&EPA
United States
Environmental Protection
Agency
Office of Monitoring Systems
and Quality Assurance
Washington DC 20460
June 1983
Research and Development
Direct Measurement of
Volatile Organic
Compounds in
Breathing-Zone Air,
Drinking Water, Breath,
Blood, and Urine
-------
OPIS-TBCHHICA1 IBFORHAttOH CERTER
<2>lS
^ EPA-600/4-82-015
Direct Measurement of Volatile Organic Compounds
in Breathing-Zone Air, Drinking Water,
Breath, Blood, and Urine
Ruth Zweidinger, Mitch Erickson, S. Cooper,
Don Whittaker, and Edo Pellizzari
Analytical Sciences Division
Chemistry and Life Sciences Group
Research Triangle Institute
and
Lance Wallace
Office of Monitoring Systems and Quality Assurance
Office of Research & Development
US EPA
Headquarters and Chemical Libraries
EPA West BJdg Room 3340
Mailcode 3404T
1301 Constitution Ave NW
Washington DC 20004
202-566-0556
-------
DISCLAIMER
This study was designed to test methods of measuring individual
exposure; it was not designed and cannot be used to characterize geo-
graphical areas or populations beyond the actual study groups them-
selves. No epidemiological conclusions regarding health effects of
measured exposure levels can be drawn from this study.
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
ii
-------
FOREWORD
Physical, chemical, and biological measurements of environmental
quality are necessary to determine not only the extent of environmental
damage but also the effects of environmental protection programs. The
Office of Monitoring Systems and Quality Assurance has the responsibility
of developing new monitoring methods, evaluating and improving existing
methods, carrying out field monitoring programs, and assuring the quality
of the environmental data collected by the Agency.
The present study was a pilot effort to evaluate new methods for
measuring personal exposure to a number of toxic compounds in air, water,
breath, and blood. The methods for collecting air and breath samples
appear to be particularly effective, and are now being employed in
large-scale studies.
H. Matthew Bills
Acting Director
Office of Monitoring Systems
and Quality Assurance
iii
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge Dr. Andrew J. Johnson of Lamar
University and Dr. William McDonnell of the University of North Caro-
lina for their willing help in locating student volunteers. To the
volunteers themselves, who carried air monitors, collected water samples,
and gave blood, - breath, and urine samples, we are deeply indebted.
iv
-------
ABSTRACT
Methods for determining individual human exposure to volatile organic
compounds (VOC) encountered during normal daily activities were field-
tested on university student volunteers in two geographical areas. The
following equipment and analytical protocols were tested:
A personal air quality monitor employing the synthetic adsorbent
Tenax-GC® to collect organic vapors for later analysis by gas
chromatography/mass spectrometry (GC/MS).
0 A specially-designed spirometer for collecting samples of expired
human breath on duplicate Tenax-GC® cartridges for later GC/MS
analysis.
0 A purge and trap analytical protocol for determining VOC levels in
blood and urine.
Results included the following:
0 The personal monitor and spirometer proved feasible for collecting
abundant quantitative data on most of the 15 target organic vapors.
0 Air exposures to many VOC varied widely, sometimes over 3 orders
of magnitude, among students on the same campus that had been
monitored over the same time period and day.
0 A log-linear relationship between breathing-zone air exposures
and concentrations in exhaled breath was suggested for three
chemicals: tetrachloroethylene, 1,1,1-trichloroethane, and
vinylidene chloride.
0 The analytical protocols for blood and urine gave different re-
sults in different laboratories. The cause of this problem is
being investigated.
0 Air was the main route of exposure for all target compounds except
the two trihalomethanes (chloroform and bromodichloromethane),
which were transmitted mainly through water.
0 Estimated total daily intake through air and water of the target
organics ranged from 0.3 to 12.6 mg, with 1,1,1-trichloroethane
at the highest concentrations in both geographic areas.
-------
CONTENTS
Foreword ill
Acknowledgments • iv
Abstract v
List of Tables vii
List of Figures ix
1. Conclusions • 1
2. Recommendations 2
3. Introduction 3
4. Program Objectives 6
5. Sampling and Analysis 10
6. Results 30
7. Discussion 53
8. References 82
Appendix: Data Collection Instruments Used at Lamar University.... 83
vi
-------
TABLES
Number Title
1 Chemicals Sought in Study 7
2 Samples Collected at Lamar University 11
3 Sampling Protocol for Personal Air
Samples—Lamar University 12
4 Meteorological Conditions at Lamar University 13
5 Samples Collected at University of North
Carolina at Chapel Hill 18
6 Sampling Protocol for Personal Air Samples—
University of North Carolina 19
7 Blanks and Controls—Lamar University 27
8 Blanks and Controls—University of North Carolina 28
9 Percent Recovery of Selected Test Substances in Air
and Breath Control Samples (Lamar University).... 31
10 Percent Recovery of Selected Test Substances in Air
and Breath Control Samples (University of
North Carolina) 32
11 Quantities of Target Compounds Measured in Tap
Water Blanks and Controls—Lamar University
(ng/mL) 33
12 Quantities of Target Compounds Measured in Tap
Water Blanks and Controls for Chapel Hill (ng/mL) 34
13 Quantities of Target Compounds Measured in Urine
Blanks and Controls—Lamar University (ng/mL) ... 35
14 Quantities of Target Compounds Measured in Urine
Blanks and Controls—University of North Carolina
(ng/mL) 36
15 Quantities of Target Compounds Measured in Blood
Plasma Blanks and Controls—Lamar University
(ng/mL) 38
16 Quantities of Target Compounds Measured in Blood
Plasma Blanks and Controls for Chapel Hill
(ng/mL) 39
17 Results of Water Blind Study 40
18 Results of Blood Blind Study 41
19 Characteristics of Student Volunteers: Lamar Univer-
sity 42
20 Characteristics of Student Volunteers: University of
North Carolina 43
21 Estimated Levels of Selected Vapor-Phase Organics
in Breathing-Zone Air of Lamar University Students
(ug/m3) 45
22 Estimated Levels of Selected Vapor Phase Organics
in Breathing-Zone Air of University of North
Carolina Students (ug/m3) 46
23 Estimated Levels of Selected Vapor Phase Organics
in Exhaled Breath of Lamar University Students .. 48
vii
-------
TABLES (con't)
Number Title Page
24 Estimated Levels, of Selected Vapor Phase Organics
in Exhaled Breath—University of North Carolina
Students 49
25 Quantities of Target Compounds Found in Tap Water
(ng/mL)—Lamar University 51
26 Quantities of Target Compounds Pound in Tap Water
(ng/mL)—Chapel Hill 52
27 Summary Statistics for Estimated Levels of
Selected Vapor-Phase Organics—Lamar University ... 54
28 ' Summary Statistics for Estimated Levels of
Selected Vapor-Phase Organics—University of
North Carolina 56
29 Percent of Individual Air Exposures Supplied by
Selected Vapor-Phase Organics—Both Groups 59
30 Spearman Correlation Coefficients for Selected
Vapor-Phase Organics in Breathing-Zone Air
and Exhaled Breath—Lamar University 60
31 Spearman Correlation Coefficients for Selected
Vapor-Phase Organics in Breathing-Zone Air
and Exhaled Breath—University of North Carolina . . 62
32 Significant Spearman Correlation Coefficients for
Selected Vapor-Phase Organics in Breathing-Zone
Air and Exhaled Breath of 17 Students at Lamar
University and University of North Carolina 63
33 Comparison of Breath-Air Regressions Using
Different Conventions for Assigning Values to
"Trace" and "Non-Detectable" Categories 65
34 Correlations of Toxics in Air & Breath (Lamar
University) 70
35 Correlations of Toxics in Air & Breath (University
of North Carolina 71
36 Correlations of Toxics in Air & Breath (Both
colleges combined 72
37 Comparison of Spearman and Pearson Correlation
Coefficients of Air and Breath Values—Both Groups 73
38 Breath/Air Ratios for Selected Vapor-Phase
Organics—Both Groups 74
39 Summary Statistics for Estimated Levels of Selected
Vapor-Phase Organics—Lamar University Student
Study 76
40 Summary Statistics for Estimated Levels of Selected
Vapor-Phase Organics—University of North Carolina 77
41 Estimated Daily Intake of Three Selected Compounds
from Water Compared to Air—Both Groups (mg) 79
42 Estimated Daily Intake of 10 Volatile Organic
Compounds Through Air and Water—Both Groups (mg) • 80
viii
-------
FIGURES
Number
1 Map of Beaumont, Texas, including Lamar University
area 8
2 Schematic of Spirometer for Collection of Breath
Samples 16
3 Headspace Purge and Trap Apparatus for Milk, Urine, and
Blood 24
4 Purge and Trap Apparatus for Water 25
5 Geometric Mean Concentrations of Seven Volatile Organics
in Air and Exhaled Breath of Two Student Groups .... 47
6 ; Tetrachloroethylene in Exhaled Breath Compared to Mean
Breathing-Level Concentrations 66
7 1,1,1-Trichloroethane in Exhaled Breath Compared to Mean
Breathing-Level Concentrations 67
8 Vinylidene Chloride in Exhaled Breath Compared to Mean
Breathing-Level Concentrations 68
ix
-------
SECTION 1
CONCLUSIONS
Sampling and analytical methods for measuring volatile organic com-
pounds in breathing-zone air, drinking water, and exhaled breath appear
to be suitable for use in field studies' of the general population.
However, more work needs to be done on the analj'tical methods for blood
and urine before they are acceptable for field use.
Five vapor-phase organic compounds were ubiquitous, occurring in 100%
of both the air and breath samples in Lamar University and the Univer-
sity of North Carolina. These organics were benzene, chloroform,
tetrachloroethylene, 1,1,1-trichloroethane, and dichlorobenzene isomer.
Three of the five ubiquitous compounds (tetrachloroethylene, vinylidene
chloride, and 1,1,1-trichloroethane) displayed an approximate log-linear
relationship between air and breath levels. (That is, the logarithm of
the breath concentration appears to vary directly with the logarithm
of the concentration in breathing-zone air).
Estimated total daily intake through air and water of the volatile
organics measured ranged from 0.3-12.6 mg, with a geometric mean value
of about 1.8 mg. Roughly one-third to one-half of this amount was
supplied by 1,1,1-trichloroethane in air.
Seven of the 10 most prevalent compounds measured by the personal air
quality monitors exhibited high variability (2-3 orders of magnitude)
despite being measured by students spending most of their time on a
single campus at the same time of day on two consecutive days. If
this preliminary observation is confirmed, it calls into question the
practice of assigning similar exposures to a "cohort" of people in a
given small geographic area.
Two organics were found in appreciable quantities in drinking water:
chloroform and bromodichloromethane. For both, the estimated daily
intake through water was about three times that through air. The sum
of the two trihalomethane concentrations in drinking water exceeded
100 parts per billion in all 38 samples.
The high breath-air ratios of chloroform noted for University of North
Carolina students compared to Lamar students is attributable to their
greater exposure to chloroform through drinking water. However, the
high breath-air ratios for two other compounds—tetrachloroethylene
and dichlorobenzene isomer—could not be explained by levels in drinking
water.
1
-------
SECTION 2
RECOMMENDATIONS
Further work is needed on the purge-and-trap analytical protocol for
blood and urine.
The observed great variability in personal exposure to organics needs
further study. If validated, it should be considered in planning
future environmental exposure studies.
Further studies relating dose to exposure are required to validate
or modify the log-linear relationship postulated in the present
study.
The anomalously high levels in breath of dichlorobenzene and tetrach-
loroethylene compared to measured exposure levels should be further
investigated.
-2-
-------
SECTION 3
INTRODUCTION
Few studies (1,2) have attempted to measure individual human expo-
sure or body burden of volatile organic compounds (VOC) at normal ambient
concentrations. Yet this information is crucial in arriving at decisions
of great economic consequence concerning the regulation of these substan-
ces. The present study is a pilot effort to evaluate currently available
methods required to determine individual human exposure and body burden
of a number of VOC including several suspected carcinogens.
METHODS FOR MEASURING EKPOSURE
Air. Until now, methods for determining individual exposure to most
organic vapors in air have been lacking. The main reasons have been
0 lack of an adsorbent capable of collecting organic vapors and
allowing quantitative recovery;
° lack of small quiet personal monitors capable of accompanying
persons on their daily routines (3).
Both of these deficiencies appear to be well along toward a solution.
For the past few years, a synthetic polymer called Tenax GC® has been
undergoing laboratory and field tests to determine its performance char-
acteristics (4-6). This polymer has an adsorption affinity for hundreds
of organic compounds. By pumping ambient air across a cartridge of Tenax
GC® and then inserting the cartridge into a thermal desorption until on a
gas chromatograph/mass spectrometer (GC/MS)/computerized system, several
-3-
-------
hundred compounds can be identified and several dozen quantified at the
part per billion (ppb) level or below.
By combining the Tenax GC® cartridge with a small quiet pump, a
personal monitor is created capable of measuring individual exposures
over an 8-hour period. The present study is the first field test of a
Tenax-based personal monitor to measure exposure to several compounds
simultaneously.
Other Routes. Besides air, drinking water, food and beverages may
be major routes of exposure to volatile organics. Analysis methods are
adequate for drinking water, but are rudimentary for food and beverages.
Therefore drinking water was included in this study, but food and bevera-
ges were not. (A more complete study of individual exposure through air,
drinking water, some food groups and beverages, and house dust, has been
recently completed by the U.S. Environmental Protection Agency) (7).
METHODS FOR CALCULATING BODY BURDEN
As with exposure, measurement methods for calculating body burden of
many VOC have only recently been developed (1,2,7-9). Among these methods,
two were selected for field testing in this study:
0 Breath analyses. A specially designed spirometer employing
Tenax GC® cartridges (1,2,7) was used to analyze the exhaled
breath of volunteer subjects for the same organic vapors measu-
red by the personal monitors.
0 Blood and urine analyses. A recently developed analytical proto-
col for purgeable volatile organics (8) was used to measure con-
centrations of organic compounds in the blood and urine of the
volunteer subjects.
-------
Exposure/Body Burden Relationships
By combining measurements of exposure and body burden made for the
same individual, it was hoped that relationships between exposure and
body burden would be suggested for at least some compounds. The existence
of such relationships would be of immense practical value in estimating
levels of previous exposure in persons whose body burden had been deter-
mined, or, vice-versa, in estimating body burden of persons whose exposure
had been measured.
-------
SECTION 4
PROGRAM OBJECTIVES
The main objective of the study was to field-test the following
methods for measuring human exposure to VOC:
0 A personal air quality monitor to sample breathing-zone air;
0 A specially designed spirometer to sample exhaled breath;
0 Analytical protocols for measuring VOC in air, tap water, and
breath, blood and urine.
Depending on the success of the various methods, a second objective
was to compare levels of VOC in breathing-zone air and drinking water with
levels of the same compounds in the human body fluids tested. Table 1
lists all compounds studied.
'Selection of Sites.
Two areas were selected for study: a petrochemical manufacturing
center in Texas and a nonindustrial community in North Carolina. Volun-
teers from local universities were sought to reduce the variability
associated with age or occupation.
Beaumont, Texas was selected to represent the petrochemical area.
Lamar University is bordered on the north and south by oil storage tank
farms, and on the northwest .by the urban area of Beaumont. (Fig. 1).
Winds from the south cross over major refineries and petrochemical plants
before reaching the University. Students who live on campus would be
exposed 24 hours of the day. Both on- and off-campus residents were
studied.
—6—
-------
Table 1, CHEMICALS SOUGHT IN STUDY
1.
2.'
3.
4.
5.
6.
7.
8.
9.
10.
11.
12,
13.
14.
15.
16.
17.
18.
19.
20.
21.
Chemical
Chloroform
Chlorobenzene
Tetrachloroethylene
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
1,1, 2-Trichloroethane
Trichloroethylene
Carbon tetrachloride
Broroodichlorome thane
Vinylidene chloride
1 , 1-Dichloroethane
1 ,2-Dichloropropane
Dibromochlorome thane
Ethylene dibromide
m-dichlorobenzene
o-dichlorobenzene
Benzene
Bromoform
1,1, 2 ,2-Tetrachloroethane
Hexachloro-1 , 3-butadiene
1 ,2-Dichloroethylene
Air and
Breath3
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Drinking Water,
Blood, and Urine
X
X
X
X
X
X
X
X
X
Bloodb
(Lamar only)
(Dec. 79)
X
X
X
X
X
X
X
X
X
X
X
X
X
(March 80)
X
X
X
X
X
X
X
X
X
X
X
X
X
Analyzed at Research Triangle Institute.
Analyzed at University of Miami.
Chemicals could not be separated by gas chromatographic column employed.
-------
at
ttt!
LAMAR
UNIV.
V
Figure 1. Map of Beaumont, Texas, including Laraar University area.
Note the many oil storage tanks and oil fields north and
south of the University.
-------
Chapel Hill, N.C. was selected to represent the non-industrial area.
Chapel Hill has essentially no heavy industry and is located near no
industrial cities. .Both on- and off-campus residents of the University
of North Carolina were included in the study population.
Selection of Subjects.
At both universities, potential participants
1) had to be currently enrolled;
2) could not be enrolled in a course allowing direct contact with
organic chemicals (e.g., chemistry, anatomy, biology, or hospital
laboratories);
3) could not be employed in occupations allowing exposure to organic
chemicals (e.g., chemical plants, service stations, garages); and
4) could not engage in hobbies allowing potential exposure to organic
chemicals (e.g., painting, gardening, refinishing furniture or
developing photographs).
A questionnaire developed by the University of Miami Medical School
was administered to each student to determine factors that might be re-
lated to exposure, such as residence on or off campus, dietary habits,
hobbies, parents' occupations. The human rights committee at the Univer-
sity of Miami Medical School approved the questionnaire (reproduced in
Appendix A) for use in studies of human exposure to VOC.
In all, 17 students were selected: 11 at Lamar University (five
sampled on March 4, 1980 and six on the following day); and six at
UNC (three students sampled on June 10, 1980 and three on the following
day). Each participant signed a consent form and received a small
incentive when sampling was completed.
—9—
-------
SECTION 5
SAMPLING AND ANALYSIS
SAMPLE COLLECTION—Lamar University
Tap water, air, breath, blood, and urine samples were collected from
the 11 Lamar University Students (Table 2) according to specific sampling
procedures (1).
Each morning an air monitor and water sample bottles were distributed
to each participant with printed instructions on how to collect the tap
water samples and what to expect during the sampling. Each participant
carried the monitors for a 5-9 hour period (Tables 3 and 4). At the end
of the day, air monitors and tap water samples were turned in and breath,
blood, and urine samples were collected.
Water. The tap water samples were collected by each individual
immediately after he or she drank. Each participant was provided with
three 125 mL glass amber bottles with Teflon® lined caps. The partici-
pants were asked to fill at least one but not necessarily all three
within the course of the day. It was preferred that one of the samples
be from the subject's home or dormitory. Bottles were filled to overflow-
ing and tightly capped. They were turned in at the end of the day and
placed on ice. The samples were kept cold until analyzed.
Air« A personal sampler employing Tenax GC® polymer to collect
organic compounds was used to collect all air samples in the study. The
sampler consisted of an MSA Model C-200 pump and an attached Tenax cart-
-10-
-------
Table 2. SAMPLES COLLECTED AT LAMAR UNIVERSTY, BEAUMONT, TX,
Participant
No.
30001
30002
30003
30004
30005
30011
30012
30013
30014
30015
30016
1980
Date
3/4
3/4
3/4
3/4
3/4
3/5
3/5
3/5
3/5
3/5
3/5
Air
1
1
1
1
1
1
1
1
1
1
1
Breath3
1
1
1
1
1
1
1
1
1
1
1
Waterb
1
2
1
2
3
3
3
3
3
2
3
Blood
1
1
1
1
1
1
1
1
1
1
1
Urine
1
1
1
1
1
1
1
1
1
1
1
3Breath samples were collected in duplicate.
Each participant was asked to collect a water sample each time he or she
drank—hence the range of 1, 2 or 3 samples.
11
-------
Table 3- SAMPLING PROTOCOL FOR PERSONAL AIR SAMPLES - LAMAR UNIVERSITY
Participant
No.
30001
30002
30003
30004
30005
30011
£ 30012
30013
30014
30015
30016
Date
3/4/80
3/4/80
3/4/80
3/4/80
3/4/80
3/5/80
3/5/80
3/5/80
3/5/80
3/5/80
3/5/80
Temperature
26.1
26.1
25.5
26.1
26.1
25.3
26.6
26.6
26.6
26.6
26.1
Pump
No.
1
2
3
4
5
1
2
3
4
5
2D
Sampling
Start
0754
0818
0818
0839
0848
0746
0800
0815
0858
0904
0920
Starting
Sampling Flow
End (mL/min)
1637
1706
1459
1550
1527
1420
1544
1517
1632
1657
1440 49.2
Ending Average Volume
Flow Flow Sampled
(mL/min) (mL/min) (L)
25 . 1
24.4
19.4
24.6
21.1
14.9
•24.6
19.9
26.2
25.3
48.0 48.6 15.6
Pump Nos. 1-5 were MSA's, count rates were calibrated in mL/min, and total counts x rate gave volume
sampled. 2D = Dupont sampler.
-------
Table 4. METEOROLOGICAL CONDITIONS AT LAMAR UNIVERSITY:* MARCH 4-5, 1980
Time
(hrs)
6
7
8
9
10
11
12
31
14
15
16
17
18
Temperature
(3/4/80)
57
58
58
59
59
60
61
60
61
61
61
61
61
(3/5/80)
62
62
63
67
67
65
63
65
66
65
65
63
Relative Humidity (%)
(3/4/80)
81
78
87
90
90
(3/5/80)
93
87
68
55
61
Wind Direction
(3/5/80)
163
157
179
199
154
150
158
156
156
158
172
162
164
(3/5/80)
234
219
223
234
304
325
340
352
348
344
350
341
Wind Speed (mph)
(3/4/80)
6
6
5
7
6
8
9
9
10
7
6
6
6
(3/5/80)
4
4
4
5
6
10
10
11
11
11
10
7
Precipitation (inches)
(3/4/80)
-
-
-
T
T
T
.01
.02
T
.03
.16
.03
T
(3/5/80)
_
-
-
-
-
-
-
-
-
-
-
-
-
Ozone Levels: 3/4/80 .02 to .04 pprn
3/5/80 .00 to .06 ppm
Fog and Thunderstorm on March 4th; Fog on March 5th
*Source: Texas Air Control Board: Lamar University station in Beaumont
-------
ridge. Participants carried the sampler with them during a 5-9 hour
period while they attended classes, ate lunch, commuted, or carried out
other normal daily activities on the campuses of their universities.
The sampler pump flow rate was adjusted to about 0.05 L/min, provi-
ding a sampling volume of about 25 L of air. This volume was chosen to
avoid exceeding the breakthrough volumes for most of the compounds select-
ed for study. (Breakthrough volume is defined as that volume at which
50% of the compound is lost through the exit of the sampling cartridge).
These breakthrough volumes ranged from 15 to 2000 liters (at 70 °F) for
the amount (1.5g) of Tenax employed.
The sampling tubes consist of a glass tube 10 cm long with an inner
diameter of 1.5 cm. Eight cm of 35/60 mesh Tenax particles were placed
between glass wool plugs which provided support at both ends. The Tenax
is prepared in the following way: virgin Tenax (or material to be recycled)
, is extracted in a Soxhlet apparatus for a minimum of 18 hours with metha-
nol followed by n-pentane. Then it is dried for 3-5 hours in a vacuum
oven at 120° and at a pressure of 28 inches of water. It is then meshed
in a clean air room to provide a 35/60 particle size range. Cartridge
samplers are then filled with the Tenax and conditioned by passing helium
(first purified in a liquid-nitrogen-cooled cryogenic trap) at a rate
of 30 mL/min through the cartridges for 120 minutes at a temperature of
270°C. To avoid recontamination of the Tenax sorbent bed, the conditioned
cartridges are transferred to Kimax® culture tubes, immediately sealed
using Teflon© - lined caps, and cooled. The culture tubes were then
placed inside plain sealable paint cans.
-------
Breath. Breath samples were collected on Tenax GC cartridges via a
spirometer (Fig. 2). The valves in the Douglas mouthpiece were replaced
with Tedlar. A bubbler filled with distilled deionized water was placed
in-line with the air tank to humidify the air for the comfort of the
participants.
The subject was seated in a comfortable chair, and the mouthpiece
height adjusted to a convenient level. A long spring clamp was used to
seal the air flow from Bag A to the mouthpiece. To prevent room air
contamination, a plug was placed in the mouthpiece opening until the
test began. Air flow from the pure-air tank was started, and when the
50-L Bag A was about half full, the clamp and plug were removed, the
nose clipped, and the subject began to breathe on the apparatus. After
a minute or two, the Nutech sampler pump was started with the flow at
approximately 7 L/min. The flow was adjusted to approximate the individ-
ual subject's respiration rate. The subject was asked to breathe until
about 75 L of exhaled breath passed through Bag B. The subject was then
asked to stop, and the remainder of Bag B's contents were sampled. The
Tenax cartridges were removed and stored in culture tubes. Bag B was
then removed and flushed with helium to decontaminate it for future use.
A clean bag was replaced in position for the next subject. The mouth-
piece was sterilized by swabbing with alcohol after each use. The Tenax
cartridges were desiccated over CaS(>4 placed in the bottom of a culture
tube and covered with glass wool.
Blood. Blood was drawn in the afternoon by a qualified phlebotomise
at the Student Health Center. Four 7-mL vacutainers (Kimble - Terumo Kt
200SKA) containing an anti-coagulant were filled from an arm vein. The
-15-
-------
ULTRAPURE
AIR TANK
TEFLON
CONNECTORS
TEDLAR
BAG A
TENAX GC
CARTRIDGES
TEDLAR
BAG B
DOUGLAS
VALVE AND
MOUTHPIECE
Figure 2. Schematic of spirometer for collection of breath samples.
16
-------
containers were immediately chilled on ice, centrifuged and placed back
in the ice. When chilled, the plasma was drawn off using a Pasteur pipet
and transferred to several 4-mL glass vials fitted with Teflon® lined rub-
ber septum caps. 'Each vial contained 4 mL of plasma and was kept cold
until analyzed. One vial from each subject was packed on ice and shipped
overnight to the University of Miami. The remaining vials were returned
to Research Triangle Institute (RTI) to be analysed.
Urine. Urine samples were collected immediately after the blood
was drawn. Wide-mouth 240-ml amber bottles, sea-led with a Teflon®
lined cap were used. Samples were placed on ice and kept cold until
analyzed.
University of North-Carolina
During June, 1980, tap water, air, breath, blood and urine samples
were collected at UNC (Table 5). Participants were screened and selected
by EPA staff under the direction of Dr. William McDonnell of the Health
Effects Research Laboratory's Inhalation Toxicology unit on the University
of North Carolina campus. Samples were also collected in this building.
The criteria and methods of sampling remained the same as for the
Lamar University study in Beaumont, Texas. Six students participated as
subjects, and two more supplied blood and urine for blanks and controls.
(Sampling protocols for air are displayed in Table 6.)
ANALYSIS PROCEDURES
All samples were analyzed at Research Triangle Institute (RTI),
except for the blood samples, which were sent to the University of Miami.
Breath and Air. For analysis, the cartridges were placed in a preheat-
ed thermal desorption chamber and purged with helium gas (15-20 mL/min.)
-17-
-------
Table 5. SAMPLES COLLECTED AT U. NORTH CAROLINA: CHAPEL HILL, NC
Participant
No.
40001
40002
40003
40011
40012
40013
Date
6/10/80
6/10/80.
6/10/80
6/11/80
6/11/80
6/11/80
Air
1
1
1
1
1
1
Breath3
1
1
1
1
1
1
Water
2
2
2
3
2
2
Blood
1
1
1
1
1
1
Urine
1
1
1
1
1
1
o
Samples were collected in duplicate sets.
18
-------
Table 6. SAMPLING PROTOCOL FOR PERSONAL AIR SAMPLES - UNIVERSTY OF NORTH CAROLINA
Participant
No.
40001
40002
40003
40011
40012
40013
Temperature
Date (°C)
6/10/80 21.1
6/10/80
6/10/80
6/11/80 21.1
6/11/80
6/11/80
Relative Pump
Humidity No.
46% 1
o /a
2,4
3
58% 1
4
3
Sampling
Start
0747
0746
1002
0747
0734
0752
0742
Sampling
End
1535
0830
1601
1612
1444
1539
1539
Volume ,
Sampled Wind
(liters) Direction
27 . 0 WSW
19.1
23.7
24.3 NNE
23.7
22.4
Wind
Velocity
(knots)
10
9
Monitor dropped by subject, stopped functioning - second monitor employed.
MSA sampler employed, 0.56 ml/count established by calibration; total counts multiplied by rate gave
volume sampled.
-------
into a liquid nitrogen capillary trap (5). After the desorption was
completed, the temperature on the capillary loop was rapidly raised (faster
than 100°C/min) and the carrier gas introduced the vapors onto the high
resolution GC column. The glass capillary column was programmed from
ambient temperature to 240°C at 4°C/min and held at the upper limit for a
minumum of 10 minutes. After all the components eluted, the column was
cooled to ambient temperatures and the next sample was run.
The helium carrier gas was precisely controlled at 2.25 mL/min by a
Xylan mass flow controller. A jet separator connected the glass capillary
column to the mass spectrometer (a Varian Mat CH-7 or LKB 2091). The
mass spectrometer (resolution 1500-2000) was equipped with single ion
monitoring capability and interfaced with a Varian 620/L or PDF 11/04
computer, respectively.
The mass spectral data were processed by scanning the original spectra
ancl extracting the reconstructed ion chromatogram (RIG). The intensity
was plotted against the spectrum number. Identities of constituents in
the sample were established by comparing mass spectra obtained with those
in the National Bureau of Standards (NBS) reference libraries.
Concentrations in air or breath of the target compounds were determin-
ed by the following technique (4). An external standard (hexafluoro-ben-
zene or perfluorotoluene) was loaded at a known concentration onto each
sample cartridge before analysis. The absolute peak height or area of
the chemical to be measured was then compared to the peak height or area
of the standard, by selecting characteristic ions in the mass spectra.
These heights or areas are proportional to the number of moles of each
substance. The proportionality constant, called the relative molar
-20-
-------
response (RMR), is known or can be determined experimentally for any com-
pound (10). The mass of the chemical on the cartridge was then determined
from the following equation:
Mu = AUWUMS/ASWSRMR
where subscripts _u and _s_ stand for the chemical of interest and the
standard respectively, M= mass, W= molecular weight, RMR = the relative
molar, response of the unknown to the standard, and A = the response (area
or peak height) of the characteristic ion.
The mean concentration of the substance in the air near the subject's
breathing zone during his exposure period can then be determined by
dividing the mass per cartridge by the volume of air drawn across the
cartridge, provided that the sampling volume does not exceed the .break-
through volume (at sampling temperature) for the given compound. Even
if the breakthrough volume has been exceeded, the mean concentration
during the latter part of the exposure period (corresponding to the
time required to pump the breakthrough volume of air across the cartridge)
can be estimated by dividing the mass of the compound on the cartridge by
the breakthrough volume. The calculated mass can be corrected for
efficiency (i.e. thermal desorption plus storage losses), which varies
by compound and by the length of time between collection and analysis
(11). Thus the mean concentration can be calculated by C = M/V, where
V = the volume of air sampled or the breakthrough volume, whichever is
less, and M is the mass on the cartridge, corrected for recovery effi-
ciency.
Drinking Water, Blood Serum and Urine
Tap water, blood serum, and urine samples were analyzed by a purge/
-21-
-------
trap/desorb method based on that of Bellar and Lichtenberg (see ref. 7).
The GC conditions were as follows: carrier gas flow rate of 30 mL/min,
transfer line temperature 200°C, and inlet temperature 140°C. The column
temperature was programmed at 60°C for 2 min, then Increased at a rate of
lO°C/min to 157°C. Both a Hall Electrolytic Conductivity Detector and a
flame ionizatlon detector were operated simultaneously to detect both the
halogenated compounds and benzene. The column was 1.8m x 2mm 0.2% Carbo-
wax 1500 on 60/80 Carbopack C. Standards, blanks and controls were
interspersed throughout the analysis period. The standards were prepared
fresh daily and transferred to smaller containers that were stored in the
refrigerator until used.
Tap water. —A 5-mL water sample was transferred to a 5-mL purge
device (Bellar and Lichtenberg design) via a glass syringe. The water was
purged onto a Tenax GC trap at a flow/rate of 40 mL/min for 12 minutes.
The compounds were then desorbed onto the analytical column. The purge
device was rinsed several times with distilled water after each sample.
Urine. —The purge device was placed in a sand bath maintained at a
temperature of 115°C during the entire analysis. A 1 mL aliquot of 1%
aqueous antifoam was added to the 5 mL purge device and purged off-line
at a flow/rate of 20 mL/min for about 20 minutes. A 2 mL aliquot of
urine was then added to the purged antifoam and purged on-line for 15 min
at a flow rate of 20 mL/min. To prevent contamination of the Tenax trap
by steam, a small glass vapor trap was installed between the purging device
and the trap (12). After the purge/trap period was complete the compounds
were transferred to the analytical column. The purge device was cleaned
with several rinsings of distilled water between samples.
-22-
-------
Blood serum. —Foaming was a major problem encountered with the blood
serum samples. Even with the 1 mL of 1% antifoam solution added to the
serum, foam migrated into the vapor trap, necessitating dismantling and
cleaning between each sample. This proved excessively time consuming so
a larger purge device (25mL) was used and the amount of 1% antifoam was
increased to 1.5mL. The purge device was placed in a constant temperature
sand bath maintained at 115°C during the analysis. The antifoam solution
was purged off-line for approximately 20 minutes with the flow-rate of
20 mL/min. A 0.5 mL aliquot of blood serum was injected into the purge
device and purged onto the trap for 15 min at a flow rate of 20 mL/min.
After the purge/trap period was complete, the compounds were transferred
to the analytical column. The purge device was cleaned with several
rinsings of distilled water.
Broad Spectrum Analysis.
Blood serum, urine and tap water samples (two each) were selected to
be purged onto Tenax GC cartridges for broad spectrum analysis by GC/MS.
Samples showing the largest number and amount of compounds detected by
the purge/trap/desorb analysis were selected for the broad spectrum analy-
sis. The blood and urine samples were purged by means of a headspace purge
apparatus (Fig. 3). Four mL of blood serum and 25 mL of urine were
diluted to 50 mL with purged distilled water. The samples were stirred
while maintained at a temperature of 50°C. The samples purged for 90 min
at a flow rate of 25 mL/min.
The conditions for the tap water purge were the same, but the purge'
apparatus used was of the through-solution type (Figure 4). A 50 mL
sample of tap water was purged. The moisture trapped in the Tenax
-23-
-------
THERMOMETER
-20to150°C
THERMOMETER
ADAPTER
with O-ring
$10/18
TENAX CARTRIDGE
HELIUM
PURGE
HELIUM
INLET
TUBE
LIQUID LEVEL
100 ml ROUND
BOTTOM FLASK
MAGNETIC
STIRRING BAR
Figure 3. Headspace purge and trap apparatus for urine and blood.
24
-------
THERMOMETER
ERLENMEYERFLASK
125ml
FRITTED FLASK
TENAX CARTRIDGE
TEFLON ADAPTER
GLASS WOOL PLUG
HELIUM PURGE
7 mm O.D., 1 mm I.D.
10 mm O.D.
Figure 4. Purge and trap apparatus for water.
25
-------
cartridge was dried by means of CaSCty (previously heated to 400°C for 3
hrs.) added to the culture storage tubes. They were desiccated overnight
and then transferred to culture tubes without CaSO^ for storage in the
freezer until analysis.
Quality Control
The blanks and controls were prepared one day prior to the sampling
trips (Tables 7 and 8). Lab blanks and controls remained at freezer
temperatures (-20°C) in the laboratory during the field trip while the
field blanks and controls were transported to and from the field, alongside
the samples.
The air controls were spiked with the following seven compounds via
a permeation system: chloroform, 1,2-dichloroethane, 1,1,1-trichlorethane,
carbon tetrachloride, trichloroethylene, bromodichloromethane, and benzene.
The breath blanks were run on the spirometer with the mouthpiece
plugged and the intake air forced through the exhaled bag. The controls
were collected in the same manner, except that each cartridge was spiked
with about 500 ng of each of the compounds listed.
Tap water blanks were prepared from distilled water. The bottles
were filled to overflowing, capped with Teflon® lined caps, and refri-
gerated. Controls were spiked to give a concentration of 10 ng/mL of each
of the following nine compounds: chloroform, 1,2-dichloroethane, 1,1,1-
trichloroethane, carbon tetrachloride, trichloroethylene, tetrachloroethy-
lene, chlorobenzene, jo-dichlorobenzene, and bromodichloromethane. They
were filled, capped and stored in the same manner as the blanks.
The blood plasma blanks and controls were prepared using combined
plasma from four individuals. The plasma was collected as described in
-26-
-------
Table 7. BLANKS AND CONTROLS--LAMAJR UNIVERSITY
Sample
type
Air
Breath
Water
Blood
Urine
Lab
Blank
2
2
2
1
1
Lab
Control
1
lb
2
1
1
Field
Blank
2a
4
3
4
2
Field
Control
2
4
3
4
2
Lost due to mishandling in field.
Invalid due to improper preparation.
27
-------
Table 8. BLANKS AND CONTROLS—UNIVERSITY
OF NORTH CAROLINA
Sample
Matrix
Air
Breath
Water
Blood
Urine
Lab
Blank
2
2
2
2
Lab
Control
2a
2
2
Field
Blank
2
2
2
2
Field
Control
2a
2
2.
Controls are for both air and breath samples.
28
-------
the section on sampling. The plasma was stored in 4mL vials and capped
with Teflon® lined rubber septum caps. Controls were spiked at a con-
centration of 10 ng/mL with the same compounds as the water.
Urine blanks and controls were prepared from a composite of urine
from three people. The urine controls were also spiked with the above
compounds at a concentration of 10 ng/mL. Both blanks and controls were
stored in clean 120-mL bottles filled with 100 mL of urine, sealed with
Teflon® lined caps, and refrigerated (+4°C).
-29-
-------
Section 6
RESULTS
Quality Control Results
Air and Breath. Percent recoveries for the air and breath analyses
generally ranged between 95-140% (Tables 9 and 10). Coefficients of
variance decreased to less than 10% for 1,1,1-trichloroethane and trich-
loroethylene between the Laraar and UNC visits, but remained between
20-30% for the other chemicals.
Water. Percent recoveries for tap water control samples at Lamar
University varied widely, from 25% for tetrachloroethylene to >100% for
chloroform (Table 11). The similarity of field controls to laboratory
controls indicated that the problem did not lie in storage or transport
of the sample. Adjustments in the analytical protocol led to the greatly
improved percent recoveries observed at Chapel Hill three months later
(Table 12). Mean recoveries ranged between 92% and 118% for the seven
spiked compounds. The precision (i.e., relative standard deviation, or
coefficients of variation) ranged between 4 and 12% at Chapel Hill. (The
peaks for trichloroethylene and 1,1,2-trichloroethane overlap, so it was
not possible to determine whether one or both compounds were present, nor
to quantiate either compound).
Urine. As with the water samples, recovery efficiencies for urine
samples showed sharp improvement between the Lamar and UNC visits.
(Tables 13 and 14). From a range of 12-57% observed at Lamar University
-30-
-------
Table 9. PERCENT RECOVERY OF SELECTED TEST SUBSTANCES
(LAMAR UNIVERSITY STUDY)
IN AIR AND BREATH CONTROL SAMPLES
Air
Compound
Benzene
Chloroform
1,2-Dichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
Tetrachloroethylene
Bromodichlorome thane
Chlorobenzene
m-Dichlorobenzene
Observed (ng/cart.)
X ± S.D. (C.V.)
570
270
750
505
606
694
286
255
567
± 177
± 91
± 285
± 175
± 155
± 142
± 90
± 49
± 174
(31)
(33)
(38)
(35)
(26)
(21)
(32)
(19)
(31)
Actual
(ng/cart.)
528
200
581
368
605
536
227
219
490
Percent
108
135
129
137
100
129
126
116
116
± 33
± 45
± 49
± 47
± 26
± 26
± 40
± 22
± 35
Breath
Observed (ng/cart.)
X ± S.D. (C.V.)
643
219
707
501
615
572
249
218
499
± 265
± 74
± 243
± 115
± 140
± 171
± 68
± 90
± 124
(41)
(34)
(34)
(23)
(23)
(30)
(27)
(41)
(25)
Percent
122
109
122
136
102
107
110
99
102
± 50
± 37
± 41
± 31
± 23
± 32
± 30
± 41
± 25
-------
Table 10. PERCENT RECOVERY OF SELECTED TEST SUBSTANCES IN AIR AND BREATH CONTROL SAMPLES
(UNIVERSITY OF NORTH CAROLINA STUDY)
Air
Compound
Benzene
Chloroform
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
Tetrachloroethylene
Bromodichlorome thane
Carbon tetrachloride
Observed (ng)
X ± S.D. (C.V.)
450
208
548
600
889
852
292
508
± 77
± 59
± 236
± 31
± 56
± 276
± 98
± 39
(17)
(28)
(43)
(5)
(6)
(32)
(33)
(8)
Actual
(ng)
530
200
567
455
742
556
225
343
Breath
Observed (ng)
Percent X ± S.D. (C.V.)
85
104
97
132
119
153
130
148
± 14 501 ± 291 (58)
±29
± 42 644 ± 74 (11)
±7 456 ± 143 (31)
±7 828 ± 41 (5)
±50
±43
± 11 484 ± 145 (30)
Percent
94 ± 55
-
113 ± 13
100 ± 31
111 ± 6
.-
-
141 ± 42
-------
Table 11. QUANTITIES OF TARGET COMPOUNDS RECOVERED IN TAP WATER BLANKS
AND CONTROLS FOR LAMAR UNIVERSITY STUDIES (ng/mL)
Sample
Water Blank
(T0)
Field Blank
Lab Blank
Field Control
Lab Control
CFa DE MCF
1.2 ± 2
0.3 ±0.1
3.0 ± 4
9.4 ± 4 (94)c 8.7 ± 0.5 (87) 3.9 ± 2 (37)
11 ± 4 (109) 9.3 i I (93) 4.6 ± 0.7 (46)
CT BCM
0.5
-
7.6
5.5 (55)
3.4 ± 0.1 (34) 5.8 t 0.1 (58)
TCE PERC
-
-
-
NQ 2.4 ± 0.2 (26)
NQ 2.6 ± 0.1 (25)
CB
-
-
'
7.5 ± 0.2 (76)
7.5 ± 1 (75)
aCF = chloroform, DE = 1,2-dichloroethane, MCF = 1,1,1-trichloroethane, CT = carbon tetrachloride, BCM = bromodichloromethane,
TCE = trichloroethylene/or 1,1,2-trichloroethane, PERC = tetrachloroethylene, CB = chlorobenzene.
- not detected.
Mean ± S.D. (percent recovered).
-------
Table 12. QUANTITIES OF TARGET COMPOUNDS RECOVERED IN TAP WATER BLANKS
AND CONTROLS FOR UNC STUDIES (ng/mL)
Sample CFa DE
Field Blank Tb
Lab Blank 1.3+0.5
Field Control 11 ± 1.4 (98)c 10 + 0.9 (100)
Lab Control 9.8 ± .3 (88) 9.2 + 0.1 (92)
aSee Table 11 for codes.
bT = trace.
MCF BCH TCF PERC CB
_
.
12 ± 1 (118) 10 ± 1 (102) 11 + 0 (108) 12.5 ± ,2 (125) 10 ± 0.5 (102)
10.8 ± 0.1 (107) 9.5 ± 0 (95) 10.4 ± .2 (104) 12+1 (107) 10 + 0.5 (100)
Mean ± S.D. (percent recovered).
-------
OJ
Ul
Table 13. QUANTITIES OF TARGET COMPOUNDS MEASURED IN
URINE BLANKS AND CONTROLS--LAMAR UNIVERSITY (ng/mL)
Sample
Lab Control
Lab Blank
Field Blank-1
Field Blank-2
Field Control-1
Field Control-2
Chloroform
9.3 (93)
-
0.3
-
5.0 (47)a
4.7 (47)
1,2-Dichloro-
ethane
9.0 (90)
-
-
-
5.7 (57)
5.7 (57)
1,1,1-Trichloro-
ethylene
5.3 (53)
-
-
-
3.2 (32)
3.2 (32)
Carbon tetrachloride
and/or bromodi-
chloromethane
4.6 (46)
-
-
-
3.0 (30)
3.0 (30)
Trichloroethylene
and/or 1,1,2-
trichloroethane
2.9 (29)
-
.
-
1.2 (12)
1.2 (12)
Tetrachloro-
ethylene
4.0 (40)
-
-
-
1.3 (13)
1.4 (14)
Chlorobenzene
13 (130)
-
-
-
2.6 (26)
4.3 (43)
Numbers in parenthesis are percent recoveries of spiked compounds (with appropriate blank substracted).
-------
Table 14. QUANTITIES OF TARGET COMPOUNDS MEASURED IN
URINE BLANKS AND CONTROLS--CHAPEL HILL (ng/mL)
Sample
Field Blank-1
Field Blank-2
Lab Blank-1
Lab Blank-2
Field Control-1
Field Control-2
Lab Control-1
Lab Control-2
Chloro-
form
0.2
3.0
3.0
8.8 (86)b
12 (90)
10 (100)
12 (90)
1,2-Dichloro-
ethane
-
-
10 (100)
11 (110)
11 (110)
11 (110)
1,1,1-Trichloro-
ethane
ND
ND
0.7
0.7
12 (120)
12 (120)
15 (143)
13 (123)
Bromodichloro-
rae thane
ND
ND
ND
ND
9.7 (97)
12 (120)
10 (100)
12 (120)
Trichloro-
ethylene
0.1
0.2
0.4
0.3
9.7 (96)
11 (108)
12 (116)
11 (107)
Tetrachloro-
ethylene
Ta
ND
T
ND
9.0 (90)
12 (120)
12 (120)
13 (130)
Chloro-
benzene
1.9
ND
ND
ND
9.0 (71)
9.6 (96)
10 (100)
11 (110)
Compound found in trace quantities.
Numbers in parenthesis represent percent recoveries of spiked compounds (with appropriate blank subtracted).
Median absolute recovery and percent recovery calculated independently.
-------
mean recoveries of the seven spiked compounds improved to 92-126% at UNC.
However, for all compounds in both study areas, field controls showed
lower recoveries than laboratory controls, indicating a possible effect
of transportation. At UNC, the precision was again high, ranging between
5 and 15%.
Blood. Blood controls for the Lamar University student study showed
generally poor or variable recovery efficiencies, ranging from 20-140%
(Table 15). For the UNC student study the efficiencies increased greatly,
ranging from 80-165% (Table 16). Five of the seven spiked compounds were
recovered at levels significantly higher than prepared, indicating the
probable presence of these chemicals in the pooled blood samples donated
as controls. This .problem of accounting for the endogenous background
has not been solved.
In addition to the quality control samples, three blind quality
assurance samples were prepared for both blood and water. These samples
were encoded prior to submission to the analyst. The results indicate
variable recoveries except for chlorobenzene (Tables 17 and 18). Neither
chloroform nor 1,1,1-trichloroethane could be recovered with good precision.
Moreover, false positive identifications occurred consistently for te-
trachloroethylene. The cause of these problems is being investigated.
Questionnaire Results
Demographic characteristics of the student volunteers from Lamar
University and University of North Carolina are displayed in Tables 19
and 20.
Field Results
Air. Of 15 compounds sought, six were found in 100% of the samples:
-37-
-------
Table 15. QUANTITIES OF TARGET COMPOUNDS MEASURED IN
BLOOD PLASMA BLANKS AND CONTROLS--LAMAR UNIVERSITY (ng/mL)
Sample
Lab Blank
Lab Control
Field Blank-3
Field Blank-4
Field Control-3
Field Control-4
System Blankc
System Blank
System Blank
Chloro- 1,2-Dichloro-
form ethane
9.0
17 (80)a
16
16
21 (50)
20 (40)
„
-
-
21
28 (70)
31
30
33 (20)
32 (20)
_
-
-
1,1,1-Trichloro- Bromodichloro- Trichloro- Tetrachloro- Chloro-
ethane methane ethylene ethylene benzene
13 - - 0.5
25 (120) 9.2 (92) 7.3 (73) 5.6 (51) 12 (120)
Tb - 17 -
T 1.2 - T -
8.0 (80) 9.2 (90) 19 (20) 14 (140) 10 (100)
4.4 (44) 3.2 (20) 17 (170) 12 (130) 9.6 (96)
_
_
_
Numbers in parenthesis are percent recoveries of spiked compounds (with appropriate blank subtracted).
bTrace,
CO °Prepurged distilled water blank with 0.5 ml of antifoam added.
-------
Table 16. QUANTITIES OF TARGET COMPOUNDS MEASURED IN
BLOOD PLASMA BLANKS AND CONTROLS--CHAPEL HILL (ng/mL)
Distilled
Water Blank
Distilled
Water Blank
Field Blank-2
Field Blank-3
Lab Blank-1
Blank 40015;!
Blank 40017
Field Control-2
Field Control-3
Lab Contcol-1
Control 40014
Control 4001 c
Mean
SD
cv (%)
Median
Chloro- 1
form
ND
ND
14
4,6
14
14
14
28 (70)d
23 (92)
30 (80)
29 (76)
41 (81)
30 (80)
7 (8)
23 (10)
29 (80)
,,2-Dichloro-
ethane
ND
ND
ND
ND
ND
ND
ND
26 (130)
18 (90)
36 (130)
26 (130)
35 (105)
26 (117)
6 (19)
23 (16)
26 (130)
1,1,1-Tri-
chloro-
ethane
ND
ND
22
21
22
22
21
41 (95)
39 (90)
42 (100)
37 (75)
60 (117)
44 (95)
9 (15)
20 (160)
41 (95)
Bromodi-
chloro- •
methane
ND
ND
ND
ND
ND
ND
ND
28 (140)
36 (130)
29 (134)
28 (140)
40 (120)
30 (135)
6 (10)
20 (7)
28 (140)
Trichloro-
ethylene
ND
ND
ND
ND
ND
T
ND
32 (160)
31 (155)
33 (165)
30 (150)
47 (141)
35 (154)
7 (9)
20 (6)
32 (155)
Tetrachloro-
ethylene
20
15
20
20
20
46
41
47
41
66
48
10
21
46
ND
ND
(130)
(130)
(135)
(105)
(138)
(218)
(13)
(10)
(130)
Chloro-
benzene
ND
ND
Ta
ND
ND
ND
ND
36 (180)
30 (150)
36 (180)
35 (175)
47 (141)
37 (165)
6 (18)
16 (11)
36 (175)
Mean
34 (128)
30 (120)
35 (133)
32 (121)
48 (120)
(124)
(6)
(5)
(121)
SD
8 (37)
8 (29)
8 (35)
6 (38)
11 (22)
CV
24 (29)
27 (24)
23 (26)
19 (31)
23 (18)
Median
32 (130)
30 (130)
33 (135)
30 (130)
47 (120)
Blank 40015 corresponds to Control 40014.
Blank 40017 corresponds to Control 40016.
Control 40016 was spiked at a higher concentration than the other controls.
Numbers in parenthesis are percent recoveries. The concentration of the blanks have been taken into account.
Compound found in trace quantities.
Absolute recovery and percent recovery calculated independently.
-------
-p-
o
Table 17. RESULTS OF WATER BLIND STUDY
Blank
Compound
Benzene
Chloroform
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Carbon tetrachloride
Bromodi chl orome thane
Trichloroethylene and/or
1,1, 2-Trichloroethane
Tetrachloromethylene
Chlorobenzene
Spike
(ng/mL)
0
0
0
0
0
0
0
0
0
0
Found
(ng/mL)
NDb
ND
ND
ND
ND
ND
ND
ND
ND
ND
Spike
(ng/mL)
0
15.5
6
14.1
16.7
0
0
0
17.0
11.6
Spike -1
Found Recovery
(ng/mL) (%)
ND
15.6 100
ND
20 142
13.4 80
ND
ND
9.2 54
12.4 107
Spike
(ng/mL)
0
1.5
0
134
ND
0
0
0
0
0
Spike -2
Found
(ng/mL)
ND
1.7
ND
202
ND
ND
0.02
ND
ND
ND
Recovery
(%)
116
151
Samples prepared with known levels of the compounds as indicated and encoded prior to submission for
analysis.
Not detected.
-------
Table 18. RESULTS OF BLOOD BLIND STUDY
Blank
Compound
Benzene
Chloroform
1 , 2-Dichloroethane
1 , 1 , 1-Trichloroethane
Carbon tetrachloride
Bromodichlorome thane
Trichloroethylene and/or
1,1, 2-Trichloroethane
Tetrachloroethylene
Chlorobenzene
Mean
Spike
(ng/mL)
0
0
0
0
0
0
0
0
0
Found
(ng/mL)
NDb
ND
ND
ND
ND
ND
ND
28
ND
Spike
(ng/mL)
0
31
0
28.2
0
0
23.2
0
11.6
Spike -1
Found
(ng/mL)
0
8.8
0
24
0
0
_c
20
9.5
Recovery Spike
(%) (ng/mL)
0
30 32.8
0
85 28.2
0
0
23.2
0 0
8.2 23.2
66
Spike -2
Found
(ng/mL)
0
31.6
0
15
0
0
-
7.8
18.6
Recovery
96
53
0
80
76
Three samples prepared with known levels as indicated and encoded prior to submission for analysis.
Not detected.
"Not quantitated - interferences.
3
Not included in calculation.
-------
Table 19. DEMOGRAPHIC CHARACTERISTICS OF LAMAR UNIVERSITY VOLUNTEERS
Participant
Code Sex Race Age
D
H
I
K
Residence Possible Exposure Activities
M W 20 Off Worked in pohto-lab developing
(Groves) (9/78-12/78); father refinery
worker
W
W
W
W
21
21
21
Off Used pesticides in yard
(Port Neches)
Off
(Beaumont)
Dorm
22
18
Off
(Beaumont)
Dorm
Ex-smoker; household includes
smoker, chemical plant worker
Family members work at petro-
leum plant; pesticides sprayed
in garden; Raid; Lysol; liquid
starch
Uses liquid paper
Swims regularly;
smokers
M W 24 Dorm Sprayed with diesel fuel
(30 min) prev. week; painted
house four months previously
M W 29 Dorm None
M W 21 Dorm Swins regularly; dorm neighbors
painted indoors recently; dry
cleaning job over Christmas
M W 21 Dorm Swims regularly; worked at
chemical plant 5/79-8/79;
exposed to smokers
M W 23 Dorm Pumped gas in last 24 hours;
uses hair spray twice a week
42
-------
Table 20. DEMOGRAPHIC CHARACTERISTICS OF UNC VOLUNTEERS
Participant
Code
U
Sex Race Age
M W 21
Residence
Off
(Carrboro)
Possible Exposure Activities
Swims regularly; lifeguard;
swam within last 24 hours;
taking cortisone.
V M W 20 Off Swims regularly; was employed
(Chapel Hill) (dishwasher) in Dept. of Anes-
thesiology for 4 months; mother
works in antibody screening
room at Red Cross; uses bleach
when mopping floors at night.
W M W 19 Off Swims regularly; used pesti-
(Carrboro) cide in last 72 hours; uses
spray starch, deodorant;
smokers at home.
X M W 19 On Lifeguard; swam in last 24
hours; pumped gas in last 72
hours; uses spray deodorant.
Y M W 22 Off Inhaled Methyl chloride for
another EPA experiment; used
liquid paper in last 24 hours.
Z M W 19 On Cook; swims regularly; swam in
last 24 hours; uses spray deo-
dorant; smokers in household.
43
-------
benzene, chloroform, trichlorothylene, tetrachloroethylene, 1,1,1-trich-
loroethane, and dichlorobenzene isomer; and four others in more than
50% of the samples: vinylidene chloride, ethylene dichloride, bromodich-
lororaethane and ^-dichlorobenzene (Tables 21 and 22).
Six of these 10 compounds showed high variability, ranging over 2-3
orders of magnitude: trichloroethylene, tetrachloroethylene, 1,2-dichlo-
roethane, 1,1,1-trichloroethane, dichlorobenzene isomer, and vinylidene
chloride.
Geometric means for one compound—1,1,1-trichloroethane—exceeded 50
ug/m^ in each student group. Geometric means for seven other compounds
generally fell between 1 and 10 ug/m^ for each group (Fig. 5). No signi-
ficant difference in concentration was noted between the two student
groups for any compound.
Breath. Results of the breath sampling are displayed in Tables 23
andA24. Five compounds were found in 100% of the samples; benzene,
chloroform, tetrachloroethylene, 1,1,1—trichloroethane, and dichloro-
benzene isomer ; two others were found in more than 50% of the samples:
trichloroethylene and vinylldene chloride.
Five of these seven compounds showed highly variable levels trich-
loroethylene, tetrachloroethylene, 1,1,1-trichloroethatie, m-dichloroben-
zene, and vinylidene chloride. Two showed low variability: benzene and
chloroform. Geometric means for these seven compounds ranged from about
1-15 ug/m^. Significant differences were noted between the two student
groups in breath concentrations for three of the seven compounds: chloro-
form, vinylidene chloride, and 1,1,1-trichloroethane. In all three cases,
UNC geometric means were higher.
-44-
-------
Ui
Table 21. ESTIMATED LEVELS OF SELECTED VAPOR-PHASE ORGANICS IN AMBIENT AIR
ASSOCIATED WITH HUMAN PARTICIPANTS--LAMAR UNIVERSITY STUDENT STUDY (yg/m3)
Compound
Benzene
Chloroform
Vinylidene chloride
1 , 1-Dichloroethane
1 ,2-Dichloroe thane
1 , 1 , 1-Trichloroe thane
Trichloroethylene
1 , 2-Dichloropropane
Tetrachloroethylene
Broaodichloroae thane
Dibroaochloroaethane
Ethylene dibroaide
Chlorobenzene
Dichlorobenzene isoaer
o-Dichlorobenzene
30001
9.5
1.4
416
1.8
11
592
19
-
174
1.6
-
-
-
8.4
30002
2.5
1.5
1.4
-
0.49
22
7.5
-
162
1.5
-
-
-
73
30003 30004
11 2.9
3.8 8.3
76 1.0
0.93
0.32
1,069 8.5
26 1.6
-
7.2 5.4
3.2
-
-
-
8.0 6.9
2.4
30005
3.6
5.2
1.1
-
0.52
8.3
0.90
-
5.6
1.0
-
-
0.47
2.5
0.38
Participant No.
30011 30012
4.8 5.8
4.0 4.8
7.0
-
1.0 0.94
31 62
3.8 2.0
-
5.0 30
1.1
-
-
-
6.4 3.0
0.20
30013 30014 30015
8.3 3.2 5.3
6.0 3.2 1.9
5.7 2.1 4.6
-
0.95 0.72 0.71
72 12 67
63 0.99 2.4
.
718 4.5 172
3.7 0.84
-
-
-
23 ' 1.8 4.3
30016 iODa
386 0.08
4.8 0.08
0.12
0.12
]3 0.12
40 0.16
3.7 0.16
0.20
9.3 0.24
0.24
0.24
0.28
2.1 0.16
33 0.20
0.20
QLb
0.40
0.40
0,60
0.60
0.60
0.80
0.80
1.00
1.20
1.20
1.20
1.40
0.80
1.00
1.00
Liait of Detection (LOD) was defined as S/N - 4 for m/z ion selected for quantification, all values in M8/"3-
Quantifiable Liait (QL) was defined as 5 x LOD or S/N - 20, all values in M8/»3-
-------
Table 22. ESTIMATED LEVELS OF VAPOR-PHASE ORGANICS IN AMBIENT AIR
FOR SEVERAL HUMAN SUBJECTS—UNIVERSITY OF NORTH CAROLINA
AT CHAPEL HILL STUDY
Participant Number
Compound
Benzene
Chloroform
Vinylidene chloride
1 , 1-Dichloroethane
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
1 ,2-Dichloropropane
Tetrachloroethylene
Broraodichloromethane
Dibromochloromethane
Ethylene dibromide
Chlorobenzene
Dichlorobenzene isomer
o-Dichlorobenzene
40001
14
7.8
14
a
-
165
9.7
-
2.5
-
-
-
-
35
1.5
40002 40003 40011
7.5 3.8 3.2
5.1 3.7 3.2
27 9.8 3.5
_
1.1 0.42
194 93 14
2.2 10.8 4.6
- - -
2.6 2.1 1.2
_
_
_
0.17
0.58 0.46 0.29
0.32
40012
4.2
17
5.7
-
0.45
70
2.2
-
4.3
4.3
-
-
0.18
15
0.27
40013
3.0
2.2
7.0
-
0.63
57
183
-
127
-
-
-
-
0.63
0.14
a
- = not detected.
46
-------
Geometric Mean Concentrations of Seven Volatile Organics in Air
and Exhaled Breath of Two Student Groups
9 LamarUniv. (N = ID
A Univ. North Carolina (N = 6)
Concentration
t jug/m3)
100
80
60
SU
40
30.
20
10
8
6
5
4
3
0.8
0.6
0.5
0.4
0.3
0.2
0.1
Air
Breath
IN=5)
IN"5)
B
./
/
B/
v
^ A
\
\
V
\ \
1
|
u
£
•g
5
£
Figure 5. Geometric mean concentrations of seven volatile organics
in air and exhaled breath of two student groups.
47
-------
Table 23. ESTIMATED LEVELS OF SELECTED VAPOR PHASE ORGANICS IN BREATH--
LAMAR UNIVERSITY STUDENT STUDY (yg/m3)
Participant Number
Compound
Benzene
Chloroform
Vinyl ideoe chloride
1 , 1-Dichloroe thane
1,2-Dichloroethtne
1,1, l-Tcichloroeth*ne
Tcichloroetbyleoe
1 , 2-Dichloropropane
Tetrachloroethylene
-P-
oo
Bromodichlorome thane
Dibroaochlorome thane
Ethylene dibroeide
Cblorobenxene
Dichlorobenzene isomer
o-Dichlorobenzene
30001
2.9 t
1.1
T'
15 ±
2.8
-b
-
161 ±
16
-
-
69 t
5.4
-
-
T
-
30002 30003 30004
1.4 ± 0.7 ± 1.71 ±
0.2 0.0 0.28
T T T
0.08 26 t 0.08
8.0
-
-
0.66 ± 93 ± T
0.16 21
1.11 ± T T
0.04
-
96 1 9.8 ± 13 ±
0.20 1.0 1.5
-
-
.
31 ± T 1
2.5
-
30005 30011
0.99 t 1.8 ±
0.33 0.13
t T
2.9 0.08
-
-
T T
T
-
13 ± T
0.71
-
-
.
T T
-
30012 30013 30014 30015 30016
0.15 0.15 0.20 0.52
2.48 T T T T
0.08 0.5 5.8 ± 3.8 t 0.08
T 1,6 1.2
-
T - T
2.5 T 1.7 1 6.5 ± T
0.46 0.75
T 1.45 ± T 1.07 ± T
0.10 0.04
.
24 161 ± 1.0 176 ± T
31 0.91
-
,
-
1.3 ± 20 ± 6.1 ± 23 ± T
6.2 3-0
.
too
0.11
0.11
0.16
0.16
0.16
0.22
0.2
0.27
0.33
0.33
0.33
0.38
0.22
0.27
0.27
QL
0.55
0.55
0.82
0.82
0.82
1.10
1.10
1.37
1.65
1.65
1.65
1.92
1.10
1.37
1.37
T = trace amount.
-------
Table 24. ESTIMATED LEVELS OF VAPOR PHASE ORGANICS IN BREATH OF HUMAN SUBJECTS-
UNIVERSITY OF NORTH CAROLINA AT CHAPEL HILL STUDY
Compound
Benzene
Chloroform
Vinylidene chloride
1 , 1-Dichloroethane
1 ,2-Dichloroethaoe
1 , 1 , 1-Trichloroethane
Trichloroethylene
40001
1.0 ± 0.05
2.8 ± 1.5
4.5 ± 0.39
_b
-
23 ± 3.6
1.1 ± 0.12
40002
1.4 ± 0.23
3.0 ± 0.16
14 ± 1.3
-
-
48 ± 11
0.55 ± 0.19
Participant Number
40003 40011 40012
1.4 ± 0.18 1.1 ± 0 NCa
5.1 ± 2.8 1.8 ± 0.28 NC
5.5 ± 1.3 3.9 ± 0.09 7.7 ± 0.05
_
0.37
19 ± 5.7 6.1 ± 0.04 8.5 t 1.8
1.2 ± 0.36 0.65 ± .05 0.49 t 0.11
40013
1.5 ± .41
1.7 ± 0.41
7.9 ± 0.65
-
0.48
13 ± 0.34
32 ± 0.70
3.4 ±0.44 3.3 ±0.19
1 ,2-Dichloropropane
Tetrachloroethylene
BroaodichloroBethane
Dibromochloroaethane
Ethylene dibromide
Cblorobenzene
Oichlorobenzene Isooer 0.54 + .03 4.5 + 0
o-Dichlorobenzene
4.3 ± 0.76 8.8 ± 1.1 7.5 ± 0.68 48 ± 5.5
0.27
2.2 + 0.56 0.92 + 0
5.3 + 0.71
1.1 •»• 0.36
*NC = Missing values.
- = not detected,
Q
Given value is below the limit of detection.
-------
Water. Ten VOC were measured in tap water (Tables 25 and 26). UNC tap
water showed consistently higher mean chloroform values than the Lamar
University sources (220 ppb to 150 ppb), except for the one water sample
at the Port Neches home of one of the Lamar University students. (This
was the only sample not taken from the Beaumont water supply). Tetrach-
loroethylene values were also higher In the UNC supplies. Bromodichloro-
methane values were similar in the two supplies (20 ppb at Lamar; 17 ppb
at UNC). Total trihalomethanes exceeded the standard of 100 ppb in all
38 water samples from the two areas.
All of the tap water samples contained chloroform and bromodichlorome-
thane. Some samples contained small amounts of tetrachloroethylene, chlo-
robenzene and either trichloroethylene or 1,1,2-trichloroethane. No
benzene, carbon tetrachloride, 1,2-dichloroethane, vinylidene chloride,
or 1,1,1-trichloroethane was detected.
Blood '.and Urine. Because of the quality control difficulties discussed
above, the analytical results of the blood and urine tests will not be
listed or discussed.
-50-
-------
Table 25. QUANTITIES OF TARGET COMPOUNDS FOUND IN TAP WATER (ng/aL), LAMAR UNIVERSITY
Number
1-30001
1-30002
2-30002
1-30003
1-30004
2-30004
1-30005
2-30005
1-30011
2-30011
3-30011
1-30012
2-30012
3-30012
1-30013
2-39913
3-30013
1-30014
2-30014
3-30014
1-30015
2-30015
1-30016
2-30016
3-30016
LOD
Chloro-
fora
110
260
130
550
160
99
140
120
120
120
110
120
170
110
140
160
130
160
130
110
110
140
120
120
120
1.0
Carbon Tetra-
1,1,1-Tri- chloride and/
1,2-Dichloro- chloro- or Bromodi-
ethane ethane chloromethane
-a - 16
18
23
44
25
18
25
22
20
18
23
22
26
17
18
22
18
20
18
13
7.4
18
13
13
14
0.6 0.2 0.4
Trichloroethylene
and/or 1,1,2-Tri-
chloroetbane
NCb
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
1.3
Tetrachloro- Chloro-
ethylene benzene
_
-
-
Tc
-
_
.
-
~ *" j
0.2 0.2°
-
-
.
-
* *
O.ld
_
.
_
_ _
0.2d 0.2d
_
-
™ J "*
O.ld
1.1 0.6
*- - not detected.
bNC = »issing data.
T = t
d
T - trace aaxmat.
Given value is below liait of detection.
-------
Table 26. QUANTITIES OF TARGET COMPOUNDS FOUND IN TAP WATER (CHAPEL HILL) (ng/mL)
to
Sample
40001-1
40001-2
40002-1
40002-3
40003-1
40003-2
40011-1
40011-2
40011-3
40012-1
40012-2
40013-1
40013-2
Mean
SD
CV (%)
Median
LODC
Chloro-
form
260
260
220
250
200
230
200
210
220
210
210
180
200
220
23
10
220
0.05
1,2-Dichloro-
ethane
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
-
-
0.06
1,1,1-Trichloro-
e thane
NDa
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
-
-
0.1
Broraodichloro-
me thane
20
19
17
18
18
18
17
16
17
17
16
15
15
17
2
12
17
0.1
Trichloro-
ethylene
2.8
3-0
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.3
0.6
1
170
ND
0.05
Tetrachloro-
ethylene
3.8
3.8
1.8
1.8
1.8
1.8
1.7
1.8
1.8
1.8
1.8
1.3
1.3
2.0
0.8
40
1.8
0.05
Chloro-
benzene
ND
ND
ND
ND
1.4
ND
ND
ND
1.5
ND
ND
1.5
ND
0.4
0.6
150
ND
ND
Not detected.
bMean of all values (ND - 1/2 LOD).
cLimit of detection (S/N = 3).
-------
SECTION 7
DISCUSSION
SUMMARY STATISTICS: AIR AND BREATH
The air and breath data are summarized in Tables 27 and 28. Each
statistic was computed using the measurements for all subjects. For
those compounds not detected, values were estimated as 1/2 LOD. Those
compounds detected as trace levels were estimated as 1/2 (QL+LOD). The
numbers displayed as ( , ) indicate how many of the samples were below
the limit of detection and at trace levels, respectively. The median,
as well as the arithmetic mean, is provided because of the skewness of
the data. The standard deviations, minumum and maximum values suggest
large variation in the data for most compounds. In most cases, the
standard deviations are larger than the means and the magnitude of this
relationship is reflected in the coefficients of variation:
(CV = Standard deviation x 100%)
mean
The concentrations of some chemicals reached unexpectedly high
levels, both in air and in human breath, compared to recent studies
(1,7,9) of ambient levels. These levels are far below the workplace
standards set by the Occupational Safety and Health Administration; how-
ever, their chronic effects are unknown and could be of significant public
health concern.-
The great variability exhibited by seven of the 10 most prevalent
-53-
-------
Table 27. SUMMARY STATISTICS FOR ESTIMATED LEVELS OF SELECTED VAPOR-PHASE ORGANICS—LAMAR UNIVERSITY
Benzene
Chloroform
Vinyl idene chloride
1 , 1-Dichloroe thane
1 ,'2-Dichloroe thane
1,1, 1-Trichloroe thane
Trichloroethylene
1,2-Dichloropropane
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
I8
Detected
100
100
100
100
82
55
18
0
91
18
100
100
100
82
0
0
X at
Trace
0
0
0
91
0
9
0
0
9
18
0
45
0
54
0
0
Meanb
40.30
1.61
4.00
0.42
46.84
4.88
0.30
0.08
2.72
0.12
180.35
24.37
11.88
0.59
0.10
0.13
Std. Dev.
118.87
0.62
2.13
0.68
124.43
8.19
0.57
0.00
4.66
0.10
339.47
53.14
18.90
0.43
0.00
0.00
Median
5.16
1.70
4.04
0.22
2.14
0.33
0.06
0.08
0.72
0.08
40.00
0.66
3.67
0.44
0.10
0.13
Range
2.46-
386.56
0.72-
2.95
1.39-
8.33
0.22-
2.48
0.06-
416.07
0.08-
25.17
0.06-
1.83
0.08-
0.08
0.06-
12.80
0.08-
0.33
8.27-
1,069.04
0.44-
161.50
0.90-
63.39
0.11-
1.45
0.10-
0.10
0.13-
0.13
%> .
10 Mg/«3
18
0
0
0
18
18
0
0
18
0
82
18
27
0
0
0
(continued)
-------
Table 27. (continued)
Tetrachloroethylene
Bromodichloromethane
Dibromochloromethane
Ethylene Dibromide
Chlo robenzene
Dichlorobenzene I some r
o-Dichlorobenzene
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Detected
100
100
64
0
0
0
0
0
18
0
100
100
27
0
% at
Trace
0
J8
0
0
0
0
0
0
0
0
0
54
0
0
Meanb
117.63
51.36
1.23
0.16
0.12
0.16
0.14
0.19
0.30
0.11
15,56
7.73
0.34
0.13
Std. Dev.
212.38
65.61
1.24
0.00
0.00
0.00.
0.00
0.00
0.08
0.00
21.52
11.31
0.67
0.00
Median
9.26
13.25
1.00
0.16
0.12
0.16
0.14
0.19
0.61
0.11
6.95
0.56
0.10
0.13
Kange
4.54-
718.20
0.66-
176.32
0.12-
3.71
0.16-
0.16
0.12-
0.12
0.16-
0.16
0.14-
0.14
0.19-
0.19
0.08-
2.12
0.11-
0.11
1.83-
73.47
0.55-
30.67
0.10-
2.40
0.13-
0.13
10 Mg/m3
45
64
0
0
0
0
0
0
0
0
27
27
0
0
n = 11.
All compounds measured in (Jg/m3.
-------
Table 28.
SUMMARY STATISTICS FOR ESTIMATED LEVELS OF SELECTED VAPOR PHASE ORGANICS
UNIVERSITY OF NORTH CAROLINA AT CHAPEL HILL STUDY
Benzene
Chloroform
Vinylidene chloride
1 , 1-Dichloroethane
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
I ,2rDichloropropqne
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
BreatTi
Detected
100
100
100
10
(n=5)
100
100
0
0
67
33
100
100
100
100
0
0
% at
Trace
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Meanb
5.89
1.29
6.59
2.86
11.21
7.28
0.06
0.08
0.45
0.19
99.07
19.47
35.33
5.94
0.10
0,13
Std. Dev.
4.14
0.22
5.66
1.36
8.66
3.73
0.00
0.00
0.39
0.18
68.06
15.14
72.16
12.60
0.00
0.00
Median
4.00
1.38
4.41
2.84
8.40
6.59
0.06
0.08
0.43
0.08
81.81
15.97
7.13
0.86
0.10
0.13
Range
2.95-
13.65
1.00-
1.51
2.25-
17.46
1.70-
5.06
3.53-
27.29
3.94-
14.12
0.06-
0.06
0.08-
0.08
0.06-
1.09
0.08-
0.48
14.45-
193.77
6.10-
47.63
2.17-
182.43
0.49-
31.66
0.10-
0.10
0.13-
0.13
10 Mg/»3
17
0
17
0
33
17
0
0
0
0
100
67
33
17
0
0
(continued)
-------
Table 28. (continued)
Tetrachloroethylene
Bromodichlorone thane
DibroDochlorome thane
Ethylene Dibroaide
Chlorobenzene
Dichlorobenzene Isooer
o-Di chlorobenzene
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Air
Breath
Xs
Detected
100
100
17
17
0
0
0
0
33
17
100
100
50
0
X at
Trace
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Meanb
23.34
12.52
0.83
0.29
0.12
0.16
0.14
0.19
0.11
0.14
8.66
2.42
0.40
0.13
Std. Dev.
50.94
17.40
1.73
0.30
0.00
0.00
0.00
0.00
0.05
0.0
14.14
2.01
0.55
0.00
Median
2.56
5.92
0.12
0.16
0.12
0.16
0.14
0.19
0.08
0.11
0.61
1.53
0.18
0.13
Range
1.22-
127.30
3.30-
47.74
0.12-
4.36
0.16-
0.91
0.12-
0.12
0.16-
0.16
0.14-
0.14
0.19-
0.19
0.08-
0.18
0.11-
0.27
0.29-
34.96
0.54-
5.30
0.10-
1.52
0.13-
0.13
X >
10 M8/»3
17
17
0
0
0
0
0
0
0
0
33
0
0
0
0=6 unless otherwise noted.
All compounds are neasured in
-------
compounds indicates that human exposures to many VOC may vary so widely
even in a very small geographical area that it is not generally possible
to characterize a geographical cohort with unifrom exposures. If this
preliminary conclusion is validated by future studies, it would have
serious implications for epidemiological studies, which have traditionally
assigned similar exposure histories to residents of a given region.
At Lamar University, four of the five students with the highest air
exposures and breath levels lived off-campus. The reason for this result
is unknown.
Table 29 shows the relative contribution of each of 12 volatile
organics to the air exposure of each of the 17 subjects. In both study
areas, 1,1,1-trichloroethane is the main contributor, supplying over half
the total intake at UNC, and more than a third at Lamar. The relative
importance of benzene, chloroform, and vinylidene chloride was also very
similar at each location, ranging between 4% and 8%. However, tetrach-
loroethylene was far more important at Lamar University than at UNC (35%
to 7%), while trichloroethylene was relatively more important at UNC
(13% to 4%). Dichlorobenzenes also seemed more common at Lamar than at
UNC.
Bromodichloromethane, chlorobenzene and 1,1-dichloroethane contribut-
ed less than 2% to total air exposure at Lamar, and even less at UNC.
CORRELATIONS BETWEEN AIR AND BREATH SAMPLES
Spearman correlation coefficients were computed between the Lamar
air and breath samples for all compounds (Table 30). Of particular
interest are the correlations between air and breath for a given compound,
shown in the highlighted diagonal of the correlation matrix. Three of
-58-
-------
Table 29. PERCENT OF INDIVIDUAL AIR EXPOSURE SUPPLIED BY
SELECTED VAPOR-PHASE ORGANICS--BOTH GROUPS
Ul
VD
Participant
Laaar University
30001
30002
30003
30004
30005
30011
30012
30013
30014
30015
30016
Mean
Standard Deviation
ONC
40001
40002
40003
40011
40012
40013
Mean
Standard Deviation
Benzene 1
0.76
0.86
0.86
7.1
8.8
8.6
4.7
0.93
11.0
2.0
*
4.7
3.8
5.6
3.1
3.0
11
2.4
0.8
4.3
±3.3
*Outlier — not included in
Chloroform 1
0.11
0.51
0.30
20
13
7.1
3.9
0.67
11.0
0.7
6.1
5.8
±6.2
3.1
2.1
3.0
11
9.6
0.6
4.9
±3.9
Vinylidene chloride
1 , 1-Oichloroethane
34 0.14
0.48
6.0 0.07
2.4
27
-
5.7
0.67
7.2
1.7
-
7.8
±11.0
5.6
11
7.8
12
3.2
1.8
6.9
±3.8
1,2-Dichloroethane
0.88
0.17
-
0.73
1.2
1.8
0.7
0.11
2.4
0.26
16
2.2
±4.4
_
-
0.88
1.3
0.23
0.16
0.43
±0.49
1,1,1 -Trichloroethane |
48
8.0
85
21
20
55
51
8.0
41
25
51
38
±22
66
81
74
47
40
15
54
±23
Trichloroethylene 1
1.5
2.6
2.1
3.9
2.2
6.8
1.6
7.0
3.4
0.9
4.7
3.5
±2.0
3.9
0.9
8.8
15
1.2
48
13
±16
Tetrachloroethylene
14
59
0.57
13
14
87
25
80
16
67
12
35
±30
1
1.1
1.7
4.0
2.4
34
7.4
±12
o
o
•fH
CO
0.13
0.52
-
7.8
2.4
-
0.90
0.40
2.8
-
-
1.4
±2.3
.
-
-
.
2.5
-
0.4
-
Chlorobenzene
B-Dichlorobenzene
0
27
0
17
1.2 6
11
2
2
6
1
2.7 42
11
±12
14
0
0
0.7 1
0.1 8
0
0.1 4
±5
.67
.63
.0
.5
.5
.2
.6
.2
.4
.0
.5
.16
.0
.3
o-Dichlorobenzene I
.
-
-
5.9
1.0
-
0.17
-
-
•
-
0.6
*
0.6
0.1
O.I
0.3
0.2
0.03
0.2
±0.2
calculations .
-------
Table 30. CORRELATIONS BETWEEN AIR AND BREATH FOR ESTIMATED
LEVELS OF SELECTED VAPOR-PHASE ORGANICS--LAMAR UNIVERSITY
Air
Benzene
Chloroform
Vinylidene Chloride
1 , 1-Dichloroethane
1,2-Dichloroe thane
1 , 1 , 1-Trichloroe thane
Trichloroethylene
1,2-Dicaloropropane
Tetrachloroethylene
Bronodichloroaethane
DibromochloromeLhane
Ethylene Oibrooide
Cblorobenzeae
Dichlorobenzeae Isoner
o-Dichlorobenzene
Benzene
.04
-.16
.24
.07
.44
.27
.12
0
.4
.29
0
0
-.43
.03
.03
E
o
Chlorol
.10
.20
.30
-.15
.10
.10
-.20
0
.10
.10
0
0
-.14
-.30
.38
ene Chloride
-ri
*-*
C
.31
-.46
.67
*
.69
*
-.18
.47
.18
0
.11
-.15
0
0
-.18
-.24
-.33
.hloroethane
a
.H
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
o
.c
ft
a
i
.15
.07
.22
-.22
.07
.37
.30
0
.60
.11
0
0
-.22
.07
-.28
.
r-l
ft
I
•H
.H
.29
-.74
**
.82
*•*
.71
-.12
.63
*
.25
0
.32
-.14
0
0
-.47
-.08
-.32
Breath
>roe thy lene
Trichlc
-.15
.01
-.02
-.43
-.14
.18
.41
0
.39
.12
0
0
-.37
.40
-.38
hloropropane
r4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
loroethylene
2
H
-.12
-.19
.53
.03
-.17
.30
.25
0
.80
**
.49
0
0
-.36
.16
-.02
o
1
0
J-
u
fiQ
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
chlorome thane
I
•H
O
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
e Dlbromlde
&>
1-
4J
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
enzene
Chlorob
0
0
0
0
0
0
0
0
0
,0
0
0
0
0
0
orobenzene
|
-.36
-.25
.20
-.40
-.14
.04
.11
0
.43
.19
0
0
-.40
.08
-.32
orobenzene
j:
u
s
01
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Spearman correlation coefficient.
n = 11 for each coefficient.
*=p<.05; ** = p <.01.
60
-------
the five compounds found at Lamar in sufficient numbers above trace levels
to allow meaningful correlations were significantly correlated (p, the pro-
bility of a chance correlation, is less than 0.05). Correlations signi-
ficantly different from zero are denoted by * (p<.05) and ** (p<.01).
Spearman correlation coefficients were also computed between air and
breath samples for the UNC students. (Table 31). Although the number of
samples was exceedingly small (N=6), two correlations were significant at
p<0.01. One of the compounds (1,1,1-trichloroethane; £=0.94), was also
significant at Lamar (£=0.63).
Since no significant differences were observed between the two stu-
dent groups with respect to air concentrations, they were combined into a
single group (N=17). and studied further for correlations between chemicals
in air and breath (Table 32). Four of the five chemicals prevalent in the
breath samples showed significant correlations with their concentrations
in air; only benzene showed little correlation betwe'en breathing-level
exposure and breath clearance.
Breath/Air Relationships.
Regressions were run for several compounds relating breath levels to
air levels. Logarithms were employed because of the wide range of con-
centrations. To determine the effect of assigning different numerical
values to the "Trace" and "Non-Detect" (ND) categories, two approaches
were used:
1) The "Standard" approach of assigning the Non-detectable (ND)
category a value of 1/2 the Limit of Detection (LOD) and the
"Trace" category a value halfway between the LOD and the Quan-
tifiable Limit (QL);
-61-
-------
Table 31. SPEARMAN CORRELATION COEFFICIENTS BETWEEN AIR
AND BREATH FOR ESTIMATED LEVELS OF
SELECTED VAPOR PHASE ORGANICS--BOTH GROUPS
Benzene
Chloroform
Vinylidene chloride
1 , 1-Dichloroethane
1,2-Dichloroethane
1 , 1 , 1-Trichloroethane
Trichloroethylene
1 , 2-Dichloropropane
Tetrachloroethylene
Bromodichloromethane
Dibromochloromethane
Ethylene dibromide
Chlorobenzene
Dichlorobenzene isomer
o-Dichlorobenzene
* p < .01
**p < .05
UNC
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
n
Lamar
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
rAir-Breath
UNC
.70
.60
.48
.00
.44
.94**
.94**
.00
.20
.20
.00
.00
.31
.03
.00
Lamar
.04
.20
.67*
.00
.07
.63*
.41
.00
.80**
.00
.00
.00
.00
.08
.00
62
-------
Table 32. SIGNIFICANT SPEARMAN CORRELATION COEFFICIENTS FOR VOLATILE ORGANIC COMPOUNDS OBSERVED IN
BREATHING-ZONE AIR AND IN EXHALED BREATH OF 17 STUDENTS AT LAMAR UNIVERSITY
AND UNIVERSITY OF NORTH CAROLINA
Breath
Air
Benzene
Methyl Chloroform
Tetrachloro-
ethylene
trichloro-
ethylene
Vinylidene
Chloride
Benzene
1,1,1-Trichloroethane
Tetrachloroethylene .54
Trichloroethylene
Vinylidene Chloride
1,2-Dichloroethane .54
,74
.73
.88
.53
.62
.77
-------
2) The "Maximum" approach of setting ND values equal to the LOD and
Trace values equal to the QL. (No "Minimum" approach was used
since setting an ND value equal to 0 would give a logarithm of
negative infinity.) The results of these approaches are compared
in Table 33.
The table indicates that little difference occurs in the regression
parameters as a result of the two approaches. For all three chemicals
50% or more of the variance in breath levels was explained by the preceding
air exposures. The F values are all above the 99% level.
The simple log-linear model appears capable of predicting breath levels
to within a factor of 3 or 4, given the air exposures for the preceding
eight hours.
Tetrachloroethylene: CB=10'23 CA -72±-13
1,1,1-Trichloroethane: Cg=10~'98 CA -91±-23
Vinylidene chloride: CB=10~'24 CA -71±-17
A linear relationship between the logarithms of air exposures and
breath concentrations of several compounds is .suggested in Figures 6—8.
If this preliminary observation is confirmed by future studies, an
exposure-dose relationship could be established for some compounds.
This would allow exposures to be estimated from a single non-invasive
test lasting just 15 minutes.
Correlations Within Air and Breath Media
Of the 15 chemicals sought in air and breath, eight in air and five
in breath were above trace levels often enough to include in a study of
correlations within and between media. Spearman rank correlations were
determined for all possible pairs of these chemicals in air, in breath,
-64-
-------
Table 33. COMPARISON OF BREATH-AIR REGRESSIONS USING DIFFERENT
CONVENTIONS FOR ASSIGNING VALUES TO
"TRACE" AND "NON-DETECTABLE" CATEGORIES
Approach
Aa B3
F
R2 S.E.
S.E. (B)
Tetrachloroethylene
Standard
c
.23 .72
31
.68 .44
.13
39
1,1,1-Trichloroethane
Standard
Maximum
Standard
Maximum
-.98
-.77
-.23
. -.13
.91
.84
Vinylidene
.71
.67
16
17
Chloride
17
18
52
73
53
5'5
.54
.48
.67
.56
.23
.20
.17
.16
Log (Concentration in Breath) = A + B log (Concentration in Air).
Standard Approach: ND = 1/2 LOD; Trace = 1/2 (LOD + QL).
"Maximum Approach: ND = LOD; Trace = QL.
65
-------
Tetrachloroethylene In Exhaled Breath Compared to Mean Breathing-Level
Concentrations Averaged Over the Preceding 6—9 Hours for Two Student Groups
H-
TO
C
if
(D
CT>
H
fl>
cr- rt
i-j i-j
m B
PI o
rt 3-
p- l-l
H. o
3 i-4
OP o
i fi>
H> rt
m cr
<: •<
n> M
Mg
o n>
o
P H-
O 3
rt
n
to
rt
O CL
3
to cr
• ^
(D
n
o
•a
n>
a.
rt
O
3
m
Breath
Concentration
in tig/m^
100
10
Trace (T)
Not 1
Detectable (NO)
Larnar Univ.
Univ. of N. Carolina
Quantifiable Limit
10
100
1000
Mean Air Concentration i
-------
"3
H
n>
cr-
H M
o> i
(a t-j
rt r-j
P* H-
H- D
» er
OQ H>
I O
M H
(D O
< (D
O P
o n>
P
O H-
§5
rt (B
H («!
to sr
ft (13
O
P
w
ft)
a-
a4
I-!
m
(U
rt
p-
i
a.
Brea'h
Concentration
in ug/m3
1,1,1-Trichloroethane in Exhaled Breath Compared to Mean Breathing-Level Concentrations
Averaged Over the Preceding 6-9 Hours for Two Student Groups
100
10
Trace (T)
Quantifiable Limit
• Lamar Univ.
A Univ. of N. Carolina
10 100
Mean Air Concentration in^jg/nv*
1000
-------
oo
•n
H-
(TO
c
fl>
00
o-l
f» H-
rr o,
D* n>
H- O
3 TO
OQ
I O
I-1 ET
(D M
< O
ft) tl
I-1 H-
D-
O fB
§ H.
n a
•to o4
rt
cr
D
O
I
ro
a.
(6
§
Breath
Concentration
in ug/rn-*
10
Trace (T) . .
Not 0.1
Dctuctabln
(NO)
Vinylidene Chloride in Exhaled Breath Compared to Mean Breathing-Level Concentrations
Averaged Over the Preceding 6 to 9 Hours for Two Student Groups
A •
Quantifiable Limit
* Lamar University
Univ. of N. Carolina
0.1
1 10
Mean Air Concentration in
100
1000
-------
and between air and breath for each group separately and for both groups
combined (Tables 34, 35, 36). For the combined groups, 16 of the 78
possible correlations were significant at the p <.05 level, a result
obtainable by chance less than once in ICH^ tries. A striking result was
the relation between vinylidene chloride and 1,1,1-trichloroethane, which
displayed the strongest correlations in air, in breath, and also between
air and breath. These chemicals occur together in the manufacture of
vinylidene chloride, and so might be expected to correlate in a manufactu-
ring area. However, they also correlated in the nonmanufacturing area.
Hence an atmospheric chemistry relationship may also exist.
As a check on the Spearman calculations and also on the suitability
of employing the logarithms of the observed chemical concentrations in
air and breath, Pearson correlation coefficients were calculated for the
logarithms of the concentrations. Fifteen of the 16 significant Spearman
correlations were also significant when calculated by the Pearson method
(Table 37).
Breath-Air Ratios
The concentrations of seven VOC in air were compared to their con-
centrations in exhaled breath for the two student groups (Table 38).
Assuming constant concentrations in air and breath, and no additional
exposures through water or food, the breath-to-air ratio of a compound
is simply the additive inverse of the absorption factor for that compound.
(That is, if one breathes out 30% of what he breathes in, then 70% of
the quantity has been absorbed).
Breath-air ratios are similar between the two student groups for
benzene (0.3), vinylidene chloride (0.8) and trichloroethylene (0.2).
-69-
-------
Table 34. SPEARMAN CORRELATION COEFFICIENTS FOR SELECTED VAPOR-PHASE
ORGANICS IN AIR AND BREATH: LAMAR UNIVERSITY
*1.
2.
3.
4.
*5.
6.
7.
8.
9.
10.
11.
*12.
13.
*14.
Air vs.
1,1, 1-Trichloroethane
Vinylidene chloride
1 , 1 , 1-Trichloroethane
m-Dichlorobenzene
1,1, 1-Trichloroethane
1 , 1 , 1-Trichloroethane
Trichloroethylene
Breath vs .
n-Dichlorobenzene
1,1, 1-Trichloroethane
m-Dichlorobenzene
Air vs.
Tetrachloroethylene
Vinylidene chloride
Vinylidene chloride
1,1, 1-Trichloroethane
Spearman
Rank
Air Correlation N
Benzene
Benzene
Trichloroethylene
Trichloroethylene
Vinylidene chloride
Tetrachloroethylene
Tetrachloroethylene
Breath
Trichloroethylene
Vinylidene chloride
Tetrachloroethylene
Breath
Tetrachloroethylene
1,1, 1-Trichloroethane
Vinylidene chloride
1,1, 1-Trichloroethane
.88
.79
.79
.79
.72
.64
.61
.79
.67
.67
.80
.82
.67
.63
10
10
11
11
11
11
11
11
11
11
11
11
11
11
P
.0008
.006
.004
.008
.012
.035
.047
.004
.024
.025
.003
.002
.024
.038
*Also significant (p < 0.05) at University of North Carolina.
70
-------
Table 35. SPEARMAN CORRELATION COEFFICIENTS FR SELECTED VAPOR-PHASE
ORGANICS IN AIR AND BREATH: UNIVERSITY OF NORTH CAROLINA
*1.
*2.
3.
4.
*5.
6.
*7.
8.
9.
10.
Air vs.
1,1, 1-Trichloroethane
1,1, 1-Trichloroethane
Chloroform
Breath vs.
Tetrachloroethylene
Air vs.
Vinyl idene chloride
Trichloroethylene
1,1, 1-Trichloroethane
1,1, 1-Trichloroethane
Benzene
Vinylidene chloride
Spearman
Rank
Air Correlation N
Vinylidene chloride
Benzene
Benzene
Breath
1,1, 1-Trichloroethane
Breath
1,1, 1-Trichloroethane
Trichloroethylene
1,1, 1-Trichloroethane
Tetrachloroethylene
Tetrachloroethylene
Tetrachloroethylene
.94
.83
.83
-.83
1.0
1.0
0.94
0.94
0.89
0.83
6
6
6
6
6
6
6
6
6
6
P
.005
.042
.042
.042
.000
.000
.005
.005
.014
.042
*Also significant (p < 0.05) at Lamar University.
71
-------
Table 36. SPEARMAN CORRELATION COEFFICIENTS FOR SELECTED VAPOR-PHASE
ORGANICS IN AIR AND BREATH: BOTH GROUPS
1.
2.
3.
4.
5.
6.
7.
8.
9:
10.
11.
12.
13.
14.
15.
16.
Air vs .
Vinylidene chloride
Benzene
Benzene
Trichloroethylene
Trichloroethylene
Breath vs.
Vinylidene chloride
Benzene
Air vs.
Vinylidene chloride
Vinylidene chloride
1,1, 1-Trichloroethane
Tetrachloroethylene
1,1, 1-Trichloroethane
Trichloroethylene
Tetrachloroethylene
Ethylene dichloride
Benzene
Spearman
Rank
Air Correlation N
1,1, 1-Trichloroethane
1,1, 1-Tri chloroethane
Vinylidene chloride
1,1, 1-Trichloroethane
Vinylidene chloride
Breath
1,1, 1-Trichloroethane
Tetrachloroethylene
Breath
1,1, 1-Trichloroethane
Vinylidene chloride
1,1, 1-Trichloroethane
Tetrachloroethylene
Vinylidene chloride
Trichloroethylene
Benzene
Benzene
1,1, 1-Trichloroethane
.86
.81
.69
.57
.49
.81
.44
.88
.77
.74
.73
.62
.53
.54
.54
.46
17
16
16
17
17
17
16
17
17
17
17
17
17
16
16
16
P
.001
.001
.002
.008
.024
.001
.045
.001
.001
.001
.001
.004
.013
.015
.016
.038
72
-------
Table 37. COMPARISON OF SPEARMAN AND PEARSON CORRELATION COEFFICIENT
FOR AIR AND BREATH VALUES--BOTH GROUPS
Correlation
Coefficient r
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Correlation
Vinylidene chloride (air) vs. 1,1,1-Trichloroethane (breath)
Vinylidene chloride (air) vs. 1,1,1-Trichloroethane (air)
Vinylidene chloride (breath) vs. 1,1,1-Trichloroethane (breath)
Benzene (air) vs. 1,1,1-Trichloroethane (air)
Vinylidene chloride (air) vs. Vinylidene chloride (breath)
1,1,1-Trichloroethane (air) vs. 1,1,1-Trichloroethane (breath)
Tetrachloroethylene (air) vs. Tetrachloroethylene (breath)
Vinylidene chloride (air) vs. Benzene (air)
Vinylidene chloride (breath) vs. 1,1,1-Trichloroethane (air)
Trichloroethylene (air) vs. 1,1,1-Trichloroethane (air)
1,2-Dichloroethane (air) vs. Benzene (breath)
Tetrachloroethylene (air) vs. Benzene (breath)
Trichloroethylene (air) vs. Trichloroethylene (breath)
Vinylidene chloride (air) vs. Trichloroethylene (air)
Benzene (air) vs. 1,1,1-Trichloroethane (breath)
Benzene (breath) vs. Tetrachloroethylene (breath)
Spearman
.88
.86
.81
.81
.77
.74
.73
.69
.62
.57
.54
.54
.53
.49
.46
.44
Pearson*
.82
.73
.81
.70
.73
.72
.82
.72
.52
.50
.59
-56
.61
NS**
.54
.42
Probability due
to Chance p
Spearman
.001
.001
.001
.001
.001
.001
.001
.002
.004
.008
.016
.015
.013
.024
.038
.045
Pearson
<.0005
<.0005
<.0005
.001
<.0005
.001
<.0005
.001
.016
.020
.006
.009
.004
HS**
.012
.048
* Calculated using logarithms of observed concentrations
**Not Significant (p > .05)
-------
Table 38. BREATH/AIR RATIOSa FOR SELECTED VOLATILE ORGANICS FOR
STUDENT VOLUNTEERS FROM TWO GEOGRAPHICAL AREAS
Participant Number
Lamar University
1
2
3
4
5
6
7
8
9
10
11
UNC
1
2
3
4
5
6
Arithmetic Mean
(± Standard Deviation)
Lamar University
UNC
Combined
Benzene
0.3
0.6
0.07
0.6
0.3
0.4
0.4
0.2
0.3
0.4
-
0.07
0.2
0.4
0.3
-
0.5
0.31 ± .18
0.30 + .17
0.31 ± .17
Chloroform
0.2
0.2b
O.lb
0.04b
0.06b
0.08b
0.5
0.05b
O.lb
0.2b
0.07b
0.4
0.6
1.4
0.6
.
0.8
0.15 ± .14
0.72 ± .38
0.33 ± .35
Viuylidene
Chloride
0.04
0.06C
0.3
0.08C
2.5
.
o.oic
0.09b
2.7
0.8
-
0.3
0.5
0.6
1.1
1.4
1.1
0.74 ± 1.1
0.83 ± .42
0.78 ± .86
1,1,1-Trichloro-
e thane
0.3
0.03
0.09
0.07b
0.07b
0.02b
0.04
0.01b
0.1
0.1
0.02b
0.1
0.2
0.2
0.4
0.1
0.2
0.08 ± .07
0.23 ± .11
0.13 ± .11
Trichloro-
ethylene
0.01°
0.2
0.03b
0.4b
0.1C
0.2b
0.3b
0.02
0.7b
0.4
0.2b
0.1
0.2
0.1
0.1
0.2
0.2
0.23 ± .21
0.17 ± .06
0.21 ± .17
Tetrachloro-
ethylene
0.4
0.6
1.4
2.4
2.4
0.2b
0.8
0.2
0.2
1.0
O.lb
1.4
1.2
2.1
7.2
1.7
0.4
0.88 ± .83
2.4 ± 2.5
1.4 ± 1.7
m-Dichloro-
benzene
O.lb
0.4
O.lb
O.lb
0.3b
O.lb
0.4
0.8
3.3
5.4
0.03b
0.02b
7.8
4.7
3.1b
3.5
1.7b
1.0 ± 1.7
3.5 ± 2.6
1.9 ± 2.4
Calculated using T - 1/2 (LOD + QL); ND - 1/2 LOD.
Breath value - trace or below quantifiable limit.
Breath value below limit of detection.
-------
For three other compounds, however, (1,1,1-trichloroethane, tetrachloro-
ethylene, and dichlorobenzene isomer) the UNC mean breath-air ratios
are several times greater than those of Lamar. Moreover, for two of
these compounds (tetrachloroethylene and dichlorobenzene isomer) the
UNC breath-air ratio is greater than 1. This anomaly is most likely a
result of the high variance and small number of subjects (N=6) at UNC.
Comparison of Lamar University Participants with University of North
Carolina Participants—Air and Breath
Differences between the Lamar and UNC geometric means for each of
the 15 compounds in both air and breath were examined for significance
by performing a t-test on the natural logs of the data. No significant
differences were found between the two study groups for any of the com-
pounds in air. However, among breath samples chloroform was higher in the
UNC group (t = 6.34, degrees of freedom (df) = 14, p<.00001). This
considerable difference between UNC and Lamar breath levels is readily
explainable as the result of the higher chloroform levels in UNC drinking
water. Other compounds with higher mean values in the UNC group were
vinylidene chloride (t=3.01, df=11.4, p<.05) and 1,1,1-trichloroethane
(t = 2.71, df = 13.4, p <.05). Caution must be exercised in evaluating
these results because of the small sample sizes and the high variability
and skewed nature of these sample distributions. For example, the mean
value of the Lamar 1,1,1-trichloroethane breath concentrations is actually
larger than the corresponding UNC mean (Lamar = 24.4 mg/nr, UNC =»
19.5), due to the occurrence of two extremely large concentrations in
the Lamar group.
-75-
-------
ON
Table 39. SUMMARY STATISTICS FOR ESTIMATED LEVELS OF SELECTED VAPOR PHASE ORGANICS
IN TAP WATER—LAMAR UNIVERSITY STUDENT STUDY
Chloroform
I , 2-Dichloroethane
1,1, 1-Trichloroethane
Bromodichlorome thane
%
Detected
100
0
0
100
% at
Trace
0
0
0
0
a
Mean
172.38
0.30
0.10
21.21
Standard
Deviation
126.56
0
0
8.38
Median
130.00
0.30
0.10
20.33
Range
117.00-550.00
0.30-0.30
0.10-0.10
13.33-44.00
and/or Carbon
Tetrachloride
Trichloroethylene
and/or 1,1,2-Tri-
chloroethane
Tetrachloroethylene
Chlorobenzene
9
0
9
0
0.95
0.30
1.33
0
0.55
0.30
0.55-4.95
0.30-0.30
All compounds are measured in ng/mL.
-------
Table 40. SUMMARY STATISTICS FOR ESTIMATED LEVELS OF
SELECTED ORGANICS IN TAP WATER—CHAPEL HILL
Chloroform
1,2-Dichloroethane
1,1, 1-Trichloroethane
Bromodichlorome thane
Trichloroethylene
Tetrachloroethylene
Chlorobenzene
Detected
100
0
0
100
23
100
23
% at
Trace
0
0
0
0
0
0
0
Mean '
220
-
-
17
0.6
2.0
0.4
Standard
Deviation
23
-
-
2
1
0.8
0.6
Median
220
-
-
17
ND
1.8
ND
Range
180-260
ND
ND
15-20
ND-3.0
1.3-3.8
ND-1.5
All compounds are measured in ng/ml.
3Mean of all values (ND = 2.00).
-------
Summary Statistics — Drinking Water
Summary statistics for the Lamar and UNC drinking water data are given
in Tables 39 and 40. For each participant with more than one water sample,
mean values were calculated for each compound and replace the raw values.
Each statistic is computed using the actual or estimated values for all
participants except where data are missing. The "percent detected" column
indicates the percentage of measurements greater than the limit of detec-
tion (including trace amount). The medium and the arithmetic mean are
provided as measures of central tendency. In general, the sample distri-
butions for water show much less variability and skewness than for air
and breath.
Spearman coefficients were computed for the air-water and water-breath
concentrations of the target compounds; no significant correlations were
observed.
Estimated Total Daily Intake. Only two of the compounds measured in tap
water samples contributed significantly to the total daily intake from
air and drinking water of the volunteers (Table 41). Assuming daily in-
takes of 10 cubic meters of air and one liter of water, drinking water
accounted for 79 percent of the chloroform intake and 76 percent of the
bromodichloromethane intake. By contrast, drinking water contributed
only 7 percent of the daily intake of tetrachloroethylene for the UNC
students, and even less for the Lamar students.
The estimated daily intake of all volatile organic compounds from
air and drinking water is listed in Table 42 for each subject. The air
values range from 0.3-12.4 mg/day, with a geometric mean of about 1.6
-78-
-------
Table 41. ESTIMATED DAILY INTAKE* OF SELECTED COMPOUNDS FROM WATER COMPARED TO AIR (ug/day)
Chloroform
Subject
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Air
13.9
15.2
28.3
83.3
52.2
40.4
48.4
60.2
31.7
18.6
48.2
78.1
50.9
37.4
31.9
175
22.5
Water
185
130
550
130
130
117
133
143
133
125
120
255
235
215
210
210
190
Percent
Intake
from Air
7
10
5
39
29
26
27
29
19
13
29
23
18
15
13
45
11
Bromodichloromethane
Air
16.3
15.2
ND**
32.1
10.0
ND
11.4
37.1
8.4
ND
ND
ND
ND
ND
ND
43.6
ND
Water
17
23
44
22
24
20
22
19
17
13
13
20
18
.18
17
16
15
Percent
Intake
from Air
49
40
<3
59
29
<6
34
66
33
<9
<9
<6
<7
<7
<7
73
<8
Tetrachloroethylene
Air
1,750
1,620
72.1
54.1
56.5
49.8
301
7,180
>45.4
1,720
92.6
24.8
26.5
20.6
12.2
43.2
1,270
Water
ND***
ND
T
ND
ND
0.07
ND
0.03
ND
0.01
0.03
3.8
1.8
1.8
1.8
1.8
1.3
Percent
Intake
from Air
100
100
95
>98
>98
100
100
100
>98
100
100
86
94
92
87
96
100
* Assuming 10 m3/day respiration rate and 1 I/day ingestion rate
** ND = < 1.2 ug/10m3
***ND = < 1.1 ug/1
-------
Table 42. ESTIMATED DAILY INTAKE* OF 10 VOLATILE
ORGANIC COMPOUNDS THROUGH AIR AND WATER FOR 17 SUBJECTS
Subject
30001
30002
30003
30004
30005
30011
30012
30013
30014
30015
30016
40001
40002
40003
40011
40012
40013
Air
ug/day
12,400
2,700
12,000
400
300
550
1,250
9,000
300
2,600
1,140**
2,470
2,390
1,250
300
1,240
3,800
Water
200
150
600
150
150
140
160
160
150
140
130
280
260
240
240
230
210
Total
12,600
2,850
12,600
550
450
690
1,410
9,160
450
2,740
1,270
2,750
2,650
1,490
540
1,470
4,010
Percent
from
Air
98
95
95
73
67
80
89
98
67
95
90
90
90
84
56
84
95
*Assuming 10 ms/day and
air and water.
**0mitting one outlier.
1 liter/day intake rates for
80
-------
mg/day and a geometric standard deviation of about 3.5. The correspon-
ding geometric mean -for the water intake is 0.2 mg/day. Thus air is the
major contributor to total daily intake of the volatile organics measured.
Summary.
This report documents the first field effort of a continuing exposure
monitoring program at the United States Environmental Protection Agency.
Sampling equipment and analytical protocols were tested on 17 subjects
at two universities. The sampling equipment (personal monitors and a
specially designed spirometer) and the analytical protocols worked well
for the air, breath, and tap water samples. However, difficulties were
encountered with the purge and trap analytical .protocol for blood and
urine. These difficulties are being investigated further.
The results for air, breath, and tap water indicate that the concept
of making direct measurement of individual human exposure to a significant
number of volatile organic compounds is feasible. This first effort has
resulted in a number of findings, including particularly the wide range
of exposures among a homogenous group of subjects, and the apparent direct
relationship between the logarithms of the amounts of certain compounds
inhaled and exhaled. These findings could not have been made using stan-
dard approaches of ambient monitoring.
-81-
-------
REFERENCES
1. E.D. Pellizzari, M. D. Erickson, and R. Zweidinger, Formulation
of a Preliminary Assessment of Halogenated Organic Compounds
in Man and Environmental Media (USEPA, Wash. 1979), pp. 143-163.
2. R.A. Zweidinger &t_ J^L, Measurement of Benzene Body Burden of Po-
tentially Environmentally Exposed Individuals (USEPA, Wash. D.C.,
1980).
3. L.A. Wallace, in Environmental Monitoring: Supplement (Vol. IV-A)
of Analytical Studies for the U.S. Environmental Protection Agency)
(National Academy of Sciences, Wash. B.C., 1977); L.A. Wallace, in
Conference Proceedings: 4th Joint Conference on Sensing of Environ-
mental Pollutants (Amer. Chem. Soc., Wash. B.C., 1978) p. 390; L.A.
Wallace, in Proceedings of the Symposium on the Development and
Usage of Personal Monitors for Exposure and Health Effect Studies,
ed. D.T. Mage and L.A. Wallace (USEPA, Research Triangle Park, NC,
1979) p. 7.
4. E.D. Pellizzari, The Measurement of Carcinogenic Vapors in Ambient
Atmospheres (USEPA, Research Triangle Park, NC 1977).
5. E.D. Pellizzari, Analysis of Organic Air Pollutants by Gas Chromato-
graphy and Mass Spectroscopy (USEPA, Research Triangle Park, NC 1979).
6. E.D. Pellizzari and J.E. Bunch, Ambient Air Carcinogenic Vapors;
Improved Sampling and Analytical Techniques and Field Studies,
(USEPA, Research Triangle Park, NC 1979).
7. E.D. Pellizzari &t_ al^, Preliminary Study on Toxic Chemicals in
Environmental and Human Samples, Parts I and II (USEPA, Wash. B.C.,
1980).
8. A.J. Peoples et_ al, Bull. Env. Cont. Tox. 23:244, 1979.
9. B.K. Krotozynski, G. Bruneau, and H.J. O'Neill, J. Anal. Tox.
3, 225-34, 1979.
10. Tables of RMRS can be consulted in Pellizzari (note 5 above).
11. Recovery efficiencies are tabulated in Pellizzari (note 7 above).
12. A.J. Peoples, University of Miami, personal communication.
-82-
-------
APPENDIX
DATA COLLECTION INSTRUMENTS USED AT LAMAR UNIVERSITY
-88-
-------
LAHAR UNIVERSITY
BLOOD TEST QUESTIONNAIRE
DATE:
NAME
NUMBER Or YEARS YOU HAVE LIVED IN THE SEAUMOKT AR£A_
AGE DATE OF BIRTH
RACE SEX
DO YOU LIVE ON THE UNIVERSITY CAMPUS?
IF YES, HCW LONG?
IF YOU DO NOT LIVE ON THE UNIVERSITY CAMPUS, GIVE TOUR HOME ADDRESS
CO YOU CONSIDER YOURSELF ATHLETIC?_
IF YES, WHAT SPORT ACTIVITY?
HOW FREQUENTLY DO YOU WORKOUT?
CO YOU WORK WITH CHEMICALS?
AROUND A53CRAFTS?
AROUND ORY CLEANERS?
AROUND DEGREASING SOLVENTS (AS IN AUTO SHCP)?_
DO YOU OR1KX DECAFFEINATED COFFEE REGULARLY?
Date OT For?,: 2/21/30
84
-------
UNIVERSITY OF MIAMI
MIAMI, FLORIDA 33177
DEPARTMENT OF EPIDEMIOLOGY DIVISION OF CHEMICAL EPIDEMIOLOGY
AND PUBLIC HEALTH 1565S S,W. »27th AVENUE
SCHOOL OF MEDICINE MIAMI, FLORIDA 3X177
(305) 2S5-J300
INFORMED CONSENT - DRINKING WATER STUDY
All water contains a number of chemicals. Most chemicals appear by nature
depending on the geographical location of the water source. Others are introduced
as a result of various water purification and treatment methods.
The University of Miami School of Medicine's Department of Epidemiology and
Public Health is conducting a study to determine if the chemicals which are present
in Dade County's water are also present in the blood of the residents of the county
who drink that water.
You can help in this research study by allowing us to draw a small amount of
blood (20 cc., the amount usually collected for other blood tests) which will be
analyzed for i ts chemical content.
There is no adverse effect on the body and the only discomfort which you Bight
feel is the needle pride. Occassionally a small bruise mark may be noted which will
disappear in a .short time. The blood will be drawn by qualified personnel using
sterile equipment.
We will be glad to answer any questions you may have regarding this research
project.
To participate, please sign the form below:
I, (Name)
(Address)
agree to. participate in the University of Miami Drinking Water Research Project and
authorize to withdraw approximately 20 cc.
of blood for chemical analysis.
(Witness) (Date)
Date of Form - 7/20/79
RECEIPT
I have received a payment of $10.00 (cash) for participating in the University of
Miami's Drinking Water Research Project (EPA Grant No. R806833-01).
NAME
ADDRESS
SOCIAL SEC. NO,
DATE PAYEE
A prjvirt, ind«p«nd«oc, inltnutionaj umVtro'tjr
An «qu*l opportunify/a/nrmacrwi Action cmpjoywr
85
-------
Interview Date
Class
Mo. Day Year
12-13 14-15 16-17
Interviewer
18
I.D.5?
Card =•
1 2345 67
Control I.D.S
Information obtained from 1. personal
,L
9 10 11
2. relative or friend only
3. records only
4. both other person, records
Same
_19. Sex 1. M. 2.F
20-21
Ag«:
Marital status l.s
22
_2.M
3,W 4.D
-5-Sep..
23 24
Total Number in
Household:
Age:
Age:
Age:_
_.\ge:
25
26
Race: 1. White :_
5.Other:
>. Black: 3. Hispanic: 4.Oriental:
(Specify!.
Education:
1. 0-€j I. 7-9: 3. 10-12: 4. College: 1-2 :_
5. 3-4: 6. Graduate school: years 9. Or.known_
Estimate Family Income:
JJumber of contributors:
1.Under 55,000:
4.515,000 - $19,999:.
2.55,000 - 59,999: 3. 510,000 - 514,999_
5. 520,000 or above: 9.
86
-------
DR32TKTSG WATER QUESTIONNAIRE -2-
Addresses: (Present and last 5 years)
Prom To
Location City, State Mo. Yr. Mo. Yr
1.
7.
3.
4.
5.
|
29
Residence in Hater area: 1. Less than 5 years
2. 5 or more years
Code for 30-34:
(1) None, (2). Leas than half, (3) Half or more, (4) Ml, (9) Unknown
Source of water supply used in drinking and cooking:
Municipal Supply: Name:
Address:_
Well Water: Address:
nName (supplier)
Bottle Water: ' Address:
32
Other (specify): vi dress :_
Do you use any method of water treatment?
34 Tyoe: 1. Water softener: 1. Filters: 3. Other:
4. None
Comments:
Any comments on tapwater: (clarity, tasta, odor, color,- sediment,
pressure)
Do you have a pool?
-5 (1) yes, svrim regularly (2) yes, don't use pool much (3) no, but
use other pools regularly (4) no, but occasional pool use,
(5) no pool, rarely use others (9) no information
•3
nDo you use a pest-control service on a regular basis?
(1)*never, (2) in yard only, (3) in house only, (4) in both yard
36 and house, (9) unknown
Name and address of service:
87
-------
How many meals do yc-j eat each day? At home
Elsewher*
37.
Fluid Intake; (Approximate' daily amounts in ounces)
Amount (oi) Amount (oz) in
Type of Fluid _J Last 24 Bours —
1. lap Water
2. ' Coffee
3. Tea
4. Soft Drinks
5. MilX (Cocoa)
6. Soups, "broths, etc.
38
Comments
7. Juices (Fruit.
vegetable)
8. Bottled or Boiled Water
9. Beer
10. Wine
11. Spirits
Approximate total
Daily Fluid Intake
39 40 41 42 "A3 44
Do you drink decaffeinated coffee regularly? l="yes, 2=no
a
45
SinoXino Habits:
Have you usec:
Code:"
1 = use now
2 = formerly,
not now
3 = never
Cioarettes ;| )
48
How 1one
(years)
I* not now, wher
(last: vear)
Cigars :
Pipe :
Chewing
Tobacco:
Snuff:
nu
53
| j
58
{ I
63
CZI
66
49 50
54 55
59 60
64 65
51
52
56
57
61
62
5s 70
71
72
88
-------
AT2?. QUESTICNNAISS -4-
Cicarette
Frequency:
7.3
(1) Never, (2) Less than 1/2 sack/day (3) More than 1/2 sack
per day, no more than 1 pack (4) Between 1 and 2 packs, (S) More
than 2 packs (6) Smoker, no information on frequency (9) no information
Occu'oationa.I Eistorv;
Employer Position From To
Address Type of Work Mo. Yr. Mo, Yr.
SELF:
Present:
Past Occupations:
2.
3.
4.
5.
SPOUSE;.
Present:
1.
Past Occupations:
2.
3.
4.
5.
Chemical exposure: (occupation)
1 (1) self presently occupationally exposed (2} previously (3) never
I (9) unknown
a(l) spouse or other family member presently occupatior.ally exposed
(2) previously, (3) never (9) unknown
89
-------
DRINKING WATER QUESTIONNAIRE
Exposure to Chemicals:
Ever been exposed to:
1 = yes
2 » no
3 = doesn't know
L_J chloroform
8
j) carbontetrachloride
11 •
Otrichloroethylene
| | tetrachloroethylene
17
dry-cleaning solvents
other volatile halocarbons
Type Code:
1 = acute
2 = chronic
3 = sporadic
D
9
12
n
is
D
21
g
D
i
IDfl
2345
Card *
EEQ
67
Tine Code:
1 = within last
24 hours
2 » within last
72 hours
g
g
n
16
a
22
[J halathane
26
(_J other anesthetic
29
|1 ethylene dichloride
32
I | pesticides and sprays
D
27
D
30
D
28
D
31
d
33
g
g
a
paint thinners
I I degreasers
41
n
40
n
43
90
-------
-6-
Exposure to Chemicals:
Ever been exposed to:
1 =
2 =
3 =
44
n
47
u
SO
3
56
3
59
u
52
3
3
D
71
74
77
!0
yes
no
doesn't know
gasoline, naphtha
chlorox, purex bleach
other chlorine-containing bleaches
chlorine pool chemicals , HC1 •
cyanuric acid (Duochlorin)
fingernail polish
fingernail polish remover
hair spray
other aerosols
liquid paper
cough syrup
toothpaste
other
Type Code:
1 = acute
2 * chronic
3 » sporadic
g
D
48
D
51
g
g
D
60
D
63
g
D
69
D
72
g
D
78
D
81
Time Code:
1 « within last
24 hours
2 » within last
72 hours
g
a
49
a
52
g
g
a
61
a
64
g
g
a
73
g
a
79
a
32
91
-------
DRJSKDJG WATER QUESTIONNAIRE
ID a
Health History;
(put circled number in box)
1 Present Health Status: 1. excellent - no health problems
_____ 2- Generally good - no complaints 3. generally good -
8 acute problem (surgery) 4. poor health 5. autopsy
Height:
Present weight:
Ibs.
14 IS 16
9 10 11 12 13
Recent (last 6 months) weight change:
Gain (amount) Loss (amount)
Reason: (1) illness (2) planned (exercise, special diet)
(3) unplanned
Are you presently taking any medications? yes(l).
_No(2).
18
If yes, please complete:
Kind (Name, if Known)
Prescribed
Condition
Amount per day,
week, etc.
1.
2.
3.
4.
5.
6.
Past Health History: (please put number in box)
19
19. When did you see a physician last? (1) less than a week
' ago (2) 1-4 weeks (3) 1-5 months (4) 6-17 months
(5) at least 18 months (9) unknown
20. Reason:
21.
(1)
(3)
acute problem, current (2)
acute problem,
past
chronic problem (4) physical exam, irrouni
sations, etc.
(9) unknown
Have you been hospitalized?
If Yes, reason:
(1) Yes (2) No
(9) unknown
92
-------
DRINKING WATER QUESTIONNAIRE
Have you ever required a doctor's care for the following?
Duration of Ill-
Condition Code: (Dyes (2) no When(yr.) ness (days)
(9) Unknown
22. Eyes I I
23. Ears, Nose, Throat f I
24. tipper Respiratory I |
25. Gastro-intestinal. f I
26. Hypertension I I
Z7. Circulatory Problem [ I
28. Heart Condition f I
29. Stroke f I
30. Endocrine (Diabetes, etc). [ |
31. Muscular-skeletal problem I I
32. Genito-urinary I I
33. (Females) .
Obstetrical I I
34 Gynecological
35. Neurological
36. Nervousness; Emotional
37. Tumors, Growths
38. Allergies
39. Skin problems
(dermatitis)
40. Dental problems I I
41. Liver (hepatic) problems I I
Comments:
Serua Chloroform Level | | ) ) )
42 43 4445
aU.S. GOVERNMENT PRINTING OFFICE: 1983/659-095/1954
93
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
tage and
ees Paid
nvironmental
rotection
Age'
EPA-335
Official Business
Penalty for Private Use, S300
Special Fourth-Class Rate
Book
in
i — i
o
i
(N
co
o
o
w
Please make all necessary changes on the above label.
detach or copy, and return to the address in the upper
left-hand corner
If you do not wish to receive these reports CHECK HERE c.
detach, or copy this cover, and return to the address in the
upper left-hand corner
EPA-600A-82-015
------- |