United States
Environmental Protection
Agency
Atmospheric Research and Exposure
Assessment Laboratory
Research Triangle Park NIC 27711
Research and Development
EPA/600/S3-90/049 Sept. 1990
©EPA Project Summary
Measurements of Exhaled Breath
Using a New Portable Sampling
Method
Lance A. Wallace and William C. Nelson
This report documents the develop-
ment and demonstration of a new breath
sampling method for volatile organic
compounds (VOCs). The project, part of
EPA's Total Exposure Assessment
Methodology (TEAM) Program, was
aimed at improving the existing field
method for sampling exhaled breath. The
new method was tested on four subjects
exposed to elevated chemical levels in six
microenvironments (hardware stores,
swimming pools, garages, etc.).
Repeated breath measurements before
and after exposure provided data on the
uptake and excretion of 20 VOCs.
This Project Summary was developed
by EPA's Atmospheric Research and Ex-
posure Assessment Laboratory, Re-
search Triangle Park, NC, to announce
key findings of the research project that
is fully documented in a separate report
of the same title (see Project Report or-
dering information at back).
Introduction
EPA's TEAM Study (1) of human exposure
to VOCs has always included measure-
ments of exhaled breath to determine time-
integrated dose and to confirm that
exposure measurements have included all
important routes of exposure. The original.
method used in all TEAM Studies between
1979 and 1987 employed a van-mounted
spirometer (2). The principle of the method
was to collect about 40 L of expired air in a
bag and then pull the air across two Tenax
cartridges for later analysis by gas
chromatography-mass spectrometry (GC-
MS). The subject breathed clean air sup-
plied by a cylinder in the van. Normal
breathing was employed, and all the exhaled
air was collected; thus the exhaled air was a
mixture of alveolar air and air from the upper
airways, or "deadspace" area.
The above method was employed to col-
lect breath samples from about 800 people
in eight cities during the 1979-1987 period.
The method had several important ad-
vantages, including (1) transportability-the
van drove to people's homes to reduce the
burden on them of supplying a sample; and
(2) simplicity of breathing technique-un-
trained persons from 5 to 85 had little difficul-
ty giving samples.
However, three important disadvantages
of the method were also identified:
(1) Although the time to provide 40 L of
breath was only about 5-6 minutes, the total
cycle time (time to complete one breath
sample and be ready to collect another) was
about 20 minutes. In situations where
repeated breath measurements are
desirable, this represents an irreducible limit
on frequency of collection.
(2) The breathing technique resulted in a
mixture of alveolar air with the clean inhaled
air that failed to penetrate the alveolar region»
("deadspace air"). Thus the actual alveolar
concentration would be higher than the
measured 'concentration by an unknown
factor, depending on the relative proportion
of deadspace air for each subject.
(3) The amount of bulky equipment re-
quired (cylinder of clean air, 40-L bags,
pumps) reduced the ability to collect
samples at any time and place desired.
Therefore EPA decided to develop a new
method for sampling exhaled breath. The
performance.goals of the new method were
as follows:
(1) Reduce the sampling time to 1-2
minutes and the cycle time to 5 minutes.
G§> Printed on Recycled Paper
-------
(2) Collect alveolar air predominantly.
(3) Be portable with no power require-
ments.
Following development and laboratory
testing of the method, it was employed in
field studies of exposure to common
mlcroenvlronments with suspected high
concentrations of certain toxic or car-
cinogenic VOCs. Following exposure,
volunteers supplied repeated breath
samples over a period of 2-4 hours. The
goals of the study were to measure con-
centrations of a number of VOCs in common
mlcroenvlronments, and determine the up-
take of these VOCs in the body and their
subsequent excretion.
Results
Development of the Method
The new method includes the following
fundamental components:
(1) A charcoal cartridge to clean ambient
air as a source of inhaled air. This eliminates
the need to provide a separate source of
clean air.
(2) A critical-orifice canister to collect a
known volume of expired air in 1-2 minutes.
This eliminates the need for a pump and the
associated power requirements.
(3) A long narrow tube to isolate alveolar
air from deadspace air for the majority of the
breath cycle time.
The complete device is mounted in an
aluminum suitcase and weighs 10.5 kg, in-
cluding two 1.8-L evacuated canisters (Fig-
ure 1). It can be carried by one person, set
up in less than five minutes, and collect
alveolar breath samples in less than two
minutes. All components are attached to an
aluminum plate mounted on pivoting slides.
These slides allow the entire mounting plate
to slide out horizontally and elevate vertically
to six different mounting heights to accom-
modate children and adults of all heights.
Flgun 1. Portable splrometer for the collection of VOCs in alveolar breath.
-------
The procedure for collecting a breath
sample is as follows: following adjustment of
the system to the correct height, the par-
ticipant dons pinch-type nose clamps to
prevent nose-breathing, and places his/her
mouth tightly over a previously sterilized
mouthpiece unit and breathes as normally
as possible. (Due to resistance from the
valves and the breath containment tube,
breathing is probably slower and deeper
than normal-this should enhance the
proportion of alveolar air collected.) As the
participant inhales, air is pulled through two
charcoal-filled respirator cartridges
mounted in parallel. Four full breaths are
taken before sample collection begins in
order to clear the spirometer and the partic-
pant airways of ambient air trace chemicals.
The inhaled air passes through the unidirec-
tional inhale valve and into the lungs; ex-
haled air passes through the unidirectional
exhale valve and the sampling port into the
breath containment tube. Following com-
pletion of the fourth breath, the canister valve
is opened and pressure-driven flow com-
mences through the fixed needle orifice.
The orifice is designed to collect 1.4 L of air
in about 1.5 minutes, at which time the
canister valve is closed.
The breath containment tube was
designed to collect over 95% alveolar air. As
the participant exhales, the deadspace air
passes rapidly by the sampling port, into the
tube and out the other end. At the comple-
tion of an exhalation, all the air remaining in
the breath containment tube is alveolar. This
alveolar air is then sampled by the canister
during the remainder of the breath cycle
(resting time plus the inhalation portion of
the next breath). Although the deadspace
air is briefly sampled by the canister during
the fraction of a second that it passes by the
sampling port, the actual volume sampled is
only about 1 -2% of the volume of alveolar air
sampled during the remainder of the breath-
ing cycle.
The respirator cartridges (Survivaire®)
were tested to determine background levels
and breakthrough volumes of 11 VOCs
selected to provide a variety of different clas-
ses and volatilities. Background levels of all
chemicals with the exception of
tetrachloroethylene and methylene chloride
were below the quantifiable limit. Since
these chemicals were at high levels in
laboratory air, and since later spirometer
blanks indicated that the filter cartridges
themselves were not contaminated, the true
source appears to have been residual
laboratory air in the system. Estimated
quantifiable limits in a 150-ml sample were
less than one /*g/m3 for 10 VOCs, and
ranged from 1.6-4.4 ^g/m3 for seven addi-
tional compounds. The breakthrough
volume for dichloromethane was deter-
mined to be 320 L per cartridge. The
cartridges were also tested to determine the
effects of storage time and intermittent reuse
of the type expected for field sampling con-
ditions. Intermittent use over a period of five
days resulted in identical breakthrough
volume of 320 L. Since two cartridges are
used in parallel, they would not need to be
replaced until 640 L of air had passed
through them. Assuming 20 L inhaled
during two minutes, this corresponds to
about 30 breath samples.
The system was also tested to determine
whether VOCs are adsorbed on any interior
surfaces. Only two of 26 chemicals showed
evidence of adsorption: n-dodecane and
4-phenylcyclohexene. "Carryover experi-
ments (collection of a low-concentration
mixture following a high-concentration one)
identified the same two chemicals as show-
ing evidence of adsorption followed by
release during later use of the system.
Therefore it is expected that compounds
more volatile than n-dodecane can be col-
lected successfully by the system in normal
use. To collect the less volatile compounds
successfully, a small amount of "condition-
ing" of the system (about two minutes of
breathing through the device by the par-
ticipant before collection of the sample) may
improve recoveries.
Breath Measurements Using the
New Method
Previous TEAM Studies have indicated
that consumer products and personal ac-
tivities, particularly in enclosed spaces
(microenvironments) provide the major
sources of exposure to many VOCs (3).
Only limited data are available on the
thousands of consumer products and
hundreds of different microenvironments
where exposure can occur. Even fewer data
are available showing the uptake and excre-
tion of VOCs during and after exposure in
these microenvironments. Therefore a
study was planned to screen a number of
possible high-exposure microenviron-
ments, consumer products and personal ac-
tivities. A small number of these would then
be selected for study of exposure, uptake,
and excretion of a number of target com-
pounds for several volunteer subjects.
A total of 24 target chemicals were
selected for study (Table 1). The chemicals
were selected on the basis of their toxic,
mutagenic, or carcinogenic properties; high
Tab/e 1. Target Chemicals for Screening and Breath Exposure Study Samples
Compound Canister Tenax
Vinyl chloride
Isopentane
Vinylidene chloride
n-Pentane
Dichloromethane
2-Methylpentane
Chloroform
1,1,1-Trichloroethane
Carbon tetrachloride
Benzene
Trichloroethylene
n-Octane
Toluene
n-Nonane
Tetrachloroethylene
V
V
V
V
V
V
V
V
V
V
V
V
V
Ethylbenzene
p-Xylene (or m-;
Styrene
o-Xylene
a-Pinene
n-Deoane
Limonene
p-Dichlorobenzene
n-Dodecane
V
•
-------
Tab/o 2. Mlcroenvlronment Screening Sample Locations for Canister Air Sampling
Microenvlronment
Photocopier room
High volume photocopy/print center
Room painting (oil based paints)
Matal shop
Woodshop
Wood staining area
Home No. 1 with moth crystals
Home No. 2 with moth crystals
Office with one heavy smoker
Indoor swimming pool
Furniture stripping shop
Hardware store No. 1
Hardware store No. 2
Interior decorating store No. 1
Interior decorating store No. 2
Beauty school No. 1 ,
Beauty school No. 2
Laundromat
Bar/nightclub with smoking
Driving and smoking during rush hour
Outdoors at a truckstop
Auto and mower refueling
Inside a new pickup truck cab
Home garage, morning
Home garage, evening after driving in car
Commercial repair garage
Body and repair shop
Paint and body shop
Home with diapers soaking in bleach
Mass spectrometer laboratory facility
Laboratory recently re-roofed
Packaging facility with much styrofoam
Sample Collection
Duration
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
1h
<1 min
20 min
<1 min
<1 min
<1 min
<1 min
<1 min
<1 min
12 h
<1 min
<1 min
<1 min
Full Scan or MID a
GCIMS Analysis
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
FS
MID
MID
MID
MID
MID
MID
MID
MID
MID
MID
MID
MID
MID
MID
*FS** full scan, MID — multiple ion detection.
Table 3. Consumer Product Emission Samples Collected on Tenax Using a Dynamic Headspace Purge
Product Name
Test
Temperature
Headspace Volume
Analyzed
Alrwlck® Solid Room Deodorizer (lemon scent)
Wood Plus9 Polish (lemon scent)
Johnny Fresh9 Bathroom Bowl Cleaner (pine scent)
Old English9 Furniture Polish
Renuzit Roomate® Liquid Air Freshner
40°C
30°C
30°C
26°C
26°C
26°C
0.23 L
0.45 L
0.48 L
0.30 L
0.23 L
0.22 L
production volumes; or their prevalence in
homes and buildings.
Phase /; Screening
Microenvironments
A tola! of 32 microenvironments (Table 2)
and six consumer products (Table 3) were
selected for screening. Air samples were
collected in evacuated canisters in each
location and were analyzed by GC-MS.
Tenax cartridges were employed in three of
the microenvironments and also for
headspace analysis of the consumer
products, since sources of elevated
a-pinene, limonene, and para-dichloroben-
zene were being sought, and these target
chemicals are not sufficiently volatile to be
recovered efficiently from the canisters.
Concentrations of the target VOCs in each
microenvironment as measured by the
canisters are shown in Table 4. Many of the
microenvironments had elevated levels of
individual VOCs, often exceeding
100^g/m3. Of the 24 target VOCs, only
three (vinyl chloride, vinylidene chloride, and
carbon tetrachloride) were not found above
10 ^tg/m3 in any of the 32 microenviron-
ments.
Nine microenvironments were inves-
tigated for nontarget chemicals (Table 5).
Tenax results for the consumer products
and the three microenvironments are
provided in Tables 6 and 7.
-------
Table 4. Air Concentrations (ftgfm3) in Microenvironment Screening Canister Samples
Compound
Vinyl chloride
Isopentane
n-Pentane
Vinylidene chloride
2-Methylpentane
Dichloromethane
Chloroform
1, 1, 1-Trichloroethane
Carbon tetrachlorlde
Benzene
Trichloroethylene
Toluene
n-Octane
Tetrachloroethylene
Ethylbenzene
m,p-Xylene
n-Nonane
o-Xylene
Styrene
n-Decane
p-Dichlorobenzene
n-Dodecane
Compound
Vinyl chloride
Isopentane
n-Pentane
Vinylidene chloride
2-Methylpentane
Dichloromethane
Chloroform
1,1, 1-Trichloroethane
Carbon tetrachloride
Benzene
TricHloroelhylene
Toluene
n-Octane
Tetrachloroethylene
Ethylbenzene
m,p-Xylene
n-Nonane
o-Xylene
Styrene
n-Decane
p-Dichlorobenzene
n-Dodecane
Photocopier
Room
NO*
ND
ND
ND
2
20
7
Z
ND
3
35
8
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Home No. 2
with Moth
Crystals
ND
3
3
ND
3
77
ND
34
ND
2
ND
61
1
ND
47
180
5
11
ND
9
>540
3
Photocopy &
Print Center
ND
ND
180
ND
2
10
50
5
ND
6
ND
9
ND
ND
1
5
2
4
ND
' ND
ND
ND
Office
with One
Smoker
ND
ND
66
ND
ND
39
36
7
ND
9
ND
21
ND
ND
1
7
ND
ND
ND
ND
ND
NC
Oil-Based
Painting
ND
ND
150
ND
ND
25
77
3
ND
ND
5
20
16
ND
24
88
230 .
39
ND
1200
ND
46
Indoor
Swimming
Pool
ND
24
15
ND
7
ND
240
2
ND
6
ND
7
1
ND
3
10
2
4
ND
4
18
ND
Metal
Shop
ND
ND
62
4
12
23
36
21000
ND
ND
8
130
27
1200
4
11
26
4
ND
63
ND
NCb
Furniture
Stripping
Shop
ND
10
6
3
26
7100
2
280
ND
4
120
2500
29
23
120
430
61
160
68
180
ND
35
Wood
Shop
ND
ND
ND
ND
ND
5
ND
140
ND
ND
15
120
53
100
90
200
8200
75
ND
1500
ND
NC
Hardware
Store
No.1
ND
29
16
2
41 .
900
ND
210
ND
9
ND
650
80
27
590
1700
290
110
38
570
39
57
Wood
Staining
ND
ND
1100
ND
58
2
ND
18
ND
10
5
2700
350
2
11
30
340
11
2
810
ND
NC
Hardware
Store
No. 2
ND
630
180
ND
120
100
1
46
ND
34
6
250
50
6
17
64
200
23
7
390
ND
25
Home No. 1
with Moth
Crystals
ND
56
28
ND
1
3
14
ND
ND
8
ND
26
ND
ND
7
13
ND
9
ND
ND
22
NC
-------
TtW« 4 (Continued)
Compound
Vinyl chloride
Isopentane
n-Pentane
Vfnyltdone chloride
2-Mothylpontane
DIchloromothane
Chloroform
1,1,1-Trichloroethane
Carbon tetrachloride
Benzene
Trichloroethytone
Toluene
n-Ocfana
Telrachloroethylene
Ethylbenzene
m.p-Xy/ane
n-Wonano
o-Xyfens
Styrono
n-Docario
p-D!chlorobenzene
r\-Dodocane
Compound
Vinyl chloride
Isopontana
n-Pontane
Vlnylldene chloride
2-Mothylpentane
Dlchloromethane
Chloroform
1,1,1-Trichloroethane
Carbon tetrachloride
Benzene
Trichtoroethylene
Toluene
n-Octena
Totrachloroethylene
Ethylbenzene
m.p-Xy/ena
n-Nonane
o-Xylone
Styrono
n-Docane
p-Dlchlorobanzene
n-Dodocane
Interior
Decorating
Store No. 1
ND
35
19
ND
12
240
ND
22
ND
9
ND
310
21
9
28
93
380
22
6
700
ND
NC
Truckstop
Outdoors
NC
80
32
ND
18
ND
ND
1
ND
8
ND
21
2
ND
5
16
2
6
ND
2
NC
NC
Interior
Decorating
Store No. 2
ND
9
5
ND
5
74
ND
12
ND
3
ND
37
53
ND
7
26
190
11
ND
590
90
ND
Auto&
Mower
Refueling
1
>1500
>3600
1
>1900
NC
NC
2
NC
>380
,ND
920
22
ND
110
340
20
120
13
10
NC
NC
Beauty
School
No.1
ND
21
10
ND
3
17
20
72
ND
15
12
240
2
ND
5
16
6
5
7 .
14
3
6
Inside
New Truck
Cab
ND
11
8
1
15
7
2
160
3
3
1
240
3
2
27
140
8
68
33 •
45
NC
NC
Beauty
School
No. 2
ND
43
11
ND
3
ND
6
8
ND
8
7
320
ND
4
2
8
3
2
ND
2
3
2
Home
Garage
AM
ND
250
120
ND
62
2
1
3
ND
30
ND
120
4
ND
26
93
4
32
6
5
NC
NC
Laundromat
ND
11
11
ND
3
6
36
2
ND
4
ND
6
, ND
17
1
3
ND
1
ND
ND
2
ND
Home
Garage
P.M.
ND
>370
222
ND
110
1
1
2
3
53
ND
160
7
ND
32
110
7
40
10
8
NC
NC
Bar/Club
with
Smokers
ND
74
27
ND
22
6
6
3
ND
20
ND
54
2
1
10
31
6
13
6
7
NC
NC
Commercial
Repair
Garage
ND
79
28
ND
19
4
ND
1
ND
10
ND
36
2
ND
7
22
3
10
ND
8
NC
NC
Rush Hour
Driving with
Smoking
ND
61
30
ND
24
5
2
5
NC
52
ND
120
3
6
23
„ 72
3
23
17
3
NC
NC
Body&
Repair
Shop
ND
88
28
1
23
4
ND
68
1
10
ND
520
7
16
56
>210
56
71
46
56
NC
NC
-------
Table 4. (Continued)
*
Compound
Vinyl chloride
Isopentane
n-Pentana
Vinylidene chloride
2-Methylpentene
Dichloromethane
Chloroform
1,1, 1-Trichloroethane
Carbon tetrachloride
Benzene
Trichloroethylene
Toluene
n-Octane
Tetrachloroethylene
Ethylbenzene
m,p-Xylene
n-Nonane
o-Xylene
Styrene
n-Decane
p-Dichlorobenzene
n-Dodecane
aND = not detected.
b NC = not calculated.
PaintS,
Body
Shop
ND
260
110
ND
61
7
1
3
ND
68
ND
2100
35
ND
67
220
36
.80
19
5
NC
NC
Home
Diapers
in Bleach
ND
20
16
ND
ND
41
94
ND
ND
4
ND
11
ND
ND
1
7
2
2
2
3
NC
NC
! Mass Spec.
Laboratory
Facility
NC
4
56
ND
9
450
49
13
1
3
5
180
5
1
1
4
1
1
1
1
NC
NC
Laboratory >
with New
Roof
ND
4
4
2
2
>1400
3
53
ND
2
1
3
ND
ND
1
2
2
1
ND
37
NC
NC
Packaging
Facilitywith
Styrofoam
ND
ND
ND
ND
ND
97
100
ND
ND
ND
ND
14
ND
ND
ND
14
ND
ND
1
ND
NC
NC
Phase II: Microenvironmental
Exposures and Breath Sampling
Based on the results from the screening
phase, six microenvironments (the furniture
stripping shop, the wood/metal shop, the
indoor swimming pool, hardware store #1,
a home garage with fuel handling and wood
staining, and a home with moth crystals and
wood polish selected from the consumer
product evaluation) were selected for the
exposure study. Four volunteers (Table 8)
were asked to spend 2-4 hours in one or
more of the selected microenvironments,
followed by 4 hours in a nearby location
where repeated breath samples could be
collected to follow the decline of the com-
pounds in the body.
A total of 10 separate exposure experi-
ments were carried out. In each case, per-
sonal air samples were collected for the
volunteers during the 12-hour period prior to
exposure to identify any unplanned ex-
posure. Air samples were also collected
during the exposure period, and in the loca-
tion where the breath samples were col-
lected. Breath samples were collected just
before the exposure period and for the four-
hour post-exposure period using the new
alveolar breath system and the older "whole-
breath" system. About 12 alveolar and 8-9
whole breath samples were collected during
the 4-hour post-exposure period, with a
higher frequency of collection (every 10
minutes for the alveolar samples, every 20
minutes for the whole breath samples)
during the early part of the period (when the
steepest decline in breath concentration was
expected).
Results were analyzed to determine
whether participants had had unplanned ex-
posures or unexplained elevated breath
concentrations prior to the exposure period.
For cases where both previous exposures
and breath concentrations were negligible,
the data from the post-exposure breath con-
centrations were analyzed to determine the
best-fit decay curve.
A simple pharmacokinetic model has pre-
viously been developed to describe the
breath data collected in the TEAM Study (4).
The main feature of the model is that it is
based on a multicompartment mass-
balance set of differential equations. The
first compartment is identified with the blood,
and additional compartments with succes-
sively "deeper" body systems, such as ves-
sel-rich tissues, muscle, and fat. The
number of compartments may be selected
according to the situation, and range typical-
ly from 1 to 3. One important feature of the
model is the existence of an intrinsic
"residence time" associated with each com-
partment. Knowledge of these residence
times is essential if breath measurements
are to be used to estimate previous ex-
posures. For the case of a negligible air
concentration, the alveolar concentration at
any time t following exposure is given by:
CALV - 2Aj6
where i indexes the compartment, the Aj are
determined by the initial concentrations in
each compartment, and the n are functions
of the intrinsic residence times associated
with each compartment.
If the residence times differ sharply be-
tween compartments, the model simplifies
to
-t/r,
CALV = 2Aie
where r\ is the residence time of the /th
compartment.
Previous chamber studies of washout
times following exposure have indicated that
the residence times associated with the
second and third compartments are on the
order of 1 -3 hours and 6-8 hours, respective-
ly, for a number of target VOCs. However,
no direct measurements of the residence
-------
Table 5. List of Nontarget Compounds Present in Selected Screening Canister Samples
Furniture Stripper
Interior Decorating
Store #1
Interior Decorating
Store #2
trichlorolluoromethane
trimothylsilanol
2-methylhexane
3-mothylhexane
acetic acid, 2-methylpropyl
ester (tent.) *
ethylcyclohexane (tent)
trimethylcyciohexane /so.b
3-mothytoctane
butanolc acid, 2-methylpropyt
ester (tent.)
docane, branched chain (tent)
4-mothylnonane
1,2,4-trimethylbonzene (tent.)
1-ethyt-2-methytbenzene (tent.)
trimethylbenzene /so.
2-methyl-1,3-butadiene
(tent.)
3-methylhexane (tent.)
trimethylcyciohexane /so.
ethylcyclohexane
trimethylcyciohexane
2-methyloctane
3-methyloctane
methylethylcyclohexane
propylcyclohexane
4-cyclohexadecane (tent)
2-methylheptane
3-methylheptane
dimethylcyclohexane /so.
dimethylcyclohexane
ethylcyclohexane (tent.)
trimethylcyclohexane /so.
2-methyloctane
3-methyloctane
ethylmethylcyclohexane
alkylcyclohexane (tent.)
alkane, branched
n-undecane
alkylcyclohexane
Hardware Store #7
Hardware Store #2
Beauty School #1
1,2-pentadlene (tent)
mothylcyclopantane
2-mothylhexane
3-mothylhexane
methylethylhexane (tent)
ketone (tent)
2-melhylheptane
3-methylheptane (tent)
acolic acid, 2-methylpropyl ester
aldehyde or ketone (tent)
1,3,5-trlmothyIcydohexane
2-methyloctane (tent.)
3-methyloctane
trans-1-ethyl-4-methylcyclohexane
(toni.)
methylnonane /so. (tent.)
(1-mothytethyl)-benzene
mathylnonane /so. (tent)
trlmathylbenzene /so. (tent)
1,3'Cyclopentadiene,5-(1-methyl-
propylidene) (tent)
ketone (tent)
trichlorofluoromethane
pentene /so.
alkane (tent)
alkane (tent.)
hexane
methylcyclohexane (tent.)
dimethylpentane (tent.)
methylhexane
branched alkane
branched alkane
alkyl cyclopentane (tent.)
branched alkane
alkyl cyclohexane
trimethylcyciohexane
methyloctane /so.
methyloctane /so.
branched alkene /so.
alkyl cyclohexane
alkane, branched
alkane, branched
alkane, branched
trichlorofluoromethane
pentadiene
2-ethylcyclobutanol (tent)
cyclic alkens or diene (tent)
n-undecane
Beauty School #2
Swimming Pool
Laundromat
With Perma Pure Dryer:
trichlorolluoromethane
pentadiene (tent)
decang, branched (tent)
docane, branched (tent)
undacane, branched (tent.)
alkane, branched
alkane, branched
Without Perma-Pure Dryer:
acetic add, anhydride (tent)
acetic add, butyl ester
dimethyl disulfide (tent.)
ester (tent)
* tent « Tentative GCJMS identification.
/so.»Isomer.
-------
Table 6. Qualitative Results of the GC/MS Analysis of Product Headspace Samples for cc-Pinene and LJmonene
Product Name
Pine-Sol® (19% pine oil)
Airwick9 Solid Room Deodorizer (lemon scent)
Wood Plus9 Polish (lemon scent)
Johnny Fresh9 Bathroom Bowl Cleaner (pine scent)
Old English9 Furniture Polish
Renuzit Roomate9 Liquid Air Freshner
a-Pinene
Present
Present
Present
Absent
Absent
Present
LJmonene
Present
Present
Present
Absent
Absent
Present
Potential
Interfering
Compounds
Many
Many
Few
_a
-
Many
" Not applicable since target compounds were absent
Table 7. Air Concentrations (fig/m3) in Microenvironmental Screening Tenax Samples
Compound
n-Octane
m,p-Xylene
Styrene
o-Xylene
n-Nonane
a-Pinene
p-Dichlorobenzene
n-Decane
LJmonene
n-Dodecane
Furniture Stripping
Shop
26
280
35
110
71
, 11
2
120
2
25
Hardware Store
No.1
29
620
15
230
110
24
3
100
5
1
Wood Shop
110
180
3
80
730
34
NDa
770
12
68
a NO = not detected.
Table 8. Participant Characteristics and Approximate Alveolar Spirometer Breathing Rates
Participant
Number
1
2
3
4
Sex
Male
Male
Male
Female
Age
35
31
32
25
Height
178cm
168cm
185 cm
180cm
Weight
82kg
57kg
79kg
61kg
Alveolar Spirometer
Breathing Rate
(breaths/min)
4.8 .
5.2
5.6
8.0
time associated with the first compartment
have been possible, due to the stow cycle
time (20 minutes) of the breath sampling
system then available. However, indirect
observations from these chamber studies,
and theoretical considerations using the
model above with observed chemical and
biological data, have suggested that the
residence time associated with the first com-
partment (the blood) is very short (on the
order of a few minutes for very volatile com-
pounds, and 25-30 minutes for relatively
nonvolatile compounds such as
tetrachloroethylene.
In view of the 4-hour sampling period for
the post-exposure breath measurements, it
is expected that only the first two (or possibly
three) compartments would be observable
in the decay curves. Therefore a simple
biexponential curve was fit to both the al-
veolar and whole breath data. As a check,
othertypes of curves were also fitto the data,
including single exponential, inverse,
logarithmic, and power functions. In all
cases, the biexponential curve provided the
best fit, with typical R2 values of 99.9% (Fig-
ure 2).
The half-lives calculated from the one- and
two-compartment models are displayed for
alveolar breath (Table 9) and whole breath
(Table 10). (The half-life is the product of the
residence time and the natural logarithm of
2.) The half-lives calculated for the first com-
partment in the two-compartment model are
2-20 minutes, in excellent agreement with
the values predicted earlier. The half-lives
for the second compartment are typically 1 -3
hours, again in good agreement with pre-
viously measured values.
No consistent correlation of measured
half-lives with exposure level was noted.
-------
1000
Mlcrogramsl
Cubic Motor '
too
10
Two Compartment Fit
Chemical = DCM Method = Alveolar
100
Minutes
200
•300
1000 H
Mtcrogramsl
Cubic Moter
100 .
10
Two Compartment Fit
Chemical = DCM Method = Whole Breath
100
200
Minutes
300
Figure 2. Ln-Hnoar display of decay data measured for dlchloromethane in alveolar (A) and whole (B) breath. The solid curve indicates a
curve defined by data showing an ideal fit to a two compartment model.
10
-------
Table 9. Calculated Half-Lives for Halogenated Hydrocarbons in Alveolar Breath
Compound
Exposure Cone.
(ftglm3)
Participant
One Compartment Model
One Compart.
tie (h)
Two Compartment Model
First
ti/2(h)
Second
Halogenated Hydrocarbons
Vinylidene chloride
Dichloromethane
Dichloromethane
Dichloromethane
Dichloromethane
Dichloromethane
Chloroform
1, 1, 1-Trichloroethane
1, 1, 1-Trichloroethane
1, 1, 1-Trichloroethane
1, 1, 1-Trichloroethane
1, 1, 1-Trichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
Tetrachloroethylene
Tetrachloroethylene
Tetrachloroethylene
56
5000
470
460
320
220
600
16000
340
200
200
200
140
77
280
190
150
1
1
2
1
3
4
2
1
2
2
• 1
4
1
1
2
3
4
2.97
0.60
0.40
1.07
1.65
1.86
': 0.72
0.88
1.33
4.30
0.99
3.39
1.00
0.65
2.42
0.85
2.06
0.12
0.13
0.10
0.78
0.08
0.17
0.08
0.10
0.13
0.00
0.17
0.17
0.08
0.20
0.18
0.11
CF»
11.60
1.80
1.07
1C'
1.14
2.07
1.58
1.90
2.60
3.81
3.18
6.08
1.80
1C
3.70
1.67
CF
81C = insufficient concentration change; model reflects insufficient change in concentration to calculate a half-life over this time interval.
CF = convergence failure; residuals failed to converge in 50 steps during iterative computation.
Calculated Half-Lives for Aromatic Hydrocarbons in Alveolar Breath
One Compartment Model
Exposure Cone.
Compound (ftglm3) Participant
Aromatic Hydrocarbons
One Compart
Two Compartment Model
First Second
Benzene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Ethylbenzene
Ethylbenzene
Ethylbenzene
Ethylbenzene
m,p-Xylene
m,p-Xylene
m,p-Xylene
m,p-Xylene
m,p-Xylene
m,p-Xylene
o-Xylene
o-Xylene
o-Xylene
430 a
5700
1200 *
640
640
510
460
320
280
2600 a
360
150
150
1700s
1600
560
560
230
160
700 *
440
790
.1
1
1
2
1
1
3
2
4
1
2
2
1
1
2
2
1
3
4
1
2
2
1.68
0.82
1.84
1.53
1.06
1.15
1.13
0.52
1.64
2.46
0.22
1.70
1.02
1.60
0.92
. 0.64
0.45
0.08
0.58
0.67
0.25
1.61
0.14
0.10
0.05
0.07
0.08
CF"
0.05
0.27
CF
0.03
0.08
0.04
0.08
CF
0.03
0.13
0.11
0.03
0.08
0.11
0.08
0.04
3.38
1.82
2.64
1.88
1.68
CF
4.05
3.23
CF
2.90
2.12
2.49
1.43
CF
1.10
2.42
2.15
2.16
2.12
2.94
1.17
9.95
a Exposure concentrations from the garage experiments are approximate.
bCF= convergence failure; residuals failed to converge in 50 steps during iterative computation.
11
-------
Tabl» 9 (continued) Calculated Half-Lives for Aliphatic Hydrocarbons in Alveolar Breath
One Compartment Model,
Exposure Cone. One Compart.
Compound (jiglm3) Participant ti/z(h)
Two Compartment Model
First
tiK (h)
Second
Aliphatic Hydrocarbons, Straight-Chain
nfentano
nfontane
n-Octane
n-Octano
n-Nonane
n-Atonana
n-Atonana
n-Nonana
n-Atonano
n-Atonana
n-Docano
n-Docana
n-Docana
n-Docana
n-Docana
n-Docana
n-Undocane
3400"
340
320"
39
12000*
210
210
180
130
110
14000"
360
360
260
210
170
5600s
Aliphatic Hydrocarbons, Branched-Chain
1
1
1
2
1
2
1
2
3
4
1
2
1
2
3
4
1
0.70
1.15
0.67
0.87
1.37
1.13
0.68
0.08
0.21
0.61
1.54
0.22
0.17
0.08
0.27
0.11
0.28
0.08
0.07
0.19
0.17
0.02
0.06
0.15
0.02
0.04
CF°
0.37
0.08
0.04
0.07
0.19
0.05
0.07
2.34
2.07
2.84
1C"
1.73
2.01
2.06
0.48
1.53
CF
16AO
1.39
1.06
1C
2.82
1C
1.36
Isopentane
2-Mothylpontano
2-Mothylhexane
3-Mothylhexane
3-Methylhexane
2-Mothyloctane
Ethylcyclohexane
10000s
2000s
340s
410"
39
5400s
900"
1
1
1
1
1
1
1
0.65
0.86
0.26
0.39
0.42
0.60
0.89
0.08
0.21
0.13
0.13
CF
0.28
0.19
2.33
3.18
3.16
2.54
CF
2.48
2.53
* Exposure concentrations from the garage experiments are approximate.
* 1C <* insufficient concentration change; model reflects insufficient change in concentration to calculate a half-life over this time interval.
0 CF = convergence failure; residuals failed to converge in 50 steps during iterative computation.
Also, no clear differences in measured half-
lives among the participants could be dis-
cerned. However, the data are quite limited
for this purpose.
Alveolar Values Compared to
Whole Breath
Since deadspace air volume is usually
considered to be about a third of the volume
of a normal breath, a simplistic assumption
would suggest that the alveolar concentra-
tions measured in this study would be about
50% higher than the whole breath concentra-
tions. However, a comparison of alveolar to
whole breath concentrations displayed the
anticipated behavior for only two or three of
16 VOCs (Table 11). The reasons forthis are
presently not clear; however, it is important
to determine the relation between alveolar
and whole breath samples in order to inter-
pret more meaningful the whole breath
measurements made in previous TEAM
Studies. The relative impact of factors such
as changed breathing patterns resulting
from the increased effort of forced expiration
orthe effect of time lapse between inhalation
of clean air and expiration need further inves-
tigation using controlled experimental con-
ditions and a rigorous quality assurance
program.
Summary and Conclusions
A new portable alveolar breath sampling
method suitable for environmental (ppb)
concentrations of many VOCs has been
developed and tested in the laboratory and
in field experiments. The system can be
carried and set up by a single technician,
requires no power, and collects 98-99% al-
veolar breath samples in 1-2 minutes from
untrained participants of any age above 5.
Organic compounds more volatile than
n-dodecane are recoverable with 95+% ef-
ficiency. Less volatile compounds can also
be measured using a slightly longer (2
minutes) conditioning period. The ability to
collect many samples in rapid succession
following exposure should greatly improve
our understanding of the uptake and decay
characteristics for a large number of VOCs.
Thirty-two common microenvironments
(homes, cars, garages, shops) were
screened for elevated concentrations of 24
VOCs. Many of these microenvironments
were found to have elevated concentrations
of multiple VOCs. Six microenvironments
were selected for exposure studies involving
four volunteers. Breath concentrations of 21
VOCs were sufficiently elevated to allow cal-
culating residence times in the blood and a
second compartment of the body. These
measurements are among the first that have
allowed direct observation of decay times for
blood concentrations resulting from ex-
posure to common environmental sources.
The measured decay times agree well with
theoretical predictions of a pharmacokinetic
model developed in conjunction with earlier
TEAM Study results.
12
-------
Table 10. Calculated Half-Lives for Aromatic and Halogenated Hydrocarbons in Whole Breath
One Compartment Model
~- Exposure Cone. One Compart
Compound (ftg/m3) Participant
Aromatic Hydrocarbons
Two Compartment Model
First
Second
Benzene
Toluene
Toluene
Toluene
Ethylbenzene
Ethylbenzene
m.p-Xy/ene
m.p-Xy/ene
m,p-Xy/ene
o-Xylene
o-Xylene
Halogenated Hydrocarbons
430 "
5700*
1200
320
2600 *
360
1700*
7600
240
700 *
440
1
1
1
2
1
2
1
2
1
1
2
1.30
1.03
1.24
1.03
1.25
0.95
1.06
1.05
0.55
1.21
0.93
0.46
0.32
0.31
CF1
0.48
0.25
0.52
0.18
'0.25
0.53
0.08
4.12
2.28
1.86
CF
2.45
2.17
4.98
1.60
2.52
6.02
1.48
Dichloromethane
Dichlorornethane
1, 1, 1-Trichloroethane
1,1,1-Trichloroethane
Tetrachloroethylene
p-Dichlorobenzene
5000
470
340
140
280
9400
1
2
2
1
2
2
0.95
0.55
1.33
1.10
2.13
1.57
0.40
0.33
1.38
0.52
1.68
0.53
7.98
5.40
IC°
1C
1C
21.00
" Exposure concentrations from the garage experiment are approximate.
bCF = convergence failure; residuals failed to converge in 50 steps during iterative computation.
01C = Insufficient concentration change; model reflects insufficient change in concentration to calculate a half-life over this time interval.
Calculated Half-LJveS'for Aliphatic and Cyclic Hydrocarbons in Whole Breath
One Compartment Model
Exposure Cone. One Compart.
Compound (ftg/m3) Participant ti/2(h)
Two Compartment Model
First
Second
ti/2(h)
Aliphatic Hydrocarbons, Straight-Chain
n-Pentane
n-Octane
n-Nonane
n-Decane
n-Undecane
3400 "
320 *
12000*
14000 *
5600 *
0.88
0.95
0.74
0.88
0.86
0.26
0.00
0.42
0.00
0.19
2.32
0.61
5.55
0.88
1.61
Aliphatic Hydrocarbons, Branched-Chain
Isopentane 10000 *
2-Methylpentane 2000 *
2-Methylhexane 340a
3-Methylhexane 400 *
2-Methyloctane 5400*
Cyclic Hydrocarbons
Ethylcyclohexane 900 *
a-Pinene 97
Limoneneb 160
1
2
2
0.89
1.02
0.87
0.88
0.96
0.99
0.79
0.16
0.24
0.26
0.29
0.31
0.57
0.40
0.13
0.33
2.85
2.25
3.47
3.57
5.20
6.64
1.60
58.70
* Exposure concentrations from the garage experiments are approximate.
Participant was exposed to limonene at the end of the period over which breath samples were provided.
13
-------
Table 11. Percent Difference Between Alveolar and Whole Breath Organic Compound Concentrations at 12, 60, and 185 Minutes
Post-Exposure a
72 M/n
Concentrations
Compound
Isopentane
n-Pentana
2-Mothylpentane
2-Mothylhexane
3-Mothylhaxane
Bonzono
Toluene
n-Octano
Ethylcyclohexane
3-Mothyloctane
Ethylbonzene
p-Xyleno
n-Nonane
o-Xylene
n-Docane
n-Undocane
Whole
180
99
50
44
27
24
40
8.6
20
120
38
23
230
8.5
160
36
Alveolar
140
87
66
38
25
17
35
5.4
24
131
36
27
154
8.6
143
61
60Min
Concentrations
Whole
71
42
25
11
11
13
22
4.3
9
62
21
11
90
4.6
71
15
Alveolar
64
43
29
13
9.2
10
23
2.4
11
52
26
18
90
4.1
83
23
185 Mm
Concentrations
Whole
38
19
12
8.8
6.3
6.2
8.8
1.5
4.9
25
8.5
4.7
37
1.9
28
6
Alveolar
42
26
15
7.5
5.7
6.5
15
1.5
5.5
16
17
7.3
50
2.6
55
11
(Alveolar-Whole)/
Alveolar
T=12
-28.6
-13.8
24.2
-15.8
-8.0
-41.2
-14.3
-59.3
16.7
8.4
-5.6
14.8
-49.4
1.2
-11.9
41.0
r=eo
-10.9
2.3
13.8
15.4
-19.6
-30.0
4.3
-79.2
18.2
-19.2
19.2
38.9
0.0
-12.2
14.5
34.8
T=185
9.5
26.9
20.0
-17.3
-10.5
4.6
41.3
0.0
10.9
-56.3
50.0
35.6
26.0
26.9
49.1
45.5
* Concentration (ftglm3) for whole breath at 12 and 185 minutes and alveolar breath at 60 minutes were as measured. The corresponding
data point In whole or alveolar breath was calculated using the equation of best fit from StatPlan as in Table 6-10 or 6-12.
Both alveolar and whole breath samples were collected into 6 L canisters and analyzed in the same manner.
Recommendations
The breath sampling method could be
further miniaturized. Additional VOCs com-
monly found in breath (e.g., ethane and
acetylene) should be tested for applicability
to this method. Extension of the method to
polar compounds would also be desirable.
Investigating more fully the factors affecting
the fraction of whole breath represented by
alveolar air will enable the whole breath
measures collected in previous TEAM
Studies to be better interpreted. Additional
study of VOC concentrations in other com-
mon mlcroenvlronments will help fill in our
knowledge of how and where people are
exposed to VOCs. Additional exposure and
breath decay experiments for the same and
additional VOCs will provide information
needed to estimate exposures from breath
measurements and vice versa. The effect of
physiological characteristics (body build,
exercise, breathing rate, etc.) on residence
time in the blood and other compartments
should be studied. The pharmacokinetic
model should be tested on a set of different
participants exposed to the same chemicals
to determine the usefulness of the model.
References
1. Wallace, L. A., (1987). TheTEAMStudy,
Volume I: Summary and Analysis. U.S.
EPA, Washington, DC 20460. EPA 600/6-
87/0023. NTIS PB 88-100060.
2. Pellizzari, E., Zweidinger, R., and Shel-
don, L (1985) "Breath Sampling" in Environ-
mental Carcinogens: Selected Methods of
Analysis, Vol. 7, L. Fishbein and I. O'Neill
(Eds). IARC Publication #68, World Health
Organization, Lyon, France, p. 399.
3. Wallace, Lance A., Pellizzari, E.D.,
Hartwell, T.D., Davis, V., Michael, L.C., and
Whitrnore, R.W. (1989) 'The Influence of Per-
sona! Activities on Exposure to Volatile Or-
ganic Compounds" Environ. Res. 50:37-55.
4. Gordon, S.M., Wallace, Lance A., Pel-
lizzari, E.D., and O'Neill, H.J., (1988).
"Breath Measurements in a Clean-Air Cham-
ber to Determine 'Wash-out' Times for
Volatile Organic Compounds at Normal En-
vironmental Concentrations," Atmos En-
viron. 22:2165-2170.
14
-------
-------
Lance A. Wallace and William C. Nelson, the EPA Project Officer, are with the
Atmospheric Research and Exposure Assessment Laboratory, U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711.
The complete report, entitled "Measurements of Exhaled Breath Using a New
Portable Sampling Method," (Order No. PB 90-250 135/AS; Cost: $39.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Research and Exposure Assessment Laboratory
U,S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA PERMIT NO. G-35
Official Business
Penally for Private Use $300
EPA/600/S3-90/049
------- |