United States Office of Mobile Source Air Pollution Control EPA-460/3-81-029
Environmental Protection Emission Control Technology Division September 1981
Agency 2565 Plymouth Road
Ann Arbor, Michigan 48105
Air
c/EPA
Nitrosamines In
Vehicle
Interiors
-------
EPA-460/3-81-029
Nitrosamines In Vehicle Interiors
by
Lawrence R. Smith
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
Contract No. 68-03-2884
Task Specification 2
EPA Project Officer: Robert J. Garbe
Task Branch Technical Representative: Thomas M. Baines
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
2565 Plymouth Road
Ann Arbor, Michigan 48105
September 1981
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from
the Library Services Office (MD-33), Research Triangle Park, North
Carolina 27711; or, for a fee, from the National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by
Southwest Research Institute, 6220 Culebra Road, San Antonio, Texas,
in fulfillment of Task Specification 2 of Contract No. 68-03-2884.
The contents of this report are reproduced herein as received from
Southwest Research Institute. The opinions, findings, and conclusions
expressed are those of the author and not necessarily those of the
Environmental Protection Agency. Mention of company or product names
is not to be considered as an endorsement by the Environmental Protec-
tion Agency.
Publication No. EPA-460/3-81-029
ii
-------
FOREWORD
This project was conducted for the U. S. Environmental Protection
Agency, 2565 Plymouth Road, Ann Arbor, Michigan 48105, by the Department
of Emissions Research of Southwest Research Institute, 6220 Culebra Road,
San Antonio, Texas 78284. Nitrosamine analyses in the project were con-
ducted by the New England Institute for Life Sciences, 125 Second Avenue,
Waltham, Massachusetts 02154. This project, authorized by Contract
68-03-2884, Task 2, was initiated on June 5, 1980 and completed May 5,
1981. The EPA Project Officer was Mr. Robert J. Garbe and the EPA Task
Technical Officer was Mr. Thomas M. Baines, both of the Emission Control
Technology Division, Environmental Protection Agency. Ms. Laurie Gallagher
of the ECTD, EPA provided major assistance in vehicle acquisition and
sample collection in Ann Arbor, Michigan. Dr. David Fine of the New
England Institute for Life Sciences was in charge of the sample analyses
for the project. The SwRI Project Leader and principal researcher for
the project was Dr. Lawrence R. Smith. The SwRI Task Technical Super-
visor was Harry E. Dietzmann, and the SwRI Project Manager was Charles T.
Hare. Key technical personnel involved at SwRI included D. R. Terrazas
and J. C. Chessher. This project was identified within Southwest
Research Institute as Project No. 11-5830-002.
iii
-------
ABSTRACT
Researchers in the nitrosamine field were contacted on their
views of the TEA analyzer and ThermoSorb/N Air Samplers for nitrosamine
analysis. Gas samples were taken from vehicle interiors to determine
the effects of vehicle type, vehicle age, mode of operation, and ambient
conditions on interior nitrosamine levels. A total of fifty-eight
vehicles were sampled in the program. Occupant exposure levels were
estimated using test vehicle data.
iv
-------
SUMMARY
The objectives of this project were to determine whether or not a
nitrosamine analytical procedure which has the consensus backing of the
scientific community exists, and then to provide gas samples from vehicle
interiors to be analyzed for nitrosamines. Various test vehicles arid
test sequences were selected to determine influences of vehicle type,
vehicle age, mode of operation, and ambient conditions on interior
nitrosamine levels and to determine occupant exposure levels.
The response from the scientific community for the most part indi-
cated an acceptance of the TEA analyzer and ThermoSorb/N traps. Re-
searchers using both the TEA analyzer and ThermoSorb/N traps felt that
both are the best available for routine analysis work.
Of the fifty-six vehicles sampled in the program by SwRI, detec-
table levels of nitrosamines were found in forty-seven. Concentrations
of detectable nitrosamines in vehicle interiors ranged from 0.01 to
0.63 Ug/m3. N-nitrosodimethylamine (NDMA) was found in all forty-seven
vehicles with detectable levels of NDMA at concentrations ranging from
0.01 to 0.39 yg/m3. N-nitrosodiethylamine (NDEA), N-nitrosomorpholine
(NMOR), N-nitrosodipropylamine (NDPA), and N-nitrosodibutylamine (NDBA)
were found in 12 or fewer vehicles at levels ranging from 0.01 to 0.16
yg/m3. Both NDMA (ND-0.24 yg/m3) and NMOR (ND-0.360 yg/m3) were found
in two vehicles sampled by the New England Institute for Life Sciences.
Test results indicated NDMA dependence on mileage and time, with
NDMA levels decreasing slightly with time and mileage; NDMA dependence
on temperature, with higher temperatures giving higher levels of NDMA;
the presence of detectable levels of NDMA in vehicles during operation;
and similar levels of NDMA in vehicles closed overnight versus vehicles
closed for only a short time. Nitrosamines were found in passenger cars,
station-wagons, passenger and cargo vans, pickup trucks, and in new and
used heavy-duty trucks. Nitrosamines were not detected in motor homes.
On the average, the daily intake of nitrosamines from vehicle interiors
is estimated to be less than that from a can of beer or from a strip of
bacon.
v
-------
TABLE OF CONTENTS
Page
FOREWORD iii
ABSTRACT iv
SUMMARY V
LIST OF FIGURES ix
LIST OF TABLES X
I. INTRODUCTION 1
A. Project Objectives 1
B. Approach and Scope 1
II. GENERAL EQUIPMENT AND TEST PROCEDURES 3
A. ThermoSorb/N Air Sampler 3
B. Sample Pumps 4
C. Nitrosamine Analysis: GC-TEA 5
D. Confirmational LC-TEA Analysis 5
E. Confirmational Mass Spectral Analysis 5
F. Collection and Analysis Procedure for Determining
Nitrosation Capacity 6
G. Analysis of Air Samples for Amines 6
H. Test Sequence to Determine Influence of Vehicle Age
at SwRI - Task II 7
I. Test Sequence to Determine Influence of Vehicle Age
at NEILS - Task II 7
J. Test Sequence to Determine Influence of Operation -
Task III 8
K. Test Sequence to Determine Influence of Ambient
Conditions - Task IV 10
L. Test Sequence for Additional Assessment - Task V 11
III. Results 13
A. Task I - Sampling and Analysis Methodology 13
B. Task II - Influence of Vehicle Age 23
C. Task III - Influence of Operation 28
D. Task IV - Influence of Ambient Conditions 33
E. Task V - Additional Assessment 36
F. Task VI - Exposure Assessment 43
vii
-------
TABLE OF CONTENTS (Cont'd)
Page
IV. DISCUSSION OF THE RESULTS 45
REFERENCES 47
APPENDICES
A. "Instructions for Monitoring"
B. Description of the GC-MS System
(as Reported to SwRI by NEILS)
C. Description of the Nitrosation Capacity
Determination (as Reported to SwRI by NEILS)
D. Description of the Nitrosatable Amine
Determination (as Reported to SwRI by NEILS)
E. Task II Trunk and Cargo/Back Seat Sampling Results
viii
-------
LIST OF FIGURES
Figure
1
2
3
4
5
6
7
Page
ThermoSorb/N Air Sampler 3
P-2500 and P-4000 Sample Pumps Connected to
ThermoSorb/N Air Samplers 4
Trap Being Clamped to Sun Visor in Heavy-Duty Truck 12
Plot of NDMA Concentration in Ford Pinto and Datsun
310 Versus Time 27
Plot of NDMA Concentration in Ford Pinto and Datsun
310 Versus Distance Driven 27
NDMA Concentration as a Function of Temperature:
Toyoto Corolla, Chevrolet Citation, Datsun 310 GX,
Ford Mustang (Autom. Trans.), Dodge St. Regis 38
NDMA Concentration as a Function of Temperature:
Mercury Marquis, Dodge Diplomat, Chrysler Newport,
Ford Mustang (Man. Trans.) 38
ix
-------
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
14
18
21
24
24
25
25
26
28
28
29
31
34
35
37
39
LIST OF TABLES
List of Researchers
Nitrosation Potential Analyses
Amine Analyses
Background and Blank Nitrosamine Samples
Mass Spectral Analyses
Description of Vehicles Sampled in SwRI Task II
Screening Process
Nitrosamine Sampling Results for SwRI Task II
Screening Process
Test Results for Nitrosamine Sampling - Task II (SwRI)
Description of Vehicles Sampled in NEILS Task II
Test Results for Nitrosamine Sampling - Task II (NEILS)
Vehicles Sampled for Influence of Operation
Test Results for Nitrosamine Sampling-Task III
Vehicles Sampled at Ambient Temperature (72-73°F)
Test Results for Sampling at 72°F-Task IV
Test Results for Sampling at 40, 72 and 100°F
Vehicles Sampled for Additional Assessment
Test Results for Nitrosamine Sampling-Task V
x
-------
I. INTRODUCTION
The presence of nitrosamines in car interiors was first determined
by the New England Institute for Life Sciences. Nitrosamines were found
in 37 of 38 new 1979 cars at concentrations ranging from 0.07 to 2.9
Ug/m3.(If2)* The estimated exposure to man was reported to be similar
to that from bacon and beer. This report describes an effort to further
assess the magnitude and scope of the problem of nitrosamines in all
types of vehicle interiors, as well as evaluate some of the variables
influencing nitrosamine levels.
A. Project Objectives
The objectives of this project were to determine whether or not a
nitrosamine analytical procedure which has the consensus backing of the
scientific community exists, and then provide samples from a variety of
vehicle interiors to be analyzed for nitrosamines. A variety of vehicles
and test situations were selected to determine influences of vehicle age,
mode of operation, and ambient conditions on interior nitrosamine concen-
trations. Occupant exposures were to be estimated using the resulting
data.
B. Approach and Scope
To effectively accomplish the project objectives, the project was
carried out in six tasks. The first task consisted of a peer review of
the sampling and analytical methodology used by the New England Institute
for Life Sciences (NEILS) to determine nitrosamine levels in car interiors.
If this methodology was found acceptable to the scientific community, it
would be used in the program for sample collection and analysis. Due to
the relatively brief duration of the project and the relatively long set-
up time needed for the method, samples were to be analyzed by NEILS if
the methodology was found acceptable.
Also, during each of the following tasks, several samples were taken
to determine: 1) level of nitrosation potential of the air, 2) level of
nitrosatable amines in the air, and 3) level of nitrosamines in ambient
air. These samples were collected using methodology developed at NEILS.
After collection at SwRI, samples were sent to NEILS for analysis.
Additional nitrosation potential information was obtained by monitoring
the N0X levels in selected circumstances with a TECO chemiluminescent
N0X analyzer. Additional nitrosatable amine information was obtained
*Numbers in parentheses designate references at end of report.
-------
by collecting selected organic amines sampled in dilute sulfuric acid,
and analyzing the samples with a gas chromatograph (GC) equipped with a
nitrogen phosphorus detector (NPD) and an ascarite precolumn.
In the second task, four vehicles {two at Southwest Research Institute
and two at NEILS) were monitored once a month over a six month period
to determine the effects of time on interior nitrosamine levels. Mileage
was also recorded as an additional variable at SwRI. To determine the
influence of operation, twelve vehicles were sampled before, during and
after operation in the third task. One vehicle was sampled: 1) imme-
diately before operation, 2) at four sequential times during operation,
traps sampled at 0, 1, 5, and 10 minutes into the trip, and 3) immediately
after the trip. In the fourth task, EPA assisted SwRI by obtaining 15 new
vehicles in Ann Arbor. These vehicles were sampled at ambient temperature
at the EPA-Ann Arbor lab by SwRI. Nine of the vehicles with the highest
nitrosamine levels were selected for additional testing, and were sampled
at 40° and 100°F in EPA's cold room. Twenty-five vehicles were sampled
in the fifth task as an additional assessment to increase the data base
for interior nitrosamine concentrations. Vehicles sampled in this task
included motor homes, new and used heavy-duty trucks, new vans, pickup
trucks, and station wagons. In the sixth task, an exposure assessment
was carried out for several population categories using the data generated
in the previous tasks. The following situations were investigated;
1) continuous exposure, 2) worst case commuter exposure, and 3) typical
case commuter exposure.
2
-------
II. GENERAL EQUIPMENT AND TEST PROCEDURES
This section contains a description of the test equipment and pro-
cedures used to collect the nitrosamine samples as well as a brief re-
view of the analytical procedures used by NEILS to analyze for nitros-
amine, nitrosation potential and nitrosatable amines.
A. ThermoSorb/N Air Sampler
All nitrosamine samples were collected using ThermoSorb/N Air
Samplers which were developed by the Thermo Electron Corporation. The
air sampler cartridge is shown in Figure 1.
The ThermoSorb/N Air Sampler cartridge has been reported in the literature
to be free of artifact formation and capable of retaining 100% of preloaded
nitrosamines.(3)
The air sample is pulled through the cartridge with the use of a battery
operated sample pump. The cartridges contain a solid sorbent of metal sili-
cates, which have been activated in a reducing atmosphere, for the collection
of nitrosamines; as well as a nitrosation inhibitor and an amine trap to pre-
vent artifact formation in the cartridge. The cartridges are constructed
of medical-grade polyethylene with 100 mesh stainless steel screens at the in-
let and outlet, and with standard Luer fittings to facilitate solvent elution
Figure 1. ThermoSorb/N air sampler.
3
-------
of the trapped nitrosamines. During testing, flow rates of 2 to 4 liters/
minute were used to collect the samples. The cartridges are reported to
efficiently collect nitrosamines at these flow rates. The cartridges were
used and handled as directed by Thermo Electron in their "Instructions for
Monitoring" (Appendix A) .
B. Sample Pumps
The sample pumps used in this program were DuPont Constant Flow
Samplers (Models P-2500 and P-4000). These pumps are battery powered,
and are capable of moving a constant volume flow rate of air through a
collection device. Both models possess an automatic flow control system
which maintains a constant air flow rate within ±5% over pressure drop
changes up to 15 inches of water column. These automatic flow control
systems are necessary when sampling the interior of a vehicle for several
hours with all windows and doors closed. The P-2500 has a sampling range
of 1000 to 25000 cc/min and the P-4000 has a sampling range of 500 to
4000 cc/min. Both can operate for eight hours before recharging. Figure
2 shows the P-2500 and the P-4000 sample pumps connected to ThermoForb/N
Air Samplers. The pumps were calibrated before and after testing with a
ThermoForb/N Air Sampler attached. A stopwatch and a bubble tower were
used to determine the air flow.
Figure P-2500 and P-4000 sample pumps
connected to ThermoSorb/N Air Samplers
4
-------
C. Nitrosamine Analysis: GC-TEA
When received at NEILS, the ThermoSorb/N Air Samplers were immediately
backflushed with a 25/75 solution of methanol in dichloromethane at a rate
of 0.5 mil/min. The first 1.5 to 1.8 mil of this eluate was saved for chro-
matographic analysis. This elution step has been previously described
in the literature.(3)
Two analyzers were used at NEILS for the GC-TEA analysis of the
nitrosamine samples. The first GC-TEA analyzer system consisted of a
Shimadzu Mini 2 GC with a temperature programmer interfaced to a TEA
analyzer (Model 502). The TEA signal was recorded using a Hewlett-
Packard intergrating recorder (Model 3308A). The GC column consisted of
a 14' x 1/8" stainless steel column packed with 10% Carbowax 20M containing
0.5% KOH on Chromosorb HP (80-100 mesh). The column was operated at 150°C
for isothermal separations, or programmed at 6°/min from 130° to 200°C.
Argon gas (flow rate 15 m£/min) was used as the carrier gas. The TEA
analyzer was operated with a specially designed sorbent dry trap instead
of the usual cryogenic. For maximum sensitivity 15 of each sample was
injected into the GC rather than the usual 4-5 y£. The second analysis
of the Task II samples were analyzed with this system.
The second GC-TEA analyzer system used for nitrosamine analysis at
NEILS consisted of a Hewlett-Packard Model 5710A with a Model 7671 Auto-
sampler interfaced to a TEA Model 543 analyzer. The TEA signal was
recorded using a Spectra-Physics 4060 Printer Plotter/SP4000 CPU system.
The GC column consisted of a 1/4" x 6' glass column packed with 10%
Carbowax 20M containing 0.5% KOH on Chromosorb WHP (80-100 mesh). The
column was operated with a temperature program from 140°C to 170°C at
4°C per minute. Argon was the carrier gas with a flow rate of 20 mVmin.
Sample injection volumes ranged from 4.5 to 10
D. Confirmational LC-TEA Analysis
The LC-TEA system used to confirm the GC-TEA analyses consisted of a
Waters 6000A solvent delivery system with a U6K Universal injector inter-
faced to a TEA Model 502 Analyzer operating in the HPLC mode. The signal
was recorded using a Soltec dual-pen recorder. The column used was a
10 ]_im Lichosorb Si 60 (4.6 x 250 mm). The mobile phase was 7% acetone in
isooctane, and a flow rate of 2.0 cc per minute was maintained.
E. Confirmational Mass Spectral Analysis
To confirm the GC- and LC-TEA analyses, gas chromatography-high
resolution mass spectral analyses (GC-MS) were performed. Due to the
small quantities of nitrosamines collected in individual ThermoSorb
air samplers, it was necessary to combine several samples in order to
obtain sufficient quantities of nitrosamines for mass spectral analyses.
Three composite samples were obtained in this manner.
5
-------
The composite samples were diluted ten-fold with distilled water
and extracted three times with dichloromethane (DCM), The DCM extracts
were dried over anhydrous sodium sulfate and concentrated to one milli-
liter using a Kuderna-Danish evaporator at 54°C. The concentrated ex-
tracts were re-examined by GC-TEA before mass spectral analysis.
The GC-MS system used to analyze the composite samples consisted of
a glass capillary column connected directly to the ion source of the mass
spectrometer. The glass capillary column was 20 m x 0.25 mm I.D. and
coated with Carbowax 20M. The mass spectrometer was an AEl MS50 instrument
operated at a resolution of 10,000 (10% valley). A detailed description
of the mass spectrometer can be found in Appendix B.
F. Collection and Analysis Procedure for Determining Nitrosation
Capacity
The nitrosation potential of the ambient air was determined in the
program by the use of sample cartridges similar to the ThermoSorb/N air
samplers used for nitrosamine collection. The cartridges were filled
with a solid sorbent coated with morpholine. Air samples were pulled
through the cartridges at a flow rate of 1 liter per minute for 50 minutes.
Any NC>2 present in the air sample reacts with the morpholine in the car-
tridge to produce N-nitrosomorpholine. The amount of N-nitrosomorpholine
present in the cartridge has been found by NEILS to be proportional to
the concentration of N02 in the sampled air. The analysis of the N-nitro-
somorpholine in the cartridges was conducted by GC-TEA as if it were from
a ThermoSorb/N air sampler. A more detailed description of the nitrosation
capacity determination can be found in Appendix C. This description of
the analysis procedure was provided by NEILS to SwRI with sample analyses.
G. Analysis of Air Samples for Amines
The amine air samples were collected using ThermoSorb/Amine Air
Cartridges. The methods for collection and analysis of amines were
similar to those used for nitrosamines on ThermoSorb/N air cartridges.
After sample collection, the cartridges were backflushed with water,
with the first milliliter collected for analysis. Aliquots of this eluate
were then analyzed by injecting into a gas chromatograph interfaced to a
prototype TEA Analyzer operating in the nitrogen mode. The analyzer used
for the study was a Varian 3700 GC operating from 60 to 220°C at 10° C/min,
with a hold at 60°C for 1 minute, interfaced to a catalytic oxidative
pyrolytic furnace. A 2 mm I.D. x 6' glass column packed with Carbopak B
was used for the GC separation. Argon was the carrier gas with a flow
rate of 20 cc/min. Detection of the NO radical formed from the high tem-
perature oxidation of the amines was by a standard TEA analyzer. A more
detailed description of this system can be found in Appendix D. This
description of the procedure was provided by NEILS to SwRI with sample
results.
6
-------
A minimum number of amine samples were collected and anayzed at SwRI.
These samples were collected by bubbling air samples through glass impingers
containing 0.01 N sulfuric acid, and analyzing a portion of the collecting
solution with the aid of a gas chromatograph equipped with an ascarite-
loaded pre-column and a nitrogen-phosphorus detector (NPD). A complete
description of this procedure is available in the literature.(4)
H. Test Sequence to Determine Influence of Vehicle Age at SwRI - Task II
The vehicles tested in the SwRI portion of this task were employee-
owned, and were tested during SwRI working hours (8:00 AM to 5:00 PM).
Four vehicles were sampled in an initial screening process, and two of
these four vehicles were sampled once a month over a six month time period.
A description of the vehicles tested in this task can be found in Section
III-B of this report. All vehicles were tested in this task using the
fo1lowing sequenee:
1. Vehicle driven into the Emissions Lab (8:00 AM to 8:30 AM)
2. Vehicle allowed to stand for 2 to 2 1/2 hours with the hood
raised and the windows open. Fan used to blow air across
engine during this period.
3. Interior air blown out with a fan to remove all traces of
nitrosamines that could have previously accumulated.
4. Hood, windows, and doors closed.
5. Vehicle allowed to stand for 3 1/2 hours at lab temperatures
(70-76°F).
6. After soak period, driver's side door opened and sample pump
and trap placed in vehicle, trap clamped to driver's sun visor
and pump started (door open approximately 10 seconds) .
7. Vehicle interior sampled for 2 to 2 1/2 hours at a flow rate
of 2 to 3.3 liters per minute.
8. Door opened, pump stopped, pump and trap removed from vehicle.
I. Test Sequence to Determine Influence of Vehicle Age at NEILS - Task II.
The two vehicles tested in NEILS' portion of the task were employee-
owned, and tested during NEILS working hours. Nitrosamine data was avail-
able for both vehicles from previous test work. A description of the
vehicles can be found in Section III-B of this report. The vehicles were
tested using the following sequence:
1. Vehicles driven to NEILS parking lot by owner.
7
-------
2. All windows and doors closed. Vehicles allowed to stand for
1-2 hours at ambient temperature.
3. After soak period, two sample pumps and traps placed in
vehicle and pumps started. One pump and trap placed in front
seat and second placed in trunk or cargo area.
4. Vehicle interior sampled at ambient temperature (27-90°F) for
4 hours at a flow rate of 2 liters per minute.
5. Pump stopped, pump and trap removed from vehicle.
J. Test Sequence to Determine Influence of Operation - Task III
Twelve vehicles were tested in Task III to determine the influence
of operation on interior nitrosamine concentrations. The vehicles tested
included 2 and 4 door sedans, 2 and 4 door hatchbacks, a station wagon, a
pickup truck, a van and a heavy-duty truck. The vehicles tested in the
task are described in detail in Section III-C of this report. All twelve
vehicles were rented locally, and had less than 13,000 miles on the
odometer. The vehicles were sampled at three test points: 1) immediately
before operation, 2) during operation, and 3) immediately after operation.
The twelve vehicles were sampled immediately before operation using
the following sequence:
1. Vehicle driven into the Emissions Lab.
2. Vehicle interior air blown out with a fan to remove all traces
of nitrosamines that could have previously accumulated.
3. All doors and windows closed.
4. Vehicle allowed to stand overnight (14 to 16 hours) at lab
temperature (70-76°F).
5. After overnight soak period, the driver's side door opened,
pump and trap placed in the vehicle, trap clamped to the driver's
sun visor and pump started (door open 10 to 30 seconds).
6. Vehicle interior sampled for 2 to 2 1/2 hours at a flow rate of
2 to 4 liters per minute.
7. Door opened, pump stopped, pump and trap removed from vehicle.
8. Door closed.
All of the vehicles with the exception of a heavy-duty truck (International
Transtar II) and a Mercury Marquis were tested during operation using the
following sequence:
8
-------
1.
Five to fifteen minutes after completion of before-operation
test, driver enters vehicle with pump and trap.
2. All doors and windows closed for duration of test.
3. Trap clamped to the driver's sun visor.
4. Engine and pump started simultaneously.
5. Vehicle driven over predetermined course for 165 to 235 minutes
while sampling at 2 to 4 liters per minute. Course covered
approximately 125 miles with average course speed of 35 miles
per hour. (Air conditioner used intermittently during test for
first six vehicles. Air conditioner and heater off for duration
of test for last six vehicles).
6. Vehicle returned to Emissions Lab, pump and engine stopped
simultaneously.
7. Driver exits vehicle and closes door.
The testing for the International was identical except for the driving
course. The International was driven over a different course to better
simulate the route of a heavy-duty truck. The course for the International
consisted of predominately highway driving, and covered 150 miles. The
average course speed during the test was 50 miles per hour. For the
Mercury Marquis, four samples were taken during operation instead of the
usual one. The four samples were taken using the previously described
sequence with the following variations:
1.
Engine and pump No.
1 started simultaneously.
2.
Pump No. 2
started
1 minute into the trip.
3.
Pump No. 3
started
5 minutes into the trip.
4.
Pump No. 4
started
10 minutes into the trip.
5.
Engine and
four pumps stopped simultaneously.
The twelve vehicles were tested immediately after operation as follows:
1. Five to fifteen minutes after completion of during operation
test, driver's side door opened, pump and trap placed in vehicle,
trap clamped to the driver's sun visor and pump started (door
open 10 to 30 seconds).
2. Vehicle interior sampled for 2 to 2 1/2 hours at a flow rate of
2 to 4 liters per minute.
9
-------
3. Door opened, pump stopped, pump and trap removed from vehicle.
K. Test Sequence to Determine Influence of Ambient Conditions - Task IV
To determine the effect of ambient temperature, fifteen vehicles
were sampled in an initial screening process at the EPA-Ann Arbor lab.
A description of the vehicles can be found in Section III-D of this
report. The fifteen vehicles were tested at ambient lab temperatures
(72-73°F) in the EPA lab soak area as follows:
1. Vehicle brought into EPA soak area.
2. Vehicle allowed to stand at lab temperature for 5 to 15 hours
with driver's window open.
3. Interior air blown out with fan to remove all traces of
nitrosamines that could have previously accumulated.
4. All windows and doors closed.
5. Vehicle allowed to stand for 2 to 2 1/2 hours at lab temperature
(72-73°F).
6. After soak period, driver's side door opened, pump and trap
placed in vehicle, trap clamped to the driver's sun visor
and pump started (door open approximately 10 seconds).
7. Vehicle sampled for 1 1/2 hours at a flow rate of 2 liters
per minute.
8. Door opened, pump stopped, pump and trap removed from vehicle.
Nine of the fifteen vehicles were selected for testing at 40° and
100°F. The vehicles were tested in the EPA cold room or in the EPA tem-
perature-controlled evaporative shed. Both rooms have facilities for
below- and above-ambient-temperature vehicle testing. The nine vehicles
were tested in a manner similar to the ambient testing. The following
sequence was used for testing at 40° and 100°C.
1. Vehicle brought into cold room or evaporative shed.
2. Vehicle allowed to stand at 40°/100° for 2 to 10 hours with
all windows open.
3. Interior air blown out with fan to remove all traces of nitro-
samines that could have previously accumulated.
4. All windows and doors closed.
5. Vehicle allowed to stand for 2 1/2 hours at 40°/100°.
10
-------
6. After soak period, driver's side door opened, pump and trap
placed in vehicle, trap clamped to the driver's sun visor and
pump started (door open approximately 10 seconds).
7. Vehicle sampled for 1 1/2 hours at a flow rate of 2 liters per
minute.
8. Door opened, pump stopped, pump and trap removed from vehicle.
L. Test Sequence for Additional Assessment - Task V
Twenty-five vehicles were tested as an additional assessment in Task
V. The vehicles tested included three used heavy-duty trucks, six new
heavy-duty trucks, six vans, five station wagons, and four motor homes.
A list of the vehicles tested can be found in Section III-E of this
report. With the exception of the four motor homes, two vans (Volkswagen
and GMC vans), and one station wagon (Chrsyler Lebaron), all of the
vehicles were tested at the SwRI Emissions Lab. The motor homes, two
vans, and the one station wagon were tested on dealer lots.
The vehicles were sampled at the Emissions lab using the following
test sequence:
1. Vehicle driven into the Emissions lab.
2. Vehicle allowed to stand for one hour with windows open.
3. Interior air blown out to remove all traces of nitrosamines
that could have previously accumulated.
4. All windows and doors closed.
5. Vehicle allowed to soak overnight (15 hours) at lab temperature
(70-76°F).
6. After soak period, the driver's side door was opened and the
pump and trap were placed in the vehicle, trap clamped to
driver's sun visor and pump started (door open 10 to 30
seconds). Figure 3 shows a sample cartridge being clamped
to the sun visor of a heavy-duty truck.
7. Vehicle interior sampled for 2 1/2 to 4 hours at a flow rate
of 2 to 4 liters per minute.
8. Door opened, pump stopped, pump and trap removed from vehicle.
11
-------
Figure 3. Trap being clamped to sun visor in heavy-duty truck
The four motor homes, the GMC and Volkswagen vans, and the Chrysler
Lebaron station wagon were tested on the dealer lots using the following
test sequence:
1. Vehicle blown out with fan to remove all traces of nitrosamines
that could have previously accumulated.
2. All windows and doors closed.
3. Vehicle allowed to stand overnight (13-18 hours) at ambient
temperature (40 to 70°F).
4. After soak period, the driver's side door opened, pump and trap
placed in vehicle, trap clamped to driver's sun visor and pump
started (door open approximately 10 seconds).
5. Vehicle interior sampled for 2 hours at a flow rate of 2 to 4
liters per minute.
6. Door opened, pump stopped, pump and trap removed from vehicle.
12
-------
III. RESULTS
This section describes the results for each of the six tasks in this
program. Task I includes the peer review of the NEILS methodology for
analyzing nitrosamines in car interiors and the results for determining
the nitrosation potential of the ambient air, the levels of nitrosatable
amines in the ambient air, and background levels of nitrosamines in the
ambient air. Tasks II through IV include the results for the vehicles
sampled to determine the effects of time (vehicle age), operation, and
temperature on interior nitrosamine levels. Task V includes the results
for vehicles sampled as an additional assessment. The vehicle types
sampled in this task include new and used heavy-duty trucks, new vans,
new motor homes, and new station wagons. Task VI gives a rough estimate
of the exposure of several population categories to nitrosamines from
vehicle interiors.
A. Task I - Sampling and Analysis Methodology
To insure the technical integrity of the data that would be gener-
ated in subsequent tasks, a peer review of the methodology used by
Dr. Fine, et al, of the New England Institute for Life Sciences (NEILS),
for sampling and analysis of nitrosamines was conducted. It was tenta-
tively planned to use this methodology in the program. This peer review
was conducted with the aid of other researchers (EPA and industry) who
have had experience with the measurement of nitrosamines. If the method-
ology was found to be acceptable to the scientific community, then samples
would be collected from vehicle interiors and sent to NEILS for analysis.
This analysis was to be carried out at NEILS due to the relatively brief
duration of the project and the relatively long set-up time needed for
the methods.
To determine if the methodology was acceptable, fifteen researchers
in the nitrosamine field were contacted by letter. These fifteen re-
searchers included people who have evaluated the ThermoSorb/N Air Sampler
as to recovery, efficiency, artifact formation, and other performance
characteristics; as well as other researchers who have had experience
with nitrosamine analysis. Individuals on the list include those known
to the author who have had experience in the nitrosamine field and those
recommended by Dr. Fine of NEILS and Dr. Robert Lyle of Southwest Research
Institute. Dr. Lyle has had experience with nitrosamine analysis and is
familiar with workers in the field. The letters sent to the researchers
asked for comments on the TEA analyzer and the ThermoSorb/N Air Sampler.
There were eleven responses to the letters. Table 1 lists the researchers
that were contacted and their organizations, the researchers that respon-
ded, and whether or not the researchers have had experience with the TEA
analyzer and/or the ThermoSorb/N Air Samplers. The responses of the con-
tacted researchers are described in the following paragraphs.
13
-------
TABLE 1. LIST OF RESEARCHERS
Researchers Contacted
Dr. Steve Prescott
Warren Hendricks
Dr. Bertold Speigelhalder
Dr. Pieter Schuler
(Response by
Dr. E. Ellen)
Steve Swarin
Dr. Edo Pellizzari
Dr. Tom Guley
Dr. Roy Zweidinger
(and Dave Dropkin)
Gene Weaver
(Response by Mike Kelly)
Dr. Richard Leoppky
Dr. S. R. Tannenbaum
Dr. Sidney Mirvish
Dr. Larry Keefer
(Response by Tom Hansen)
Dr. Wm. Lijinsky
Dr. Christopher Michedja
Organization Response
Air Products No
OSHA No
Inst, fur Toxicol.
und Chem., Germany Yes
Rijksinstitunt voor
de Volksgeyondkeid,
Netherlands Yes
General Motors Yes
Research Triangle
Institute Yes
Goodyear Tire &
Rubber Co. Yes
EPA-RTP Yes
Ford Motor Co. Yes
University of
Missouri at
Columbia
Massachusetts
Institute of
Technology Yes
Eppley Institute for
Research in Cancer Yes
National Institute
of Health Yes
Frederick Cancer
Research Center Yes
Frederick Cancer
Research Center No
Experience
Experience with
with TEA ThermoSorb/N
Analyzer
Trap
Yes
Yes
Yes
No
Yes
No
No
Yes
Yes
Yes
No
Yes
No
Yes
Out of Country
Yes
Yes
Yes
Yes
Yes
Yes
No
14
-------
Dr. Steve Prescott, Air Products, Warren Hendricks, OSHA, and
Dr. Christopher Michedja, Frederick Cancer Research Center did not
respond. Dr. Richard Leoppky, University of Missouri, was out of the
country at the time of the review. Dr. Edo Pellizzari, Research Triangle
Institute, and Dr. Roy Zweidinger (and Dave Dropkin), EPA-RTP did not
have experience with the TEA analyzer or the ThermoSorb/N Air Sampler.
Dr. Pellizzari, while not familiar with potential TEA and ThermoSorb/N
problems, pointed out that they have been discussed by other investi-
gators in recent publications, and suggested a careful search in ana-
lytical journals for the past two years for more information. Mike Kelly
(responding for Gene Weaver), Ford, did not have experience with the TEA
analyzer. However, Ford has collected samples in ThermoSorb/N traps and
sent them to Dr. Fine at NEILS for analysis. In general, Mike Kelly was
satisfied with Dr. Fine's methodology, however, he questioned the pre-
cision of the method at low nitrosamine concentrations and noted that Ford
has observed blank ThermoSorb traps that were reported to contain trace
amounts of nitrosamines. Ford has also noted in replicate testing varia-
tions of ±50 percent at low nitrosamine levels.
Dr. Sidney Mirvish, Eppley Institute for Research in Cancer, saw no
need to check Dr. Fine's method since it was published in a reputable
journal and peer-reviewed. Tom Hansen (responding for Dr. Larry Keefer),
National Institute of Health, had used a GC-TEA and a LC-TEA. He thought
the TEA sensitivity was very good (much better for GC-TEA than for LC-TEA)
and was happy with the overall performance of the instrument. Mr. Hansen
had little experience with the ThermoSorb/N traps, but thought that more
experiments needed to be done to determine whether or not artifact for-
mation took place on the traps. Dr. William Lijinsky, Frederick Cancer
Research Center, had also worked with the TEA analyzer, but had no
experience with the ThermoSorb/N traps. Dr. Lijinsky thought that the
TEA analyzer was a very sensitive, useful instrument that correlated well
with other GC methods of detection such as the nitrogen/phosphorus
detector (NPD) and the flame ionization detector (FID). He noted that
there are several compounds that interfere with the method, such as
nitrite and nitrate esters which are present in smog.
Dr. S. R. Tannenbaum has had several years experience with the TEA
analyzer and found it relatively trouble free with few difficulties.
Dr. Tannenbaum has used the TEA analyzer for routine analysis as opposed
to mass spectroscopy which is only used for confirmation. He felt the
TEA is a better analyzer for quantification than mass spectroscopy.
Dr. Tannenbaum has had little experience with the ThermoSorb/N traps,
however he has not encountered any problems with their use.
In Dr. E. Ellen's lab (responding for Dr. Pieter Schuler),
Rijksinstitunt voor de Volksgeyondkeid, Netherlands, two TEA analyzers
have been used for nitrosamine analyses, one coupled to a GC and one to
a HPLC. In 1976 and 1977 they performed numerous analyses using both
the GC-TEA and GC-High Resolution Mass Spectroscopy. In most cases they
15
-------
found good agreement between the two methods, both qualitatively and
quantitatively. They have used several methods to collect nitrosamines;
wet trapping (KOH solution or low temperature traps), activated char-
coal, and ThermoSorb/N traps. The ThermoSorb/N sampling tubes gave better
results than any other system that they have used. They observed no
artifact formation of NDMA on the traps when sampling air containing 1 ppm
NO/NO2 and dimethylamine. For seven nitrosamines (NDMA, NDEA, NDPA, NDBA,
NPYR, NPIP, and NMOR) they found recoveries of 80-110% when sampling 120
liters of air containing 200 ng of each nitrosamine. The other methods
for trapping nitrosamines had low recoveries and/or severe artifact
formation.
Dr. Bertold Spriegelholder, Inst, fur Toxicol und Chem., Germany,
was very satisfied with the TEA and the ThermoSrob/N traps. He reported
that the TEA instruments were easy to maintain and operate. Dr. Spriegel-
holder has never observed a negative mass spetrcoscopy analysis when he
had found a positive TEA response. Originally, his lab analyzed every
sample by both the TEA analyzer and a mass spectrometer. They found that
the results from the two methods usually did not differ by more than
10-20 percent, and that the TEA was much easier to use for routine sample
analysis. Nitrosamine recoveries using the ThermoSorb/N Air Sampler in
his lab ranged from 90 to 100 percent. He also noted that when the traps
were eluted with solvent, some sulfamic acid was removed from the traps.
The material caused problems by plugging syringes in automatic samplers.
They now filter all samples to prevent this problem.
Dr. Thomas Burley, Goodyear Tire and Rubber Co., has had excellent
performance from both the ThermoSorb/N traps and the TEA instrument for
over a year. He has found that the instrument gives linear response in
the ppb range. Dr. Gurley also found that nitrosamine standards must be
stored in light-protected vessels in the freezer to maintain sample inte-
grity. His lab has found the detection limits for several nitrosamines
to be as follows: 14 nanograms NDMA, 20 nanograms NDEA, and 48 nanograms
NMOR per trap. Dr. Gurley feels that these limits are conservative and
can be improved upon by optimizing instrument parameters. He has also
found that the ThermoSorb/N traps give results that correlate well with
the KOH (1.0 N) impinger traps. Dr. Gurley feels that "...the sampling
technique and the analytical methodology are excellent approaches to
this complex analytical problem."
Steve Swarin, General Motors, has had considerable experiences with
the TEA analyzer. He has found that the instrument a) has little down
time, b) is a good instrument for a technical level person to operate,
c) gives a short analysis time compared to the mass spectrometer, d)
correlates well with the mass spectrometer, ±20 percent, e) has good
repeatability, and f) can detect nitrosamine levels from 0.002-0.2 ppb
when 500 liters of air are sampled. Mr. Swarin has found that the
ThermoSorb/N traps are: a) easy to use, b) stable for 3 to 4 weeks,
c) have no problems when sampling at 3 liters/minute, d) have good
16
-------
trap-to-trap reproducibility, and e) can be eluted with one milliliter
of acetone to give 98 percent of the nitrosamines on the trap.
The above responses for the most part indicate an acceptance of the
TEA analyzer and ThermoSorb/N traps by the scientific community. Re-
searchers using both the TEA analyzer and ThermoSorb/N traps feel that
both are the best available for routine analysis work. Due to the
general positive nature of these responses, the ThermoSorb/N Air Samplers
were used to collect nitrosamine samples from vehicle interiors and were
sent to NEILS for analysis by GC-TEA.
In order to determine if sampling artifacts are being formed and to
obtain information as to the origin of the nitrosamines, the level of
nitrosation potential of the air, the level of nitrosatable amines in
the air, and the level of nitrosamines in the ambient air were determined
during the course of Tasks II, III, IV, and V.
The nitrosation potential of the ambient air (potential of the ambient
air to convert amines to nitrosamines) was determined using ThermoSorb/N
Air Samplers which were spiked at NEILS with the easily-nitrosated amine,
morpholine. Samples were collected by SwRI and returned to NEILS for
analysis of N-nitrosomorpholine. NEILS had previously found that the
nitrosation potential of the air expressed in ppm NO2 was related to the
amount of N-nitrosomorpholine found in the ThermoSorb/N traps. NEILS
also found that NO levels did not decrease after passage through car-
tridges containing morpholine, suggesting that NO does not contribute
to the formation of NMOR. The following formula related the ppm NO2 in
the ambient air to the mg of N-nitrosomorpholine in the ThermoSorb/N trap.
_ Mol. Wt. NO2 x Ug of formed NMOR
ppm 2 ~ Mol. Wt. NMOR liters of air sampled
24.45 x 760 (mm Hg) x T(°C)+273° y
Mol. Wt. N02 Pexp (mm Hg) 298°K
1
0.07*
~Experimentally 6-8% of the NO2 in the gas stream combines with the
surface-held morpholine to form N-nitrosomorpholine when the traps
have air containing N0X drawn through them at 1 £./min for 50 minutes.
Twenty-one nitrosation potential samples and two blanks were col-
lected and/or analyzed during the course of the program. Table 2 lists
the location from which the nitrosation potential samples were taken,
the ng of NMOR found in the sample when analyzed by NEILS, the concentra-
tion of NO2 in the ambient air using the above formula, and the ppm N0X
in the ambient air analyzed by the use of a chemiluminescent analyzer at
SwRI.
17
-------
TABLE 2 . NITROSATION POTENTIAL ANALYSES
Location
EPA Soak Area
Blank
EPA Cold Room (40°)
EPA Cold Room (100°)
EPA Modified Cold
Room (38°)
SwRI Soak Area
Ford Mustang (After
Operation
Chevrolet Chevette
(Before Operation)
International Transtar
II
GMC Astro
Blank
SwRI Soak Area*3
Oldsmobilg Cutlass
Cruiser
Ford Granada
(During Operation)
SwRI Soak Areac
Mercury Marquisc
Date
6/7/80
6/7/80
6/12/80
6/12/80
6/14/80
8/11/80
10/2/80
10/3/80
1/9/81
1/13/81
1/28/81
2/11/81
2/11/81
2/18/81
2/20/81
2/20/81
ng NMOR
per cartridge
377
793
177
365
223
440
1400
1270
1900
2120
1530
118000
6720
1540
2540
2260
ppm N02a
calculated from
trap analysis
0.020
0.005
0.024
0.012
0.028
0.048
0.043
0.089
0.126
1.161
0.063
0.093
0.156
0.136
ppm NO
(measured)
< 1
< 1
< 1
< 0.5
< 0.5
18
-------
TABLE 2 (Cont'd). NITROSATION POTENTIAL ANALYSES
Location
Date
Ford Pinto
(sixth sampling)
Winnebago Chieftain
Dealer Lot
San Antonio^
Itasca Sunflyer^
Ford Fairmonte
Oldsmobile Station
Wagone
Oldsmobile Station
Wagone
2/23/81
3/17/81
3/18/81
3/18/81
4/23/81
4/27/81
4/27/81
ng NMOR
per cartridge
2130
1570
1820
1990
600
450
510
ppm NO2
calculated from
trap analysis
0.145
0.105
0.110
0.134
0.018
0.014
0.015
ppm NOx
(measured)
ppm N02 calculated from formula
ppm
M.W. N02
x
yg of formed NMOR
M.W. NMOR liters of air sampled
24.45 x 760 x T + 273
M.W. N02 P 298
oToT*
*Experimentally 6-8% of the N02 in gas stream combines with the surface held
morpholine to form N-nitrosomorpholine when the devices have air containing
NOx drawn through them at 1 &/min for 50 min.
^Samples taken simultaneously
Sample volume 300JI instead of 508,
Samples taken simultaneously
^Samples taken simultaneously
Samples taken at NEILS by NEILS
19
-------
The NO2 levels in vehicle interiors were found to range from 0.014
to 0.145 ppm NO2, while ambient air samples outside the vehicles were
found to range from 0.005 to 1.161 ppm NC^. However, the sample which
gave 1.161 ppm NO2 was inadvertantly sampled for 300 liters instead of
the usual 50 liters. This longer sampling period may negate the validity
of the NO2 calculation. Disregarding this value, the NO2 levels ranged
from 0.005 to 0,156 ppm. Two blank samples were taken during the course
of the program. The analyses of these two blanks at NEILS gave appre-
ciable amounts of NMOR in each sample, 793 ng in the June 7th sample and
1530 ng in the January 28th sample. This indicates that there is some
conversion of morpholine to NMOR with time (increase from 793 to 1530 ng
NMOR) in a closed cartridge. Also the presence of 793 ng NMOR in the
June 7th blank (higher than all samples taken in June) indicates there
are problems with conversion of morpholine to NMOR during morpholine
spiking or during sample analysis. The chemiluminescent NOx measurements
were accurate to only ±0.3 ppm NOx and indicated NOx levels less than
1 ppm. This is in agreement with the NO2 values calculated from the
NMOR levels.
The level of nitrosatable amines in the air was determined by
collecting amine air samples in ThermoSorb/amine air cartridges at
SwRI and EPA-AA and then sending the cartridges to NEILS where they were
backflushed with 2 m£ of 1 N KOH to elute the amines, and analyzed by
injecting a portion of the eluate into a prototype TEA analyzer operating
in the nitrogen mode. A minimum number of samples were collected in
impingers containing dilute sulfuric acid, and analyzed using a gas chro-
matograph (GC) equipped with an ascarite-loaded precolumn and a nitrogen
phosphorus detector (NPD). These samples were collected simultaneously
with those collected in air cartridges and were analyzed at SwRI. This
method is described in detail in the EPA report, "Analytical Procedures
for Characterizing Unregulated Pollutant Emissions from Motor Vehicles.
During the program, eighteen amine air samples and one blank car-
tridge were sent to NEILS for analysis. Table 3 lists the location and
the date the samples were taken, the volume of air sampled, the mass
(in ng) of amines found in each cartridge, the concentration of amines
(in yg/m3) determined using the sulfuric acid impinger and GC-NPD. Ten
sample cartridges were found to contain dimethylamine at levels ranging
from 460 to 5400 ng. These values give ambient air concentrations that
range from 1 to 31 yg DMA/m3. Morpholine was detected in four cartridges
at levels ranging from 240 to 10930 ng (1 to 23 yg/m3) and n-propylamine
was detected in one cartridge (266,000 ng; 839 yg/m3). Five samples con-
tained one or two unknown nitrogen-containing compounds at levels ranging
from 640 to 128,000 ng. These values were determined by quantification
against morpholine. The Fairmont and Chevette samples which contained
the n-propylamine and the highest levels of unknown compounds were inad-
vertantly analyzed for nitrosamines before the amine analysis. Therefore,
these two samples were handled in a manner differently from the other
samples. Four of the samples were collected in duplicate, one with the
20
-------
TABLE 3. AMINE ANALYSES
Location
EPA Soak Area
EPA Cold Room (105°F)
EPA Cold Room (38°)
EPA Modified Cold
Room (38°)
SwRI Soak Area
Fairmont (Before
Operation)^
Chevette (Before
Operation)5
International
Transtar
GMC Astro
Blank
Date
6/7/81
6/11/80
6/13/80
6/14/80
8/11/80
10/1/81
10/3/81
1/8/81
1/13/81
1/28/81
Volume
Sampled (liters)
163
177
194
182
270
317
409
210
111
ng Amine
per Cartridge5
1900 DMAb
5400 DMA
240 M0Rc
2700 DMA
1600 DMA
1860 MOR
5400 DMA
266,000 nPrA
5400 Unkn-le
23,400 Unkn-2e
720 DMA
128,000 Unkn-1
86,400 Unkn-2
6800 DMA
8,600 Unkn-1
17,200 Unkn-2
720 DMA
1,980 Unkn-1
ND
Amine levels
Ug/m3
12 DMA
31 DMA
1 MOR
14 DMA
9 DMA
7 MOR
17 DMA
839 nPrA
17 Unkn-1
74 Unkn-2
2 DMA
313 Unkn-1
211 Unkn-2
32 DMA
41 Unkn-1
82 Unkn-2
6 DMA
18 Unkn-1
ND
SwRI Amine
Analysis ]ig/m3
14 MMA1
2 DMA
-------
TABLE 3 (Cont'd). AMINE ANALYSES
location
Oldsmobile Cutlass
Cruiser
Datsun 310 GX
SwRI Soak Area
Ford Super Wagon
Ford Granada
(After Operation)
Ford Pinto
Winnebago Chieftain
Ford Fairmont*1
Oldsmobile Station
Wagon'1
Volume
Date Sampled (liters)
2/11/81
2/16/81
2/17/81
2/17/81
2/18/81
2/23/81
3/17/81
4/23/81
4/27/81
355
315
371
480
310
347
245
480
480
ng Amine
per cartridge
ND
ND
ND
640 Unkn-2
ND
ND
ND
460 DMA
1090 MOR
460 DMA
3050 MOR
Amine levels
ug/m3
ND
ND
ND
1 Unkn-2
ND
ND
ND
1 DMA
23 MOR
1 DMA
6 MOR
SwRI Amine
Analysis Ug/m3
ND
0.1 DMA
aDetection limit 100 ng of each amine per cartridge
^DMA = dimethylamine
cMOR = morpholine
nPrA = n-propylamine
®UnKn-l, UnKn-2, unknown compounds quantified against morpholine
MMA = monomethylamine
^These two samples were inadvertantly analyzed at NEILS for nitrosamines before the amine analysis
^Samples taken at NEILS by NEILS
-------
air cartridges, and one with sulfuric acid. In the International Transtar,
SwRI (using the sulfuric acid collection media and the GC-NPD for analysis)
found 14 yg/m3 monomethylamine, while the cartridge analyzed at NEILS indi-
cated 32 yg/m3 dimethylamine. In the GMC Astro, both methods indicated
dimethylamine (6 yg/m3 NEILS, 2 yg/m3 SwRI). SwRI found 0.1 yg/m3 dimethyl-
amine in the Ford Granada where none was detected in the corresponding
air cartridge. However, the detection limit for the cartridge is calcu-
lated to be only 0.3 yg/m3 for the sample.
To determine if nitrosamines were present in the ambient air or were
present as impurities in the analysis, fourteen background and blank nitros-
amine cartridges were collected and analyzed during the program. The re-
sults of these analyses are presented in Table 4. Of the thirteen, only
three samples were found to contain nitrosamines (one contained 9 ng NDMA,
in only one of two analyses performed on the sample; the second contained
5.4 ng NDMA which was only 0.4 ng above the minimum detection value; and
the third contained 18 ng NMOR). The three traps were taken inside the
EPA cold room at 105°F with two vehicles in the room; in the SwRI soak
area with no test vehicles present; and in the SwRI soak area with two
test vehicles present.
For additional confirmation of the presence of NDMA in vehicle
interiors, three composite samples were analyzed with the aid of gas
chromatography-high resolution mass spectroscopy. The following vehicle
interior samples were combined to make up the three composite samples:
Sample #1 - Composite of 12 samples from heavy-duty truck
interiors
Sample #2 - Composite of 20 samples from passenger car
interiors
Sample #3 - Composite of 25 samples from station wagon
interiors.
The composite samples were analyzed with the use of a GC-TEA before
mass spectral analysis. The mass spectral analysis confirmed the
presence of NDMA in vehicle interiors and was in good agreement with
the GC-TEA analysis (Table 5).
B. Task II - Influence of Vehicle Age
This section describes the results for sampling four vehicles (two
at SwRI and two at NEILS) once a month for 6 months during the course of
the program. At SwRI, four vehicles were sampled in an initial screening
process. A description of these four vehicles is presented in Table 6.
The table lists the vehicle, the date the vehicle was manufactured, the
mileage on the vehicle odometer during sampling, and a description of the
spare tire and the vehicle interior. These parameters may influence the
vehicle interior nitrosamine concentrations and are therefore recorded.
The results for sampling during the SwRI screening process are presented
in Table 7. Of the four vehicles, the Datsun 310 GX and the Ford Pinto
23
-------
TABLE 4. BACKGROUND AND BLANK NITROSAMINE SAMPLES
Location
EPA Soak Area
EPA Soak Area
Blank
EPA Cold Room
EPA Modified
Cold Room
EPA Cold Room
Blank
SwRI Soak Area
SwRI Soak Area
SwRI Soak Area
SwRI Soak Area
SwRI Soak Area
SwRI Soak Area
Blank
Date
6/06/80
6/07/80
6/07/80
6/11/80
6/11/80
6/12/80
6/14/80
8/06/80
10/30/80
1/20/81
1/27/81
1/29/81
2/12/81
4/27/81
Temp.
73°
72°
105°
101°
100°
81°
74°
73°
74°
75°
70°
Volume
Sampled
90 I
242 I
178 I
164 I
178 I
301 I
289 I
356 i
315 i
288 i
240 i
GC-TEA Analyses
First Second
ND
ND
ND
ND
ND
ND
ND
ND
ND
5.4 ng NDMA
18. ng NMOR
ND
ND
ND
ND
ND
ND
9 ng NDMA
ND
ND
ND
second analysis not conducted
TABLE 5. MASS SPECTRAL ANALYSES
Sample
Composite #1
Composite #2
Composite #3
GC-TEA
NDMA yg/m&
0.19
0.29
0.25
GC-MS
NDMA yg/m£
0.2
0.3
0.3
24
-------
TABLE 6. DESCRIPTION OF VEHICLES SAMPLED IN SwRI
TASK II SCREENING PROCESS
Vehicle
Datsun 310 GX
Date
Manufactured
March 1980
Mileage
03685
Spare Tire
Bridgestone,
steel belted
radial
Interior
Red Cloth/Vinyl
Blue Stripe
Oldsmobile Cutlass
Supreme January 1980
03469 Uniroyal,
Temporary Use
Black Vinyl
Ford Pinto
December 1979
(Canada)
Goodyear,
04038 Temporary Use
(Canada)
Orange Cloth/
Vinyl
Pontiac Trans Am October 1970
10229
Goodyear
Tan Cloth/
Vinyl
TABLE 7
Vehicle
Datsun 310 GX
Oldsmobile Cutlass
Supreme
Fort Pinto
Pontiac Trans Am
NITROSAMINE SAMPLING RESULTS FOR SwRI
TASK II SCREENING PROCESS
Date
8/06/80
8/06/80
8/06/80
8/11/80
Volume
Sampled
(liters)
456
381
277
355
ng NDMAa
per cartridge
24
12
13
10
NDMA
cone,ng/m^
0.053
0.031
0.047
0.028
NDMA was the only detectable nitrosamine found in the cartridges
25
-------
gave the highest level of NDMA, and were selected for further testing.
Both had approximately 0.05 yg/m3 NDMA in the interior. Table 8 lists
the results for sampling the Datsun 310 GX and the Ford Pinto six times
over a period of 6 1/2 months. The results in Table 8 indicate a general
trend of decreasing NDMA with mileage and time. This trend can be seen
TABLE 8.
TEST
RESULTS FOR
NITROSAMINE SAMPLING - TASK
II (SwRI)
Volume Sampled
ng DNMA
NDMA Levels
Sample
Date
Mileage
(liters)
cartridge
yg/m3
Pinto
8/6
4038
277
13
0.047
Pinto
9/24
5121
314
17a
0.054
Pinto
10/30
5939
319
7
0.022
Pinto
11/25
—
500
19
0.038
Pinto
1/7
—
318
10
0.031
Pinto
2/23
8769
317
11
0.035
Datsun
310
8/6
3685
456
24
0.053
Datsun
310
9/30
7120
254
8
0.031
Datsun
310
10/31
9429
315
0.022
Datsun
310
11/25
—
395
11
0.028
Datsun
310
1/8
13639
317
6
0.019
Datsun
310
2/16
16355
594
12
0.020
^Confirmed by LC-TEA to be 24 ng NDMA/cartridge
Confirmed by LC-TEA to be 10 ng NDMA/cartridge
Sample also contained 10 ng of diethylnitrosamine NDEA
more clearly in Figures 4 and 5. Figure 4 is a plot of NDMA concentration
versus time for the Datsun and Pinto. For the Datsun, the plot indicates
a large initial drop of NDMA with time and a leveling off at 0.02 yg/m3
after approximately 80 days. The Pinto plot is more scattered, but shows
an overall decrease of NDMA with time. Figure 5, a plot of NDMA concen-
tration versus miles driven after the initial test indicates an initial
drop in NDMA concentration followed by a leveling off at 0.02 lig/m3 after
approximately 6,000 miles for the Datsun. The plot for the Pinto, as in
the time plot, is scattered, but indicates a decrease in NDMA concentration
with miles driven.
Table 9 gives a limited description of the two vehicles sampled with
time at NEILS. Both vehicles had been sampled in an earlier program at
NEILS and therefore data is available for the two vehicles over periods
of 17 and 21 months. Table 10 lists the results for sampling the front
seat area of the two vehicles. The vehicles were also sampled in the
trunk (Fairmont) and in the cargo/back seat (Oldsmobile) areas. This
data is presented in Appendix E. The results in Table 10 reveal no trend
26
-------
0.06
Cn
3.
C
O
•H
¦P
(0
4J
C
-------
for the nitrosamine data with time. However, the vehicle temperature
was not held constant from test to test and the changes in temperature
appear to have introduced a more dominant variable into the testing.
Section I1I-D of this report presents data indicating that nitrosamine
levels are dependent on temperature.
TABLE 9. DESCRIPTION OF VEHICLES SAMPLED IN NEILS TASK II
Vehicle Model Year Body Style Interior
Ford Fairmont 1979 4-door
Oldsmobile Cutlass
Supreme 1979 station wagon tan cloth seats
TABLE 10. TEST RESULTS FOR NITROSAMINE SAMPLING-TASK II (NEILS)
Nitrosamine Levels, Ug/m3
Sample
Date
Temp., °F
NDMA
NMOR
Fairmont
11/2/79
45
0.059
ND
Fairmont
1/30/81
34
0.017
ND
Fairmont
2/17/81
57
0.035
ND
Fairmont
3/18/81
27
0.029
ND
Fairmont
4/23/81
64
0.083
0.098
Oldsmobile
7/23/79
90
0.244
0.360
Oldsmobile
6/17/80
75
0.180
—
Oldsmobile
2/17/81
57
0.058
ND
Oldsmobile
3/18/81
27
ND
ND
Oldsmobile
4/27/81
61
0.029
0.013
C. Task III - Influence of Operation
In this task twelve vehicles were selected and sampled to determine
the influence of operation on nitrosamine levels in vehicle interiors.
The twelve vehicles selected for sampling in this task are described
in Table 11. Data of manufacture, mileage, spare tire description and
interior description are included in the table. All twelve vehicles
were rented locally and had less than 13,000 miles on the odometer.
The vehicles were sampled at three test points: 1) immediately before
operation, 2) during operation, and 3) immediately after operation.
28
-------
TABLE 11. VEHICLES SAMPLED
Vehicle
Oldsmobile
Cutlass Supreme
Buick Skylark
Chevrolet Citation
Hatchback
Chevrolet Chevette
Hatchback
Ford Mustang
Ford Fairmont
Body
Style
2-door
4-door
2-door
4-door
2-door
4-door
Date
Manufactured
June 1980
March 1980
August 1979
March 1980
May 1980
August 1980
Chevrolet
Custom Deluxe
Mercury Zephyr
Mercury Marquis
Pickup
Station
Wagon
Dec. 1979
Dec. 1980
4-door Sept. 1980
International
Transtar II
Ford Super
Wagon
Cab over
w/sleeper
Van
Nov. 1980
Aug. 1980
Ford Granada
4-door
Sept. 1980
DR INFLUENCE OF OPERATION
Mileage
04021
Spare Tire
Interior
Firestone Temp. Use Green Vinyl
10002
12843
07385
General Temp. Use
Uniroyal Hideaway
Tan Cloth/Vinyl
Brown/Gray Cloth
Vinyl
Firestone Temp. Use Red Vinyl
07893
00672
04653
Firestone Temp. Use
Michelin Radial
(made in Italy)
N/A
Tan Cloth/Vinyl
Red Vinyl
Blue Vinyl
00349
03007
01381
Firestone Steel-
Belted Radial
General Dual Steel
Radial
N/A
Blue Vinyl
Red Cloth/Vinyl
Tan Vinyl
06208
Goodyear
Blue Vinyl
06506 Goodyear Polysteel Tan Vinyl
Radial
-------
To test the vehicles before operation, a vehicle was driven into
the Emissions Lab, the air in the interior of the vehicle was blown out
with a fan to remove any previous accumulation of nitrosamines, all doors
and windows were closed, and the vehicle was allowed to soak overnight
(14 to 16 hours) at lab temperature (70-76°F). After the overnight soak
period, the driver's side door was opened, the pump and trap were placed
in the vehicle, the trap was clamped to the driver's sun visor and the
pump started. The vehicle interior was sampled for 2 to 2 1/2 hours at
a flow rate of 2 to 4 liters per minute.
To test the vehicle during operation, a driver entered each vehicle
with a second pump and trap five to fifteen minutes after completion of
the previous test. The trap was attached as before and all doors and
windows were closed for the duration of the test. The vehicle and pump
were started simultaneously. The vehicle was driven over a predetermined
course for 165 to 235 minutes while sampling at 2 to 4 liters per minute.
The course covered approximately 125 miles with an average course speed of
35 miles per hour. Upon returning to the Emissions Lab, the pump and
engine were turned off simultaneously. The International Transtar II
was driven over a different course to better simulate the route of a
heavy-duty truck. The course in this case consisted of predominantly
highway driving with only three stops. It covered approximately 150
miles with an average speed of 50 miles per hour. The air conditioner
was used intermittently during the testing of the first six vehicles.
The last six vehicles were tested with all windows closed and with the
air conditioner and heater off for the duration of the test. Four samples
were taken during the operation of the Mercury Marquis instead of the
usual one. In this case, the first sample pump and the vehicle were
started simultaneously, and the second, third, and fourth pumps were
started at 1, 5, and 10 minutes into the trip. All four pumps and the
engine were stopped simultaneously at the end of the trip.
To test the vehicle after operation, the vehicle was driven back
into the Emissions Lab and a third pump and trap were placed in the
vehicle. This test was conducted as described in the "before operation
test." This test began five to fifteen minutes after the completion of
the previous test.
The results for samples from the twelve vehicles are listed in
Table 12. All twelve vehicles were found to contain detectable concen-
trations of NDMA before operation, at levels of 0.013 to 0.141 ]ig/m^.
No other nitrosamines were detected for the before- and after-operation
testing. The average levels of NDMA before operation (0.049 \jg/m3 for
12 vehicles) and after operation (0.049 yg/m^ for 11 vehicles) were equal,
indicating a relatively rapid equilibrium of NDMA between the materials
in the vehicle interior and the interior air. During operation, five
vehicles were found to contain nitrosamines. Two vehicles contained
only NDMA, one contained NDMA and NDEA, one contained NDMA and NMOR,
and one contained NDMA, NDEA, and NMOR. The presence of NDEA and NMOR
is not fully understood, since these nitrosamines were not found before-
30
-------
TABLE 12.
TEST RESULTS FOR NITROSAMINE SAMPLING-TASK III
Volume
Sampled
Nitrosamine
Analysis
Nitrosamine
levels
Vehicle
Test Condition
(liters)
(ng NDMA/cart.)
(yg/m )
Oldsmobile
Before Operation
355
10
0.028 NDMA
Cutlass Supreme
a
During Operation
744
ND
ND
After Operation
353
4
0.011 NDMA
Buick Skylark
Before Operation
412
8
0.019 NDMA
During Operation
714
ND
ND
After Operation
415
8
0.019 NDMA
Chevrolet Citation
Before Operation
570
12
0.021 NDMA
During Operation
810
ND
ND
After Operation
570
13
0.023 NDMA
Chevrolet Chevette
Before Operation
500
14
0.028 NDMA
During Operation
696
ND
ND
After Operation
501
8
0.016 NDMA
Ford Mustang
Before Operation
435
9
0.021 NDMA
During Operation
592
ND
ND
After Operation
444
14
0.032 NDMA
Ford Fairmont
Before Operation
458
6
0.013 NDMA
During Operation
693
ND
ND
After Operation
413
ND
ND
Chevrolet Pick-Up
Before Operation
252
14
0.055 NDMA
During Operation
631
ND
ND
After Operation
346
14
0.040 NDMA
Mercury Zephyr
Before Operation
455
64b
0.141 NDMA
Station Wagon
During Operation
655
20C
0.031 NDMA
After Operation
250
42
6.9NDEA 0.011 NDEA
e
0.168 NDMA
31
-------
TABLE 12 (Cont'd) TEST RESULTS FOR NITROSAMINE SAMPLING-TASK III
Vehicle
Mercury Marquis
International
Transtar II
Ford Superwagon
Ford Granada
Test Condition
Before Operation
During Operation
(duration of test)
Volume
Sampled
(liters)
456
708
During Operation
(one min. into trip) 641
During Operation
(five min. into trip)358
During Operation
(ten min. into trip) 334
After Operation 431
Before Operation 247
During Operation 708
After Operation 252
Before Operation 252
During Operation 579
After Operation 247
Before Operation 247
During Operation 776
After Operation 248
Nitrosamine
Analysis
(ng NDMA/cart.)
25
ND
11
18 NMOR
ND
Nitrosamine
levels
(Ug/m3)
0.055 NDMA
ND
0.017 NDMA
0.028 NMOR
ND
7.2 0.022 NDMA
8.2 NDEA 0.025 NDEA
18
15
5.5
14
22
30
21 NMOR
18
15
8.5
14
0.042 NDMA
0.061
0.008
0.056
0.087
0.052
0.036
0.073
0.061
0.011
0.056
NDMA
NDMA
NDMA
NDMA
NDMA
NMOR
NDMA
NDMA
NDMA
NDMA
a
Not detected - limit of detection per cartridge 5 ng for NDMA, 6.5 ng for NDEA,
k and 9 ng for NMOR
Confirmed by LC to be 45 ng NDMA/cartridge
^ Confirmed by LC to be 22 ng NDMA/cartridge
Confirmed by LC to be 6.5 ng NDEA/cartridge
Confirmed by LC to be 40 ng NDMA/cartridge
32
-------
or after-operation. NDEA and NMOR were found in only one of four traps
for the vehicle that contained both NDEA and NMOR. The levels of NDMA
found during operation for the Zephyr station wagon, the Marquis (average
of the four traps), and the Granada were approximately one fifth of those
found before and after operation. The levels were higher for the Super-
wagon, one half, and lower for the International Transtar II, one seventh.
If the first six vehicles tested had NDMA levels during operation one
fifth or less of the before- and after-operation average, then the levels
would be below the detection limits of the methodology and would not be
detected. No NDMA was detected during operation for the first six
vehicles. If the Chevrolet Pick-up contained one seventh or less of the
before and after operation average, then no NDMA would be detected (no
NDMA was detected).
D. Task IV - Influence of Ambient Conditions
This section describes the results of testing new car interiors to
determine the influence of ambient temperature conditions on nitrosamine
levels in the car interior. In an initial screening process, fifteen new
vehicles were obtained by EPA and brought to the EPA Ann Arbor lab. With
the aid of EPA, SwRI sampled the fifteen vehicles at ambient lab tempera-
ture, 72-73°F. A description of the fifteen vehicles is given in Table 13.
For testing at ambient temperature, the vehicles were allowed to stand at
lab temperature for 5 to 15 hours with the driver's window open. The air
in the interior of the vehicle was then blown out with a fan to remove
all traces of nitrosamines that could have accumulated at a previous
temperature. The vehicles were then closed and allowed to soak for 2 to
2 1/2 hours at lab temperature (72-73°F). After the soak period, the
driver's side door was opened, the pump and trap were placed in the
vehicle, the trap was clamped to the driver's sun visor and the pump
started (door open approximately 10 seconds). The vehicle was sampled
for 1 1/2 hours at a flow rate of 2 liters per minute.
The results for sampling the fifteen vehicles are listed in Table 14.
Nine of the vehicles were selected for testing at 40 and 100°F as a result
of nitrosamine detection in the first nitrosamine analysis (column 3).
The nitrosamine cartridges were reanalyzed by NEILS at a later date.
The results for this reanalysis are listed in column 4. In the first
set of analyses, the Citation and St. Regis gave measurable amounts of
NDMA and the LTD did not, whereas in the second set of analyses, the
LTD gave a measurable amount of NDMA and the Citation and St. Regis did
not. Because the selection of the vehicles for further testing was
based on the initial analysis, the Citation and St. Regis were selected
for further testing and the Ford LTD was not selected. Table 14 also
lists LC-TEA confirmational analyses for seven of the samples and cal-
culated NDMA concentrations in yg NDMA per cubic meter of air. The cal-
culations are based on the average of the first and second GC-TEA analyses
or on only one of the two analyses when the second analysis gave no
detection.
33
-------
TABLE 13. VEHICLES SAMPLED AT AMBIENT TEMPERATURE (72-73°F)
Vehicle
Plymouth Horizon/
Hatchback
Ford Pinto/
Hatchback
Ford LTD
Honda Accord
Chevrolet Caprice
Classic
Oldsmobile Cutlass
Supreme
Dodge St. Regis
Ford Mustang/
Hatchback
Dodge Diplomat/
Wagon
Chevrolet Citation/
Hatchback
Chrysler Newport
Mercury Marquis
Ford Mustang/
Hatchback
Datsun 310 GX
Toyota Corolla
Sport Coupe
Date
Manufactured Mileage
Spare Tire
Interior
May 1980
March 1980
March 1980
April 1980
00062
01816
01678
00017
Firestone Temp. Use
Goodyear Temp. Use
Firestone Steel
Belted Radial
Bridgestone Steel
Belted Radial
December 1979 07325 Uniroyal Hideaway
December 1979 02121
January 1980 00013
January 1980 01748
Uniroyal
Goodyear Temp. Use
B.F. Goodrich Temp. Use
February 1980 00077 Goodyear Temp. Use
January 1980 10670
December 1979 00212
March 1980 00151
February 1980 00022
January 1980 00427
March 1980 00202
General
Goodyear Temp. Use
Goodyear Radial
Firestone, Limited
Use, Space Saver
Bridgestone
Bridgestone
Steel Belted Radial
Tan Vinyl
Red Vinyl
Beige Cloth/Vinyl
Tan Cloth/Vinyl
Blue Vinyl
Tan Cloth/Vinyl
Cream Cloth/Vinyl
Blue Cloth/Vinyl
Tan Cloth/Vinyl
Tan Vinyl
Tan Cloth/Vinyl
Beige Cloth/Vinyl
Blue Cloth/Vinyl
Blue Cloth/Vinyl
Cream/Brown Vinyl
-------
TABLE 14. TEST RESULTS FOR SAMPLING AT 72°F-TASK IV
ng NDMA/Cartridge ng NDMA/Cartridge ng NDMA/
Vol. Samp. First Analysis Second Analysis3 Cartridge Levels,
Vehicle (liters) GC-TEA GC-TEA LC-TEAb (yg/m3)c
Plymouth Horizon/
Hatchback
213
NDd
ND
ND
Ford Pinto/
Hatchback
188
ND
ND
ND
Ford LTD
181
ND
23
19
0.127
Honda Accord
195
ND
ND
ND
Chevrolet Caprice
Classic
190
ND
ND
ND
Oldsmobile Cutlass
Supreme
174
ND
ND
ND
Dodge St. Regis
188
4.6
ND
0.024
Ford Mustang/
Hatchback
207
5
18
8
0.056
Dodge Diplomat/Wagon
207
9.6
6
0.038
Chevrolet Citation/
Hatchback
184
8.4
ND
0.046
Chrysler Newport
196
9.4
15
8
0.062
Mercury Marquis
207
11
26
23
0.089
Ford Mustang/
Hatchback
185
12
12
20
0.065
Datsun 310 GX
188
16
19
11
0.093
Toyota Corolla
Sport Coupe
183
16
33
30
0.134
aSecond analysis of same sample cartridges by GC-TEA, detection limit 5 ng/cartridge
^Confirmational analysis performed on samples by LC-TEA
cLevels based on the average of the first and second GC-TEA analyses or on one of the two analyses when the
second analyses gave no detection
^No detection
-------
The testing of the nine vehicles at 40° and 100°F was performed in
a manner similar to that described for the ambient tests, with the follow-
ing differences. The vehicles were allowed to stand at 40 or 100°F for
2 to 10 hours (ambient 5-15 hours) with all windows open, and after
blowing out the interior air and closing all windows and doors, the
vehicles were allowed to soak for 2 1/2 hours before testing (ambient
2 to 2 1/2 hours).
The nine vehicles that were selected for further study are listed
in Table 15, along with the volume of air sampled from the interior,
results from duplicate analyses of the ThermoSorb cartridges, and cal-
culated NDMA concentrations. The data in Table 15 appear to indicate
NDMA dependence on temperature, with higher temperatures giving higher
levels of NDMA. This dependence on temperature can be seen more clearly
in Figures 6 and 7. The two figures show increases in NDMA concentrations
with temperature for all nine vehicles.
E. Task V - Additional Assessment
Twenty-five vehicles were sampled in this task, widening the nitros-
amine data base to include new and used heavy-duty trucks, new vans,
motor homes, station wagons, and pickup trucks. The vehicles sampled in
this task are listed in Table 16. With the exception of the four motor
homes and the Chrysler Lebaron station wagon, all vehicles tested were
found to contain one or more nitrosamines at levels of 0.01 to 0.12 yg/m3
(Table 17). The four motor homes contained no detectable concentrations
of nitrosamines. The Chrysler LeBaron station wagon (1 mile on odometer)
contained 0.38 yg/m3 NDMA and 0.16 yg/m3 N-nitrosodibutylmaine (NDBA).
The three used heavy-duty trucks contained only NDMA, at levels of
0.022 to 0.029 yg/m3. The Ford 9000 and the White Road Boss had conven-
tional cabs, while the Freightliner had a "cab over" with a sleeper.
This testing indicated that used heavy-duty trucks, over three years old
and with over 200,000 miles on the odometer, can still contain measurable
amounts of nitrosamines. The six new heavy-duty trucks contained higher
levels of NDMA, 0.026 to 0.119 yg/m3, with an average of 0.067 yg/m3.
Three trucks, the two Freightliners and the Mercedes, contained only
NDMA, while the two Internationals and the GMC were found to also contain
NDEA and/or NMOR at levels of 0.014 to 0.104 yg/m3. Total nitrosamine
levels in the six trucks ranged from 0.026 yg/m3 for the Mercedes to
0.151 yg/m3 for the International S1800, with an average level of 0.091
yg/m3. The Mercedes' interior contained little non-metallic material
other than the seat. This could explain the lower nitrosamine levels
for the Mercedes.
Hie six vans tested in the program all contained NDMA at levels
ranging from 0.017 to 0.101 yg/m3, with an average concentration of
0.041 yg/m3. The Chevrolet Van 30 (mini school bus manufactured by
Wayne Corp), the Dodge Sportsman, and the Volkswagen Vanagon contained
only NDMA, while the Ford Club Wagon and the Ford Superwagon also
36
-------
TABLE
15. TEST
RESULTS FOR
SAMPLING AT 40, 72 and
100°F
First Nitrosamine
Second Nitrosamine
Temp.
Vol. Samp.
Analysis
Analysis
NDMA Leve
Vehicle
°F
(liters)
ng NDMA/Cartridge
ng NDMA/Cartridge
(yg/m3)
Dodge St. Regis
45
190
trace NMOR
NDb
ND
Dodge St. Regis
73
188
4.6
ND
0.024
Dodge St. Regis
94
181
trace
14
0.077
Chevrolet Citation
40
178
trace NMOR
ND
ND
Chevrolet Citation
72
184
8.4
ND
0.046
Chevrolet Citation
103
177
44
44
0.249
Dodge Diplomat
40
188
trace
ND
ND
Dodge Diplomat
73
207
9.6
6
0.038
Dodge Diplomat
100
188
ND
28
0.149
Ford Mustang (Man. Trans)
41
182
ND
ND
ND
Ford Mustang
72
207
5
18
0.056
Ford Mustang
100
184
trace
11
0.060
Ford Mustang (Auto. Trans)
42
192
ND
ND
ND
Ford Mustang
72
185
12
12
0.065
Ford Mustang
98
188
23
32
0.146
Chrysler Newport
45
200
ND
ND
ND
Chrysler Newport
73
196
9.4
15
0.062
Chrysler Newport
94
186
trace
23
0.124
Mercury Marquis
42
179
ND
ND
ND
Mercury Marquis
72
207
11
26
0.089
Mercury Marquis
99
180
26
30
0.156
Datsun 310 GX
41
184
ND
ND
ND
Datsun 310 GX
72
188
16
19
0.093
Datsun 310 GX
103
186
31
49
0.215
Toyota Corolla
41
190
ND
ND
ND
Toyota Corolla
72
183
16
33
0.134
Toyota Corolla
100
195
51
79
0.333
Levels based on the average of the first and second analyses or on only one of the
^two analyses when the second analysis gave a trace amount or no detection
No detection, detection limit for second analysis 5 ng NDMA/Cartridge
-------
ro
e
tn
3.
c
o
•H
¦P
m
w
4J
c
0)
u
S
0
u
Q
s
0.3 -
0.2 -
0.1 -
1 Toyota Corolla
2 Chevrolet Citation
3 Datsun 310 GX
4 Ford Mustang (Autom.
5 Dodge St. Regis
ND
Temperature,
Figure 6. NDMA concentration as a function of temperature:
Toyoto Corolla, Chevrolet Citation, Datsun 310 GX,
Ford Mustang (Autom. Trans.), Dodge St. Regis.
0.3 r-
00
tji
3.
G
0
•H
-P
US
u
4J
G
0)
o
c
o
o
§
0.2 ~
0.1
ND
1 Mercury Marquis
2 Dodge Diplomat
3 Chrysler Newport
4 Ford Mustang (Man. Trans.)
Temperature,
Figure 7. NDMA concentration as a function of temperature:
Mercury Marquis, Dodge Diplomat,
Chrysler Newport, Ford Mustang (Man. Trans.).
38
-------
TABLE 16. VEHICLES SAMPLED
Date
Body Style Manufactured
Vehicle
Freightliner
Ford 9000
White Corp.
Road Boss #2
Freightliner
International
S 1800
Mercedes 116
Diesel
International
Transtar II
GMC Astro
Freightliner
Ford Club Wagon
Ford Super
Wagon
Chevrolet Van 30
Dodge Sportsman
Volkswagen
Vanagon
GMC Vandura 25
Ford Custom F150
Cab over w/sleeper
Conventional Cab
Conventional Cab
Cab over w/sleeper
Conventional Cab
Conventional Cab
Cab over w/sleeper
Cab over w/sleeper
Cab over w/sleeper
4 seat van
Van
Mini Bus
Van
Van
Van
Pickup
March 1978
1978
1979
July 1980
September 1980
May 1980
October 1980
October 1980
September 1979
August 1980
October 1980
March 1980
May 1980
August 1980
October 1980
July 1980
FOR ADDITIONAL ASSESSMENT
Mileage Spare Tire
217,616 N/A
103,879 N/A
124,470 N/A
46 N/A
3,377 N/A
250 N/A
44 N/A
26 N/A
12,987 N/A
2,252 General, Jumbo Steel
Radial
436 Goodyear
14,422 N/A
1,828 Goodyear
15 N/A
21 Uniroyal
6,461 N/A
Interior
Dark Red Vinyl
Brown Vinyl
Brown Vinyl
Dark Red Vinyl/Cloth
Brown Vinyl
Red/Black Vinyl/Cloth
Tan Vinyl
Black Vinyl
Dark Red Vinyl
Blue Vinyl/Cloth
Brown Vinyl
Green/Tan Vinyl
Blue Vinyl/Cloth
Tan Vinyl/Cloth
Tan Vinyl
Blue Vinyl
-------
TABLE 16 (Cont'd). VEHICLES SAMPLED FOR ADDITIONAL ASSESSMENT
Vehicle
Body Style
Chevrolet Impala Station Wagon
Chevrolet Caprice
Oldsmobile
Cutlass Cruiser
Chrysler LeBaron
Mercury Marquis
Winnebago
Sportscoach
Winnebago
Chieftain
Itasca Sunflyer
Winnebago
Sportsman
Station Wagon
Station Wagon
Station Wagon
Station Wagon
Motor Home
Motor Home
Motor Home
Motor Home
Date
Manuf actured
September 1980
September 1980
October 1980
January 1981
October 1980
October 1980
December 1980
September 1980
October 1980
Mileage Spare Tire
3,530 Goodyear Convenience
Spare
6,383 Goodyear Convenience
Spare
4,214 Uniroyal Hideaway
1 General
2,625 General
2,082 N/A
1,924 N/A
4,846 N/A
1,115 N/A
Interior
Tan Vinyl
Tan Vinyl
Tan Vinyl
Red Vinyl
Off-White Vinyl
Orange/Tan/Cloth/
Vinyl
Brown Cloth
Blue/Beige Cloth
White/Orange/Brown
Cloth/Vinyl
-------
TABLE 17. TEST RESULTS FOR NITROSAMINE SAMPLING-TASK V
Vol. Nitrosamine Confirmational Nitrosamine
Samp. Anal. GC-TEA Anal. LC-TEA Levels
Vehicle
Freightliner (3/78)
Ford 9000
White Road Boss
Freightliner (7/80)
International S1800
Mercedes 116 Diesel
International
Transtar II
GMC Astro
Freightliner (9/79)
Ford Club Wagon
Ford Superwagon
Chevrolet Van 30
Dodge Sportsman
Volkswagen Vanagon
GMC Vandura 25
(liters) (ng/Cartridge) (ng/Cartridge)
yg/m-
697
796
835
768
828
770
683
607
805
536
376
376
648
464
456
Ford Custom F150
364
17 NDMA
23 NDMA
18 NDMA
34 NDMA
39 NDMA
86 NDEA
20 NDMA
81 NDMA
9.7 NDEA
9.7 NMOR
68 NDMA
9.7 NMOR
42 NDMA
28 NDMA
29 NMOR
38 NDMA
13 NDEA
7.3 NDMA
11 NDMA
8 NDMA
19 NDMA
7.7 NDEA
21 NDPA
25 NMOR
17 NDMA
23 NMOR
a
a
a
30 NDMA
30 NDMA
55 NDEA
19 NDMA
47 NDMA
ND
ND
92 NDMA
27 NMOR
30 NDMA
25 NDMA
29 NMOR
39 NDMA
12 NDEA
a
a
a
17 NDMA
ND
ND
45 NMOR
a
a
0.024 NDMA
0.029 NDMA
0.022 NDMA
0.044 NDMA
0.047 NDMA
0.104 NDEA
0.026 NDMA
0.119 NDMA
0.014 NDEA
0.014 NMOR
0.112 NDMA
0.016 NMOR
0.052 NDMA
0.052 NDMA
0.054 NMOR
0.101 NDMA
0.035 NDEA
0.019 NDMA
0.017 NDMA
0.017 NDMA
0.042 NDMA
0.017 NDEA
0.046 NDPA
0.055 NMOR
0.047 NDMA
0.063 NMOR
41
-------
TABLE 17 (Cont'd). TEST RESULTS FOR NITROSAMINE SAMPLING-TASK V
Vehicle
Chevrolet Impala
Chevrolet Caprice
Oldsmobile Cutlass
Cruiser
Chrysler LeBaron
Mercury Marquis
Winnebago Sports-
coach
Winnebago Chieftain
Itasca Sunflyer
Winnebago Sportsman
Vol.
Samp.
(liters)
360
376
626
472
Nitrosamine
Anal. GC-TEA
Confirmational Nitrosamine
Anal. LC-TEA Levels
312
275
480
480
253
(ng/Cartridge) (ng/Cartridge)
28 NDMA
12 NDMA
15 NMOR
13 NDMA
183 NDMA
14 NDEA
75 NDBA
27 NMOR
35 NDMA
31 NMOR
ND
ND
ND
ND
24 NDMA
a
a
116 NDMA
8.3 NDEA
36 NDBA
26 NMOR
a
a
a
a
a
yg/rti-
0.078 NDMA
0.032 NDMA
0.040 NMOR
0.021 NDMA
0.388 NDM
0.030 NDEA
0.159 NDBA
0.057 NMOR
0.112 NDMA
0.099 NMOR
ND
ND
ND
ND
aNo LC-TEA confirmational analysis conducted
42
-------
contained NMOR. The GMC Vandura, the only cargo van tested, contained
NDMA, NDEA, NMOR, and N-nitrodipropylamine (NDPA). This was the only
instance in which NDPA was detected in a vehicle interior during the
entire program. However, NDPA was not detected in the LC-TEA confirma-
tional analysis. Total nitrosamine levels in new vans ranged from 0.017
to 0.160 yg/m3, with an average of 0.076 yg/m3.
One pickup truck was tested in this task, a Ford Custom F150. The
Ford pickup contained both NDMA and NMOR at levels of 0.047 and 0.063
yg/m3, respectively.
Five station wagons were tested in this task. All five wagons con-
tained NDMA at levels ranging from 0.021 to 0.388 yg/m3 NDMA. The average
NDMA level was 0.126 yg/m3. The Caprice and Marquis also contained NMOR
at 0.040 and 0.057 yg/m3, respectively. The Chrysler LeBaron contained
four nitrosamines, NDMA, NDEA, NDBA, and NMOR. All four were confirmed
by LC-TEA. The total nitrosamines ranged from 0.021 yg/m3 for the Olds-
mobile Cutlass to 0.634 yg/m3 for the Chrysler Lebaron. The average
nitrosamine concentration for the five station wagons was 0.203 yg/m3.
Four motor homes were tested in the task. None of the four had a
detectable level of nitrosamines in the vehicle interior. This was the
only vehicle group tested in the program in which no detectable levels
of nitrosamines were found.
F. Task VI - Exposure Assessment
This section gives the results for estimating exposure levels of
several population categories to nitrosamines from vehicle interiors.
The respiration rate of an average male during operation of a vehicle
has been estimated to be 5 to 10 liters of air per minute.(5) The upper
value, 10 £/min, has been used in all of the following estimates to
simplify calculations. The assumption that 100% of the inhaled nitros-
amines were absorbed by the body was also made.
If a commuter is exposed to nitrosamines for 3 hours/day in a moving
vehicle and for 10 minutes a day in a vehicle at rest, the maximum daily
exposure can be calculated using the data from Tasks III and V. The
maximum concentration of nitrosamine in a vehicle interior was 0.63 yg/m3
(Task V) and the maximum level in a moving passenger car was found to be
one fifth of the at-rest concentration (Task III), or 0.13 yg/m3 when
applied to this maximum case. Using this data, the amount of nitros-
amines a commuter is exposed to per day can be calculated as follows:
(0.63 yg nitrosamines ^ 10 I „ 0.001 m3
mass nitrosamines = ^^j
jjl.O x 10 min^ + ^0.2 * 3 hr x 6°h"in)j
= 0.3 yg nitrosamines
43
-------
If the average nitrosamine concentration (not including the 4 motor homes)
found in the project, 0.08 yg/m3, is used in the calculation, then the
exposure to a commuter would be:
(0.07 x 10 x 0.001) [(1.0 x 10) + (0.2 x 3 x 60)] = 0.04 yg nitrosamines
The data in Tasks III and V can be used to calculate other daily
exposure categories, such as the truck driver. A maximum exposure case
would be a truck driver who operates his truck for 16 hours and sleeps
the remaining 8 hours in the truck. The highest nitrosamine concentration
for a heavy-duty truck was found to be 0.15 yg/m3 (Task V). The concen-
tration in a moving truck was found to be one seventh of the at-rest con-
centration (Task III) . Using these data the maximum esqposure of a truck
driver would be:
(0.15 x io x 0.001) [(1.0 x 8 x 60) + (0.143 x 16 x 60)] = 0.9 yg
nitrosamines
If the average nitrosamine concentration in heavy-duty truck interiors
is used, 0.09 yg/m3, then the daily exposure would be 0.6 yg of nitrosamines.
The eight hours sleeping in the truck contributes 0.4 yg of this value.
The average interior nitrosamine concentration for a van at-rest is
0.08 yg/m3 (Task V). A van in operation contains approximately one-half
the at-rest concentration (Task III) . If the driver of a van is exposed
to 7 1/2 hours of nitrosamines in a moving vehicle and 30 minutes in a
vehicle at-rest, then the exposure would bes
(0.08 x io x 0.001) [(1.0 x 30) + (0.5 x 7.5 x 60)] = 0.2 yg nitrosamines
A worst-case situation for daily exposure would exist for a person re-
maining in a closed, non-moving vehicle for 24 hours. If the highest
nitrosamine concentration for a vehicle interior obtained in the program
is used (0.63 yg/m3), maximum exposure would be:
0.63 x 10 x 0.001 x 24 x 60 = 9 yg nitrosamines
44
-------
IV. DISCUSSION OF THE RESULTS
The response from the scientific community for the most part
indicates an acceptance of the TEA analyzer and the ThermoSorb/N traps
for nitrosamine analysis and collection. Researchers who have used
both the TEA analyzer and the ThermoSorb/N traps feel that both are
the best available for routine nitrosamine analysis.
Of the fifty-eight vehicles sampled in the program, forty-nine
contained NDMA (0.014 to 0.388 yg/m3), fourteen contained NMOR (trace
to 0.360 yg/m3), eight contained NDEA (0.011 to 0.104 yg/m3), one
contained NDPA (0.046 yg/m3), and one NDBA (0.159 yg/m3). Total nitros-
amine levels ranged from 0.01 to 0.63 yg/m3. The concentrations in
general are lower than those found by Fine for new 1979 passenger cars,
0.07 yg/m3 to 2.91 yg/m3.d'2)
Testing in the program indicated NDMA dependence on mileage and
time with NDMA levels decreasing slightly with time and mileage. It
showed NDMA dependence on temperature, with higher temperatures giving
higher levels of NDMA. Detectable levels of NDMA were found in vehicles
during operation (e.g., one fifth of the at-rest concentration for a
passenger car if all windows are closed and the air conditioner and
heater off). Similar levels of NDMA were found in vehicles sampled
after being closed overnight or sampled immediately after operation.
NMOR and NDEA were detected randomly in the program, and no trends were
observed for these nitrosamines.
Nitrosamines were found in passenger cars, station wagons, pass-
enger and cargo vans, pickup trucks, and in new and used heavy-duty
trucks. Nitrosamines were not detected in four motor homes tested in
the program.
The nitrosation potential and level of nitrosatable amines of the
ambient air were monitoted throughout the program. No detectable nitros-
amine trends with NO2 or amines were observed in the program. Nitros-
amines were found in three of eleven background nitrosamine samples
taken during the program. In one background sample, NDMA was detected
in only one of two duplicate analyses; in the second sample, the randomly
appearing nitrosamine NMOR was detected; and in the third background
sample, NDMA was detected at the detection limit of the analytical
procedure.
Other common sources of nitrosamines in the environment include
beer and bacon. Dr. Fine at NEILS has estimated that one can of beer
contains 1 yg of nitrosamines, and one strip of bacon contains 0.1 yg
of nitrosamines.d/2) If these values are accurate and if inhalation
and ingestion are comparable routes of exposure, then the average daily
exposure of a commuter would be greater from either a can of beer or a
strip of bacon than from a vehicle interior (0.04 yg of nitrosamines).
45
-------
REFERENCES
1. Rounbehler, D. P., et al, Fd Cosmet. Toxicol., Vol. 18, pg 147, 1980.
2. Walker, E. A., et al (Ed.), N-NITROSO- Compounds: Analysis, Formation
and Occurrence, International Agency for Research on Cancer, Lyon,
IARC Scientific Publication No. 31, pg 541, 1980.
3. Rounbehler, D. P., et al, Anal. Chem., Vol. 52, pg 273, 1980.
4. Dietzmann, H. E., et al, "Analytical Procedures for Characterizing
Unregulated Pollutant Emissions from Motor Vehicles," EPA Report
600/2-79-017, February 1979.
5. Private Communication with Jeffrey Waugh, Southwest Foundation for
Research and Education, San Antonio, Texas.
47
-------
APPENDICES
A. "Instructions for Monitoring"
B. Description of the GC-MS System
(as Reported to SwRI by NEILS)
C. Description of the Nitrosation Capacity
Determination (as Reported to SwRI by NEILS)
D. Description of the Nitrosatable Amine
Determination (as Reported to SwRI by NEILS)
E. Task II Trunk and Cargo/Back Seat Sampling
Results
-------
APPENDIX A
"Instructions for Monitoring"
-------
ThermoSorb/N Air Sampler
Instructions for Monitoring
Introduction
Air is drawn through a proprietary sorbent with a suitable
air sampling pump. The N-nitroso compounds are absorbed
with high efficiency. After sampling is complete, the sorbent
is eluted with solvent to remove the N-nitroso compounds. The
solvent is then analyzed by combined gas-liquid chromato-
graphy with TEA Analyzer. Detection limits of better than 0.05
/jg/nrW are possible when sampling for one hour at 2.0 L/min.
The ThermoSorb/N air sampler contains:
ThermoSorb N
Catalogue Number 6533
Contents
20 ThermoSorb/N air samplers in foil pouches
20 Foil pouch clips
20 Data log work sheets
1 Instruction sheet
Other Equipment Needed for Monitoring:
1. Air sampling pump (high flow or low flow)
2. Battery charger
3. Pump calibration — soap bubble tower
4. Tubing, 1/4 in., flexible
5. Stopwatch
6. Scissors
Preparation Before Sampling:
fftermoSorb/N
. j-gy'-P1' •' tor 111" '"liecbon
' '!l>" in air
1. Remove the ThermoSorb/N air sampler from the foil pouch.
Use scissors to cut open the foil pouch. Save the foil pouch
for re-use.
2. Remove the red end caps from the inlet and outlet ports.
The red caps can be stored on the ThermoSorb/N air
sampler in the brackets under the "AIR IN" sign.
Catalogue Number 6525
Contents
10 ThermoSorb/N air samplers in foil pouches
10 Foil pouch clips
10 Data log work sheets
10 Mailing envelopes for
10 Analyses at the Analytical Services
Laboratory of Thermo Electron Corporation A-2
1 Instruction sheet
ThermoSorb N
-------
3. Label the ThermoSorb/N air sampler with the peel-off "Air
Sampler" label provided on the "Data Log" worksheet. The
molded clip of the ThermoSorb/N air sampler provides a flat
surface to affix the label.
4. Attach the ThermoSorb/N air sampler to the sampling pump
using an appropriate length of 1/4 in. flexible tubing.
5. Calibrate the pump with the ThermoSorb/N air sampler at-
tached. Use a stopwatch and bubble tower to determine the
air flow. A 2.0 L/min flow rate is suggested for general
monitoring. Flow rates for the ThermoSorb/N air sampler
can vary between 0.2 Umin to 4.0 L/min without affecting
collection efficiency.
6. Record the air flow and all other appropriate data on the
"Data Log" worksheet. The "Data Log" worksheet can be
readily applied to a laboratory notebook page.
Sampling:
1. Attach the ThermoSorb/N air sampler
3. If high concentrations of nitrosamines are expected (i.e.,
over 1500 ^g) use another ThermoSorb/N air sampler as a
"back-up") section.
4. Note changes in monitoring conditions (the "Air Sampler"
label on the "Data Log" worksheet can be used for notes in
the field), for example:
a) Obstructions in the ThermoSorb/N air sampler.
b) Changes in flow rate.
c) Changes in ambient temperature, barometric pressure, or
relative humidity.
5. Remove the ThermoSorb/N air sampler from the monitoring
site.
After Sampling:
1. Calibrate the pump with the ThermoSorb/N air sampler
attached.
2. Detach the ThermoSorb/N air sampler from the pump.
3. Replace the red end caps on the inlet and outlet ports of
the ThermoSorb/N air sampler.
4. Record the appropriate data, be sure to include the flow
rate after sampling.
a) In the breathing zone of the worker to be monitored. The
molded clip attaches easily to pockets or collars.
b) Near the process to be monitored. The molded clip pro-
vides a flat surface so that the ThermoSorb/N air sampler
can be easily oriented toward the area of interest.
c) On the pump, in the area to be monitored.
2. Sample for an appropriate period of time. 100L of air total
volume is the recommended sample size.
a) For time weighted averages (TWA's) use 0.2 L/min for 8hrs.
b) For process sampling use 2.0 L/min for 50 min or 4.0
Umin for 25 min.
.Affix the "Analysis Vial" label from the "Data Log"
worksheet to the ThermoSorb/N air sampler. This can be
done on the side of the "AIR IN" sign. The laboratory will
then be able to label the analysis vial with a label numbered
the same as the ThermoSorb/N air sampler.
A-3
ThermoSorb N
-------
ThermoSorb//y
> ° "i air
imts
; ¦^«nvMAar!'J41
¦
6. Replace the ThermoSorb/N air sampler in the foil pouch.
Fold the pouch and seal it with the clip provided.
7. Place the sealed foil pouch into a mailer and submit for
analysis.
8. Affix the "Mailer Label" from the "Data Log" worksheet as
a seal on the back flap of the mailer.
Analysis by Thermo Electron:
"I.The Analytical Services Laboratory at Thermo Electron
Corporation will report results of analysis within 3 working
days after the sample is received.
2. The Laboratory reports results in nanograms.
3. Calculate the concentration of the substance monitored by
dividing the results in nanograms by the average volume of
air sampled in liters. Results will be in Hg/m •
For information on analysis, request publication IS-33,
"ThermoSorb/N Air Sampler Analysis Instructions".
For further information:
Call: (617)890-8700
Write: Thermo Electron Corporation
Analytical Instruments
Waltham, MA 02154 U.S.A.
Telex: 92-3473
Ask for Order Entry, Analytical Instruments, regarding
purchase of additional ThermoSorb/N air samplers.
Ask for ThermoSorb/N Customer Service for questions
regarding air monitoring.
Ask for Analytical Services Laboratory for questions
regarding analysis of the ThermoSorb/N air samplers.
A-4
-------
APPENDIX B
Description of the GC-MS System
(as Reported to SwRl by NEILS)
-------
Description of the GC-MS System
The GC-MS System consisted of a glass capillary column coupled
directly to the ion source of the mass Spectrometer. The column was
20m x 0.25mm i.d. glass, coated with Carbowax 20M. The mass Spectro-
meter was an AE1 MS50 instrument operated at a resolution of 10,000
(10% valley). It was fitted with a peak matching unit and a Daly
detector. Hexapole focusing enabled the slits to be fully extended.
The accelerating voltage was 8KV and ion current 3001JA. The mass
Spectrometer was calibrated using a standard solution of N-rnitrosodi-
methylamine (NDMA) . 0.3 V& sample was injected onto the column via an
SGE splitless injection unit. NDMA was detected by monitoring the
parent ion (m/e 74.0480) with reference to the fragment m/e 69.9986
of perfluorokerosine. The detection limit of the GC-MS was 1 yg/£
of injected material, representing 0.3 pg NDMA.
B-2
-------
APPENDIX C
Description of the Nitrosation Capacity Determination
(as Reported to SwRI by NEILS)
-------
Description of the Nitrosation Capacity Determination
These samples were collected using an experimental air collector
consisting of a ThermoSorb air cartridge filled with a solid sorbent
coated with morpholine. These air samples are usually collected using
battery driven air pumps with an air flow rate maintained as close to
1 %/min as possible.
In previous experiments in the laboratories of NEILS, 15 of these
collectors had been tested in atmospheres containing from 0.5 to 3,0 ppm
of equal amounts of NO and N02/ i.e., 0.25 ppm N02 to 1.5 ppm N02. Air
was drawn through these collectors at 1 £/min for 50 min at 5 different
concentrations of N0X. All tests were run in triplicate. The average
amount of N-nitrosomorpholine formed on the three cartridges for each
concentration of NO2 were plotted against the concentration of the
square of the NO2 [N02]2. The resulting plot was linear with a slope
of 0.81 and had a correlation coefficient of 0.99. The total amount
of NMOR formed indicated that 6-8% of the available N02 reacted with
the morpholine under the test conditions. While the tests were in
progress, measurements of the N0X in the air stream were made at both
the entrance and exit of the cartridges. It was observed that the NO
levels did not decrease after passage through the cartridge suggesting
that this oxide of nitrogen does not contribute to the formation of the
resulting NMOR. These cartridges are being developed to aid in deter-
mining the comparative amounts of nitrosating agents in industrial
atmospheres. However, the nitrosation capacity of airborne oxides of
nitrogen have been found to vary with humidity and it is possible
that other factors such as ozone may effect the results. While the
results from this sampling experiment are only indicative of the
absolute amount of N02 in the atmosphere, they are useful in determining
the qualitative capacity of any given atmosphere to form N-nitrosomorpholine.
The methods for analyzing these cartridges are identical to that used for
the ThermoSorb/N air cartridges.
As a first approximation, the N02 levels that experimentally produce
the observed amounts of N-nitrosomorpholine eluted from these air sampling
devices are:
PPM m2 = w ¦oa/t * x m „ ii™ x
.. M.W. NOo yq of formed NMOR
"9 2/^ - M.W. NMOR * t of air sampled
*Experimentally 6-7% of the N02 in the gas stream combines with the
surface held morpholine to form N-nitrosomorpholine when the devices
have air containing NOx drawn through them at 1 £/min for 50 min.
Example: yg of NMOR formed on collector = 0.36
I of air sampled = 52
_n 46 x 0.36 24.45 , . 1 <\j - -
Then PPM NO? = —— — x —-r— x 1 x 1 x ¦ ¦ • = 0.021
*¦ 116 x 52 46 0.07
C-2
-------
APPENDIX D
Description of the Nitrosatable Amine Determinations
(as Reported to SwRI by NEILS)
-------
Description of the Nitrosatable Amine Determination
The methods used in this study for sampling and analysis of airborne
amines have not been previously described. Amine air samples are col-
lected using ThermoSorb/Amine air cartridges. The methods used to
examine the cartridges for amines are similar to those used for the
analysis of nitrosamines on ThermoSorb/N air cartridges. Following air
sampling, the trapped amines are extracted from the cartridge by back-
flushing with methanol with the first 3 mil collected for analysis.
Aliquots of this eluate are then examined by injecting on a Gas Chroma-
tograph interfaced to a prototype TEA Analyzer operating in the nitrogen
mode. The prototype TEA Analyzer (nitrogen mode) used for this study
consisted of a Shimadzu Mini 2 GC operating at from 110°C to 180°C, with
a flow of 20 cc/min of argon gas through a 1/8" x 12" stainless steel
column packed with either 10% Pennwalt 223 on Gas ChromeR or Chromosorb
103 interfaced to a catalytic oxidative pyrolytic furnace. Detection
of the NO radical formed from the high temperature oxidation of the
amines was by a standard TEA Analyzer. Tests of this sampling and
detection method using dimethylamine and morpholine, have demonstrated
it to be linear from 0.8 ug/m3 to 200 yg/m3 with an average recovery of
about 90% at all spiking levels. In all respects the detection of amines
by this method is similar in linearity and detection limit to that of the
standard GC-TEA method for N-nitroso compounds. As a further control,
some of the "amine air samples" are examined by a standard Varian GC
equipped with an alkaline flame ionization detector (AFID).
D-2
-------
APPENDIX E
Task II Trunk and Cargo/Back Seat
Sampling Results
-------
APPENDIX E. TASK II TRUNK AND CARGO/BACK. SEAT SAMPLING RESULTS
Nitrosamine Levels, yg/m^
Vehicle
Location3
Date
Temp.°F
NDMA
NMOR
Fairmont
T
11/2/79
45
0.046
ND
Fairmont
T
1/30/81
34
0.027
ND
Fairmont
T
2/17/81
57
0.075b
0.038"
Fairmont
T
3/18/81
27
0.040
ND
Fairmont
T
4/23/81
64
0.152d
0.040'
Oldsmobile
B
7/23/79
90
0.470
0.480
Oldsmobile
B
7/31/79
86
0.115
0.511
Oldsmobile
B
6/17/80
75
0.637
3.000
Oldsmobile
B
2/17/81
57
0.058
ND
Oldsmobile
B
3/18/81
27
0.012
ND
Oldsmobile
B
4/27/81
61
0.031e
0.044'
aLocation whe re sample w.
^Confirmed by LC-TEA to
cNot detected by LC-TEA
^Confirmed by LC-TEA to
eConfirmed by LC-TEA to
is monitored: T, trunk;
be 0.075 ug NDMA/m3
be 0.140 yg NDMA/m3
be 0.035 yg NDMA/m3
B, backseat/cargo area
E-2
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-460/3-81-029
3. RECIPIENT'S ACCESSION*NO.
4. TITLE AND SUBTITLE
NITROSAMINES IN VEHICLE INTERIORS
5. REPORT DATE
September 1981
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Lawrence R. Smith
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG\NIZATION NAME AND ADDRESS
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-03-2884
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Mobile Source Air Pollution Control
2565 Plymouth Road
Ann Arbor, Michigan 48105
13. TYPE OF REPORT AND PERIOD COVERED
Final-June 1980-Mav 1981
14. SPONSORING AGENCY CODE
16. SUPPLEMENTARY NOTES
16. ABSTRACT
Researchers in the nitrosamine field were contacted on their views of
the TEA analyzer and ThermoSorb/N Air Samplers for nitrosamine analysis.
Gas samples were taken from vehicle interiors to determine the effects
of vehicle type, vehicle age, mode of operation, the ambient conditions
on interior nitrosamine levels. A total of fifty-eight vehicles were
sampled in the program. Occupant es^osure levels were estimated using
test vehicle data.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Held/Group
Air Pollution
Motor Vehicle
Nitrosamines
Nitrosamine Analysis
Nitrosamine Collection
Occupant Exposure
Vehicle Interiors
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO, OF PAGES
70
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |