digestion in Teflon bombs) for total elemental analysis. The
other filter section was extracted with water and filtered to provide water-soluble samples.
Measurements of 50 major and trace elements were conducted using high-resolution magnetic
sector inductively coupled plasma mass spectrometry (SF-ICPMS, Thermo-Finnigan Element 2).
Details of the ICP-MS analysis are presented in previous studies (Zhang et al., 2008).

To further demonstrate the ability of the coarse PM Cu online monitor to operate
continuously with minimal operator intervention, the sampler was run at the PIU sampling site
for 6 consecutive days in November 2014 (i.e. November 3rd to November 9th). Coarse PM
slurry samples were collected every 3 hours. The copper ISE was calibrated every day automated
by a one-point standard check (using 100 ppb Cu standard solution). The sensing surface was
polished and entire set of calibration was performed manually every two days. The ISE-
determined total Cu concentrations were also plotted with prevailing wind direction data reported
by a monitoring site of the South Coast Air Quality Management District (AQMD), located
approximately 3 km north of our (PIU) sampling site in downtown Los Angeles during the
sampling period.

6.3. Results and discussion
6.3.1. Laboratory evaluation of collection efficiency of modified BioSampler
The particle collection efficiency of the modified BioSampler was evaluated with 0.75, 1,
1.5, 3 and 6 µ m PSL reference aerosols. The total collection efficiency (E
c
) is calculated by the
mass concentrations collected downstream (C
up
) and that collected upstream (C
down
) of the
modified BioSampler (following the methodology described earlier in this manuscript), which is
presented in the following equation:
102 | P a g e

E
c
=1-C
down
/C
up
[1]

The wall loss (E
w
) is determined using the mass concentrations of the residuals on the
modified BioSampler impaction surfaces (C
w
) divided by C
up
:

E
w
=C
w
/C
up
[2]

Results of these tests are presented in Table 1. Generally, under lower flow rate
operations (i.e. 5 and 9 L/min), the total collection efficiency of PSL particles was above 60%,
however a significant portion was lost on the impaction wall and not captured into the slurry.
Since the BioSampler is working as a ―Swirling Aerosol Collector‖, particles are captured by
inertial (centrifugal) acceleration into the aqueous layer underneath the acceleration nozzles of
the sampler (Lin et al., 1997; Willeke et al., 1998). If the jet velocity of each nozzle is not high
enough to induce sufficient centrifugal acceleration to the exiting aerosol (i.e. the case at lower
flow rates), a significant portion of particles escape collection and/or adhere to the BioSampler
walls, which results in relatively high wall losses (i.e. 10-37%), at 5 and 9 L/min (Table 6.1).
The PSL particles are likely more hydrophobic than typical ambient PM, therefore the wall
losses measured in these tests may represent a high end of particle losses. For a flow rate of 10
L/min, the total collection efficiencies for all particles tested are generally above 80% (above 95%
for particles in the coarse PM size range), with relatively low wall losses of less than 5-10%. As
a result, 10 L/min was chosen as the optimal sampling flow rate during all following laboratory
and field evaluation tests. Figure 6.3 presents the comparison between the collection efficiency
tests performed in the current study using a 10 L/min flow rate and the previous study by
Willeke at al. (i.e. tests under 12.5 L/min flow) (Willeke et al., 1998), both using 20 mL of water
as the collection liquid. Our results are generally consistent with the data reported by Willeke at
al., reaffirming the very good collection performance of the modified BioSampler as discussed
above.

103 | P a g e

Table 6.1 Collection efficiency of PSL particles
Flow rate PSL particle size Total collection efficiency Wall loss
Fraction collected in
suspension
Water
evaporation
rate (mL/hr)
5 L/min
0.75 μm 66± 2% 37± 3% 29± 4%
0.7± 0.2
1 μm 75± 3% 35± 4% 40± 5%
1.5 μm 80± 3% 30± 2% 50± 4%
3 μm 95± 2% 15± 3% 80± 4%
6 μm 95± 2% 25± 2% 70± 3%
9 L/min
0.75 μm 76± 2% 21± 2% 55± 3%
2.0± 0.3
1 μm 84± 4% 18± 2% 68± 5%
1.5 μm 90± 3% 17± 3% 73± 4%
3 μm >99% 12± 2% 88± 2%
6 μm >99% 10± 2% 90± 2%
10 L/min
0.75 μm 83± 3% 9± 2% 74± 4%
2.6± 0.4
1 μm 91± 2% 7± 2% 84± 3%
1.5 μm 98± 2% <5% 93± 2%
3 μm >99% <5% >95%
6 μm >99% <5% >95%

104 | P a g e

Figure 6.3 Comparison of the BioSampler collection efficiency tests with a previous study by
Willeke et al (Willeke et al., 1998)

As shown in Table 6.1, the average water evaporation rates under different flow rates
were also evaluated. At the nominal BioSampler flow rate of 12.5 L/min the pressure drop across
the BioSampler is about 0.5 atm based on both theoretical calculation and experimental
measurements, which results in a water evaporation rate of roughly 5-6 mL/hour (Willeke et al.,
1998). In the current study, the reduced flow rate of 10 L/min decreases the rate of water
evaporation to about 2-2.5 mL/hour (Table 6.1). Therefore, in long term, continuous field
measurements, 2.5 mL of ultrapure water is added into the modified BioSampler each hour to
replenish the water evaporation.

105 | P a g e

Incorporated into the PSL tests with the 3 and 6 µ m particles were experiments to directly
evaluate recovery of PSL from the slurries collected by the modified BioSampler. These data
enabled us to better define the actual suspension collection efficiency of the sampler (i.e., the
percentage of particles collected within the 20 mL water suspension of the modified BioSampler)
and provide further evidence of sampler efficacy. Results of these tests are presented in Table 6.2.
Overall, very good agreement between the recoveries of the modified BioSampler and the
theoretical fraction collected in the slurry (i.e. total collection efficiency subtracted by wall
losses) was observed (the measured average recovery was 91% compared to a theoretical value
of 95%). Though some uncertainty remains due to analytical challenges of the evaporation-
extraction procedure discussed earlier, the recovery collection efficiencies of the modified
BioSampler show excellent agreement with the estimated fraction collected in suspension.

Table 6.2 Particle recovery of modified BioSampler

BioSampler
recovery
Theoretical Fraction collected in
suspension
PSL 3µ m 90± 2% >95%
PSL 6µ m 92± 3% >95%

6.3.2. Field evaluations
6.3.2.1. Chemical results for coarse PM

samples
A comparison of the total and water-soluble elements (22 and 18 species, respectively)
measured from the filters with that determined in the modified BioSampler slurry samples is
presented in Figure 6.4(a-b). A few species (i.e. Pd and etc.) were excluded from water-soluble
analysis since their signal-noise ratio did not rise above our threshold criterion. Generally,
excellent linear regressions were observed for both total and water-soluble metals and elements,
with slopes close to 1 and with high correlation coefficient ( R
2
) values (i.e. R
2
= 0.97 for total
metals and R
2
= 0.98 for water-soluble metals), showing very good comparability of this novel
monitor and traditional sampling methods (i.e. filter collection). These outcomes are consistent
with a previous study (Daher et al., 2011) and reaffirm the modified BioSampler as a robust
aerosol collector for field measurement of PM chemical composition.

106 | P a g e

(a)

(b)
107 | P a g e

Figure 6.4 (a-b): Coarse PM metals and trace elements comparison between filter and slurry
samples by SF-ICPMS: a. total metal and trace elements; b. water-soluble metal and trace
elements. Data in these figures represent the average of different samples. Error bars represent
the standard deviation of multiple samples. Copper is indicated by arrow in both figures. (Note:
dash line represents Y=X line)

A comparison of Cu measurement obtained by the copper ISE (both total after
acidification and water-soluble without acidification) and SF-ICPMS for total and water-soluble
Cu was performed (presented in Figure 6.5). In general, very good agreement between the
copper ISE and SF-ICPMS measurements were observed (R
2
= 0.89 (total); R
2
= 0.91 (water
soluble)). It is noted that in some samples the concentration of Cu were found in the non-linear
range of the copper ISE (i.e. <50 ppb). However, a more detailed calibration was prepared
separately at this lower range to achieve accurate measurements. These signals for actual
samples are much higher than the background signal/detection limit of this cupper ISE (i.e. down
to several ppb). Moreover, the close to 1 slope in the linear regression, shown in this figure,
indicate that very accurate measurements can be achieved by the copper ISE even in the lower
concentration ranges.

108 | P a g e

Figure 6.5 Coarse PM Cu measurements comparison between copper ISE and SF-ICPMS results.
Error bars represent the standard deviation of multiple measurements. (Note: Dash line
represents the Y=X line)

6.3.2.2. Field operation of the online coarse Cu monitor
The online coarse Cu monitor was deployed at the PIU and operated for 6 consecutive
days (total copper concentrations shown in Figure 6.6) to evaluate the system’s ability to operate
for prolonged periods. In Figure 6.6, the measured total Cu concentration data were plotted with
the prevailing wind direction data obtained from the nearby AQMD site, as mentioned above.
Overall, the measured total Cu concentration in coarse PM (CPM) during the test period varied
between 16.7 and 83.3 ng/m
3
, which is generally consistent with coarse PM Cu concentrations
reported in previous studies in the same location (Cheung et al., 2011; Wang et al., 2013b).
Meanwhile, a dominant contribution to Cu concentration from southwest wind during the
daytime is observed, indicating a potential source of Cu in coarse PM from the I-110 freeway
located about 120 m to the southwest of the sampling site. The diurnal trend of coarse PM Cu
concentration is also comparable to those reported in an earlier study conducted in a similar
109 | P a g e

period in 2010 (Cheung et al., 2011), as shown in Figure 6.7. Generally, the coarse PM Cu
concentration reaches higher levels in the overnight and morning periods, and decreases during
midday and afternoon periods. Cheung et al attributed the higher overnight concentrations of
coarse PM gravimetric mass as well as species associated with mineral and road dust (such as Cu)
to the lower atmospheric dilution, combined with particle re-suspension due to the turbulence
induced by nearby traffic from the I-110 freeway. Overall, the field operation data clearly
demonstrates the monitor’s ability to operate for multiple consecutive days. Although in this
study the sampler was only operated continuously for a relative short sampling period (i.e around
a week). These field data show the utility of the real-time measurement that cannot be obtained
by long-term sampling such as filter collection. The ability of this monitor to provide reliable Cu
concentrations in short time intervals will be instrumental in determining sources of these species
in future studies.

Figure 6.6 Field continuous total Cu measurements VS. dominate wind direction.

110 | P a g e

Figure 6.7 Comparison of the average diurnal concentrations of coarse PM Cu to those reported
previously in the same sampling location by Cheung el al. (2011) in winter 2010

6.4. Summary and Conclusions
This study describes the development and evaluation of a new monitor for online, in-situ
measurement of total and water-soluble Cu in ambient coarse PM. This technique could achieve
near-continuous measurements (at 2-4 hour intervals) for at least 6 days without any obvious
limitations in its operation and with minimal supervision. Overall, this monitor is very robust
with potential applications in long-term, in-situ measurement of many other chemical species in
ambient PM, and successfully extends the equivalent technology in previous study measuring
PM
2.5
to coarse PM range. Although this study reports only Cu measurements, applications of
this system could be extended to other important metal species by deploying different types of
ISEs, and eventually provide more comprehensive measurement and evaluation of important
redox active metal species in ambient PM.

111 | P a g e

6.5. Acknowledgements
This study was supported by South Coast Air Quality Management District (AQMD)
through award number #25138-1397-01, and the US-National Institute of Allergy and Infectious
Diseases through grants #5R01AI065617-14 to the University of Southern California (USC).The
research described herein has not been subjected to the agency’s required peer and policy review
and therefore does not necessarily reflect the views of the agency, and no official endorsement
should be inferred. Mention of trade names or commercial products does not constitute an
endorsement or recommendation for use.

112 | P a g e

Chapter 7 Development and Evaluation of a Novel Monitor for
Online Measurement of Iron, Manganese, and
Chromium in Ambient Particulate Matter (PM)

This chapter is based on the following publication:
Wang, D., Sowlat, M. H., Shafer, M. M., Schauer, J. J., & Sioutas, C. (2016). Development and
evaluation of a novel monitor for online measurement of Iron, Manganese, and Chromium in
ambient particulate matter (PM), Science of the Total Environment 565, 123-131.

7.1. Introduction
Numerous studies have attempted to investigate the sources, transport, and chemical
speciation of redox-active trace metals, including iron (Fe), manganese (Mn), and chromium (Cr)
in ambient particulate matter (PM) (Chester and Stoner, 1974; Fang et al., 2015; Harrison et al.,
2012). These metals may significantly contribute to the adverse human health impacts of
aerosols via oxidative stress mechanisms (e.g. formation of Reactive Oxygen Species (ROS)),
which can initiate activate inflammatory cascades in cells and tissues (Kelly, 2003; Martinet et
al., 2004; See et al., 2007). Importantly, the oxidative stress toxicity of PM-associated metal
species critically depends on the metals’ oxidation state (Valko et al., 2005). Association
between these metals and ROS activity in different particle size ranges of ambient PM are
consistently observed in various locations globally (Daher et al., 2012; Ntziachristos et al., 2007;
Saffari et al., 2014; Shafer et al., 2010; Valavanidis et al., 2005; Wang et al., 2013b). Previous
studies have shown that major sources of these transition metals in ambient PM are mechanical
wear processes in automobiles, such as from brake linings, tire wear, and abrasive emissions
from tire-pavement interactions (Amato et al., 2011; Harrison et al., 2012; Schauer et al., 2006).
These metals are also constituents of crustal material and thus may be sourced from soil dust
(Chester and Stoner, 1974; Chow et al., 1994; Hueglin et al., 2005). Air pollution sources as well
as atmospheric chemical processes produce PM on fine time scales (i.e., few hours) and one must
capture their temporal profiles to adequately assess and understand the dynamics of atmospheric
PM emission, dispersion, and fate. Assessing atmospheric processing in the content of redox
cycling of transition metals remains a heated research topic, which will provide significant
113 | P a g e

insights on chemical reactions of aerosol during atmospheric transport, as well as controls on PM
induced toxicity. Therefore, technologies for analyzing trace metals, and more importantly their
chemical forms (i.e. total/water-soluble forms and/or different oxidation states), in ambient PM,
in high time resolution, will greatly assist in characterizing the chemical composition of PM, as
well as enabling a better understanding of the potential sources of toxicity of these metals in
ambient PM.

Past efforts to develop field-deployable, higher time resolution tools for characterization
of the elemental composition of ambient PM have focused on the technologies of aerosol time-
of-flight mass spectrometry (ATOFMS) and X-ray fluorescence (XRF). ATOFMS was one of
the earliest developed tools for determining the size and chemical composition of individual
atmospheric particles; however, this technology lacks the sensitivity to detect several important
metals in ambient air and is semi-quantitative at best for measurement of total metals only (Gard
et al., 1997). And, like most mass-spectrometry-based tools, it is insensitive to metal oxidation
state and chemical species. Another commercially available technique for semi-continuous
measurement of metals/elements in ambient PM is based on X-ray fluorescence (XRF)
(Creatchman, 1999; S. Park et al., 2014). Though capable of detecting a number of major and
minor elements in filter-collected PM, the XRF technique provides information on total element
levels only – it is insensitive to oxidation state and clearly is unable to distinguish between
potentially soluble species and insoluble species. Additionally calibration/quantification in
complex, variable, aerosol matrices can be challenging. There is great need, however, to measure
the speciation, in particular oxidation state, of important redox active metals (i.e. Fe, Mn and Cr),
as the biological pathways underlying the adverse health effects of ambient PM are speciation-
driven. Both the XRF and ATOFMS techniques cannot provide this speciation information that
is critical in providing insights on the potentially bio-available state of these redox active metals.

In contrast, spectrophotometry has been applied to quantify metals with a very high
degree of accuracy in aqueous solutions by means of suitable metal-specific chromophoric
reagents with high specific absorptivity and water solubility (Eller, 1994; Morgan and Stumm,
1965; Stookey, 1970). This approach has been used for quantifying water-soluble metals in
ambient aerosol extracts by deploying a long light path liquid waveguide capillary cell (LWCC)
114 | P a g e

to provide the necessary sensitivity (Ashley et al., 2003; Majestic et al., 2007, 2006). Recently
the spectrophotometry approach has been incorporated into a few online systems in which
ambient PM are directly collected as slurry samples, and metal concentrations in the slurry are
sequentially measured. For example, Rastogi et al. (2009) developed a prototype online system
for detecting water-soluble Fe (II) in ambient PM
2.5
(i.e. particles with aerodynamic diameter
less than 2.5 µ m), while Khlystov and Ma (2006) reported measurements of Cr (VI) and Cr (III)
in ambient PM using a new online system.

In this study, we developed an online system for measuring total concentrations of Fe,
Mn, and Cr in ambient PM
2.5
. In this system, ambient PM
2.5
are collected as slurry samples using
a high flow rate Aerosol-Into-Liquid Collector (Wang et al., 2013a), and metal concentrations
are subsequently determined by spectrophotometry via a Micro Volume Flow Cell (MVFC).
Before entering the MVFC, slurries are mixed with acid (i.e., HCl for Fe/Mn, and HNO
3
for Cr)
to dissolve water insoluble species of the metals and make it possible to measure their total
concentration. Oxidation/reduction reagents are then added to ensure that the metals are in the
specific oxidation state required for the metal-specific analytical reagents (i.e., Fe-ferrozine, Mn-
formaldoxime (FAD), and Cr-diphenycarbazide (DPC)) for absorptivity measurements.
Although our goal in this study was to measure total concentrations of the target metals, this
system can be readily modified for measuring water-soluble forms of the target metals by
passing the collected aerosol suspension through an in-line liquid filter prior to conducting the
MVFC measurements. Moreover, determination of native metal oxidation (i.e., Fe(II), Mn(II),
and Cr(VI)) state speciation can be achieved by taking measurements without introducing the
oxidation/reduction reagents. The new metal monitor was deployed in the field for continuous
operation, and system performance and measurement accuracy were evaluated by comparing the
online data with parallel off-line, time-integrated filter measurements.

7.2. Methodology
7.2.1. Reagents and Standards
All the acids used were trace metal grade (VWR) and the water used was produced by a
Millipore A-10 water purification system (EMD Millipore, Billerica, MA). All the lab wares (i.e.,
beakers, vials, and volumetric flasks) used in preparing the regents and standard solutions were
115 | P a g e

made of polypropylene and thoroughly cleaned (with 4 M hydrochloric acid (HCl) followed by
rinsing with water) before use. For all three metals, standard solutions were prepared spanning
concentration ranges corresponding to the typical levels measured in slurries from ambient PM
collections (Wang et al., 2014b, 2013b). For Fe measurements, the standard solutions (range 0.5-
100 ppb) were prepared by serial dilution of 1000 ppm Fe (II) stock solution (acidified with 4 M
HCl) prepared gravimetrically from (NH
4
)
2
Fe(SO
4
)
2
(ACS) salt. The Fe analytical reagent
(Ferrozine) was prepared by adding 133 mg of ferrozine (Sigma) to 50 mL of water containing
65 µ L of 4M HCl (Rastogi et al., 2009). The hydroxylamine hydrochloride (HA) solid (Sigma),
with a purity of 99.9999%, was prepared by dissolving 19.3 mg HA into 50 mL of water
(Majestic et al., 2006). The pH adjustment was also done by adding appropriate amounts of 3 M
sodium hydroxide (NaOH) solution, prepared from reagent grade NaOH solid (AMRESCO). For
Mn measurements, the working standards (0.5-10 ppb) were prepared by diluting 100 ppm Mn
(II) stock solution prepared using MnCl
2
(ACS) salt (acidified with 4 M HCl). The manganese
analytical reagent (formaldoxime; FAD) solution was prepared by dissolving 20 g of HA in 450
ml of water; 10 ml of 37% formaldehyde solution was then added and made up to 500 ml with
water. Appropriate amounts of 3 M NaOH solution was added to reach the desired pH for the
working standards (Majestic et al., 2007). For Cr measurements, the standards (0.5-10 ppb) were
prepared by serial dilution of 100 ppm of Cr (VI) stock solution (acidified with 4 M HNO
3
)
prepared by using K
2
CrO
7
(ACS) salt. The chromium analytical reagent (diphenylcarbazide,
DPC) was prepared by dissolving 167 mg of the reagent in 100 ml of acetone, which was then
mixed with 1.67% H
2
SO
4
solution at 1:1 volume ratio (Khlystov and Ma, 2006). To prepare 0.1%
H
2
O
2
solution, 0.143 ml of H
2
O
2
was diluted to 100 ml by adding 0.1 M NaOH solution.

7.2.2. Aerosol-Into-Liquid Collector
The Aerosol-Into-Liquid Collector was used as the PM sampler. This sampler operates at
a sampling flow rate of 200 L/min and collects ambient PM
2.5
directly as concentrated slurry
samples for spectrophotometry analysis. Details of the Aerosol-Into-Liquid Collector design and
operation are reported elsewhere (Wang et al., 2013a). Briefly, sampled air is first drawn into a
saturator tank to be mixed with saturated water vapor at approximately 30° C. The particle-vapor
mixture then passes through a condensational growth section where it is cooled down to about
20 ° C; the resulting super-saturation condenses ultrapure water vapor onto the incoming particles,
116 | P a g e

which grow to 3-4 µ m droplets. The grown droplets are then separated from the air stream by
inertial impaction and accumulated as concentrated slurry samples in the bottom reservoir of the
impactor. The liquid volume of the collected slurry is determined by a micro liquid level detector
(ML 101, Cosense Inc., Hauppauge, New York) by measuring the distance between the sensor
and the liquid level. After each collection period, the slurry is transferred through the liquid
outlet at the bottom of the impactor to the measurement module of the monitor for sequential
metal analysis. Ultrapure water was transferred into the saturator tank periodically by a timer-
controlled peristaltic pump (Mityflex mode 907, Anko product Inc., Bradenton, FL, using Tygon
tubing) to make up for the water consumptions during the continuous operations.

7.2.3. Micro Volume Flow Cell (MVFC)
A 10-cm path-length, optical flow cell MVFC (FIA-ZSMA-ML-100-TEF, Ocean Optics,
Inc., Dunedin, FL), with an internal volume of 60 µ L, was used for the spectrophotometry
detection. The long optical path of the MVFC provides high measurement accuracy within the
working concentration ranges, and has the advantage of reducing clogging due to bubble
formation and back-pressure issues that have been observed with the LWCC (Majestic et al.,
2007, 2006). In the developed system, a DT-MINI-2-GS deuterium and tungsten halogen light
source (Ocean Optics, Inc., Dunedin, FL) was used to provide light (200-900 nm) through an
optical fiber cable (QP450-1-XSR, Ocean Optics, Inc., Dunedin, FL) into the MVFC. The light
passed through the MVFC was then directed by another optical fiber into the spectrometer
(USB4000-UV-VIS, Ocean Optics, Inc., Dunedin, FL). The concentration of metals was
determined at the wavelength of maximum absorptivity for each analytical reagent (i.e., Fe-
ferrozine at 562nm, Mn-FAD at 450nm, and Cr-DPC at 540nm). In all the measurements, 700
nm was considered as a non-absorbing/background wavelength and used for baseline correction
of the absorption measurements. It should be noted that the field measurements of these metals
were performed separately, i.e., the reagents and chemicals used for the measurement of different
metals were not all added to one sample.

7.2.4. System configuration and operation
The configuration of the developed online metal monitor system is presented in Figure
7.1. As shown in the figure, this system utilizes the Aerosol-Into-Liquid Collector as the PM
117 | P a g e

collection module, and the MVFC spectrophotometry components as the metal analysis module.
It should be noted that in the current study, the three target metals were not measured
simultaneously due to limitations in the availability of instrument modules; therefore, the three
metals were measured by switching between the reagents manually. However, simultaneous
measurements can be achieved by splitting the flow of the collected slurry samples and
deploying multiple MVFC spectrophotometry units in parallel, thus enabling separate and
simultaneous detection of each metal.

Ambient PM
2.5
was collected into the Aerosol-Into-Liquid Collector every 2 hours, and
the collected samples were then automatically transferred to a vial (polypropylene, Thermal
Scientific, Rockwood, TN) for acid digestion and pH adjustment by a computer-controlled
syringe pump (Model C 3000, 3K step, TriContinent Inc., Grass Valley, CA, using FEP
laboratory tubing) with a 4-way distribution valve and a 12.5 mL PTFE syringe. During the acid
digestion procedure, concentrated acid (i.e., 4 M HCl for Fe and Mn measurements, and 4M
HNO
3
plus 1% H
2
O
2
for Cr measurements) was mixed with the slurry sample at a volume ratio
of 1:20 (acid:sample), resulting in a pH value of approximately 0.5, to achieve total metal
measurements. For Fe and Mn measurements, after acid digestion with a residence time of 10
min, the pH of the slurry was adjusted to the desired level for the following reagent reactions (i.e.,
pH of 5-7 for Fe-ferrozine reaction and 7‒8.5 for Mn-FAD reaction) by adding appropriate
amounts of 3 M NaOH solution. The Cr-DPC reaction is optimal near pH=1, so neutralization is
not required for the Cr measurements. In the acid digestion/neutralization unit, a pH/temperature
probe (Orion 9107BNMD, Thermo Fisher Scientific Inc., Waltham, MA) connected to a pH
meter (Orion Star A214, Thermo Fisher Scientific, Waltham, MA) was deployed to monitor the
pH of the slurry. After the digestion/neutralization procedure, the slurry, the metal reagent, and
the oxidation/reduction reagent were transferred by peristaltic pumps (Mityflex mode 907, Anko
product Inc., Bradenton, FL, using Tygon tubing) and mixed through a double T-junction (Table
7.1). Since the metal-analytical reagents are oxidation state specific, the addition of an
oxidation/reduction reagent is essential to convert the possible different oxidation states of the
target metal to a uniform state (i.e., Fe(II), Mn(II) and, Cr(VI)) for the following reactions. For
Fe measurements, Fe(III) is converted to Fe(II) by adding the reducing reagent HA solution at a
volume ratio of 1%. For Mn measurements, the HA and ethylenediaminetetraacetic acid (EDTA)
118 | P a g e

solutions were both added at a volume ratio of 1% to convert higher oxidation states (III, IV, VII)
states to Mn (II) and to eliminate interference from Fe, respectively. For Cr measurements, no
oxidation reagent is needed in this procedure, because Cr(III) is oxidized to Cr(VI) by adding
H
2
O
2
. A summary of the reagent additions for each metal is presented in Table 7.1. After adding
the appropriate reagents, the mixture was drawn into a serpentine reactor (1 m length, 0.75 mm
ID PFA tubing, Global FIA, Inc., Fox Island, WA) followed by a self-made, 100-turn reaction
coil (0.75 mm ID PEEK tubing) to achieve complete mixing. The flow was stopped for 10 min to
ensure complete formation of the metal complex. The sample was then pumped through the
MVFC, stopped again before the trailing air slug entered, and held there for 2 min to determine
sample absorbance. During continuous field operations, the Aerosol-Into-Liquid Collector, the
MVFC, and all other sample pathways were rinsed with ultrapure water right after each sample
collection and measurement, and cleaned with 2 M HCl every two days. A calibration standard
was run every 2 days to confirm measurement accuracy during the continuous measurements.

Figure 7.1 Schematic of the developed on-line metal monitoring system.

119 | P a g e

Table 7.1 Summary of the reagents used in different procedure for metal detections. (Percentage
values in parentheses represent volume ratio)
Metals Acid digestion Oxidation/reduction Metal reagent
Fe
4M HCl (5%)
HA (1%) Ferrozine (1%)
3M NaOH (adjust pH to 5-7)
Mn
4M HCl (5%) HA (1%)
FAD (1%)
3M NaOH (adjust pH to 7‒8.5) EDTA (1%)
Cr
4M HNO
3
(5%)
N/A DPC (1%)
1% H
2
O
2
(1%)

It should be noted that although in the current study the system was designed to measure
total metal concentrations by including acid digestion and oxidation/reduction units, the
instrument’s modular design can be readily re-configured to measure water-soluble fractions of
target metals by replacing the acid digestion step with an in-line liquid filtration unit (e.g, a 0.20
µ m polypropylene syringe filter) to remove the water-insoluble fractions (Rastogi et al., 2009).
Moreover, oxidation state speciation of the target metals in the water-soluble fraction can also be
readily achieved by taking measurements with/without the addition of the oxidation/reduction
reagent. For example, in order to quantify both Fe(II) and Fe(III) concentrations in water-soluble
fraction, the absorption measurements can first be taken with the addition of HA solution to
provide a total water-soluble Fe measurement (in which all Fe(III) species are reduced to Fe(II)
and detected as Fe(III)+Fe(II)) and then without HA (in which only Fe(II) species are detected).
Subsequently, the Fe (III) concentration can be calculated by subtracting the measured
concentration without addition of HA from measurements with HA added to the sample. Similar
procedure/calculations can be applied to Mn and Cr measurements as well for oxidation state
speciation determinations.

7.2.5. Laboratory evaluation tests
The performance of the on-line metal monitor was evaluated by collecting and measuring
samples of lab-generated (NH
4
)
2
Fe(SO
4
)
2
and MnCl
2
aerosols with known mass concentrations
(Cr tests were not conducted due to lab safety concerns). Metal salt aerosol particles were
generated by atomizing solutions of known concentrations of the salts using a HOPE nebulizer
(B&B Medical Technologies, Carlsbad, CA) and directed into the system as feeding aerosols.
120 | P a g e

Size distributions of the generated aerosols were determined using a Scanning Mobility Particle
Sizer (SMPS 3936, TSI Inc., Shoreview, MN) coupled with a Condensation Particle Counter
(CPC 3022A, TSI Inc., Shoreview, MN). The mass concentration of the generated aerosols was
calculated by multiplying the densities of the salts (i.e., 1.86 g/cm
3
and 2.01 g/cm
3
for Fe and Mn
salts, respectively) by their air volume concentrations (obtained from size distributions and
assuming that particles are spherical), as shown in Equation [1] below. The measured metal
concentration in the slurry collected from the generated aerosols was then compared with the
calculated reference concentration, which was determined based on the mass concentration of the
generated aerosol, the sampled air volume, and the volume of the slurry samples, as presented in
Equation [2] below:

Mass conc. (generated) = Density of salt (generated) × Volume of particles (generated) [1]
Metal conc. (calculated) = Mass conc. (generated) × Volume of Air / Volume of slurry [2]

7.2.6. Site description and field evaluation tests
The metal monitor system was subsequently deployed at the Particle Instrumentation
Unit (PIU) of the University of Southern California (USC) for field evaluation tests in September
2015 for a period of one month. The PIU site is located about 150 m downwind of the I-110
freeway, in an urban area near downtown Los Angeles, California. Ambient PM
2.5
was collected
semi-continuously (i.e., with a 2-hr time resolution) and the target metal concentrations in the
collected slurry samples were then measured online, resulting in 12 data points per day. The
reagents were manually switched every week to enable the measurement of each of the three
metals during the one month of field evaluation tests. To provide an independent measure of the
metals concentrations in the concentrated slurries, aliquots of the collected slurry samples were
also preserved and kept frozen at –20 ° C after the sampling and analyzed for major and trace
elements using magnetic sector inductively coupled plasma mass spectrometry (SF-ICPMS,
Thermo-Finnigan Element 2) (Zhang et al., 2008). In parallel with the continuous operation of
the metal monitor, time-integrated filter samples were collected via the Versatile Aerosol
Concentration Enrichment System (VACES) (Kim et al., 2001c, 2001f) in order to compare
metal concentrations between off-line filter and online measurements. Briefly, a third
condensation tube coupled with a virtual impactor (VI), having a flow rate of 100 L/min, was
121 | P a g e

added to the saturator tank to concentrate the grown droplets into the 5-L/min minor flow of the
VI. The concentrated droplets then passed through a diffusion dryer (Model 3062, TSI Inc.,
Shoreview, MN) to remove the excess water, and subsequently collected onto 37-mm Teflon
filters (Teflo, Pall Corp., Life Sciences, 1-μm pore, Ann Arbor, MI). Details of the filter
collection procedure using VACES system has been reported in previous studies (Wang et al.,
2014). The time resolution for the field on-line measurements was 2 hr. The time-integrated
filter samples were collected over a 4 hr time period, to collect sufficient PM mass for the SF-
ICPMS analysis. The online data were averaged over the same time periods with the off-line
filter samples to allow direct comparison. Additionally, the system was modified and operated
for on-line measurements of the water-soluble fraction and specific oxidation states of the target
metals.

7.3. Results and Discussion
7.3.1. Calibration
The spectrophotometry module was calibrated for each metal by running standard
solutions within working ranges similar to the anticipated concentration levels of the slurry
samples collected from ―real-world‖ ambient PM. The results of the calibration tests for the three
target metals are summarized in Table 7.2. The calibration results showed a robust linear relation
between absorbance versus metal concentration, with slopes of the linear regression being
0.0078± 0.0001, 0.0151± 0.0003, and 0.0209± 0.0004 (1/ppb) for Fe, Mn, and Cr, respectively. All
three of the calibration curves have R
2
values of higher than 0.99, and the calibration curves were
well reproduced during multiple calibrations conducted on different days. During the calibration
tests, the limit of detection (LOD) for each metal was also determined. Following the same
method reported in previous studies (Khlystov and Ma, 2006; Majestic et al., 2007; Rastogi et al.,
2009), the LOD of each metal was estimated as 3 times the standard deviation of the field
method blank samples, determined by injecting ultrapure water together with metal reagents into
the MVFC and measuring the corresponding absorbance. The field method blanks were also
tested in real field operations after system rinsing to account for potential carryovers throughout
the system between subsequent samples. Our measurements indicated that the LODs for the
developed systems are approximately 0.3, 0.2, and 0.2 ppb (corresponding to average ambient
concentration of 0.13, 0.08, and 0.08 ng/m
3
) for Fe, Mn, and Cr measurements, respectively.
122 | P a g e

Given that the metal concentration levels in the actual samples are much higher than the obtained
LODs, the measurement signal-to-noise (S/N) ratio of the developed system is very high for
ambient PM measurements. Considering the metal concentration levels in slurry samples
collected under ―typical‖ ambient PM concentrations as reported in previous studies (Daher et al.,
2011; Wang et al., 2013b), the S/N ratio of the system is about 400, 60, and 28 for Fe, Mn and Cr
measurements, respectively, which are generally comparable to the S/N ratios in other
technologies such as XRF under the similar sampling time resolution (U.S. Environmental
Protection Agency, 2012).

Table 7.2 Calibration results for different metals
Metal Ranges (ppb) Slope (1/ppb) R
2
LOD (ppb)
Fe 0.5-100 0.0078± 0.0001 0.9991 0.3
Mn 0.5-10 0.0151± 0.0003 0.9986 0.2
Cr 0.5-10 0.0209± 0.0004 0.9978 0.2

7.3.2. Laboratory aerosol generation tests
Figure 7.2(a-b) presents the comparisons between calculated concentrations in the
slurries versus the actual measurements by spectrophotometry, overall indicating very good
agreement between generated and measured concentrations (slopes and R
2
values are quite close
to 1). It should be noted that the concentrations generated in these tests were much higher than
those of actual ambient slurry samples, since we generated ―pure metal salt particles‖ as the input
aerosols. Moreover, for the highest concentration levels, the collected slurries were diluted with
ultrapure water to the working ranges mentioned above in order to prevent saturation in the
absorbance measurements. However, the wide concentration ranges covered in these tests,
together with the consistent agreement observed, further indicates that the on-line metal monitor
could achieve a robust collection efficiency as well as measurement accuracy based on the
controlled laboratory experiments.

123 | P a g e

(a)

(b)
Figure 7.2 (a-b) Lab aerosol generation and collection results: a) Fe tests; b) Mn tests. Error bars
correspond to one standard deviation of multiple tests.

124 | P a g e

7.3.3. Comparison between online measurements and offline sample results
To provide further validation of the accuracy of the online measurement system, data
from the online measurements were compared with that from time-integrated samples (i.e., both
parallel filter collections and slurry samples analyzed by SF-ICPMS) for all three of the target
metals, the results of which are presented in Figure 7.3(a-c). Overall, the online measurements
are very comparable to those of parallel filters and slurries for all three of the metals; linear
regressions between online and offline data indicated a very good overall agreement, with slopes
ranging between 0.85 and 1.09 and R
2
values between 0.76 and 0.94. These results attest to the
accuracy and reliability of the performance of the developed on-line metal monitoring system in
determining the target metal concentrations in ambient PM. Moreover, the agreements between
our online measurements and referenced SF-ICPMS results further confirm that the chemical
treatment procedures, including acid digestions and addition of oxidation/reduction reagents,
were adequate to ensure that all significant metal-containing PM components (including water-
insoluble species) could be brought into solution and made available for spectrophotometry
detection, thus achieving a robust total metal measurement. It should, however, be noted that
these data pertain to fine PM, and the efficacy of the acid treatment for larger size PM fractions,
which contain more crustal material and thus more acid-resistant phases, has not yet been
demonstrated. As mentioned in previous sections, one of the critical advantages of the developed
system is that it can be easily modified by replacing the acid digestion procedure with an in-line
filtration module to remove the water-insoluble fraction of PM, thereby making it possible to
achieve on-line measurement of the water-soluble fraction of these metals. Alternatively, both
total and water-soluble measurements can be made by splitting the slurry with one stream
diverted into the digestion module and the other to a filtration module. Additionally, the system
can be further modified by bypassing the oxidation/reduction step, so that oxidation state-
specific measurements can be performed. This flexibility provides a powerful tool to directly
address critical speciation questions in aerosol-driven human toxicology.
125 | P a g e

(a)

(b)
126 | P a g e

(c)
Figure 7.3 (a-c) Comparison between online metal measurements and off-line parallel samples
analyzed by SF-ICPMS analysis for a) Fe; b) Mn; and c) Cr. Error bars represent one standard
deviation of multiple online measurements during parallel sample collection period. Data in
black color represent slurry samples analyzed by SF-ICPMS, whereas data in grey represent
filter samples. Dashed line represents Line Y=X.

7.3.4. Field deployment of the developed metal monitor
The online metal monitor was deployed at an urban site near Downtown Los Angeles (i.e.,
the PIU) for a month-long period in Sep. 2015. Ambient PM
2.5
was continuously sampled and
metal concentrations were measured with a 2-hr time resolution for 5-7 consecutive days for
each species (shown in Figure 7.4(a-c)). In each of these roughly week-long experiments, the
system operated automatically with only very limited involvement (i.e., downloading the data
and running calibration standards) by the field staff. Overall, the collection and measurement
performance of the novel trace-metal monitoring system are robust, without any obvious
shortcomings during the continuous operation. The measured concentrations of all three of the
metals show relatively high diurnal variability during the test period, which ranges between 4.8-
127 | P a g e

65.6, below detection limit to 10.0, and below detection limit to 6.6 ng/m
3
for Fe, Mn, and Cr,
respectively. These ranges are consistent with the time-integrated PM
2.5
concentrations of the
same metal species reported in previous studies at the same location (Daher et al., 2011; Wang et
al., 2013b). Concurrent PM
2.5
mass concentration data measured at a nearby monitoring site
located approximately 3 km to the north of the PIU site operated by the South Coast Air Quality
Management District (AQMD) are also included in Figure 7.4(a-c). Overall, Fe and Mn
concentrations follow similar diurnal trends as that observed in PM
2.5
concentrations, showing a
peak in the morning period (i.e., 8 am to 12 pm), on average 60% and 28% higher than the daily
average for Fe and Mn, respectively. Moderate levels (close to daily average) of these two metals
are observed during the afternoon period (i.e., 12 pm to 4 pm), while their concentrations
decrease to 75-80% of the daily average value during nighttime (i.e., 4 pm to 8 am). Since Fe and
Mn commonly originate from traffic-related sources, such as vehicular abrasion (Harrison et al.,
2012; Schroeder et al., 1987), the observed peaks in the morning are likely attributed to a
combination of intense traffic and lower atmospheric mixing heights, while the lower
concentrations during night time might be a result of the lower traffic volume during that period.
Unlike Fe and Mn, Cr measurements display a slightly different diurnal trend, with the highest
levels observed in the morning period (60% higher than the daily average), and the lowest levels
observed in the afternoon period (roughly 25% lower than the daily average). It should be noted
that these field measurements, each lasting for about a week, were conducted to test the ability of
this system to provide near-continuous measurements for each metal for prolonged field
operation and measurements while requiring minimal supervision. Obviously longer sampling
campaigns during different seasons would be required to draw more robust conclusions on the
sources and diurnal variability of these metals. Pilot feasibility demonstration measurements of
water-soluble fractions as well as oxidation states of the target metals (presented in Figure 7.5 (a-
c) and Figure 7.6) showed that the average ratios of water-soluble fractions to total
concentrations for Fe, Mn and Cr were 68± 15%, 56± 33%, and 59± 20%, respectively. Moreover,
water-soluble Fe(II) contributed to about 64± 23% of water-soluble Fe based on limited field
observations. We acknowledge the fact that these tests were carried out for a relative short period
(i.e. 36-48 hours each) and should therefore be regarded mainly as feasibility demonstration tests
corroborating the ability of the monitor to measure both total and water-soluble fractions of the
target metal species, as well as their specific oxidation states. Our results demonstrate the ability
128 | P a g e

of the developed metal monitor to provide reliable near-continuous measurements of three
important metal species in ambient PM
2.5
, as well as detecting their water-soluble fractions and
specific oxidation states, which are essential elements in studying the redox cycling capability of
these metals.

(a)

(b)

(c)
Figure 7.4 (a-c): Continuous online metal measurements during field deployment: a) Fe
measurements; b) Cr measurements; c) Mn measurements. Few gaps in Cr and Mn
measurements are due to data points below detection limit.

129 | P a g e

(a)

(b)

(c)
130 | P a g e

Figure 7.5 (a-c): Near-continuous (2 hour) measurements of total and water-soluble
concentrations of target metals: (a) Fe measurements; (b) Mn measurements; and (c) Cr
measurements.

Figure 7.6 Field measurements of oxidation states of Fe

7.4. Summary and conclusions
In this study, a novel monitor was developed and evaluated for online measurements of
three important metals (i.e. Fe, Mn and Cr) in ambient PM
2.5
. This monitor utilizes the Aerosol-
Into-Liquid Collector as the sampling module, and target metal concentrations are sequentially
determined by means of spectrophotometry. Results from laboratory and field evaluation tests
demonstrated that overall, this monitor has very high collection efficiency and measurement
accuracy. Field tests indicated that the monitor is capable of continuous operation for at least 5
consecutive days with minimal supervision. More importantly, the modular design of the system
makes it possible to detect both total and water-soluble fractions as well as specific oxidation
states of the target metals. These unique features of this technology make it a potentially
important tool in developing a better understanding of the sources, formation mechanisms, and
transport of Fe, Mn and Cr in the atmosphere.

7.5. Acknowledgements
This study has been supported by the South Coast Air Quality Management District
(AQMD) through award number #11527, and the National Institutes of Health (NIH) through
Grants R01AI065617-13, and R21AG040753-02 to the University of Southern California (USC).
131 | P a g e

The research described herein has not been subjected to the agency’s required peer and policy
review and, therefore, does not necessarily reflect the views of the agency, and no official
endorsement should be inferred. Mention of trade names or commercial products does not
constitute an endorsement or recommendation for use. The authors also acknowledge the support
of USC Viterbi Graduate School fellowship.

132 | P a g e

Chapter 8 Conclusions and recommendations for future research
8.1. Conclusions
This thesis focuses on the development and evaluations of a series of novel technologies
in characterizing physical, chemical and toxicological properties of ambient particulate matter
(PM). All these novel tools presented in this dissertation are well-designed and thoroughly
evaluated both in laboratory experiments and in real world field operations. Results from
laboratory and field evaluation tests clearly demonstrate that overall techniques developed in this
thesis have very good system performance, in terms of both collection efficiency and
measurement accuracy, making them very promising tools in advancing the current knowledge
of PM sampling and measurements technologies, as well as providing significant insight on
sources, formation and toxicological cycling of specific ambient PM components.

As described in Chapter 2-7, the two-stage VI system (Chapter 2) is appropriate for
providing a robust and consistent high concentration enrichment for different types of particles, a
feature that makes it a potentially valuable tool in applications requiring highly concentrated
particles especially at low flow rates (on the order of 1-2 L/min). The Aerosol-Into-Liquid
Collector (Chapter 4), on the other hand, is capable of high volume collection of fine and
ultrafine particles directly into aqueous suspensions. Therefore, it is highly adaptable and
versatile for potential applications in long-term, in-situ measurement of PM chemistry by
overcoming the drawbacks of traditional sampling methods, both in terms of detection limit and
time resolution. Based on the successful development of this new developed PM collector, novel
monitors for online, in-situ measurement of important metals species (i.e. Cu, Fe, Mn and Cr) in
ambient size fractioned PM were further developed (Chapter 5-7). All these metal monitors have
the unique ability to could achieve near-continuous collection and measurements for at least 4 to
7 days without any obvious shortcomings in its operation. More importantly, measurements of
both total and water-soluble fractions, as well as specific oxidation states of target metals can be
achieved in these metal monitors, a significant novelty that can advance a better understanding of
the sources, formation mechanisms, and transport of these toxic relevant metals in the
atmosphere.

133 | P a g e

The developed techniques will potentially create the foundation for transformative
changes in air pollution monitoring that will more directly connect monitoring goals with the
protection of public health. Such technologies will provide significant insights on the
measurements of detailed speciation (i.e. the total and water-soluble fractions, as well as the
oxidation states) of the redox active metals in PM, which is critical as being the biologically
relevant fraction of PM for many toxicological pathways. The deployment of these tools with
online sampling and analysis methods will avoid artifacts associated with collection of
particulate matter on filter and impactors substrates, and will allow the investigations of
diurnal/temporal changes in PM components and chemistry that is expected to change with
sources and atmospheric aging and redox cycling. These advances, in themselves, will greatly
advance the capabilities of atmospheric pollution monitoring and efforts to understand the
chemical and biologically active components of PM. Ultimately, the use of biologically active
PM components to connect sources of ambient PM will create better connections with human
health studies and accelerate the understanding of how PM impacts public health and the
promulgation of effective control strategies to better protect public health.

8.2. Recommendations for future research
8.2.1. Limitations of current studies
While this dissertation contributes to the current state of scientific knowledge on PM
collection and measurement technologies, a consideration of the limitations of the presented
work is necessary to further advance our knowledge on characterization of PM properties and
related methodologies.

Although the developed the technologies are versatile and robust in system operations
and measurements accuracies, the online measurements achieved in this dissertation were
restricted to several important redox active metals. Other chemical groups which are also
relevant to PM induced toxicity, for example carbonaceous components or organic compounds,
are currently under evaluated (Saffari et al., 2014b; Schauer et al., 2003). Previous studies have
demonstrated that there are relatively large spatial and temporal variations of PM-bound organic
components in urban atmospheric environment (Fine et al., 2004; Saarikoski et al., 2008), which
134 | P a g e

is generated from both primary emissions and secondary formations in the atmosphere (Seinfeld
and Pandis, 2006). Therefore, there are still great needs of advance measurement technologies
with high time resolutions in detecting organic components in ambient PM. In current studies,
since the developments are mostly focused on measurements of transition metals, investigations
based only on metal measurements may not provide rigorous and comprehensive conclusions on
possible atmospheric processes and formation mechanisms of PM that influence the PM induced
toxicities.

Another point of consideration lies on some potential sampling artifact in the
technologies developed in this thesis, especially the positive sampling artifact of organic carbon
(as discussed in Chapter 4) coming from condensation of gas phase organic vapors. Although
detailed experiments have been designed and conducted to evaluate the level of positive artifact
of OC/WSOC in slurry samples collected by Aerosol-Into-Liquid Collector, and indeed the
levels of artifact are not really significant, improvements on the current system configurations
are still needed if one attempt to eliminate such sampling artifacts. In other particle-into-liquid
collectors, a denuder is usually deployed after the aerosol inlet to remove the gas phase organic
vapors in sampled air, and therefore such positive artifact can be eliminated (Weber et al., 2001).
However, this is not a possible option for current Aerosol-Into-Liquid Collector, mostly due to
its high operating flow rate (200L/min). More efforts are still needed in addressing the potential
issues on positive sampling artifact for volatile organic compounds in technologies described in
this thesis.

8.2.2. Recommendations for future research
The technologies described in this dissertation have provided value implications to
current development of advanced PM sampling and measurement technologies. Although the
results are limited to measurements of a certain group of metals, researchers looking to discover
new PM sampling and measurements technologies can use these tools as a foundation for similar
future development. For example, in addition to the online PM chemistry measurement, several
studies have also attempted to develop continuous measurements of PM oxidative potential using
mostly, if not exclusively, abiotic, non-cellular methods (Fang et al., 2015; Miljevic et al., 2010),
which can, at least in theory, be used in a high-throughput screening approach to detect redox
135 | P a g e

active PM species. Similar assays and detection methods can be readily combined with the
advance PM sampling system developed in current studies. Moreover, cell-based assays,
especially macrophage models, can also be used for continuous measurements of PM induced
oxidative stress in a cellular assay. Eventually the tight coupling of online chemical speciation
with oxidative stress measurements may open new possibilities to understand the biological
active components of PM and how these biologically active components change with
atmospheric processing and with sources variation.

Another aspect for possible future research is applications of investigations of spatial
and/or temporal variations of target metals and their specific chemical forms (i.e. water-soluble
fractions and oxidation states) by deploying these monitors for long-term sampling campaigns.
The unique abilities of these monitors in prolonged unattended operations and determination of
target metals speciation make them powerful tools for such applications, which were previously
restrained by conventional methodologies. By obtaining semi-continuous measurements for these
metals with high time resolutions, researchers could attempt to investigate the diurnal variations
of these metals in atmosphere, and further link these observations to obtain a better
understanding of the sources, formation mechanisms, and transport of these toxic relevant metals
in the atmosphere.
136 | P a g e

Bibliography:

Amato, F., Pandolfi, M., Moreno, T., Furger, M., Pey, J., Alastuey, A., Bukowiecki, N., Prevot,
A.S.H., Baltensperger, U., Querol, X., 2011. Sources and variability of inhalable road dust
particles in three European cities. Atmos. Environ. 45, 6777–6787.
doi:10.1016/j.atmosenv.2011.06.003
Araujo, J.A., Barajas, B., Kleinman, M., Wang, X., Bennett, B.J., Gong, K.W., Navab, M.,
Harkema, J., Sioutas, C., Lusis, A.J., Nel, A.E., 2008. Ambient Particulate Pollutants in the
Ultrafine Range Promote Early Atherosclerosis and Systemic Oxidative Stress. Circ. Res. 102,
589–596.
Ashley, K., Howe, A.M., Demange, M., Nygren, O., 2003. Sampling and analysis considerations
for the determination of hexavalent chromium in workplace air. J. Environ. Monit. 5, 707.
doi:10.1039/b306105c
Becker, S., Mundandhara, S., Devlin, R.B., Madden, M., 2005. Regulation of cytokine
production in human alveolar macrophages and airway epithelial cells in response to ambient air
pollution particles: Further mechanistic studies. Toxicol. Appl. Pharmacol., Living in a Safe
Chemical World. Proceedings of the 10th International Congress of Toxicology 11-15 July, 2004,
Tampere, Finland 207, 269–275. doi:10.1016/j.taap.2005.01.023
Becker, S., Soukup, J., 2003. Coarse(PM 2.5-10 ), Fine(PM 2.5 ), and Ultrafine Air Pollution
Particles Induce/Increase Immune Costimulatory Receptors on Human Blood-Derived
Monocytes but not on Alveolar Macrophages. J. Toxicol. Environ. Health A 66, 847–859.
doi:10.1080/15287390306381
Buhr, S.M., Burr, M.P., Fehsenfeld, F.C., Holloway, J.S., Karst, U., Norton, R.B., Parrish, D.D.,
Sievers, R.E., 1995. Development of a semi-continuous method for the measurement of nitric
acid vapor and particulate nitrate and sulfate. Atmos. Environ. 29, 2609–2624.
doi:10.1016/1352-2310(95)00171-T
Campbell, A., Oldham, M., Becaria, A., Bondy, S.C., Meacher, D., Sioutas, C., Misra, C.,
Mendez, L.B., Kleinman, M., 2005. Particulate Matter in Polluted Air May Increase Biomarkers
of Inflammation in Mouse Brain. NeuroToxicology 26, 133–140.
doi:10.1016/j.neuro.2004.08.003
Castranova, V., Ma, J.Y., Yang, H.M., Antonini, J.M., Butterworth, L., Barger, M.W., Roberts,
J., Ma, J.K., 2001. Effect of exposure to diesel exhaust particles on the susceptibility of the lung
to infection. Environ. Health Perspect. 109, 609–612.
Chang, M.C., Sioutas, C., Kim, S., Gong Jr., H., Linn, W.S., 2000. Reduction of nitrate losses
from filter and impactor samplers by means of concentration enrichment. Atmos. Environ. 34,
85–98. doi:10.1016/S1352-2310(99)00308-8
Charrier, J.G., Anastasio, C., 2012. On dithiothreitol (DTT) as a measure of oxidative potential
for ambient particles: evidence for the importance of soluble transition metals. Atmos Chem
Phys Discuss 12, 11317–11350. doi:10.5194/acpd-12-11317-2012
137 | P a g e

Chen, B.T., Yeh, H.C., 1987. An improved virtual impactor: Design and performance. J. Aerosol
Sci. 18, 203–214. doi:10.1016/0021-8502(87)90056-5
Chester, R., Stoner, J.H., 1974. The distribution of Mn, Fe, Cu, Ni, Co, Ga, Cr, V, Ba, Sr, Sn, Zn,
and Pb, in some soil-sized particulates from the lower troposphere over the world ocean. Mar.
Chem. 2, 157–188. doi:10.1016/0304-4203(74)90013-9
Cheung, K., Daher, N., Kam, W., Shafer, M.M., Ning, Z., Schauer, J.J., Sioutas, C., 2011.
Spatial and temporal variation of chemical composition and mass closure of ambient coarse
particulate matter (PM10–2.5) in the Los Angeles area. Atmos. Environ. 45, 2651–2662.
doi:10.1016/j.atmosenv.2011.02.066
Cho, A.K., Sioutas, C., Miguel, A.H., Kumagai, Y., Schmitz, D.A., Singh, M., Eiguren-
Fernandez, A., Froines, J.R., 2005. Redox activity of airborne particulate matter at different sites
in the Los Angeles Basin. Environ. Res. 99, 40–47. doi:10.1016/j.envres.2005.01.003
Chow, J.C., Watson, J.G., Fujita, E.M., Lu, Z., Lawson, D.R., Ashbaugh, L.L., 1994. Temporal
and spatial variations of PM2.5 and PM10 aerosol in the Southern California air quality study.
Atmos. Environ. 28, 2061–2080. doi:10.1016/1352-2310(94)90474-X
Claiborn, C.S., Larson, T., Sheppard, L., 2002. Testing the metals hypothesis in Spokane,
Washington. Environ. Health Perspect. 110, 547–552.
Creatchman, M., 1999. Elemental Analysis of Airborne Particles. CRC Press.
Cruz, C.N., Pandis, S.N., 1999. Condensation of Organic Vapors on an Externally Mixed
Aerosol Population. Aerosol Sci. Technol. 31, 392–407. doi:10.1080/027868299304110
Daher, N., Hasheminassab, S., Shafer, M.M., Schauer, J.J., Sioutas, C., 2013. Seasonal and
spatial variability in chemical composition and mass closure of ambient ultrafine particles in the
megacity of Los Angeles. Env. Sci Process. Impacts 15, 283–295. doi:10.1039/C2EM30615H
Daher, N., Ning, Z., Cho, A.K., Shafer, M., Schauer, J.J., Sioutas, C., 2011. Comparison of the
Chemical and Oxidative Characteristics of Particulate Matter (PM) Collected by Different
Methods: Filters, Impactors, and BioSamplers. Aerosol Sci. Technol. 45, 1294–1304.
doi:10.1080/02786826.2011.590554
Daher, N., Ruprecht, A., Invernizzi, G., De Marco, C., Miller-Schulze, J., Heo, J.B., Shafer,
M.M., Shelton, B.R., Schauer, J.J., Sioutas, C., 2012. Characterization, sources and redox
activity of fine and coarse particulate matter in Milan, Italy. Atmos. Environ. 49, 130–141.
doi:10.1016/j.atmosenv.2011.12.011
Davis, D.A., Akopian, G., Walsh, J.P., Sioutas, C., Morgan, T.E., Finch, C.E., 2013a. Urban air
pollutants reduce synaptic function of CA1 neurons via an NMDA/NȮ pathway in vitro. J.
Neurochem. 127, 509–519. doi:10.1111/jnc.12395
Davis, D.A., Bortolato, M., Godar, S.C., Sander, T.K., Iwata, N., Pakbin, P., Shih, J.C., Berhane,
K., McConnell, R., Sioutas, C., Finch, C.E., Morgan, T.E., 2013b. Prenatal Exposure to Urban
Air Nanoparticles in Mice Causes Altered Neuronal Differentiation and Depression-Like
Responses. PLOS ONE 8, e64128. doi:10.1371/journal.pone.0064128
138 | P a g e

DeCarlo, P.F., Kimmel, J.R., Trimborn, A., Northway, M.J., Jayne, J.T., Aiken, A.C., Gonin, M.,
Fuhrer, K., Horvath, T., Docherty, K.S., Worsnop, D.R., Jimenez, J.L., 2006. Field-Deployable,
High-Resolution, Time-of-Flight Aerosol Mass Spectrometer. Anal. Chem. 78, 8281–8289.
doi:10.1021/ac061249n
Delfino, R.J., Gillen, D.L., Tjoa, T., Staimer, N., Polidori, A., Arhami, M., Sioutas, C.,
Longhurst, J., 2011. Electrocardiographic ST-Segment Depression and Exposure to Traffic-
Related Aerosols in Elderly Subjects with Coronary Artery Disease. Environ. Health Perspect.
119, 196–202. doi:10.1289/ehp.1002372
Delfino, R.J., Sioutas, C., Malik, S., 2005. Potential Role of Ultrafine Particles in Associations
between Airborne Particle Mass and Cardiovascular Health. Environ. Health Perspect. 113, 934–
946.
Dellinger, B., Pryor, W.A., Cueto, R., Squadrito, G.L., Hegde, V., Deutsch, W.A., 2001. Role of
Free Radicals in the Toxicity of Airborne Fine Particulate Matter. Chem. Res. Toxicol. 14,
1371–1377. doi:10.1021/tx010050x
Demokritou, P., Gupta, T., Ferguson, S., Koutrakis, P., 2003. Development of a High-Volume
Concentrated Ambient Particles System (CAPS) for Human and Animal Inhalation
Toxicological Studies. Inhal. Toxicol. 15, 111–129. doi:10.1080/08958370304475
Demokritou, P., Gupta, T., Koutrakis, P., 2002. A High Volume Apparatus for the
Condensational Growth of Ultrafine Particles for Inhalation Toxicological Studies. Aerosol Sci.
Technol. 36, 1061–1072. doi:10.1080/02786820290092230
Dick, C.A.J., Stone, V., Brown, D.M., Watt, M., Cherrie, J.W., Howarth, S., Seaton, A.,
Donaldson, K., 2000. Toxic and inflammatory effects of filters frequently used for the collection
of airborne particulate matter. Atmos. Environ. 34, 2587–2592. doi:10.1016/S1352-
2310(99)00476-8
Docherty, K.S., Stone, E.A., Ulbrich, I.M., DeCarlo, P.F., Snyder, D.C., Schauer, J.J., Peltier,
R.E., Weber, R.J., Murphy, S.M., Seinfeld, J.H., Grover, B.D., Eatough, D.J., Jimenez, J.L.,
2008. Apportionment of Primary and Secondary Organic Aerosols in Southern California during
the 2005 Study of Organic Aerosols in Riverside (SOAR-1). Environ. Sci. Technol. 42, 7655–
7662. doi:10.1021/es8008166
Dominici F, Peng RD, Bell ML, et al, 2006. FIne particulate air pollution and hospital admission
for cardiovascular and respiratory diseases. JAMA 295, 1127–1134.
doi:10.1001/jama.295.10.1127
Donaldson, K., Brown, D., Clouter, A., Duffin, R., MacNee, W., Renwick, L., Tran, L., Stone,
V., 2002. The Pulmonary Toxicology of Ultrafine Particles. J. Aerosol Med. 15, 213–220.
doi:10.1089/089426802320282338
Dreher, K.L., Jaskot, R.H., Lehmann, J.R., 1997. Soluble Transition Metals Mediate Residual
Oil Fly Ash Induced Acute Lung Injury. J. Toxicol. Environ. Health 50.
139 | P a g e

Eatough, D.J., Long, R.W., Modey, W.K., Eatough, N.L., 2003. Semi-volatile secondary organic
aerosol in urban atmospheres: meeting a measurement challenge. Atmos. Environ., James P.
Lodge, Jr. Memorial Issue. Measurement Issues in Atmospheric Chemistry 37, 1277–1292.
doi:10.1016/S1352-2310(02)01020-8
Eller, P.M., 1994. NIOSH Manual of Analytical Methods. DIANE Publishing.
Fach, E., Waldman, W.J., Williams, M., Long, J., Meister, R.K., Dutta, P.K., 2002. Analysis of
the biological and chemical reactivity of zeolite-based aluminosilicate fibers and particulates.
Environ. Health Perspect. 110, 1087–1096.
Fang, T., Guo, H., Verma, V., Peltier, R.E., Weber, R.J., 2015a. PM2.5 water-soluble elements
in the southeastern United States: automated analytical method development, spatiotemporal
distributions, source apportionment, and implications for heath studies. Atmos Chem Phys 15,
11667–11682. doi:10.5194/acp-15-11667-2015
Fang, T., Verma, V., Bates, J.T., Abrams, J., Klein, M., Strickland, M.J., Sarnat, S.E., Chang,
H.H., Mulholland, J.A., Tolbert, P.E., Russell, A.G., Weber, R.J., 2015b. Oxidative potential of
ambient water-soluble PM2.5 measured by Dithiothreitol (DTT) and Ascorbic Acid (AA) assays
in the southeastern United States: contrasts in sources and health associations. Atmospheric
Chem. Phys. Discuss. 15, 30609–30644. doi:10.5194/acpd-15-30609-2015
Fang, T., Verma, V., Guo, H., King, L.E., Edgerton, E.S., Weber, R.J., 2014. A semi-automated
system for quantifying the oxidative potential of ambient particles in aqueous extracts using the
dithiothreitol (DTT) assay: results from the Southeastern Center for Air Pollution and
Epidemiology (SCAPE). Atmos Meas Tech Discuss 7, 7245–7279. doi:10.5194/amtd-7-7245-
2014
Fine, P.M., Chakrabarti, B., Krudysz, M., Schauer, J.J., Sioutas, C., 2004. Diurnal Variations of
Individual Organic Compound Constituents of Ultrafine and Accumulation Mode Particulate
Matter in the Los Angeles Basin. Environ. Sci. Technol. 38, 1296–1304. doi:10.1021/es0348389
Gard, E., Mayer, J.E., Morrical, B.D., Dienes, T., Fergenson, D.P., Prather, K.A., 1997. Real-
Time Analysis of Individual Atmospheric Aerosol Particles: Design and Performance of a
Portable ATOFMS. Anal. Chem. 69, 4083–4091. doi:10.1021/ac970540n
Garshick, E., Laden, F., Hart, J.E., Rosner, B., Smith, T.J., Dockery, D.W., Speizer, F.E., 2004.
Lung Cancer in Railroad Workers Exposed to Diesel Exhaust. Environ. Health Perspect. 112,
1539–1543. doi:10.1289/ehp.7195
Gasser, M., Riediker, M., Mueller, L., Perrenoud, A., Blank, F., Gehr, P., Rothen-Rutishauser, B.,
2009. Toxic effects of brake wear particles on epithelial lung cells in vitro. Part. Fibre Toxicol. 6,
1–13. doi:10.1186/1743-8977-6-30
Gatto, N.M., Henderson, V.W., Hodis, H.N., St. John, J.A., Lurmann, F., Chen, J.-C., Mack,
W.J., 2014. Components of air pollution and cognitive function in middle-aged and older adults
in Los Angeles. NeuroToxicology 40, 1–7. doi:10.1016/j.neuro.2013.09.004
140 | P a g e

Geller, M.D., Biswas, S., Fine, P.M., Sioutas, C., 2005. A new compact aerosol concentrator for
use in conjunction with low flow-rate continuous aerosol instrumentation. J. Aerosol Sci. 36,
1006–1022. doi:10.1016/j.jaerosci.2004.11.015
Gong, S.L., 2003. A parameterization of sea-salt aerosol source function for sub- and super-
micron particles. Glob. Biogeochem. Cycles 17, 1097. doi:10.1029/2003GB002079
Gupta, K.C., D’Arc, M.J., 2000. Performance evaluation of copper ion selective electrode based
on cyanocopolymers. Sens. Actuators B Chem. 62, 171–176. doi:10.1016/S0925-
4005(99)00362-7
Harrison, R.M., Jones, A.M., Gietl, J., Yin, J., Green, D.C., 2012. Estimation of the
Contributions of Brake Dust, Tire Wear, and Resuspension to Nonexhaust Traffic Particles
Derived from Atmospheric Measurements. Environ. Sci. Technol. 46, 6523–6529.
doi:10.1021/es300894r
Harrison, R.M., Yin, J., Mark, D., Stedman, J., Appleby, R.S., Booker, J., Moorcroft, S., 2001.
Studies of the coarse particle (2.5–10 μm) component in UK urban atmospheres. Atmos. Environ.
35, 3667–3679. doi:10.1016/S1352-2310(00)00526-4
Haynes, W.M., 2013. Handbook of Chemistry and Physics, 94th ed. CRC Press, Ann Arbor,
Michigan.
Hays, M.D., Cho, S.-H., Baldauf, R., Schauer, J.J., Shafer, M., 2011. Particle size distributions of
metal and non-metal elements in an urban near-highway environment. Atmos. Environ. 45, 925–
934. doi:10.1016/j.atmosenv.2010.11.010
Heal, M.R., Hibbs, L.R., Agius, R.M., Beverland, I.J., 2005. Total and water-soluble trace metal
content of urban background PM10, PM2.5 and black smoke in Edinburgh, UK. Atmos. Environ.
39, 1417–1430. doi:10.1016/j.atmosenv.2004.11.026
Hinds, W.C., 2012. Aerosol Technology: Properties, Behavior, and Measurement of Airborne
Particles. John Wiley & Sons.
Hjortenkrans, D.S.T., Bergbä ck, B.G., Hä ggerud, A.V., 2007. Metal Emissions from Brake
Linings and Tires: Case Studies of Stockholm, Sweden 1995/1998 and 2005. Environ. Sci.
Technol. 41, 5224–5230. doi:10.1021/es070198o
Hueglin, C., Gehrig, R., Baltensperger, U., Gysel, M., Monn, C., Vonmont, H., 2005. Chemical
characterisation of PM2.5, PM10 and coarse particles at urban, near-city and rural sites in
Switzerland. Atmos. Environ. 39, 637–651. doi:10.1016/j.atmosenv.2004.10.027
Hu, S., Polidori, A., Arhami, M., Shafer, M.M., Schauer, J.J., Cho, A., Sioutas, C., 2008. Redox
activity and chemical speciation of size fractioned PM in the communities of the Los Angeles-
Long Beach harbor. Atmos Chem Phys 8, 6439–6451. doi:10.5194/acp-8-6439-2008
Invernizzi, G., Ruprecht, A., Mazza, R., De Marco, C., Močnik, G., Sioutas, C., Westerdahl, D.,
2011. Measurement of black carbon concentration as an indicator of air quality benefits of traffic
restriction policies within the ecopass zone in Milan, Italy. Atmos. Environ. 45, 3522–3527.
doi:10.1016/j.atmosenv.2011.04.008
141 | P a g e

Jalava, P.I., Salonen, R.O., Pennanen, A.S., Happo, M.S., Penttinen, P., Hä linen, A.I., Sillanpä ä ,
M., Hillamo, R., Hirvonen, M.-R., 2008. Effects of solubility of urban air fine and coarse
particles on cytotoxic and inflammatory responses in RAW 264.7 macrophage cell line. Toxicol.
Appl. Pharmacol. 229, 146–160. doi:10.1016/j.taap.2008.01.006
Jalava, P., Salonen, R.O., Hä linen, A.I., Sillanpä ä , M., Sandell, E., Hirvonen, M.-R., 2005.
Effects of Sample Preparation on Chemistry, Cytotoxicity, and Inflammatory Responses Induced
by Air Particulate Matter. Inhal. Toxicol. 17, 107–117. doi:10.1080/08958370590899550
Jansen, K.L., Larson, T.V., Koenig, J.Q., Mar, T.F., Fields, C., Stewart, J., Lippmann, M., 2005.
Associations between Health Effects and Particulate Matter and Black Carbon in Subjects with
Respiratory Disease. Environ. Health Perspect. 113, 1741–1746.
J ONDOV, W.K., 1992. Tracing fly ash emitted from a coal-fired power plant with enriched
rare-earth isotopes: An urban scale test. Atmospheric Environ. Part B Urban Atmosphere 26,
453–462. doi:10.1016/0957-1272(92)90052-T
Karlsson, A., Irgum, K., Hansson, H.-C., 1997. Single-stage flowing liquid film impactor for
continuous on-line particle analysis. J. Aerosol Sci. 28, 1539–1551. doi:10.1016/S0021-
8502(97)00013-X
KKarthikeyan, S., Joshi, U.M., Balasubramanian, R., 2006. Microwave assisted sample
preparation for determining water-soluble fraction of trace elements in urban airborne particulate
matter: Evaluation of bioavailability. Anal. Chim. Acta, Asianalysis VIII Papers presented at the
8th Asian Conference on Analytical Chemistry (Asianalysis VIII, Taipei, Taiwan, 16-20 October
2005. 576, 23–30. doi:10.1016/j.aca.2006.05.051
Kelly, F.J., 2003. Oxidative stress: its role in air pollution and adverse health effects. Occup.
Environ. Med. 60, 612–616. doi:10.1136/oem.60.8.612
Khlystov, A., Ma, Y., 2006. An on-line instrument for mobile measurements of the spatial
variability of hexavalent and trivalent chromium in urban air. Atmos. Environ. 40, 8088–8093.
doi:10.1016/j.atmosenv.2006.09.030
Khlystov, A., Zhang, Q., Jimenez, J.L., Stanier, C., Pandis, S.N., Canagaratna, M.R., Fine, P.,
Misra, C., Sioutas, C., 2005. In situ concentration of semi-volatile aerosol using water-
condensation technology. J. Aerosol Sci. 36, 866–880. doi:10.1016/j.jaerosci.2004.11.005
Kidwell, C.B., Ondov, J.M., 2004. Elemental Analysis of Sub-Hourly Ambient Aerosol
Collections. Aerosol Sci. Technol. 38, 205–218. doi:10.1080/02786820490261726
Kidwell, C.B., Ondov, J.M., 2001. Development and Evaluation of a Prototype System for
Collecting Sub-Hourly Ambient Aerosol for Chemical Analysis. Aerosol Sci. Technol. 35, 596–
601. doi:10.1080/02786820152051445
Kim, S., Jaques, P.A., Chang, M., Barone, T., Xiong, C., Friedlander, S.K., Sioutas, C., 2001a.
Versatile aerosol concentration enrichment system (VACES) for simultaneous in vivo and in
vitro evaluation of toxic effects of ultrafine, fine and coarse ambient particles Part II: Field
evaluation. J. Aerosol Sci. 32, 1299–1314. doi:10.1016/S0021-8502(01)00058-1
142 | P a g e

Kim, S., Jaques, P.A., Chang, M., Froines, J.R., Sioutas, C., 2001b. Versatile aerosol
concentration enrichment system (VACES) for simultaneous in vivo and in vitro evaluation of
toxic effects of ultrafine, fine and coarse ambient particles Part I: Development and laboratory
characterization. J. Aerosol Sci. 32, 1281–1297. doi:10.1016/S0021-8502(01)00057-X
Kleinman, M.T., Araujo, J.A., Nel, A., Sioutas, C., Campbell, A., Cong, P.Q., Li, H., Bondy,
S.C., 2008. Inhaled ultrafine particulate matter affects CNS inflammatory processes and may act
via MAP kinase signaling pathways. Toxicol. Lett. 178, 127–130.
doi:10.1016/j.toxlet.2008.03.001
Landreman, A.P., Shafer, M.M., Hemming, J.C., Hannigan, M.P., Schauer, J.J., 2008. A
Macrophage-Based Method for the Assessment of the Reactive Oxygen Species (ROS) Activity
of Atmospheric Particulate Matter (PM) and Application to Routine (Daily-24 h) Aerosol
Monitoring Studies. Aerosol Sci. Technol. 42, 946–957. doi:10.1080/02786820802363819
Lin, C.-C., Chen, S.-J., Huang, K.-L., Hwang, W.-I., Chang-Chien, G.-P., Lin, W.-Y., 2005.
Characteristics of Metals in Nano/Ultrafine/Fine/Coarse Particles Collected Beside a Heavily
Trafficked Road. Environ. Sci. Technol. 39, 8113–8122. doi:10.1021/es048182a
Li, N., Hao, M., Phalen, R.F., Hinds, W.C., Nel, A.E., 2003a. Particulate air pollutants and
asthma: A paradigm for the role of oxidative stress in PM-induced adverse health effects. Clin.
Immunol. 109, 250–265. doi:10.1016/j.clim.2003.08.006
Li, N., Sioutas, C., Cho, A., Schmitz, D., Misra, C., Sempf, J., Wang, M., Oberley, T., Froines, J.,
Nel, A., 2003b. Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage.
Environ. Health Perspect. 111, 455–460.
Li, N., Wang, M., Bramble, L.A., Schmitz, D.A., Schauer, J.J., Sioutas, C., Harkema, J.R., Nel,
A.E., 2009. The Adjuvant Effect of Ambient Particulate Matter Is Closely Reflected by the
Particulate Oxidant Potential. Environ. Health Perspect. 117, 1116–1123.
doi:10.1289/ehp.0800319
Li, N., Xia, T., Nel, A.E., 2008. The role of oxidative stress in ambient particulate matter-
induced lung diseases and its implications in the toxicity of engineered nanoparticles. Free Radic.
Biol. Med. 44, 1689–1699. doi:10.1016/j.freeradbiomed.2008.01.028
Lin, X., Willeke, K., Ulevicius, V., Grinshpun, S.A., 1997. Effect of Sampling Time on the
Collection Efficiency of All-Glass Impingers. Am. Ind. Hyg. Assoc. J. 58, 480–488.
doi:10.1080/15428119791012577
Li, R., Navab, M., Pakbin, P., Ning, Z., Navab, K., Hough, G., Morgan, T.E., Finch, C.E., Araujo,
J.A., Fogelman, A.M., Sioutas, C., Hsiai, T., 2013. Ambient ultrafine particles alter lipid
metabolism and HDL anti-oxidant capacity in LDLR-null mice. J. Lipid Res. 54, 1608–1615.
doi:10.1194/jlr.M035014
Majestic, B.J., Schauer, J.J., Shafer, M.M., 2007. Development of a Manganese Speciation
Method for Atmospheric Aerosols in Biologically and Environmentally Relevant Fluids. Aerosol
Sci. Technol. 41, 925–933. doi:10.1080/02786820701564657
143 | P a g e

Majestic, B.J., Schauer, J.J., Shafer, M.M., Turner, J.R., Fine, P.M., Singh, M., Sioutas, C., 2006.
Development of a Wet-Chemical Method for the Speciation of Iron in Atmospheric Aerosols.
Environ. Sci. Technol. 40, 2346–2351. doi:10.1021/es052023p
Malm, W.C., Schichtel, B.A., Pitchford, M.L., Ashbaugh, L.L., Eldred, R.A., 2004. Spatial and
monthly trends in speciated fine particle concentration in the United States. J. Geophys. Res.
Atmospheres 109, D03306. doi:10.1029/2003JD003739
Manoli, E., Voutsa, D., Samara, C., 2002. Chemical characterization and source
identification/apportionment of fine and coarse air particles in Thessaloniki, Greece. Atmos.
Environ. 36, 949–961. doi:10.1016/S1352-2310(01)00486-1
Marple, V.A., Chien, C.M., 1980. Virtual impactors: a theoretical study. Environ. Sci. Technol.
14, 976–985. doi:10.1021/es60168a019
Marple, V.A., Willeke, K., 1976. Impactor design. Atmospheric Environ. 1967 10, 891–896.
doi:10.1016/0004-6981(76)90144-X
Martinet, W., De Meyer, G.R.Y., Herman, A.G., Kockx, M.M., 2004. Reactive oxygen species
induce RNA damage in human atherosclerosis. Eur. J. Clin. Invest. 34, 323–327.
doi:10.1111/j.1365-2362.2004.01343.x
Midander, K., Cronholm, P., Karlsson, H.L., Elihn, K., Mö ller, L., Leygraf, C., Wallinder, I.O.,
2009. Surface Characteristics, Copper Release, and Toxicity of Nano- and Micrometer-Sized
Copper and Copper(II) Oxide Particles: A Cross-Disciplinary Study. Small 5, 389–399.
doi:10.1002/smll.200801220
Miljevic, B., Fairfull-Smith, K.E., Bottle, S.E., Ristovski, Z.D., 2010. The application of
profluorescent nitroxides to detect reactive oxygen species derived from combustion-generated
particulate matter: Cigarette smoke – A case study. Atmos. Environ. 44, 2224–2230.
doi:10.1016/j.atmosenv.2010.02.043
Misra, C., Fine, P.M., Singh, M., Sioutas, C., 2004. Development and Evaluation of a Compact
Facility for Exposing Humans to Concentrated Ambient Ultrafine Particles. Aerosol Sci. Technol.
38, 27–35. doi:10.1080/02786820490247605
Misra, C., Kim, S., Shen, S., Sioutas, C., 2002. A high flow rate, very low pressure drop
impactor for inertial separation of ultrafine from accumulation mode particles. J. Aerosol Sci. 33,
735–752. doi:10.1016/S0021-8502(01)00210-5
Montaser, A., 1998. Inductively Coupled Plasma Mass Spectrometry. Wiley.
Morgan, J.J., Stumm, W., 1965. Analytical Chemistry of Aqueous Manganese. J. Am. Water
Works Assoc. 57, 107–119.
Morgan, T.E., Davis, D.A., Iwata, N., Tanner, J.A., Snyder, D., Ning, Z., Kam, W., Hsu, Y.-T.,
Winkler, J.W., Chen, J.-C., Petasis, N.A., Baudry, M., Sioutas, C., Finch, C.E., 2011.
Glutamatergic Neurons in Rodent Models Respond to Nanoscale Particulate Urban Air
Pollutants in Vivo and in Vitro. Environ. Health Perspect. 119, 1003–1009.
doi:10.1289/ehp.1002973
144 | P a g e

Mozurkewich, M., 1993. The dissociation constant of ammonium nitrate and its dependence on
temperature, relative humidity and particle size. Atmospheric Environ. Part Gen. Top. 27, 261–
270. doi:10.1016/0960-1686(93)90356-4
Ning, Z., Geller, M.D., Moore, K.F., Sheesley, R., Schauer, J.J., Sioutas, C., 2007. Daily
Variation in Chemical Characteristics of Urban Ultrafine Aerosols and Inference of Their
Sources. Environ. Sci. Technol. 41, 6000–6006. doi:10.1021/es070653g
Ning, Z., Moore, K.F., Polidori, A., Sioutas, C., 2006. Field Validation of the New Miniature
Versatile Aerosol Concentration Enrichment System (mVACES). Aerosol Sci. Technol. 40,
1098–1110. doi:10.1080/02786820600996422
Nishida, S., Teramoto, K., Kimoto-Kinoshita, S., Tohda, Y., Nakajima, S., Tomura, T.T.,
Irimajiri, K., 2002. Change of Cu,Zn-superoxide Dismutase Activity of Guinea Pig Lung in
Experimental Asthma. Free Radic. Res. 36, 601–606. doi:10.1080/10715760210872
Ntziachristos, L., Froines, J.R., Cho, A.K., Sioutas, C., 2007. Relationship between redox
activity and chemical speciation of size-fractionated particulate matter. Part. Fibre Toxicol. 4, 5.
doi:10.1186/1743-8977-4-5
Orsini, D.A., Ma, Y., Sullivan, A., Sierau, B., Baumann, K., Weber, R.J., 2003. Refinements to
the particle-into-liquid sampler (PILS) for ground and airborne measurements of water soluble
aerosol composition. Atmos. Environ., James P. Lodge, Jr. Memorial Issue. Measurement Issues
in Atmospheric Chemistry 37, 1243–1259. doi:10.1016/S1352-2310(02)01015-4
Orsini, D.A., Rhoads, K., McElhoney, K., Schick, E., Koehler, D., Hogrefe, O., 2008. A Water
Cyclone to Preserve Insoluble Aerosols in Liquid Flow—An Interface to Flow Cytometry to
Detect Airborne Nucleic Acid. Aerosol Sci. Technol. 42, 343–356.
doi:10.1080/02786820802072881
Pakbin, P., Ning, Z., Eiguren-Fernandez, A., Sioutas, C., 2011. Modification of the Versatile
Aerosol Concentration Enrichment System (VACES) for conducting inhalation exposures to
semi-volatile vapor phase pollutants. J. Aerosol Sci. 42, 555–566.
doi:10.1016/j.jaerosci.2011.06.002
Pancras, J.P., Landis, M.S., Norris, G.A., Vedantham, R., Dvonch, J.T., 2013. Source
apportionment of ambient fine particulate matter in Dearborn, Michigan, using hourly resolved
PM chemical composition data. Sci. Total Environ., Atmospheric Mercury, Air Pollution, and
Associated Effects on Health: A Festschrift to Professor Jerry Keeler. 448, 2–13.
doi:10.1016/j.scitotenv.2012.11.083
Park, S., Cho, S., Jo, M., Gong, B., Park, J., Lee, S., 2014. Field evaluation of a near–real time
elemental monitor and identification of element sources observed at an air monitoring supersite
in Korea. Atmospheric Pollut. Res. 5, 119–128. doi:10.5094/APR.2014.015
Pastuszka, J.S., Rogula-Kozłowska, W., Zajusz-Zubek, E., 2010. Characterization of PM10 and
PM2.5 and associated heavy metals at the crossroads and urban background site in Zabrze,
Upper Silesia, Poland, during the smog episodes. Environ. Monit. Assess. 168, 613–627.
doi:10.1007/s10661-009-1138-8
145 | P a g e

Petit, J.-E., Favez, O., Sciare, J., Crenn, V., Sarda-Estève, R., Bonnaire, N., Močnik, G., Dupont,
J.-C., Haeffelin, M., Leoz-Garziandia, E., 2015. Two years of near real-time chemical
composition of submicron aerosols in the region of Paris using an Aerosol Chemical Speciation
Monitor (ACSM) and a multi-wavelength Aethalometer. Atmos Chem Phys 15, 2985–3005.
doi:10.5194/acp-15-2985-2015
Phan, H.N., McFarland, A.R., 2004. Aerosol-to-Hydrosol Transfer Stages for Use in Bioaerosol
Sampling. Aerosol Sci. Technol. 38, 300–310. doi:10.1080/02786820490426183
Pope, C.A., Burnett, R.T., Thurston, G.D., Thun, M.J., Calle, E.E., Krewski, D., Godleski, J.J.,
2004. Cardiovascular Mortality and Long-Term Exposure to Particulate Air Pollution
Epidemiological Evidence of General Pathophysiological Pathways of Disease. Circulation 109,
71–77. doi:10.1161/01.CIR.0000108927.80044.7F
Pope, C.A., Dockery, D.W., 2006. Health Effects of Fine Particulate Air Pollution: Lines that
Connect. J. Air Waste Manag. Assoc. 56, 709–42.
Pope, C.A., III, Burnett, R.T., Krewski, D., Jerrett, M., Shi YuanLi, Calle, E.E., Thun, M.J.,
2009. Cardiovascular mortality and exposure to airborne fine particulate matter and cigarette
smoke: shape of the exposure-response relationship. Circulation 120, 941–948.
doi:10.1161/CIRCULATIONAHA.109.857888
Prabhakar, G., Sorooshian, A., Toffol, E., Arellano, A.F., Betterton, E.A., 2014. Spatiotemporal
distribution of airborne particulate metals and metalloids in a populated arid region. Atmos.
Environ. 92, 339–347. doi:10.1016/j.atmosenv.2014.04.044
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., Hü glin, C.,
Jones, A.M., Kasper-Giebl, A., Kiss, G., Kousa, A., Kuhlbusch, T.A.J., Lö schau, G., Maenhaut,
W., Molnar, A., Moreno, T., 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, 1308–1320. doi:10.1016/j.atmosenv.2009.12.011
Rastogi, N., McWhinney, R.D., Akhtar, U.S., Urch, B., Fila, M., Abbatt, J.P.D., Scott, J.A.,
Silverman, F.S., Brook, J.R., Evans, G.J., 2012. Physical Characterization of the University of
Toronto Coarse, Fine, and Ultrafine High-Volume Particle Concentrator Systems. Aerosol Sci.
Technol. 46, 1015–1024. doi:10.1080/02786826.2012.686674
Rastogi, N., Oakes, M.M., Schauer, J.J., Shafer, M.M., Majestic, B.J., Weber, R.J., 2009. New
Technique for Online Measurement of Water-Soluble Fe(II) in Atmospheric Aerosols. Environ.
Sci. Technol. 43, 2425–2430. doi:10.1021/es8031902
Ris, C., 2007. U.S. EPA Health Assessment for Diesel Engine Exhaust: A Review. Inhal.
Toxicol. 19, 229–239. doi:10.1080/08958370701497960
146 | P a g e

Risom, L., Mø ller, P., Loft, S., 2005. Oxidative stress-induced DNA damage by particulate air
pollution. Mutat. Res. Mol. Mech. Mutagen., Linking Toxicology to Epidemiology: Biomarkers
and New Technologies 592, 119–137. doi:10.1016/j.mrfmmm.2005.06.012
Ritz, B., Wilhelm, M., 2008. Ambient Air Pollution and Adverse Birth Outcomes: Methodologic
Issues in an Emerging Field. Basic Clin. Pharmacol. Toxicol. 102, 182–190. doi:10.1111/j.1742-
7843.2007.00161.x
Ritz, B., Yu, F., Fruin, S., Chapa, G., Shaw, G.M., Harris, J.A., 2002. Ambient Air Pollution and
Risk of Birth Defects in Southern California. Am. J. Epidemiol. 155, 17–25.
doi:10.1093/aje/155.1.17
Robinson, A.L., Donahue, N.M., Shrivastava, M.K., Weitkamp, E.A., Sage, A.M., Grieshop,
A.P., Lane, T.E., Pierce, J.R., Pandis, S.N., 2007. Rethinking Organic Aerosols: Semivolatile
Emissions and Photochemical Aging. Science 315, 1259–1262. doi:10.1126/science.1133061
Rundle, C.C., 2000. A beginners guide to ion-selective electrode measurements. Nico2000 Ltd
Lond.
Saarikoski, S., Timonen, H., Saarnio, K., Aurela, M., Jä rvi, L., Keronen, P., Kerminen, V.-M.,
Hillamo, R., 2008. Sources of organic carbon in fine particulate matter in northern European
urban air. Atmospheric Chem. Phys. 8, 6281–6295. doi:10.5194/acp-8-6281-2008
Saffari, A., Daher, N., Shafer, M.M., Schauer, J.J., Sioutas, C., 2014a. Global Perspective on the
Oxidative Potential of Airborne Particulate Matter: A Synthesis of Research Findings. Environ.
Sci. Technol. 48, 7576–7583. doi:10.1021/es500937x
Saffari, A., Daher, N., Shafer, M.M., Schauer, J.J., Sioutas, C., 2013. Seasonal and spatial
variation of trace elements and metals in quasi-ultrafine (PM0.25) particles in the Los Angeles
metropolitan area and characterization of their sources. Environ. Pollut. 181, 14–23.
doi:10.1016/j.envpol.2013.06.001
Salameh, D., Detournay, A., Pey, J., Pé rez, N., Liguori, F., Saraga, D., Bove, M.C., Brotto, P.,
Cassola, F., Massabò , D., Latella, A., Pillon, S., Formenton, G., Patti, S., Armengaud, A., Piga,
D., Jaffrezo, J.L., Bartzis, J., Tolis, E., Prati, P., Querol, X., Wortham, H., Marchand, N., 2015.
PM2.5 chemical composition in five European Mediterranean cities: A 1-year study.
Atmospheric Res. 155, 102–117. doi:10.1016/j.atmosres.2014.12.001
Saleh, R., Khlystov, A., Shihadeh, A., 2012. Determination of Evaporation Coefficients of
Ambient and Laboratory-Generated Semivolatile Organic Aerosols from Phase Equilibration
Kinetics in a Thermodenuder. Aerosol Sci. Technol. 46, 22–30.
doi:10.1080/02786826.2011.602762
Sapkota, A., Chelikowsky, A.P., Nachman, K.E., Cohen, A.J., Ritz, B., 2010. Exposure to
particulate matter and adverse birth outcomes: a comprehensive review and meta-analysis. Air
Qual. Atmosphere Health 5, 369–381. doi:10.1007/s11869-010-0106-3
147 | P a g e

Sardar, S.B., Fine, P.M., Sioutas, C., 2005. Seasonal and spatial variability of the size-resolved
chemical composition of particulate matter (PM10) in the Los Angeles Basin. J. Geophys. Res.
Atmospheres 110, D07S08. doi:10.1029/2004JD004627
Savi, M., Kalberer, M., Lang, D., Ryser, M., Fierz, M., Gaschen, A., Rička, J., Geiser, M., 2008.
A Novel Exposure System for the Efficient and Controlled Deposition of Aerosol Particles onto
Cell Cultures. Environ. Sci. Technol. 42, 5667–5674. doi:10.1021/es703075q
Schauer, C., Niessner, R., Pö schl, U., 2003. Polycyclic Aromatic Hydrocarbons in Urban Air
Particulate Matter: Decadal and Seasonal Trends, Chemical Degradation, and Sampling
Artifacts. Environ. Sci. Technol. 37, 2861–2868. doi:10.1021/es034059s
Schauer, J.J., Lough, G.C., Shafer, M.M., Christensen, W.F., Arndt, M.F., DeMinter, J.T., Park,
J.-S., 2006. Characterization of metals emitted from motor vehicles. Res. Rep. Health Eff. Inst.
1–76; discussion 77–88.
Schroeder, W.H., Dobson, M., Kane, D.M., Johnson, N.D., 1987. Toxic Trace Elements
Associated with Airborne Particulate Matter: A Review. JAPCA 37, 1267–1285.
doi:10.1080/08940630.1987.10466321
Schwarz, J., Kaden, H., Pausch, G., 2000. Development of miniaturized potentiometric nitrate-
and ammonium selective electrodes for applications in water monitoring. Fresenius J. Anal.
Chem. 367, 396–398. doi:10.1007/s002160000367
See, S.W., Wang, Y.H., Balasubramanian, R., 2007. Contrasting reactive oxygen species and
transition metal concentrations in combustion aerosols. Environ. Res. 103, 317–324.
doi:10.1016/j.envres.2006.08.012
Seinfeld, J.H., Pandis, S.N., 2006. Atmospheric Chemistry and Physics, Second. ed. Wiley-Inter
Science.
Shafer, M.M., Perkins, D.A., Antkiewicz, D.S., Stone, E.A., Quraishi, T.A., Schauer, J.J., 2010.
Reactive oxygen species activity and chemical speciation of size-fractionated atmospheric
particulate matter from Lahore, Pakistan: an important role for transition metals. J. Environ.
Monit. 12, 704. doi:10.1039/b915008k
Sheesley, R.J., Schauer, J.J., Bean, E., Kenski, D., 2004. Trends in Secondary Organic Aerosol at
a Remote Site in Michigan’s Upper Peninsula. Environ. Sci. Technol. 38, 6491–6500.
doi:10.1021/es049104q
Shinyashiki, M., Eiguren-Fernandez, A., Schmitz, D.A., Di Stefano, E., Li, N., Linak, W.P., Cho,
S.-H., Froines, J.R., Cho, A.K., 2009. Electrophilic and redox properties of diesel exhaust
particles. Environ. Res. 109, 239–244. doi:10.1016/j.envres.2008.12.008
Shi, T., Schins, R.P.F., Knaapen, A.M., Kuhlbusch, T., Pitz, M., Heinrich, J., Borm, P.J.A., 2003.
Hydroxyl radical generation by electron paramagnetic resonance as a new method to monitor
ambient particulate matter composition. J. Environ. Monit. JEM 5, 550–556.
148 | P a g e

Sillanpä ä , M., Geller, M.D., Phuleria, H.C., Sioutas, C., 2008. High collection efficiency
electrostatic precipitator for in vitro cell exposure to concentrated ambient particulate matter
(PM). J. Aerosol Sci. 39, 335–347. doi:10.1016/j.jaerosci.2007.12.001
Sioutas, C., Kim, S., Chang, M., 1999. DEVELOPMENT AND EVALUATION OF A
PROTOTYPE ULTRAFINE PARTICLE CONCENTRATOR. J. Aerosol Sci. 30, 1001–1017.
doi:10.1016/S0021-8502(98)00769-1
Stone, E.A., Hedman, C.J., Sheesley, R.J., Shafer, M.M., Schauer, J.J., 2009. Investigating the
chemical nature of humic-like substances (HULIS) in North American atmospheric aerosols by
liquid chromatography tandem mass spectrometry. Atmos. Environ. 43, 4205–4213.
doi:10.1016/j.atmosenv.2009.05.030
Stookey, L.L., 1970. Ferrozine---a new spectrophotometric reagent for iron. Anal. Chem. 42,
779–781. doi:10.1021/ac60289a016
Sullivan, A.P., Weber, R.J., Clements, A.L., Turner, J.R., Bae, M.S., Schauer, J.J., 2004. A
method for on-line measurement of water-soluble organic carbon in ambient aerosol particles:
Results from an urban site. Geophys. Res. Lett. 31, L13105. doi:10.1029/2004GL019681
Takegawa, N., Miyazaki, Y., Kondo, Y., Komazaki, Y., Miyakawa, T., Jimenez, J.L., Jayne, J.T.,
Worsnop, D.R., Allan, J.D., Weber, R.J., 2005. Characterization of an Aerodyne Aerosol Mass
Spectrometer (AMS): Intercomparison with Other Aerosol Instruments. Aerosol Sci. Technol. 39,
760–770. doi:10.1080/02786820500243404
Thorpe, A., Harrison, R.M., 2008. Sources and properties of non-exhaust particulate matter from
road traffic: A review. Sci. Total Environ. 400, 270–282. doi:10.1016/j.scitotenv.2008.06.007
Turpin, B.J., Huntzicker, J.J., Hering, S.V., 1994. Investigation of organic aerosol sampling
artifacts in the los angeles basin. Atmos. Environ. 28, 3061–3071. doi:10.1016/1352-
2310(94)00133-6
Turpin, B.J., Lim, H.-J., 2001. Species contributions to PM2.5 mass concentrations : Revisiting
common assumptions for estimating organic mass. Aerosol Sci. Technol. 35, 602–610.
U.S. Environmental Protection Agency, 2012. U.S. EPA, Environmental Technology
Verification Report: Copper Environment Services LLC Xact 625 Particulate Metals Monitor.
USEPA, U.S.E.P.A., 1990. National Ambient Air Quality Standards (NAAQS).
Valavanidis, A., Fiotakis, K., Bakeas, E., Vlahogianni, T., 2005. Electron paramagnetic
resonance study of the generation of reactive oxygen species catalysed by transition metals and
quinoid redox cycling by inhalable ambient particulate matter. Redox Rep. 10, 37–51.
doi:10.1179/135100005X21606
Valavanidis, A., Fiotakis, K., Vlachogianni, T., 2008. Airborne particulate matter and human
health: toxicological assessment and importance of size and composition of particles for
oxidative damage and carcinogenic mechanisms. J. Environ. Sci. Health Part C Environ.
Carcinog. Ecotoxicol. Rev. 26, 339–362. doi:10.1080/10590500802494538
149 | P a g e

Valko, M., Morris, H., Cronin, M.T.D., 2005. Metals, Toxicity and Oxidative Stress. Curr. Med.
Chem. 12, 1161–1208. doi:10.2174/0929867053764635
Verma, V., Ning, Z., Cho, A.K., Schauer, J.J., Shafer, M.M., Sioutas, C., 2009a. Redox activity
of urban quasi-ultrafine particles from primary and secondary sources. Atmos. Environ. 43,
6360–6368. doi:10.1016/j.atmosenv.2009.09.019
Verma, V., Polidori, A., Schauer, J.J., Shafer, M.M., Cassee, F.R., Sioutas, C., 2009b.
Physicochemical and Toxicological Profiles of Particulate Matter in Los Angeles during the
October 2007 Southern California Wildfires. Environ. Sci. Technol. 43, 954–960.
doi:10.1021/es8021667
Verma, V., Rico-Martinez, R., Kotra, N., King, L., Liu, J., Snell, T.W., Weber, R.J., 2012.
Contribution of Water-Soluble and Insoluble Components and Their Hydrophobic/Hydrophilic
Subfractions to the Reactive Oxygen Species-Generating Potential of Fine Ambient Aerosols.
Environ. Sci. Technol. 46, 11384–11392. doi:10.1021/es302484r
Verma, V., Shafer, M.M., Schauer, J.J., Sioutas, C., 2010. Contribution of transition metals in
the reactive oxygen species activity of PM emissions from retrofitted heavy-duty vehicles.
Atmos. Environ. 44, 5165–5173. doi:10.1016/j.atmosenv.2010.08.052
Volckens, J., Dailey, L., Walters, G., Devlin, R.B., 2009. Direct Particle-to-Cell Deposition of
Coarse Ambient Particulate Matter Increases the Production of Inflammatory Mediators from
Cultured Human Airway Epithelial Cells. Environ. Sci. Technol. 43, 4595–4599.
doi:10.1021/es900698a
Wang, D., Kam, W., Cheung, K., Pakbin, P., Sioutas, C., 2012. Development of a Two-Stage
Virtual Impactor System for High Concentration Enrichment of Ultrafine, PM2.5, and Coarse
Particulate Matter. Aerosol Sci. Technol. 47, 231–238. doi:10.1080/02786826.2012.744446
Wang, D., Pakbin, P., Saffari, A., Shafer, M.M., Schauer, J.J., Sioutas, C., 2013a. Development
and Evaluation of a High-Volume Aerosol-into-Liquid Collector for Fine and Ultrafine
Particulate Matter. Aerosol Sci. Technol. 47, 1226–1238. doi:10.1080/02786826.2013.830693
Wang, D., Pakbin, P., Shafer, M.M., Antkiewicz, D., Schauer, J.J., Sioutas, C., 2013b.
Macrophage reactive oxygen species activity of water-soluble and water-insoluble fractions of
ambient coarse, PM2.5 and ultrafine particulate matter (PM) in Los Angeles. Atmos. Environ. 77,
301–310. doi:10.1016/j.atmosenv.2013.05.031
Wang, D., Shafer, M.M., Schauer, J.J., Sioutas, C., 2014. Development of a Technology for
Online Measurement of Total and Water-Soluble Copper (Cu) in PM2.5. Aerosol Sci. Technol.
48, 864–874. doi:10.1080/02786826.2014.937478
Weber, R.J., Orsini, D., Daun, Y., Lee, Y.-N., Klotz, P.J., Brechtel, F., 2001. A Particle-into-
Liquid Collector for Rapid Measurement of Aerosol Bulk Chemical Composition. Aerosol Sci.
Technol. 35, 718–727. doi:10.1080/02786820152546761
150 | P a g e

Wilhelm, M., Ritz, B., 2005. Local Variations in CO and Particulate Air Pollution and Adverse
Birth Outcomes in Los Angeles County, California, USA. Environ. Health Perspect. 113, 1212–
1221.
Willeke, K., Lin, X., Grinshpun, S.A., 1998. Improved Aerosol Collection by Combined
Impaction and Centrifugal Motion. Aerosol Sci. Technol. 28, 439–456.
doi:10.1080/02786829808965536
Xia, T., Korge, P., Weiss, J.N., Li, N., Venkatesen, M.I., Sioutas, C., Nel, A., 2004. Quinones
and aromatic chemical compounds in particulate matter induce mitochondrial dysfunction:
implications for ultrafine particle toxicity. Environ. Health Perspect. 112, 1347–1358.
Xia, T., Kovochich, M., Brant, J., Hotze, M., Sempf, J., Oberley, T., Sioutas, C., Yeh, J.I.,
Wiesner, M.R., Nel, A.E., 2006. Comparison of the Abilities of Ambient and Manufactured
Nanoparticles To Induce Cellular Toxicity According to an Oxidative Stress Paradigm. Nano
Lett. 6, 1794–1807. doi:10.1021/nl061025k
Zhang, X., Liu, J., Parker, E.T., Hayes, P.L., Jimenez, J.L., de Gouw, J.A., Flynn, J.H.,
Grossberg, N., Lefer, B.L., Weber, R.J., 2012. On the gas-particle partitioning of soluble organic
aerosol in two urban atmospheres with contrasting emissions: 1. Bulk water-soluble organic
carbon. J. Geophys. Res. Atmospheres 117, D00V16. doi:10.1029/2012JD017908
Zhang, Y., Schauer, J.J., Shafer, M.M., Hannigan, M.P., Dutton, S.J., 2008. Source
Apportionment of in Vitro Reactive Oxygen Species Bioassay Activity from Atmospheric
Particulate Matter. Environ. Sci. Technol. 42, 7502–7509. doi:10.1021/es800126y
Zhao, Y., Bein, K.J., Wexler, A.S., Misra, C., Fine, P.M., Sioutas, C., 2005. Field evaluation of
the versatile aerosol concentration enrichment system (VACES) particle concentrator coupled to
the rapid single-particle mass spectrometer (RSMS-3). J. Geophys. Res. Atmospheres 110,
D07S02. doi:10.1029/2004JD004644
Zheng, L.-F., Wei, Q.-Y., Cai, Y.-J., Fang, J.-G., Zhou, B., Yang, L., Liu, Z.-L., 2006. DNA
damage induced by resveratrol and its synthetic analogues in the presence of Cu (II) ions:
Mechanism and structure-activity relationship. Free Radic. Biol. Med. 41, 1807–1816.
doi:10.1016/j.freeradbiomed.2006.09.007
Zielinski, H., Mudway, I.S., Bé rubé , K.A., Murphy, S., Richards, R., Kelly, F.J., 1999. Modeling
the interactions of particulates with epithelial lining fluid antioxidants. Am. J. Physiol. - Lung
Cell. Mol. Physiol. 277, L719–L726.

Abstract (if available)

Abstract There has been a large body of epidemiological and toxicological studies indicating the strong associations between exposures to ambient particulate matter (PM) and adverse health outcomes. Some specific chemical component in ambient PM, including but not restricted to airborne transition metals, has been hypothesized to be mostly responsible for generating excess cellular oxidation stress and eventually result in PM induced health risks. Due to the complex physical property and chemical composition of ambient PM, assessing which PM constituents are linked to adverse health outcomes as well as the exact mechanisms leading to these outcomes remains an active topic of research. Therefore, novel technologies for characterizing ambient PM with high time resolutions are in great needs, which will assist in investigations of the physical properties and chemical composition of PM, as well as enabling a better understanding of toxicological properties of ambient PM. ❧ This dissertation focuses on the development and evaluations on novel techniques in determining the physical, chemical and toxicological properties of ambient PM. As the first part of this dissertation, a two-stage particle concentration enrichment system was developed to provide highly concentrated particles at low flow rate (i.e. 1.5 L/min). This system can enrich particle concentration by a factor roughly of 100-120 without altering their physical and chemical properties. Secondly, as a principle investigation of the new PM collection technology, the relative contributions of water-soluble and water-insoluble portions of ambient PM to cellular redox activity were investigated. Results from this study indicated that both water-soluble and water-insoluble portions of PM played important roles in influencing potential cellular toxicity. Next, a novel Aerosol-Into-Liquid Collector was developed to provide concentrated slurries of fine and/or ultrafine PM, in which both water-soluble and water-insoluble components were well preserved in the collected slurry samples. This new aerosol collection system could achieve an excellent collection efficiency (over 90%), and has the unique ability to be continuously operated unattended for at least 4 to 5 days without any obvious shortcomings in its operation. Following the successfully development of Aerosol-Into-Liquid Collector, this new PM sampler was further developed into a novel monitor for online, in-situ measurement of copper (Cu) in ambient fine PM. Evaluations of the Cu monitor indicates a very good agreement for total and water-soluble Cu concentrations obtained online by this monitor, with measurements performed by inductively coupled plasma mass spectrometry (ICP-MS) as a reference method, suggesting the excellent performance of this Cu monitor in aspect of collection efficiency and measurement accuracy. This technology is then extended to coarse PM by utilizing two virtual impactors combined with a modified liquid impinger (BioSampler) as PM collector. Lastly, a prototype atmospheric aerosol monitor was developed and evaluated for online measurement of other three toxicologically relevant redox-active metals (Fe, Mn, and Cr) in ambient PM2.5 based on the developed instrument described in the previous parts of this dissertation. ❧ The novel techniques developed in this dissertation will greatly advance the capabilities of atmospheric pollution monitoring to understand the physical properties, as well as the chemical and toxicological active components of ambient PM. Moreover, such technologies will provide significant insights on developing a better understanding of the sources, formation mechanisms, and transport of PM in the atmosphere. Ultimately, air pollution monitoring goals can be more directly linked with the protection of public health. More effective and targeted control strategies to better protect human health can thus be implemented.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Size-resolved particulate matter (PM) in urban areas: toxico-chemical characteristics, sources, trends and health implications

PDF

Physico-chemical characteristics and sources of ambient PM mass and number concentrations and their associated toxicity, and development of novel techniques for high time-resolution measurement o...

PDF

Impact of urban source emissions on ambient particulate matter (PM) composition and associated toxicity in local and regional scales using source appointment models

PDF

Investigating the temporal trends, sources, and toxicity of ambient particulate matter (PM) in different metropolitan environments, and development of a novel aerosol generation setup for inhalat...

PDF

Assessing indoor and outdoor air quality by characterizing the physicochemical and toxicological properties of particulate matter using…

$Chemical and toxicological characteristics and historical trends of size-fractioned particulate matter from traffic-related emissions in Los Angeles$

PDF

Chemical and toxicological characteristics and historical trends of size-fractioned particulate matter from traffic-related emissions in Los Angeles

PDF

Influence of atmospheric processes on the physico-chemical properties of primary particles and their impact on public exposure

PDF

Oxidative potential of urban atmospheric particles: spatiotemporal trends and associations with source-specific chemical components

PDF

Measurement and methods of assessing the impact of prevalent particulate matter sources on air quality in southern California

PDF

Toxicity of urban particulate matter: long-term health risks, influences of surrounding geography, and diurnal variation in chemical composition and the cellular oxidative stress response

PDF

Temporal, spatial and toxicological characteristics of coarse particulate matter in an urban area and relation to sources and regulations

PDF

Characterization of black carbon: from source to evolution of physical and optical properties in the atmosphere

PDF

Investigation of physico-chemical characteristics of particulate matter from vehicular sources

PDF

Toxicological characteristics of particulate matter in an urban environment and their linkage to the source-specific constituents

$Physico-chemical properties and source apportionment of size-fractionated airborne particulate matter in urban areas with implications for public health$

PDF

Physico-chemical properties and source apportionment of size-fractionated airborne particulate matter in urban areas with implications for public health

$Exposure assessment and source apportionment of size fractions of airborne particulate matter$

PDF

Exposure assessment and source apportionment of size fractions of airborne particulate matter

PDF

Analysis of sources and profiles of organic carbon in ambient particulate matter across fine and coarse sizes and introducing an optical technique for real-time urban dust measurement

PDF

Particulate matter (PM) exposure for commuters in Los Angeles: chemical characterization and implications to public health

PDF

Investigating the role of urban emission sources on redox-active PM compounds and the chemical analysis of the standardized diesel exhaust particles

PDF

Chemical and toxicological characteristics of particulate matter in urban environments with a focus on its sources, associated health impacts and mitigation policies

Asset Metadata

Creator Wang, Dongbin (author)

Core Title Development of novel techniques for evaluating physical, chemical and toxicological properties of particulate matter in ambient air

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Environmental Engineering

Publication Date 06/14/2016

Defense Date 05/10/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aerosol-into-liquid collector,Metals,OAI-PMH Harvest,online measurements,particulate matter

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Sioutas, Constantinos (committee chair), Finch, Caleb (committee member), Henry, Ronald (committee member)

Creator Email dongbinw@usc.edu,wangdongbin1987@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-251116

Unique identifier UC11280531

Identifier etd-WangDongbi-4428.pdf (filename),usctheses-c40-251116 (legacy record id)

Legacy Identifier etd-WangDongbi-4428.pdf

Dmrecord 251116

Document Type Dissertation

Format application/pdf (imt)

Rights Wang, Dongbin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA