 

 

 

 

 

 

 

 

 

 

. . s.

I
331K.-

fauna-bu.

ﬂan.
Q.
~

nu ‘ 66‘s..“

., .
.23?me
:ﬁ 8.. v.3
.mfnagﬁ

in s
3.: 1:1.

I‘F‘.~ ‘Iou
. vsﬂiszdun ll.
, ‘5 int-‘15.

2:73

 

 

11.35313 . LIBRARY

MICI‘tIN‘“ State

2030' I University

 

This is to certify that the
dissertation entitled

SAMPLING AND MASS SPECTROMETRY APPROACHES
FOR THE DETECTION OF DRUGS AND FOREIGN
CONTAMINANTS IN BREATH FOR HOMELAND SECURITY
APPLICATIONS

presented by

Audrey Noreen Martin

has been accepted towards fulﬁllment
of the requirements for the

Doctoral degree in Chemistry

 

 

 

 

 

MSU is an Afﬁnnative Action/Equal Opportunity Employer

.o-.----.-_- - _ _ _.

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K./Prolecc&Pres/ClRC/DaleOue indd

 

-a-‘u-

SAMPLING AND MASS SPECTROMETRY APPROACHES FOR THE
DETECTION OF DRUGS AND FOREIGN CONTAMINANTS IN BREATH FOR
HOMELAND SECURITY APPLICATIONS
By

Audrey Noreen Martin

A DISSERTATION
Submitted to
Michigan State University
in partial ﬁIlﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemistry

2009

 

ABSTRACT
SAMPLING AND MASS SPECTROMETRY APPROACHES FOR THE DETECTION
OF DRUGS AND FOREIGN CONTAMINANTS IN BREATH FOR HOMELAND
SECURITY APPLICATIONS
By
Audrey Noreen Martin
Homeland security relies heavily on analytical chemistry to identify suspicious materials
and persons. Traditionally this role has focused on attribution, determining the type and
origin of an explosive, for example. But as technology advances, analytical chemistry
can and will play an important role in the prevention and preemption of terrorist attacks.
More sensitive and selective detection techniques can allow suspicious materials and
persons to be identiﬁed even before a ﬁnal destructive product is made. The work

presented herein focuses on the use of commercial and novel detection techniques for

application to the prevention of terrorist activities.

Although drugs are not commonly thought of when discussing terrorism, narcoterrorism
has become a signiﬁcant threat in the let century. The role of the drug trade in the
funding of terrorist groups is prevalent; thus, reducing the trafficking of illegal drugs can
play a role in the prevention of terrorism by cutting off much needed ﬁinding. To do so,
sensitive, speciﬁc, and robust analytical equipment is needed to quickly identify a
suspected drug sample no matter what matrix it is in. Single Particle Aerosol Mass
SPeCtTOInetry (SPAMS) is a novel technique that has previously been applied to
biological and chemical detection. The current work applies SPAMS to drug analysis,
identifying the active ingredients in single component, multi-component, and multi-tablet

drug samples in a relatively non-destructive manner. In order to do so, a sampling

 

 

apparatus was created to allow particle generation from drug tablets with on-line
introduction to the SPAMS instrument. Rules trees were developed to automate the

identiﬁcation of drug samples on a single particle basis.

A novel analytical scheme was also developed to identify suspect individuals based on
chemical signatures in human breath. Human breath was sampled using an RTubeTM and
the trace volatile organic compounds (VOCS) were preconcentrated using solid phase
microextraction (SPME) and identiﬁed using gas chromatography — mass spectrometry
(GC-MS). Modiﬁcations to the sampling apparatus allowed for increased VOC
collection efﬁciency, and reduced the time of sampling and analysis by over 25%. The
VOCS are present in breath due to either endogenous production, or exposure to an
external source through absorption, inhalation, or ingestion. Detection of these
exogenous chemicals can provide information on the prior location and activities of the
subject. Breath samples collected before and after exposure in a hardware store and nail
salon were analyzed to investigate the prior location of a subject; breath samples
collected before and after oral exposure to terpenes and terpenoid compounds,
pseudoephedrine, and inhalation exposure to hexamine and other explosive related
compounds were analyzed to investigate the prior activity of a subject. The elimination
of such compounds from the body was also monitored. In application, this technique
may provide an early warning system to identify persons of interest in the prevention and

preemption stages of homeland security.

 

 

ACKNOWLEDGEMENTS

As my road to the point of writing this dissertation was unusual, there are many people

that deserve my deepest thanks for their help along the way.

1 must start by thanking my research advisor, Dan Jones. When you received a call from
a stranger out of the blue that I wanted to talk to you about joining your research group
and promptly moving to California to do my work at LLNL, you didn’t run the opposite
direction. That belief ﬁom day one that we could make this arrangement work was, and
is, much appreciated. Your going outside the box for me has allowed me to become what

I am today, and I will always be forever grateful for that. Thank you.

To George Farquar and Matthias Frank, my other research advisors — your support and
efforts made it possible for me to return to, and stay aﬂoat at, LLNL. Without you both,
this research wouldn’t exist, so thank you. Thanks especially to George, you made my
initial summer at LLNL enjoyable enough that I wanted to return, provided valuable
mentorship for the past 2 years, and didn’t make my life miserable as I got closer and

closer to graduation.

Thank you to my dear friends, the Musketeers that are always there to cheer me up when
I’m upset, to make me laugh, to deep-fry anything we can lift, and to remind me that
there IS, in fact, a life outside the lab worth enjoying. Thanks to my St. Louis ladies who
have been supportive of my being in school for the last two decades of my life, and have

provided me with many words of wisdom to help me hang in there.

iv

Thank you to my family — you’re always there to silently support my decisions, and have
helped me with some of the uphill battles I have faced over the last few years. Dad,
you’ve always encouraged me to be the best I can be, and helped me to determine what
that is. Mom, you’re always supportive and interested in whatever I’m doing, and that
keeps me going when I feel like nothing is going right. Alan, Ciara and Aaron, you’ve
inspired me to keep on trucking through grad school, even if it is by way of teasing me
for not being a ‘real’ doctor. John and Andi, you’ve helped me realize what the priorities

in my life should be, each in your own way. . . and thank you for NOT being doctors.

To Devan — who would have known that a quick summer internship in California would
lead me to you? For the past few years you’ve been my best friend, my voice of reason,
you’ve supported me, written MatLAB scripts for me, kept Southwest Airlines in
business with me, listened to me as I talked your ear off on days when I didn’t encounter

any people at work, and shown me how happy I can be. Thank you for being you.

Finally, I would like to acknowledge the funding sources that made my research possible.
Part of this work was performed as a Department of Homeland Security (DHS) Fellow
through a program administered by the Oak Ridge Institute for Science and Education
(ORlSE). The development of the SPAMS system at was ﬁmded through DARPA and
TSWG in the Department of Defense, as well as through internal ﬁmding at LLNL. This
work was performed under the auspices of the US. Department of Energy by Lawrence
Livermore National Laboratory in part under Contract W-7405-Eng-48 and in part under

Contract DE-AC52-O7NA27344. LLNL—TH-410574.

TABLE OF CONTENTS

 

 

 

 

LIST OF TABLES ix
LIST OF FIGURES _ xi
LIST OF EQUATIONS xxx
CHAPTER 1: Introduction ........ - -- -- 1
1-1 : Motivations .......................................................................................................... 1

1-2: Overview .............................................................................................................. 2
1-3: References ............................................................................................................ 4
CHAPTER 2: Single Particle Aerosol Mass Spectrometry of Drug Samples ............. 5
2-1: Motivations and Introduction ................................................................................ 5
2-2: Background on Drug Analysis .............................................................................. 6
2-3: Background of SPAMS ........................................................................................ 9
2-4: SPAMS Description ........................................................................................... 12
2-5: Laser Optics ....................................................................................................... 16
2-6: Solid Sampling Materials .................................................................................... 17
2-7: Data Collection and Analysis .............................................................................. 18
2-8: Drug Material Preparation .................................................................................. 21
2-9: Drug Analysis Using SPAMS ............................................................................. 22
2-9-1: BayerTM Aspirin .......................................................................................... 22
2-9-2: Chloroquine Diphosphate ............................................................................ 25
2-9-3: Clemastine Fumarate ................................................................................... 28
2-9-4: BenadrleM (Diphenhydramine) ................................................................... 29
2-9-5: Loratadine ................................................................................................... 30
2-9-6: Generic Ibuprofen ........................................................................................ 31
2—9-7: Generic Pseudoephedrine and Phenylephrine ............................................... 33
2-9-8: BayerTM Aspirin versus CertiﬁedTM Aspirin ................................................. 35
2-9-9: Tylenol Extra StrengthTM (Acetaminophen) ................................................. 37
2-9-10: Tylenol ColdTM (Acetaminophen + Diphenhydramine) .............................. 39
2-9-11: Tylenol PMTM - Caplet versus Tablet (Acetaminophen + Diphenhydramine)
.............................................................................................................................. 41
2-9-12: Advil Cold and SinusTM (Ibuprofen + Pseudoephedrine) ............................ 43
2-9-13: Tylenol Severe CongestionTM (Acetaminophen + Dextromethorphan +
Guaifenesin + Pseudoephedrine) ............................................................................ 43
2-9-14: Multiple Tablets: Phenylephrine and Aspirin ............................................. 47
2-9-15: Multiple Tablets - Separate Vials: Phenylephrine and Aspirin ................... 48
2-9-16: Multiple Tablets: Phenylephrine, Aspirin, Ibuprofen, Loratadine,
Pseudoephedrine .................................................................................................... 49
2-9-17: Emptied Vial ............................................................................................. 51
2-9-18: In Situ Sampling: Chloroquine Diphosphate and Loratadine ...................... 52

vi

2-10: Conclusions ...................................................................................................... 54

 

2-1 1: References ........................................................................................................ 56
CHAPTER 3: Breath Analysis — Method Developments -- _ ......... -61
3-1: Motivations and Introduction .............................................................................. 61
3-2: Background on Breath Analysis .......................................................................... 62
3-3: Breath Collection Methods ................................................................................. 65
3-4: Volatile and Non-Volatile Components .............................................................. 67
3-5: Breath Analysis Techniques ................................................................................ 67
3-6: Issues in Breath Analysis .................................................................................... 69
3-6-1 : Preconcentration .......................................................................................... 69
3-6-2: Contamination ............................................................................................. 71
3-7: Background of Gas Chromatography-Mass Spectrometry ................................... 73
3-7—1: Gas Chromatography Theory ....................................................................... 73
3-7-2: Mass Spectrometry Theory .......................................................................... 76
3-7-3: GC-MS Description ..................................................................................... 80
3-8: Instrument Description ....................................................................................... 82
3-8-1: Agilent System ............................................................................................ 82
3-8-2: Shimadzu System ........................................................................................ 83
3-9: Materials ............................................................................................................ 83
3-10: Subjects ............................................................................................................ 84
3-11: Original Sample Collection (EBC) .................................................................... 84
3-12: GC-MS Method Development .......................................................................... 85
3-12-1: GC Method Optimization .......................................................................... 85
3-12-2: SPME Exposure Time Optimization .......................................................... 88
3-13: Comparison of Breath to Background Air ......................................................... 89
3-14: Reproducibility Studies ..................................................................................... 92
3-14-1: Repeatability of Measurement ................................................................... 92
3-14-2: Single Subject, Single Day ......................................................................... 92
3-14-3: Single Subject, Multiple Days .................................................................... 94
3-15: Optimization of Breath Sample Collection ........................................................ 95
3-15-1: Exhaled Breath Condensate (BBC) and Exhaled Breath Vapor at -80°C
(EBV-80°C) Collection .......................................................................................... 95
3-15-2: Exhaled Breath Vapor — Room Temperature (EBV-RT) Collection ........... 96
3-15-3: Comparison of Collection Methods ............................................................ 97
3-16: Efforts to Control the Inhaled Air of a Subject ................................................ 100
3-16-1: RTubeTM Modiﬁcation with PVC Tubing ................................................ 100
3-16-2: Nail Salon Inhalation with PVC Tubing ................................................... 101
3-16-3: RTubeTM Modiﬁcation with Respirator Cartridge .................................... 103
3-16—4: Nail Polish Inhalation with Respirator Cartridge ...................................... 104
3-16-5: Lemonade Inhalation with Respirator Cartridge ....................................... 106
3-16-6: Gasoline Inhalation with Respirator Cartridge and Synthetic Lung Substitute
Development ........................................................................................................ 1 07
3-17: Conclusions .................................................................................................... 110
3-18: References ...................................................................................................... l 12

vii

CHAPTER 4: Measuring Chemical Exposure using Breath Analysis .................... 120

 

4-1: Motivations and Introduction ............................................................................ 120
4-2: Background ...................................................................................................... 121
4-2-1: Industrial/Commercial Exposure ................................................................ 123
4-2-2: Exposure via Ingestion — Terpenes and Terpenoid Compounds .................. 124
4-2-3: Research Directions ................................................................................... 125
4-3: Materials and Methods ..................................................................................... 126
4-3-1 : Materials ................................................................................................... 126
4-3-2: Methods .................................................................................................... 127
4-4: Subjects ............................................................................................................ 128
4-5: Nail Salon Study ............................................................................................... 128
4-5-1: Headspace Analysis of Manicured Hands .................................................. 133
4-6: Hardware Store Study ....................................................................................... 134
4-7: Lemonade Study ............................................................................................... 135
4-7-1: Reproducibility of Single Subject Drinking Original Lemonade ................ 139
4-7-2: Concentration Dependence of Terpene Signal in Breath ............................ 144
4-8: Drug Studies ..................................................................................................... 151
4-8-1: Sudafed ..................................................................................................... 151
4-9: Explosive Studies ............................................................................................. 154
4-9-1 : Hexarnine .................................................................................................. 154
4-9-2: Field Studies .............................................................................................. 158
4-10: Conclusions .................................................................................................... 159
4-1 1 : References ...................................................................................................... 161
CHAPTER 5: Conclusions and Thoughts for the Future 164
5-1: Conclusions ...................................................................................................... 164
5-1-1: Single Particle Aerosol Mass Spectrometry ............................................... 164
5—1-2: Breath Analysis ......................................................................................... 165
5-2: Thoughts for the Future .................................................................................... 166
5-2-1: Single Particle Aerosol Mass Spectrometry ............................................... 166
5-2-2: Breath Analysis ......................................................................................... 168
5-3: References ........................................................................................................ 1 75

APPENDIX A — Additional Rules Trees for Drug Identiﬁcation Using SPAMS....176

APPENDIX B - Additional Terpene and Terpenoid Data from Lemonade
Consumption 181

viii

LIST OF TABLES

Table 2- 1: Structures and molecular weights of drug compounds studied. ..................... 22

Table 3- 1: Description of method parameters for methods tested to optimize the
separation of breath samples. ......................................................................................... 86

Table 3- 2: Compounds and corresponding retention times tentatively identiﬁed in a
SPME sample of room air. ............................................................................................. 91

Table 3- 3: MarkerLynx method parameters for peak area extraction ............................. 93

Table 3- 4: Breath VOCS and corresponding retention times tentatively identiﬁed in a
typical EBV-RT sample (no known exposure). .............................................................. 99

Table 3- 5: Peak area of several compounds tentatively identiﬁed in total ion current
chromatograms of breath samples collected as EBV-80°C or EBV-RT versus peak area of
BBC total ion current chromatograms. These peaks are indicated by the arrows in Figure
3- 9. ............................................................................................................................. 100

Table 3- 6: Reduction in peak area for several ingredients of nail polish detected in
human breath obtained with respirator or PVC tubing. Peak areas were calculated by
integration of the XIC for each compound, and are displayed relative to the XIC peak
area value obtained without modifying the inhaled air source. ..................................... 105

Table 3- 7: Reduction in peak area for the monoterpenes detected in EBV-RT samples
obtained over lemonade. Peak areas were calculated by integration of the XIC for each
compound, and are displayed relative to the peak area value obtained without modifying
the inhaled air source. .................................................................................................. 107

Table 3- 8: Reduction in peak area for the tentatively identiﬁed compounds detected in
EBV-RT samples generated with a synthetic lung substitute respirated over gasoline.
Peak areas were calculated by integration of the TIC for each compound, and are
displayed relative to the peak area value obtained without modifying the inhaled air
source. The additional peak area reduction demonstrated by the use of two respirators
was calculated compared to those values when using one respirator to demonstrate the
additional beneﬁt of the respirator series. ..................................................................... 109

Table 4- 1: Molecular weights, structures, and GC retention times (RT) of 13 terpenes
and terpenoid compounds studied. Average decay values, tau, for a single subject (n=5)
after consumption of original lemonade, modeling compound decay in breath to a single
phase decay are also shown. ........................................................................................ 138

Table 4- 2: Half-lives (tau) for single exponential decay of 13 terpenes and terpenoid
compounds in breath aﬁer original lemonade consumption by 3 subjects. .................... 143

 

Table 4- 3: Coefﬁcients and half-lives obtained after ﬁtting the uptake and decay of a-
limonene in breath after consumption of Mediterranean lemonade to Equation 4- 1 and

Equation 4- 2. .............................................................................................................. 150

Table B- 1: Fit Parameters for the terpenes and terpenoid compounds monitored in

breath. The data after lemonade consumption were ﬁt according to Equation 4-1 ........ 240
x

LIST OF FIGURES

Figure 2- 1: Schematic of the Single Particle Aerosol Mass Spectrometer. Particles are
generated using the modiﬁed glass vial and transferred to the SPAMS via conductive
tubing. After passing through a pressure ﬂow reducer with a relaxation chamber and
aerodynamic focusing lens, the pressure is lowered and the beam is focused into a linear
beam. Particles are then tracked, where the aerodynamic velocity and size are
determined. Laser induced ﬂuorescence at two wavelengths (266 and 355 nm) as well as
particle charge is then determined before the particle enters the source region of the mass
spectrometer. The particle is desorbed and ionized by an Nd:YAG laser operating at 266
nm, and the positive and negative ions created are concurrently accelerated in linearly
opposing directions into two reﬂectron time-of-ﬂight mass analyzers, where two mass
spectra are recorded for each particle. ............................................................................ 12

Figure 2- 2: Diagram of the desorption/ionization laser optics. A Nd:YAG laser is
ﬁequency quadrupled to generate a 266 nm beam which is then passed through an
aperture, adjustable half-wave plate and vertical polarizer before being imaged by a
plano-convex lens with a 10 cm focus length onto two planar mirrors which direct the
image into the source region of the mass spectrometer. .................................................. 16

Figure 2- 3: Diagram of modiﬁed glass vial used to generate particles from solid samples
for analysis by SPAMS. The plastic cap was modiﬁed using stainless steel tubing to
create a port to transfer the sample to the SPAMS as well as a second port to which a
HEPA ﬁlter was attached. The solid sample is placed in the vial which is then shaken or
vortexed; particles are generated by the collisions induced by the shaking. .................... 17

Figure 2- 4: Rules tree for classiﬁcation of ibuprofen. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘1’ indicate a positive identiﬁcation for ibuprofen;
circles labeled ‘Null’ indicate a negative response. M/z 206 corresponds to the [M]+ ion,
m/z -205 and -206 correspond to the [M]'and [M-H]' ion, m/z 161 corresponds to the [M-
COOl-l]+ ion, and m/z -161 corresponds to the [M-COOH]' ion ...................................... 20

Figure 2- 5: (a) Average single particle aerosol mass spectra of 999 aspirin particles, and
(b) ﬁve consecutively obtained aspirin particles (aerodynamic diameter displayed in
upper right of each spectrum). The single particle spectra contain all peaks used to
identify aspirin. The spectra from these particles were obtained in less than ﬁve seconds.
...................................................................................................................................... 23

Figure 2- 6: a) average dual-polarity mass spectrum of 1000 particles dislodged ﬁom a
single chloroquine phosphate tablet, b) mass spectra obtained from 5 individual particles

of chloroquine phosphate dislodged from the tablet. Particle size is notated. ................. 26

Figure 2- 7: Average mass spectrum of a single particle dislodged from a Clemastine

tablet (aerodynamic diameter 2.09 pm) .......................................................................... 28
xi

 

Figure 2- 8: Single particle aerosol mass spectrum of an individual diphenhydramine
particle (aerodynamic diameter 1.14 pm). ...................................................................... 29

Figure 2- 9: Single particle aerosol mass spectrum of an individual loratadine particle
(aerodynamic diameter 3.95 pm). Inset shows high mass peaks due to loratadine dimers.
...................................................................................................................................... 31

Figure 2- 10: Single particle aerosol mass spectrum of a randomly selected individual
ibuprofen particle (aerodynamic diameter 1.21 pm). Note the presence of the [M]+ and
[M-H]’ peaks to allow for simple identiﬁcation .............................................................. 32

Figure 2- 11: (a) Single particle aerosol mass spectrum of an individual pseudoephedrine
particle (aerodynamic diameter 1.08 pm) and (b) an individual phenylephrine particle
(aerodynamic diameter 0.96 pm). Although similar in structure and mass, these two
compounds can be differentiated using SPAMS based on molecular and fragment ion
species. .......................................................................................................................... 34

Figure 2- 12: Average mass spectra of 1000 particles of BayerTM brand aspirin (a,c,e) and
CertiﬁedTM brand aspirin (b,d,f) in three separate mass regions. Note the presence of
additional peaks in the CertiﬁedTM Brand aspirin, attributed to ﬁllers and impurities in the
sample that can be used to distinguish the two brands. ................................................... 36

Figure 2- 13: Average mass Spectrum from 50 particles dislodged from a Tylenol EX
tablet. The 50 particles shown were selected by sorting 799 particles for the highest peak
area at m/z 152. ............................................................................................................. 38

Figure 2- 14: Average mass spectra of particles dislodged ﬁ'om a.) a Benadryl tablet
containing diphenhydramine (n=999), b.) a Tylenol Cold tablet containing
acetaminophen and diphenhydramine (n=999), and c.) a Tylenol EX tablet containing
acetaminophen (n=25). Dashed lines indicate peaks in the Tylenol Cold sample that can
be attributed to either acetaminophen or diphenhydramine ............................................. 40

Figure 2- 15: Average mass spectra obtained from 997 particles dislodged from a.) a
Tylenol PM caplet and b.) a Tylenol PM tablet. While many peaks indicative of the
active ingredients, acetaminophen and diphenhydramine, are seen, other peaks are also
present in either sample, allowing differentiation between the two formulations. ........... 41

Figure 2- 16: Average mass spectrum obtained from 50 particles dislodged from an Advil
Cold and Sinus tablet. The 50 particles were selected by sorting 797 particles for the
highest mass area at m/z 166 .......................................................................................... 43

Figure 2- 17: a) Average mass spectrum of 1000 particles dislodged from a single
Tylenol Severe Congestion tablet containing pseudoephedrine, acetaminophen,
dextromethorphan, and guaifenesin as the active ingredients, b-c) mass spectra obtained
ﬁ'om a single particle dislodged from a tablet containing only pseudoephedrine or
acetaminophen, respectively, as the active ingredient, d-e) concatenated standard NIST
mass spectra of dextromethorphan and guaifenesin, respectively. Single component
tablets containing these ingredients were not readily available. ...................................... 45

xii

Figure 2- 18: Average mass spectrum obtained from a.) 249 particles dislodged from a
phenylephrine tablet, b.) 798 particles dislodged from a single phenylephrine tablet and a
single aspirin tablet placed in the same glass vial and sampled simultaneously, c.) 249
particles dislodged from a single aspirin tablet. Note that both phenylephrine and aspirin
peaks are seen in the mixed tablet spectrum. .................................................................. 47

Figure 2- 19: a) average mass spectrum obtained from 799 particles introduced ﬁom a
complex sample containing 5 single active ingredient tablets including aspirin, ibuprofen,
loratadine, phenylephrine, and pseudoephedrine, b-t) single particles selected from the
799 particles collected that were identiﬁed as containing primarily pseudoephedrine,
ibuprofen, loratadine, phenylephrine, or aspirin, respectively. ........................................ 50

Figure 2- 20: a) average mass spectrum of 500 particles introduced ﬁom the complex
sample containing 5 single active ingredient tablets including aspirin, ibuprofen,
loratadine, phenylephrine, and pseudoephedrine, b) average mass spectrum of 500
residual particles introduced ﬁom the emptied sampling vial. ........................................ 52

Figure 2- 21: a) Mass spectrum of a single particle of loratadine detected in an original
bottle containing 11 chloroquine phosphate tablets, b) average mass spectrum of 100
loratadine particles previously collected for comparison. ............................................... 53

Figure 3- 1: Diagram of the alveolar-capillary interface. Alveolar air sacks within the
bronchioles of the lung are in close contact with the blood stream via the alveolar-
pulmonary membrane. Oxygen and carbon dioxide, as well as VOCS, partition across
this membrane, making the alveolar air ﬁrnction as a headspace of the blood. (Image
originally in color) ......................................................................................................... 64

Figure 3- 2: Picture of the RTubeTM. The RTubeTM is used for BBC collection and can be
chilled with a previously frozen aluminum sleeve. (Image originally in color) ............... 65

Figure 3- 3: Diagram of the ECoScreen II. This device is used to collect BBC and can
also control the volume or time of sample collection. (Image originally in color) ........... 66

Figure 3- 4: Schematic of a gas chromatograph-mass spectrometer (GC-MS). A sample
is injected onto a capillary column held inside a temperature programmed oven. As the
sample passes through the column, components separate based on partitioning between
the mobile and stationary phases of the column, aided by the temperature ramp in the
oven. The components exit the column directly into the source region of the mass
spectrometer, where they are ionized, fragmented, and pass into the mass analyzer, where
the fragments are detected. The resulting data includes a chromatograrn with a
corresponding mass spectrum for each time point. (Image originally in color) ............... 81

Figure 3- 5: GC-MS total ion chromatograms obtained from a single breath sample
analyzed with the 4 methods shown in Table 3- 1. Method 1 was selected as the optimal
detection method. .......................................................................................................... 87

Figure 3- 6: Comparison of TIC peak area to exposure time of SPME ﬁber. The
exposure time of the SPME ﬁber is equivalent to the breath collection time for these EBV
samples. (Image originally in color) ............................................................................... 88

xiii

Figure 3- 7: Chromatograms obtained from (a) 10 minute exposure of the SPME ﬁber to
the ambient room air at the collection site, (b) a representative breath sample (BBC), and
(c) the difference between the chromatograms in a and b. Note that the majority of peaks
in breath are in much greater concentration than those in the room air. .......................... 90

Figure 3- 8: Exhaled breath condensate (BBC) and vapor (EBV) collection devices. a.)
the commercially available RTubeTM; b.) the RTubeTM was modiﬁed for EBV collection
by the addition of a plastic ﬁtting to hold a SPME ﬁber for active sampling; c.) the
additional modiﬁcation of the RTubeTM to provide an enclosed environment for SPME
sampling as well as the addition of a respirator to purify inhaled air. (Image originally in
color) ............................................................................................................................. 96

Figure 3- 9: Total ion chromatograms of breath samples from a single subject. Samples
were obtained using several collection methods: EBC with a -80 °C condenser (black
trace, bottom), EBV-80°C (green trace, middle), and EBV-RT (red trace, top). Arrows
indicate several peaks with increased signal in the EBV samples compared to the BBC
sample. Chromatograms are offset for clarity. (Image originally in color) ...................... 98

Figure 3- 10: Extracted ion chromatograms (XIC) of breath samples from a single subject
(BBC). One sample was collected 215 minutes after the subject received a manicure with
an RTubeTM modiﬁed with PVC tubing which provided inhaled air from a separate room
(black trace, bottom). Another sample was obtained 317 minutes after receiving the
manicure, and was obtained with the standard unmodiﬁed RTubeTM (red trace, top). a.)
the XIC of m/z 45, identiﬁed as a major fragment ion ﬁom isopropyl alcohol, b.) the XIC
of m/z 43, corresponding to a major fragment ion from diacetone alcohol, and c.) the XIC
of m/z 152 corresponding to the molecular ion of camphor. (Image originally in color)
.................................................................................................................................... 102

Figure 3- 11: Total ion current chromatograms of breath samples from a single subject
(EBV-RT) collected using a variety of inhaled air sources: ambient air (red trace, top),
ambient air through a respirator (blue trace, middle), and ambient air from a separate
room than sampling (green trace, bottom). The subject painted a simulated ﬁngernail
with commercial nail polish during sampling. Compounds labeled are listed ingredients
of the nail polish or byproducts of ingredients. Chromatograms are offset for clarity.
(Image originally in color) ........................................................................................... 104

Figure 3- 12: Extracted ion chromatograms (XIC m/z 93) of breath samples from a single
subject (EBV-RT). One sample was collected over 4 oz. of lemonade with no
modiﬁcation to the inhaled air (red trace, top) and another sample was collected over 4
oz. of lemonade using a respirator (blue trace, bottom). Peaks labeled are identiﬁed as
monoterpenes and monoterpene alcohols, ﬂavor ingredients present in lemonade. (Image
originally in color) ....................................................................................................... 106

Figure 3- 13: Total ion current chromatograms of ‘breath’ samples obtained using a
synthetic lung substitute over gasoline (EBV-RT). One sample was collected with no
modiﬁcation to the inhaled air (red trace, top), a second sample was collected using a
single respirator (blue trace, middle), and a ﬁnal sample was collected using two

xiv

respirators in series (green trace, bottom) Chromatograms are offset for clarity. (Image
originally in color) ....................................................................................................... 108

Figure 4- 1: GC-MS total ion chromatograms of sampled breath constituents obtained
before salon exposure (top) and 12 minutes after leaving the salon (bottom). The signal
obtained after leaving the salon was inverted for clarity. Many peaks endogenous to
breath are present in both samples. Additional peaks are present in the sample taken after
the subject left the salon, such as those indicated by the arrows. (Image originally in
color) ........................................................................................................................... 130

Figure 4- 2: (3) Selected window of GC-MS total ion chromatograms obtained before
salon exposure (black), and 12, 50, 99, 152, and 214 minutes after leaving the salon. The
source of the peak was identiﬁed as a common nail polish ingredient, camphor.
Chromatograms are slightly offset for clarity. (b) Exponential decay, R2: 0.999, of the
area of the peak due to camphor, calculated from the XIC (m/z 152) for the molecular ion
(chromatogram not shown). (Image originally in color) ............................................... 131

Figure 4- 3: (a) Selected window of GC-MS XIC for m/z 91showing the peak due to
toluene in breath before (no peak detected) and, 11, 25, 52, 79, 121, 168, 222, 285, and
313 minutes after salon exposure. (b)Decay of the area of the peak due to toluene ﬁt to an
exponential process (R2=0.960). (Image originally in color) ........................................ 132

Figure 4- 4: GC-MS total ion chromatogram obtained by exposing a SPME ﬁber to the
manicured hand inside a plastic bag 5 hours, 1 minute after salon exposure. Note that
chemicals indicative of the nail polish are still present in the headspace of the nails ..... 133

Figure 4- 5: (a.) Selected window of XICs (m/z 105) obtained ﬁom Subject A before
exposure (black) in the hardware store and 8, 70, and 125 minutes after leaving the
hardware store. (b.) Selected window of Xle (m/z 105) obtained from Subject B before
exposure (black) and 4, 28, and 37 minutes after leaving the hardware store. The source
of the peak was tentatively identiﬁed as an ingredient in furniture stain, l-methoxyethyl
benzoate. This chemical was also detected in a SPME sample of the ambient air in the
hardware store. (Image originally in color) .................................................................. 135

Figure 4- 6: Extracted Ion Chromatograms (XICs) of the m/z 93 ion in breath samples
obtained before lemonade consumption (black trace) and 3 min after original lemonade
consumption (red trace) offset for clarity. Inset shows magniﬁcation of boxed area of the
sample taken before lemonade exposure. 13 terpenes and terpenoid compounds of
interest are identiﬁed. (Image originally in color) ........................................................ 136

Figure 4- 7: (a) XICs (m/z 93) of peak identiﬁed as u-limonene in breath samples
obtained before and 3-1368 minutes after original lemonade consumption. (b.) Magniﬁed
view of XICs shown in (a) obtained 247-1368 minutes after original lemonade
consumption, c) Area of peak due to a-limonene corrected for initial area before
lemonade consumption plotted versus time after ingestion. These data are ﬁt to a single
phase exponential uptake and decay model. (Image originally in color) ....................... 140

Figure 4— 8: (a) GC-MS XICs (m/z 93) of peak identiﬁed as y-terpinene in breath samples
obtained before and 3-1368 minutes after original lemonade consumption, (b) Magniﬁed

XV

view of XICS shown in Figure 4a obtained 247-1368 minutes after original lemonade
consumption, (c) Area of peak due to y-terpinene versus time aﬁer ingestion. These data
are ﬁt to a single phase exponential uptake and decay model. (Image originally in color)
.................................................................................................................................... 142

Figure 4- 9: a—Limonene peak area versus time after consumption for original,
concentrated, and Mediterranean lemonade. Note the higher tmax for higher lemonade
concentrations as well as the appearance of the uptake curve of the a-limonene. (Image
originally in color) ....................................................................................................... 145

Figure 4- 10: Relative peak areas of 13 terpene and terpenoid compounds in (a) neat
samples of original, concentrated, and Mediterranean lemonade, and (b) breath samples
after consumption of original, concentrated, and Mediterranean lemonade. Peak areas are
normalized to the original lemonade peak area for each compound, and the breath sample
obtained at tmax was used. (Image originally in color) ................................................... 146

Figure 4- 11: Uptake and decay of a-limonene aﬁer consumption of Mediterranean
lemonade ﬁt to Equation 4- 1.Equation 4- 2 Fit parameters are shown in Table 4- 3. 148

Figure 4- 12: Uptake and decay of a-limonene after consumption of Mediterranean

lemonade ﬁt to Equation 4- 2 using the method of residuals. Fit parameters are shown in
Table 4- 3 .................................................................................................................... 149

Figure 4- 13: Selected window of the m/z 58 chromatogram obtained in SIM mode before
and after ingestion of 60 mg of pseudoephedrine. The peak has been identiﬁed as
pseudoephedrine. (a) shows all breath samples obtained, (b) shows a zoomed in view of
the samples excluding that obtained 10 min after consumption for clarity. (Image
originally in color) ....................................................................................................... 153

Figure 4- 14: (3) Selected window of GC—MS XIC (m/z 140) for breath samples (EBC)
obtained before (black) and 26 min after (red) exposure to hexamine, (b) Selected
window of XIC (m/z 140) for breath samples (EBV) obtained before (black) and 30 min
after (red) exposure to hexamine. (Image originally in color) ....................................... 156

Figure 4- 15: (a) Selected window of TIC (SIM, m/z 140, 42) ﬁ'om breath samples (EBV)
of a single subject before and after exposure to hexamine, (b) Exponential decay, R2:
0.994, of the area of the peak due to hexamine, calculated from the XIC (m/z 140)
(chromatogram not shown). (Image originally in color) ............................................... 157

Figure 4- 16: Integrated area of GC-MS XIC m/z 140 peak attributed to hexamine at
RT=7.69 min in chromatograms obtained in SIM mode (m/z 140 and 42) of breath
samples from a single subject before, during, and after several 10 min exposures (shown
in red) to hexamine. Note the increase in hexamine on the breath during exposure and
the gradual decay after exposure. Hexarnine was still detected on breath 24 hrs after the
initial exposure. (Image originally in color) ................................................................. 158

Figure, 5- 1: Diagram of a typical time capnogram, which displays the concentration of
carbon dioxide in respiratory gas. Inhalation is represented by Phase I-O, exhalation is
represented by Phases E-l, E-2, and E-3, corresponding to exhalation of deadspace (C02

xvi

ﬁ'ee) air, a mixture of dead space and alveolar air, and primarily alveolar air, respectively.
Adapted from reference 6. ........................................................................................... 171

Figure 5- 2: Diagram of possible handheld breath analyzer. Breath is exhaled into the
mouthpiece where it passes into a thermally controlled preconcentration region. After a
predetermined period of time, the temperature of the preconcentrator is quickly ramped,
desorbing the analytes of interest and allowing them to pass into the separation region,
possibly a miniaturized GC where separation occurs. The eluent then passes into a
detection system, possibly a miniaturized MS where the analytes are identiﬁed and a
signal is reported to the user (signal digitization not shown). ....................................... 172

Figure A- 1: Rules tree for classiﬁcation of pseudoephedrine. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘PSE’ indicate a positive identiﬁcation for
pseudoephedrine; circles labeled ‘Null’ indicate a negative response. .......................... 176

Figure A- 2: Rules tree for classiﬁcation of Coricidin. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘C’ indicate a positive identiﬁcation for Coricidin;
circles labeled ‘Null’ indicate a negative response. ...................................................... 177

Figure A- 3: Rules tree for classiﬁcation of loratadine. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘L’ indicate a positive identiﬁcation for loratadine;
circles labeled ‘Null’ indicate a negative response. ...................................................... 177

Figure A- 4: Rules tree for classiﬁcation of acetaminophen. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘Ac’ indicate a positive identiﬁcation for
acetaminophen; circles labeled ‘Null’ indicate a negative response. ............................. 178

Figure A- 5: Rules tree for classiﬁcation of dextromethorphan. The mass vectors probed
at each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘D’ indicate a positive identiﬁcation for
dextromethorphan; circles labeled ‘Null’ indicate a negative response. ........................ 178

Figure A- 6: Rules tree for classiﬁcation of aspirin. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘A’ indicate a positive identiﬁcation for aspirin;
circles labeled ‘Null’ indicate a negative response. ...................................................... 179

Figure A- 7: Rules tree for classiﬁcation of phenylephrine. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘PE’ indicate a positive identiﬁcation for
phenylephrine; circles labeled ‘Null’ indicate a negative response. .............................. 180

Figure A- 8: Rules tree for classiﬁcation of guaifenesin. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a

xvii

negative response. Circles labeled ‘G’ indicate a positive identiﬁcation for guaifenesin;
circles labeled ‘Null’ indicate a negative response. ...................................................... 180

Figure B- 1: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-
lfor ﬁt parameters. ...................................................................................................... 181

Figure B- 2: Integrated area of the peak attributed to a-limonene in chromatograms
obtained ﬁom the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 182

Figure B- 3: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I
for ﬁt parameters. ........................................................................................................ 182

Figure B- 4: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table 8- 1 for ﬁt parameters. ................................................................................ 183

Figure B- 5: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 183

Figure B- 6: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 184

Figure B- 7: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table 8- 1 for ﬁt parameters. ................................................................................ 184

Figure B- 8: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 185

Figure B- 9: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 185

xviii

Figure B- 10: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 186

Figure B- 11: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 186

Figure B- 12: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 187

Figure B- 13: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained ﬁ'om the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 187

Figure B- 14: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 188

Figure B- 15: Integrated area of the peak attributed to u-terpineol in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 188

Figure B- 16: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 189

Figure B- 17: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 189

Figure B- 18: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained ﬁ'om the breath of subject A on the third day of original lemonade consumption.
See Table B- 1 for ﬁt parameters. ................................................................................ 190

Figure B- 19: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 190

xix

Figure B- 20: Integrated area of the peak attributed to u-terpinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 191

Figure B- 21: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 191

Figure B- 22: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 192

Figure B- 23: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 192

Figure B- 24: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- I for ﬁt parameters. ................................................................................ 193

Figure B- 25: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 193

Figure B- 26: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 194

Figure B- 27: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 194

Figure B- 28: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 195

Figure B- 29: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table 8- 1 for ﬁt parameters. ................................................................................ 195

XX

Figure B- 30: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 196

Figure B- 31: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 196

Figure B- 32: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 197

Figure B- 33: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 197

Figure B- 34: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-l).
See Table B- 1 for ﬁt parameters. ................................................................................ 198

Figure B- 35: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 198

Figure B- 36: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 199

Figure B- 37: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 199

Figure B- 38: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 200

Figure B- 39: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 200

xxi

Figure B- 40: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 201

Figure B- 41: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 201

Figure B- 42: Integrated area of the peak attributed to a-pinene in chromatograms
obtained ﬁom the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 202

Figure B- 43: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 202

Figure B- 44: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 203

Figure B- 45: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 203

Figure B- 46: Integrated area of the peak attributed to B-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 204

Figure B- 47: Integrated area of the peak attributed to [i-pinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 204

Figure B- 48: Integrated area of the peak attributed to B-pinene in chromatograms
obtained ﬁom the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 205

Figure B- 49: Integrated area of the peak attributed to B-pinene in chromatograms
obtained ﬁom the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-l).
See Table B- l for ﬁt parameters. ................................................................................ 205

xxii

_—3L_:—

 

Figure B- 50: Integrated area of the peak attributed to B-pinene in chromatograms
obtained ﬁom the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 206

Figure B- 51: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 206

Figure B- 52: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table 8- I for ﬁt parameters. ................................................................................ 207

Figure B- 53: Integrated area of the peak attributed to li-phellandrene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 207

Figure B- 54: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained ﬁom the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 208

Figure B- 55: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 208

Figure B- 56: Integrated area of the peak attributed to a-thujene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 209

Figure B- 57: Integrated area of the peak attributed to a-thujene in chromatograms
obtained ﬁ'om the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table 8- 1 for ﬁt parameters. ................................................................................ 209

Figure B- 58: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 210

Figure B- 59: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 210

xxiii

Figure B- 60: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 211

Figure B- 61: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 211

Figure B- 62: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 212

Figure B- 63: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 212

Figure B- 64: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained ﬁom the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 213

Figure B- 65: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 213

Figure B- 66: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 214

Figure B- 67: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 214

Figure B- 68: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 215

Figure B- 69: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 215

xxiv

Figure B- 70: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained ﬁom the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 216

Figure B- 71: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- l
for ﬁt parameters. ........................................................................................................ 216

Figure B- 72: Integrated area of the peak attributed to camphene in chromatograms
obtained ﬁ'om the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 217

Figure B- 73: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 217

Figure B- 74: Integrated area of the peak attributed to a-pinene in chromatograms
obtained ﬁ'om the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table 8- 1
for ﬁt parameters. ........................................................................................................ 218

Figure B- 75: Integrated area of the peak attributed to B-pinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 218

Figure B- 76: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 219

Figure B- 77: Integrated area of the peak attributed to u-thujene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 219

Figure B- 78: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 220

Figure B- 79: Integrated area of the peak attributed to u-limonene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 220

XXV

Figure B- 80: Integrated area of the peak attributed to 4—terpineol in chromatograms
obtained ﬁom the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I
for ﬁt parameters. ........................................................................................................ 221

Figure B- 81: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained ﬁom the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 221

Figure B- 82: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 222

Figure B- 83: Integrated area of the peak attributed to a—terpinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 222

Figure B- 84: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 223

Figure B- 85: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I
for ﬁt parameters. ........................................................................................................ 223

Figure B- 86: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 224

Figure B- 87: Integrated area of the peak attributed to a-pinene in chromatograms
obtained ﬁom the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 224

Figure B- 88: Integrated area of the peak attributed to B-pinene in chromatograms
obtained ﬁom the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 225

Figure B- 89: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 225

xxvi

.Unb
-

.~.. _.

..~_§.

D' 1‘.-

..\ u.-

 

“o -4 I.

 

 

. '-
6 ~ .

T '1
‘4§4.\ -

-
'na _)'
.,___.‘
‘II. “I“
.1. \

. ‘
- ~
\e --
. g A., ‘
-

u
,.

, .1
3
'h... ‘
x

C“.
.._

v..-
e \
“-
..

. V
\N .
-y‘ ‘- .‘
>-

' y

 

. --‘_"
.“ .y
-.
b. \ ~-
.
m
_ ‘
q; n
~\. I .
I . h
. ’.‘. H
:‘W L.
a”.
. ‘_ ‘
..‘
._.
‘t \ "'u
"‘
'-
. .
‘
”I
\‘\ I
‘-

 

‘-

hm a.“
a.“

'e

‘ a

\- .

~h ‘

.5 1..

 

t "
._\ 5 t.
a,“
\1: .V
“5 lb
.‘ ’_
._.

Figure B- 90: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- l
for ﬁt parameters. ........................................................................................................ 226

Figure B- 91: Integrated area of the peak attributed to B—myrcene in chromatograms
obtained ﬁom the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- 1
for ﬁt parameters. ........................................................................................................ 226

Figure B- 92: Integrated area of the peak attributed to a-limonene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 227

Figure B- 93: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 227

Figure B- 94: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 228

Figure B- 95: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 228

Figure B- 96: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained ﬁom the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 229

Figure B- 97: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 229

Figure B- 98: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 230

Figure B- 99: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 230

xxvii

Figure B- 100: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 231

Figure B- 101: Integrated area of the peak attributed to B-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁt parameters. ................................................................................ 231

Figure B- 102: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 232

Figure B- 103: Integrated area of the peak attributed to a-thujene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 232

Figure B- 104: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- 1 for ﬁt parameters. ................................................................................ 233'

Figure B- 105: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 233

Figure B- 106: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject an on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 234

Figure B- 107: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 234

Figure B- 108: Integrated area of the peak attributed to terpinolene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption ............................................................................................................... 235

Figure B- 109: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained ﬁ'om the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 235

Figure B- 110: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 236

xxviii

Figure B- 111: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 236

Figure B- 112: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 237

Figure B- 113: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 237

Figure B- 114: Integrated area of the peak attributed to B-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 238

Figure B- 115: Integrated area of the peak attributed to B-phellandrene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 238

Figure B- 116: Integrated area of the peak attributed to u-thujene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 239

Figure B- 117: Integrated area of the peak attributed to B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption. ............................................................................................................... 239

xxix

LIST OF EQUATIONS

Equation 2- 1: Relationship between aerodynamic diameter and Stokes’ Diameter, ....... 14

Equation 2- 2: Relationship between time and mass-to-charge ratio in a time-of-ﬂight
mass analyzer, ............................................................................................................... 15

Equation 3- 1: Equation describing the partitioning of an analyte into a SPME ﬁber

coating ........................................................................................................................... 70
Equation 3- 2: Description of the equilibrium constant of an analyte as it partitions
between the stationary and mobile phases during separation. ......................................... 74
Equation 3- 3: Relationship between the partition ratio and retention time. .................... 74
Equation 3- 4: Deﬁnition of the phase ratio .................................................................... 75
Equation 3- 5: Deﬁnition of the number of theoretical plates. ........................................ 75
Equation 3- 6: Deﬁnition of the effective number of theoretical plates. .......................... 76
Equation 3- 7: Deﬁnition of the mean free path of an ion. .............................................. 76

Equation 3- 8: Approximation of the mean free path of an ion at 300 K and 3.8x10'IO
molecules/cm’. .............................................................................................................. 77

Equation 3- 9: Equations describing the electric ﬁeld formed in the center of a quadrupole

mass analyzer. ............................................................................................................... 79
Equation 3- 10: Relationship between angular frequency and RF ﬁeld frequency. ......... 79
Equation 4- 1: Exponential rise and decay with a y-offset ............................................ 141

Equation 4- 2: Model of uptake and triphasic decay for monoterpenes and terpenoids .149

XXX

CHAPTER 1: Introduction

[-1 : Motivations

Protection of the public from threats of violence in the 21St century is based upon the

principles of deterrence, prevention, preemption, crisis management, consequence
management, attribution, and response.1 Deterrence of violence involves actions that
prevent the terrorist from planning an attack.2 These actions can include fear of
retribution, as seen in the nuclear arms race of the Cold War. Prevention is needed when
deterrence fails and a terrorist group moves ahead with an attack plan. Prevention takes
place at ports of entry, border checkpoints, and in many law enforcement actions that can
stop or mitigate a planned attack.3 Preemption follows closely with prevention, and aims
to stop an attack before it occurs. This often includes an action, such as a raid by law
enforcement, to disrupt the action of the terrorist.4 If an attack does occur, crisis and
consequence management become key functions in the protection of public safety.
Strong leadership is required to coordinate multiple agencies responding to a crisis and
ensure information is shared. Consequence management attempts to mitigate the effects
of an attack. This occurs simultaneously with crisis management, and for example,
would involve the coordination and actions of public health personnel after a biological
attack. Attributing a terrorist attack involves identifying and conﬁrming the parties who
participated in the events This involves not only the forensic analysis of the scene of the
attack, but also analysis of money and paper trails that may identify silent partners.6 The
ability to quickly and effectively attribute an attack also feeds back into deterrence, as
being identiﬁed as the attackers may prevent an organization from planning an attack.

Finally, response to violence is the ﬁnal principle of protecting the public. It includes not

only focusing on the current identiﬁed threat, but also preventing potential criminals and
terrorists ﬁom engaging in ﬁrture violent activity. All of the principles rely on studying
the aggressors, i.e., the social science of studying the psychology of terrorist groups,
identifying the aggressors, i.e., law enforcement efforts, as well as the science and

technology of understanding and identifying the threats.

The science and technology necessary for protecting the public from violence relies
heavily on analytical chemistry to identify suspicious materials and persons.
Traditionally this role has been more in the attribution phase, determining the type and
origin of an explosive, for example. But, as technology has advanced and new
techniques have been created, analytical chemistry can and will play an important role in
the prevention and preemption of terrorist attacks. More sensitive and selective detection
techniques can allow suspicious materials to be identiﬁed even before a ﬁnal destructive
product is made. The work presented herein focuses on the use of commercial and novel

detection techniques for application to the prevention of terrorist activities.
1-2: Overview

Chapter 2 will discuss advances in the detection of drugs using Single Particle Aerosol
Mass Spectrometry (SPAMS), which is a rapid, sensitive, and information-rich analytical

tool. Although drugs are not commonly thought of when discussing terrorism,
narcoterrorism has become a signiﬁcant threat in the 21St century. The role of the illicit
drug trade in the ﬁmding of terrorist groups is prevalent, thus, reducing the trafﬁcking of

illegal drugs can play a role in the prevention of terrorism by cutting off much needed

funding. To do so, sensitive, speciﬁc, and robust analytical equipment is needed to

V I
5.3..

 

 

 

 

I»
h' \J‘
. q.
u. 0
7 D.
“r‘ v
.‘ “
r.
l
‘ 2!.
._‘.L :o
L, .
v‘)‘;
"\I I}.
n .
3“
‘2; _,
*‘1:‘ ‘5

quickly identify a suspected drug sample no matter what matrix it is in. SPAMS is a
novel technique that has previously been applied to biological and chemical detection;
this chapter will discuss efforts to apply the technique to the identiﬁcation of a variety of

constituents in drug tablets.

Chapters 3 and 4 will discuss a novel analytical scheme to identify potential terrorists and
criminals using a commercial gas chromatography-mass spectrometry (GC-MS) system.
Human breath is sampled, breath constituents are concentrated, and the sensitivity of the
GC-MS allows for trace volatile organic compounds (VOCS) to be identiﬁed in the
breath. These VOCS are present in breath owing either to their production in vivo, or
exposure to an external source of VOCS through absorption, inhalation, or ingestion.
Those chemicals arising from exposure can provide information about the prior location
and activities of the subject; thus, activities such as synthesis of explosive materials may
be detected through analysis of substances in breath. In application, this technique may
provide an early warning system to identify persons of interest in the prevention and
preemption stages of homeland security. Chapter 3 will discuss the development of the
analytical method, including instrument settings, breath sampling procedures and
optimization, as well as repeatability of the analysis. Chapter 4 will present ﬁndings
from several experiments when a subject had a known chemical exposure. Information

on the uptake and decay of several VOCs is included in this chapter.

Chapter 5 will summarize conclusions from this body of work, as well as milestones that
have been achieved. Information about the potential applications of the techniques

developed in these studies will be included.

1-3: References

(1)

(2)

(3)

(4)

(5)
(6)

Combating Terrorism: Protecting the United States - Part 1; Hearing Before the
Subcomm on National Security, Veterans' Affairs, and International Relations of
the House Comm. on Gov. Reform, 107th Cong. (March 23, 2002) (Statement of
Col. R. J. Larson (USAF, Ret.) Director, ANSER Institute for Homeland

Security).

Deterrence, Protection, and Preparation: The New Transportation Security
Imperative, Special Report 270; Transportation Research Board, 2002.

Smith, J. M. In 10th Annual Research Results Conference: Colorado Springs, CO,
2003.

Countering the Changing Threat of International Terrorism; National
Commission on Terrorism, 2000.

R. J. Larsen; David, R. A. Homeland Security 2000, 1-8.
Colby, E. A. Expanded Deterrence In Policy Review, 2008.

CHAPTER 2: Single Particle Aerosol Mass Spectrometry of Drug Samples

2-1: Motivations and Introduction

The manufacture, sale, and use of legal and illegal drugs are major issues in the world
today. The science of pharmaceutical development and quality control/quality assurance
has consistently been a pursuit of the chemist. Analytical techniques that can identify the
active and inactive ingredients of legal drugs, including over-the-counter and prescription
medications, are necessary as it critical to know the components in the drug formulation
and optimize the efﬁcacy of a treatment. Counterfeit prescription drug synthesis and sale
is increasingly becoming a major problem,7 especially with the increase in the number
and availability of online pharmacies and rising drug prices. For example, a study that
analyzed anti-malarial formulations marketed as containing chloroquine as the active
ingredient obtained from street vendors in Cameroon determined that 32% of the samples
analyzed were counterfeit and contained no chloroquine.8 The lack of active ingredient
in such drugs can cause the spread of diseases such as malaria, and well as create new

health concerns.

The motivation for the analysis of illegal drugs is more criminal in nature. The active
and inactive ingredients are still important to identify, but many byproducts, cutting
agents, and contaminants are usually present in illegal drugs, as synthesis is often done
with off the shelf ingredients and apparati, in clandestine locations. These ingredients
may confound the analysis, thus insensitivity to matrix effects is an important quality of
an analytical technique. Another aspect of detecting illicit drugs is that the criminal may
attempt to destroy evidence before analysis, such as rinsing the drugs down the drain or

destroying them in some manner. Analysis of residual particles would then be

advantageous to determine what the prior contents of a bottle or distillation ﬂask were, or

the identity of a powder on the ﬂoor or table of a suspected. drug house.

Drug trade contributes to the threat of terrorism as proﬁts from drug exportation and sale
often ﬁrnd organizations that commit violence against the public. For example, the
Revolutionary Armed Forces of Colombia, FARC, is accused of being a narcoterrorist
organization, monetarily beneﬁting from the production and sale of cocaine.9 It is
important to have an analytical technique that can identify illicit drugs with speed,
sensitivity, and selectivity in any environment, including border crossings to prevent the

importation of illegal drugs into the United States and combat the drug trade.

This chapter will describe efforts made to use the Single Particle Aerosol Mass
Spectrometer to detect and identify drugs. For safety and regulatory reasons, over-the-
counter drugs are used in lieu of illicit drugs, but the results are transferable across
different classes of drugs. Several over-the-counter drugs such as pseudoephedrine are
used as starting materials in the synthesis of illegal drugs, thus their identiﬁcation is
important for identifying processes used in clandestine synthesis of drugs. Also, many
over-the-counter drug formulations are used as cutting agents in counterfeit and illegal
drugs, so their identiﬁcation would aid investigators identify tablets and their sources. An

anti-malarial drug is also analyzed as proof-of-concept for counterfeit drug detection.

2-2: Background on Drug Analysis

Sensitive and speciﬁc analysis of drug tablets has a broad range of applications including

pharmaceutical quality control and forensic analysis of counterfeit drugs or illicit

substances. In a pharmaceutical context, it is important to identify the drug ingredient as
well as any impurities that affect quality control. In forensic investigations, it is
important to identify the drug ingredients and impurities while maintaining the physical
integrity of a tablet. Both industries depend on fast and accurate analytical methods in

order to process a large sample load.

Preliminary forensic tests of a tablet rely ﬁrst on visual inspection to identify any
differences in shape, color, or markings from an authentic pill, or similarities to other
drug seizures for prosecution reasons.7’ ’0 Conﬁrmatory testing requires that the active
ingredient be identiﬁed and characterized. In fact, the World Health Organization
(WHO) states that “the principal requirement for a suitable screening procedure is the
identiﬁcation of the active drug substance”.10 To this end, forensic laboratories as well as
pharmaceutical companies are faced with an ever increasing sample load, stressing the
importance of fast high-throughput analysis. Thus, a simple, fast detection system with
little sample preparation that can identify the active ingredient in a tablet while

preserving the integrity of its structure and visual identiﬁers would be advantageous.

Several methods have been used to separate and/or detect drug samples alone, as
mixtures, or in complex samples. Capillary electrophoresis” '2 and infrared (IR)
spectroscopy have often been used for drug detection, although sample preparation can
be time-consuming.13 "5 To circumvent this issue, near IR reﬂectance spectroscopy has
been used to directly analyze drug tabletsn’ '5 While sensitive, the main limitation to IR
is the training required for the interpretation of the complex spectra the technique yields.
Gas and liquid chromatography (GC and LC)'6'18 have both been used for drug analysis

when coupled to mass spectrometry (MS). The most common technique currently in use

for drug analysis is GC-MS. This technique has the advantage of providing orthogonal
data from one analytical scheme which ﬁrlﬁlls the Scientiﬁc Working Group for the
Analysis of Seized Drugs (SWGDRUG) recommendations for forensic analysis of seized
drugs.19 The main drawbacks of GC-MS are that an extraction step is required which
lengthens the time of analysis, uses consumable reagents, destroys the physical integrity
of the solid sample, many drugs require derivatization because of their low volatility, and
the chromatography itself has an inherent time requirement that can lead to an individual
analysis taking over ten minutes.16 Chromatography is also used to analyze drugs in

16:20:21 and biological samplesn’ ’8 have been analyzed

complex samples. Water samples
using LC-MS/MS, high performance LC-MS, GC with a nitrogen-phosphorus detector,
and GC-MS. Antonio and Heath demonstrated the use of GC-MS in the determination of

several drugs in river sediments after extraction and the addition of a derivatizing agent to

increase the volatility of the drugs. ’6

Because of the sample preparation time required for GC-MS, other options for sample
introduction into a mass spectrometer have been explored. The recently developed
ionization methods direct analysis in real time (DART), desorption atmospheric pressure
chemical ionization (DAPCI) and direct electrospray ionization (DESI) have been used in
conjunction with mass spectrometry to identify active ingredients in prescription, non-
prescription and illicit drugs by placing the tablet in the ionization stream of the
instrument.”30 Although many groups break or scrape the tablets before analysis” 27' 3 1
in order to sample the contents of the center of a coated tablet, little to no sample

preparation is necessary. Molecular fragmentation is commonly minimal; the ions

predominantly obtained are [M+H]+ or [M-H]' which allow for facile identiﬁcation of the

active ingredientszz’ 30 Van Berkel et al. have coupled DESI—MS to TLC separation of
extracted drug ingredients creating an orthogonal approach to drug analysis.32 While
these techniques are promising advances in drug analysis, they do not sample the
heterogeneous body of a tablet, only the small area that interacts with the ion or droplet

stream, which may allow a drug component to go undetected.

Single Particle Aerosol Mass Spectrometry (SPAMS) is an on-line detection system that
individually analyzes single micrometer-sized particles and provides real-time data

including sizing and simultaneously obtained positive and negative ion mass spectra.”35

SPAMS has been applied in a variety of environments,” 34

and is capable of low
concentration detection of target particles in a high background environment. It was
hypothesized that SPAMS would also provide a fast, simple, sample preparation-free
method of analyzing drugs in tablet form. The application of SPAMS to the rapid
detection of over-the-counter drug samples is described herein. Although SPAMS has
previously been applied primarily to aerosolized liquids, this work shows the direct

analysis of solid particles without the need for solvent extraction, and demonstrates the

applicability of SPAMS to drug detection and identiﬁcation.

2-3: Background of SPAMS

Single Particle Aerosol Mass Spectrometry (SPAMS) is a technique based on the original
work of Davis in the 19708.36 Davis’ system allowed aerosol particles to impinge on a
heated Rhenium ribbon causing ions to be formed and released for analysis in a magnetic
sector mass analyzer; the maximum particle yield, deﬁned as the number of particles

producing ions as compared to the number of particles per cm3 of air, was reported as

0.3%.36 Over the past several decades, many modiﬁcations and improvements have been
made to Davis’ original system, which have been reviewed by Noble and Prather,37
resulting in the current commercially available aerosol mass spectrometer from TSI Inc.
(Shoreview, MN). This instrument incorporates two time of ﬂight mass analyzers which
simultaneously analyze both the positive and negative ions produced in the source
region}3 with a particle sizing region that calculates particle size based on the timing of
scattering effects of 2 continuous wave (CW) lasers through which each particle passes.
Since the 19905, work at Lawrence Livermore National Laboratory (LLNL) has focused
on modifying and improving this commercial instrument to ﬁrnction as a sensitive and
selective biological agent detection system under the name of Bioaerosol Mass
Spectrometry (BAMS). Through this work the particle sizing region on the front end of
the instrument, as well as the operating software, was modiﬁed, incorporating up to 6
CW lasers allowing a more accurate determination of the aerodynamic diameter of each
particle.35 This added a degree of selectivity, because particles of a size of interest could
be selectively ionized in the source region, preventing excessive amounts of non-useful
data ﬁom taking up disk space as well as time to be thoroughly analyzed. A laser
induced ﬂuorescence (LIF) region was also added, allowing the ﬂuorescence of each
particle excited at 266 nm to be determined. Advances in data acquisition and storage
have also been made. In the past two years, an additional LIF stage was incorporated to
also measure the ﬂuorescence of each particle excited at 355 nm, as well as a stage to
determine the charge on each particle. These stages can be operated to pre-select
particles of interest for mass spectrometric analysis. In the current conﬁguration, particle

yield, when deﬁned in the same manner as in Davis’ experiments of the 19705, can be as

10

high as 90%. The operating software for the LLNL system is very powerful. After
‘training’ the system on a particular target particle type by collecting many spectra of a
standard sample, alarm ﬁles can be made and the instrument will record an alarm
whenever a user-set threshold number of particles identiﬁed by the software as containing
that speciﬁc compound described by the alarm ﬁle in a set period of time are detected.
This allows the instrument to operate autonomously with little user interaction for air

monitoring studies.

Although originally applied to the detection of such defense threat analytes as Bacillus
spores?"44 viruses, 4’ toxin simulants,41 fungi,39 and Mycobacteria,44 work at LLNL has
expanded the application of the BAMS system to include explosives,” chemical warfare
agent simulants,46 and even stretched into medical applications.“ 47 Because of this
broader sample application, the name BAMS was ofﬁcially changed to SPAMS45 to

present a more universal detection system.

A signiﬁcant advantage of SPAMS is its ability to analyze solid and liquid samples with
no sample preparation. Vapor phase samples can also be analyzed, but do require some
sample preparation to create particles large enough to detect.48 It is also able to detect a
small number of particles with a particular chemical signature in the midst of a large
number of particles that do not share the same chemistry. This ability is aided by the
capacity to pre-select particles for mass spectrometric analysis using the ﬂuorescence

and/or charge stages.

11

2-4: SPAMS Description

A schematic of the SPAMS instrument used at LLNL is shown in Figure 2- 1.39

   
  
  

Pressure Flow Reducer
Relaxation Chamber

Tracking Region
(660 nm diode lasers)
Fluorescence/Charge Stages

I I I “mam—miw-J "*3-

III+ + + r“

 

 

 

 

 

 

-||l
“in.

-I I 1
Negative Ions

 

 

 

 

Positive Ions

 

Reﬂectron Reﬂectron
TOF-MS . TOF-MS
Microchannel 266 nm In laser
Plates

Figure 2— I .' Schematic of the Single Particle Aerosol Mass Spectrometer. Particles are
generated using the modiﬁed glass vial and transferred to the SPAMS via conductive
tubing. After passing through a pressure ﬂow reducer with a relaxation chamber and
aerodynamic focusing lens, the pressure is lowered and the beam is focused into a linear
beam. Particles are then tracked, where the aerodynamic velocity and size are
determined. Laser induced ﬂuorescence at two wavelengths (266 and 355 nm) as well as
particle charge is then determined before the particle enters the source region of the
mass spectrometer. The particle is desorbed and ionized by an Nd: YAG laser operating
at 266 nm, and the positive and negative ions created are concurrently accelerated in
linearly opposing directions into two reﬂectron time-of-ﬂight mass analyzers, where two
mass spectra are recorded for each particle.

Particles released from the analyte in the solid sampling vial (discussed below) are
carried to the SPAMS via 3/8” conductive tubing and enter the pressure flow reducer
which reduces the pressure of the analyte stream for introduction to the higher vacuum
chambers by passing the particles into two facing conical apertures, thereby cutting the
ﬂow signiﬁcantly. On the other side of the apertures the particles lose their directionality
in the relaxation chamber. From this chamber, the particles pass into the aerodynamic
focusing lens, which centers the particles into a thin stream by passing them through a
series of nine consecutively thinner lenses with oriﬁces of decreasing diameter. The

particles then pass into the tracking region of the SPAMS, where their aerodynamic

diameter (dp,a) is determined. As the particles travel through this region of the

instrument, the larger particles move more slowly than the smaller particles; thus size can
be extrapolated from a calculation of the particle’s speed. The tracking region (~10‘4
Torr) contains up to 6 diode lasers (660 nm, continuous wave) each paired with a channel
photomultiplier tube (CPMTs, part c962, Perkin Elmer Optoelectronics, Santa Clara,
CA). As the particle travels through the laser beams, the light is scattered from its normal
path, and a change in laser light intensity is detected by the CPMTs. The time of each
scattering event is recorded, and the series of scattering events can be used to determine
the particle’s velocity and therefore aerodynamic size using a ﬁeld-programmable gate
array. Incorporating 6 lasers provides several scattering events per particle that would
allow for the deconvolution of velocities of two particles traveling through the tracking
region simultaneously. The size measurement is calibrated by analysis of spherical
standards (polymeric beads, Bangs Laboratories, Inc., Fishers, IN, and Duke Scientific

Corp, Fremont, CA) of known size. Aerodynamic diameter, and not actual diameter, is

13

used to describe aerosol particles. Aerodynamic diameter is the diameter of a unit-
density spherical object that moves with the same velocity through a ﬂuid49 and relates to

Stokes’ diameter, which assumes a particle is spherical, by Equation 2- l:
dim = dag/6,,

Equation 2- I : Relationship between aerodynamic diameter and Stokes’ Diameter,

where dp,a is the aerodynamic diameter of a particle, dp93 is the Stokes’ diameter of the

particle, and SP is the density of the particle.

After size is determined, the particle enters the source region of the mass spectrometer
(~106 Torr). The previously determined particle velocity is used to calculate the exact
time the particle will be centered in the source region, and internal electronics trigger the
desorption/ionization (D/I) laser to ﬁre at this precise moment. The laser, a Q-switched
frequency-quadrupled Nd:YAG (Ultra CFR, Big Sky Laser Technologies, Inc., Bozeman,
MT), operates at 266 nm with 7 ns FWHM pulses and can ﬁre at up to 20 Hz with
ﬂuences between 0.05-8.4 nJ/umz. When ﬁred, the laser imparts a signiﬁcant amount of
energy to the particle, desorbing and ionizing it to create both positive and negative ions.
These ions are simultaneously accelerated by an electric ﬁeld in linearly opposing
directions into two linear reﬂectron time-of-ﬂight (TOF) mass analyzers,50 each 536 mm
in length. The TOF mass analyzer operates by determining an ion’s mass/charge by
measuring the time between acceleration and arrival at the detection plate as it travels

through a ﬁeld-free drift region.” 5 l The reﬂectron aspect of the TOP was developed to

14

improve the resolution of the time of arrival of ions of equal masses which have slightly
different ﬂight times owing to the variation in their ionization time, initial position, and
initial kinetic energy.51 It involves the addition of retarding plates at the end of the ﬂight
tube which deﬂect the ions and reverse their direction to pass back through the ﬂight tube
to ring microchannel plate detectors (MCP, Burle Technologies, Inc., Lancaster, PA)
surrounding the exit from the source region. Ions that were imparted with larger kinetic
energies or were closer to the ﬂight tube during ionization and acceleration travel deeper
into the retarding ﬁeld than do those with lower kinetic energies or those that were
further ﬁ'om the ﬂight tube; thus, the faster ions are slowed slightly, allowing both types

of ions to reach the detector simultaneously. Time is then related to the mass-to-charge

ratio (m/z) by Equation 2- 2: 5'

t2 _ md2
ZZBVS

 

Equation 2- 2: Relationship between time and mass-to-charge ratio in a time-of-ﬂight

"lass analyzer,

Where In is the mass of the ion (kg), (1 is the distance the ion travels to the detector (m), z

‘3 the charge on the ion, e is the charge on an electron (C), and V5 is the accelerating

voltage (V). The signal ﬁ'om the MCPs is digitized (Signatec, Inc., Newport Beach, CA,
lDqDAMOOO) at 500 MHz, 8 bits with 256 arbitrary units (a.u.) on a 333 mV scale, yielding

a mass spectrum from both polarities of ions, providing a signiﬁcant amount of

111formation on each single particle.

15

2—5 .' Laser Optics

Figure 2- 2 shows a schematic of the optics used to create the D/I laser beam in the

source region.

Mass Analyzer
Vertical Polarizer

Frequency Doublers
Nd:YAG l:l| G

 

 

. Mirrors

/' Plano-Convex

Adjustable Lens (10 cm)
Aperture Half-Wave Plate

Figure 2- 2." Diagram of the desorption/ionization laser optics. A Nd: YAG laser is
frequency quadrupled to generate a 266 nm beam which is then passed through an
aperture, adjustable half-wave plate and vertical polarizer before being imaged by a
Plano-convex lens with a 10 cm focus length onto two planar mirrors which direct the
Image into the source region of the mass spectrometer.

The Nd:YAG laser beam, originally 1064 nm in wavelength, is passed through 2
t‘requency doublers to form a 266 nm beam. The beam then passes through an aperture
and a half-wave plate, which is manually rotatable to control the polarization of the beam.
OTlly vertically polarized light is then passed through the vertical polarizer; thus, the half-
Vvaye plate and polarizer together control the energy of the transmitted beam. The beam
paSses through a 10 cm plano-convex lens which creates an image of the aperture that is

r'Eﬂected into the source region on 2 planar mirrors. The position of the mirrors is

controlled by manual calipers which manipulate the position of the image on the x-y
plane. The resulting beam is ~ 330 pm in diameter in the center of the source region, and
maintains a relatively ﬂat-topped energy proﬁle,41 although weaker areas do exist,
particularly near the beam edges. The energy of each laser pulse is measured at the back
of the source region using a digital laser energy meter (Coherent, Inc., Santa Clara, CA,
J 25LP—MUV). The laser energy is averaged over 50 particles to give statistically
signiﬁcant values for energy, with the pulse-to-pulse variation generally being <5%.
Laser ﬂuence is often reported, and is equivalent to the laser energy per beam area, which

in the current system is 0.086 mmz.

2- 6 .’ Solid Sampling Materials

Figme 2- 3 shows the modiﬁed glass vial used to introduce solid samples, particularly

drug tablets, to the SPAMS.

Stainless
Steel Outlet\‘ 1
to SPAMS "

 
  
 

Stainless Steel
Tubing Modiﬁed
with a HEPA Filter

Figure 2- 3: Diagram of modified glass vial used to generate particles ﬁom solid samples

for analysis by SPAMS. The plastic cap was modified using stainless steel tubing to

create a port to transfer the sample to the SPAMS as well as a second port to which a

HEPA ﬁlter was attached. The solid sample is placed in the vial which is then shaken or
O"texed; particles are generated by the collisions induced by the shaking.

l7

 

 

\h’

l‘_‘

l
".3

s...

 

“r—'§I
‘h

‘ I
“u I.
l
.
‘h
\..1‘
§ ‘ I

A ~20 mL glass vial with a plastic cap was modiﬁed using stainless steel tubing to create
a port to allow connection of 3/8” conductive tubing to connect to the SPAMS
instrument. A second piece of steel tubing was also modiﬁed into the plastic cap to allow
connection of a HEPA ﬁlter to the vial. Particles are created by manually shaking or

vortexing the vial, causing collisions between the vial contents and the vial walls.

2- 7: Data Collection and Analysis

Operation of the SPAMS system and recording of the data obtained is performed using a
program developed at LLNL by Vincent Riot. Data can be collected using all analysis
stages or any combination thereof, with the minimum requirement that two tracking
lasers are used for accurate velocity determination. The current experiments used only
the sizing (3 tracking lasers) and mass spectral regions, generating smaller data ﬁles. The
software can be set to acquire a speciﬁc number of mass spectral ‘hits;’ a ‘hit’ occurs
when mass spectral data is obtained that surpasses a user-deﬁne threshold, typically ~120
mV on the 333 mV scale. Hundreds if not thousands of particles per sample are
analyzed, typically collecting 1000 mass spectral ‘hits’. Once the data is obtained,
spectral analysis occurs using another program developed in house by Paul Steele, Gary
Armstrong, and David Fergenson, ‘lrene,’ which was written in MatLAB. By using
‘Irene,’ spectra can be averaged, sorted by peak area at a particular m/z value, integrated,

and statistically analyzed.

The mass spectrum for each particle is analyzed using ‘Irene’ to determine its
composition by comparison against user-generated alarm ﬁles. The alarm ﬁles used with

SPAMS are typically created using an algorithm based on pattern matching, where the

18

 

350-order vector of the mass spectrum of a single particle is compared to the average
vector from the mass spectra of a known material contained in the alarm ﬁle.52‘ 5" Further
reﬁnement of the alarm ﬁles is then done by creating rules trees to further discriminate
particles. In a rules tree, each ‘branch’ is created by probing a speciﬁc mass vector to
determine if it is greater than or less than a speciﬁc value. If the particle passes onto a
speciﬁc branch, another mass vector may be probed, i.e. a second ‘branch’, etc. until
sufﬁcient differentiation has occurred. Rules trees analyze normalized vector values so
that an increase in overall signal from the MCPs does not affect the classiﬁcation.
Contributions from mass vectors typically seen in all particles, such as m/z 23 and 39,
due to sodium and potassium, respectively, are eliminated to facilitate comparisons of

mass spectra.

Analysis of the heterogeneous drug samples provided many particles that contained
numerous active ingredients, as discussed below, precluding accurate pattern matching
without creating alarm ﬁles and thus a training set requiring a standard mixture for every
possible combination of drug ingredients. Therefore, alarm ﬁles for this study were
created using only rules trees, where speciﬁc m/z values, or mass vectors, could be used
to categorize the components of the particle. A rules tree was created to identify each
drug using m/z values corresponding to characteristic peaks seen in the mass spectra of
each compound. Using a unique alarm ﬁle for each drug allowed detection of any
combination of l, 2, 3, 4, etc. of these compounds in a single particle, as well as reduced
confounding by any ﬁller materials and inactive ingredients. When discriminated with a
Single alarm ﬁle, each particle is thus processed through all ‘branches’ of the rules tree,

and is identiﬁed as containing that speciﬁc drug, or not containing that speciﬁc drug

19

(Null). All particles are then subjected to the rules tree for the second drug, etc., until all
alarms have been run, allowing a single particle to be identiﬁed by multiple alarm ﬁles.
The rules trees used for the identiﬁcation of ibuprofen is shown in Figure 2- 4, and

subsequent rules trees for other drug components are included in Appendix A.

   

    
  

AllParticles
m/z
206 m/z
' , N -205,-206
i. It . y Y
' .t' m/z m/z
-16l,-162 -205,-206
Y
m/z Y m/z m/z
161,162 461 161

Y

N
a . ~ 0
o in

Figure 2- 4: Rules tree for classiﬁcation of ibuprofen. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘1 ’ indicate a positive identiﬁcation for ibuprofen;
circles labeled ‘Null’ indicate a negative response. M/z 206 corresponds to the [M]+ ion,
m/z -205 and -206 correspond to the [ M f and [M-Hf ion, m/z 16] corresponds to the [M-
COOH]+ ion, and m/z -161 corresponds to the [M—COOH] ion.

   

F

It should be noted that although only mass vectors are used to create the alarm ﬁles in
this work, particle size, ﬂuorescence (266 nm and 355 nm excitation), and/or charge
could be incorporated into the alarm ﬁle in the future to further enhance the conﬁdence in

identiﬁcation after an alarm is reported to further the selectivity.

20

2-8: Drug Material Preparation

Single ingredient over-the-counter medications containing acetaminophen
(analgesic/antipyretic), aspirin (NSAID), Clemastine (antihistamine), diphenhydramine
(antihistamine, sedative), ibuprofen (NSAID), loratadine (antihistamine), phenylephrine
(decongestant), or pseudoephedrine (decongestant) were obtained as tablets from
commercial grocery chains. Tablets containing chloroquine (antimalarial) were obtained
ﬁom a commercial pharmacy in Belize City, Belize. Multi-ingredient medications were
also obtained from commercial grocery chains: Tylenol ColdTM and Tylenol PMTM,
containing acetaminophen and diphenhydramine, Tylenol Severe Congestion”,
containing acetaminophen, diphenhydramine, pseudoephedrine, and guaifenesin

(expectorant), and Advil Cold and Sinus”, containing ibuprofen and pseudoephedrine.

For the single tablet experiments, a single tablet was placed in the modiﬁed glass vial and
attached to the SPAMS using 3/8” conductive tubing. The glass vial was either manually
shaken or placed on a vortexer (Fisher Scientiﬁc, Inc.); the collisions of the tablet with
the sides of the glass vial created sufﬁcient particles for analysis. In fact, manual shaking
typically generated and transferred more particles to the SPAMS instrument than could
be analyzed at the 20 Hz frequency of the DH laser. Placing the vial on the vortexer at a
low setting provided a steadier stream of particles to the instrument, without constant
saturation of the sampling rate of the instrument. For the multiple tablet experiments,
several tablets were simultaneously placed in the vial, each containing a different active
ingredient. These tablets were then removed from the vial and the emptied vial again
analyzed. A ﬂuence of 2.7 nJ/um2 (SD=O.1) was used for all samples. The drugs studied

are shown in Table 2- 1, including their molecular weights and chemical structures.

21

 

Table 2- 1: Structures and molecular weights of drug compounds studied.

 

 

 

 

 

 

 

 

Drug Structure Drug Structure
Acetaminophen OH Diphenhydramine
MW]51.06 O O MW255.16
H O 0
Aspirin O /0
MW 1 80.16 k Guaifenesin jg
0 MW 198.09 HO/T\O
O OH H
Chloroquine Cl Ibuprofen
MW 319.18 MW 206.28
H O
N N
\ / on
ER
Chlorpheniramine \ Loratadine
MW 274.12 I N MW 382.15
/
\1‘
Cl
Clemastine l Phenylephrine
MW343.l7 0 MW 167.21
\
\

O o H OH
Dextromethorphan ﬂ 0\ Pseud oephed rine H H
MW271J9 O MWI65.23 N\

/N .

 

 

 

 

 

 

2-9: Drug Analysis Using SPAMS

2-9-1: BayerTM Aspirin

The resulting dual-polarity mass spectrum obtained from a single BayerTM tablet

containing 325 mg of aspirin is shown in Figure 2- 5.

22

 

(a [M-GisCOT [M'CH3COO]+
200JIZM-C02-C2H202'H]-[M-H]' [MCOOHF [M-OH]+
150 \ [rumor [@1130]+
100_ 94‘[C2H302] \ [CsHsl+

l_

 
 
 
 

[2M-CH3COO-

[ZM-C02\ Cszo-H]
CH3cg-coonr‘

\

50‘

250_Jl?-:1914 I I I.

(b)Five Single "02 “m
2oo‘Particle Spectra
150—

IOOJ
50a

 

 

 

 

 

2533 ' A ' 1.21pm
200-
150-
100«

50
,1 1.1.1.

250- ' ‘ 1.69 pm I
200-
150-
100-

so-

0‘ - gmm . .

250‘ - 1.54 pm
200-
150+

5°: ..1 L n

0 a r ‘
250‘ 3.49 pm

 

 

 

 

Ion Signal (a.u.)

 

 

 

 

 

 

 

-300 -200 -100 0 100 200
Mass-to—Charge Ratio

Figure 2- 5: (a) Average single particle aerosol mass spectra of 999 aspirin particles,
and (b) ﬁve consecutively obtained aspirin particles (aeroaynamic diameter displayed in
upper right of each spectrum). The single particle spectra contain all peaks used to
identify aspirin. The spectra from these particles were obtained in less than ﬁve seconds.

23

Figure 2- 5(a) shows the average mass spectrum of 999 particles generated from the
tablet obtained at a rate of 1 spectrum every 1.26 seconds. The average aerodynamic
diameter of the particles was 1.54 pm (RSD 71.7%). The large size distribution (0.59 —
8.71 pm) was expected owing to the method of particle formation, and particle size did
not affect the relative ion abundances. The major ions observed in the positive and
negative mass spectra are characteristic of the aspirin (acetylsalicylic acid) present in the
tablet. Peaks at m/z -l79, -151, and -137, due to [M-H]', [M-H-CO]', and [M-CH3CO]',
respectively, and major peaks in the negative ion spectrum, and have been previously
documented in solid tablet analysis using DESI in negative ion mode to analyze the

broken surface of a tabletzs’ 26

The spectrum also contains several signiﬁcant peaks
hypothesized to arise from the formation and fragmentation of an acetylsalicylic acid
dimer formed via hydrogen-bonding ([2M-COz-C2HzO-H)]' and [2M-COz-C2HzOz-H]', at
m/z -273, and -257, respectively) which can also aid in identiﬁcation of the active
ingredient. The main peaks in the positive ion spectrum are present at m/z +121, +135,
and +163 ([M-CH3COO]+, [M-COOHF, and [M+H-HZO]+, respectively) the latter of
which has been previously described using DART, DESI and DAPCI in positive ion
mode to analyze the broken surface of a tablet.3 1 Small peaks attributed to the dimer are

also seen in the positive ion spectrum at m/z +213, 241, 255, and 283, assigned as [2M-

CH3COO-CH3CO-COOH1’L, [2M-2CH3COO-H]+, [2M-CH3COO-COOH-H]+, and [2M-

COOH-CH3-OH]+, respectively. Figure 2- 5(b) shows the mass spectra obtained from

ﬁve consecutive single particles of the aspirin tablet. Note the presence of all identifying

peaks in the single spectra as well as the consistency in the signal obtained, allowing the

presence of aspirin in the table to be conﬁrmed from a single particle. The speed with

24

 

which the mass spectra are obtained is demonstrated by the fact that the spectra shown in
the ﬁgure were obtained in less than ﬁve seconds. Software developed in-house has the
capability to identify a sample in real-time based on single spectra and could be trained to

identify and alarm when a speciﬁc drug was detected.53

2-9-2: Chloroquine Diphosphate
Figure 2- 6a shows the average dual-polarity mass spectrum obtained from a single tablet
containing 250 mg of an anti-malarial drug, chloroquine diphosphate. 1000 particles were

analyzed in 182 s.

25

 

2507 a) Average Spectrum [H 3P04-2H]'

 

 

 

200 ‘ "=10“ -1755 C" CH3CH(CH2)3N(C2H5)2+ M+H+
- - + - +
150 [2113904311] P206 P03- CH2N(C2H5)2 [2H3P04-154HC]1;+2H1
100 - [M-CH3CH2-Cl]' / +
50‘ [M—CH CH 1' -132 [M-N(C2H\5)2]
3 2
- I” L- ALI L] J 1 11111.. L]. Jﬁl

I I r

o . .
250 - b) Five Single 2-04 11m

200 - Particle Spectra
150 ~
100 -

522.4% Wt...

,—
250 - 2.05 pm

 

 

 

 

 

 

 

 

[on Signal (a.u.)

 

150' I
100‘ '
50+ J
0 .4 L...L._ AAAAA .l 1 J . - .LJ 1.4.... ASLL. A l .14... __..

250 - ' ' ' ' ' ' 1.29 11m
200 4
150 -

 

 

 

 

 

250 - 1.76 pm

 

 

 

 

 

 

 

 

 

-300 -200 -100 0 100 200 300
Mass-to-Charge Ratio

Figure 2- 6: a) average dual-polarity mass spectrum of 1000 particles dislodged from a

single chloroquine phosphate tablet, b) mass spectra obtained from 5 individual particles
of chloroquine phosphate dislodged from the tablet. Particle size is notated.

26

The average aerodynamic diameter of the particles was 1.88 pm (RSD 43.7%). Most
peaks in the positive and negative polarities can be attributed to the parent or fragment
ions of the drug compound (7-chloro-4-((4-(diethylamino)-l-methy1butyl)amino)-
quinoline = M, MW 319.18). The [M+H]+ ion is the most abundant ion in the positive
spectrum at m/z 320, followed by the [CH2N(C2H5)2]+ fragment at m/z 86. Other
fragments due to the loss of chlorine and addition of hydrogen (m/z 286), the loss of
N(C2H5)2 (m/z 247), and [CH3CH(CH2)3N(C2H5)2]+ (m/z 142) are also seen in the
positive spectrum. The ﬁnal major fragment seen in the positive ion spectrum is due to
the phosphate salt of the drug, [2H3.PO4-OH]+ at m/z 179. Several peaks are also seen in
the negative ion spectrum that can be attributed to phosphate salts. P206} PO3', [2H3PO4-
H]', [2H3PO4-H3O]', and H2P04' are seen at m/z -158, -79, -195, -177, and -97,
respectively. Peaks indicative of chloroquine are also seen in the negative spectrum; loss
of an ethyl group is seen at m/z -290. All of these fragment ions are consistent with the
presence of the active ingredient, chloroquine, in the sample. Figure 2- 6b shows the
dual-polarity mass spectra from ﬁve individual particles containing chloroquine.
Although there is some variation in the intensity of the peaks, the characteristic peaks are
present in all spectra, and chloroquine can be easily recognized as a component of the
tablet. The high quality and reproducibility of these mass spectra allow accurate rules
trees to be created, which, when used with the software developed in house, provide
automated identiﬁcation of the compound with none of the alarm ﬁles for the other drugs

tests provided a positive alarm.

27

 

2-9-3: Clemastine Fumarate

An allergy medication containing 1.34 mg of Clemastine fumarate was then studied. An
equivalent value of 1 mg of Clemastine free base was present in the tablet. The average
aerodynamic diameter of 999 particles was 1.82 pm (RSD 45.4%). Figure 2- 7shows the
dual-polarity mass spectrum obtained ﬁom a single particle of the tablet. This particle

was 2.09 pm in diameter.

 

  
   

 

 

 

 

 

250- - -
A \ [igiizcmmcng- or [F-H20]‘ O‘CH2)2C4H7NC"3]+
3200- W4CH2hC4H7NCH3l' [C4H7NCH3IT/ [M or M+H1+
E 150- CH3C0(CH2)2]'
ﬂ ' +
5‘100- [M-O(CH2)2C4H7NCH3]‘ ' [NI-0(CH2)2C4H7NCH31
‘3 -195
3 50 - i went

0 .. she-Lg] _'_ _ .A ”MAL .

' ' D I l T
-200 -100 0 100 200 300

Mass-to-Charge Ratio

Figure 2- 7: Average mass spectrum of a single particle dislodged from a Clemastine
tablet (aerodynamic diameter 2. 09 pm).

Many ﬁ'agments are present, owing to the fragmentation of the bonds connecting the 3
rings of the molecule. A series of ﬁagments attributed to the consecutive breakage of C-
C bonds on the alkane chain connecting the pyrrolidine ring to the molecule are present:
[C4H7NCH3]+, [CH2C4H7NCH313 [M'(CH2)2C4H7NCH3]3 [0(CHz)2C4H7NCH31+ and [M'
()(CH2)2C4H7NCH3]+ and [M-H-O(CH2)2C4H7NCH3]', at m/z 84, -98, -231, 128, 215,
and -215, respectively. The fragment at m/z -71 is formed by the removal of both the
aromatic rings and the pyrrolidine ring, [CH3CO(CH2)2]’. The chlorine atom is also

removed from the molecule, leading to fragment ions at m/z -35, Cl', and m/z 308,

28

 

assigned as [M-Cl]+. A peak assigned as a molecular ion is seen in the positive spectrum;
however, limitations in the data acquisition range prevent accurate mass calibration in
this region, so it is unknown whether the peak is at m/z 344, assigned as the [M]+ ion, or
m/z 345, due to [M+H]+. This peak, as well as all peaks due to ions containing the
chlorine atom, shows a consistent and appropriate isotopic distribution. A peak due to
the fumarate salt is also seen in the negative ion spectrum: [F-H]' at m/z -115. An
additional peak at m/z -98 may be due to [F -HzO]', although this is more probably due to

the Clemastine fragment described above.

2-9-4: BenadrylTM (Diphenhydramine)

Next, an antihistamine and sedative BenadrylTM tablet containing 25 mg
diphenhydramine was studied using SPAMS. The average aerodynamic diameter of 999
particles was 1.81 pm (RSD 80.6%). Figure 2- 8 shows the mass spectrum of an

individual particle generated from a single tablet.

 

   
  
   
 

 

 

 

 

 

250‘ 0' [M+H]+
; 200 ‘ [(éH3hNCHf \ |\" [M-(CH3)2N(CH w )1()]+
3 I (CH3)2N(CHz)2]+ ' ‘ ‘ + [2M+K]+?
3' 150‘ - [( CH ) N(CHI) 01+ [M+(CH3)2NCH2]
E0 100‘ / 3 2 2 2 [TVI+(CHs)2N(CHz)2]+
E 50 - [TrtittC6Hs)2CH]+
0 _.
0 100 200 300 400 500
Mass-to-Charge Ratio

Figure 2- 8: Single particle aerosol mass spectrum of an individual diphenhydramine
particle (aeroaynamic diameter 1.14 pm).

29

 

ting

A»..;.-a

\ -" -
1 ’\ .
a. .

.
us. «9

.L"

Di. ......‘.__

 

 

, I
J a
l
w." ‘7 r
e
" i'_
.

_. .~
‘-s ' ..
‘u.

' \
i
1

. n
m 4"! .‘

a‘L .
~-

i'

Minimal structural information is present in the negative ion spectrum; only the fragment
attributed to [CN]' is identiﬁed. A peak attributed to Hle is also seen, with the
corresponding chlorine isotope ratio at m/z -71, -73, and -75. However, many peaks
indicative of diphenhydramine are present in the positive ion spectrum The most
prominent peaks are assigned as loss of the side-chain from the parent molecule:
[(CH3)2N(CH2)2O]+, [(CH3)2N(CH2)2]+, and [(CH3)2NCH2}+ at m/z +88, +72, and +58, as
well as [M-(CH3)2N(CH2)ZO]+ (m/z +167); the latter three have been previously reported
using GC-MS (+EI),54 and LC-MS—MS (+C1D).55 Peaks corresponding to the addition of
these fragments to the molecular species as well as the protonated intact molecule are
also present: [M+(CH3)2NCH2]+ (m/z +313), [M+(CH3)2N(CH2)2]+ (m/z +327),
[M+(C6H_<,)2CH]+ (m/z +422) and [M+H]+ (m/z +256). Two peaks separated by 38 mass
units are seen in the high mass region of the positive mass spectrum. These peaks have
been tentatively attributed to the protonated dimer and potassiated dimer, which would
yield m/z +511 and +549, respectively, but the mass calibration at these high masses is
unreliable, so ﬁnal assignments cannot be made. The peaks seen in the mass spectrum of

diphenhydramine allow for the positive identiﬁcation of diphenhydrarrrine in the tablet.

2-9-5: Loratadine

An antihistamine tablet containing 10 mg of loratadine was analyzed using SPAMS. The
average aerodynamic diameter of 999 particles was 1.67 pm (RSD 80.9%). A mass
spectrum of an individual particle generated from the single tablet is shown in Figure 2-

9.

3O

Inn Sign"! (a.u-)

a.
- F a
.2‘x‘
.. 5
I
u. ‘
‘i .
“31'“...
Ada.

. “ \
Nib “\

 

 

 

 

 

 

 

232504 C- [CN]- 10 2M+H-2Cl]+[2M—+H-c1]
é ZOO‘M-OCHOCZHs-C4H3N-C1T 0 [M-C4H8N02]'[M +H1+
Ta 150— '59 690 710 730 750 770
a -71 30
9100— +
a: +1. 10‘ [C3H6021 245 [M+H-CI]+ .
E 50“ I ; /[C3H5N02]:l 21 I
0_ . i . I I
-200 -100 0 100 200 300
Mass-to-Charge Ratio

Figure 2- 9: Single particle aerosol mass spectrum of an individual loratadine particle
(aerodynamic diameter 3. 95 ,um). Inset shows high mass peaks due to loratadine dimers.

The negative ion spectrum shows several peaks characteristic of loratadine, including
those assigned as the ﬁ'agments [M-OCHOCsz-C4H3N-Cl]’, and [CN]', at m/z -221 and
-26, respectively. The positive ion spectrum also shows peaks due to several fragments,
including [M-C4H3NOZ]+ at m/z +280; however, more signiﬁcantly, peaks at m/z +348
and +383, assigned as [M+H-Clj+ and [M+H]+ are seen. The [M+H1+ has been
previously reported.”’ 25 The inset in Figure 2-9 shows the high mass region between
m/z +680 and +780. Although the intensity is low, signiﬁcant peaks are seen with
sufﬁcient signal-to-noise ratios at m/z +695 and +730, possibly due to loratadine dimers
[2M+H-2Cl]+ and [2M+H-Cl]+. These peaks are the highest mass peaks that have been

clearly resolved on the instrument used in this study.

2-9-6: Generic Ibuprofen

Figure 2- 10 shows the mass spectrum of an individual particle generated from a single

tablet containing ibuprofen (200 mg). Although 999 spectra were obtained to ensure the

31

consistency of the data recorded, the single particle spectrum is representative of the data

obtained and was randomly selected.

 

250— [MW -89 [M-COOHF/
2001 \\[M-C00H1' [C6H4C3H6lﬂ [Mr

150‘ [M-C5H101'\

    

[C6Hs]+

 

[on Signal (a.u.)

 

 

 

 

 

 

-400 -300 -200 -100 0 100 200
Mass-to-Charge Ratio

Figure 2- 10: Single particle aerosol mass spectrum of a randomly selected individual
ibuprofen particle (aerodynamic diameter 1.21 am). Note the presence of the [M ]+ and
[M-Hf peaks to allow for simple identification

The average aerodynamic diameter of the 999 particles was 1.83 um (RSD 65.3%), again
with the large size distribution due to the method of particle formation. The aerodynamic
diameter of the particle displayed was 1.21 pm. The negative ion spectrum is dominated
by peaks at m/z -205 and -161, due to [M-H]' and [M-COOH]', respectively.” 26’ 3' A
peak at m/z -411 is attributed to the deprotonated dimer, [2M-H]', also seen by Williams
and Scribens using DESI with a quadrupole-TOFMS.26 The positive ion spectrum
contains many peaks, including those due to the [M]+ ion as well as peaks due to the loss
of -OH, and -COOH21 (m/z +206, +189, +161). A slight peak due to the [M+H]+ ion is
also seen in some mass spectra; however the [M]+ ion does dwarf this peak in all the
spectra. The pseudomolecular ion of ibuprofen was also not seen in studies using LC-
ESI-MS for the analysis of several drug samples. The presence of the [M]+ and [M-H]‘

peaks allows for the simple identiﬁcation of ibuprofen in the sample using either the

32

positive or negative ion spectrum; the dual-polarity nature of the data provides additional

conﬁrmation of the active ingredient.

2-9-7: Generic Pseudoephedrine and Phenylephrine

Several decongestant tablets were analyzed using SPAMS. Pseudoephedrine is a
decongestant commonly sold under the brand name of SudafedTM. Since the enactment
of the Combat Methamphetamine Epidemic Act of 2005, the quantity of pseudoephedrine
purchased by an individual has been restricted to no more than 3.6 g per day and no more
than 9 g in any 30 day period due to its use as a starting material for methamphetamine
synthesis.56 This legislation has increased the promotion and availability of alternative
decongestants, such as phenylephrine (commonly sold as Sudafed PET”). Figure 2- 11
shows a mass spectrum of a single particle generated from (a) a pseudoephedrine (30 mg)

containing tablet and (b) a phenylephrine (10 mg) containing tablet.

33

 

' -\ .\

A.oo--v-=-un..-.’.

‘

.?t

e D‘.‘

L . a u n... .C .\ . an
.i. .l. . . .-i \.
l , . . J . - . .qi .. 4 are
a . 1 .\ r .. 1
- . s h .5 Li
. a .14 .. . n. C .T I an
.e . p r. .. .1. 9.. ~ 1 LE
.. . . ‘1 i . . I
.14. UK a... m.» 1.“. v. h In \ . 4

«U
t .
W‘s
nlu
61*
.IU

he \1

LL.

 

 

 

      
        
 

 

 

 

 

250 - a.) Pseudoephedrine / C1" [CH3NHC2H5]+ [M+H]+
200 - C3H7N01' [M-OH]+
[C4H9N1’
'50 ‘ - [CNl' +
A100 q [M+C4H9N] Na
=3 l -80 [211/[+11]+
a _ K+ ‘
V 50 - o- 225 /
E [M‘ICI] .64
50 O .— , , _._,. _._¥ ~-v#22_-___-_ __________ _ ___ _ _‘__ I_ _ __n ____ A A
a 1 1
32001 [C311 NO]‘ [M-onr‘; :
”01 c H m- =1
100 J [M 2 8 I
t l
50 - 1 g
0 _______.—___—_—____——__._—______
-200 -150 -100 -50 O 50 100 150 200 250 300
Mass-to-Charge Ratio

Figure 2- I 1: (a) Single particle aerosol mass spectrum of an individual pseudoephedrine
particle (aerodynamic diameter 1.08 pm) and (b) an individual phenylephrine particle
(aerodynamic diameter 0. 96 pm). Although similar in structure and mass, these two
compounds can be diﬁferentiated using SPAMS based on molecular and fragment ion
specres.

Molecular masses of pseudoephedrine and phenylephrine are 165.23 and 167.21 Da,
respectively. The average aerodynamic diameter of 999 particles of each drug was 1.81
pm (69.5% RSD) for pseudoephedrine and 1.23 pm (58.0% RSD) for phenylephrine.
The positive ion spectra of the two substances show similar fragment ions and losses
owing to the similar structures of the compounds. Peaks corresponding to the base peaks
in the NIST standard mass spectra of pseudoephedrine ([CH3NHC2H5]+, m/z 58) and
phenylephrine ([CH3NHCH2]+, m/z 44) are present in high yield in the current spectra;
however, peaks assigned to [M+H]+ and [M-OH]+ are also seen in both drug samples,

have been previously reported?”0

and provide more conﬁrmatory identiﬁcation of the
drug. Pseudoephedrine also shows evidence of dimer formation with [2M+H]+ (m/z

+331). The negative ion spectra of these two compounds again show similarities. Peaks

34

due to [M+Cl]‘ are seen in both compounds, as well as peaks attributed to [C3H7NO]'
(m/z -73) and [CN]' (m/z -26). Several other ﬁ'agments are also present that contribute to

the signature of each drug.

2-9-8: BayerTM Aspirin versus CertiﬁedTM Aspirin

SPAMS can also be used to identify differences between brands of tablets containing a
single drug. Although intended to have the same medicinal effects, differences in
manufacturing processes, equipment, and inactive ingredients in the tablets impart a
different chemical signature to the tablets. Figure 2- 12 compares three regions of the
mass spectra obtained from two tablets containing aspirin: the average mass spectrum of
999 particles dislodged from a BayerTM brand tablet (325 mg aspirin) and a CertiﬁedTM

brand tablet (81 mg aspirin).

35

- u
.\

.A....uav-=.

. D
1II\
in. p.
u d I o a
\ .

.i. 1 -\.

crud-I an...

<
n g .

~.-.:v-=-ui

t; .I

Q .

‘1. sun'-

\ t u u k \ .II1- w ‘A . IIW as ~ .\ rt 1’1\
\ n . 1‘ a u . \. .L. {II F” .1?

.111 . . 1 . . (1 .114
A.=.=v~o-:u‘.’. nun-h 1.1.,111 W

. H .
.,. u 1.4
.5 .K r:
rIJ. linil

 

_ a.) Bayer Aspirin

 

 

 

In

 

_ b.) Certified Aspirin

 

 

 

 

 

 

 

I.
It

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\ . 1.....1 Id
0 _ I I I f I T T I I I
-600 -550 -500 -450 -400 -350 ~300 -250 -200 -150 -100
5 Mass-to-Charge Ratio
4 c.) Bayer Aspirinl
3 -1
2 .4
’1‘ 1
=3
e:
:7 5 311 (1.) Certiﬁed
g 44 Aspirin
.29 3 4 143
m
339
S 2 . 1
—
1‘ W W
0 l I I I I I I I I I
160 1 80 200 220 240 260 280 300 320 340
7 Mass-to-Charge Ratio
1 e.) Bayer Aspirin
0.8
0.6
’1‘
5. 0.4
a AIL
.3 0.2 W _ Nev “ . -
g 2,...
.9 O f.) Certified Aspirin
‘0 0.8‘
S
1—1 0.6-
_ 481 513523 600
0.4 567 611 628 656
0.2“
0 r l l r I r
400 450 500 550 600 650
Mass-to-Charge Ratio

Figure 2- 12: Average mass spectra of 1000 particles of BayerTM brand aspirin (a, c, e)
and CertifiedTM brand aspirin (b, d,f) in three separate mass regions. Note the presence
of additional peaks in the CertifiedTM Brand aspirin, attributed to ﬁllers and impurities in

the sample that can be used to distinguish the two brands.

36

’1‘ J" ' ”7.:
" 3&383 . t u

it.»

1 1

.0. ~ ‘0'-
'8 a

1'... ‘. Sun.
I

" ‘ ",'.3 ma,
wh"..~ 5 bag
.

:0; II \t “’3

I63 "L. .,‘,\C‘

'96}-

~ \1,‘

S - . _‘

‘u'q‘ 1y I.
. ‘ y"

‘02" 'v 6
‘Lu. .,

‘ -'~ J.‘ \

~13.

I'vu"?
>.\t g \~‘._
"5311‘

'3-.,'
w‘ "A‘ '-
““‘-3 iii

 

 

 

The peaks previously discussed (Figure 2- 5) as being indicative of aspirin are present in
both spectra; however other features in the spectra differ. The labeled peaks in the ﬁgure
indicate peaks (S/NZS) present only in the CertiﬁedTM brand tablet. In fact, these peaks
are not present even in single particle spectra of the BayerTM brand aspirin sorted for that
particular mass. Although small in amplitude, the presence of these peaks can be used to
qualitatively distinguish between these two brands of aspirin. These peaks are
hypothesized to be due to fragments originating from an unknown ﬁller material in the

tablets.

2-9-9: Tylenol Extra StrengthTM (Acetaminophen)

Many varieties of tablets can be analyzed using SPAMS. Tablets with a harder shell
coating to slow tablet dissolution may take a longer analysis time, as the active ingredient
may or may not be present in this shell. Several over the counter and prescription
medications are coated with an enteric coating to aid in digestion and to ensure the tablet
reaches the appropriate body compartment before releasing the active ingredient (e. g.
stomach or small intestine). If this coating does not contain the active ingredient,
analysis may take a longer time; particles containing the active ingredient must be
dislodged from the tablet, so many spectra may be obtained that do not contain the active
ingredient, formed ﬁom this coating. One such medication studied was Tylenol Extra

StrengthTM (EX) geltabs, containing 500 mg acetaminophen. Figure 2- l3shows (a) the

average mass spectrum obtained from 799 particles (da=2.33 um, RSD 52.7%) dislodged

from 3 Tylenol EXTM tablet and (b) the average mass spectrum obtained from the 50

37

 

particles having the highest mass spectral signal at m/z 152, corresponding to the [M+H]+

ion ofa

Ion Signal (a.u.)
9.? i

l

.5
O

N
O
i

cetaminophen.

 

_ a.) n=799

A4

b.) 11:50 '

[M-CH3C(O)]'
- \
[NI-CH3-01'

[MOHI'
-192

 

O
l

~200

[CH3COJ'\ -26
-60

-72 Cl‘
96.84

I O-
1 I

I I 1 1

 

-100 -50

Na

 

 

C2H3]+

1c [TﬂOMF

[C5H4l+
/

481

50

 

123

100

[M+H-CH3C(O)]+

[Mm]+
1M+Na1+

_i 1__ -
I I
150 200

 

 

 

 

 

Mass-to-Charge Ratio

Figure 2- 13: Average mass spectrum from 50 particles dislodged from a Tylenol EX
tablet. The 50 particles shown were selected by sorting 799 particles for the highest peak
area at m/z 152.

Although peaks due to the active ingredient are seen in the averaged spectrum (n=799),
the intensity is low, impeding the facile identiﬁcation of the substance. However, after
sorting the spectra for a particular peak known to be caused by acetaminophen, the active
ingredient is easily distinguished ﬁom the background material. Thus, even coated
tablets can be analyzed successfully with SPAMS. Other ions indicative of
acetaminophen are also seen in the spectra. The sodiated ion, [M+Na]+ is seen at m/z
174, as well as the fragment [M+H-CH3C(O)]+, caused by cleavage of the C-N bond, at
m/z 109, [CH3C(O)NH]+ at m/z 58, is assigned as a fragment from the aeetamide group,

and, and [C2H3]+ at m/z 27 is likely a fragment of the aromatic ring. The negative ion

38

 

W~‘uu

Lu".

I
D
.-

1;
n
”o:

- .1,
55h.

Lu
v
”'11". .
”1. 111M“
.. .b
.._‘
-' C-
:i;'-TJ «-
‘Wgs ,‘
9 ""‘
a
5.”. - T .
s.’ c 1-l
.l““l 4

 

 

 

mass spectrum also has identifying peaks: [M-H]‘, [M-OH]', [M-O-CH3]', [M-CH3C(O)]',

and [C4]’ at m/z -150, -l34, -120, ~108, and -48, respectively. Other peaks, attributed to
anionic carbon clusters (C1,), are seen in the negative spectrum at -96, -84, -72, -60, —48,

and -36.

2-9-10: Tylenol ColdTM (Acetaminophen + Diphenhydramine)

SPAMS can also be used to identify multiple ingredients in a single tablet. Several

multicorrrponent tablets were analyzed and presented below. Figure 2- 14b shows the
average mass spectrum obtained from 999 particles (da= 1.57 pm, RSD 44.9%) dislodged

from 3 Tylenol ColdTM tablet.

39

 

 

 

        
 
   

 

_ a.) Benadryl
200 58 Diphenhydramine
150‘ _71 l 67 n=999
100~ ‘73

256
I b.) Tylenol Col
Diphenhydramine+ Acetaminophe
' n=99

I
I
A A A

' c.) 131E111» EX

 

-- .—--_..----

 

 

 

g I I I I E I E I Acetaminophen
150- = i i i ' i , i 5 1 n=25
100. ' 5 I I i E '
' 5 : 152
50" 3
0_ 134 AL -- g .
-200 -150 -100 -50 0 50 100 150 200 250 300

Mass-to-Charge Ratio

Figure 2- 14: Average mass spectra of particles dislodged from a.) a Benadryl tablet
containing diphenhydramine (n=999), b.) a Tylenol Cold tablet containing
acetaminophen and diphenhydramine (n=999), and c.) a Tylenol EX tablet containing
acetaminophen (n=25). Dashed lines indicate peaks in the Tylenol Cold sample that can
be attributed to either acetaminophen or diphenhydramine.

Tylenol ColdTM contains 500 mg of acetaminophen and 12.5 mg of diphenhydramine.
Diphenhydramine is the active ingredient in BenadrleM, so the average mass spectrum of
999 particles from a BenadrleM tablet is shown (Figure 2- 14a) as well as the average
mass spectrum of 25 particles from a Tylenol EXTM tablet (Figure 2- 14c). Dashed lines
indicate the peaks in the Tylenol ColdTM spectrum that have been attributed to either
diphenhydramine or acetaminophen. Numerous peaks are present due to each compound;
especially of note are the [M+H]+ ions ﬁ'om each drug, at m/z 152 for acetaminophen,

and m/z 256 for diphenhydramine. The presence of two active ingredients does not affect

4O

 

 

the ability to identify these substances. The acetaminophen alarm ﬁle identiﬁed 953
particles as containing acetaminophen. Other peaks are also seen in the Tylenol Cold
spectrum that cannot be attributed to either active ingredient. These peaks are

hypothesized to be due to the additional ﬁller compounds in the tablet.

2-9-11: Tylenol PMTM — Caplet versus Tablet (Acetaminophen + Diphenhydramine)

Figure 2- 15 shows mass spectra obtained from two varieties of Tylenol PM — a caplet

and a tablet.

 

   
     

 
 
 
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200
III-193M“ PM Caplet 1\[M-H]‘ [CH3CONH]+ or [(CH3)2N(CH2)20]+ +
_n— - +H + "— +H-CH CO
150 [M-H+CH2C(O)]' [MCH3C0] [cnzcoy [M I 1(c1[i:)21~1(c1132)zo]]+
[MOHI' '72 (D) [D-(CH3)2N CH2)201+
AIOO‘m-mmcnpzr \ -60 A) [Ni-0111+
- - +
3W \ -264(M) \ [Mm‘lymnt
:: _ _ [D+H]+ [n+xr‘
a - 1 AI/[D’EILJ “1111.1. __ _ _ 216 3, / .
EB b.)'l‘ylenolPMTablet ' i ii i 5 1 1
11 11 1 11 11111111111
1..-,,.,,-1 1 1 1 9391' 111 1 1 1 11 1
501 \E 408 I I III II II I III
I ll [ALI ‘ '1' k. i? -
-300 -200 -100 0 100 200 300
Mass-to-ChargeRatio

Figure 2- 15: Average mass spectra obtained from 997 particles dislodged from a) a
Tylenol PM caplet and b.) a Tylenol PM tablet. While many peaks indicative of the
active ingredients, acetaminophen and diphenhydramine, are seen, other peaks are also
present in either sample, allowing diﬂerentiation between the two formulations.

41

 

Both formulations also contain both acetaminophen (500 mg) and diphenhydramine (25
mg). Again, peaks attributed to the acetaminophen: [M+Na]+, [M+H]+, [M-OH]+, [M+H-
CH3O]+, [CH3C(O)NH]+, [CH3CO]', [M+CH3CO]', [M-CHg-OI, [M-OH]’, [M-H]' and
[M-H+CH2C(O)]‘ are seen at m/z 174, 152, 134, 109, 58, -48, -108, -120, -l34, -150, and
-192 respectively and those attributed to diphenhydramine, [D+K]+, [D+H]+, [D-
(CH3)2N(CH2)20]+, [(CH3)2N(CH2)2O]+, [(CH3)2N(CH2)2]+, [(CH3)2NCH2]+, and [D-H]',
at m/z 294, 256, 167, 88, 72, 58, -254, and -271, respectively are seen in both samples.
The tablet formulation also contains peaks at m/z -271 attributed to [D+O]', -208
unattributed, and -93, which is also unattributed but also seen in a Benadryl tablet,
containing only diphenhydramine. The caplet formulation contains additional peaks at
m/z -264, -72, and -60 which were also seen in the Tylenol EX sample attributed to
acetaminophen and m/z 310, which was seen in the Benadryl tablet. This variation in
peaks present allows the different formulations of the same drug produced by a single
brand to be distinguished, highlighting another ability of SPAMS. The alarm ﬁles for
both acetaminophen and diphenhydramine were run again all particles. 853 particles in
the caplet were identiﬁed as containing acetaminophen, and 390 were identiﬁed as
containing diphenhydramine. In the tablet, 898 particles were identiﬁed as containing
acetaminophen, while 584 particles were identiﬁed as containing diphenhydramine.

Thus, many particles must contain both drug components.

42

————--._ _—.——-. F.

 

 

2-9-12: Advil Cold and Sinus” (Ibuprofen + Pseudoephedrine)

Figure 2- 16 shows an average mass spectrum of 50 particles from a multicomponent

tablet, Advil Cold and Sinus”. This tablet contains 200 mg ibuprofen and 30 mg

pseudoephedrine. The 50 particles shown were chosen by sorting 797 particles (da=l.68

um, RSD 48.6%) by peak area at m/z 166, and selecting the top 50 particles. While
many peaks have been attributed to either ibuprofen (l) or pseudoephedrine (P), including
[l]+, [P+H]+, and [I-H]' at m/z 206, 166, and -205, many additional peaks are seen that
have not been attributed. Again, these peaks, such as those at m/z -264, -97, 56, and 64

are thought to be due to the unknown ﬁller compounds in the tablet.

 

140 . . _79
at 5" [1-111' 4 9 I [l-C(CH3)C02H-CH3]+
leOn sorted b m/z 166 -

K+

N x
= 804 -88 -45

a“ 60- -97 [Pm] [Pm]+
g 40‘ I \ [11+ [1+Na]+

0‘ L} A

ﬂ: I I
-300 -200 -100 0 100 200 300
Mass-to-Charge Ratio

 

 

 

 

 

 

 

 

 

 

Figure 2- 16: Average mass spectrum obtained from 50 particles dislodged from an Advil
Cold and Sinus tablet. The 50 particles were selected by sorting 797 particles for the
highest mass area at m/z 166.

2-9-13: Tylenol Severe Congestion” (Acetaminophen + Dextromethorphan +
Guaifenesin + Pseudoephedrine)

A single Tylenol Severe Congestion” tablet containing four active ingredients:
acetaminophen (325 mg), dextromethorphan (15 mg), guaifenesin (200 mg), and

pseudoephedrine (30 mg), was analyzed. Figure 2- 17a shows the average dual-polarity

43

 

mass spectrum obtained from 1000 particles (da= 1.46 pm, RSD 49.0%) from this tablet.

Figure 2- 17b-e show spectra of the individual components for comparison.

44

 

200 ~
150 -
100 -1
50 —

[on Signal (a.u.)

 

 

 

258 -
200 -
150 4
100 J

50 -

Ion Signal (a.u.)

 

 

 

0*

 

 

'\
0 I1 1
l*.

a) Average Spectrum
n=1000

A

LllLainJL-

 

 

b) Single Particle] of
Pseudoephedrine
Only Tablet

I

 

 

 

ll lkA a l J‘LL __ ‘_ .L-‘h. ;_

I I

 

250 -1
200 -1
150 1
100 4

50 4

Ion Signal (a.u.)

100‘

50-

Rel. Intensity (%)

 

 

1L

 

50-

Rel. Intensity (%)

 

 

 

 

 

 

I
-300

I

-200

-100

I .1 .IL ml- .I. .I...1 1....

 

 

 

 

[kW-III -~

 

 

T T
c) Single Particle of
Acetaminophen
Only Tablet
*

d) NIST Reference
Spectrum (El) of Pure
Dextromethorphan

A

e) NIST Reference
Spectrum (El) of Pure
Guaifenesin

 

0 100

Mass-to-Charge Ratio

200 300

Figure 2- I 7: a) Average mass spectrum of 1000 particles dislodged from a single

Tylenol Severe

Congestion

tablet containing pseudoephedrine,

acetaminophen,

dextromethorphan, and guaifenesin as the active ingredients, b-c) mass spectra obtained
from a single particle dislodged from a tablet containing only pseudoephedrine or
acetaminophen, respectively, as the active ingredient, d-e) concatenated standard NIS T
mass spectra of dextromethorphan and guaifenesin, respectively. Single component
tablets containing these ingredients were not readily available.

45

 

 

 

. e
'1") ‘

. not .5‘

.14.“...-
. -.

rh$~ﬂ~euﬂ

I
l \- 41 1
>5. 5..

04 A
I. ‘L .
1”" “I"

7“
.J_-I"

 

.,

IDQDT1.‘3,.
. ls.“

1

1.,

‘51 1‘3' \
MI ‘1‘3:

A tablet containing pseudoephedrine as the active ingredient as well as a tablet containing
acetaminophen as the active ingredient were analyzed to create the spectra shown in
Figure 2- 17b—c; Single component tablets containing dextromethorphan or guaifenesin
were not available during the study, therefore standard 70 eV electron ionization (El)
spectra from the NIST reference library are shown (Figure 2- 17d-e), and concatenated to
display all ions as potentially-formed positive and negative ions. Note that peaks
attributed to the molecular or fragment ions of each component are present in the average
(Figure 2- 17a). Pseudomolecular ions of each compound are present and indicated by
the arrows: [M-H]' and [M+H]+ for acetaminophen, [M+H]+ for dextromethorphan,
[M+H]+ and M' for guaifenesin, and [M+H]+ for pseudoephedrine. After analysis of
1000 particles, the mass of the tablet was reduced by only 0.025%, and. all markings were

clearly visible on the tablet.

The 1000 particles were then processed through the rules trees for all the drug
compounds in this study. 953 particles produced a positive identiﬁcation for
acetaminophen, 571 for pseudoephedrine, 907 for dextromethorphan, and 804 for
guaifenesin. No false positives were obtained for aspirin, loratadine, phenylephrine,
ibuprofen, or chloroquine. This averages to 3.2 drug compounds per particle. The
heterogeneous nature of this multi-ingredient tablet allowed for varying combinations of
drug compounds to be detected in single particles and demonstrates that the ingredients

of such a heterogeneous tablet can be rapidly identiﬁed using SPAMS.

46

'lDll Slullul (u.ea_. )
I I

2-9-14: Multiple Tablets: Phenylephrine and Aspirin

Next, SPAMS was tested to determine if multiple tablets could be simultaneously
analyzed. . A single phenylephrine tablet (10 mg) and a single generic aspirin tablet (81
mg) were placed in a single vial and introduced to the SPAMS system. Figure 2- 18
shows the average of 798 mass spectra obtained from this sample (Figure 2- 18b) and a
spectrum ﬁ'om 249 particles of phenylephrine alone (Figure 2- 18a) and aspirin alone

(Figure 2- 18c).

 

 

 

 

 

 

   

 

 

 

 

 

 

 

a.) Phenylephrine 3 5 44
80‘ n=249 '
60~ 158
-71 '
40-I -201 157
20'1 -I66 '122 150
G L A A. J. . LA; #2 _ __ A _
I I I I I l T I
A e o l I : 1 l :
€120- b.)Asprrin +Phenylephrrrrei 5 5 5 5 5 1
51001 =7” 5 :: 5 5 E : : 5
5,801 1 11 1 1 1 1 1 1
m 60— I ii I I I I I I
= 40- 1 =1 e a 1 ‘ :
'-' 20- ' E: 5 ' J :
20 .L AAitJL} l I 3AM i'uJJJ .l._? L 'L‘ A A L l
ciAspirin 55 : : 51137 55 i 5
15a r249, 1 g 1 -1191 i; 1 1
t i i i : I 55 163 1
we a s 1 s 1 s a: =
50— E I 4725 I I 1215 i
i 1 I 229 'B 11 I
0 .3584:314 2 6 L ﬁAJ .A M AIAJAA LL- 1; I L It A L? 135.. .. 238. A L L I
-300 -200 -100 0 100 200 300

Mass-to-Charge Ratio

Figure 2- 18: Average mass spectrum obtained from a.) 249 particles dislodged from a
phenylephrine tablet, b.) 798 particles dislodged from a single phenylephrine tablet and a
single aspirin tablet placed in the same glass vial and sampled simultaneously, c.) 249
particles dislodged from a single aspirin tablet. Note that both phenylephrine and
aspirin peaks are seen in the mixed tablet spectrum.

47

 

..
‘u .u' .w ‘J
in. 1A.»

0 ‘l
”*3” l 'H\

-n'....'il.| i‘. i..-
.

' I ‘ 1
q.« ..
" z . 3‘ “I "'
”Evy-nub .
o

  

-3
-l... .m...\:\
.

. . U I
\;-O.0‘J‘ N‘ .
‘H....i.»L ‘\ ‘I

l t _
‘GJC ‘ '- 8“ . Q
-- us. we.» h

a .
n~ . -
. ‘5.“ "' o
l" -;H_‘I | 5
l
. ‘5 "I
3‘ ‘ 3‘.
‘ ‘ ‘ ~\
‘ 1
l
-u‘>5. .‘v— .
a A 5.
:\ ‘5' I
531‘" V“ .‘
~t. , >4‘
. "~ 15‘
‘

 

‘5'»
1:5,; 5 ~
kiSL) 1D
a >§
hi 'v
p
‘5‘5‘e ’.
5 «‘17:
"1A. '
r37."

The presence of each component is seen in peaks attributed to the molecular and
fragment ions of each compound, all of which have been previously identiﬁed and are
indicated by the dashed lines. When the data are analyzed on a particle by particle basis,
most particles can be identiﬁed as containing only one drug, but approximately 5% are
identiﬁed by the rules trees as containing both compounds. It is thought that these are
indeed individual single particles formed by a collision inside of the glass vial and are not
due to several particles being concurrently desorbed/ionized in a single pulse, as the
probability of more than one particle being present in the D/I laser beam when it is ﬁred
is 0.5%, based on a sample containing 1000 particles/ml. and sampling rate of 1 L/min.“
This hypothesis is addressed further in Section 2-9-15. Visual analysis of the tablets aﬁer
sampling reveals small particles present on the outside of the tablets, thought to be
residual particles from the other tablet in sample. Particles formed during tablet-tablet

collisions may contain fragments from both tablets.

2-9-15: Multiple Tablets - Separate Vials: Phenylephrine and Aspirin

In order to test the hypothesis of tablet-tablet collisions causing the multi-component
particles, 2 tablets were simultaneously introduced to the instrument using 2 separate
sampling vials. A single tablet of loratadine was placed in a modiﬁed glass vial, while a
single tablet of pseudoephedrine was placed in a second vial. Both vials were connected
to the SPAMS inlet using a y-shaped connector for the conductive tubing. By placing the
tablets in separate vials, tablet collisions are eliminated as a source of this phenomenon,
while maintaining all other parameters as consistent. Mass spectra were collected for 498

particles using this setup, and analyzed on a particle by particle basis. No spectra were

48

h.-

[I

1.3? "P? "3'
\s.. nus ek-

. I
n u 1 . .-
UII‘5 .- V

_—..‘-b¥ b. o‘- v
. .

 

‘1' .‘

.l-..

.
”I'D. ”.0

"' \-| L.

 

seen that containing peaks indicative of both drugs, which supports the hypothesis of

mixed drug particles forming within the sampling vial.

2-9-16: Multiple Tablets: Phenylephrine, Aspirin, Ibuprofen, Loratadine,
Pseudoephedrine

It is also important to ensure that drug components in a bulk sample containing more than
1 or 2 types of tablets could be accurately detected and identiﬁed. Thus, a greater
number of tablets, 5, were placed in a single vial: aspirin (81 mg), ibuprofen (200 mg),
loratadine (10 mg), phenylephrine (10 mg), and pseudoephedrine (30 mg) and
simultaneously analyzed. Figure 2- 19a shows the average dual-polarity mass spectrum

of 799 particles.

49

 

 

 

 

 

 

 

 

    

 

 

     
    

 

 

 

20
a) Average Spectrum
15 .. ‘ . I n=799
105 O ‘ -
* I
O t .0 O
5 J I
‘ O
0 A I I I I r r
25° ‘ b) Single Partic
200 4 identiﬁed as containin
Pseudoephedrin
150 - I
100 -
50 «
o .
25° ‘ c) Single Partici
200 - Identified as containin
Ibuprofen
Q 150 "
:I
5' 100 1
:
a 50 -
a 0
g 250 - d) Single Particle
"' 200 - identified as containing
150 r
100 ~
50 -5
0 AM- - _-_ -_
250 " e) Single Particle
200 ‘ identified as containing
550 q Phenylephrige
100 ‘
50 " A AL . a
0 1 I I I I I I '— I
250 - DSingie Particle
200 - identified as containing
A irin
150 - Sp ‘

 

 

 

 

 

—300 -200 -l 00 0 100 200 300

Mass-to-Charge Ratio

Figure 2- 19: a) average mass spectrum obtained from 799 particles introduced from a
complex sample containing 5 single active ingredient tablets including aspirin, ibuprofen,
loratadine, phenylephrine, and pseudoephedrine, b-ﬂ single particles selected from the
799 particles collected that were identiﬁed as containing primarily pseudoephedrine,
ibuprofen, loratadine, phenylephrine, or aspirin, respectively.

50

The presence of each component is seen in peaks attributed to the parent and fragment
ions of each compound, all of which have been previously identiﬁed. Figure 2- 19b-f
shows single particles of the 799 particles analyzed. These particles were selected as
representative of each drug component. Again, many particles do contain only one drug,
but many others seem to contain remnants of several of the drug tablets. This is again

hypothesized to be due to the tablet-tablet collisions forming mixed component particles.

The alarm algorithm also successfully identiﬁed the components of this drug mixture.
534 particles were identiﬁed as containing aspirin, 606 as loratadine, 13 as
phenylephrine, 581 as pseudoephedrine, and 206 as ibuprofen. No false positives were
detected for dextromethorphan, guaifenesin, acetaminophen, or chloroquine. This
averages to 2.4 identiﬁed components per particle, thought to be due to the tablet

collisions.

2-9-17: Emptied Vial

The prior contents of a vial can also be detected with this system. Aﬁer analysis, the ﬁve
tablets were removed from the sampling vial, and the empty vial reanalyzed. While the
number of residual particles dislodged from the empty vial was limited, over 500
particles were analyzed. Figure 2- 20b shows the resulting average mass spectrum from
500 particles. Figure 2- 20a shows 500 randomly selected particles from the vial
containing the ﬁve tablets for comparison. Although the ion abundances vary, all the
identifying peaks are still detected for each drug tablet. 328, 348, 7, 353, and 155

particles were identiﬁed as containing aspirin, loratadine, phenylephrine,

51

.n.‘

pseudoephedrine, or ibuprofen, respectively, with no false positives, again, indicating

formation of multi—component particles.

 

     
  
 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

20 a.) Average Spectrum
15 _ A O n=500
O
10 4 ’ ’
2 A
o 5 "
{3; A
a o-
.29 ' b.) Empty via
2 30 _ Average Spectru
3 ' n=50
20 -
10 " l
0 4 T I I L—_ I I f
-300 -200 -100 0 100 200 300

Mass-to-Charge Ratio

Figure 2- 20: a) average mass spectrum of 500 particles introduced from the complex
sample containing 5 single active ingredient tablets including aspirin, ibuprofen,
loratadine, phenylephrine, and pseudoephedrine, b) average mass spectrum of 500
residual particles introduced from the emptied sampling vial.

2-9-18: In Situ Sampling: Chloroquine Diphosphate and Loratadine

The SPAMS technique is also successﬁrl when used to sample tablets in situ, i. e. in their
actual commercial bottle. 11 chloroquine phosphate tablets remaining in their original
bottle were sampled by taping the modiﬁed vial cap to the opening of the jar. Mass
spectra of 100 particles dislodged from these 11 tablets were analyzed in 64 seconds
(ﬂuence=2.4 nJ/pmz); all spectra were equivalent to those shown in Figure 2- 6 and all
particles were identiﬁed by the rules trees as only containing chloroquine, with no false
positives. Finally, a single loratadine tablet was added to the eleven chloroquine

phosphate tablets in this container to simulate the presence of a single illicit tablet in a

52

container of licit materials. The loratadine tablet accounted for 1.2% of the weight of the
tablets in the container. Using this same sampling method, particles were again identiﬁed
as chloroquine, with mass spectra matching those in Figure 2- 6. However, a particle
identiﬁed as loratadine was detected 117 seconds after sampling began. Figure 2- 21a
shows the mass spectrum obtained from this single loratadine particle and Figure 2- 21b
shows the averaged mass spectrum previously obtained ﬁom 100 loratadine particles
sampled from a single tablet for comparison. All identifying peaks are present, especially
the [M-H]' and [M+H]+ ions which greatly facilitate identiﬁcation. The rules tree for

loratadine identiﬁed this single particle, with no false identiﬁcations for the other drugs

 

 

 

 

 

 

 

studied.
605a)srngIe Loratadine Particle from Bottle [m]- [M+H]+
50‘ . Cl +
40- [M-H-C2H50H] [M-C2H5-HCI]' [M-C4H8N02]
’7 30- / [Mczusoqowcug- [C3H602]+ 5
=1 [M-Hl' I + [M+H-Cl]+
% 20‘ [C4H7N02]'\ x+ ,lC3H5N021
g 10
.92." 60
(g b) Average of 100 Loratadine Spectra
a 50‘
~
40-
30‘
20‘
10‘
0 I I I F I ﬂ I
—300 -200 -l 00 0 100 200 300
Mass-to-Charge Ratio

Figure 2- 21: a) Mass spectrum of a single particle of loratadine detected in an original
bottle containing 11 chloroquine phosphate tablets, b) average mass spectrum of 100

loratadine particles previously collected for comparison.

53

 

2-10: Conclusions

As a technique that has previously proven successful for the detection of biological,
explosive, and chemical warfare agent samples, it was hypothesized that SPAMS would
also be an ideal detection system for drugs. A novel sample formation and introduction
system had to be developed for this application; the modiﬁed glass vial was an efﬁcient,
simple, and effective method for creating particles from solid samples due to their
collisions with each other and the sides of the glass vial, and the slight vacuum of the
SPAMS system allowed for the particles to be pulled into the system with no additional
power requirements. Many single component over-the—counter medications were
analyzed and effectively characterized using SPAMS, including acetaminophen, aspirin,
Clemastine, diphenhydramine, ibuprofen, loratadine, phenylephrine, and
pseudoephedrine. These tablets were analyzed without destroying other forensically
useful information, such as their shape, color, and markings. Chloroquine was also
identiﬁed in tablets, demonstrating the ability of SPAMS to be used for the identiﬁcation
of counterfeit drugs. Additionally, multi-component medications were also analyzed,
showing indicative peaks from all active ingredients. These studies also included
chlorpheniramine, dextromethorphan, and guaifenesin. Multiple single component
tablets were then simultaneously analyzed to show the ability of SPAMS to serve as a
general screening device. While many particles analyzed contained one drug, many also
contained multiple drugs, and ﬁirther experiments showed that this was caused by tablet-
tablet collisions in the glass vial, and not by multiple particles being simultaneously
desorbed/ionized in the source region. The study also highlights the ability of SPAMS to

detect drug compounds even after they have been removed from the container. This has

54

 

 

forensic applications to situations in which a suspected drug sample was discarded before
seizure, leaving only the emptied container as possible evidence. Enough residual
particles were present in the emptied vial to allow for the collection and analysis of over
500 particles. Proof of concept in situ analysis experiments were performed by attaching
the modiﬁed glass vial cap to a commercial chloroquine bottle. The chloroquine was still
effectively sampled and analyzed, reducing consumables and analysis time. Finally, a
single loratadine tablet placed in a bottle with multiple chloroquine tablets was
successﬁilly detected, even though it only made up 1.2% weight of the sample,
demonstrating that SPAMS is a useﬁrl technique for detecting a single illicit tablet in a
large batch of licit materials. SPAMS has been successfully demonstrated as a useful and
effective drug detection technique with over-the—counter medications, and it is
hypothesized that it will also be an effective system for sensitive and rapid illicit drug
detection. As such, SPAMS may be a powerful tool for border control and customs
applications to provide almost instantaneous identifying information about the

components in drug tablets.

55

I50 _‘ -'_'— ﬁi‘

 

 

11 .° References

)

Newton, P. N.; Green, M. D.; Fernandez, F. M.; Day, N. P. J.; White, N. J. The

Lancet Infectious Diseases 2006, 6, 602-613.

145—250.

(E5i)

(fi)

(53)

(f3)

(7)

(8)

(9)

(10)

ill)

(12)

(13)

(14)

Basco, L. K. The American Journal of Tropical Medicine and Hygiene 2004, 70,

Foreign Terrorist Organizations, Ofﬁce of the Coordinator for Counterterrorism,
Washington, DC, 2008.

Counterfeit drugs: Guidelines for the development of measures to combat
counterfeit drugs, Geneva: World Health Organization, 1999.

Azhagvuel, S.; Sekar, R. J. Pharm. Biomed. Anal. 2007, 43, 873-878.
Tanyanyiwa, J .; Hauser, P. C. Anal. Chem. 2002, 74, 6378—6382.

Borer, M. W.; Zhou, X. J.; Hays, D. M.; Hofer, J. D.; White, K. C. J. Pharm.
Biomed. Anal. 1998, 17, 641-650.

Bondesson, L.; Mikkelsen, K. V.; lLuo, Y.; Garberg, P.; Agren, H. Spectrochimica
Acta Part A—Molecular and Biomolecular Spectroscopy 2007, 66, 213-224.

Meza, C. P.; Santos, M. A.; Romanach, R. J. Aaps Pharmscitech 2006, 7, E1-E9.

Antonie, J.; Heath, E. Anal. Bioanal. Chem. 2007, 387, 1337-1342.
Hardt, J. Fresenius'Journal of Analytical Chemistry 2001, 371, 787-790.

Tseng, Y. L.; Shieh, M.-H.; Kuo, F.-H. Forensic Science International 2006, 15 7,
149-155.

Ma, W.; Liu, X. Y.; Pawliszyn, J. Anal. Biochem. 2006, 385, 1398-1408.

Hernando, M. D.; Heath, E.; Petrovic, M.; Barcelo, D. Anal. Bioanal. Chem.
2006, 385, 985-991.

56

 

 

15) Cahill, J. D.; Furlong, E. T.; Burkhardt, M. R.; Kolpin, D.; Anderson, L. G. J.
Chromatogr. A 2004, 1041, 171—180.
16) Takats, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Science 2004, 306, 471-
473.
(17) Cody, R. B.; Laramee, J. A.; Nilles, J. M.; Durst, H. D. JEOL News 2005, 40, 8-
12.
(18) Wood, M.; Laloup, M.; Samyn, N.; del Mar Ramirez Fernandez, M.; de Bruijn, E.
A.; Maes, R. A. A.; De Boeck, G. J. Chromatogr. A 2006, 1130, 3-15.
(19) Chen, H.; Talaty, N. N.; Takats, Z.; Cooks, R. G. Anal. Chem. 2005, 77, 6915-
6927.
(20) Williams, J. P.; Scrivens, J. H. Rapid Commun. Mass Spectrom. 2005, I9, 3643-
3650.
(21) Leuthold, L. A.; Mandscheff, J. F.; Fathi, M.; Giroud, C.; Augsburger, M.;
Varesio, E.; Hopfgartner, G. Rapid Commun. Mass Spectrom. 2006, 20, 103-110.
(22) Rodriguez-Cruz, S. E. Rapid Commun. Mass Spectrom. 2006, 20, 53-60.
(23) Weston, D. J.; Bateman, R.; Wilson, I. D.; Wood, T. R.; Creaser, C. S. Anal.
Chem. 2005, 7 7, 7572-7580.
(24) Cody, R. B.; Laramee, J. A.; Durst, H. D. Anal. Chem. 2005, 7 7, 2297-2302.
(25)

Williams, J. R; Patel, V. J.; Holland, R.; Scrivens, J. H. Rapid Commun. Mass
Spectrom. 2006, 20, 1447-1456.

(26) VanBerkel, G. J .; Ford, M. J.; Deibel, M. A. Anal. Chem. 2005, 7 7, 1207-1215.

(27) Gard, E.; Mayer, J. E.; Morrical, B. D.; Dienes, T.; Fergenson, D. P.; Prather, K.
A. Anal. Chem. 1997, 69, 4083-4091.

(28) Noble, C. A.; Prather, K. A. Environ. Sci. Technol. 1996, 30, 2667-2680.

57

—

 

9) Sinha, M. P. Rev. Sci. Instrum. 1984, 55, 886-891.

0) Davis, W. D. Environ. Sci. Technol. 1977, 11, 587-592.

51) Noble, C. A.; Prather, K. A. Mass Spectrom. Rev. 2000, 19, 248-274.

(32) Czerwieniec, G. A.; Russell, S. C.; Tobias, H. J .; Pitesky, M. E.; Fergenson, D. P.;

Steele, P.; Srivastava, A.; Horn, J. M.; Frank, M.; Gard, E. E.; Lebrilla, C. B.
Anal. Chem. 2005, 77, 1081-1087.

(33) Fergenson, D. P.; Pitesky, M. E.; Tobias, H. J.; Steele, P. T.; Czerwieniec, G. A.;
Russell, S. C.; Lebrilla, C. B.; Horn, J. M.; Coffee, K. R.; Srivastava, A.; Pillai, S.
P.; Shih, M.-T. R; Hall, H. L.; Ramponi, A. J.; Chang, J. T.; Langlois, R. G.;
Estacio, P. L.; Hadley, R. T.; Frank, M.; Gard, E. E. Anal. Chem. 2004, 76, 373-
378.

(34) Srivastava, A.; Pitesky, M. E.; Steele, P. T.; Tobias, H. J.; Fergenson, D. P.; Horn,

J. M.; Russell, S. C.; Czerwieniec, G. A.; Lebrilla, C. B.; Gard, E. E.; Frank, M.
Anal. Chem. 2005, 77, 3315-3323.

(35) Steele, P. T.; Srivastava, A.; Pitesky, M. E.; Fergenson, D. P.; Tobias, H. J .; Gard,
E. E.; Frank, M. Anal. Chem. 2005, 7 7, 7448-7454.

(36) Steele, P. T.; Tobias, H. J.; Fergenson, D. P.; Pitesky, M. E.; Horn, J. M.;

Czerwieniec, G. A.; Russell, S. C.; Lebrilla, C. B.; Gard, E. E.; Frank, M. Anal.
Chem. 2003, 75, 5480-5487.

(37) Tobias, H. J.; Pitesky, M. E.; Fergenson, D. P.; Steele, P. T.; Horn, J.; Frank, M.;
Gard, E. E. J. Microbiol. Methods 2006, 67, 56-63.

(38) Tobias, H. J.; Schafer, M. P.; Pitesky, M.; Fergenson, D. P.; Horn, J.; Frank, M.;
Gard, E. B. Applied and Environmental Microbiology 2005, 71, 6086-6095.

(39)

Martin, A. N.; Farquar, G. R.; Gard, E. E.; Frank, M.; Fergenson, D. P. Anal.
Chem. 2007, 79, 1918-1925.

(40)

Martin, A. N.; Farquar, G. R.; Frank, M.; Gard, E. E.; Fergenson, D. P. Anal.
Chem. 2007, 79, 6368-6375.

58

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(48)

(49)

(50)

(51)

(52)

(53)

Adams, K. L.; Steele, P. T.; Bogan, M. J.; Sadler, N. M.; Martin, S. 1.; Martin, A.
N.; Frank, M. Anal. Chem. 2008, 80, 5350-5357.

Martin, A. N., Michigan State University, East Lansing, MI, 2006.

Dockery, D. W.; Pope III, C. A. Annual Review of Public Health 1994, I5, 107-
132.

Wiley, W. C.; McLaren, I. H. Rev. Sci. Instrum. 1955, 26, 1150-1157.

de Hoffman, E.; Stroobant, V. Mass Spectrometry: Principles and Applications;
John Wiley & Sons, Ltd.: New York, 2002.

Riot, V.; Coffee, K.; Gard, E.; Fergenson, D.; Ramani, S.; Steele, P., Waltham,
MA 2006; Proceedings of the Fourth IEEE Workshop on Sensor Array and
Multichannel Signal Processing; 98-101.

Steele, P. T.; Farquar, G. R.; Martin, A. N.; Coffee, K. R.; Riot, V. J.; Martin, S.
1.; Fergenson, D. P.; Gard, E. E.; Frank, M. Anal. Chem. 2008, 80, 4583-4589.

Hasegawa, C.; Kumazawa, T.; Lee, X. P.; Fujishiro, M.; Kuriki, A.; Marumo, A.;
Seno, H.; Sato, K. Rapid Commun. Mass Spectrom. 2006, 20, 537-543.

Zhu, L.; Chen, X. Y.; Zhang, Y. F.; Yu, H. L.; Zhong, D. F. Journal of
Chromatography B-Analytical Technologies in the Biomedical and Life Sciences
2005, 820, 175-182.

U.S. Drug Enforcement Agency, 2005; Vol. Title VII, P.L. 109-177.

Beyer, J.; Peters, F. T.; Kraemer, T.; Maurer, H. H. J. Mass Spectrom. 2007, 42,
150-160.

Marin, A.; Barbas, C. J. Pharm. Biomed. Anal. 2004, 35, 1035-1045.

Ming, M.; Fang, F.; Yulan, S.; Shuangiin, C.; Han, L. Journal of
Chromatography B-Analytical Technologies in the Biomedical and Life Sciences
2007, 846, 105-111.

59

(54) Nirogi, R. V. S.; Kandikere, V. N.; Shukla, M.; Mudigonda, K.; Maurya, S.;
Komameni, P. Rapid Commun. Mass Spectrom. 2006, 20, 3030-303 8.

(55) Farquar, G. R.; Steele, P. T.; Mchmpsey, E. L.; Lebrilla, C. B.; Tobias, H. J.;
Gard, E. E.; Frank, M.; Coffee, K. R.; Riot, V. J.; Fergenson, D. P. J. Aerosol Sci
2008, 39, 10-18.

60

CHAPTER 3: Breath Analysis — Method Developments

3-1: Motivations and Introduction

Although breathing is an involuntary action of the human body that oﬁen occurs
unnoticed, it is one of the ways that the human body interacts with the world. Gas
exchange occurs each second, with oxygen being inhaled ﬁom the air, into the lung,
transferring into the blood, and being delivered to muscles and organs throughout the
body. Carbon dioxide is then removed from the body in the reverse action. However,
more than Oz-COz exchange is occurring at this blood-air barrier. Other compounds,
both volatile and non-volatile, are constantly being brought into and expelled from the
body. These compounds may range from being inhaled and exhaled with no interaction
with the body, or may be inhaled, interact with the body and then be exhaled, or they may '
even be inhaled and metabolized before elimination. Some compounds are also produced
in the body and subsequently exhaled. These compounds may provide insight into the

processes regularly occurring in the body, from metabolism to oxidation.

The non-invasive nature of breath analysis has generated signiﬁcant interest in using
proﬁles of breath constituents to derive information about physiological functions and
interactions with the environment. Applications including medical diagnostics,
assessment of environmental contamination, and physiological monitoring are common
interests in the breath analysis community.62 Homeland security applications of breath
analysis have been suggested.62' 63 Investigation of whether breath analysis could be used
to detect the prior activity of an individual, or even biometric applications have gained

momentum recently.

61

Whatever the targeted application of breath analysis, fundamental studies are necessary to
ensure the reliability of data obtained from breath analysis in such applications. These
fundamental studies must identify the problems and weaknesses in breath analysis, with
the aim of solving some of these problems. This chapter will describe efforts made to
develop a breath analysis scheme capable of breath sampling, preconcentration, analysis,
and interpretation. Each of these segments of the scheme present individual challenges
and options, many of which are addressed. Several breath collection modiﬁcations are
presented, and several detection challenges are discussed. The next chapter will address
more applications of breath analysis once a reliable system and method is established

herein.

3-2: Background on Breath Analysis

Breath analysis has existed in a qualitative form for thousands of years. Hippocrates
described the process of diagnosing disease based on the odor of the breath of patients in
the 4‘h century BC.64 In the 18th century, the French scientist Antoine Lavoisier and
colleagues measured the breath of guinea pigs to discover the presence of carbon dioxide
in exhaled breath using a precipitation reaction.“ 65 The 19‘h century brought additional
information on exhaled breath, as colorimetric tests were used to detect acetone in the
breath of a diabetic patient and alcohol in the breath of a subject after consumption.65
The detection of alcohol in breath grew in the 20‘h century, as the automotive industry
boomed and impaired driving became an issue. Bogen’s work in 1927 to quantify the
ethanol in the blood by measuring the ethanol in the breath began the development of

breathalyzers for legal and medical purposes.(’6 The larger scale application of breath

62

analysis garnered renewed interest with the work of Pauling in 1971. In this work,
Pauling et al. described that several hundred volatile organic compounds (VOCs) were
present in breath in low, but measurable quantities.67 Since this work, these VOCs have

been the subject of much research in medical and environmental realms.

From a medical perspective breath analysis is appealing owing to the non-invasive
diagnostic nature of the technique. Thus far, over 3000 VOCs have been detected in
human breath,68 and variations in particular markers have been studied in relation to

asthma,”73 chronic obstructive pulmonary disease,” 75 breast cancer,”78 lung cancer,”

73, 86, 87 88-90 91, 92

85 lung infection, transplant rejection, and diabetes. Breath samples are

easily collected, and have been simpliﬁed to the point that the subjects can collect the
breath privately in their own homes.93 From an environmental perspective, breath
analysis has focused on detecting exposure to pollutants and industrial chemicals.94']°1
Several Environmental Protection Agency (EPA) studies have measured the breath of
subjects throughout the United States, documenting exposure to chemicals such as

automotive exhaust and cigarette smoke.97’ '00'103

No matter what the application, breath analysis relies on the principle that the lung
alveolar air can be treated as the headspace of the blood stream, given the constant
partitioning of volatile organic compounds (VOCs) between the two compartments via

68. 104

the alveolar pulmonary membrane (Figure 3- 1).

63

 

Alveolar Alveolus

Pulmonary m

 

  

 

 

 

 

 

 

Figure 3- 1: Diagram of the alveolar-capillary interface. Alveolar air sacks within the
bronchioles of the lung are in close contact with the blood stream via the alveolar-
pulmonary membrane. Oxygen and carbon dioxide, as well as VOCS, partition across
this membrane, making the alveolar air function as a headspace of the blood. (Image
originally in color)

For medical diagnostics, the compounds of interest are the endogenous compounds in

68, 105 For

breath that are formed by metabolic processes taking place in vivo.
environmental applications the other compounds present in breath, including exogenous
compounds that are inhaled, are of interest. Once inhaled, these compounds can be
passed across the alveolar membrane and be distributed throughout the body via the
bloodstream. These compounds can then be cleared via many pathways, including

respiration. Breath analysis therefore presents an interesting and non-invasive method of

detecting chemical exposure as well as monitoring the clearance of such chemicals.

64

3-3: Breath Collection Methods

Several commercial and laboratory-based devices have been used for breath collection.
The RTubeTM (Respiratory Research, Inc., Charlottesville, VA) is one such device, a
commercially available, FDA approved, disposable exhaled breath condensate (EBC)
collection system (Figure 3- 2). The RTubeTM consists of a polypropylene tube ﬁtted to a
polyethylene mouthpiece. The subject inhales through the mouthpiece via a one-way
valve (valve A) ﬁtted with a saliva trap to prevent sample contamination. The subject
then exhales through the mouthpiece into a polypropylene collection tube via a second
one-way valve (valve B). The polypropylene collection tube is placed inside an
aluminum sleeve cooled in a freezer (-20 or -80 °C) prior to collection (not shown). In
this conﬁguration, breath passing through the collection tube condenses and collects at

the base of polypropylene tube.

. 3.
ﬁll,
a
e

Valve B i

 

Valve A

Figure 3- 2: Picture of the RT ubeTM. The RT ubeTM is used for BBC collection and can be
chilled with a previously frozen aluminum sleeve. (Image originally in color)

65

The ECoScreen H (VIASYS Healthcare, Yorba Linda, CA) is a third generation EBC
collection system that also incorporates a spirometer to control either volume or time
during collection (Figure 3- 3). The subject exhales through the mouthpiece into a
pneurnotachograph, where the rate of airﬂow is measured, and the breath is then passed
into a polyethylene collector, stored in a temperature-controlled unit. A unique feature of
the ECoScreen II is the optional use of a second collector, thus allowing separation of
alveolar air ﬁ'om dead space air. A portable version is also available, the ECoScreen

Turbo, which uses a Peltier cooler to maintain the temperature of the breath collectors.

 

Valve 0
Valve 2 Pneumotachogreph
om 2 Valve 1 / Mouthpiece
Outlet 1
Collector 1 Collector 2
\\‘
‘

 

 

 

 

 

 

 

 

.’ ,4
Figure 3- 3: Diagram of the ECoScreen II. This device is used to collect BBC and can
also control the volume or time of sample collection. (Image originally in color)

Summa canisters are also used for breath collection. A Summa canister is a stainless
steel container whose inner surfaces are passivated by electropolishing with chemical

deactivation. Summa canisters are held under vacuum before sample collection, so they

may be stored for up to 30 days before use. These canisters were used for the extensive

experiments of the EPA’s environmental studies.98’ 100402: ‘06'109

3-4: Volatile and Non- Volatile Components

Breath analysis has focused on both the volatile and non-volatile components in breath.
Though they have various sources, both classes of components provide interesting
information about the subject. The Volatile components in breath range from oxygen,
nitrogen, and carbon dioxide, which compose over 99% of inhaled air, to the low
concentrations (ppbv and below) of acetone, methanol, and many alkanes and methylated
alkanes. The non-volatile components of breath are often proteins, oxidation or nitration
products, cytokines, eicosanoids, and mediators.“0 One of the most abundant
constituents of breath is water, which is exhaled in the form of respiratory droplets;
aerosol particles can become trapped in these droplets, allowing them to be expelledm
Most breath analysis techniques focus on either the volatile or non-volatile components.
Work by Zenobi et a1. is of interest because the extractive electrospray ionization used
with mass spectrometry allows the detection of volatile and non-volatile components

simultaneously from the same breath ofair.112

3-5: Breath Analysis Techniques

Many techniques have been pursued for analysis of exhaled breath including enzyme-

115,

linked immunosorbent assay (ELISA),”3’ ”4 colorimetric tests, 116optical

spectroscopy,”6’ “7 high performance liquid chromatography, 72’ ”6 ion mobility

67

spectrometry,87 liquid chromatography-mass spectrometry (LC-MS),72’ 74’ 75 GC with

. . . . - - 2 5
ﬂame ionization detection (GC-FID),]9’ “8 ”'0 and GC-MS.68‘ 76‘ 79 84’ 86‘ 88‘ 89‘ 9 ‘99’ 100’ 105-

120427 Ekips Technologies has developed a system approved by the FDA for research
studies called the Breathmeter.7O The Breathmeter uses tunable diode later spectroscopy
to detect both organic and inorganic components in breath. The mid-IR absorption of
many compounds allows for internal calibration and wide-range detection with laser
tuning. Aerocine Inc. markets the ﬁrst FDA-approved system called NIOX Flex which
uses chemiluminescence to monitor NO in exhaled breath with a detection limit of 2
ppb.128 Toda et al. developed a breath isoprene detection system based on its
chemiluminescence after reaction with ozone with a limit of detection of 0.6 ppbv.129
Plodinec and Wang used cavity ring-down spectroscopy for the detection of breath
acetone in a single breath.130 Amirav et al. demonstrated the use of gas chromatography
— electrolyzer-fed FID (GC-EFID) for breath detection of ethanol, isoprene, pentane, and
acetone.131 This system generated its own combustion gases from water and thus was gas
cylinder-free making it more ﬁeld-portable.'3' Sacks et al. used GC-GC for breath

detection, achieving limits of detection in the ppt range.132 The most common breath

detection system currently used is gas chromatography-mass spectrometry (GC-MS).

Gas chromatography-mass spectrometry (GC-MS) has shown great potential for breath
analysis, due to its selectivity and sensitivity.“ 76‘ 79‘84' 86’ 88’ 89‘ 92‘ 99’ 100‘ 104‘ 120427 Indeed, a
breath analysis apparatus using GC-MS, Heartsbreath, developed by Menssana Research,
Inc. has recently been approved by the FDA as a test for heart transplant rejection. 89

Chemical exposure has also been detected using GC-MS.'OO'102’ 125‘ '33 Soﬁa et al. studied

toluene in the exhaled breath condensate (EBC) of individuals exposed to the chemical

68

via inhalation.99 A relationship between concentration in the environment and EBC was
determined, and the toluene concentration in the EBC decreased with time after the
exposure ended.99 However, no efforts were made to ensure that additional exposure, e. g.

from clothing, did not inadvertently occur during sampling.

3-6: Issues in Breath Analysis

3-6-1 : Preconcentration

A major difficulty in accurate breath analysis is the low concentrations of the VOCs that
are present in breath samples (ppbv and below). This issue is compounded by issues
including the dilution of VOCs by extraneous air as well as loss of VOCs during
condensation if EBC is to be collected. To circumvent this issue, several techniques have
been developed to preconcentrate a breath sample before analysis. Gordin and Amirav
have developed the ‘Snifprobe’ as a novel preconcentration method.134 The Snifprobe
consists of a small length of capillary or porous layer open tubular (PLOT) column that
preconcentrates the sample. Breath was sampled for 5 5 through the Snifprobe, after
which the entire column segment was placed inside a direct/dirty sample introduction
device (DSI) which was then inserted into the GC injector for thermal desorption.134 A
limit of detection for ethanol was stated in the low ppb (v/v) level. '34 Several researchers
have also used a sorbent material to sample from a bag or chamber containing a breath
sample.” 8" 83’ 100’ ”8’ 127. 129 In this manner, preconcentration takes place after breath

' ' ' ' 22 . 2 . 3 ,
collection. Other groups have created in-line preconcentration methodsloo" ’ ‘25 1 6 ‘ 5

‘36 Phillips, for example, has developed a breath collection apparatus (BCA) which

incorporates a sorbent trap in-line with the breath flow so that VOCS are preconcentrated

69

- 68. 76. 79, so. 82. 86.89.92. 122-124
as the breath sample IS collected. '

Several generations of this
instrument have been developed, the latest being the BCA 6.0 (Menssana Research,

lnc.).'37 A promising tool for breath VOC preconcentration is solid phase

microextraction (SPME) which was included in a recent review.94

SPME was developed in 1989 by Belardi and Pawliszyn138 as a method to quickly,
inexpensively, and easily preconcentrate chemicals of interest. The SPME ﬁber is
typically an inch-long piece of fused silica coated with various polymeric materials
depending on the target compounds for preconcentration. The SPME ﬁber is housed in a
standard injection needle, so introduction to the GC is simple, and desorption occurs in
the heated injection port. Compounds are absorbed or adsorbed to the ﬁber from the
sample. The SPME ﬁber can be directly exposed to a liquid sample, where the analyte
will partition between the liquid phase of the sample and the liquid phase of the SPME
ﬁber coating, or exposed in the headspace of a sample, where the analyte partitions
between the sample, the gaseous phase above the sample, and the SPME ﬁber coating.
SPME can be used quantitatively as the amount of analyte absorbed/adsorbed to the ﬁber

is proportional to the concentration of the analyte in the sample.

An analyte partitions into a SPME ﬁber according to Equation 3- 1:139'140

KfszVsoC

"2f Kfo+KhS Vh +V

 

Equation 3- 1: Equation describing the partitioning of an analyte into a SPME ﬁber
coating.

70

Where nf is the amount of analyte extracted into the SPME ﬁber, Kfs is the distribution
constant of the analyte between the liquid sample and the SPME coating, Vf is the
volume of the SPME coating, VS is the volume of the liquid sample, C0 is the
concentration of the analyte in the sample, Khs is the distribution constant of the analyte
between the gaseous phase and the SPME coating, V}, is the gas volume of the vial, and

VS is the volume of the liquid sample.

SPME sampling can be passive, where a breath sample is obtained and later exposed to a
SPME ﬁber, or active, where a breath sample is exposed to a SPME ﬁber as it is
obtained.105 Wang et al. used passive sampling to preconcentrate breath samples
obtained in Tedlar® sampling bags, obtaining sub ng/mL detection limits for many
VOCs.119 Mutti et al. used a similar technique, collecting breath samples into Teﬂon
bulbs into which a SPME ﬁber was inserted for preconcentration, with limits of detection
on the order of 10'12 M.84 Grote and Pawliszyn have used SPME for active breath
samplingm The SPME ﬁber was inserted into an inert tube serving as a mouthpiece,

and acetone, isoprene, and ethanol were analyzed using GC-MS.I41

3-6-2: Contamination

Because of the inﬂuence of exogenous VOCs on the air exhaled by a subject, some
emphasis has been placed on ensuring the inhaled air supply of a subject is clean. One
solution is to provide a closed source of air for the subjects" 83‘ 96' 127. ”2’ ”3 Although

effective, this method is reagent consuming and makes the testing apparatus irnportable.

71

Others have required subjects to sit in the sampling environment for a set period of time

. . . , , . 7
to allow equilibration With the envrronmental air.92‘ ”9‘ “‘4

This is very time consuming
for the subjects and requires a single sampling location to compare groups of subjects.
Others still have attempted to clean the ambient air using charcoal ﬁlters133 ' 143 Many
studies do not cleanse the inhaled air supply;‘9’ 72. 75, 85, 37. 113. 116. 118-120. 129. 136. 144-147
commonly a background sample of the ambient air is sampled at the same time as the
subject who has reached equilibrium with the room air donates a breath sample and
calculation of the alveolar gradient (concentration in the breath — concentration in the
ambient air) is thought to eliminate any effects of contaminated inhaled air.“ 76’ 79’ 80' 82’ 86'
89’ 92‘ 122’ 123‘ '26 124‘ 148 A positive alveolar gradient suggests that VOC was manufactured
in the body while a negative alveolar gradient indicates the source of the VOC is external
to the body.149 While this technique has advantages in the determination of kinetics and
metabolomics, it does require that equilibrium exists for all VOCs in the body and the
ambient air — a condition that may be difﬁcult to achieve in select situations. Also,

important information about occupational and incidental chemical exposure can be

gained by studying the real-tirne exposure of a subject.

This work presents the use of a commercial respirator cartridge or PVC tubing for
purifying the inhaled air. This reduces the amount and concentration of exogenous
substances a subject inhales during sample collection, but can also reduce exposure via

clothing or skin after exposure.

72

3- 7: Background of Gas Chromatography-Mass Spectrometry

Gas Chromatography-Mass Spectrometry (GC-MS) is an orthogonal detection technique
that combines the separation ability of chromatography with the speciﬁcity of mass
spectrometry. GC is used to separate complex mixtures into more sirnple or pure
components as a method of sample clean-up and/or to prepare more pure materials for
sample analysis. The comparison of the retention time of a compound on 2 columns of
differing polarities with that of an unknown sample is a presumptive method of
compound identiﬁcation. Mass spectrometry is used to analyze the components in a
sample, allowing identiﬁcation based on the fragmentation patterns of the analyte. Gas
chromatography was ﬁrst coupled to mass spectrometry in 1958 and has evolved over the
last 5 decades into a common bench-top technique used in many standard operating

procedures in environmental, forensic, and research laboratories. '50

3-7-1: Gas Chromatography Theory

Gas chromatography was ﬁrst described by Martin and Synge in 1941 and demonstrated
in 1952 by Martin and James.15 1’ '52 Gas chromatography, or gas-liquid chromatography,
relies on a gaseous mobile phase carrying an analyte through a column ﬁtted with a liquid
stationary phase supported by a solid material. Gas-solid chromatography has also been
described, but has not garnered as much success or attention as gas-liquid

chromatography. As the solute passes through the column, it partitions between the

stationary and mobile phases. The equilibrium constant of this process, KD, depends on

73

the solute, the stationary phase, and the temperature, and can be relate to the partition

ratio (k) and phase ratio ([3) of the separation according to Equation 3- 2.

K : [statzonary(lzquid) phase] = k ,5
D [mobile( gas) phase]

 

Equation 3- 2: Description of the equilibrium constant of an analyte as it partitions
between the stationary and mobile phases during separation.

The partition ratio is a measure of how long an analyte spends in the stationary phase,

and is related to the retention time, tR, which is the time an analyte spends in the column

before elution. The standard compound, methane, is used as a comparison for partition
ratios, since methane is considered to not be retained by a stationary phase, and spends all
the time in the column in the mobile phase, and thus has the minimum retention time

possible in a column. Thus,

 

Equation 3- 3: Relationship between the partition ratio and retention time.

where k is the partition ratio, tR is the retention time of the analyte, tM is the retention

time of methane and tR’ is the time an analyte spends in the mobile phase. The phase

ratio (,6) can be deﬁned as the ratio between the volume of gas in the column and the

volume of the liquid stationary phase,

74

 

Equation 3- 4: Deﬁnition of the phase ratio.

where r is the radius of the column and df is the thickness of the ﬁlm of the stationary

phase.

The efﬁciency of a chromatographic column can be expressed in terms of theoretical
plates, or effective theoretical plates. The use of the term ‘theoretical plates’ refers to the
early ﬁactionating distillation columns which indeed had discrete plates to perform the
separation; chromatographic columns of today do not have discrete plates, so ‘theoretical
plates’ is used to describe the efﬁciency in a relative manner. Theoretical plates (11) were

originally used to deﬁne efﬁciency:

5 2
n = 5.54[Wi]
H

Equation 3- 5: Deﬁnition of the number of theoretical plates.

with WH being the width at half height of a peak. Effective theoretical plates (N) are

used to describe the actual time that an analyte is in the mobile phase of the column, thus

modifying Equation 3- 5 to be:

75

t.
N = 5.54 WR—
H

Equation 3- 6: Deﬁnition of the effective number of theoretical plates.

3-7-2: Mass Spectrometry Theory

The roots of mass spectrometry are based in the late 19th century with the analysis of
positive ions using magnetic deﬂection by Wien.150 Although there are many types of
mass spectrometry, all mass spectrometers detect ions, positive or negative. A sample
must be introduced to the system (sample introduction), ions must be formed (ionization
region), separated (mass analyzer), and detected (detector), and the data must be

processed and output into an understandable form (readout).

There are a variety of sample introduction methods, all requiring introduction of a sample
to the mass spectrometer in the gas phase, which must be held under high vacuum in
order to reduce ion-molecule collisions. These collisions may cause the ions to lose their
charge, thus losing their ability to be detected, cause ions to undergo reactions, or deﬂect
ion trajectories. Therefore to reduce these ion collisions, the mean free path of an ion is

important. The mean free path (L, cm), is deﬁned as:

1

((2pn0'2

Equation 3- 7: Deﬁnition of the mean free path of an ion.

L:

76

where p is the pressure (Pa), n is the number of moles per cm3 and o is the collision
diameter (cm). The mean free path of an ion should be at least 1 m in most mass
spectrometry applications, which requires a pressure of 66 nbar or lower as shown in

Equation 3- 8:

Equation 3- 8: Approximation of the mean free path of an ion at 300 K and 3.8x10'm
molecules/cm3.

approximated for 300 K and o =3.8x10"0 molecules/cm3. Often chromatographic
methods are used for sample introduction as they allow for compound separation and
sample cleanup. Other options include bleeding a gaseous sample into the ionization
region, or novel methods such as Direct Electrospray Ionization (DESI) or Direct
Analysis in Real Time (DART) that combine the sample introduction, gasiﬁcation, and

ionization steps.

The ionization region produces ions from the sample for analysis. Several ionization
sources are available, most commonly electron ionization (E1) or chemical ionization (C1)
are used in GC-MS, but other ionization sources exist for other applications. In electron
ionization, the sample is ionized by interactions with electrons. The electrons are
produced by heating a ﬁlament, and accelerated towards an anodic plate. As the
electrons travel through the source, they encounter the gaseous sample, and if the
wavelength of the electron approaches the bond length of the molecules in the sample,

energy transfer may occur, and an electron is expelled from the molecule, creating a

77

positive ion. Typical electron energy of 70 eV is used for the ionization of organic
molecules, which maximizes the ionization efﬁciency to ~O.1% on average because of
the similar de Broglie wavelength of the electrons and bond length of the organic
molecules (~ 0.14 nm). Only the positive molecular ion is formed using El, as the energy
of the electrons is too high for electron capture to occur. Fragmentation of the molecule

is common using EI, creating fragment ions which are also detected.

Another common ionization method is chemical ionization (CI) which involves the
introduction of a reagent gas, such as methane, which is ionized using electron ionization.
The ions produced will then collide and react with other molecules of reagent gas,
creating a series of ions in the source region. These reagent gas ions will then react with
the sample molecules, ionizing them in a ‘soft’ ionization process that causes minimal
ﬁagmentation. Deposition of energy into analyte ions is much less than for El because it

depends on the thermodynamics of the ion-molecule reaction.

After ions are produced they must be separated before detection. This separation occurs
in the mass analyzer, which is commonly a quadrupole mass analyzer for GC-MS and is
the mass analyzer used in the instruments used in these experiments. A quadrupole mass
analyzer consists of 4 parallel hyperbolic rods, although circular rods are more commonly
used. An electric ﬁeld consisting of a quadrupolar oscillating ﬁeld superimposed on a

constant electric ﬁeld exists in the center of the analyzer, following:

78

(1)0 =(U—Vcoscot)

—(I)0 =—(U-Vcosa)t)

Equation 3- 9: Equations describing the electric ﬁeld formed in the center of a
quadrupole mass analyzer.

where (D0 is the potential applied to the rods, 0) is the angular frequency, and U is the

constant potential and V is the peak amplitude of the RF voltage. The angular frequency

is related to the frequency of the RF ﬁeld by Equation 3- 10:
a) = 279’

Equation 3- 10: Relationship between angular frequency and RF ﬁeld frequency.

where f is the ﬁ'equency of the RF ﬁeld. Ions created in the source region are accelerated
along the z-axis via focusing lenses into the space between the rods. The electric ﬁeld
induces forces on the ions in the x and y directions, causing the ion to move away from
and towards the rods. If the ion does not change directions quickly enough with the
oscillation of the RF ﬁeld, it will collide with the rod and be discharged. If the ion
direction oscillates with the RF ﬁeld the entire time it is within the rods, it will pass
through to the detector with its charge intact. Thus, by modifying the electric ﬁeld, i.e.
controlling U and V, ions of a speciﬁed mass-to-charge ratio can be selectively passed
through to the detector. For detection of a wide range of m/z, U and V must be scanned
to allow ions of successive m/z to be passed through the mass analyzer. Generally, the

UN ratio is kept constant to allow this successive detection.

79

The ions that pass through the mass analyzer are converted to electrons by collisions with
a conversion dynode, and these secondary electrons are detected using an electron
multiplier, typically a continuous dynode electron multiplier. In an electron multiplier,
the charged particle collides with a metal or lead oxide coated surface of low work
function, inducing the emission of 1-3 electrons. These electrons are then accelerated
onto another plate, inducing an additional 1-3 electrons each. This process is repeated
numerous times until a gain in signal of about 105 is realized. The current induced in this
process is measured as a voltage drop across a resistor, and this voltage is converted to a
digital value and stored on a computer. A mass calibration ﬁinction assigns m/z values to
each ion based on the U and V potentials that are applied coincident with the arrival time

of each ion.

3-7-3: GC-MS Description

A schematic of a GC-MS instrument used at LLNL is shown in Figure 3- 4.

80

 

Intensity

 

. / dd

/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 i:-
A I e A .5 __.—-y 'g
- ' o
- g a
< m/z
Retention Time \ 5'
M
. a
8
oven 5‘
m/z

 

 

 

 

Figure 3- 4: Schematic of a gas chromatographmass spectrometer (GC-MS). A sample
is infected onto a capillary column held inside a temperature programmed oven. As the
sample passes through the column, components separate based on partitioning between
the mobile and stationary phases of the column, aided by the temperature ramp in the
own The components exit the column directly into the source region of the mass
spectrometer, where they are ionized fragmented, and pass into the mass analyzer,
where the ﬁ'agments are detected The resulting data includes a chromatogram with a

corresponding mass spectrum for each time point. (Image originally in color)

The sample is introduced to the instrument through a temperature controlled injection
Port using a syringe (liquid samples) or a solid phase microextraction (SPME) fiber,
discussed previously, where a solid preconcentrating material is injected via syringe
needle and the adsorbed sample is thermally desorbed from the material. The sample is
vaporized and carried onto the capillary column by a ﬂow of carrier gas, typically helium.
The column is stored in a temperature controlled oven, and the initial temperature is
ty'l)ically held at a point lower than that of the injection port, causing the vaporized
SaIl'lple to condense at the head of the column. The carrier gas carries the sample through
the column and the analytes partition between the mobile (carrier gas) and stationary

p11'c‘tses (column coating) to different extents. Compounds that partition more into the

81

 

mobile phase are transported through the column faster than compounds retained more in
the stationary phase, causing the separation to occur. The temperature of the oven is
often increased with time, aiding volatilization of less volatile analytes. The end of the
column is positioned inside the ionization chamber of the mass spectrometer, where the
pressure is held in the range of 108-10’5 Torr. Each analyte is ionized via electron
ionization (EI) as it elutes from the column. The resulting ions are accelerated through
the quadrupole mass analyzer and are detected by the electron multiplier. The signal is
digitized and the data readout provided through the commercial software that also

operates the instrument.

The current study demonstrates an in-house modiﬁcation of a standard exhaled breath
condensate collection device, the RTubeTM (Respiratory Research, Inc., Charlottesville,
VA)- This modiﬁcation reduces the time of analysis from sampling to results to 27
minutes, as well as eliminates the need for a freezer, making the sampling device more
ﬁeld-ﬁiendly, and increases the sensitivity. This study also shows the need for
puriﬁcation of inhaled air in some cases and describes additional RTubeTM modiﬁcations

that meet this need.

3 - - . .
8~ Instrument Description

3-8‘ 1 Z Agilent System
An Agilent (Santa Clara, CA) 6890N GC equipped with a SPME inlet sleeve and an
Agilent J&W DB-5MS column (30 m x 0.25 mm x 0.25 pm, equivalent to 5% phenyl,

950/0 methylpolysiloxane) was used for the separation of all compounds. Helium

(99 999%, Praxair, Inc., Danbury, CT) was used as a carrier gas and the GC was operated

82

 

 

in constant pressure mode (10 psi). The injector temperature was held at 250 °C, and the
oven temperature was held at 40°C for 1 min, then ramped at 20°C/min to 300°C, where it
was held for 3 min. Detection was carried out with an Agilent 5973 MS detector (Santa
Clara, CA), using 70 eV electron ionization and operated in scan mode (m/z 40 — 450) at
3-54 scans/s, with the source held at 230°C and the mass analyzer (quadrupole) held at

1 5 0°C. ChemStation software was used for data collection and analysis.

3-8—2: Shimadzu System
A Shimadzu (Pleasanton, CA) QP2010 with a SPME inlet sleeve and ﬁtted with an

Agilent J&W DB-SMS column was used. The same method was used as in the Agilent

system. GCMS Solutions software was used for data collection and analysis.

Compounds were identiﬁed based on comparison of their mass spectra with those in the
NIST Mass Spectral Library (RMatch >700, NIST MS Search 2.0, NIST, Gaithersburg,
MD) as well as mass spectra published in the literature, and comparison of retention time

with pure chemicals when available.

3‘9 -' Materials

Frozen lemonade concentrate (Safeway/Vons Lemonade, Safeway, Inc., Pleasanton, CA)
Was purchased from a local supermarket and reconstituted according the manufacturer’s
instructions. Samples were used the day of reconstitution after reaching room
ternperature (24 °C). Nail Polish (ORLY Nail Color, Orly International, Inc., Los

A“gEIes, CA) was obtained from commercial sources and used as is. Gasoline (89 octane

83

 

automotive grade) was obtained from a local gas station. D-Camphor was obtained ﬁom
J -T- Baker, Inc. (Phillipsburg, NJ). A 65.3 mM solution was made in a 1:4 solution of
H 20 :methanol and stored in a glass vial. (1S)-(-)-B-pinene (99%) was obtained from Alfa
Aesar (Ward Hill, MA) and +/- a-limonene (95%) was obtained from TCI America, Inc.

( Portland, OR).

3 -1 0: Subjects

Fourteen healthy adults were recruited as subjects for this study. All subjects had no
known respiratory ailments and did not report any routine chemical exposure on a
questionnaire. The Institutional Review Board (IRB) at Lawrence Livermore National
Laboratory (LLNL) as well as the IRB at Michigan State University approved the study
and formal written informed consent was obtained from each subject prior to

participation.

3-11 .- Original Sample Collection (EBC)

Exhaled breath condensate (EBC) was originally collected as per the RTubeTM
man“ facturer’s instructions. The aluminum sleeve was cooled in a -80°C freezer
oveI‘Ilight, and then placed in the insulating sleeve on the RTubeTM body. The subject
held the RTubeTM and inhaled and exhaled through the mouthpiece, causing the exhaled
breath to condense on the walls of the cooled polypropylene tube. Breath samples were

coneCted for 10 minutes, after which the mouthpiece apparatus was removed ﬁom the

RT‘lbeTM, and the polypropylene tube placed on a plunger which caused the liquid

84

 

condensate to be collected in a small volume area of the tube. The liquid was then
divided into 500 pL aliquots in microcentrifuge vials and the excess vials refrigerated. A
single EBC aliquot was sampled using a SPME ﬁber under direct exposure for 10
minutes which was then injected into the GC-MS. The SPME ﬁber was predesorbed
during a ‘blank’ GC-MS run each morning before any breath sampling occurred. After
collection the RTubeTM was washed with methanol followed by water, followed by Di.
water and dried before reuse. Studies demonstrated there was no carryover effect from

reusing the RTubeTM after cleaning.

3 - I 2 : GC-MS Method Development

3- 1 2— 1: GC Method Optimization

Experiments were performed to optimize the GC—MS method used for analysis. Mass
Spectra were acquired over m/z 40-450. The temperature ramp of the GC oven was
modiﬁed to create different methods to optimize the separation of a breath sample. EBC
Was 1lsed in order to have a single sample that could be analyzed using each method.

Folll‘ methods were created and are shown in Table 3- 1.

A Single EBC sample was collected from the subject, and split into 4 0.5 mL aliquots. A
SPME ﬁber was directly exposed to the liquid condensate for 10 min. Figure 3- 5 shows
total iOn chromatograms obtained using each method. Overall peak area is consistent
thl‘o‘lghout methods 1-3, with 8.3% RSD for the total integrated area under the curve;
ho“Vever, the total area under the curve for method 4 was approximately 3 times larger

than for the other methods. The peak area for the peak attributed to acetone, at

apProximately 1.4 min, revealed an RSD of only 4.9% across all 4 methods. Integration

85

 

 

using the ChemStation Software (RTE Integrator, 6000 count area minimum) identiﬁed
6 1 , 48, 52, and 11 peaks, respectively for each method. Method 1 was therefore selected

to provide an optimal separation method based on number of peaks detected and time of

analysis.

Table 3- 1: Description of method parameters for methods tested to optimize the L .
'1

separation of breath samples.

 

 

 

 

 

 

 

 

 

Method 1 Method 2 Method 3 Method 4
Starting Temp 40°C (1 min) 40°C (1 min) 40°C (1 min) 40°C (1 min)
Ramp Rate 20°C/min 15°C/ min 10°C/min 1°C/ min
Final Temp 300 (5 min) 300 (5 min) 300 (5 min) 300 (5 min)

 

86

 

 

 

200000 i Method 1

   
         

t ‘ Ll "lli

2.0 4.0 6.10

 

 

llllllll V
' 8.0 -

Method 2

10.0 120’ 14.0 16.0

 

.—

 

. ILhiJi liLiI Li YIliLliL - 5

(2.0 4.0 6.0 8.0 10.012.014011607180200 22.0
2000004 ‘

: ,ILU. l.

 

 

1600001 Method 3

 

 

i ii. L1 lillil illili Ill 1i ll till 1111.

0 =—‘- . . .
, 5.0 10.0 15.0 20.0 25.0 30.0
200000

i: MethOd 4

a
a
E

.5
12mw5

Abundance (8.11. )

ego

 

.A

50.0 100.0 11570.0 ' 2071.0 250.0
Retention Time (min)

1:314" e 3- 5: GC-MS total ion chromatograms obtained from a single breath sample
"afyZed with the 4 methods shown in Table 3- 1. Method 1 was selected as the
Opt""01 detection method.

87

Z. '_ -“.. I

 

3-12-2: SPME Exposure Time Optimization
The SPME PDMS/DVB ﬁber coating was selected based on the literature using SPME

for breath analysism’ ‘53 and the need to preconcentrate mostly semi-volatile compounds,
and was conﬁrmed based on conversations with other researchers in the breath analysis
field.154 The exposure time was optimized in two sets of experiments, one using BBC
and one using EBV. Figure 3- 6 shows the peak area of three peaks detected in EBV
collected for 0.5, l, 2, 5, and 10 min. These compounds, an unidentiﬁed branched alkane
(RT=7.83 min), decanal (RT =7.29 min), and an unidentiﬁed alcohol (RT=10.67 min),
were selected to represent an alkane, an aldehyde, and an alcohol, respectively. Data
points are the average of 3 replicates taken in a single day. A sampling time of 10 min

Was optimal and used for the remainder of the experiments. ‘55

40 g Aunidentiﬁed branchedalkane
l I
35 5 decanal
l O unidentiﬁed alcohol
a 30
e .1
'2 25
><
8 20 ';
2 :
E 15 4
On 3
10 5
5 ivii l
0 - -i e-
0 2 4 6 8 10 12

Time of SPME Exposure Breath Collection (mull

Figure 3- 6: Comparison of TIC peak area to exposure time of SPME ﬁber. The
e“Inmate time of the SPMEﬁber is equivalent to the breath collection time for these EBV

s(Naples. (Image originally in color)

88

 

 

3 — 1 3 .' Comparison of Breath t0 Background Air

It is important to understand the source of inhaled air to determine whether a compound
detected on breath is exogenous or endogenous. The inhaled air was measured by
exposing a SPME ﬁber for 10 minutes on a desk in the room where breath samples were
obtained. This SPME ﬁber was then analyzed using the standard GC-MS method.
Figure 3- 7a shows the chromatogram obtained after exposure to room air. There are
several peaks present in the sample. Table 3- 2 shows the tentative identiﬁcation as well
as retention time of these peaks. Figure 3- 7b shows a representative breath sample
(EBC) collected from a single subject. Figure 3- 7c shows this same sample with the
contribution ﬁ‘om the room air subtracted. While many peaks in the room air and seen in

the breath, the concentration is much larger in breath.

89

huh-4

 

1 4 a.) Room Air
1 2
l O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l {b - ) Breath Sample
2.? 12
3
§ 10
g 8
s:
s 6 h
E
~<° 4
E 2 I II I
0
14 c- ) Breath Sample with Room Air Subtracted
12
10
8
6
4
2 l L
0 J L 1 LI 1 _
2 4 8 , , 10 , 12 14 16
Retention T1me(m1n)
F‘
ﬁgure 3“ 7: Chromatograms obtained from (a) 10 minute exposure of the SPMEﬁber to
(c) i‘hmbl-ent room air at the collection site, (b) a representative breath sample (EBC), and
in I”: 211:1}? erence between the chromatograms in a and b. Note that the majority of peaks
0

are in much greater concentration than those in the room air.

90

 

T able 3- 2: Compounds and corresponding retention times tentatively identified in a
SPAM sample of room air.

 

 

 

Peak Compound RT Peak Compound RT
1 Carbon dioxide 1.33 14 3—ethyl-Benzaldehyde 7.01
2 D imethyl silanediol 2.35 15 4-ethyl-Benzaldehyde 7. l 7
3 Butanoic acid, ethyl ester 3.42 16 Decanal 7.30
4 Hexamethyl 3 .55 17 2,4-diisocyanato- 1 -methyl- 8.52
cyc lotricyloxane benzene
5 Un identiﬁed aromatic 4.29 18 1,3-dihydro-5-methyl-2H- 8.91
benzimidazol-Z-one
6 Be nzaldehyde 5. 12 19 1,3-dihydro-5-methyl-2H- 8.95
benzimidazol-Z-one
7 OCtamethyl 5.18 20 Unidentiﬁed compound 9.04
Cchotetrasiloxane
8 6—methyl-5-Hepten-2-one 5.28 21 l-tridecanol 9.35
9 He)‘Eanoic acid, ethyl ester 5.40 22 6-undecylamine 12.00
10 2‘(Z—ethoxyethoxy)- 5.43 23 Bisphenol A 13.59
et11211101
\
11 Benzyl alcohol 5.79 24 Unidentiﬁed Compound 16.19
‘2 NonanaI 6.42 25 2-hydroxy-l-(hydroxymethyl) 16.43
ethyl ester-octadecanoic acid
13 DeCamethyl- 6.58
Cyclopentasiloxane

 

91

 

3 - I 4: Reproducibility Studies

3— 1 4- 1 : Repeatability of Measurement

A study was undertaken to determine the injection to injection repeatability of the
measurement. Since breath itself is a variable matrix, sensitive to storage and the activity
of each subject, a standard of diethyl ethylphosphonate (DEEP) was spiked into samples
for measurement; thus, the matrix effects of breath were still present. In order to generate
a larger volume of sample, EBC was simultaneously collected from two subjects, and
combined for a total of approximately 5.7 mL of EBC. 15 uL of a 0.031 M solution of
DEEP irl methanol was spiked into 5.5 mL of this combined EBC sample. The sample
was then divided into eleven, 0.5 uL aliquots, each of which was refrigerated until
analysis by GC-MS. Samples were analyzed 19, 79, 139, 199, 260, and 441 minutes, as
well as 24-3, 48.9, 49.3, 72.9, and 96.1 hours after sample collection. Chromatograms
obtained ﬁ‘om the samples at 48.9 and 72.9 hours after exposure were eliminated because
the SPME ﬁber was exposed to the sample longer than intended. Peak areas were
calculated by manually integrating the area of the peak at 6.23 min due to DEEP in each
sample. The relative standard deviation (RSD) was determined to be 11.7%. This

measurement includes error in sample dilution, SPME ﬁber exposure time, and

Instrument variation.

3- - I . o O

14 2- Smgle Subject, Smgle Day
1 __ . . . . - ' '
ntra Subject var1at10n 1s a concern when developmg an analyt1cal method relymg on

bi - , _ . .
Ologlcal samples. Natural var1at10n between samples may cause a false pos1t1ve or false

negative test if it is not understood. This variation may occur on a short time scale, being

92

 

present within multiple samples collected in a single day, or on a longer time scale, being
present within samples collected over the course of weeks or months. In order to study
the short-term variation, EBV samples were collected ﬁom a single subject throughout
the day- l 9 samples were analyzed using the standard GC-MS method immediately after
collection- The GC-MS data ﬁles were processed using MarkerLynx, which extracts XIC
peak areas for nontargeted analytes, in this case extracting an integrated area for a single
m/z ﬁ'om each detected peak. Method parameters for XIC peak area extraction are

shown in Table 3- 3.

Table 3 — 3 .- MarkerLynx method parameters for peak area extraction

Markers at m/z 45, 73, 77, and 207 were excluded as they arise from septum and column

bleed

 

 

 

 

Retention Time Window 0-17 min
Mass Range 40—450 Da
Mass Tolerance 0.30 Da
Use Relative Retention Time No

 

Peak Width at 5% Height

1 s (automatic)

 

Peak-to-Peak Baseline Noise

0 (automatic)

 

Masses per Retention Time

1

 

 

 

 

 

 

 

 

Minimum Intensity 1% of base peak intensity
Mass Window 0.3

Retention Time Window 0.2

Noise Elimination Level 0

Deisotope Data No

Peak Area Normalization None

Peak Area Scaling None

 

 

93

’ and m/z 93 was also excluded as this is the major m/z peak in terpenes, which the

 

 

subject consumed in lemonade the day of sampling. Ninety marker compounds were
reported. The standard error of the mean (SEM) of the raw peak area of each marker for
the l 9 samples was calculated. 27.8% (25) of the markers had a SEM of <5%, 8.9% (8)
had an SEM between 5% and 10%, 10.0% (9) had an SEM between 10% and 15%,
5.6% (5) had an SEM between 15-20%, 7.8% (7) had an SEM between 20-25%, and
21.1% ( l 9) had an SEM >50%. Of the markers with an SEM <5%, 72.0% (18) had an
SEM <1 °/o. These results show that the intra-subject variation during a single day is less
than 25% for the majority of markers. The variance is lower for those markers with
lower average peak areas, with those with an SEM<1% ranging from 0157-23] a.u.,
those with SEM<5% ranging from 3.61-5.69 a.u., SEM<10% ranging from 659-298 a.u.,
SEM<150/o ranging from 11.5-364 a.u., SEM<20% ranging from 29.5-45.8 a.u.,

SEM<2so/oranging ﬁom 33.8-564 a.u., and SEM>50% ranging from 93.0-11800 a.u..

3'14'33 Single Subject, Multiple Days

lntra-subj ect variation is not only caused by the variation in the breath composition
throughout a single day, but also by the long-tenn variation in the breath. This type of
variation could be caused by more major lifestyle changes, including diet, exercise,
em’ircmn'lent, and stress.155 This long-term variation was studied by collecting 25 EBV
samples from a Single subject over the course of 1 month. Samples were analyzed using
the Standard GC-MS method immediately after collection and analyzed using
MarkerLynx as described in section 3-14-2 and Table 3- 3 with the exclusion of only the
markers at m/z 45, 73, 77, and 207. The software reported data for 233 markers. The

standard error of the mean (SEM) for the raw peak areas of 25 samples was calculated for

94

 

each marker. 43.8% (102) of the markers had a SEM of <5%, 9.4% (22) had an SEM
between 5% and 10%, 6.4% (15) had an SEM between 10% and 15%, 3.0% (7) had an
SEM between 15-20%, 3.86% (9) had an SEM between 20-25%, and 20.2% (47) had an
SEM >50%. Of the 102 markers with an SEM <5%, 58.8% (60) had an SEM <1%;
however, these 60 markers had very small average peak areas, the maximum being 0.28
a.u., and they were not present in all samples, so they may be meaningless as markers.

These results show that the intra-subject variation over a period of a month is minimal for

the majority of markers.

3-15: Optimization of Breath Sample Collection

3'15‘11 Exhaled Breath Condensate (EBC) and Exhaled Breath Vapor at -80°C (EBV-
80°C) Co llection

The RTUb eTM was slightly modiﬁed to allow for simultaneous vapor collection above the
BBC A plastic ﬁtting was constructed from a second RTubeTM (Figure 3- 8b) which
attached directly to the top of the RTubeTM and served as a mount for a SPME sampler.
A SPME ﬁber (65 um PDMS/DVB Stableﬂex ﬁber, Sigma-Aldrich, St. Louis, MO),
previously thermally conditioned in the GC injection [port per manufacturer’s
i“Structiolls, was ﬁtted into this mount and extended into the polypropylene tube, the end
0f the ﬁber extending approximately 3.7 cm into the tube. The subjects breathed at
normal frequency and tidal volume through the mouthpiece for 10 01' 15 min, typically
yielding 0.7-2.5 mL of EBC. Alter collection, the exhaled breath vapor (EBV-80°C)
conected on the SPME ﬁber was analyzed immediately by GC-MS, and a standard

volume, 0.5 mL, of EBC was aliquoted into a disposable polypropylene microcentrifuge

tube (Fisher Scientiﬁc, Pittsburgh, PA). Alter analysis of the EBV-80°C was complete,

95

L“

the SPME ﬁber was directly inserted into the aliquot of BBC for 10 minutes (direct

exposure), followed by injection into the GC—MS.

   

35w. e 3 ~ 8: Exhaled breath condensate (EBC) and vapor (EBIO collection devices. a.)

combiercially available RT ubem; b.) the RT ubeTM was modified for EBV collection
Zyddt'he addition of a plastic ﬁtting to hold a SPAM ﬁber for active sampling; c.) the
samlnz?nal modification of the RT ubeTM to provide an enclosed environment for 3PM
c at 01:.)‘ng as well as the addition of a respirator to punﬁi inhaled air. (Image originally in

3‘15‘21 Exhaled Breath Vapor — Room Temperature (EBV-RT) Collection

EBV‘RT was collected using the modiﬁed RTubeTM (Figure 3- 8b) at room temperature,

usmg 110 aluminum condenser. The subject breathed at a normal frequency and tidal

volume through the mouthpiece of the RTubeTM for 10 min while the SPME ﬁber was

exposed inside the polypropylene tube. Aﬁer collection, the sample was immediately

analyzed.

3-15-3 : Comparison of Collection Methods

Breath samples were successfully collected from all subjects (n=4) yielding EBC (-80 °C
condenser), EBV-80°C (vapor collection while using -80 °C condenser), and EBV-RT
(vapor collection at room temperature). To reduce confounding effects such as diet,
ambient intake air, and activity, all samples were obtained from each subject within one
hour. All EBV-80°C and EBV-RT samples were analyzed immediately after collection;
all EBC Samples were sampled using SPME shortly after collection and then immediately
injected- Noticeable warming of the aluminum condenser occurred during sampling, as
has been previously documented.114 A typical EBC collection yielded 0.7-2.5 mL EBC
during the 10 minute collection time. EBV-RT collection typically yielded < 200 uL

condensate which was discarded. Figure 3- 9 shows example total ion chromatograms

(TIC) from breath samples of a single subject collected using EBC, EBV-80 °C, and
EBV-RT-

97

 

  
 

1 L L.ltn1..l n r ALL-L

 

 

L L L l. l A A.; I n“. A- A L A f
3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
Time (min)

Figure 3— 9: Total ion chromatograms of breath samples from a single subject. Samples
were obtained using several collection methods: EBC with a -80 “C condenser (black
trace, bottom), EBV-80°C (green trace, middle), and EBV-RT (red trace, top). Arrows
Indicate several peaks with increased signal in the EBV samples compared to the EBC
”MP1 e- Chromatograms are oﬂlset for clarity. (Image originally in color)

The EBV-RT samples yielded an average of 23% more peaks than the EBC samples from
the same subject using the same integration parameters, several of which are indicated by
the arr 0W5. Table 3- 4 shows a list of observed compounds and their retention times

tentatively identiﬁed in EBV-RT breath samples.

98

 

Table 3- 4: Breath VOCS and corresponding retention times tentatively identiﬁed in a
typical EB V -R T sample (no known exposure).

 

 

 

 

 

 

 

 

 

Peak Compound RT Peak Compound RT

1 Carbon dioxide 1.3 38 a-tcrpinene 6.3
2 Acetone 1.54 39 a-p-dimethylstyrene 6.34
3 Acetic acid 1.85 40 Nonanal 6.43
4 Allyl methyl sulﬁde 2.5 41 D-fcnchyl alcohol 6,64
5 Unidentiﬁed Compound 2.63 42 4,7,7-Trimethylbicyclo 6.76

[4. l .0]heptan-3-ol

6 (Z)- l -(methylthio)—1-propene 2.82 43 L-i80puleg01 6.89
7 Methyl isobutyl ketone 2.86 44 Camphor 6.92
8 Formic acid 3.0 45 Menthone 6.96

9 Toluen e 3. l 3 46 Isopregol 7.0
10 4,4-di methyl-Z-pentene 3.4 47 Ethyl benzoate 7.05
11 Butyl ester acetic acid 3.55 48 4-isopropyl-l- 7.13

methylcyclohexanol
12 3-fural dehyde 3 .77 49 4-terpineol 7. l 8
l 3 4'hydroxy-4-methyl-2-pentanone 3 .86 50 a-terpineol 7.29
14 2-ful‘anmethanol 3.97 5 l 4-ethyl octane 7.35
i 0”xylene 4.08 52 S-hydroxymethylfurfural 7.43
‘6 P‘Xylene 4.18 53 Undecane 7.58
17 Unidentiﬁed Aromatic 4.31 54 3-methyl-dodecane 7.67
‘8 Styrene 4.4 55 Tridecane 7.75
19 Ethylene glycol monoisobutyl ether 4.49 56 2,6,10-m'methyl-dodecane 7.82
20 2-methyl-bicyclo[3.1.0]hex-2-ene 4.75 57 Unidentiﬁed Alkane 7.85
21 6’6'dimethyl'2' 4.84 58 lsomenthol acetate 8.02
22 me"hylenebicycloB. l . l ]heptane

23 Benzaldehyde 5.13 59 Unidentiﬁed Alkane 8.54

24 Phenol 5.19 60 Unidentiﬁed Alkane 8.6
25 G‘methyl-S-hepten-Z-one 5.29 61 Unidentiﬁed Alkane 8.63
26\Cyclofenchene 5.30 62 Unidentiﬁed Alkane 8.81

27 beta-pinene 5.34 63 Unidentiﬁed Alkane 8.9
28 ISOpentyl ether 5.46 64 Dihydropseudoionone 9.18
29 a1pha-terpinene 5.66 65 2,6-dimethyl heptadecane 9.25
3O 2ettthyl-l -hexanol 5.7 66 l-dodecanol 9.35
Merle 5.73 67 Unidentiﬁed Compound 9.48
32 Q-limonene 5.78 68 Hexyloctylether 9.53
33 Eucalyptol 5.83 69 Butylated hydroxytoluene 9.62
34 2-methyl tridecane 5.96 70 Diethyl phthalate 10.1
35 4,6-dimethy1-undecane 6.01 71 Heptadecane 10.2
36\gamma-terpmene 6.04 72 2,2-d1methyl-3-octanone 10.8
Isooctylvrnyl ether 6.16 73 2,2-d1methyl-4-Octen-3-ol 1 1. 1

37 2,5-furandicarboxaldehyde 6.19

99

The vapor samples also provided increased peak areas for many peaks. For example, 01-
limonene showed a ten-fold increase in peak area when a vapor phase sample was
collected and analyzed. The area increase of several such peaks is shown in Table 3- 5.
Thus, for many compounds, vapor phase collection of exhaled breath provides increased
signal and increased information.

Table 3- 5: Peak area of several compounds tentatively identified in total ion current

chromatograms of breath samples collected as EB V-800C or EB V—RT versus peak area of

EBC total ion current chromatograms. These peaks are indicated by the arrows in
Figure 3- 9-

 

 

 

 

 

 

 

 

 

 

 

 

Factor Increase in Measured
Peak Area vs. EBC-80°C
Compound R.T. (min) EBV-80°C EBV-RT
Allyl methyl sulﬁde 2.50 5.0 2.1
\
(Z)-1 -(methy1thio)- 1 -pr0pene 2.82 9.9 3.7
_\
a-Lunonene 5.78 6.8 10.2
Ethyl benzoate 7.05 1.6 4.6
LT'\
r1decone 7.75 2.0 8.4
’u\
n1dentiﬁed Alkane 7.85 2.1 9.0
Butyloted Hydroxytoluene . 9.62 4.9 8.4

 

 

 

 

3- .
1 6' Eﬁorts to Control the Inhaled Air of a Subject

346-1; Rrobem Modiﬁcation with PVC Tubing

A method to provide clean air to a subject was developed using approximately twelve
feet of FDA-approved clear PVC tubing (0.5 in I.D., 0.75 in OD, VWR, Inc., West

Chester, PA) ﬁtted to the base of valve A on the modiﬁed RTubeTM (Figure 3- 8b) using

100

 

 

a plastic ﬁtting. Tubing was stretched into a different room (air space) than that in which
the samples were obtained. This provided inhaled air ﬂee from contamination from the
skin or clothing of the individual, and eliminated contamination by the localized
environment of the subject. The subject breathed at normal frequency and tidal volume
through the mouthpiece while holding their nose for 10 min. After collection, the EBV-

RT collected on the SPME ﬁber was analyzed immediately.

The headspace of the plastic tubing was sampled using SPME in order to understand any
background that may be present in the tubing. A SPME ﬁber was placed inside the
tubing for 10 minutes, and analyzed using GC-MS. The major compounds detected were
2-ethyl—l-hexanol, and butylated hydroxytoluene, a solvent and stabilizer used in the
production of plastics, respectively. An increase in the peak area of these compounds in
chromatograms obtained using the plastic tubing to provide inhaled air relative to
samples obtained without the tubing was found, but both compounds are present in many
products that an individual is exposed to on a daily basis; thus they are normally found in
breath obtained with or without the tubing. These peak areas decrease over time as more
air is passed through the plastic tube, thus, the tube was ﬂushed with compressed air

before use.

3-16-2: Nail Salon Inhalation with PVC Tubing

Breath samples (EBC) from a single subject with a fresh manicure were obtained at
regular intervals over the course of a single day (7 hours) using the RTubeTM modiﬁed
with PVC tubing. Immediately alter the manicure, the concentration of several

compounds, tentatively identiﬁed as isopropyl alcohol, camphor, and diacetone alcohol,

10]

increased and then exponentially decayed with time (data not shown). Isopropyl alcohol
and camphor were listed as ingredients on the nail polish used; diacetone alcohol was not
a listed ingredient, but is commonly formed via an aldol condensation of acetone,156
which was a listed ingredient. The breath sample obtained 215 minutes aﬁer the
manicure showed minimal detectable signal ﬁ'om these compounds. An additional
sample was then taken 317 minutes post-manicure using an unmodiﬁed RTubeTM (no
tubing). This chromatogram again showed the increased concentrations of isopropyl

alcohol, camphor, and diacetone alcohol. Extracted ion chromatograms for each of these

peaks are shown in Figure 3- 10.

 

 

 

 

 

 

 

a. b. 2000 c.
6 M l ) A 10000l ) )
3 5 M l 8000, 1600‘ — EBC ~ standard
E 4 M - 6000. 1200. collection
5 3 M - . — EBC— PVC
'8 4000- 800‘ tubing
3 2 M ' 4
OJ ..........

 

 

 

 

O 0,
1.50 1.54 1.58 3.82 3.86 6.90 6.92
Retention Time (min)

Figure 3- 10: Extracted ion chromatograms (XIC) of breath samples ﬁ'om a single
subject (EBC). One sample was collected 215 minutes after the subject received a
manicure with an RT ubem modified with PVC tubing which provided inhaled air from a
sepa'ate room (black trace, bottom). Another sample was obtained 31 7 minutes aﬂer
receiving the manicure, and was obtained with the standard unmodified RT ubem (red
trace, top). a.) the MC of m/z 45, identiﬁed as a major fragment ion ﬂow: isopropyl
alcohol, b.) the XIC of m/z 43, corresponding to a major fragment ion ﬁom diacetone
alcohol, and c.) the XIC of m/z 152 corresponding to the molecular ion of camphor.
(Image originally in color)

102

By holding the RTubeTM to obtain the breath sample, the subject’s hands (i.e. painted
ﬁngernails) were in close proximity to the mouth/nose, inadvertently causing the subject
to inhale a higher concentration of compounds directly from the nail polish. The use of
the PVC tubing to provide an air source removed from the subject’s person eliminated

these localized situational effects.

Alveolar gradients were calculated for isopropyl alcohol, camphor, and diacetone alcohol
in both breath samples. The breath sample collected using the traditional RTubeTM
method (317 min after exposure) yielded positive values for all three compounds,
implying that they were manufactured inside the bodym’ ‘24‘ 149 However, the sample
obtained using the RTubeTM modiﬁed with PVC tubing (215 min after exposure) yielded
negative alveolar gradient values for camphor and diacetone alcohol, indicating they were
not synthesized in the body. The alveolar gradient for isopropyl alcohol remained

positive, but was 96.3% smaller than in the contaminated sample.

3-16-3: RTubeTM Modiﬁcation with Respirator Cartridge

A commercial organic vapor respirator cartridge with a P100 Particulate Filter (99.97%
minimum ﬁlter efﬁciency, Lab Safety Supply 1nc., Janesville, WI) was ﬁtted to the
plastic ﬁtting holding valve A via a Viton® o-ring (approximately 2.5 cm diameter) and a
machined brass ﬁtting. Plastic caps, the polypropylene tube, and valve B from a second
RTubeTM were used to isolate the SPME ﬁber from the outside air as shown in Figure 3-
8c. The subject breathed at normal frequency and tidal volume through the mouthpiece

while holding their nose for 10 min. Aﬁer collection, the EBV-RT collected on the SPME

103

ﬁber was analyzed immediately. A second respirator can also be attached in series to

provide additional ﬁltration in selected studies described below.

3-16-4: Nail Polish Inhalation with Respirator Cartridge

Breath samples were then collected from a single subject while painting a standardized
1.5 in x 1.5 in cardboard square covered in packaging tape with nail polish. The
cardboard was used to simulate a fingernail while preventing the possibility of dermal
absorption of the compounds. Breath samples (EBV-RT) were obtained i) with no
modiﬁcation to the inhaled air, ii) with the inhaled air passing through an organic vapor
respirator, and iii) with the inhaled air passing through PVC tubing from a separate room.

Figure 3- 11 shows the chromatograms obtained in each case.

 

 

 

 

 

 

 

 

 

 

 

_._ E BV-RT without respirator
TOIUCI'IC __ EBV-RT with respirator
28 M .. \ Butyl acetate » l1 BV-R'l \ch tubing
A 24 M j Dnsopropoxy
=5 20 M j methane Mesityl Oxide (diacetone alcohol dehydration)
3 16 M ' \ . . Camphor
3 12 M . Ethyl acetate N ‘/ Diacetone alcohol (acetone condensation) j \
a 8 M J .__1 l M L A/ A A - A A - L
S 4M* L IL L . .4 .
-° . ﬁ‘. . n." f A- .* Niki- .‘
< 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50
Retention Time (min)

Figure 3- I 1: Total ion current chromatograms of breath samples from a single subject
(EB V-RI) collected using a variety of inhaled air sources: ambient air (red trace, top),
ambient air through a respirator (blue trace, middle), and ambient air from a separate
room than sampling (green trace, bottom). The subject painted a simulated ﬁngernail
with commercial nail polish during sampling. Compounds labeled are listed ingredients
of the nail polish or byproducts of ingredients. Chromatograms are oﬂset for clarity.
(Image originally in color)

104

It is immediately apparent that many foreign compounds are present in the breath
samples. The ingredients of the nail polish used were listed as butyl acetate, ethyl
acetate, toluene, nitrocellulose, tosylamide/formaldehyde resin, dibutyl phthalate,
isopropyl alcohol, stearalkonium hectorite, camphor, phthalic anhydride/trimellitic
anhydride/glycols copolymer, benzophenone-l, and guanine; many of these compounds
are seen in the chromatograms. Table 3- 6 shows the relative reduction in XIC peak area

of several compounds caused by the use of the respirator or PVC tubing.

Table 3- 6: Reduction in peak area for several ingredients of nail polish detected in
human breath obtained with respirator or PVC tubing. Peak areas were calculated by
integration of the HC for each compound, and are displayed relative to the XIC peak
area value obtained without modijjting the inhaled air source.

 

 

 

 

 

 

 

 

 

 

 

 

Peak Area Reduction (%)
Compound Extracted Ion (m/z) With Respirator With Tubing
Isopropyl Alcohol 45 9O 96
Ethyl Acetate 88 92 94
Diisopropoxy methane 73 89 96
Toluene 91 94 98
LMesityl oxide 83 94 99
Butyl acetate 56 92 98
Diacetone alcohol 43 96 94
Camphor (l) 152 95 100
Camphor (2) 152 97 99

 

 

 

 

 

 

There is a trend that compounds with a higher vapor pressure are less effectively removed

from the intake air by the respirator (e.g. VP of isopropyl alcohol = 30.75 torr; VP of

105

157, 158

camphor = 0.15 torr) . The large reductions in peak area (> 90%) demonstrate the

advantages of controlling the inhaled air of a subject. Two additional subjects repeated

the experiment using a respirator with similar results (peak area reduction RSD 0-1 7%).

3-16-5: Lemonade Inhalation with Respirator Cartridge

Figure 3- 12 shows the extracted ion chromatograms (m/z 93) of breath samples obtained
ﬁ'om a single subject inhaling directly above a glass containing approximately 125 mL of
room temperature lemonade, which contains many ﬂavor and aroma compounds. Breath

samples were obtained with and without the use of a respirator to cleanse the inhaled air.

 

 

 

 

 

 

 

 

— EBV-RT without respirator
21.2 M a-limone/n; ‘ . — EBV-RT with respirator
a; 1.0M , y-Terplnene
:0 3 M . 0. -Terp1nene a-terpineol
° ' B-pinene - Terpinolene '
g 0.6 M a -p1nene B myrcene 4-te meol
§04M \ \A/ H / K
.55 0.2M . - - - U
0

 

4:60 4:80 5200 5.20 5140 5160 5.80 6.00 6.20 6.40 6.60 6:80 7100 7220 7140
Retention Time (min)

Figure 3- 12: Extracted ion chromatograms (XIC m/z 93) of breath samples ﬁom a single
subject (EB V-R T). One sample was collected over 4 oz. of lemonade with no
modiﬁcation to the inhaled air (red trace, top) and another sample was collected over 4
oz. of lemonade using a respirator (blue trace, bottom). Peaks labeled are identiﬁed as

monoterpenes and monoterpene alcohols, ﬂavor ingredients present in lemonade. (Image
originally in color)

A large reduction in peak area is seen for all compounds present. The peak area percent

reduction relative to the sample without the respirator is shown in Table 3— 7. These data

106

show the effectiveness of the respirator for removing naturally-derived compounds, such

as monoterpenes, from the inhaled air.

Table 3- 7.“ Reduction in peak area for the monoterpenes detected in EB V-RT samples
obtained over lemonade. Peak areas were calculated by integration of the XIC for each
compound, and are displayed relative to the peak area value obtained without modiﬁiing
the inhaled air source.

 

 

 

 

 

 

Compound Peak Area .Reduction (%) Compound Peak Area .Reduction (%)
With Respirator With Respirator

a—pinene 82 3-carene 91

B—pinene 93 y-terpinene 98

B-myrcene 95 Terpinolene 98

a-terpinene 96 4-terpineol 100

a—limonene 97 a-terpineol 99

 

 

 

 

 

 

3-16-6: Gasoline Inhalation with Respirator Cartridge and Synthetic Lung Substitute
Development

Tests were then performed to determine the efﬁciency of the respirator-modiﬁed
RTubeTM when obtaining breath samples in a very dirty environment. Automobile-grade
gasoline was used to simulate dirty sampling conditions. To prevent unnecessary
exposure to harmﬁil chemicals a human subject was not used to collect these data; a
polyethylene synthetic lung substitute was developed to simulate human respiration. The
synthetic lung substitute was fabricated to approximate adult tidal volume (500 mL)159

and was pumped to simulate the inhalation and exhalation of a normal adult respiratory

pattern over the gasoline and through an RTubeTM.

107

Figure 3- 13 shows a portion of the total ion chromatogram of this synthetic breath
sample obtained using the synthetic lung substitute at a constant and steady respiration
rate. Many peaks are present in this chromatogram - numerous compounds have been

tentatively identiﬁed in Table 3- 8.

 

   
   
 

 

 

 

 

  
 

 

 

 

 

 

 

‘— -
—— EBV-RT without respirator \ “xylene m xylene
: EBV-RT with respirator / p-xylene
f? 60 M EBV-RT with two respirators teluene \
3 .
3 50 M meth 1 methyl
8 40 M y cyclohexane n-octane Propylbenzene
cyclopentane‘ x. \
30 M .
g 20 M l
< 10 M ‘r
1.50 2.00 2.50 3.00 3.50 4.00 4.50
Retention Time (min)

Figure 3- 13: Total ion current chromatograms of ‘breath ’ samples obtained using a
synthetic lung substitute over gasoline (EB V-R 7). One sample was collected with no
modification to the inhaled air (red trace, top), a second sample was collected using a
single respirator (blue trace, middle), and a ﬁnal sample was collected using two

respirators in series (green trace, bottom) Chromatograms are offset for clarity. (Image
originally in color)

Additional synthetic breath samples were obtained by attaching a single respirator
cartridge to the intake valve of the RTubeTM and repeating the synthetic lung substitute’s
respiration over gasoline, as well as attaching two respirator cartridges in series by
removing the particulate ﬁlter from one cartridge in order to attach the second. These
chromatograms are also shown in Figure 3- 13. The percent reduction in peak area of the
compounds detected in the sample obtained with one respirator cartridge versus no
respirator cartridge is shown in Table 3- 8. The peak area was reduced by 56-98% when a

single respirator is used, speciﬁcally with a greater decrease in peak area for smaller

108

molecular weight compounds.

additional reductions in peak areas were determined versus the sample obtained with one

respirator cartridge (Table 3- 8).

Table 3- 8: Reduction in peak area for the tentatively identified compounds detected in
EB V-RT samples generated with a synthetic lung substitute respirated over gasoline.
Peak areas were calculated by integration of the TIC for each compound, and are
displayed relative to the peak area value obtained without modifying the inhaled air
source. The additional peak area reduction demonstrated by the use of two respirators
was calculated compared to those values when using one respirator to demonstrate the

additional benefit of the respirator series.

With the addition of the second respirator cartridge,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

RT Compound Area Reduction (%) Additional Area Reduction (%)
One Respirator Two vs. One Respirator

1.44 Ethanol 93 86
1.48 Unidentiﬁed compound 97 82
l .5 1 2-methylbutane 97 102
1.56 Pentane 96 82
l .61 2-methyl-2-butene 94 94

1 .65 2,2-dimethy1butane 91 95

l .72 2-methy1pentane 97 59

1 .75 2,3-dimethylbutane 97 60
1.77 Unidentiﬁed alkane 98 100
1.81 3-methylpentane 98 59
1.85 2,2-dimethylheptane 98 34
1.88 Hexane 97 52
1.91 2,3-dimethyl-2—butene 97 68
2.03 2-cyclopropylpropane 98 22
2.05 Methylcyclopentane 97 70
2. 1 8 l-methylcyclopentene 96 49
2.27 Unidentiﬁed alkane 96 42
2.30 2,3-dimethylpentane 96 70
2.34 3-methylhexane 87 81
2.44 2,2,3,3-tetramethylbutane 95 68
2.52 Heptane 9O 61
2.76 Methylcyclohexane 92 50
2.78 3-methylheptane 91 109
2.84 Ethylcyclopentane 93 69
2.90 l ,2,4-Tiimethylcyclopentane 92 77
2.97 2,3,4-Trimethylpentane 91 68
3.04 2,3,3-Trimethylpentane 90 68

 

 

 

 

109

 

Table 3-8 (cont ’d).

 

 

 

 

 

 

 

 

 

 

 

 

 

3.15 Toluene 77 60
3.25 3,5,5-trimethyl-1-hexene 86 64
3.28 1,3-dimethylcyclohexane 87 75
3.36 1-ethyl-3-methylcyclopentane 83 56
p.38 (Z’Z'Dlmethyl 81 72
propylidene)cyclopropane

3.43 n-octane 68 66
3.79 2,5-dimethylheptane 71 71
4.10 o—xylene 61 66
4.29 iii-xylene 68 79
4.43 p¢xylene 58 70

.73 Isopropylbenzene 64 75
5.04 Propylbenzene 59 74
5.07 4-methylnonane 56 59

 

 

 

 

 

 

Depending on the concentration of contaminants in the ambient air, the respirator is
useﬁil in reducing the amount of carryover of the contamination into a breath sample, and
more than one can be used in extremely dirty environments. While not a perfect solution
to entirely eliminating background components from a breath sample, this method allows

for fast and facile sampling and analysis, with no power (freezer) requirements.

3-17: Conclusions

The data ﬁ'om this study demonstrate that real-time sampling of EBV is an effective and
advantageous method for breath analysis. As an active sampling method, collection is
fast; because sampling and preconcentration occur concurrently, the overall collection
and analysis time of a breath sample is reduced ~30% compared to EBC collection.
Sampling and SPME exposure time were optimized to yield the highest signal from the

GC-MS. Comparisons were made between the traditional EBC collection at -80°C, EBV

110

collection at -80°C, and EBV collection at room temperature. This study has shown that
EBV is more sensitive than EBC, yielding 23% more compounds than traditionally
collected EBC, as well as up to a lO-fold increase in peak area for many compounds.
This modiﬁcation is simple and thus maintains the disposable and inexpensive nature of
the RTubeTM. The sampling technique is ﬁeld portable as it requires minimal user
training and has zero power requirements or need for a ﬁozen condenser. Also
demonstrated herein was the need to control the quality and cleanliness of inhaled air
when collecting a breath sample in a dirty environment. The use of PVC tubing or a
respirator cartridge was established as a simple and effective method of providing inhaled
air yielding breath samples with 89-100% reduction in peak area for compounds inhaled
from nail polish, and 82-100% reduction in peak are for terpenes inhaled from lemonade.
The use of multiple respirator cartridges was also demonstrated, showing an additional
reduction in peak area of 22-109% when a simulated breath sample was obtained over
gasoline. Caution must be paid to the localized situational environment created at a
subject’s point of inhalation; compounds may be inadvertently inhaled at concentrations
greater than that in the general ambient air causing error in alveolar gradient calculations.
Collection and analysis of EBV using the methods described in this study will provide

valuable information for future breath analysis.

111

3 - 1 8: References

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

Breath Analysis for Biomedicine and National Security Sensor Design Issues and
Strategies for Biomarker Discovery, Davis, CA 2008.

DARPA Workshop: Breath Signatures, Livermore, CA 2008.

Stix, G. Sci. Am. 2003, 289.

Phillips, M. Sci. Am. 1992, 267, 74-79.

Bogen, E. California and Westem Medicine 1927, 26, 778-783.

Pauling, L.; Robinson, A. B.; Teranish.R; Cary, P. Proc. Natl. Acad Sci. U. S. A.
1971, 68, 2374-&.

Phillips, M.; Herrera, J.; Krishnan, S.; Zain, M.; Greenberg, J.; Cataneo, R. N. J.
Chromatogr. B 1999, 729, 75-88.

Mukhopadhyay, R. Anal. Chem. 2004, 76, 273A-276A.

Namjou, K.; Roller, C. B.; McCann, P. J. IEEE Circuits & Devices Magazine
2006, 22, 22-28.

Hunt, J. F.; Fang, K.; Malik, R.; Snyder, A.; Malhotra, N.; Platts-Mills, T. A. E.;
Gaston, B. Am. J. Respir. Crit. Care Med. 2000, 161, 694-499.

Corradi, M.; Folesani, G.; Andreoli, R.; Manini, P.; Bodini, A.; Piacentini, G.;
Carraro, S.; Zanconato, S.; Baraldi, E. Am. J. Respir. Crit. Care Med. 2003, 167,
395-399.

Wood, L. G.; Gibson, P. G.; Garg, M. L. Eur. Respir. J. 2003, 2], 177-186.

Corradi, M.; Rubinstein, 1.; Andreoli, R.; Manini, P.; Caglieri, A.; Poli, D.;
Alinovi, R.; Mutti, A. Am. J. Respir. Crit. Care Med. 2003, I 67, 1380-1386.

112

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

Gonzalez-Reche, L. M.; Kucharczyk, A.; Musiol, A. K.; Kraus, T. Rapid
Commun. Mass Spectrom. 2006, 20, 2747-2752.

Phillips, M.; Cataneo, R. N.; Ditkoff, B. A.; Fisher, P.; Greenberg, J.;
Gunawardena, R.; Kwon, C. S.; Tietje, 0.; Wong, C. Breast Cancer Res. Treat.
2006, 99, 19-21.

Phillips, M.; Cataneo, R. N.; Ditkoff, B. A.; Fisher, P.; Greenberg, J.;
Gunawardena, R.; Kwon, C. S.; Rahbari-Oskoui, F .; Wong, C. The Breast Journal
2003, 9, 184-191.

Phillips, M.; Cataneo, R. N‘.; Greenberg, J .; Gunawardena, R.; Naidu, A.; Rahbari-
Oskoui, F. J. Lab. Clin. Med. 2000, 136, 243—249.

Phillips, M.; Altorki, N.; Austin, J. H. M.; Cameron, R. B.; Cataneo, R. N.;
Greenberg, J .; Kloss, R.; Maxﬁeld, R. A.; Munawar, M. 1.; Pass, H. I.; Rashid, A.;
Rom, W. N.; Schmitt, P. Cancer Biomarkers 2007, 3, 95-109.

Phillips, M.; Cataneo, R. N.; Cummin, A. R. C.; Gagliardi, A. J.; Gleeson, K.;
Greenberg, J.; Maxﬁeld, R. A.; Rom, W. N. Chest 2003, 123, 2115-2123.

Gordon, S. M.; Szidon, J. P.; Krotoszynski, B. K.; Gibbons, R. D.; Oneill, H. J.
Clin. Chem. 1985, 31, 1278-1282.

Phillips, M.; Gleeson, K.; Hughes, J. M. B.; Greenberg, J.; Cataneo, R. N.; Baker,
L.; McVay, W. P. The Lancet 1999, 353, 1930-1933.

Oneill, H. J.; Gordon, S. M.; Oneill, M. H.; Gibbons, R. D.; Szidon, J. P. Clin.
Chem. 1988, 34, 1613-1618.

Poli, D.; Carbognani, P.; Corradi, M.; Goldoni, M.; Acampa, 0.; Balbi, B.;
Bianchi, L.; Rusca, M.; Mutti, A. Respiratory Research 2005, 6.

Di Natale, C.; Macagnano, A.; Martinelli, E.; Paolesse, R.; D'Arcangelo, G.;
Roscioni, C.; Finazzi-Agro, A.; D'Amico, A. Biosens. Bioelectron. 2003, 18,
1209-1218.

113

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

Phillips, M.; Cataneo, R. N.; Condos, R.; Erickson, G. A. R.; Greenberg, J.; La
Bombardi, V.; Munawar, M. 1.; Tietje, O. Tuberculosis 2007, 87, 44-52.

Ruzsanyi, V.; Baumbach, J. I.; Sielemann, S.; Litterst, P.; Westhoff, M.; Freitag,
L. J. Chromatogr. A 2005, 1084, 145-151.

Phillips, M.; Boehmer, J. P.; Cataneo, R. N.; Cheema, T.; Eisen, H. J.; Fallon, J.
T.; Fisher, P. E.; Gass, A.; Greenberg, J.; Kobashigawa, J .; Mancini, D.; Rayburn,
B.; Zucker, M. J. Am. J. Cardiol. 2004, 94, 1593-1594.

Phillips, M.; Boehmer, J. P.; Cataneo, R. N.; Cheema, T.; Eisen, H. J.; Fallon, J.
T.; Fisher, P. E.; Gass, A.; Greenberg, J.; Kobashigawa, J.; Mancini, D.; Rayburn,
B.; Zucker, M. J. J. Heart Lung Transplant. 2004, 23, 701-708.

Sobotka, P. A.; Gupta, D. K.; Lansky, D. M.; Costanzo, M. R.; Zarling, E. J. J.
Heart Lung Transplant. 1994, 13, 224-229.

Deng, C.; Zhang, J .; Yu, X.; Zhang, W.; Zhang, X. J. Chromatogr. B 2004, 810,
269-275.

Phillips, M.; Cataneo, R. N.; Cheema, T.; Greenberg, J. Clin. Chim. Acta 2004,
344, 189-194.

Exhaled Breath Condensate; Respiratory Research, lnc.: Charlottesville, VA,
2005.

Amorim, L. C. A.; Cardeal, Z. d. L. J. Chromatogr. B 2007, 853, 1-9.

Chinery, R. L.; Gleason, A. K, 1993; Vol. 13, pp 51-62.

Giardino, N. J.; Gordon, S. M.; Brinkman, M. C.; Callahan, P. J.; Kenny, D. V.
Epidemiology 1999, 10, S72-S72.

Lindstrom, A. B.; Pleil, J. D. J. Air Waste Manage. Assoc. 1996, 46, 676-682.

Lindstrom, A. B.; Pleil, J. D.; Berkoff, D. C. Environ. Health Perspect. 1997, 105,
636-642.

114

(38)

(39)

(40)

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(48)

(49)

(50)

(51)

Maniscalco, M.; De Laurentiis, G.; Pentella, C.; Mormile, M.; Sanduzzi, A.;
Carratu, P.; Soﬁa, M. Biomarkers 2006, 11, 233-240.

Raymer, J. H.; Pellizzari, E. D.; Thomas, K. W.; Cooper, S. D. J. Expo. Anal.
Environ. Epidemiol. 1991, I, 439-451.

Wallace, L. A. Annual Review of Energy and the Environment 2001, 26, 269-301.

Wallace, L. Environ. Health Perspect. 1996, 104, 1129-1136.

Wallace, L.; Pellizzari, E.; Gordon, S. J. Expo. Anal. Environ. Epidemiol. 1993, 3,
75-102.

Giardina, M.; Olesik, S. V. Anal. Chem. 2003, 75, 1604-1614.

Giardina, M.; Olesik, S. V. Anal. Chem. 2003, 75, 1604-1614.

Pleil, J. D.; Lindstrom, A. B. Journal of Chromatography B-Biomedical
Applications 1995, 665, 271-279.

Pleil, J. D.; Fisher, J. W.; Lindstrom, A. B. Environ. Health Perspect. 1998, 106,
573-580.

Pleil, J. D; Smith, L. B.; Zelnick, S. D. Environ. Health Perspect. 2000, 108,
183-192.

Wallace, L. A.; Nelson, W. C.; Pellizzari, E. D.; Raymer, J. H. J. Expo. Anal.
Environ. Epidemiol. 1997, 7, 141-163.

"Exhaled Breath Condensate and Other Markers in Exhaled Air" In Pediatric
Pulmonary Function Testing; Eber, H. 1., Ed.; Karger, 2005; Vol. 33, pp 190-202.

Schneideler, 1.; Manke, H. G.; Schwulera, U.; Imacker, P.; Hammerle, H. Am.
Rev. Respir. Dis. 1993, 148, 778-784.

Chen, H. W.; Wortmann, A.; Zhang, W. H.; Zenobi, R. Angewandte Chemie-
International Edition 2007, 46, 580-583.

115

(52)

(53)

(54)

(55)

(56)

(57)

(58)

(59)

(60)

(61)

(62)

(63)

(64)

(65)

(66)

Cunningham, S.; McColm, J. R.; Ho, L. P.; Greening, A. P.; Marshall, T. G. Eur.
Respir. J. 2000, 15, 955-957.

Soyer, O. U.; Dizdar, E. A.; Keskin, 0.; Lilly, C.; Kalayci, 0. Allergy 2006, 61,
1016-1018.

Reinhold, P.; Jaeger, J .; Schroeder, C. Biomarkers 2006, 11, 118-142.

Vass, G.; Huszar, E.; Barat, E.; Valyon, M.; Kiss, D.; Penzes, L.; Augusztinovicz,
M.; Horvath, I. Am. J. Respir. Crit. Care Med. 2003, 167, 850-855.

Skeldon, K. D.; Patterson, C.; Wyse, C. A.; Gibson, G. M.; Padgett, M. J.;
Longbottom, C.; McMillan, L. C. J. Opt. A 2005, 7, $376-$384.

Lord, H.; Yu, Y. F.; Sega], A.; Pawliszyn, J. Anal. Chem. 2002, 74, 5650-5657.

Ma, W.; Liu, X. Y.; Pawliszyn, J. Anal. Biochem. 2006, 385, 1398-1408.

Yu, H.; Xu, L.; Wang, P. J. Chromatogr. B 2005, 826, 69-74.

Lawson, L. D.; Wang, Z. J. J. Agric. Food Chem. 2005, 53, 1974-1983.

Carpenter, C. T.; Price, P. V.; Christman, B. W. Chest 1998, 114.

Phillips, M. Anal. Biochem. 1997, 247, 272-278.

Phillips, M.; Cataneo, R. N.; Greenberg, J.; Grodman, R.; Gunawardena, R.;
Naidu, A. Eur. Respir. J. 2003, 21, 48-51.

Phillips, M.; Greenberg, J.; Cataneo, R. N. Free Radic. Res. 2000, 33, 57-63.

Raymer, J. H.; Pellizzari, E. D.; Thomas, K. W.; Kizakevich, P.; Cooper, S. D. J.
Expo. Anal. Environ. Epidemiol. 1992, 2, 67-83.

Statheropoulos, M.; Agapiou, A.; Georgiadou, A. J. Chromatogr. B 2006, 832,
274-279.

116

(67)

(68)

(69)

(70)

(71)

(72)

(73)

(74)

(75)

(76)

(77)

(78)

(79)

(80)

(81)

(82)

Wallace, L. A.; Pellizzari, E. D. Environ. Health Perspect. 1995, 103, 95-98.

NIOX Flex: Technical Specification; Aerocrine AB: Solna, Sweden.

Ohira, S. 1.; Li, J. H.; Lonneman, W. A.; Dasgupta, P. K.; Toda, K. Anal. Chem.
2007, 79, 2641-2649.

Walsh, J. Spectroscopy 2004, I 9, 49-50.

Frishman, G.; Tzanani, N.; Amirav, A. Field Anal. Chem. Tech. 2001, 5, 107-115.

Libardoni, M.; Stevens, P. T.; Waite, J. H.; Sacks, R. J. Chromatogr. B 2006, 842,
13-21.

Raymer, J. H.; Thomas, K. W.; Cooper, S. D.; Whitaker, D. A.; Pellizzari, E. D. J.
Anal. T oxicol. 1990, 14, 337-344.

Gordin, A.; Amirav, A. J. Chromatogr. A 2000, 903, 155-172.

Sanchez, J. M.; Sacks, R. D. Anal. Chem. 2003, 75, 2231-2236.

Sanchez, J. M.; Sacks, R. D. Anal. Chem. 2006, 78, 3046-3054.

Menssana Research, lnc.:Breathtaking Technology,

Berlardi, R.; Pawliszyn, 1. Water Pollution Research Journal of Canada 1989, 24,
179-191.

Zhang, Z.; Pawliszyn, J. Anal. Chem. 1993, 65, 1843-1852.

Louch, D.; Mothlagh, S.; Pawliszyn, J. Anal. Chem. 1992, 64, 1187-1199.

Grote, C.; Pawliszyn, J. Anal. Chem. 1997, 69, 587-596.

Phillips, M.; Greenberg, J. Clin. Chem. 1992, 38, 60-65.

117

(83)

(84)

(85)

(86)

(87)

(88)

(89)

(90)

(91)

(92)

(93)

(94)

(95)

(96)

Wallace, L.; Buckley, T.; Pellizzari, E.; Gordon, S. Environ. Health Perspect.
1996, 104, 861-869.

Effros, R. M.; Biller, J.; Foss, B.; Hoagland, K.; Dunning, M. B.; Castillo, D.;
Bosbous, M.; Sun, F .; Shaker, R. Am. J. Respir. Crit. Care Med. 2003, 168, 1500-
1505.

Wanner, A.; Mendes, E. S.; Atkins, N. D. J. Appl. Physiol. 2006, 100, 1674-1678.

Muhlberger, F .; Streibel, T.; Wieser, J.; Ulrich, A.; Zimmermann, R. Anal. Chem.
2005, 77, 7408-7414.

"An Improved Method for the Biological Monitoring of Volatile Compounds" In
Environmental Health in Central and Eastern Europe; Donnelly, K. C., Cizmas,
L. H., Eds.; Springer: Dordrecht, The Netherlands, 2006, pp 65-71.

Phillips, M.; Greenberg, J.; Awad, J. J. Clin. Pathol. 1994, 47, 1052-1053.

Phillips, M.; Greenberg, J .; Sabas, M. Free Radic. Res. 1994, 20, 333-337.

De Hoffman, E.; Charette, J.; Stroobant, V. Mass Spectrometry: Principles and
Applications; John Wiley & Sons: New York, 1996.

James, A. T.; Martin, A. 11. P. Biochem. J. 1952, 50, 679-690.

Martin, A. J. P.; Synge, R. L. M. Biochem. J. 1941, 35, 1358-1368.

Merrick, W., Massachusetts Institute of Technology, Cambridge, MA, 2004.

Oser, H.; SRI International: Davis, CA, 2008.

Hyspler, R.; Crhova, S.; Gasparic, J.; Zadak, Z.; Cizkova, M.; Balasova, V.
Journal of Chromatography B: Biomedical Sciences and Applications 2000, 739,
183-190.

Ahluwalia, V. K.; Parashar, R. K. Organic Reaction Mechanisms, 2nd ed.; Alpha
Science International Ltd.: Harrow, UK, 2005.

118

(97) Knovel Critical Tables; Knovel, 2003.
(98) Wickstrom, E. 1988; INCHEM.

(99) Palsson, B. P.; Hubbell, J. A.; Plonsey, R.; Bronzino, J. D. Tissue Engineering;
CRC Press: San Diego, CA, 2003.

119

CHAPTER 4: Measuring Chemical Exposure using Breath Analysis

4-1: Motivations and Introduction

l-Iuman exposure to environmental chemicals is a reality; the presence of chemicals in the
work place, in food, and in association with common human activities, such as visiting a
gas station, or using moth balls for clothing storage, puts all humans in contact with many
chemicals. While exposure to some of these chemicals may be harmless, some
compounds elicit toxicity in individuals exposed to high doses. Rapid assessment and
monitoring of chemical exposure is important in order to diagnose the exposure and
select the appropriate medical treatment to minimize its physiological effects. The ideal
technique would provide instantaneous identiﬁcation of the chemical in question, provide
information on its concentration within the subject, monitor the rise and fall of this
concentration, provide information about the route of exposure, and monitor the subject
to ensure no further exposure is occurring. If the chemical of interest is toxic, this
information can be used to determine the appropriate treatment plan for the subject; if the
chemical is non-toxic, it is still useﬁil to monitor its uptake and elimination in the body in
order to learn about metabolic and excretion processes. If the substance is a chemical of
interest for forensic and homeland security applications, the presence of the chemical
documents that exposure has occurred. However, it would also be useful to gain some
information about the time and extent of exposure. Detection of chemicals unique to a
particular environment may also provide information on the prior location of an

individual and could be used to place a subject at a particular location for forensic

purposes.

120

This chapter will describe several studies designed to test the abilities of breath analysis
for use as a means to detect chemical exposure. This chemical exposure is targeted in an
effort to document occupational exposure, determine the chemical signature of unique
locations, or determine the prior location or activity of an individual. These experiments
will demonstrate that breath analysis is an effective technique for all of these purposes,
and the implications of sensitive analysis of breath constituents will also be discussed in

regards to a ﬁeld study for the detection of industrial chemicals.

4—2: Background

Regardless of the exposure route (e. g. dermal, inhalation, oral), breath analysis provides
an alternative method of monitoring exposure of humans to chemicals. The non-invasive
nature of breath sampling allows a greater number of samples to be collected from a
subject owing to the higher tolerance of subjects to the collection method compared to
subject willingness to provide blood samples, which is commonly painful. Another
advantage of breath collection is that the biological waste is limited. Research at the US
Environmental Protection Agency (EPA) has shown that exhaled breath can be an
accurate surrogate for blood analysis when monitoring chemical exposure to
trichloroethene.1 Exhaled breath comes ﬁom the lungs, where the inspired air difﬁises
through tissues and into the blood. Chemicals in the blood strive for equilibrium between
the alveolar air and the capillaries in the lung, passing through the pulmonary alveolar
membrane in order to achieve this. Therefore, volatile compounds that are in circulation

in the bloodstream can be detected by sampling exhaled breath.

121

Breath analysis has been extensively reported in the literature for medical applications,
but additional studies focused on the exogenous compounds in breath — speciﬁcally those
due to a chemical exposure}9 have been reviewed by Wallace.10 Many studies of
chemical exposure focus on the dermal and inhalation routes of exposure. Raymer et al.
studied the effect of inhalation exposure in several microenvironments, including
common work places including a woodworking shop; 21 VOCs were monitored and their
elimination ﬁom the body was described by a two-compartment model of the body.”
Work at the US EPA has focused on methods development for monitoring inhalation
exposures in several microenvironments in part through the Total Exposure Assessment
Methodology (TEAM) program”’ 13 and in part through controlled exposures.”17 The
TEAM program developed breath sampling methods to analyze the VOCs in human
breath of subjects in various cities in the United States, testing levels of VOCs in their
normal environments as well as after exposure in several environments or to several
household products, and again ﬁt the acquired data of elimination using a two-
cornpartment model of the body.13 Dermal exposure was also probed in these studies, for
example, comparisons were made between the levels of chloroform in breath samples
collected while the subjects showered with and without rubber suits, isolating inhalation
exposure versus inhalation and dermal exposure, where the levels detected were twice
that detected while the rubber suits were worn.12 Several studies have also researched
trihalomethane exposure from sources such as chlorinated water in showers or swimming
pools and modeled the uptake and decay of such compounds with a 3-compartment
model.1‘6’ 18‘ 19 These studies provided insight into the VOCs detected in breath, common

environmental sources of exposure, as well as common routes of exposure.

122

4-2-1: Industrial/Commercial Exposure

One example of industrial chemical exposure occurs in the nail salon. Nail salons are
poorly regulated, and safety procedures are recommended, but not enforced, and there is
a general lack of concern of chemical exposure by salon technicians.20 Chemical
exposure in the nail salon can occur from the nail polish, nail polish remover, acrylic
nails and other sources. Workers in the salon may wear a mask, but generally not one
that would seal against the face and prevent inhalation exposure. During the application
of acrylic nails, ethyl methacrylate and silica exposure can be high for the technicians,
whose breathing zone is 6-12 in from the ﬁngernails. The ﬁling and sanding of such
nails can produce much dust containing acrylic monomers, ﬁberglass, and silica.
Exposure to such chemicals has been shown to cause asthma, contact dermatitis, and

even silicosis?" 22

A study by the Illinois Department of Public Health assessed
exposures of several nail technicians in salons throughout Illinois. In this study, none of
the subjects exceeded the National Institute for Occupational Safety and Health (NIOSH)
Recommended Exposure Limits (REL) or the Occupational Safety and Health
Administration (OSHA) Permissible Exposure Limits (PEL) for respirable silica dust

exposures,23 but the proximity of the technician to the chemicals as well as an intense

smell of ethyl methacrylate was noted.

The EPA has also issued comments about the health and chemical exposures of nail salon
technicians. They identify acetone, benzoyl peroxide, butyl acetate, butyl methacrylate,
camphor, dibutyl phthalate, ethyl acetate, ethyl cyanoacrylate, ethyl methacrylate,
formalin, hydroquinone, isobutyl methacrylate, methacrylic acid, 4-methoxyphenol,

methyl ethyl ketone, methyl methacrylate, poly (ethyl/methyl methacrylate), titanium

123

dioxide, toluene, and tosylamide formaldehyde resin as common ingredients in nail care
products.24 Exposure may occur through inhalation or absorption via the skin and
ﬁngernails, and may cause medical trouble including asthma, cancer, lung ﬁbrosis, or
shortness of breath. Monitoring exposure to such chemicals would provide valuable
information about the risks posed to the health of the subject, as well as the amount of

time spent in a particular location.

Other industrial and commercial locations also pose the risk of dangerous chemical
exposure. Manufacturing plants and well as retailers of chemicals may present high
levels of solvents in vapor or liquid form. These locations are regulated by OSHA and
NIOSH, but chemical signatures unique to these locations may be present and may

provide information on the prior location of an individual if detected on breath.

4-2-2: Exposure via Ingestion — Terpenes and Terpenoid Compounds

Terpenoid compounds are natural organic compounds, which are derived from several
isoprene building blocks, (C5H3)n. Terpenoids and terpenes are present in many natural
products, including fruit oils, ﬂowers, and turpentine. Several studies have described the
elimination of monoterpenes from the human body after exposure to these compounds.
Levin et al. showed that 8% of inhaled a-pinene was eliminated unmetabolized in
exhaled breath, with less than 0.001% excreted via the kidneys, thus leaving the majority
of inhaled a-pinene stored in the body or excreted as a metabolite.25 Falk-Fillipson
reported that the respiratory clearance of 3-carene was signiﬁcantly higher when the
subject was exposed to the pure monoterpene, rather than a mixture of monoterpenes in

turpentine.26 The decay of three monoterpenes in blood was also ﬁt to a 3-compartment

124

model.26

Such a model describes how the organs and processes of the body are
simpliﬁed; a l-compartment model assumes that a substance is uniformly distributed
throughout the body via the same kinetic processes. A 2-compartment model assumes
that the organs with highest blood ﬂow form the central compartment and have uniform
rapid distribution of a substance, then the rest of the body forms the peripheral
compartment which experiences a slower rate of distribution. Likewise, a 3-compartment
model assume that the central compartment distributes into two compartments with
slower, and different, kinetics.27 Wallace et al. ﬁt the decay of signal in exhaled breath
after inhalation exposure to a-limonene to a 3-compartment model, but were unable to
gather information on the ﬁrst compartment decay due to its short half-life.l4 Few studies
have described the pharmacokinetics of monoterpenes and terpenoid compounds after
oral ingestion, especially after consumption of a mixture of compounds. When orally
dosed, ﬁrst-pass metabolism in the liver affects the amount of unmetabolized compound
present in the bloodstream.27 Chen et al. orally dosed rates with a-limonene, ﬁtting the

resulting decay curve in blood to a biexponential model.28

It has been reported that
although the monoterpenes are metabolized in the body, these metabolites were detected

only in the urine and only the original monoterpenes were detected in the blood.29

4-2-3: Research Directions

Due to the ubiquitous nature of VOCs in several environments, detection of these VOCs
in the breath of a subject may provide insight into their prior location or activities.
Although visiting a nail salon or hardware store is not an activity of forensic interest,

these systems may be used as surrogates for exposure to VOCs in more illicit

125

environments, or to more toxic VOCs. The accessible nature of these environments
makes them ideal to study the presence and elimination of such VOCS in the breath of a
subject. The results from these studies can then be used to predict the uptake and
elimination of other VOCS of interest in the breath of subjects that have been exposed.
Similarly, while the detection of terpenes in breath is not indicative of any criminal
activity, these compounds are easily accessible and provide further insight into the uptake
and elimination of VOCs in breath. As presented below, the detection of terpenes in
breath allows many data points to be acquired during the elimination of the compounds
from the body, allowing greater information on the half-lives and elimination processes

of these chemicals in breath to be elucidated.

4 -3 .' Materials V and Methods

4-3-1 : Materials

D-Camphor was obtained from J.T. Baker, Inc. (Phillipsburg, NJ). A 65.3 mM solution
was made in a 1:4 (vzv) solution of HzOzmethanol and stored in a glass vial. Hexamine
(hexamethylenetetramine) was obtained from Chem Service, Inc. (West Chester, PA).
Frozen lemonade concentrate (Safeway/Vons Lemonade, Safeway, Inc., Pleasanton, CA)
was purchased from local sources. Original lemonade was mixed according to the
manufacturer’s instructions. Concentrated lemonade was mixed using half the water
required by the manufacturer’s instructions. Mediterranean lemonade was made by
mixing 10 thinly sliced lemons, 2 cups of sugar, and 5 cups of water. (18)-(-)-B-pinene

(99%) was obtained from Alfa Aesar (Ward Hill, MA) and +/- a—limonene (95%) was

126

11 iii

".3

a ..5

r0
.

n)

my

obtained from TCI America, Inc. (Portland, OR). Serial dilutions of both compounds

were made using deionized water.

4-3-2: Methods
The RTubeTM (Respiratory Research, Inc., Charlottesville, VA), was used to collect all

breath samples. The subject inhales via a mouthpiece through a one-way valve that
prevents salivary contamination of the EBC or EBV sample. The subject then exhales
through the same mouthpiece via a second one-way valve at the base of a polypropylene
tube which serves as the BBC or EBV collection media. The polypropylene tube is used
at room temperature (EBV) or cooled by a previously frozen (-80 C) aluminum sleeve
(BBC). The subject breathes at a normal tidal volume and frequency though the

RTubeTM. Subjects in the environmental exposure study held their nose during collection.

When EBC is collected, 0.5 mL of EBC was aliquoted into a disposable polypropylene
microcentrifuge tube (Fisher Scientiﬁc, Pittsburgh, PA). A previously thermally
conditioned SPME ﬁber (65 um PDMS/DVB Stableﬂex ﬁber, Sigma-Aldrich, St. Louis,
MO), was inserted directly into this aliquot of EBC for 10 minutes, followed by injection
into the GC-MS. When EBV is studied, the breath sample is simultaneously collected

and preconcentrated on the SPME ﬁber, which is then injected into the GC-MS.

An Agilent (Santa Clara, CA) 6890N GC with an Agilent 5973N mass spectrometer
(Santa Clara, CA), or a Shimadzu (Pleasanton, CA) QP2010 GC-MS were used for
analysis. In both instruments, the Helium carrier gas was held at a ﬂow rate of 1.3 L min’

1 (99.999%, Praxair, Inc., Danbury, CT). Both instruments were operated using 70 eV

127

 

electron ionization, 3.54 scans s", operated in scan mode (m/z 40 - 450). ChemStation
software was used for data collection and analysis with the Agilent system; GCMS
Solutions software was used with the Shimadzu system for data processing. An Agilent
J&W DB-SMS column (30 m x 0.25 mm x 0.25 pm, equivalent to 5% phenyl, 95%

methylpolysiloxane) was used for all experiments.

Compounds were identiﬁed by comparing mass spectra obtained in these experiments
with those found in the NIST Mass Spectral Library (RMatch >700, NIST MS Search
2.0, NIST, Gaithersburg, MD) and the literature, as well as comparison of retention time

and mass spectra obtained from pure chemicals when available.

4-4: Subjects

Thirteen healthy adults with no known respiratory ailments were recruited for this study.
Subjects were asked to report any chemical exposure (none) and gave formal written
consent prior to sample collection. The Institutional Review Boards (IRB) at Lawrence
Livermore National Laboratory (LLNL) and Michigan State University approved

experimental protocols used for this study.

4-5.‘ Nail Salon Study
In order to provide an isolated inhaled air source and prevent contamination from
localized chemical exposures (i.e. chemical residues remaining on clothing or

skin/fmgernails), the RTubeTM was modiﬁed as follows. Approximately twelve feet of

FDA-approved plastic tubing (0.5 in I.D., 0.75 in OD, VWR, Inc., West Chester, PA)

128

 

 

was attached to the base of the intake valve on the RTubeTM using a plastic ﬁtting. This
tubing was extended into a different air space (room) than that in which the subject
donated a breath sample. The breath sample (EBC) was obtained using the standard
procedure with the aluminum sleeve cooled to -80 C, or EBV was obtained using the

modiﬁed procedure with no aluminum sleeve.

A 10 minute breath sample was obtained from a subject early in the morning. The
subject then visited a nail salon; after approximately 45 minutes in the salon, the subject
returned to the sampling area and gave additional 10 minute breath samples over a 4 hour
period. Figure 4- 1 shows the chromatogram obtained from the breath sample obtained
before visiting the salon versus that obtained 12 minutes after leaving the salon (inverted
for comparison). The majority of peaks observed in the GC-MS total ion chromatograms
are present in both samples, and are attributed to endogenous compounds in the subject’s
breath; however additional peaks, such as those highlighted by the arrows, are present in
the sample obtained after salon exposure. The use of the plastic tubing adapter connected
to the RTubeTM during breath collection provided inhaled air from a different room. This
prevented inadvertent chemical exposure of the subject while obtaining a breath sample,
as chemicals from the nail salon may have absorbed onto the clothing of the subject, as
well as remained on the frngemails of the subject. Previous data have shown that this
approach provides effective prevention of subject inhalation of the local chemical source

(substances applied to ﬁngernails).

129

 

 

 

 

 

 

 

BeforeSalon
27 2.51
53; 2.0‘
”O 1.51
'73 1.0'
8 0.5' J l
g 0 I“ 'WIV'*W‘1r-111'~v 47v -——w—~.,_.—~.rT—~
3 -0.5' l I
.D .
<9 -1'0 12MinAfterSalon
g -151

 

0.0 210 410 6.0 8.0 10.0 1210 121.0 16.0
Retention Time (min)

Figure 4— I .' GC-MS total ion chromatograms of sampled breath constituents obtained
before salon exposure (top) and 12 minutes after leaving the salon (bottom). The signal
obtained after leaving the salon was inverted for clarity. Many peaks endogenous to
breath are present in both samples. Additional peaks are present in the sample taken
after the subject left the salon, such as those indicated by the arrows. (Image originally in
color)

Figure 4- 2 shows a narrow retention time window in the chromatograms generated ﬁom
breath samples from a human subject collected before visiting the salon, and 12, 50, 99,
152, and 214 minutes after leaving the salon. There is a sharp increase in the total ion
chromatogram signal in this region immediately after salon exposure, which decays back
to near baseline over several hours. The source of this peak was identiﬁed as camphor, a
common ingredient in nail polish, and listed as an ingredient in the particular polish used
on the subject. Camphor was also identiﬁed in a SPME sample obtained from the
ambient air in the salon, conﬁrming that the subject’s inhaled air contained this
compound. The area of the peak due to camphor was calculated using the extracted ion

chromatogram (XIC) of m/z 152. The decay of the peak area can be ﬁt to an exponential

130

 

(R2=0.999), as is shown in Figure 4- 2b. The half-life of the decay was calculated to be

 

 

 

 

23.5 min.

a ) b.)
190000 7000

’5 170000 Timepost A 6000

3 150000 exposure 3 5000‘

‘0’ 130000 g l

O

5 110000 < 4000

E 90000 ‘35, 3000‘

'2 70000 $ 2000.

E 50000 E 1000. .
30000 m 0 \l\—'—l
10000 ........

6.91 6.92 6.93 0 50 100 150 200 250
Retention Tune (min) Time After Exposure (min)

Figure 4- 2: (0) Selected window of GC-MS total ion chromatograms obtained before
salon exposure (black), and 12, 50, 99, 152, and 214 minutes after leaving the salon The
source of the peak was identiﬁed as a common nail polish ingredient, camphor.
Chromatograms are slightly offset for clarity. (b) Exponential decay, R2 = 0.999, of the
area of the peak due to camphor; calculated from the XIC (m/z I 52) for the molecular ion
(chromatogram not shown). (Image originally in color)

The subject made an additional visit to the nail salon, and a polish without camphor as a
listed ingredient was applied to the subject’s ﬁngernails. Breath samples were obtained
using EBV collection with the plastic tubing supplying inhaled air from a separate
location than the subject. A single breath sample was obtained ﬁ‘om a subject early in the
morning. The subject then visited the nail salon; after approximately 45 minutes in the
salon, the subject returned to the sampling area and gave additional breath samples over a

5 hour period. Figure 4- 3a shows the retention time window in the XICS for m/z 91

131

t.

 

 

generated ﬁ'om samples obtained before visiting the salon, and 11, 25, 52, 79, 121, 168,
222, 285, and 313 minutes after leaving the salon. There is a sharp increase in signal in
this region immediately after salon exposure, which decays back to near baseline over
several hours. The source of this peak was identiﬁed as toluene, which is also a common
ingredient in nail polish and is listed on the polish used on the subject. Toluene was also
seen in the sample obtained from the headspace of the salon. The area of this peak was
calculated using the XIC for m/z 91. The decay of the peak area was ﬁt to an exponential
(R2=0.960), as is shown in Figure 4- 3b. The half-life of the decay was calculated to be
21.4 min. A small amount of camphor was detected in the sample collected 11 minutes
after exposure in the salon. Though it was not a constituent in the polish used, camphor
was still present in the air of the salon, so the subject was exposed even when care was

taken to use a polish free of camphor.

   
 

 

 

 

 

’7 a. =5 b.)
$25 ) Timepost $12 '.'
g exposure 310 1
N
§20 E
o 8
o D
515 52 6 .
E a
31° “,2 4
<1 8 - I I
g 5 a 2 I I I I
: “sun
ox 0‘ " " g 0
N 0.. 0 100 200 . 300
B 4.50 4.60 4.70 4.80 Time After Exposure (mm)

Retention Time (min)

Figure 4- 3: (0) Selected window of GC—MS 20C for m/z 91showing the peak due to
toluene in breath before (no peak detected) and, II, 25, 52, 79, 121, I68, 222, 285, and
313 minutes after salon exposure. (b)Decay of the area of the peak due to toluene ﬁt to an
exponential process (R2 =0. 960). (Image originally in color)

132

L.‘

.4
'\ .

..

4‘.

;

I.‘

i..—

..‘.au-u1- f4

4-5-1: Headspace Analysis of Manicured Hands

In order to support the hypothesis that the subject was being reexposed to the nail salon
chemicals as they outgassed ﬁom the nails and hands, studies were performed to
determine the composition of the localized headspace surrounding the hands. At 301
minutes after exposure at the nail salon, one hand of the subject was placed in a
resealable plastic bag with a SPME ﬁber for 5 minutes, allowing the VOCs outgassing

from the nails and hands to be collected. The resulting TIC is shown in Figure 4- 4.

 

 

35 ‘ I
.4 r
,5 n-butyl acetate \
,3 8 . Diacetone alcohol
f? 7 Toluene\ / Camphor
g 6: l ‘ \l
5: I
g 4 j 1 6-methyl-5-hepten-2-one '
<2 3- ‘ i
a .
.2 2. l
1 i
i ‘ big/m _)Li‘ L .L._1.4\;s .11“ Maxi ALI. 11.11 4. LA xi.
0 1 2 3 4 5 6 7 8

Retention Time (min)

Figure 4- 4: GC-MS total ion chromatogram obtained by exposing a SPME fiber to the
manicured hand inside a plastic bag 5 hours, 1 minute after salon exposure. Note that
chemicals indicative of the nail polish are still present in the headspace of the nails.

Many chemicals were measured in this sample, including toluene, n-butyl acetate,
diacetone alcohol, 6-methyl-5-hepten-2-one, and camphor, common ingredients in nail

polish and listed as hazardous ingredients in nail care products by the EPA.24 That these

133

 

chemicals remain in the headspace of the ﬁngernails over 5 hours after exposure in the
salon indicates that the products used in nail salons release numerous VOCS to which nail
salon workers and clients can be continually exposed, even outside of the salon. Thus,
the localized environment from which the inhaled air of a subject giving a breath sample

is taken is important to consider in studies such as these.

4-6: Hardware Store Study

The decay in abundance of levels of inhaled volatile chemicals has also been seen post-
exposure in hardware store exposures to VOCs. Breath samples (EBC) were obtained
from two human subjects before they spent an hour in a local hardware store. After
returning to the sampling area, 3 additional breath samples were obtained from each
subject over the course of 2 hours. Samples were obtained from subject A at 18, 70, and
125 min after leaving the hardware store; samples were obtained ﬁom subject B at 4, 28,
and 37 min after leaving the hardware store. Figure 4- Sshows a selected area of the XIC
(m/z 105) obtained from each breath sample of the 2 subjects. This peak has been
tentatively identiﬁed as l-methoxyethyl benzoate, an ingredient in furniture stain that is
sold at the hardware store. The same compound was detected in a direct SPME sample
taken of the air inside the hardware store for 10 minutes (data not shown). Given that the
compound was present in the air source of the subjects, and the breath of both individuals
shows a rise and fall in the concentration of this compound, these data show that breath
analysis using GC-MS is an appropriate way to measure and monitor chemical exposure,

with detectable signal remaining for hours after exposure. A small signal was detected in

134

 

the breath samples obtained before exposure; however this is not unexpected given the

ubiquitous nature of furniture stain in an ofﬁce/laboratory setting.

   
   

 

 

 

 

 

SubjectA Subject B
79999 36000
60000 32000
’1‘ Time post ’5
3 50000 exposme 3 23000
‘8‘“ 40000 3’ 24000
0
g S 20000
§ 30000 g 16000
.0 .o
:2: 20000 E 12000
.2. 3 8000
10000 , 4000
O ‘ 0-.----ﬁs
7.96 7.98 8.00 8.02 7.96 7.98 8.00 8.02
Retention Tune (min) Retention Tune (min)

Figure 4- 5: (a. ) Selected window of )ﬂCs (m/z 105) obtained from Subject A before
exposure (black) in the hardware store and 8, 70, and 125 minutes after leaving the
hardware store. (b. ) Selected window of MCS (m/z 105) obtained from Subject B before
exposure (black) and 4, 28, and 3 7 minutes after leaving the hardware store. The source
of the peak was tentatively identiﬁed as an ingredient in ﬁtmiture stain, 1 -methoxyethyl
benzoate. This chemical was also detected in a SPA/ﬂ? sample of the ambient air in the
hardware store. (Image originally in color)

4- 7: Lemonade Study

A control breath sample was taken prior to lemonade consumption and then many
samples were collected after lemonade consumption. A breath sample taken before
lemonade consumption does show a detectable but low-level signal in the GC-MS TIC
and XIC (m/z 93) from monoterpenes and terpenoid compounds; however, this signal is
dwarfed by the contribution of these compounds in a post-lemonade sample. Many of the

monoterpenes and terpenoids are consumed in food products or inhaled from the ambient

135

air due to their abundance in the environment from natural and synthetic sources,
presumably causing this initial signal. In fact, estimates of the average per capita
combined exposure of a-limonene, terpinolene, and B-myrcene approach 1.3 mg/day.30
Figure 4- 6 shows an XIC obtained before consuming original lemonade compared to one

obtained 3 minutes after lemonade consumption, offset for clarity.

Before Lemonade

Consumption
201 3 min After 224000 Before Lemonade

Consumption

 

l8: Lemonade Consumption 33420000
a-thujene l 6000

when, 5 12000
camphene § 8000

B-Phellandrene 4000 1‘ LL J“ L A l; 1 _ 44* A

- - 0
159mm 5 4.60 5.00 5.40 5.80 6.20 6.60 7.00 7.4

B-myrcene - -
aterpinene I 0 Retention Time (mm)

p-cymene

a-Iimonene

. IOy-terpinene “h
11 Terpinolene

f 124-16113111601 2 34M
11
1

 

 

 

Ion Ab

 

 

 

ﬂ
\OmNOxM-wa-d

Ion Abundance x 105 (a.u.)
g 3

8
11 13,13

A.

 

l3 a-terp ineol

 

 

 

A l
I

 

 

ONAQOO

 

ﬁ 7 v v ' v u v i I r r v v I v u 1 Y r 1 r I

2.00 4.00 6.00 .8100. ' ' 10:00 ' 12.00 V 14:00 ' ' 16:00 .
Retention Time (min)

Figure 4- 6: Extracted Ion Chromatograms (M Cs) of the m/z 93 ion in breath samples
obtained before lemonade consumption (black trace) and 3 min after original lemonade
consumption (red trace) offset for clarity. Inset shows magnification of boxed area of the
sample taken before lemonade exposure. I 3 terpenes and terpenoid compounds of
interest are identified. (Image originally in color)

The 13 monoterpenes and terpenoid compounds of interest are labeled, and their
structures, molecular weights, and retention times (RT) are shown in Table 4- 1. The
signal post-consumption ranges ﬁom 1.6-57 times that seen in the sample before

consumption. The inset shows a magniﬁcation of the boxed area in the ﬁgure. Although

136

 

these compounds are present in the sample before consuming lemonade, this background
contribution to the compounds’ level measured after lemonade consumption is minimal.
This background contribution is corrected for in peak area calculations for all

measurements made after lemonade consumption.

 

137

Table 4- 1: Molecular weights, structures, and GC retention times (RT) of 13 terpenes
and terpenoid compounds studied. Average decay values, tau, for a single subject (n=5)
after consumption of original lemonade, modeling compound decay in breath to a single
phase decay are also shown.

 

 

 

 

 

 

 

 

 

 

 

RT Average Tau

Compound Structure (min) “:5 (min, SEM)
a—Thujene 4.75 85.9(13.9)
MW 136.23 tauabs=1.0
a—Pinene 4.84 129.1(12.57)
MW 136.23 tauabs=2.0
Camphene 5.03 213.1 (44.6)
MW 136.23 ‘ tauabs=l .5
B—Phellandrene < > < 5.23 129.9 (25.7)1
MW 136.23 tauabs=0.l
[S—Pinene 5.30 118.2 (17.6)
MW 136.23 tauabs=l .5
B—Myrcene 5.34 81 .3(10.2)
erms M 61,561
a—Terpinene 5.66 60.2(13.8)
MW 136.23 W tauabs=0.l
p—Cymene 5.73 186.0(25 .6)1
MW 134.22 —©—< tauabs=1.0
a—Limonene 5.78 86.6 (5.7)
MW 136.23 —<:>—< tauabs=0.2
y—Terpinene 6.04 62.6 (4.3)
MW 136.23 ‘Q—< tauabs=0.0

 

 

 

 

 

 

 

 

13

00

Table 4-1 (cont ’d).

 

RT Average Tau
(min) n=5 (min, SEM)

Terpinolene O_ 6.30 56.1(12.1)
MW 136.23 tauabs=0.0
4—Terpineol O < 7.17 12.3(2.9)
MW 154.25 OH tauabs=0.0
a—Terpineol 7.29 8.6 (01 .7)
MW 154.25 H OH tauabs=0.2

1n=4, In=3

Compound Structure

 

 

 

 

 

 

 

 

 

4-7-1: Reproducibility of Single Subject Drinking Original Lemonade

Figure 4- 7a-b show the XIC region containing the peak identiﬁed as a-limonene from 19
breath samples obtained from a single subject after original lemonade consumption. The

XIC trace from the pre-consumption sample is also shown.

139

 
   
  
   
 
   

 

 

 

 

 

18 illiiiiyjniﬁr-I "4“ :mn .1111‘1
t 11“ (n.5, b) 777mm after C.)
16 75 min after 3""111111 .1111-1 20
’7 99 min ullcr .131 mm .11'11-1' ’7
g l4 123 min after 3710- ‘5' 347 mm after 3 tau=8] .6rmn
V 14711111111111? 5; ‘ 3 111111131114 :15
E: 12 172mm alter 0 398mm 11111-1 S
a 197 min after 2 8‘ 425 min after ><
D 10 22111111141111 ’< 13681111113111-1 a
8 34‘111111 .1111‘1 8 before Q4‘10
g 3 272 min alter a 6'1 <1"
5 6 3971111110114 '2 . f:
.D 321n1|n.111cr 0
< 347mm after Q 4i : 5
8 4 372111111 :1I'1cr = - g
'_ 39811111131111 2 2_ g
2 ,425111111 ruler "-80
0 ' - _ 1368 mm .rllci' 4 <
before 04W ..-r......
R5'76 - 5'71? 5'80- 5.76 5.78 5.80 0 . 5 10' 15_ 20 25
etentlon 1me (mm) Retention Time (min) Time 30911118651100 (hf)

Figure 4- 7: (a) XICs (m/z 93) of peak identified as a-limonene in breath samples
obtained before and 3-1 3 68 minutes after original lemonade consumption. (b. ) Magnified
view of XICS shown in (a) obtained 247—1368 minutes after original lemonade
consumption, c) Area of peak due to a-limonene corrected for initial area before
lemonade consumption plotted versus time after ingestion. These data are ﬁt to a single
phase exponential uptake and decay model. (Image originally in color)

In the sample obtained 3 minutes after lemonade consumption, a 77-fold increase in the
peak area above background level is observed. An additional experiment in which the
subject gargled with lemonade without swallowing and completed the described rinsing
procedure demonstrated that a minimal amount of this increase is due to residuals in the
mouth and/or throat. Figure 4- 7c shows the a-limonene peak area from each time point.
a-Limonene, as well as a-pinene and B-pinene, has previously been shown to follow a
triphasic elimination, with a short half life immediately after exposure and longer half
lives as the elimination processes progress.” 3" 32 This slow elimination is thought to be
caused by the high lipophilicity of the compounds which causes the compounds to
d.26, 33

preferentially partition in fatty tissue, assuming they are not ﬁrst metabolize

Indeed, previous reports have shown that metabolic products of monoterpenes excreted

140

through the kidneys reach maximum levels approximately 5 days after exposure, even
after hemoperfusion, suggesting storage and slow metabolism in the fatty tissues.29
While many studies have ﬁt terpene data to a two- or three- component exponential decay
process,l 1’ 28 the data presented in this study do not reveal the ﬁrst phase of elimination in
the a-limonene data as the reported half-life of this phase is 3 min,34 and the initial breath
sample in this study was obtained from 3-13 minutes aﬁer exposure. The current data ﬁts

most accurately to a model including a single exponential rise (absorption) and a single

exponential decay (elimination) model:

(—x/tau1) _ A e(—x/tauabs)
1

yzyo+Ale

Equation 4- 1: Exponential rise and decay with a y—oﬂset

where yO is the offset of the ﬁt, A1 is the concentration remaining in the body, tau] is the
half-life of the decay, and tauabS is the half-life of the absorption process. tauabS was held

constant at 0.2 min for the ﬁt, yielding a half-life (taul) of 86.6 min for a-limonene.

Figure 4- 8a-b shows the XIC region for the peak identiﬁed as containing y-terpinene
from the same breath samples previously discussed. The pre-consumption sample
contained only a trace amount of y-terpinene, so the sample obtained 3 minutes after
consumption showed a 149-fold increase in peak area compared to this value. Figure 4-
8c shows the decay in peak area for this peak ﬁt to a single exponential rise and decay

(Equation 4- l). The half-life of absorption was held constant at 0 min, and the half-life

141

of elimination was determined to be 62.6 minutes. Similar curve peak area analysis and

curve ﬁtting was performed for all peaks (See Appendix B).

    
   

3 min after
37 ' .
3.2 :1 niiii iii” 32 b) 317111111 .1011 3 5+ c_)
'3 111111.111u1 372mmal‘1er ﬁx ' .
2‘8 Wininalter 2'8 3‘1'121111111” 30. —62 6 _
123 min alter r7 131111111 Jllcr ‘ tall“ . mm
2'4 147111111 din-r :1 2-4 347111111 afrer
172 min alter v \ 372111111 .11'1cr 2.5‘
2.0 197 min utter 1 398111111 .1111‘1
1 425 min after 2.0

321111111.111‘r
i 13(1-‘411111111llc1‘

Ion Abundance x 10" (a.u.)
; .
Adjusted Peak Area x 10“ (an
5

 

 

 

 

 

 

 

cc
3 2.
l6 247111111.1l1c1 $1.6 \,
272mm alter g 1
3‘17111111.1|1c1 e312
.121 mm 11111-1 '2 1.0-
347 min after 3
0'8 .172 min 111m '2 0-8 o 5.
19811111141111 :1 I
0-4 425 min utter 3 0.4 .
1368111111 .1llc1 0
0 before 0‘ ' ' r ' ' ' ' ' ' I
6-02 .694. 69" 6.02 6.04 6.06 0 5 10 15 20 25
Retention Trme (mm) Retention Time (min) Time aﬁer ingestion (hr)

Figure 4- 8: (a) GC—MS XI Cs (m/z 93) of peak identified as y-teipinene in breath samples
obtained before and 3—1368 minutes after original lemonade consumption, (b) Magnified
view of XICS shown in Figure 4a obtained 24 7-1368 minutes after original lemonade
consumption, (c) Area of peak due to y-teipinene versus time after ingestion. These data
are ﬁt to a single phase exponential uptake and decay model. (Image originally in color)

Calculated half-lives are shown in Table 4- 1 for 5 replicates of a single subject except
where noted. The standard error of the mean (SEM) is surprisingly low for many of the
monoterpenes and terpenoids monitored. This particular subject had a consistent exercise
and diet schedule during the study which was thought to affect the consistency of the
data. The half-lives of the monoterpene alcohols were determined to be signiﬁcantly
lower than those of the other monoterpenes and terpenoids using Welch’s t-test
(p<0.001). This may be due to the higher water solubility and lower saturated vapor

35-38

pressure of the alcohols, causing them to quickly partition into the water vapor that

dominates the components in human breath. The monocyclic monoterpenes’ half-lives

142

were also signiﬁcantly lower than those of the bicyclic monoterpenes (p< 0.001), and the
aromatic monoterpene, p-cymene (0.001<p<0.01). The linear monoterpene’s half-life
was also signiﬁcantly different from that of the aromatic monoterpene (0.01<p<0.02,
Student’s t-test), which may also be due to the different solubilities of the compounds.”
37’ 39 Table 4- 2 shows the tau values calculated for 3 subjects after drinking original
lemonade. Similar relative values for half-lives of the different compounds were

observed for data obtained from 3 subjects after consumption of original lemonade.

Table 4- 2: Half-lives (tau) for single exponential decay of 13 terpenes and terpenoid
compounds in breath after original lemonade consumption by 3 subjects.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tau (n=3 subjects)

Compound Average SEM
a - Thujene 62.5 26.4
a - Pinene 56.0 29.5
Camphene 76.3 34.7
B- Phellandrene 43.4 27.9
B- Pinene 47.7 31.2
13 - Myrcene 39.6 20.7
a - Terpinene 35.7 18.8
p - Cymene 105.7 10.4
a - Limonene 43.6 22.9
7 - Terpinene 32.5 15.6
Terpinolene 32.6 15.1
4 - Terpineol 12.5 2.3

a - Terpineol 15.6 3.6

 

 

 

 

 

143

4-7-2: Concentration Dependence of Terpene Signal in Breath

Other concentrations and types of lemonade were also used to analyze the uptake and
decay kinetics of the terpenes and terpenoid compounds. Concentrated lemonade was
prepared from the same lot of frozen lemonade concentrate as the original lemonade but
with less dilution. Mediterranean lemonade was made from fresh lemons in an effort to
further increase the terpene and terpenoid concentration consumed. Figure 4- 9 shows
the a—limonene XIC peak area (m/z 93) versus time after consumption of each type of
lemonade for a single subject. The highest XIC peak area reached in a sample increases
linearly with the a-limonene concentration measured in each lemonade sample: 0.696
mM, 1.93 mM and 5.96 mM for the original, concentrated, and Mediterranean varieties,

25. 31

respectively (R2=0.993), previously seen with monoterpenes. This trend is also

apparent in the other monoterpenes and terpenoid compounds.

144

14"

p—
I»)
l

 

  
 
  

—"'— Mediterranean Lemonade
—-o— (.‘oncentrated Lemonade

—-—Ori inal Lemonade

p-
O
L

A

00
A J

 

Adjusted Peak Area x10“(a.u.)
«l- O’\

 

 

 

 

2-1
0-1 "v ?$'3=‘—.—.
'I'rﬁrr1'r'rrr't'r'u'ﬁr
0 400 800 1200 1600 2000

Time After Consumption (min)

Figure 4- 9: a-Limonene peak area versus time after consumption for original,

concentrated, and Mediterranean lemonade. Note the higher tmax for higher lemonade
concentrations as well as the appearance of the uptake curve of the a-limonene. (Image
originally in color)

The maximum peak area of each terpene and terpenoid compound measured in breath
after consumption of each lemonade sample was normalized to the maximum peak area
measured in breath after consumption of the original lemonade for comparison.
Similarly, the peak area of each terpene and terpenoid compound in the actual (neat)
lemonade samples was normalized to its area in the original lemonade. Figure 4- 10
shows these relative peak areas for the 13 compounds monitored in the actual (neat)
lemonade samples (Figure 4- 10a) as well as the maximum peak area measured in breath

after consumption of each type of lemonade (Figure 4- 10b). All terpene analytes show a

145

signiﬁcant increase in peak area in breath as the concentration of the lemonade consumed
increases, similar to the peak area increase of the terpenes measured in the neat lemonade
samples of increasing concentrations. The only exception is the terpene alcohols, 4-
terpineol and a-terpineol, where no signiﬁcant increase in signal in breath samples with
increasing concentration of lemonade consumed is seen; this is possibly due to the quick

elimination of terpene alcohols from the body.

  
  
  
    

1100j a) - Original Lemonade 620 ' b) - Original Lemonade in Breath
000 j; - Concentrated Lemonade - Concentrated Lemonade in Breath
30: -Mediterranean lemonade ‘ - Mediterranean Lemonade in Breath
20 ‘ 610 z I
50 ‘

 

 

Relative Peak Area (Corrected).

10« 301
20j
10:
00° 0 \ 0&0 \
06346" 9° 9° 9° oigoo
3 0&1? 00°090°¢3° 1°¢I9°9 cocooQQ
31$: wefeﬁ «$1035 efﬁgy V” «s

32:?

Figure 4- 10. Relative peak areas of l 3 terpene and terpenoid compounds in (a) neat
samples of original, concentrated, and Mediterranean lemonade, and (b) breath samples
after consumption of original, concentrated, and Mediterranean lemonade. Peak areas
are normalized to the original lemonade peak area for each compound, and the breath

sample obtained at tmax was used. (Image originally 1n color)

More information about the uptake of the monoterpenes and terpenoids is gleaned when
higher concentrations of lemonade are consumed. Figure 4- 9 also demonstrates that the
maximum peak area, thus, concentration in breath, is reached at later times when the

terpene concentrations in lemonade are greater. This ﬁnding also supports the conclusion

146

that residual compounds in the mouth are not causing the signal seen in the breath

samples as the signal continues to rise long after consumption. In the original lemonade,

the initial breath sample taken 3 minutes aﬁer consumption (tmax) yielded the highest (1-

limonene concentration. In the concentrated lemonade this t max was not reached. until 52

minutes after consumption, and in the Mediterranean lemonade, 73 minutes after
consumption. After exposure, a chemical is first distributed through the body, and then
eliminated. These processes are not exclusive of each other; elimination begins while the
chemical is still being distributed. In the experiment presented, chemical exposure
proceeds through ingestion, so the kinetics of an additional absorption step must be
considered. The terpenes and terpenoids are ﬁrst absorbed through the digestive tract,
passed into the blood and distributed to various tissues based on their lipid and water

0 Absorption and distribution also occur concurrently. The terpenes and

solubilities.4
terpenoids pass from the blood into the lung via the alveolar puhnonary membrane and
attempt to establish equilibrium between the two compartments. Exhalation pushes the
compounds out of the lung, which are subsequently replaced through the constant shift
towards equilibrium. The higher the concentration of lemonade consumed, the longer it
takes for the absorption and distribution processes to occur, yielding a rise followed by

the decay of the peak area due to monoterpenes and terpenoid compounds as previously

described.28 With higher concentration of monoterpenes and terpenoids, absorption and

distribution take a longer time, yielding the increased tmax, and describing that the

absorption or distribution processes have been saturated, and that ﬁirther terpenes cannot

be absorbed until elimination has removed some from circulation. The extended time

147

amdki

mentor-ﬁat

likmlr s1
imam:
1asuu

\k i tar-r“: r

U ﬁnial

1113113181:

Figure .

v

BeCﬂUSe
161mm
lhedm.

InphaSlQ

period before elimination becomes the primary process in the uptake and decay of the
monoterpenes and terpenoid compounds after consumption of Mediterranean lemonade

allows the compound uptake to become more apparent in the data, as several data points

were obtained before tmax was reached. The data from the consumption of the

Mediterranean lemonade was ﬁt to Equation 4- l, and is shown in Figure 4- 11. Fit

parameters are shown in Table 4- 3.

  
  

 

 

 

.5100 4

9

08/ 80 —

2

pg: 60 "‘

(U

8

< 40 —

£5

a: 20 —

on O . .

l ' l ' I I I l V
0 400 800 1200 1600

Time After Consumption (min)

Figure 4- I]: Uptake and decay of a-limonene after consumption of Mediterranean
lemonade ﬁt to Equation 4- I.Equation 4- 2 Fit parameters are shown in Table 4— 3.

Because of the deviation seen in this ﬁt, the data from consumption of the Mediterranean
lemonade was ﬁt to a higher order model using the method of residuals (Figure 4- 12).27

The data was ﬁt to the following equation which describes the ﬁrst order uptake, and

triphasic decay of a-limonene:

148

there 1 15 1h

Q

{US$115 01
iii: llﬁl 0de

3115 $1011 611
Eamon 4.

11131111656 d,

lgure 4.

Table 4‘ 3”;

: Ale(—x/taul) + A (—x/tau2) + A3e(—x/tau3) _ U e(—x/tauabs)

26

y

Equation 4— 2: Model of uptake and triphasic decay for monoterpenes and terpenoids

TIE-I.

where y is the peak area of the compound detected in breath, A1, A2, A3, and U are the

constants of the three phases of decay and uptake, respectively, tauabS is the half-life of

 

 

the ﬁrst order absorption process, and taul, tauz, and tau3 are the ﬁrst order fast, medium,

and slow elimination half-lives. These ﬁt values are also shown in Table 4- 3. Neither
Equation 4- 1 nor Equation 4- 2 describe the data in an acceptable manner. It is thought

that these data require a much more complicated ﬁt, which may be pursued in the future.

  

3 120-

°o

:3 80+

8

:5 40—

.54

83

On 0_ .

 

 

0 400 800 1200 1600
Time After Consumption (min)

Figure 4- 12: Uptake and decay of a-limonene after consumption of Mediterranean
lemonade ﬁt to Equation 4- 2 using the method of residuals. Fit parameters are shown in

Table 4— 3

149

 

Table 4- 3: Coefﬁcients and half—lives obtained after ﬁtting the uptake and decay of a-
limonene in breath after consumption of Mediterranean lemonade to Equation 4- I and
Equation 4— 2.

 

Fit to Equation 4-1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A tauabs tau]
3.70 x 108 57.0 120
Fit to Equation 4-2
Uptake Decay
U tauabS A, tau]
3.54 x108 55.2 1.10 x107 968
A2 tauz
1.93 x 108 171
A3 tau3
[ 1.28 x 108 365

 

 

 

 

A deviation from exponential decay is seen for many of the monoterpenes and terpenoids
across all subjects. Approximately 5 hours after lemonade consumption, the peak area of
the terpenes and terpenoids in breath plateaus or increases slightly, rather than continuing
to decrease exponentially (see Figure 4- 9 and Appendix B). After approximately 1 hour,
peak areas measured in breath samples return to the exponential decay. This is thought to
be a real effect, as all samples were analyzed using the same GC-MS method and the
deviation is consistently observed for all terpenes and across subjects. A similar plateau
phenomenon has been previously reported for several VOCS measured in breath,
speciﬁcally 1,1,1-trichloroethane,11 and was thought to be due to changes in activity,

blood ﬂow, or diet. Several subjects in this study fasted while participating, eliminating

150

 

 

immediate effects of diet variation as a cause of this effect, however changes in activity,
blood ﬂow, or changes in metabolism due to fasting are viable possibilities and will be

considered further in the ﬁiture.

4-8: Drug Studies

4-8-1 : Sudafed

Studies were performed to determine if oral ingestion of an OTC drug would be
detectable on breath. If so, this method may have applications to drug testing, or even
diagnostics in an emergency room setting where a patient may have overdosed on an
unknown drug. Two subjects participated in this study. A breath sample was collected in
the morning, followed by the subject consuming two 30 mg pseudoephedrine tablets.
Breath samples were then collected at several times after ingestion. Breath samples were
analyzed using Selected Ion Monitoring (SIM) mode of the GC-MS, monitoring m/z 58,
the ion corresponding to the base peak in the mass spectrum of pseudoephedrine. For
retention time and mass spectral conﬁrmation, a pseudoephedrine tablet was crushed and
26.6 mg of the powder was dissolved in 5 mL of water. A SPME ﬁber was directly
exposed to the water for 1 min, and analyzed using the method previously described and
well as in scan mode. The mass spectrum obtained was consistent with that in the NIST

mass spectral database, and the retention time was determined to be 8.61 min.

Results were variable across subjects and even between different exposures of a single
subject. Figure 4- 13 shows the ‘best case’ data of exposure of a single subject. Figure

4- 13a shows the chromatographic peak attributed to pseudoephedrine in breath samples

151

 

 

obtained before and at eight intervals after ingestion. Figure 4- 13b shows the same data,
without the 10 min after consumption sample, for clarity. A signiﬁcant increase in peak
area is seen in the breath sample obtained 10 minutes after pseudoephedrine consumption
relative to that seen in the breath sample obtained before consumption. Over several
hours, the area of this peak decays to values near that seen in the sample obtained before
consumption. The decay can be ﬁt to an exponential model (R2=0.9989) with a half-life
of 13.4 minutes. This is a preliminary experiment, and the reproducibility of the rise and
decay of the pseudoephedrine peak aﬁer consumption has not been determined. In
repeated experiments, the pseudoephedrine peak area in a breath sample obtained
approximately 10 minutes alter consumption was lower than that shown below, and is not
seen in samples obtained approximately an hour after consumption, thus preventing study
of the decay of the compound in these additional experiments. This could be due to
variation in diet of the subject as it was not regulated, illness of the subject as medication
was only taken when symptoms were present, or other variation in metabolism and

digestion.

‘-.'

 

— before

 
 
 
 
 
 
 

 

 

 

 

 

,56000- a) ~—10minafter ,51200 b)

3 ‘ ' — 33 min after 3 '

Br 50001 k 55 min after v 1000

o ‘ — 80 min after 8

g 4000' —- 105 min after 5 800

g ' — 134 min after g

.3 3000: — 175 min after '8 600

E 2000‘ ‘- 207 min after 8

oo ‘ o: 400

'2 1000‘ g 2 ‘2 200

a ‘ ...... 7‘ ..... 2. E

0 8.59 8.61 8.63 8.65 8.59 8.61 8.63 8.65
Retention Time (min) Retention Time (min)

Figure 4- 13: Selected window of the m/z 58 chromatogram obtained in SIM mode before
and after ingestion of 60 mg of pseudoephedrine. The peak has been identified as
pseudoephedrine. (a) shows all breath samples obtained, (b) shows a zoomed in view of
the samples excluding that obtained I 0 min after consumption for clarity. (Image
originally in color)

The signal seen in the chromatograms obtained ﬁom breath after consumption of
pseudoephedrine is 4 orders of magnitude less than that previously observed for terpenes
after lemonade consumption. The amount consumed is actually higher; 60 mg of
pseudoephedrine were consumed in this study while ~45 mg of a-limonene were
consumed in the original lemonade sample. The vapor pressure of limonene at 25 °C is
20 torr,“ much higher than that of pseudoephedrine, which is listed as a non-volatile
solid.42 Indeed, pseudoephedrine is typically derivatized prior to GC-MS analysis in
order to increase its volatility. In order for the pseudoephedrine to be detected in breath,
it must be absorbed and distributed into the blood which then passes by the alveolar
pulmonary membrane, where the pseudoephedrine must partition into the alveolar air and
be exhaled. With a low vapor pressure, partitioning into the alveolar air would be slight;
whereas the higher vapor pressure of limonene would allow signiﬁcant partitioning

across the membrane. A recent study has used breath analysis to monitor levels of

153

 

 

 

propofol, an intravenous anesthetic, during surgery by sampling breath directly from the
endotracheal tube in the patient.43 In this study, exhaled breath was analyzed and the
propofol signal was correlated to the concentration measured in blood. Although the
vapor pressure of propofol is still relatively low (0.142 torr at 20 °C), it was reported that
its low water solubility results in a measurable quantity of the drug in the gas phase near

a hydrophilic solution.43

Investigation into the detection of ibuprofen ingestion was also pursued, but no signal
was detected in breath after ingestion. The vapor pressure of ibuprofen is reported as < 1
torr.44 Ibuprofen undergoes minimal ﬁrst pass metabolism in the liver; however, it
experiences very high binding to plasma proteins.45 It is reported that only 0.3-0.5% of
ibuprofen remains unbound when a low concentration, as used in this study, is
consumed.45 A combination of little free ibuprofen being available to partition across the
alveolar pulmonary membrane, as well as the beneﬁt of derivatizing the molecule for
GC-MS analysis which was not done in these studies may be responsible for the lack of

signal in this study.

4-9: Explosive Studies

4-9-1: Hexarnine

Studies were performed to determine if explosive-related compounds could be detected in
human breath. Comparisons were made between samples collected using EBC and EBV.
A single breath sample was obtained from a subject in the morning. The subject then
poured approximately 10 g of hexamine, a common precursor in the synthesis of RDX,

between 2 100 mL plastic beakers on the laboratory benchtop for 10 min. Subsequent

154

 

 

breath samples were then collected. Samples were analyzed via GC-MS which was
operated in selected ion monitoring (SIM) mode to speciﬁcally detect the m/z 140 and 42
ions, the molecular ion and base peak in the hexamine mass spectrum, respectively, with
greater sensitivity. For retention time and mass spectrometric conﬁrmation, a sample of
vapors ﬁom the hexamine powder was obtained by suspending a SPME ﬁber in the
headspace of the jar for 1 minute and analyzing for hexamine using the method
previously described and well as with the GCMS in scan mode. The mass spectrum
obtained was consistent with that in the NIST mass spectral database, and the retention

time was determined to be 7.64 min.

Figure 4- 14a shows the peak due to hexamine in the m/z 140 XIC for EBC samples
obtained before and 26 minutes aﬁer exposure to hexamine. Figure 4- 14b shows the
peak due to hexamine in the m/z 140 XIC for EBV samples obtained before and 30
minutes after exposure to hexamine from the sample subject. Note the difference in the
y-axis for the samples. Although they both show an increase in signal after exposure to
hexamine, the signal ﬁom the EBV is much higher than that ﬁom the BBC. The
difference in signal may be partially explained by whether the hexamine is inhaled in
vapor or aerosol form. The vapor pressure of hexamine is reported to be 0.0035 mbar,46
which makes it unlikely that it would be present in the vapor phase. Aerosol particles of
hexamine may have been inhaled and then exhaled and collected by the SPME ﬁber.
Even a single particle collected in this manner would provide signiﬁcant signal. Another
possible explanation is differences in actual exposure, as the only parameter that was
controlled was the length of exposure; the actual exposure, breathing rate and depth were

not easily controlled.

155

 

 

 

160 a.) b)
$140 £2500
%120 E 2000

g 100 g 1500

g 80 g 1000

E 60 2 500

o 3

 

 

 

40 0
7.60 7.65 _ 7.70 . 7.75 7.60 7.65 7.70 7.75
Retentron Time (mln) Retention Time (min)

Figure 4- 14: (a) Selected window of GC—MS XIC (m/z 140) for breath samples (EBC)
obtained before (black) and 26 min after (red) exposure to hexamine, (b) Selected
window of 20C (m/z 140) for breath samples (EBV) obtained before (black) and 30 min
after (red) exposure to hexamine. (Image originally in color)

A study was then performed to investigate the rate at which hexamine was cleared from
the body via the breath. A single EBV sample was again obtained in the morning, and
the subject poured the hexamine back and forth between plastic beakers. Breath samples
were then obtained 1, 30, 61, and 82 minutes after exposure. Figure 4- 15a shows the
peak due to hexamine in these samples obtained by monitoring m/z 140 and m/z 42.
Peak areas were calculated from the XIC of m/z 140. Figure 4- 15b shows the change in
peak area versus time after exposure for the samples. The four samples taken after
exposure were ﬁt to an exponential decay (R2=0.994), and the half-life was determined to
be 15.6 minutes. This provides initial evidence that an RDX precursor can be detected on

breath, and may remain on breath for a period of time after exposure.

156

 

  
  
  

 

 

12000‘ a.) b.)
2 ,3 300
310000 ~ 1min after 3 250 .
g 8000‘ 30min after ‘; tau=15.6mm
E 6000- — 61min after g 200 R2=0.994
— 82min after is
150
<2 4000 before a:
2 2000- (Q) 100 . g
-2.
0 50 . o
7.50 ' 7154 ' 7.'68 ' 7.'72 ' 0' b 70 110 60 {,0

 

Retention Time (min) Time After Exposure (min)

Figure 4— 15: (a) Selected window of TIC (SIM, m/z 140, 42) from breath samples (EB V)
of a single subject before and after exposure to hexamine, (b) Exponential decay, R =
0. 994, of the area of the peak due to hexamine, calculated from the HC (m/z 140)
(chromatogram not shown). (Image originally in color)

As 10 minutes of exposure to 10 g of hexamine provides a relatively low exposure level,
a study was performed to determine if a longer exposure would provide a higher signal
from the breath samples. A single EBV breath sample was obtained from a subject in the
morning. The subject then poured ~10 g of hexamine between the plastic beakers for 10
min. An additional breath sample was collected, and the subject then repeated the 10 min
hexamine pour on the benchtop. This procedure was repeated for over 5 hours, resulting
in 8, 10 min exposures and the collection of 12 breath samples. All breath samples were
analyzed using selected ion monitoring of ions at m/z 140 and 42, the molecular ion and
base peak fragment ion in a mass spectrum of pure hexamine, respectively. Figure 4- 16
shows the increase and decay of the area of the hexamine peak in the XIC m/z 140
chromatograms over time. The red bars represent the time the subject was exposed to the
hexamine. As the subject’s exposure increased, so did the concentration of hexamine on

the breath. When exposure ceased, the concentration of hexamine began to decay. The

157

 

 

 

subject in this study was handling only 10 g hexamine in a well-ventilated environment,
while wearing goggles, nitrile gloves, and a laboratory coat for a combined time of 80
minutes; it is hypothesized that clandestine synthesis and/or handling of explosives such
as RDX would involve longer exposures to higher concentrations of the precursors in less

than ideal environments, allowing for detection on a longer timescale.

_ Hexarnine Exposure

 

 

 

 

 

0 200 400 600 800 1000 1200 1400
Time from Initial Exposure (min)

Figure 4- I 6: Integrated area of GC—MS EC m/z 140 peak attributed to hexamine at
RT = 7.69 min in chromatograms obtained in SIM mode (m/z I 40 and 42) of breath
samples from a single subject before, during, and after several I 0 min exposures (shown
in red) to hexamine. Note the increase in hexamine on the breath during exposure and
the gradual decay after exposure. Hexarnine was still detected on breath 24 hrs after the
initial exposure. (Image originally in color)

4-9-2: Field Studies

In order to demonstrate the real-world ability of this technique to detect exposure to
industrial chemicals, ﬁeld studies were performed with six volunteers who participated in
exercises designed to accurately simulate the clandestine building of devices. These

subjects participated in daily activities that included packing, preparing, and transporting

158

 

the materials. Breath samples (EBV) were collected from the subjects each morning
before these activities began, and each afternoon after the simulation was completed.
Although details of these studies are sensitive and cannot be shared, it was conﬁrmed that
exposure to these industrial materials can indeed be measured based on signatures

measured in breath samples.

4-10: Conclusions

The detection of chemicals for hours aﬁer exposure can serve as proof of concept for a
detection protocol for chemical exposure. In the data presented here, the presence of key
chemicals, such as camphor, on the breath of an individual can suggest the prior
environment of the subject, such as the nail salon. These data are also important in
measuring chemical exposures, intentional or accidental, as in the case of a chemical
spill. The environment and exposure that occurs in a hardware store are also described,
which also provides data on the chemicals that could relate an individual to that location.
Even though these locations may not be of forensic interest, this work demonstrates that
the prior location of an individual may be determined based on residual signatures of
VOCs indicative of that location in human breath. Exposure via ingestion is also
described in the uptake and elimination of terpenes and terpenoid compounds after oral
exposure to a mixed terpenoid beverage (lemonade). The elimination of each compound
is modeled and ﬁt to an exponential decay in order for comparisons to be made between
the different compound classes. The uptake and triphasic decay of (Jr-limonene on breath
was modeled after consumption of a high concentration lemonade sample. The effect of

concentration on the elimination process is also described. These data support the

159

'I

 

hypothesis that longer exposures or higher concentration exposures will increase the
signal of a chemical measured on breath, therefore providing a larger time scale of

detection.

The detection of drugs and chemicals used in the synthesis of explosives is also shown,
allowing for the monitoring of illicit activity. Field studies presented an opportunity to
test the breath analysis technique on real-world samples from industrial chemical

workers, proving the success of the technique and methods that were developed.

160

 

4 - l I : References

(1)

(2)

(3)

(4)
(5)

(6)

(7)

(8)

(9)

(10)
(11)

(12)

(13)

(14)

(15)

(16)

(17)

Pleil, J. D.; Fisher, J. W.; Lindstrom, A. B. Environ. Health Perspect. 1998, 106,
573-580.

Goldoni, M.; Catalani, S.; De Palma, G.; Manini, P.; Acampa, 0.; Corradi, M.;
Bergonzi, R.; Apostoli, P.; Mutti, A. Environ. Health Perspect. 2004, 112, 1293-
1298.

Maniscalco, M.; De Laurentiis, G.; Pentella, C.; Mormile, M.; Sanduzzi, A.;
Carratu, P.; Soﬁa, M. Biomarkers 2006, 11, 233-240.

Pleil, J. D.; Lindstrom, A. B. Clin. Chem. 1997, 43, 723-730.

Poet, T. S.; Thrall, K. D.; Corley, R. A.; Hui, X.; Edwards, J. A.; Weitz, K. K.;
Maibach, H. 1.; Wester, R. C. T oxicol. Sci. 2000, 54, 42-51.

Corley, R. A.; Gordon, S. M.; Wallace, L. A. T oxicol. Sci. 2000, 53, 13-23.

Gargas, M. L.; Tyler, T. R.; Sweeney, L. M.; Corley, R. A.; Weitz, K. K.; Mast,
T. J.; Paustenbach, D. J.; Hays, S. M. T oxicol. Appl. Pharmacol. 2000, 165, 63-
73.

Corley, R. A.; Bormett, G. A.; Ghanayem, B. I. T oxicol. Appl. Pharmacol. 1994,
129, 61-79.

Corley, R. A.; English, J. C.; Hill, T. S.; Fiorica, L. A.; Morgott, D. A. T oxicol.
Appl. Pharmacol. 2000, 165, 163-174.

Wallace, L. A. Annual Review of Energy and the Environment 2001, 26, 269-301.

Raymer, J. H.; Pellizzari, E. D.; Thomas, K. W.; Cooper, S. D. J. Expo. Anal.
Environ. Epidemiol. 1991, 1, 439-451.

Wallace, L.; Buckley, T.; Pellizzari, E.; Gordon, S. Environ. Health Perspect.
1996, 104, 861-869.

Wallace, L. A.; Nelson, W. C. Measurements of Exhaled Breath Using a New
Portable Sampling Method, Environmental Protection Agency, EPA/600/S3-
90/049, 1990.

Wallace, L.; Pellizzari, E.; Gordon, S. J. Expo. Anal. Environ. Epidemiol. 1993, 3,
75-102.

Pellizzari, E. D.; Wallace, L. A.; Gordon, S. M. J. Expo. Anal. Environ.
Epidemiol. 1992, 2, 341-355.

Wallace, L. A.; Nelson, W. C.; Pellizzari, E. D.; Raymer, J. H. J. Expo. Anal.
Environ. Epidemiol. 1997, 7, 141-163.

Lindstrom, A. B.; Pleil, J. D. Biomarkers 2002, 7, 189-208.

161

 

 

(18)

(19)
(20)

(21)

(22)

(23)

(24)

(25)

(26)
(27)

(28)
(29)

(30)

(31)

(32)

(33)

Lindstrom, A. B.; Pleil, J. D.; Berkoff, D. C. Environ. Health Perspect. 1997, 105,
636-642.

Jo, W. K.; Weisel, C. P. L., P.J. Risk Anal. 1990, 10, 575-580.

Spencer, A. B.; Estill, C. F.; McCammon, J. B.; Mickelsen, R. L.; Johnston, 0. E.
Control of Ethyl Methacrylate Exposures During the Application of Artiﬁcial
Fingernails at Moore’s University of Hair Design, Cincinnati, Ohio U. S.
Department of Health and Human Services. Public Health Services, Centers for
Disease Control and Prevention, National Institute for Occupational Safety and
Health, Engineering Control Technology Branch. Report No.: ECTB 171-22a.

Gosselin, R. E.; Smith, R. P.; Hodge, H. C. Clinical Toxicology of Commercial
Products; Williams and Wilkins Press: Baltimore, MD, 1984.

Spencer, A. B.; Estill, C. F.; McCammon, J. B.; Mickelsen, R. L.; Johnston, 0. E.
Am. Ind. Hyg. Assoc. J. 1997, 58, 214-218.

Maxﬁeld, R.; Howe, H. L. Silica Exposure in Artiﬁcial Nail Application Salons.
Epidemiologic Report Series 97:8. Springﬁeld, 11: Illinois Department of Public
Health: November 1997.

Protecting the Health of Nail Salon Workers: EPA no. 744-F-07-001. March
2007.

Levin, J.; Eriksson, K.; Falk, A.; Lof, A. Int. Arch. Occup. Environ. Health 1992,
63, 571-573.

Falk-Filipsson, A. Occup. Environ. Med. 1996, 53, 100-105.

Gibaldi, M.; Perrier, 1D. Pharmacokinetics, 2nd Edition ed.; Marcel Dekker, lnc.:
New York, 1982.

Chen, H.; Chan, K. K.; Budd, T. J. Pharm. Biomed. Anal. 1998, 17, 631-640.

Koppel, C.; Tenczer, J.; Tonnesmann, U.; Schirop, T.; Ibe, K. Arch. Toxicol.
1981, 49, 73-78.

Test Plan for Monoterpene Hydrocarbons; The Flavor and Fragrance High
Production Volume Consortia: Washington, DC, 2002.

Falk, A.; Hagberg, M. T.; Lof, A. E.; Wigaeus-Hjelm, E. M.; Zhiping, W.
Scandinavian Journal of Work, Environment &Health 1990, 16, 372-378.

Safety evaluation of certain food additives In WHO Food Additives Series, 54,
Geneva, 2006.

Falk, A.; Gullstrand, E.; Lof, A.; Wiggaeus-Hjelm, E. Br. J. Ind. Med. 1990, 47,
62-64.

162

 

 

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

(43)

(44)

(45)

(46)

Falk-Filipsson, A.; Lof, A.; Hagberg, M.; Hjelm, E. W.; Wang, Z. J. Toxicol.
Environ. Health 1993, 38, 77-88.

Copolovici, L. O.; Filella, 1.; Llusia, J.; Niinemets, U.; Penuelas, J. Plant Physiol.
2005, 139, 485-496.

Li, J.; Perdue, E. M.; Pavlostathis, S. G.; Araujo, R. Environment International
1998, 24, 353-358.

"Aqueous Solubility and Henry's Law Constants of Organic Compounds" In CRC
Handbook of Chemistry and Physics, 88th Edition (Internet Version 2008); Lide,
D. R., Ed.; CRC Press/1’ aylor & Francis Boca Raton, FL.

Niinemets, U.; Reichstein, M. Global Biogeochemical Cycles 2002, 16, 1-26.

Howard, P. H. Handbook of Environmental Fate and Exposure Data for Organic
Chemicals, Vol. V; CRC Press: Boca Raton, FL., 1997.

Refat, M.; Moore, T. J.; Kazui, M.; Risby, T. H.; Perman, J. A.; Schwarz, K. B.
Pediatr. Res. 1991, 30, 396-403.

MSDS d-limonene (Cas=5989275), California Environmental Protection Agency,
2009.

Salocks, C.; Kaley, K. B. Technical Support Document: Toxicology Clandestine
Drug Labs/Methamphetamine, California EPA, 2003.

Homuss, C.; Praun, S.; Villinger, J.; Domauer, A.; Moehnle, P.; Dolch, M.;
Weninger, E.; Chouker, A.; Feil, C.; Briegel, J.; Thiel, M.; Schelling, G.
Anesthesiology 2007, 106, 665-674.

Ibuprofen, Mallinckrodt Inc., 2002.

"Metabolism and Pharmacokinetics of Ibuprofen" In Aspirin and Related Drugs;
Rainsford, K D., Ed.; CRC Press, 2004.

"Hexamethylenetetramine" In Synthetic Nitrogen Products; Springer US, 2005,
pp 333-336.

163

 

CHAPTER 5: Conclusions and Thoughts for the Future

5-1: Conclusions

5-1-1: Single Particle Aerosol Mass Spectrometry

This dissertation has presented an evaluation of the performance of Single Particle
Aerosol Mass Spectrometry (SPAMS) for the detection of active ingredients in over—the-
counter (OTC) drug tablets. In order to analyze these solid tablets, a sampling vial was
developed that could create particles from the tablets because of collisions with the vial
walls or other tablets. This novel sampling system was coupled easily to SPAMS, and
beneﬁted from the slight vacuum at the inlet of SPAMS pulling the aerosolized particles

into the instrument.

The forensic implications of the current work are vast. The active ingredients in OTC
tablets were identiﬁed without destroying the identifying marks on the exterior of the
tablet, and even in the presence of ﬁller materials in the tablets. The alarm algorithm
developed for use with SPAMS relies on rules trees alone, and was successﬁil in
identifying the active ingredients in single ingredient and multi-ingredient tablets, as well
as a sample containing multiple tablets of different drugs. The ability for the sampling
mechanism to be used for in situ sampling out of an original drug bottle is another
forensic attribute. This suggests the case at which a new sample could be presumptively
screened for the presence of illicit material, without the need for separating each tablet,
extracting the active ingredient, and testing each one individually. The active ingredient
could even be detected from the residual particles left on the walls of a container after it
had been emptied. This means that even if a drug sample was discarded in order to

prevent detection or arrest, identiﬁcation may still be made.

164

 

 

5-1-2: Breath Analysis

Breath analysis has been developed as a method of detecting chemical exposure based on
the identiﬁcation of signature compounds in the breath. Signiﬁcant attention was paid to
developing a method for sampling the breath, testing both exhaled breath condensate
(EBC) and exhaled breath vapor (EBV) obtained at two different temperatures, using
GC-MS for analyte detection. EBV was found to be superior for the analysis of most
compounds, and thus used for many of the studies. The importance of the localized
breathing environment was demonstrated with studies involving chemicals used in a nail
salon. The rebreathing of chemicals outgassing from the ﬁngernails was avoided by
incorporating a respirator cartridge or plastic tubing to remove contaminants from the
inhaled air, or provide it from a different environment, respectively. Both methods were

effective at reducing the contamination by the localized environment.

Extensive studies were also completed to analyze the breath of a human subject after
ingestion or inhalation exposure to chemicals, in particular terpenes and terpenoid
compounds. An immediate rise in signal, followed by decay back to near baseline over
several hours was measured for 13 compounds. This decay was successfully ﬁt to a
model describing a single exponential uptake and decay. The effects of different doses of
compounds on their levels in exhaled breath were then tested by varying the
concentration of the lemonade consumed in the study. An increased signal was measured
for constituents in breath which corresponded to the increase in concentration of each
compound in the lemonade sample. Several subjects were sampled, and several
replicates were obtained from a single subject, providing relatively similar signals over

time. When more concentrated forms of lemonade were consumed, the kinetics of the

165

 

..
- on

 

uptake and distribution process of the terpenes and terpenoids could be assessed based on
levels exhaled in the breath. Fitting these data to an established model presented
substantial challenges, but these measurements provide information that can be used to

assess the exposure of individuals to volatile chemicals.

Results from this work indicate that the prior location or prior activity of a subject may be
indicated by the presence of chemicals signatures of a particular environment or activity
on the breath. The chemical signatures of volatiles present in a nail salon and hardware
store were described and detected in breath sampled for several hours after exposure.
Chemicals indicative of exposure to explosive-related compounds were also detected on
breath for several hours after even a short exposure. These results, although preliminary,
indicate that breath analysis will be a powerful technique for identifying subjects of

interest when investigating possible criminal and terrorist activities.

5-2: Thoughts for the Future

5-2-1: Single Particle Aerosol Mass Spectrometry

Single Particle Aerosol Mass Spectrometry has already been established as a reliable
detection system capable of identifying biological agents, chemical agents, radiological
material, and explosives. lts ability to operate autonomously and sound alarms in the
presence of a user-defmed number of identiﬁed particles in time has also been deﬁned.
The current work has expanded these applications to include the detection of OTC drugs.
The ﬁiture of SPAMS shows promise in expanding this drug application also to illicit

drug testing.

166

 

The detection of illicit drugs with SPAMS is a natural extension of the current work.
Forensic laboratories would beneﬁt from this ability. In the future, illicit drug samples
could be obtained from the Drug Enforcement Agency and analyzed in small quantities
using SPAMS. Alarm ﬁles could be created from these data and unknown drug samples
could be analyzed and characterized using alarm ﬁles. However, cutting agents, ﬁller
materials, synthetic impurities and byproducts may be present in the ‘real world’ drug
samples. These additives are problematic in that they may interfere with the alarm
identiﬁcation, but are useful in that they may provide information on the synthetic origin
of a sample, [1561111 for forensic attribution.l Therefore, standard cutting agents,
impurities, and byproducts should also be tested, alarm ﬁles created for these compounds,
and the illicit drug alarm ﬁles adjusted to allow identiﬁcation in these complex matrices.
Some common cutting materials in illicit drugs include pseudoephedrine, caffeine,
procaine, baking soda, evaporated milk, and salt, some of which have been previously

studied with SPAMS.

Once alarm ﬁles are operable, SPAMS could become a powerful tool for drug detection
in a forensic laboratory. The National Forensic Laboratory Information System
(NFLIS), reports that US. state and local laboratories analyzed ~l .9 million drug items in
2006.3 The Census of Publicly Funded Forensic Crime Laboratories reported in 2002 that
the overwhelming number of requests caused a backlog of 233,000 samples. The fast
analysis time, as well as the ability to perform in situ analyses of SPAMS would be

beneﬁcial in reducing this massive caseload.

Further development of SPAMS may beneﬁt from modiﬁcation of the solid tablet

sampling apparatus. The current device is made from a glass sampling vial; however, the

167

 

 

transport efﬁciency of aerosol particles can be inﬂuenced by the charge of the particles
and interactions of particles with the container. Thus, a non-glass sample introduction

device is expected to yield a higher efﬁciency of particles for SPAMS characterization.

5-2-2: Breath Analysis

This work was initially approached with a single overarching goal: develop a detection
scheme that can be used to identify an individual that has been using illicit materials (e. g.
explosives) based on characteristic compounds in their breath. This goal was set based
on the current political climate and needs of the United States military and government to
protect its citizens, soldiers, and innocent civilians, from terrorist attacks, speciﬁcally
those involving improvised explosive devices (lEDs). Unfortunately, as fast as science
and technology evolve on the defense and detection side, so too do they evolve on the
side of those who wish to use science and technology with malicious intention. The
constant battle waged by US. government-funded defense science of the 21St century is
to evolve at a faster pace than the evolution of malicious science. But evolution speed is
not the only important factor in such a research project. A fundamental understanding of
the processes being studied as well as the analytical thoroughness of the work and its
place in a larger context are also crucial. Thus, a technique developed without attention
to these factors may be doomed to failure in the future if the fundamentals are not also

studied.

This work began with a two-fold approach — seek out subjects who may have exposure to
explosives who can serve to provide initial signatures and proof of concept results to

determine if breath analysis will be successﬁrl at detecting explosive markers on breath,

168

 

VI ; m.1_- __‘I'
J
.

 

and perform more fundamental studies into what constituents are present in breath, how
they can best be detected, and what useful results can be obtained from this relatively
understudied matrix. This two-fold approach focused on both the speed of evolution and

the analytical thoroughness in parallel.

This research has developed and evaluated a breath analysis technique that has been
proven effective at detecting explosives and explosive-related compounds in human
breath after exposure to these substances. The technique uses commercial
instrumentation which would allow it to be disseminated and used in any lab with a
standard GC-MS almost instantaneously. The original goal of the project has therefore
been met. But along the way, the dual nature of the scientiﬁc approach to this problem
allowed additional useful information to be discovered from studies involving
unintentional chemical exposure at nail salons, in hardware stores, and hem dietary
intake, which contributed to the development of a stronger detection method, and added

more conﬁdence in the results obtained from the exposures to explosives.

The future of this work must then build upon current results pertinent to explosives
detection. In order to advance breath analysis for the detection of explosive makers,
enhancements to the sampling and detection system should be made to allow for a ﬁeld-
deployable device. The constant push for deployable systems requires a smaller size both
in footprint and weight, low power consumption, higher sensitivity and selectivity, and a
shorter analysis time. In order to meet these criteria, signiﬁcant changes will need to be

made to the analytical scheme.

169

has“

 

 

In order to reduce the size of the device, the detector may need to be changed. While
GC-MS is a powerful analytical tool, it does not serve well for miniaturization. Mini-
GCs have been developed and show remarkable efﬁciencies and resolving power;
however, miniaturizing a mass spectrometer has been more challenging. Although
miniature mass analyzers have been developed}5 it is the remainder of the system,
including the electronics, power supplies, and vacuum systems that are difﬁcult to‘
miniaturize while retaining functionality and the pumping speeds necessary to maintain
the required vacuum. GC-GC, and GC-DMS have been successfully miniaturized, and
may show promise for the future of breath analysis. Another aspect that will have to be
miniaturized and incorporated into a hand—held system is the preconcentrating device,
currently the SPME ﬁber. The ideal system would incorporate the preconcentrator in an
in-line manner, so the sample could be obtained, preconcentrated, and analyzed in a
single action. Options for miniaturized preconcentrators are plentiful, as even the
polymer on the SPME ﬁbers used in the current study could be incorporated in a high-
surface area manner onto different miniaturized structures. Carbon nanotubes or
micropore or micropillar technologies are also in development at various laboratories,
and these technologies may be useful for a future breath analyzer. The preconcentrator
may also be adapted to increase the sensitivity and control the selectivity of a future
instrument by selecting materials that would preferentially absorb the compounds of
interest. These preconcentrating materials could perhaps be arranged in a consecutive
fashion so that multiple materials speciﬁc to particular compounds could be used and

serve as a crude separation step in the analysis.

170

 

In order to reduce the analysis time, the time of sampling must be reduced. If this time is
to be cut, attention must be paid to what part of a breath is sampled. Figure 5- 1 shows a
typical time capnogram which can be used to describe the respiration pattern of a healthy
adult. Carbon dioxide is used as a tracer gas, and the capnogram shows at what points of

respiration the concentration of C02 is the highest in the respiratory gas.6

e > a» 9)
3' 5° 3” f.”
03“?" eggs“? Dye

 

Pco2

 

   

 

D—-——-————-- ——-

time

Figure 5- 1: Diagram of a typical time capnogram, which displays the concentration of
carbon dioxide in respiratory gas. Inhalation is represented by Phase 1-0, exhalation is
represented by Phases E-I, E—2, and E-3, corresponding to exhalation of deadspace (C02
free) air, a mixture of dead space and alveolar air, and primarily alveolar air,
respectively. Adapted from reference 6.

The present work sampled breath from all phases of respiration. It can be seen from the
ﬁgure that the highest concentration of C02, and therefore other VOCs that diffuse across
the pulmonary alveolar membrane, occurs during Phase E-3, when the exhalate contains
primarily alveolar air. If the time during which a breath sample is obtained is reduced,
efforts must be made to sample during Phase E-3, and not Phase E-l or E-2, to ensure the

sample is collected while the concentrations of VOCs are the highest. If the sensitivity

171

-r-—-—

 

 

and surface area of the preconcentration material is improved, reducing the sampling time
to this window should still provide enough signal for detection. Figure 5- 2 shows a
potential schematic for a future breath analysis device, incorporating all of these

modiﬁcations.

Mouthpiece Preconcentration Separation (GC?) Detection (MS?)

Exhaled Breath LW -

Figure 5- 2: Diagram of possible handheld breath analyzer. Breath is exhaled into the
mouthpiece where it passes into a thermally controlled preconcentration region. After a
predetermined period of time, the temperature of the preconcentrator is quickly ramped,
desorbing the analytes of interest and allowing them to pass into the separation region,
possibly a miniaturized GC where separation occurs. The eluent then passes into a
detection system, possibly a miniaturized MS where the analytes are identified and a
signal is reported to the user (signal digitization not shown).

 

 

 

This technique should not only be pursued for the detection of explosives on breath. This
work demonstrated proof of concept data for the use of breath analysis for detecting the
prior location of an individual based upon chemical signatures of breath constituents;
however, this information also demonstrates a non-invasive way to monitor chemical
exposure. Another promising avenue of research may be found in the oral exposure
studies that were pursued. This presents breath analysis as a method for assessing the
dynamics of the metabolic processes in an individual. The ability to visualize the

absorption and uptake phase of exposure would also provide valuable pharmacokinetic

172

 

 

 

data if similar behaviors were measured for other orally ingested drugs or compounds.
Using breath to monitor intravenous exposure to drugs or chemicals would also present
potential applications for breath analysis, but would require subjects to already be
undergoing medical treatment, and may be difficult to justify to an Institutional Review

Board.

Several avenues of research should be pursued to improve the breath research ﬁeld.
There is no accepted and standardized method for breath collection and analysis. This
makes it difﬁcult to compare results with those of another laboratory using a different
system. It is hoped that the future will hold some standardization practices and
recommendations. This should include the temperature and time of breath collection,
ﬂow rate of breathing, type of sampling (alveolar versus total air), material of the
collection device, and preconcentration methods. The ﬁeld would also beneﬁt from the
existence of a breath reference standard. Figures of merit and analytical method
performance are difﬁcult to establish without a standard breath, and no two human
breaths are exactly alike. Although an EBC standard would not be ideal for experiments
involving the collection of EBV, condensate may be a more reliable standard, as a real
breath would contain a mixture of vapor and suspended aerosols that would not be
amenable to storage, and condensate could be refrigerated or frozen for stability. In
whatever form, this standard would have to contain a complex matrix similar to that of
human breath, resulting in hundreds of compounds being necessary. Once a standard is
developed it would be a logical step to spike this breath sample with target compounds to

determine more reliable limits of detection. Such fundamental studies would allow data

173

 

 

to be more easily compared across research groups, and would improve the possibility of

successﬁil applications of breath analysis.

174

 

 

5-3: References

(1)

(2)

(3)

(4)

(5)

(6)

Peterson, J. L.; Hickman, M. J. In Bureau of Justice Statistics Bulletin; US.
Department of Justice, 2005.

Henry, C. M. Chemical and Engineering News 2002, 80, 34—35.

Laughlin, B. C.; Mulligan, C. C.; Cooks, R. G. Anal. Chem. 2005, 7 7, 2928-2939.

Badman, E. R.; Cooks, R. G. J. Mass Spectrom. 2000, 35, 659-671.

Fico, M.; Yu, M.; Ouyang, Z.; Cooks, R. G.; Chappell, W. J. Anal. Chem. 2007,
79, 8076-8082.

Bhavani-Shankar, K.; Philip, J. H. Anesth. Analg. 2000, 91, 973-977.

175

Le

 

 

APPENDIX A — Additional Rules Trees for Drug Identiﬁcation Using SPAMS

 
  
 

All Particles
i

m/z

166

N

      

 

   

 

N Y
m/z m/z m/z @
y 58 148 58
N m/z Y N m/z m/z
-200/-201 m/z 148 Y -71

     
  
   
  

        
    
 

      
  
 
 

S8
m/z N
Y '71 Y N m/z 3
m/z 71,-73 m/z
'200 Y 166/167
Y @
N Y

       
   
  

     

m/z
148/149

   

@N ”@

 

Y

    
   

m/z
-73

@

m/z
166/167

Y

m/z

118/117

@

   

N

  
 

 

m/z
166/165

Y

          
  

  

N

 

 

Figure A- 1: Rules tree for classification of pseudoephedrine. The mass vectors probed
at each branch division are shown, with ‘1” indicating a positive response and ‘N
indicating a negative response. Circles labeled ‘PSE ’ indicate a positive identification

for pseudoephedrine; circles labeled ‘Null’ indicate a negative response.

176

 

All Particles

i

m/z
320,321

 
     
 
   
 
  
 
    
    

N

N
. N

  

m/z
286

  
 
   
 

  
  

. -I95
m/z
m/z
m/z _79 -195
Y '32] m/z N

m/z

.. m
N

1: l y
m
m/z ,
m m <9
-133
N

159 P(attic-J

Figure A- 2: Rules tree for classification of Coricidin. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘C’ indicate a positive identification for Coricidin;
circles labeled ‘Null’ indicate a negative response.

  
   

    

N m/z
-l95

Y

   

m/z
-178

 

m/z
-178

 
 

All Particles
i

m/z
280
N

m/z
245

   
 
   
    

m/z
217

Y

N

m/z .
266 @1

m. 0 m

Figure A- 3: Rules tree for classification of loratadine. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘L’ indicate a positive identiﬁcation for loratadine;
circles labeled ‘Null’ indicate a negative response.

 

177

14

 

 

All Particles

i

m/z
152

 
     
   
      

 

Y m/z
~~ N Y -151 Y
m
" 109
Y N ‘N .
m/z {$1

  

Y m/z

m

Figure A- 4: Rules tree for classiﬁcation of acetaminophen. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘Ac’ indicate a positive identification for
acetaminophen; circles labeled ‘Null’ indicate a negative response.

‘ “xi \i-

 

All Particles

 

Figure A- 5 .' Rules tree for classification of dextromethorphan. The mass vectors probed
at each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘D’ indicate a positive identification for
dextromethorphan; circles labeled ‘Null’ indicate a negative response.

 

 

I78

All Particles

l

 
    
 
  
    

m/z
l 19,120,121,122

Y
Y

m/z m/z
—137 163,164

N m/z
EW ‘ m/z
~- -l79,-180

 

m/z -179,-180 A

-137 Y I ‘T

N e '7
m

m/z
-229

_N Y
13 6

Figure A- 6: Rules tree for classification of aspirin. The mass vectors probed at each
branch division are shown, with ‘Y’ indicating a positive response and ‘N’ indicating a
negative response. Circles labeled ‘A’ indicate a positive identiﬁcation for aspirin;
circles labeled ‘Null’ indicate a negative response.

-93

   
    

Y

 

 

179

All Particles

i

      
 

   

   

   
      

m/z
168J69
1N Y

1 .2 nﬂz Y
“ 150 nﬂz
150

nﬂz 0’ Y
N N -202 m, g

m/z _71

44
nﬂz
-157

mm

 

    

m
0N
«m1

 

44 . _
P

 

Figure A- 7: Rules tree for classiﬁcation of phenylephrine. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘PE’ indicate a positive identification for
phenylephrine; circles labeled ‘Null’ indicate a negative response.

All Particles
i

nﬂz
I99

 
    

nﬂz
Y -l98 m/z

   
 

1N Y
Figure A- 8: Rules tree for classiﬁcation of guaifenesin. The mass vectors probed at
each branch division are shown, with ‘Y’ indicating a positive response and ‘N’
indicating a negative response. Circles labeled ‘0’ indicate a positive identification for
guaifenesin; circles labeled ‘Null’ indicate a negative response.

180

 

WI

 

APPENDIX B - Additional Terpene and Terpenoid Data from Lemonade Consumption

Included herein are the data obtained from several subjects after consumption of original,
concentrated, and Mediterranean lemonade. Thirteen terpenes and terpenoid compounds
were monitored; peaks due to each compound in the TIC of each breath sample were
integrated, and the area plotted versus time after lemonade consumption. The resulting

curves were ﬁt to Equation 4-l, describing the rise and decay in each compound. See

 

Chapter 4 for additional information.

N
O
J

17.
1

Peak Areax106 (a.u.)
1,. 23
A l

 

 

1 1 1 1 1 1
O 200 400 600 800 1000 1200
Time after Consumption (min)

—

Figure B- 1: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the first day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I

for ﬁt parameters.

181

 

i—n
0‘ .
l

1

ES
1

1

Peak Area x106 (a.u.)
4}. OO
1 1

l

 

 

 

1 1 1 1 1 1 1 1
0 100 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 2: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- I for ﬁt parameters.

 

 

 

.o

r710-

:1

53

\O 8—

S

>< 6-

8 o

<‘E 4-

:4

51"}

12.. 2“

o

1 r 1 1 1 1 1
0 100 200 300 400 500 600

Time after Consumption (min)

Figure B- 3: Integrated area of the peak attributed to a-limonene in chromatograms

obtained from the breath of subject A on the third day of original lemonade consumption.

The data has been fit with an exponential rise and decay (Equation 4-1). See Table B— I
forjit parameters.

182

 

 

Momma;
llllll

Peak Area x106 (a.u.)

 

O
I l l l l I l l

200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

 

C

Figure B- 4: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been fit with an exponential rise and decay (Equation 4-1).
See Table B- I for ﬁt parameters.

C N
1 1

Peak Area x106 (a.u.)

r

 

1 1 1 1
100 200 300 400
Time after Consumption (min)

Figure B- 5: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the fifth day of original lemonade consumption.
The data has been fit with an exponential rise and decay (Equation 4-1). See Table B- 1

forfit parameters.

183

 

 

 

 

 

50—1 . . . . . . - . I .
l T l l

100 200 300 400
Time after Consumption (min)

Figure B- 6: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I
for fit parameters.

' l
A 500 4i

 

 

 

l l l l l l l
0 100 200 300 400 500 600 700

Time after Consumption (min)

 

Figure B- 7: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been fit with an exponential rise and decay (Equation 4-1).
See Table B- 1 for fit parameters.

184

 

 

 

’7 _l
:2 600

a 400 -—
(6

E 300 —
A; 200 —
100 —

Pe

 

 

 

 

0 100 200 300 400 500 600
Time after Consumption (min)

Figure B- 8: Integrated area of the peak attributed to 4-teipineol in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B- I
for ﬁt parameters.

 

 

500 -
:3; 400 —
“’2
C6
0)
a 200 -
A4
8
l _.
12.. 00 . . .
O
l l 1 l l
100 200 300 400 500

Time after Consumption (min)

Figure B- 9: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B- l for ﬁtparameters.

185

 

 

500 -.

Peak Area x103 (a.u.)

 

 

 

l l |
100 200 300 400

Time after Consumption (min)

Figure B- 10: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

A 250 -
:1
:6
M" 200 -'
S.
X
en 150 -
‘2
<3
1 _
E 00 .
G-t O Q
50 _' . ‘ . .
l l l T—
100 200 300 400

Time after Consumption (min)

Figure B- I 1: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

186

Pr-r

 

 

A700 41‘

a, 600 -
"’3 500 -

i 400 -

Q)

E 300 J
'25

q, 200 A
O-r

o

100- . 9
l l T l l l l

0 100 200 300 400 500 600 700

Time after Consumption (min)

 

 

 

Figure B- 12: Integrated area of the peak attributed to a—terpineol in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

Peak Area x106
9 .o .o
N A O\
1 1 1

 

 

 

 

W 1 1 1 1 1 1
0 100 200 300 400 500 600
Time after Consumption (min)

Figure B- 13: Integrated area of the peak attributed to a-tetpineol in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

187

 

 

 

 

lOO- . . .
T—‘T 1 r 1 1 1 1

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

Figure B- 14: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

 

 

800-.
3' .1
:6
”V 600 —

53 -

K

g 400 w

“is .

. d . . .
j l I ﬂ l
100 200 300 400

Time after Consumption (min)

Figure B- 15: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject A on the ﬁﬁh day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

188

Au—

"1

 

 

r7140-

F-ﬁ —
O N
O O
l 1

Peak Area 11103 (a.u
93°
1

O\
O
I

 

 

 

 

1 1 1 1 1 1 1
0 l 00 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 16: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

L. .

 

 

:5

:5 80—

m .

d)

a o

x 60-1 0

Cd 0

a: - - '
l T l l I' 1
0 100 200 300 400 500

Time after Consumption (min)

Figure B- I 7: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

189

 

.1
‘-

 

0\
Ln
1
O

 

:3

si60-.
"éss— '

><

gsod.

‘3 o o
x45ﬁ

8 .0 C .
9.. oo o.

40“ O...
1 1 1L 1 1 1 1 1

 

O 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 18: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
See Table B-1 for ﬁt parameters.

 

 

 

 

. o

380—

3

W:,70-—

i o

360- .

if: 50 _

£3 ‘ . '

O . 44
0 o

1 1 1 1 '
100 200 300 400

Time after Consumption (min)

Figure B- 19: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

190

I"!

 

 

40 ‘ . .
1 1 1 T—
100 200 300 400
Time after Consumption (min)

 

 

Figure B- 20: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

—

T 1 1 1 1 1
200 400 600 800 1000 1200
Time after Consumption (min)

C

Figure B- 21: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

191

 

 

6

Peak Area x10 (a.u.)
.— :— N N

O U! Q UI

I I I I

.O
U:
I

 

 

I".
I I I T I I I |

O 100 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 22: Integrated area of the peak attributed to y-tetpinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

:-‘ :— N N
O U: 0 U:
1 1 J 1

Peak Area x106 (a.u.)
.9
I

 

 

 

0 100 200 300 400 500 600
Time after Consumption (min)

Figure B- 23: Integrated area of the peak attributed to y—terpinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

192

 

 

go
O '
l

pd
u:
l

Peak Area x106 (a.u.)
.o u—5
U: C
1 1

 

 

O
l I I I l I I

0 200 400 600 800 I 000 l 200 I400 .
Time after Consumption (min) t u‘

 

_

 

Figure B- 24: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

N .N
O U:
I J

Peak Area )th6) (a.u.)
G
1

 

 

 

 

Time after Consumption (min)

Figure B- 25: Integrated area of the peak attributed to y-terpinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

193

256 —

Peak Area x103 (a.u.)

 

 

50 — “5'”..-

1 1 1 1 r 1
O 200 400 600 800 1000 1200
Time after Consumption (min)

Figure B- 26: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4—1). See Table B-1
for ﬁt parameters.

M

100—
o

 

o o
I ' g o
1 1 1 1 1 1 1 1
O 100 200 300 400 500 600 700

Time after Consumption (min)

Figure B- 27: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on. the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

194

 

 

 

(H II 1

\

will

N

O

O
I

Ln
O
l

100 _‘ Q 0

Peak Area x103 (a.u.)

 

 

.0. i..
I I I I I I l
0 100 200 300 400 500 600

Time after Consumption (min)

 

Figure B- 28: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

 

-o

370‘ .

"2

x65- 0

«s

o

260— 0 o

£5

£3 ' o

55- o o o
0 o

1 r 1 1 1
100 200 300 400 500

Time after Consumption (min)

Figure B- 29: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-l).
See Table B-1 for ﬁt parameters.

195

 

 

 

 

 

100 200 300 400
Time after Consumption (min)

Figure B- 30: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

Ox
Q

U:
C
L_ﬁi

Peak Area x103 (a.u.)
‘3
1

 

 

 

1 1 1 1
100 200 300 400
Time aﬁer Consumption (min)

Figure B- 31: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

196

 

 

 

 

Aranvl 1'! II ‘

II.-..

Peak Area (a.u.)

 

 

 

O 100 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 32: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).

See Table B-1 for ﬁt parameters.

N
O
O
O
I

1500 -*

Peak Area (a.u.

1000 ‘1

500 A

 

 

 

O I 00 200 300 400 500 600
Time after Consumption (min)

Figure B- 33 .° Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1

for ﬁt parameters.

197

 

 

A.;..J :3..< 1:3».

0b

{'0

Se

Peak Area (a.u.)

 

 

 

 

1 1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 34: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

3000 4
2500 -

2000 -‘

 

1500 -4'

Peak Area (a.u.)

1 000

 

500

 

1 1 1 1
100 200 300 400
Time after Consumption (min)

Figure B- 35 .' Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1

for ﬁt parameters.

198

 

r-r'

 

 

 

 

 

 

1 r 1
200 400 600 800 1000 1200
Time after Consumption (min)

Figure B- 36: Integrated area of the peak attributed to p—cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.

The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

~400—

U

Peak Area x103 (a
'-‘ N U)
8 8 8
1 1 4

 

 

 

O 100 200 300 400 500 600 700
Time after Consumption (min)

Figure B— 37 .' Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the second day of original lemonade

consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

199

 

 

 

 

 

 

 

0 100 200 300 400 500 600
Time after Consumption (min)

Figure B- 38: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

 

1 1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B— 39: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

200

 

 

 

H .
s.

6....

.

.\. _I\. a..- . III
1. 1.

”v A: x :U-_/\ .433..—
a

F122“

g

01m

The .

turf

73411:! .....-.< .1: C

Fi

T1

fr.

 

 

 

I I I l
100 200 300 400

Time after Consumption (min)

Figure B- 40: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

1 1 1 1
100 200 300 400
Time after Consumption (min)

Figure B- 41: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

201

 

 

 

 

 

: 1|“ (2| LL)

(ll/II;

('01

(41.11.)

1

t“
V
X
:5
<
:1
5'.
A
-_

 

 

 

Peak Area x106 (a.u.)

 

 

0 I 00 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 42: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

 

Peak Area x103 (a.u.)
§
1.;
O

 

 

1 1 1 1 1 1 1
0 1 00 200 300 400 500 600
Time after Consumption (min)

Figure B- 43 .' Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1

for ﬁt parameters.

202

 

 

 

(a.u

M 400—

Peak Area x10

~ 0

1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

 

 

—1

Figure B- 44: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

450 -

u
4:.
O
O
I

350 -
300 -
250 -
200 -
150 -

M

0 (a. .)

Peak Area x1

 

 

 

100 200 300 400
Time after Consumption (min)

Figure B- 45 .' Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

203

 

 

 

J)
I .

w
l

Peak Area x106 (a.u.)
H N
I I

 

 

—

1 T 1 1 1 1
0 200 400 600 800 1000 1200
Time after Consumption (min)

Figure B- 46: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

Peak Area x10

 

 

 

' o
1 1 1 1 1 f 1
O 100 200 300 400 500 600 700

Time after Consumption (min)

 

Figure B- 47: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

204

 

 

I’cuk Aron x10“ (a.u.)

(n n )

(v

Aron] 1! ll)

')1IIIL’

L2-

Peak Area x106 (a.u.)

.o .o
A O\
I

 

 

 

0 100 200 300 400 500 600
Time after Consumption (min)

Figure B- 48: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

A 1.6-I
=5 1.4-—
EL
oo 12"
>< 1.0-
Cd
é’os— .
.54 0.6- .
3 ,
13.. 0.4-
- I
0'2 1 1 1 1 1 1 1 1

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 49: Integrated area of the peak attributed to ﬂ-pinene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

205

I‘ﬂ-

 

 

I’m-1k Arcu xI()h(;1.u_)

Fig;
0qu
The
for]

 

(a.u.)

1

Peak Arcu x I0"

Peak Area x106 (a.u.)

 

 

 

100 200 300 400
Time after Consumption (min)

Figure B- 50: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1

for ﬁt parameters.

 

 

 

 

:2 55 ﬂ

EL
"‘2 50 —

X

345 "" .

x 40 -

CO I

On 35 _ . . O

Q 0 O
I I -I I
100 200 300 400

Time after Consumption (min)

Figure B- 51 .' Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B—1

for ﬁt parameters.

206

 

 

 

 

 

 

0 100 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 52: Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

 

 

' o
A 54 -’ o .
' ‘ O O
S
3; 52 - . .
M 50 — C

2 o
>< 48 4N
“I o
E 46 ﬂ .

44 _
E . Q
On 42 -

4O _ I I I r % I I

O I 00 200 300 400 500 600

Time after Consumption (min)

Figure B- 53 .' Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

207

‘IT 'L-'..-Llnno l

 

 

A
O

 

Peak Area x103 (a.u.)

 

 

 

 

1 1 1 1 1
0 200 400 600 800 1000 1200 I400
Time after Consumption (min)

Figure B- 54: Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

3‘50— 0
848—
246-
X

344—

£542.

£40- . . .

 

 

1 f 1 1
100 200 300 400
Time after Consumption (min)

Figure B- 55: Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

208

 

 

 

_|..

1

‘In

7:43 53:.“ Sui/x 1:9.—

F1bu'

ObILl

The .

Tor f1

72....» .A: r. ...U-~< x301

CO1:

395 I

80-

Peak Area x103 (a.u.)
8
1

4:.
O
I

 

 

 

100 200 300 400
Time after Consumption (min)

Figure B- 56: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4 -1). See Table B-1
for ﬁt parameters.

 

 

 

1 1 1 1 1 1 1 1
O 1 00 200 300 400 500 600 700
Time after Consumption (min)

Figure B- 57: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

209

 

 

 

 

 

 

 

50'-
:5
5,; 45 -
Mo 0
g 40 —
8
E 35 -
'26
a: 30 4 T.
o . 4.
1 1 1 1 T 1 1
0 l 00 200 300 400 500 600

Time aﬁer Consumption (min)

Figure B- 58: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

451 .

A
O

t.»
O
A

 

Peak Area x103 (a.u.)
b.“
1

 

 

 

1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 59: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

210

 

A
O
I

u
m
4

 

 

Peak Area x103 (a.u.)

 

 

. o 0 o
30_ _L .
. O
o . '
I "I I l
100 200 300 400

Time after Consumption (min)

Figure B- 60: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

:3: 500d
8

Mo 400 —

x1

=3 300-

200 A

 

Peak Are

100 -I .

I O
I T I I

100 200 300 400
Time after Consumption (min)

Figure B- 61: Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

211

 

 

 

 

 

 

4- o
1 1
0 100 200 300 400 500 600 700

Time after Consumption (min)

 

Figure B- 62: Integrated area of the peak attributed to ﬂ-myrcene in chromatograms
obtained from the breath of subject A on the second day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

)

N
Ur
o .
I

(a.u.

3
N
O
O
I

Peak Area x10
0

 

 

50q . . .9.
I I I I I I I

0 l 00 200 300 400 500 600
Time after Consumption (min)

 

Figure B- 63 : Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject A on the third day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

212

 

 

Peak Area x103 (a.u.)

 

 

50-

_

1 1 F 1 1 1 1
0 200 400 600 800 1 000 1 200 1400
Time after Consumption (min)

Figure B- 64: Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject A on the fourth day of original lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

100 m

Peak Area
Z7.
O
1

 

M
O
I

 

 

100 200 300 400
Time after Consumption (min)

Figure B- 65 : Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁfth day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

213

 

 

 

or
T le
for

 

 

100 —T

00
o
I

O\
O
|

 

4:.
O
I

Peak Area x106 (a.u.)
8
1

' o
1 1 1 V 1 1 1 1
O 200 400 600 800 1 000 1200 1400
Time after Consumption (min)

 

Figure B- 66: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

@70'0 JT

3600i

 

"g 500 -
X
3 400 —
<1” 300 —

 

33200—
f
100— ~ '00..

1 1 1 1 1 1 1
0 200 400 600 800 I 000 l 200 I 400
Time after Consumption (min)

 

Figure B— 6 7: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

214

 

’2‘ 800 -T

3 700 -
mo 600 -

><

a 500 —

O)

E 400 —

i6 300 —

Q)

°" 200—

 

 

. O
I L I I I I I I
0 200 400 600 800 1000 1200 1400

Time after Consumption (min)

 

Figure B- 68: Integrated area of the peak attributed to a—tetpineol in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

I

'0‘.
.1—

—
N
I _L

 

L

1

Peak Area x106 (a.u.)
O
00
I

.O
A
I

 

l

 

no.9 '
1 1 1 1 1 1 1 1

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 69: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

215

- ‘—

 

 

 

 

 

1.0 —T
”r?
3 0.8 —
\O
O
“:5 0.6—
t3
0
E 0.4-
.M
8
CL. 02- o
O
l I I l I l ﬂ 1

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

Figure B- 70: Integrated area of the peak attributed to a-tetpinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

’ I

3.
J__

23
r

 

Peak Area x106 (a.u.)
1

 

 

_ O
I I I I I I I I

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 7]: Integrated area of the peak attributed to y-tetpinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1

for ﬁt parameters.

216

 

 

 

 

22100—

3

S 80—

N

(U

G)

<2“ 60-

%: o

33 4o . '
I o to...
I j _ﬁ I I I f I

 

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 72: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

O
Q.

0.2—I ....
I I _ I I T I I

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

 

 

q

Figure B- 73 : Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

217

 

 

 

al.6—

:3

53; ..

0o1.2-

:; _

§0.8-

<3 .

%

g 0.4—I .

I n¢0"..' .
I I

 

1 1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 74: Integrated area of the peak attributed to a—pinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.

The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

' I
2 8 J
EL
cc: 6 I
v; 1
4 —
<
%
a) 2 _
o.
.
1 1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400

Time aﬂer Consumption (min)

Figure B- 75: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.

The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

218

 

”£110—

(
S
O

1

 

3
PeakAreaxlO
O\ \l 00 \O
O O O O
1 1 1 1
o
o
o
o

 

1 1 1 1 1 1 1 1
0 200 400 600 800 l 000 l 200 1 400
Time after Consumption (min)

Figure B- 7 6: Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

1

 

3

Peak Area x10 (a.u.)
u—I '— N N
UI O UI O Ut
O O O O O
1 1 1 1 1

 

 

1 1 1 1 1 f 1 1
0 200 400 600 800 1000 l 200 1400
Time after Consumption (min)

Figure B- 7 7: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

219

 

 

 

 

 

O
I I I I I I I I

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

 

 

Figure B- 78: Integrated area of the peak attributed to ﬂ-myrcene in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

Peak Area x106
0\
1

 

 

1 1 1 1 1 1 1 1
200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

O-1

Figure B- 7 9: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

220

15
a“: 200— ,

 

100— ' o o g
1 1 1 1 1 1

50 100 150 200 250 300
Time after Consumption (min)

 

Figure B- 80: Integrated area of the peak attributed to 4-terpineol in chromatograms
obtained from the breath of subject B on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

'9
I

O
00
I

Peak Area x106 (a.u.)
g> <3
A O\
1 1

O
N
I

 

. .
’ 0 i r o
I I I I I I

50 100 150 200 250 300
Time aﬁer Consumption (min)

Figure B- 81: Integrated area of the peak attributed to a-terpineol in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

221

 

 

 

 

 

’3 600 —
5‘;
.., 500 -
S
X
a 400 —
d)
2 300 —
ﬁg 0
g: 200 - '. .
'2'.

 

1 1 1 1 1 1 1 1
0 200 400 600 800 1000 1 200 1400 1 600
Time aﬁer Consumption (min)

 

Figure B- 82: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

1 1 1 1 1 1 1 1
0 200 400 600 800 1000 I 200 1400 1 600
Time aﬁer Consumption (min)

Figure B- 83 : Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

222

 

I"
m .
I

s

S,
20‘

No

2

X15—

C3

0.)

21o-

M

3

a“ 0.5‘

 

 

O
I I I I I I I I I

O 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 84: Integrated area of the peak attributed to y-teipinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

f:
:s
3; 60 -
mo
3504 .
cu
E
_ o
.M 40 .
3 .
a. 30 — ' o.
g
1 I 1 1 1 1 1 1 1
O 200 400 600 800 1000 1200 1400 1600

Time after Consumption (min)

Figure B- 85 : Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

223

 

 

 

 

 

 

 

 

 

 

1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

 

Figure B- 86: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

500 —

)

(a.u

m 400 -10

L»

O

O
I

 

Peak Area x10
N
8
1

 

 

1 r 1 1 1 1 1 1 1
0 200 400 600 800 1000 l 200 1400 l 600
Time after Consumption (min)

 

Figure B- 8 7: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

224

Peak Area x106 (a.u.)
53 F’ r' 7‘
O\ OO O N '
1 1 1 1

F’
4;
I

 

 

 

1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

 

Figure B- 88: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

 

 

 

75-— '
’3 70 e
V 65 -
$2 60 —
g 55 — .
50 — .
a o‘ .
at: 45 -‘ . ’..
40 .1 ..
‘1— 1 1 1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400 1600

Time after Consumption (min)

Figure B- 89: Integrated area of the peak attributed to ,B-phellandrene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

225

 

 

 

 

”3750—

5?;
mg 45_ .

X

C

840—

2 .

x

335—

m . '0

.‘o
O
I— I I I I I I I I

 

0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 90: Integrated area of the peak attributed to a—thujene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

A 350 —'

3300-

'“c 250 a
200 -
E 150 —
100 —
50 -

ea x1

Peak

 

 

 

1 1 1 1 1 r 1 1
0 200 400 600 800 1000 1 200 1400 l 600
Time after Consumption (min)

-—1

Figure B- 91: Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject C on the ﬁrst day of original lemonade consumption.
The data has been ﬁt with an exponential rise and decay (Equation 4-1). See Table B-1
for ﬁt parameters.

226

 

I

 

 

 

925—
3
o 20-
2
>< 15—1
at
Q)
2 10-
.M O
O
I I l I
0 500 1000 1500

Time after Consumption (min)

Figure B- 92: Integrated area of the peak attributed to a-limonene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

(a.u.)
M
O
O
1

300 —

Peak Area x10
N
8
1

 

 

J

1 1 1 1 1 1 1 1
200 400 600 800 1000 1200 1400
Time aﬁer Consumption (min)

O

Figure B- 93 : Integrated area of the peak attributed to 4-tetpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

227

A500—

(a.u
A
O
O

1

3
Peak Area x10
N b)
O O
O O
1 1

O
O
4
O
O
O

 

 

.00.
I I I I I I I I

200 400 600 800 1000 1200 1400
Time after Consumption (min)

O

Figure B- 94: Integrated area of the peak attributed to a-tetpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

N U.)
Ur O
O O
I l

Peak Area x103 (a.u.)
§ 2 §
1 1 1

LII
O
O

 

 

1 1 1 1 1 1 1
200 400 600 800 1000 1200 1400
Time after Consumption (min)

c—

Figure B- 95: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

228

 

DJ

O

O
I

Peak Area xlO3 (a.u.)
‘6
1

 

50 — .
1 j 1 1 1 1 1 1
200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

O

Figure B- 96: Integrated area of the peak attributed to a-terpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

.b
I .

w
4

Peak Area 11106 (a.u.)
"I T
O

 

 

O
I I I I f I I I

0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 97 : Integrated area of the peak attributed to y-tetpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

229

 

 

 

 

1 1 1 1
0 500 1000 1500
Time after Consumption (min)

Figure B- 98: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

.9 .o t" r"
ox oo o N '
I J I I

 

Peak Area x106 (a.u.)
.9
A
I

 

 

 

 

I I I | I I I l
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 99: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

230

 

u.)
5:;
111

¢>
I

 

Peak Area xlO6 (a.
5: <3
A O\
1 _1
k

 

53
N
I

 

i
| I I I

0 500 1000 1500
Time after Consumption (min)

 

Figure B- 100: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

71
36-
s

o 5-
O

:4-
(3

933-

Ag ze'I

o
54]»—

 

 

 

_ .
I I I r

0 500 1000 1500
Time after Consumption (min)

Figure B- 101: Integrated area of the peak attributed to ,B-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

231

 

 

 

 

1 1 1 1 1
0 200 400 600 800 1000 1200 1400
Time after Consumption (min)

 

Figure B- 102: Integrated area of the peak attributed to ,B-phellandrene in
chromatograms obtained from the breath of subject A on the ﬁrst day of concentrated
lemonade consumption. The data has been ﬁt with an exponential rise and decay
(Equation 4-1). See Table B-1 for ﬁt parameters.

 

 

 

 

250 —
2?

3 200 —
Mo

:5 150 -
Cd
‘3.

<1 100 —
x
«I
<1)

°~ 50 -

F
I 41
0 500 1000 1500

Time aﬁer Consumption (min)

Figure B— 103: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

232

 

 

3: 700 —
3 600 —
"’3 500 -
E 400 —
Q)
E 300 -
A4
3:3 200 - o
100 -1

 

 

I
l l I l I l I

O 200 400 600 800 1000 1200 1400
Time after Consumption (min)

Figure B- 104: Integrated area of the peak attributed to ﬂ-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of concentrated lemonade
consumption. The data has been ﬁt with an exponential rise and decay (Equation 4-1).
See Table B-1 for ﬁt parameters.

 

C
A1004 0‘
:3
c6 0
v 80‘. '
\o O
52
X 60‘ .
«3

o
E 40— o
.M I. '0
8 20—
9.. x“.
~ 0 o o

 

1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 105: Integrated area of the peak attributed to a-limonene in chromatograms

obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

233

 

 

=2 400-
3 J.
m , 00.
o —
3.5 300 . .
8 . O O
:5 200— an
x o ' 0
(U
Q)
m 100— ‘1." ° .
O. .

 

—

1 1 1 1 1 1 1 1
0 200 400 600 800 1 000 l 200 1400 1600
Time after Consumption (min)

Figure B- 106: Integrated area of the peak attributed to 4-tetpineol in chromatograms
obtained from the breath of subject an on the ﬁrst day of Mediterranean lemonade
consumption.

 

.
5500—
3
3L
m 400—
52
X
a 300— .
E o
.
M 200‘ ..
8 I I,” o
a“ 100-i .h. . .
.0
I J

 

1 1 1 1 1 1 1 I
O 200 400 600 800 11000 1200 1400 1 600
Time after Consumption (min)

Figure B- 107: Integrated area of the peak attributed to a-tetpineol in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

234

 

 

O

A "a

:5 800— '-

3; O
m C

o 600—

? O 0

<6

0 400— '

2 o

O

:3 200 " o

O _

9.. R.,. .~ .

. o

 

—

1 1 1 1 1 1 1 1
0 200 400 600 800 1000 l 200 1400 1 600
Time after Consumption (min)

Figure B- 108: Integrated area of the peak attributed to terpinolene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

 

800 — ,
c7 ' O
3
«1 . '
v 600 ~ ..
>< o
«s 400 -
Q) C
2 l. - '
M 5
(:3 200 — '. .
x“ C .
O

 

1 1 1 1 1 1 1 1 1
O 200 400 600 800 1 000 l 200 1400 1600
Time aﬁer Consumption (min)

Figure B- 109: Integrated area of the peak attributed to a-terpinene in chromatograms

obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

235

 

”716— 0‘

:3 _ o

3 0'

c 12—' 0
o

v— 4 .
X 8

«3 ~ 0
2 J .
<1 0 ..
% 4J

G.)

a.

L

‘00.... o

1 1 1 1 r 1 1 1
0 200 400 600 800 1000 1200 1400 1600
Time aﬁer Consumption (min)

 

_

Figure B- 110: Integrated area of the peak attributed to y-tetpinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

 

o
:i '0'
M o
3150—
>< o
3 o
_ o
2100 . .
‘cé . ~ '
Q)
m 50— o .
“Q. o . .
1 1 r 1 1 1 T 1 1

 

O 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 111: Integrated area of the peak attributed to camphene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

236

 

 

 

o
0
A30— '0.
=5 . °'°
32.5— 0
co 0
§ 2.0— .
a 5
”1.5-
2 ..
.54 _
81.0 o... .
a“ 05— '“5
[0 o . o
T 1 1 1 1 1 1 1 1

0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 112: Integrated area of the peak attributed to p-cymene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

 

6— o
A 0
=25— ',
<8
©o4_1 .
X
«:3— .
E I

2_ O
.x: '.
N
01-.
Du \.

'1- . . .
l l I l l l l I 7

 

0 200 400 600 800 i000 1200 1400 1600
Time after Consumption (min)

Figure B- 113: Integrated area of the peak attributed to a-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

237

 

 

35— '
’5 'u
C330- 0
v o O
o 25- °
3 .
x20—
315 '

‘ o
E .
%IO_ 0
33 5% \u
‘0
~ Q . .
1 1 1 1 1 1 1 ﬁ 1

 

0 200 400 600 800 1000 1200 1400 1600
Time aﬁer Consumption (min)

Figure B- 114: Integrated area of the peak attributed to ﬂ-pinene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

 

35-0’
=53.0— o
3
o 2.5-
2 o
x2.0‘
a:
215—J '.
o
x10~ 0'.
g 0
~ 0 o o
1 1 1 1 1 1 1 1 1

 

0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

Figure B- 115: Integrated area of the peak attributed to ,B-phellandrene in
chromatograms obtained from the breath of subject A on the ﬁrst day of Mediterranean
lemonade consumption.

238

(a.u.)
S I
1 1

6

1.0ﬁ Q

.o .O
0‘ 00
l l

g:
.5
_l
O

,0

Peak Area x10
0

’.

 

. o
1 1 1 1 1 1 1

0 200 400 600 800 1000 1200 1400 1600
Time after Consumption (min)

 

-4
—

Figure B- 116: Integrated area of the peak attributed to a-thujene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

 

 

 

0
A4—
: 0
cc
:3“ 0.
c O O
E O
cUZ- o
23.
< o
ﬁl-io .0 .
a '0
“no a. o . .

 

—1
_
=

1 1 1 1 1 1
0 200 400 600 800 1000 1200 1400 1600
Time aﬁer Consumption (min)

Figure B- 117: Integrated area of the peak attributed to ,B-myrcene in chromatograms
obtained from the breath of subject A on the ﬁrst day of Mediterranean lemonade
consumption.

239

Table B- I .' Fit Parameters for the terpenes and terpenoid compounds monitored in
breath. The data after lemonade consumption were ﬁt according to Equation 4-1.

 

 

 

 

 

 

SubjectA SubjectA SubjectA SubjectA SubjectA SubjectB SubjectC SubjectA
Original Original Original Original Original Original Original Concentrated
First Second Third Fourth Fifth First First First

a-limonene y0 544310 1021000 661590 491670 1089600 3790200 874680 424470
A1 21034000 16377000 10373000 11691000 11822000109160000 16273000 34604000

taul 81.564 108.4 74.521 85.105 83.484 12.349 31.422 114.03

tauka 0.2 0.2 0.2 0.2 0.2 0.2 0.2 5

4-terpineol y0 51083 80935 53062 50923 49537 163310 125720 65881
A 1 317930 517310 703930 552660 567080 634550 625420 601010

tau 1 23.896 9.2733 9.1991 9.9467 8.94281 13.349 15.902 18.076

tauka 0.000001 0.000001 0.000001 0.000001 0.000001 0.000001 113-07 0.001

a-terpineol y0 56696 87705 88101 85451 89343 170800 154040 77308
A1 252460 767850 1465000 816210 1013100 753710 1212800 542880

tau] 15.268 6.6357 7.985 7.5243 5.6561 12.365 10.641 16.807

tauka 0.01 0.000001 0.000001 0.000001 1 0.01 1E—07 0.2

terpineolene y0 64617 69237 68144 58483 64372 216010 167190 50995
A 1 209800 204870 169450 16317 115580 1652000 536590 356460

tau] 78.108 91.896 38.416 31.751 40.248 15.107 34.455 99.715

tauka 0.001 0.001 0.001 0.001 0.001 0.001 0.001 5

a-terpinene Yo 36958 5 1700 46303 44039 44389 1 10570 67407 35795
A1 131740 99934 68834 14414 41738 1009800 154460 325180

tau] 68.593 103.75 60.344 18.312 50.149 10.478 27.205 106.02

tauka 0.1 0.1 0.1 0.1 0.1 0.1 0.1 5

y-terpinene yo 141540 193150 139900 129970 216150 507560 176230 115050
A] 3274900 2588000 1674100 1945500 2033800 20981000 285420 5411600

tau] 63.624 78.745 54.905 60.109 55.54 10.638 24.82 99.381

tauka 0.001 0.001 0.001 0.001 0.001 0.001 0.001 5

 

 

 

 

 

 

 

 

 

 

240

Table B- I (cont ’d).

 

 

 

 

 

 

 

 

SubjectA SubjectA SubjectA SubjectA SubjectA SubjectB SubjectC SubjectA
Original Original Original Original Original Original Original Concentrated
First Second Third Fourth Fifth First First First

camphene Yo 17514l 484.92 189.55 -53.236 -459.81 38904 34352 20197
A1 47217 6917.7 3083.4 3642.6 3792.7 149140 44602 127360

tau] 110.64 143.19 206.24 185.91 349.4 11.6 36.913 102.43

tauka 1.5 1.5 1.5 1.5 1.5 1.5 1.5 20

p-cymene yo 82408 82563 68480 47475 -370770 261000 150230 199970
A 1 557820 408760 211550 348540 718883 1936800 589630 1785100

tau] 137.65 250.73 153.15 202.34 1032. 26.01 58.287 125.941

tauka l 1 1 1 1 1 1 20

a-pinene y0 93362 189450 137520 101760 103430 213250 148260 179430
A 1 867400 942450 409330 446750 377030 3934900 480570 3026700

tau] 98.043 128.93 110.21 137.0 171.21 8.8082 45.430 84.163

tauka 2 2 2 2 2 20

B—pinene yo 252950 388570 310790 24672 336370 443750 326810 107140
A1 4671600 3229900 1156400 15334 1178800 21566000 1442200 8452400

tau] 68.442 136.25 85.069 139.5 161.39 9.1774 35.176 132.87

tauka 1.5 1.5 1.5 1. 1.5 1.5 1.5 5

B—phellandrene Yo 35710 26337 4641 259981 62417 69373 48093 49434‘
A] 24723 29399 4273.9 13625 41760 89102 1383.9 102570

tau] 38.561 207.12 479.66 144.13 1107.3 1.6004 9.1121 62.239

tauka 0.1 0.1 0.1 0.1 0.1 0.1 0.1 10

a-thujene y0 25208' 3141 30556 29669 30051 4281 35280 33511
A 1 1.1 1 01 60926 21087 156481 13470 40425 23081 268620

tau] 61.854 131.9 82.485 99.071 54.184 8.4933 17.735 106.93

tanka 1 1 1 1 1 1 l 5

B—myrcene y0 46828 63287 59473 49877 63215 1 15540 57665 54613
A] 572360 400240 215380 196240 255320 3555100 331510 1098000

tau] 69.242 120.56 66.424 81.517 68.671 9.3712 29.104 88.270

tanka 0.1 0.1 0.1 0.1 0.1 0.1 0.1 10

 

 

 

 

 

 

 

 

 

 

241

”111111111111111111131111111111111

62 6042