Revised Draft Report
Nonoccupational Asbestos Exposure
EPA Contract No. 68-02-4254
Task No. 31
Prepared for:
U.S. Environmental Protection Agency
Office of Toxic Substances
401 M Street, SW
Washington, DC 20460
Prepared by:
Versar Inc.
6850 Versar Center
Springfield, VA 22151
September 25, 1987
-------
ACKNOWLEDGEMENTS
This project was funded by the U.S. Environmental Protection Agency,
Office of Toxic Substances, Exposure Evaluation Division, Exposure
Assessment Branch. This report was prepared by Versar Inc. of
Springfield, Virginia, in response to EPA Work Order 31 of Contract No.
68-02-4254.
Acknowledgement is given to Elizabeth Bryan, the EPA Project
Officer, and Lynn Delpire, the EPA Task Manager, who provided support and
direction throughout this effort.
A number of Versar personnel have contributed to this effort:
Program Management - Gayaneh Contos
Task Management - CI ay Carpenter
Technical Staff - Cindy Lewis
Dennis HIinka
Renaldo Jenkins
Secretarial Staff - Lynn Maxfield
Kammi Johannsen
i i i
-------
TABLE OF CONTENTS
Page No.
1. INTRODUCTION 1-1
1.1 Background 1-1
1.2 Report Scope and Organization 1-1
2. AIRBORNE ASBESTOS FROM VEHICLE BRAKES 2-1
2.1 National Asbestos Emissions 2-3
2.2 Asbestos Emissions from 24 Cities 2-11
2.3 Ambient Asbestos Concentrations Resulting from
Vehicle Brakes 2-15
2.4 Average National Ambient Asbestos Concentration from
Vehicle Brakes 2-22
2.4.1 Method 1: Population 2-22
2.4.2 Method 2: National Emissions 2-26
3. CONSUMER EXPOSURE TO ASBESTOS IN COATINGS, SEALANTS, AND
PAINTS 3-1
3.1 Roof Coatings 3-1
3.2 Waterproof/Corrosion Resistant Coatings 3-4
3.3 Driveway Sealants/Epoxy Adhesives/Glazing
Compounds 3-4
4. CONSUMER EXPOSURE TO ASBESTOS IN VINYL ASBESTOS FLOOR
TILE 4-1
4.1 Removal of the Floor Tile by the Consumer 4-1
4.2 Floor Tile Wear 4-1
5. CONSUMER EXPOSURE DURING BRAKE REPAIR 5-1
5.1 Exposed Population 5-1
5.2 Consumer vs. Occupational Exposure 5-2
5.3 Estimated Exposure 5-6
6. OTHER CONSUMER EXPOSURES 6-1
7. REFERENCES 7-1
APPENDIX A - Determination of Ambient Asbestos Concentrations
from Vehicle Brakes ........ A-1
APPENDIX B - Secondary Concentrations of Asbestos in a Garage
Occurring from Brake Repair B-l
iv
-------
LIST OF TABLES
Page No.
Table 2-1. Results Reported in Previous Studies 2-2
Table 2-2. Airborne Asbestos Emissions from Vehicles 2-4
Table 2-3. Airborne Asbestos Emissions by Vehicle Type 2-5
Table 2-4. Input Data for Emissions Estimates 2-6
Table 2-5. Cities for Which Concentrations Were Estimated ... 2-12
Table 2-6. Estimated Car, Light Truck, Heavy Truck, and
Combined Emissions for 24 Cities 2-14
Table 2-7. Estimated Annual Asbestos Concentrations from
Vehicle Brakes in 24 American Cities ............. 2-18
Table 2-8. Worksheet for Estimating National Ambient Asbestos
Levels by a Population Correlation 2-25
Table 3-1. Coatings, Seal ants, and Paints that May Contain
Asbestos and Be Used by Consumers 3-2
Table 3-2. Consumer Exposure to Asbestos in Coatings,
Seal ants, and Paints 3-3
Table 4-1. Inhalation Exposure to Asbestos from Vinyl Tile .. 4-2
Table 5-1. Estimated Number of Garages in Owner-Occupied
Housing 5-4
Table 5-2. Estimated Number of Garages Associated with Renter-
Occupied Housing 5-5
Table 5-3. Estimated Consumer Exposure from Brake Dust
Brushing 5-7
Table 5-4. Estimated Consumer Exposure During Brake Repair .. 5-9
Table 6-1. Consumer Product Categories Thought to Contain
Asbestos' 6-2
v
-------
LIST OF FIGURES
Pace No.
Figure 2-1, Modeled area of Baltimore City 2-16
Figure 2-2. Estimated asbestos concentration in the air of
10 large cities resulting from vehicular brakes . 2-19
Figure 2-3. Estimated asbestos concentration in the air of
8 medium-sized cities resulting from vehicular
brakes 2-20
Figure 2-4. Estimated asbestos concentration in the air of
6 small cities resulting from vehicular brakes .. 2-21
Figure 2-5. Average relative contribution to the total
average asbestos concentration 2-23
vi
-------
1. INTRODUCTION
In January 1986, the Office of Toxic Substances (OTS) proposed a rule
that banned the use of asbestos in certain products and phased out the
use of asbestos in all other products. The public comment period for
this rule closed on June 30, 1986, and public hearings were held
July 15-25, 1986. OTS is now charged with updating previous work, based
on recent literature references and data obtained during the review
process. This report presents updated information on certain
nonoccupational exposures to asbestos.
].1 Background
Asbestos is a fibrous si 1icate mineral valued for its noncombustible,
nonconductive, chemically inert properties. It has been used in a wide
variety of products including building components, insulation, plastics,
textiles, pipe, friction products, and fi1ters. Exposure to asbestos has
been linked with 1ung cancer, mesothelioma, and asbestosis.
In 1984, Versar prepared an asbestos exposure assessment under
Task 49 of EPA Contract 68-01-6271. Along with other information, this
exposure assessment was used in preparing the 1986 proposed rule. Based
on more recent information, however, certain aspects of this report need
to be updated. The more recent information includes (1) testimony given
in the public hearings and in public comments, (2) the asbestos NESHAP
review, (3) a survey of asbestos mining, milling, and processing
companies that was conducted by ICF Corporation, and (4) recent
literature references, which have been identified through a 1iterature
search.
1.2 Report Scope and Organization
The scope of this report is limited to specific nonoccupational
exposures designated by the Office of Toxic Substances. The topics
covered and their organization within this report are as follows:
1-1
-------
Topic Section No.
Airborne Asbestos from Vehicle Brakes 2
Consumer Exposure to Asbestos in
Coatings, Sealants, and Paints 3
Consumer Exposure to Asbestos in
Vinyl Asbestos Floor Tile 4
Consumer Exposure During Brake Repair 5
Other Potential Consumer Exposures 6
1-2
-------
2. AIRBORNE ASBESTOS FROM VEHICLE BRAKES
Both disc and drum brakes contain asbestos fibers as friction
material. When either type of brake is applied, the friction material is
mechanically abraded and thermally degraded. The abrasion of the
friction material often results in airborne asbestos. Because most of
the 170 million vehicles registered in the U.S. in 1985 were equipped
with brakes capable of emitting asbestos to the atmosphere, the quantity
of asbestos emitted by each vehicle and the contribution of brakes to
ambient asbestos levels, are of concern. Several studies have been
undertaken to identify these parameters.
The results of these studies are summarized in Table 2-1. As shown
in this table, it is difficult to compare results among studies because
of differences in focus, basic assumptions and reporting methods. Three
of the studies with similar focuses, but different methods were Williams
and Muhlbaier (1982); Jacko, DuCharme, and Somers (1973); and Cha and
Carter (1982). Although these three studies use different approaches to
the question of ambient asbestos levels resulting from vehicle brakes,
they all publish results of their studies in enough detail to allow
additional analyses. The other studies 1isted in Table 2-1 either
contain insufficient data to allow further analysis or determine factors
that do not apply to the current analysis. For these reasons, the
remainder of this section will focus on the data collected for these
three studies.
Two factors are common to all three studies: (1) transmission
electron microscopy is used to quantify asbestos levels, and (2) the
analyses performed on the data collected use vehicle registration,
mileage, and use statistics which are now outdated. To the extent
possible, this report will apply consistent methods and conversion
factors to the published data to result in up-to-date, comparable
estimates of national asbestos emissions. The arithmetic average of
these values will then form the basis for estimates of asbestos emissions
2-1
-------
6/71N
Table 2-1 Results Reported in Previous Studies
Reference
Result
Conments
Williams and Munlbaier
! i 9eL')
0,063 ng/m ambient
asbestos concentrdtion
from braking
Assumes only passenger cars contribute.
Calculated by analogy with airborne lead.
4,/00 kg/yr
Airborne asbestos emissions of ? 6 jg/km,
J
with 1.14 x 10 passenger cars, e-.icn
4
averaging 1.6 x 10 km/yr (197? figures).
Raw data from dynamometer tests.
19 ng/m toll booth
asbestos concentration
By analogy with the General Motors Sulfate
Dispersion £xperiment.
¦iocka, Dt-Chijrme, ana
vomers lly7j)
5.060 lb/yr airborne
asbestos from c lutcn and
orak. ing
Includes asbestos from disc and drum Drakes
and clutches for passenger cars, trucks, and
unset 1 ianeous vehicles. Raw data from veni-
c le tests.
Ciu ana Carter (tS6t)
0.25 ng/m ambient
asbestos concentration
from automcbiles
Uses dispersion model for St. Louis, MO, mm
emissions estimated by combining measured
results for disc brakes from representatlve
braking cycles with estimated values for drum
brakes. Accounts for autamooiles only.
cmyrutp et a 1 (.ii»ti]
to 25 ng/m
Ambient asoestos concentration resulting from
automotive friction products. Iheoret :co i
bas i s.
Af.ttfcrson et a i (!9?3j
.0?/.
Asbestos content of emissions. Dynamometer
tests.
2 iskind et a 1, (15B2)
31,000 to 42,000
3
fibers/m
Chrysotila asDestos concentrations measured
beside a Century City, California, rigrvay
Ron 1 et a 1. (1576)
2-lb/, asbestos in brake
drum dusts
Asbestos content of dust blown out of brake
during maintenance.
? 6-16.0 f ibers/m"'
Range of means 3 to 20 feet from maintenance
operation (blowing dust out of brake drums
with compressed air).
Will idini ana Muhltkt ter
(1980)
0.Q8 ng/m ambient
asbestos concentration
from braking
4,900 kg/yr total asbestos
emissions from braking
Assumes only passenger cars contribute.
Calculated by analogy with airborne lead
0 03;'. of the total brake emission is asbestos.
2-2
-------
from 24 United States cities. The city-wide emissions will be used, in
conjunction with ambient air dispersion modeling, to estimate the
asbestos concentration in these cities resulting from vehicular brakes,
finally, an average national ambient asbestos concentration from vehicle
brakes is estimated by two methods.
2.1 National Asbestos Emissions
In order to use the data compiled in the three studies mentioned
previously, background information on the studies must be known. The
methods used in each of the three studies will be outlined in this
section, followed by a detailed discussion of the manipulations performed
on these data to derive comparable asbestos emissions estimates.
Table 2-2 summarizes pertinent data from each of the three studies, along
with comparable emissions estimates based on these data. These estimates
result from equating emissions from trucks, buses, and motorcycles with
those from cars, and, therefore, misrepresent total asbestos emissions.
To avoid this problem, emission factors were used to scale automobile
emissions up or down in order to reflect emissions from trucks, buses,
and motorcycles. The results of these manipulations are presented in
Table 2-3; bases for these manipulations are summarized in Table 2-4. A
more detailed discussion of Tables 2-2, 2-3, and 2-4 follows the
descriptions of the studies conducted by Williams and Muhlbaier; Jacko,
DuCharme, and Somers; and Cha and Carter.
Wi11iams and Muhlbaier (1982) performed dynamometer tests on new and
used disc and drum brakes. They collected airborne, sedimentary, and
entrained particles, and analyzed samples for asbestos content. These
measurements allowed the researchers to determine directly such factors
as the mass of particulates emitted, the percentage of particulates that
remain airborne, and the asbestos content of airborne emissions. The
experiment was repeated for different initial velocities and deceleration
rates. By averaging the driving cycles and the emissions, they estimated
2-3
-------
? ? A n tirji ne A^hf Ma*. Inn,' irm » row Vt*i tl. lc*'.
Study (date)
Type of
test
Brake
type
Avrraae
par t ItU1flte
emission
(ma/brate/
stop)
flvL ragf
p< ¦ tertflae
of
part irulates
a irhor rie
«>)
A -rborne
Cjft iru^ale en 155 ions
fiange Average
(mg/b r a ke / (mg/b rake/
step J stop)
Asbestos content
of airborne
partiLuiates
Range
ftirbarne asbestos emission
Mean (ug/veh/stop) (ug/veh/kr) (kg'yr)
rv
i
4s»
Willtams and Dynamometer Disc
Muhlbaier (198?)
Disc
Drum
Jacko, OuCharme. Vehicle Disc
and Soners (1973)
Drum
Disc
Drum
Cha and Carter Dynamometer Disc
(198?)
3,J
D.70-1?,0
0.75-2 8
1 6
r
0 082-1 zr 0.39
0,004-0.18Jc 0 06
0.082-0.364^ 0 0Sle
0 042-0.181d 0 me
? 4?.
31
0.030
0,00095-0,18 0.0??
0.0030-0.19 0.031
0.052-0,290 0 Hr«
0.079-0.493 0.216
0,052-0.290 0.159
0.079-0.493 0.216
0.00005-0.14 0.01?
2.2
1.55
1.59
0 72
0.43
2.64
1.86
.91
0,86
1 JO9
7,000
4,900
5,000
2,300
3,400
Conversion factors used in calculating consistent units {except as noted) are; 1.2 stops/km (2 0 stops/mile) (Jacko, Oucharme and Sowers 1973),
1.64x10® autos. trucks and buses registered in ll.i in 1983 (USOOt 1985a), avg. of 10,000 miles traveled per vehicle {USD0C 1985a), 2 dri/n and 2 disc
brakes, mean (or average) values.
bHed tan values.
cResults for seven different driving cycles (e.g., city driving, severe stopping) are included.
^Severity weighting factors are available for only five of the seven driving cycles, therefore, only those five are included in this range.
eAverage value weighted by severity of braking mode. These values are assured to be snore accurate than the non-weighted averages; therefore, they are
used in the remainder of this report.
includes contribution from two drum brakes. The values for the diun brakes were estimated by applying the ratio of airborne particulate emissions from
drum and disc brakes found by Williams and Huhlbater (1982) (1.1/S.t) to the airborne asbestos emissions from disc brakes estimated in the current study.
^Assumes 5.05 stops per mile (3.03 stops'ki"). measured by author during this test (inner-city driving).
-------
Kil-k ? 3. A irJiOT rtt' t o;< f t > r.'. *rr • h> Vf«hn typ*?
Data source
Brake
type
Ave-aqp
airborne
part icu late
emissions
(mg/brake/
stop)
Mean
asbestos
content
of airborne
particulates
i>)
A'r;-3<~rie esbes-toi e* issues
Total air
borne
as!;esto;
er'siiors from veh«cular brakes
cars
buses
(ug'veh,'kiri)a' 'J
I ight Med Heavy
truck truck truck
Motor -
eye le
car
bus
t lght
t ruck
(fcg/yr)b
Med Heavy
t ruck t ruck
KatOr-
cycle total
Williams and
disc
1, 6C
0.0 21
Kyhlbaier
1.66
10.66
4.1 10 66 28.7
0.93
3,900
120
2.300
270 3,500
18 10,000
(1982)
drum
l.lc
0.031
Jacko,
disc
0.091d
0.153
OiiCharme
0.86
6.6B
2.S7 6.66 17.99
0.43
1.800
?'
1,400
170 2.200
8.3 5,690
& Somers
drum
0.099d
0.216
(1373)
Cha and
disc
0,?46e
0.01?
Carter^
1.30
6.87
2.63 6.87 18.5
0.6'.4
2./00
71
1,500
180 2,300
13 6,800
(1982)
drum
0.S13e'®
0.017
aE*cept as noted, assumes 1 2 stops/km
^Emissions estimates were based on data in Table 2-4
cMedian values.
^Average values weighted by severity of braking mode. These values were assumed to be more accurate than the non-weighted averages, and are, therefore,
the values used in the remainder of this report
e!hese values are reported as they appear in the reference, they are not consistent with those derived froo values noted in Table 2 2.
^tha and Carter (158?) report average of 5 OS stops/mile [3.03 stops/km),
Estimated by applying the ratio of airborne particulates emitted by disc brakes to that nutted by drum brakes determined by Williams and HuMbaier
(193?),
Example calculation, for light trucks, (1. lxl0"3o/fdrum brake) * (stop)] * (I ? stops/km) * (? d-urti brakes/axle) * {0, 00031) * (S/light truck/a*te)
x (>)0Bug.'g) « 4,1 ug/veh/km. Hole: Although the correct units for airborne asbestos emissions are ug/ (vehic le * km), they are reported as
ug/veh/km to be consistent with convent ion The unit less emus ton factor (2.5 for light trucks) is based on the relative weights of friction material
in truck and bus brakes as compared to that in car brakes (Jackc. DuCharnie 8, Sowers 1973). The product of this factor tires the number of a* les per
vehicle results in the er-ission factor (m (vi-h/axle) ') given in Table 7-1 and the original reference (Jacko, DuCbarme and Sowers 1973) for each
vehicle class,
-------
Table 2 4 Input P
Vehicle class
Nurber
Ccmnent
(source)
Average
distance
t rave 1 led'5
(km/yr)
torment
(source)
Brake
conflgjra*ionc
[itiss icn
factor^
(veb/axle)
-I
Ctmne it
(source)
ro
cr»
Passenger cars 13Z.108.164 1985 Regist rat ion f igures
(B Baker*)
Light truck 34.8)7.? 14 B9.3J? of the total number
of trucks registered in
198? fa 11 into the "light"
category (USOOC 1985b).
The total number of
trucks registered in 1985
was 38,989.04? (B. Baker*)
Medium truck 1,364,616 3 5 V of the total number
of trucks registered in
198? fall into the
"medium" category (USOOC
1985b). The total number
of trucks registered in
1985 was 38.989,04? (B.
Baker*).
15,815 1965 Registration figures
(B Baker*)
16.05? 1984 Figures (USDOT 1981)
16.889 1985 Figures (B Baker"]
? Disc,
2 Drun
Orun
Drum
13
(Jacko, Ouchartne.
5 Sailers 1973)
(Jacko, Ducharme,
6 Somers 1973)
(Jacko. Ducharme,
I Sows 1973 )
Heavy truck 2,768,2?? 7 IV. of the total number
of trucks registered in
198? fall into the
"heavy" category (USOOC
1985b). The total number
of trucks registered in
1985 was 38.989,04? (B.
Baker*).
44,210 198? Figures Weighted
average of values given
for light-heavy and heavy-
heavy trucks (USDOC 1985b)
Drum
35 (Jacko, Due ha me.
6 Sobers 1973)
-------
tati It- ? 4 (con inuci)
Vebic te c lass
Njuber
Connect
(source)
Ave;aue
distance
travelled'1
fM.'yr}
to"ment
(source)
Brake
configurat >onc
Emss ton
factor^
(veh/a? le J
Comment
(source)
Bus
593,5 lev. than 6 percent. For these classes (cars, medium trucks, buses, and matc-cyc les).
the 1985 registration data were used when possible, Registration data for 1985 were not available for light or heavy trucks, For light tracks, data
from three sources varied by three percent; 1984 05DOT data were used. For the category of heavy trucks, average mileage figures varied from ??,435
p.i/yr (US DOC 1985b) to 61,031 miles/yr (U5D0T 1986), The lo«er value is an average for light-heavy and heavy-heavy trucks weighted by the number of
registrations in each class. In this category, truck classes were wel 1-def ined by the USQ0C, arid fit our 3-category classification, therefore. USD0C
data were used.
c for light, medium, and heavy trucks, and buses, only the type of brake is of concern. It is assumed that, for these vehicle types, there are two
brakes per axle, and the emission factor is applied to the per axle emissions.
d This factor is based on the relative weight of friction material in vehicle brakes and the average number of axles per vehicle within each vehicle
class, The factors were given in Jacko. DaCharme, and Sorters (1973) as shown, except that no units were assigned Ihe units were added for clarity
-------
that 2.6 ug of asbestos is emitted to the air for each kilometer traveled
by a typical car. Information is available in the published reports to
refine this value. For instance, data are available on the differences
in emissions between disc and drum brakes. Applying this differentia-
tion, with the assumption that the vehicle has two disc and two drum
brakes, decreases the estimated emissions to 1.9 ug/km. Additional
analyses performed by the researchers employed 1977 vehicle registration
and use statistics to estimate nationwide emissions. Updated statistics
can be used to produce more realistic values. Results of this type of
manipulation are presented in Table 2-2.
Jacko, DuCharme, and Somers (1973) designed, built, and installed
col lection devices on the disc and drum brakes and clutch assembly of an
average automobile. The test car was then driven through predetermined
cycles that were planned to test both normal and abusive brake wear. The
emissions from each driving cycle were collected and analyzed for
asbestos fibers with a transmission electron microscope. Therefore,
knowledge of the mass of particulates emitted to the air, deposited on
the ground, or entrained in the brake assembly for each driving cycle, as
well as the asbestos content of these particulates, could be measured
directly. These data, combined with information on the duration (in
miles} of each driving cycle allowed calculation of the average airborne
asbestos emissions for each type of driving.
The first entry noted for this study on Table 2-2 shows the range of
airborne emissions and asbestos contents for all seven of the driving
cycles. The airborne asbestos emissions determined from the arithmetic
averages of the values in these ranges are artificially high because
equal weight was given to all driving cycles.
The average values noted in the next entry eliminated this problem by
weighting the different cycles by the severity factors described by the
authors. Severity factors were only available for five of the seven
2-8
-------
driving cycles; therefore, only the results from these five are used in
this analysis, Using weighted averages to estimate the airborne asbestos
emissions lowers the estimated value by more than a factor of two over
the unweighted estimates.
Cha and Carter (1982) based their dynamometer studies of asbestos
emissions from brakes on statistically determined representative braking
cycles (RBCs). They then collected and directly measured the mass of the
airborne particulates and analyzed the asbestos content with a
transmission electron microscope. The authors then weighted the
resulting figures by the frequency of occurrence of a particular RBC as
determined by their statistical evaluation. As shown in Table 2-2, these
weighted averages were used to estimate the airborne asbestos emissions.
The analysis presented in the referenced report seemed to 1ack
consistency, and is, therefore, not noted in this report. One assumption
used by Cha and Carter is, however, incorporated into the emission
estimates given in Table 2-2, i.e., that the ratio of airborne
particulate emissions from disc brakes to that from drum brakes that was
determined by Williams and Muhlbaier {1982) was also valid in the current
study. Although Cha and Carter only tested disc brakes, this ratio
allowed them to estimate values for drum brakes and calculate a per-
vehicle figure consistent with the other reports.
The assumptions used in compiling Table 2-2 are: (1) each vehicle
has four brakes, two disc and two drum, (2) average braking cycles
include 1.2 stops/km (Jacko, DuCharme, and Somers 1973), (3) asbestos is
emitted from brakes on automobiles, trucks, and buses, (4) there were 164
mi 11 ion vehicles in those categories in the U.S. in 1983 (USDOC 1985a),
and (5) each vehicle averaged 10,000 miles of travel in 1983
(USDOC 1985a). It should be stressed that the emission estimates
provided in Table 2-2 result from averaging the miles traveled by al1
vehicles, and by equating the emissions from trucks and buses with those
2-9
-------
from cars. In addition, not all vehicle classes have two disc and two
drum brakes, many have drum brakes only. Therefore, these assumptions
misrepresent total asbestos emissions.
A better estimate of asbestos emissions resulting from vehicular
brakes can be obtained by estimating and summing the individual
contributions from automobiles, light, medium, and heavy trucks, buses,
and motorcycles. Estimates of this type, based on data found in the
three referenced studies, are provided in Table 2-3. The relative
contributions from each vehicle type are determined, as in the method
described for Jacko, OuCharme, and Somers (1973), by applying emission
factors for each vehicle class to the estimated per axle automotive
emissions. A1though this method is based on imperfect emission factors,
it should provide a more realistic estimate of asbestos emissions than
those presented in Table 2-2.
In Table 2-3, whenever possible, the individual totals for cars,
%
buses, and motorcycles are based on 1985 vehicle registrations and
•k
average di stance travel led as shown in Table 2-4 . The total number
of trucks in each weight category was determined by applying a
distribution ratio derived from 1982 data (USDOC 1985b) to the 1985
number of trucks registered in the U.S.* The average distance traveled
each year by trucks in the light and medium weight categories was taken
from census data (USDOC 1985b), while that for the heavy weight category
was obtained by weighting the average distance traveled by light-heavy
and heavy-heavy (DOC classification) trucks, by the number of trucks
reported in those categories (USDOC 1985b).
The estimated total airborne asbestos emissions for al1 vehicle types
range from 5.6 to 10 metric tons per year.
Personal communication between C. Lewis (Versar, Inc.) and B. Baker
(Motor Vehicle Manufacturers' Association (313) 872-4311) January
16, 1987.
2-10
-------
2.2 Asbestos Emissions from 24 Cities
The portion of the national airborne asbestos emissions from
vehicular brakes (shown in Table 2-3) originating in each of 24 United
States cities was estimated with the aid of the National Emissions Data
System (NEDS), average mileage figures, and census data. The method used
for this estimation is described in the following section.
Twenty-four cities were chosen to represent small, medium, and large
cities in all parts of the country. The 24 cities are identified in
Table 2-5. As shown by thi s table, the ci ties studied range from 35
thousand people in E1 mira, New York, to 3 million people in Los Angeles,
California.
Once the cities were identified, it was necessary to determine the
proper portion of the national emissions to be assigned to each city.
For this effort, NEDS was used to provide data on the amount of fuel used
by different vehicle classes within each county of concern in 1983. NEDS
is maintained by EPAs Office of Air Quality Planning and Standards in
Research Triangle Park, North Carolina. Data from NEDS are available as
publications, complete printouts, and magnetic tape files. Additional
data on the use of NEDS is available in "NEDS National Emissions Data
System Information" (USEPA 1980).
The NEDS data were combined with 1983 average mileage data for each
vehicle class to estimate the vehicle miles travel 1ed by each vehicle
class within the counties. In some cases, such as Denver, Colorado, and
Napa, California, the city of concern corresponded to the county for
which NEDS data were available. In other cases, however, the city of
concern was within a 1arger county. In these cases, the vehicle mile
data derived from NEDS for the county was decreased by a factor
determined by population ratio between the city and the county.
2-11
-------
67/1H
Table 2-5. Cities for Which Concentrations Were Estimated
City
County
Population within
city limits*
Large
Baltimore, HO
Chattanooga, IN
Denver, CO
Detroit, HI
Houston, TX
Los Angeles. CA
Miami, FL
Rochester, NY
St. Louis, MO
Wichita, KS
Baltimore
HaniUon
Denver
Wayne
Harris
Los Angeles
Dade
Monroe
St. Louis
Sedgewick
786,775
163,565
492,365
1,203,339
1.595.138
2,966,850
346,865
241,741
453,085
279.272
Medium
Boise, ID
Charleston, WV
Duluth, MN
Eugene, OR
Lincoln, NE
Montgomery, AL
Odessa, IX
Portland, M£
Ada
Kanawha
St. louis
Lane
Lancaster
Montgomery
Ector
Cumber land
102,451
63,968
92.811
105,625
171.932
176,387
90,027
61.572
SffiLLT
Elmira, NY
Napa, CA
Oca la, FL
Pine Bluff, AR
Rochester, MN
Santa fe, NM
Chemung
Napa
Marion
Jefferson
Olmstead
Santa Fe
35.327
50,879
37.170
56,636
57,890
48.353
a1983 figures (US00C 1983).
bThe city sue classifications were based on the sue of the entire
standard metropolitan statistical area as reported in U5D0C (1983).
2-12
-------
Additional problems were encountered in that NEDS data classified
vehicles into four distinct categories according to fuel consumption:
{1) light duty gasoline vehicles, (2) light duty diesel vehicles,
(3) heavy duty gasoline vehicles, and (4) heavy duty diesel vehicles.
These categories did not correspond with those for which mileage
estimates were available or with those chosen for the asbestos emissions
estimates. Therefore, the amount of fuel (both gasoline and diesel) used
by light vehicles was determined, and this total number was apportioned
between cars and trucks based on the vehicle registration ratio. The
resulting four categories, (1) cars, (2) 1ight trucks, (3) heavy gasoline
vehicles, and (4) heavy diesel vehicles, corresponded to those for which
mileage figures were available. For the purpose of determining asbestos
emissions, however, the distinction between gasoline- and diesel-powered
heavy duty vehicles was not necessary because of the smal1 number of
heavy duty gasoline vehicles. Therefore, the vehicle mile data for the
two categories was summed to provide a total value for heavy duty
trucks. The three remaining categories - cars, light duty trucks, and
heavy duty trucks - correspond to those categories for which asbestos
emissions from brakes were highest (see Table 2-3). Therefore, the
vehicle miles traveled by each vehicle class within each area of concern
were determined.
This value was divided by the total vehicle miles travelled
nationally and the resulting ratio was applied to the national asbestos
vehicle emissions values (see Table 2-3) to estimate that portion of the
national emissions that was contributed by each city of concern. The
results of these manipulations are shown in Table 2-6. Each step of the
calculation is detai1ed in Appendix A.
For comparison purposes, an alternate method was also used to
estimate emissions within the city of Baltimore. In this alternate
method, data avai1 able from the Regional PIanning Counci1 were used in
conjunction with data obtained by the Maryland Air Management
2-13
-------
6/7 1 H
table 2-0. Estimated Car, Light I ruck. Heavy Truck,
and Combined Emissions for 24 Cities
City
Cars
(g/sec)
i ight
trucks
(g/sec)
Hea*y
trucks
(g/sec)
CcmOined
(g/sec)
,11 .)(*
Baltimore, Hi)
Chattanooga. TN
Denver, CO
Detroit. MI
Houston. IX
tos Angeles, CA
Miami, FL
Rochester, NY
St. loins, MO
Wichita, KS
3,93*10
6.41x10'
2.81*10"
3.36x10"
7.49x10'
2,70*10"
1.29x10
a.25*10'
1.23*10"
1.12x10"
1.tiLxlO
3.02*10'
1.32x10"
1. StixlO"
3.52*10"
1.27*10'
6.04x10"
j.B6*lQ"
S.MOxlO'
5,24x10'
7.14*10
1 .94*10*
4.75x10"
S.9HxlO*
1.63x10*
3 93x10
1.64x10"
b.27*10"
3.84*10"
2.55*10"
-4
Baltimore, MD
1,50x10
3.77x10
5.bBx10
2.43x10
Men i urn
Bolst-, ID
Charleston, WV
Djlutn, Mh
tuyent. OS
Lincoln, hE
Mont gone'y. At
Odessa, TX
Portland, ME
4.1t)*10
2.90*10"
7.04x10
6.29x10
-5
5.61x10"'
5,7CxlC"5
3 99*10"'
4,76*10"5
1.96x10
1.3CxlG
3.31*10'
2 SbxlQ
2.73x10'
2.6a*10'
i.a&xio'
2.24x10'
8.95x10
1.24x10"
9.85x10
1,33x10"
1.16*10
1.53*10"
1,48*10
9.14x10"
-5
-4
-4
Siih 11
Elmira, NY
Ntipd. CA
OCdlii, Fl
P me Bluff, AR
Rochester, MK
Stir.ta fe, NM
80x10"
81x10"
22x10
53x10"
23x10"
42x10"
-5
44*10
32x10"
45x10
24x10"
-5
52XS0"-*
i.eixio"5
83x10
32x10"
52x10"
&9xio"
14x10*
64*10
-5
For this entry, emissions were determined by an alternate method,
explanation of this method.
See text for an
2-14
-------
Administration from EPA's MOB IL3 model (USEPA 1984) to determine vehicle
miles travelled within set area source grids. These grids were chosen
according to population density and the shape of the city of Baltimore,
and they correlate reasonably well with the outline of the city. These
grids are illustrated in Figure 2-1, Emissions were then determined for
the area source grids by applying emission factors based on the per-
vehicle kilometer emissions values given in Table 2-3. This method of
evaluating eniissions takes driving patterns and speed into consideration.
2.3 Ambient Asbestos Concentrations Resulting from Vehicle Brakes
The ambient concentrations of asbestos in the air of each city were
modeled based on specific meteorological conditions of each ci ty.
Because the ambient concentration is linear with respect to emissions,
modeling was performed for each city based on an emission rate of
1 g/sec. The actual concentration was then determined by sealing this
figure according to the emissions estimated for each city. This method
allowed for flexibi1ity in the emissions figures without requiring
additional computer modeling. For instance, emi ssions from Baltimore
were estimated by an alternate method as described in the previous
section, and combined with the model results to derive a comparable
concentration estimate.
The computer model used in this task was the Graphical Exposure
Modeling System (GEMS) BOXMOD model. This model is accessed through the
GEMS Atmospheric Modeling System (GAMS) (GSC 1986) maintained by EPA's
Office of Toxic Substances. The BOXMOD model characterizes each city as
a square surface from which pollutants are emitted into a box extending
above the surface. The size of the square was estimated by assuming that
the surface area covered by the city was configured in a perfect square.
Ambient air pollutant concentrations in the box are a function of the
amount of pol 1 utants and the dispersive conditions within the box. The
BOXMOD model assumes no transport of pollutants into the box and a
uniform pollutant emission rate from the surface.
2-15
-------
Solid line represents the actual outline of Baltimore City.
Broken line represents the area modeled by the alternate raethod.
Figure 2-1. Modeled area of Baltimore City.
2-16
-------
For the purpose of this study, the model was run in a Gaussian mode
whereby pollutant concentrations within the box are assumed to be
normally distributed with highest concentrations occurring at the surface
and decreasing rapidly with height. Summary meteorological data from the
weather station nearest the particular city under study were used to
simulate local dispersion conditions. Loss of pol1utant-mass within the
urban area because of dry deposition was negligible because of the small
diameter of the asbestos fibers (virtually all fibers are less than 5 um,
with 83 percent less than 0.375 um (Rohl et al. 1976)). Precipitation
scavenging was close to zero with no loss attributable to chemical
removal processes.
Ambient concentrations estimated by these methods are presented in
Table 2-7. Ambient concentrations from brakes on all vehicle types range
3 3
from 0.005 ng/m in Lincoln, Nebraska, to a high of 0.258 ng/m in
Los Angeles, California,
The air concentration calculated for Baltimore by the alternate
method is only 19 percent of the value estimated for Baltimore by the
method used for the rest of the cities. This variance is caused by the
difference in the estimated emissions. Because this alternate method is
based on data generated by the State of Maryland specifically for
Baltimore, it cannot be readily applied to other cities. If, however,
analogous data were available and the alternate method applied, it would
not be realistic to expect the resulting concentration estimates to vary
in any consistent way with the concentrations estimated in this report.
The variance would be expected to be random because there is no reason to
expect that NEDS data would vary in any systematic way when compared to
hypothetical state-generated data.
Estimated asbestos concentrations resulting from vehicular brakes for
each of the 24 cities are illustrated in Figures 2-2, 2-3, and 2-4. Each
figure depicts the ambient asbestos concentration for the cities within
2-17
-------
6771H
Table 2-7. Estimated Annua! Asbestos Concentrations
from Vehicle Brakes in 24 American Cities
L tgnt heavy Combined
Cars trucks trucks vehicle brakes3
f ity (ng/m^j (ng/m^) (ng/tn-*) (ng/m^)
Bolt tmore, MO 0.00*5
Chattanooga, IN 0,00?
Denver, CD 0.031
Detroit. HI 0.024
Houston, IX 0.017
Los Arables, CA 0 DM)
Muntil, r L 0.044
Rochester, NY 0.021
it. Louis, M0 0,020
Wichita, hi G 011
Baltimore, MDb 0.024
>*¦». J i urn
Bai if, 10 0.009
Crwrleston, WV 0.016
Dii iuir,, Mfi 0.010
tvjijene, OS 0,022
l inco In. Nt 0.001
Montgamsr,, At 0,006
Ouessa. IX 0.014
Port lana, ML 0.019
Sma 11
t tmira, NY 0.0J3
Napa. CA 0,018
Oca la, Fl 0.0Z5
Pine bluff, Afi 0.014
Rochester, MN 0,012
Sontj Fe, NM 0,012
0.03C 0.U6 0.203
0.003 0 022 0,033
0,014 0.0S2 0.098
0.011 0 043 0.079
0 COe O.Qib 0,061
o 04; oa:a o.25b
0.021 0 05o 0.121
0.010 0.021 0.05,1
0,009 0.062 0.092
0.00L 0.025 0,041
0,006 0.003 0.039
0.QC4 0,019 0,031
0 00b 0.07? 0.104
0,005 0.014 0.029
0.010 0.047 0 07o
0,001 0.003 0.005
0.003 0.017 0.026
0.007 0.0S2 0.072
0.00S 0.036 0,064
0 CIS 0.034 0.0B2
0 008 0 015 0 041
0,012 0.026 0.0&3
0.007 0.02S 0,046
0,006 0.016 0.034
0.005 0,012 0.029
°Tfnt. column represents trie sura of the est imated concentrations from each
category (cars, light trucks, and heavy trucks). The numbers in this column
may not exactly total because of rounding.
DFor this entry, emissions were determined by an alternate method. See text
for an explanation of this method.
2-18
-------
3
Asbestos concentration (ng/m )
0.3125 -
0.258
City, In order of decreasing population
Figure 2-2. Estimated asbestos concentration in the air .of 10 large cities
resulting from vehicular brakes.
-------
Asbestos concentration (ng/tn )
0.1094 -
AL NE OR ID MN TX WV ME
City, in order of decreasing population
Figure 2-3. Estimated asbestos concentrations in the air of 8 medium-si zed
cities resulting from vehicular brakes.
-------
Asbestos concentration (ng/m )
0.0875 -
0.082
ro
i
ro
0.075
0.0625
0.05
0.0375
0.025
0.0125
0.034
Rochester,
MN
0.046
0.041
0.029
Pine Bluff,
AR
Napa,
CA
Santa Fe,
NM
0.063
City, in order of decreasing population
Figure 2-4. Estimated asbestos concentration in the air of 6 small cities
resultino from vehicular brakes.
-------
each size classification, in order of decreasing population. A general
trend can be seen loosely correlating city size with asbestos
concentration from vehicular brakes between large and small cities.
However, a definite conclusion concerning the effect of population as a
measure of ambient asbestos concentrations within a size category is not
supported.
Figure 2-5 illustrates the relative contributions of each vehicle
class to the total ambient concentration from vehicle brakes. The
quantities shown in-this figure were determined by averaging the asbestos
concentrations from each vehicle class over all the cities studied.
Figure 2-5 shows that, with more severe braking and higher asbestos
emissions for each vehicle, the total contribution from heavy trucks is
much higher than that from cars or light trucks. This distribution
exists in spite of the fewer vehicle miles travelled in the cities by
heavy trucks.
2.4 Average National Ambient Asbestos Concentration from Vehicle
Brakes
This section details two methods for estimating national ambient
asbestos concentrations from vehicle brakes. Both of these methods use
dispersion modeling described in Section 2.3 for 24 U.S. cities. The
first method provides bracketed worst- and best-case estimates of 0.087
3 3
ng/m and 0.039 ng/m , respectively. The second method provides a
3
best estimate of 0,057 ng/m .
2.4.1 Method 1: Population
Quick estimates of worst- and best-case national asbestos
concentrations can be derived based on population. For both the worst
and best cases, the cities for which modeling was conducted were grouped
into six categories,.according to their 1980 populations (US00C 1983) and
average ambient asbestos concentrations in each population category were
obtained (Appendix A). For the worst-case estimate, it was assumed that
rural areas with populations below 25,000 have the same ambient asbestos
2-22
-------
Figure 2-5. Average relative contribution to the total average asbestos concentration.
-------
concentrations as do cities and towns with populations above 25,000.
Conversely, for the best-case estimates, it was assumed that the rural
areas with populations below 25,000 had no ambient asbestos from vehicle
braking; therefore, the entire national average concentration was
contributed by areas with more than 25,000 people.
For the worst-case estimate, the average concentrations were then
weighted by the number of people in each category (in 1980) (USDOC
1985a). The weighting factor was derived by dividing the number of
people in each category by the total number of people in the categories
considered. This method, detailed in Table 2-8, results in a worst-case
3
estimate of 0.0872 ng/m , by assuming that towns and rural areas of
less than 25,000 population have the same average asbestos concentrations
as those with greater than 25,000 population. In other words, the
weighted average concentration estimated for the cities or towns with
more than 25,000 people is assumed to hold for the rest of the country.
In reality, the rural areas have relatively less asbestos released to
the atmosphere from vehicle brakes than do urban areas; therefore, a
corresponding best-case value can be derived by assuming that the rural
areas have no asbestos released from brakes. Because 55.5 percent of the
population lives in areas with fewer than 25,000 people (USDOC 1985a),
the worst-case asbestos level would be lowered by 44.5 percent, to 0.039
3
ng/m as a best-case value.
The major advantage to this method is that it provides a bracketed
estimate of national concentration. However, this method has three
distinct disadvantages. First it relies heavily on the choice and number
of cities within each population category. Because the cities were not
chosen specifically for this type of categorization, they do not
represent a statistically accurate sampling. In fact, for the 500,000 to
1,000,000 category, only one city is used to represent the entire
category. This fact could shift the weighted averages significantly.
Second, this method relies on the correlation between city population and
2-24
-------
1
677M
Kit) If i'-d. Worksheet for Est inviting National AmOient Asbestos Levels by a Popu Ut ion Corrtr Ut i on
Area sift
fpOpU Id t toil)
lotj S U.S.
population in
each category*
!in thousands 5
Area
(city)
Concent ran on
(ng/n3)
Percent
of national
populat ion
in c it >es
'25. 000 within
each group
we lgntt-n
average
concentration
(ng/m3)
1 1 , t'Ji. , U Jt'
17,530
503,0QQ-i,C00,000
2SD,000-b00,GuO
10.B14
12,1 SO
Detroit
Houston
Los Annei&s
Average
Ba Itimore
Denver
M lami
St, Louis
Wichi tii
Average
0-0?9
0.061
o.?'»a
0.133
0 109
0.0&B
0.121
0.092
2-21L
0,0a<3
17.4
0.023
10. a
12. 1
0.023
0.011
i 00, DOG-..-30,000
17.015
Chattanooga
Rochester,
Boise
Eugene
L mcoln
Montgomery
Average
0.033
0.051
0.031
0.0B0
0.005
M?5
0. 0.(6
16.9
0.006
50,000-100,GOO
1&.7BG
Charleston
Qulutft
Ooe^id
Port Und
Pme Bluff
Rochester, MN
Natlrt
Average
16.6
O.Oil
25,000-^0,000
23.436
£ Imira
Oca la
Santa Fe
Average
23.2
0.013
lota 1
IM. 759
100
0.007
"USDOC (lS*i,j).
D¥ersar (I9e7).
cThis is the subpopulatson total for areas with more than 25,000 people; the population in areas with less than
irian 25,000 people is 125,787,000. These figures are based on a total population of 226,546,000 persons in 19bQ
(USI)Oi 1985a)
2-25
-------
ambient asbestos concentration from vehicle brakes. Although the data
presented in this report show a general trend of decreasing concentration
with city size, a definite correlation is not supported. Third, the
national concentration is scaled up from city concentrations that were
derived by scaling down national emissions. This indirect route in
itself introduces uncertainty into the resulting estimate.
2.4.2 Method 2: National Emissions
The alternate method is based on ambient air modeling of 24 cities,
in conjunction with the overall national asbestos emissions given in
Table 2-3. This method has the advantage of using the emission rate as
it was initially estimated (on a national level) rather than scaling a
national estimate down to the city level (as in Section 2.2) and then
sealing it back up to the national level.
Because the modeling was initially conducted for a unit emission rate
of 1 g/sec, the final concentration figure for each city can be adjusted
for estimated emission rates. The concentration based on unity emissions
3
averaged over all the 24 cities modeled is 0.325 ug/m . Because the
concentrations were modeled based on an area source of a given size, the
average concentration over al1 24 cities has to be scaled up to a
national level. This seal ing can be accomplished by multiplying the
average concentration by the ratio of land area covered by the 24 cities
to that within the contiguous U.S. as follows:
(0.325 ug/m3) x (5,779 km2 in the 24 citiesa)/(7,813,112 km2 in the U.S.b)
x (103 ng/ug) « 0.240 ng/m3.
* See Table A-l.
b Source: USDOC (1985a).
2-26
-------
This figure, when multiplied by the average national asbestos
emission rate for all vehicle classes of 0.237 g/sec yields an
estimated national ambient asbestos concentration from vehicle brakes of
3
0.057 ng/m . As would be expected, this value falls within the range
predicted by the previous method.
Although this value is based on modeling conducted for cities with
populations over 25,000, it does not exclude the rural area of the
country from consideration because national emissions estimates are
used. The 24-city 1 imitation only affects the modeling results, which
are based on land area and meteorological conditions. The land area
restriction has been accounted for by our sealing factors. Although the
cities chosen for modeling were not selected to provide an accurate cross-
section of meteorological conditions, a wide variety of conditions are
accounted for. This method could be refined to account for differences
in meteorology types, if data were available that showed the national
land area covered by a given meteorology type, and classified the cities
that were modeled by meteorology type; however, these data are not
available.
Derived from averaging the total national emissions from all vehicle
classes over al1 three studies, as shown in Table 2-3. This figure
does not equal the combined national emissions shown in Appendix A
because the Appendix A emissions are for three vehicle classes
only: cars, light trucks, and heavy trucks. The total used here
also includes buses, medium trucks, and motorcycles.
2-27
-------
3. CONSUMER EXPOSURE TO ASBESTOS IN COATINGS, SEALANTS, AND PAINTS
The number of asbestos-containing coatings, sealants, and paints that
are available to consumers has decreased dramatically since the mid-
1970s . For example, spackle, dry wall joint compounds, caul king
compounds, and paints are no 1onger manufactured with asbestos with the
intent of selling directly to consumers (Anderson et al. 1982, Versar
1984). However, consumers can sti11 legally purchase some of these
products. The list of coatings, sealants, and paint products that still
may contain asbestos is presented in Table 3-1. Not all of the products
in these categories, however, are likely to contain asbestos.
Table 3-2 summarizes consumer exposure to asbestos in the six
coatings and seal ants product categories that still may contain
asbestos. A brief discussion on the potential exposure to asbestos from
each of these products is given below.
3.1 Roof Coatings
These are petroleum-based products that are composed of 5 to 30
percent asbestos. The asbestos is added to improve the strength, thermal
resistance, and waterproofing properties of the coating (Anderson et al .
1982). According to product label s, roof coatings can be used on metal,
composition, slate, and asphalt roofs. They are recommended for the
general repair of leaks (in the roof itself and around chimneys,
skylights, and gutters), although roof coatings may be applied as a
protective seal over an entire roof, depending on the composition of the
roof.
Essentially all exposure to asbestos in roof coatings is expected to
occur when it is appl ied. Roof coatings are generally sprayed on,
although they may be applied with a brush. Estimates of the consumer
population that does roofing work vary widely, from 841,000 to 11,193,000
people (see Table 3-2). However, since not all consumers doing roof
repair will be using roof coatings (e.g., shingle replacement) and not
3-1
-------
6792H
Table 3-1. Coatings. Sealants, and Paints that May Contain
Asbestos and be Used by Consumers
Product Source
Roof coating 1,2
Waterproof coating 1
Corrosion resistant coating 1
Driveway sealant Z
£poxy adhesive 2
Glazing compound 2
'Anderson et al. (1982).
^Izaretes 1198?).
3-2
-------
6792H
Table 3-2, Consumer Exposure to Asbestos in Coatings, Sealants, and Faints
Subpopulation
Asbestos
concentration
(f/cm3)
Breathing
rate
Fiber type (ra^/hr)
Number of
Frequency persons exposed
Hr/yr (years) (thousands }d
Roof coatings
Waterproof coatings
Corrosion resistant
coatings
Driveway sealant
Epoxy adhesive
Glazing compound
0.0 to 0 6*
NA
NA
NA
NA
NA
Chrysotile
Chrysotile
Chrysot ile
NA
NA
NA
1.3
1.3
1.3
1.3
1.3
1.3
NA
NA
NA
0.011
NA
f
3-5
NA
NA
NA
1/12*
NA
841 - 11,193
NA
NA
NA
27,770s
NA
NA - not avallable.
aSpray application (Anoerson et al. 1962).
bVersar estimate.
cHcme Pro Systems. Inc. (1905). This corresponds to the time suggested to refinisn metal roofs. It
is also assumed to be the time interval when other types of roofs will need same type of repair,
^Unless otherwise stated, these data were reported as the number of households that participate in
do-it-yourselfer activities (e.g., roofing repair). It was assumed that only one person would be
directly exposed from each household.
e01Y Consumer Purchase Profile (Farnsworth and McKittrick 1984) reported 11.193,000 households do
roof repair. The 1983 Consumer Attitude Survey, conducted by Leo J, Shapiro & Assoc. for National
Home Center News (Farnsworth and HcKittrick, 1984), had the lowest estimate; it reported 0.9 percent
of all U.S. households do roof repair. The total ntntber of U.S. households was 93,519,000 in 1983
(Bureau of Census 1985), which implies that 841,000 households do roof repair. The 1983 Hardware Age
Consumer DIY Study, conducted by Chilton Research Services (Farnsworth and McK-ittrick 1984). has a
mid-range estimate. They estimated that 2 percent of the U.S. population is involved with roof
repair. The total population in 1983 was 234.000,000 (USD0C 1985), which implies that 4,680,000
people did roof repairs.
f«,
Versar (1986b).
9Versar (1985).
3-3
-------
al1 roof coatings are likely to contain asbestos, the lower end of this
range is more reasonable.
Once the coating has dried, essentially no releases are expected
during normal use. This is because the asbestos should be securely bound
to the petroleum-based product.
3.2 Waterproof/Corrosion Resi stant Coatings
Both waterproof and corrosion resistant coatings are petroleum-based
products. They are discussed together because of similar application
practices and exposure potential.
According to product labels, both types of coatings are usually
applied with a brush to a wel1-defined smal1 area. Corrosion resistant
coating will be applied outdoors, while waterproof coatings may be
applied outdoors or indoors (e.g., basements, garages). Since these
coatings are not 1ikely to be severely agitated during application (like
the spraying of roof coatings}, the asbestos fibers in these products are
expected to remain bound in the tacky petroleum base. After the
petroleum base solidifies, essentially no asbestos releases are expected
to occur during normal use.
3.3 Driveway Sealants/Epoxv Adhesives/Glazino Compounds
Tzanetos (1987) indicates that driveway sealants, epoxy adhesives,
and glazing compounds contain asbestos, although this could not be
confirmed from an independent shelf survey. Assuming that they do
contain asbestos, however, exposure is expected to be limited by the
wetted and viscous consistency of these products.
Driveway sealant is a petroleum-based product that is used to cover
asphalt driveways to give the driveway a smooth and consistent appearance
and to prevent moisture from seeping into the asphalt, which may cause
damage as it freezes and thaws. Sealers also help prevent damage from
oxidation, ultraviolet sunlight, and gas and oil spills. The seal ant is
applied from a standing position with a 1ong-handled 36-inch squeegee.
3-4
-------
Epoxy adhesives are usually two-part systems comprising a resin and a
hardener which must be mixed by the consumer in equal parts. Epoxies are
suitable for bonding almost any two materials together including wood,
metals, ceramics, tiles, rigid piastics, pottery, and drywal1. Epoxy
adhesives will generally be used indoors for smal1 repairs.
Glazing compound is a putty-1ike substance used to seal glass into
wood windows. It is usually applied with a putty knife on the outdoor
surface of the window. After the glazing compound has cured, it is
painted for appearance and to prevent water from seeping under the
glazing compound and contacting the wood frame.
No exposure data could be found to quantify asbestos releases from
these products during application.
3-5
-------
4. CONSUMER EXPOSURE TO ASBESTOS IN VINYL ASBESTOS FLOOR TILE
In 1976, the sales of floor materials represented about 15 percent of
the total U.S. market for asbestos (Sebastian 1982). It is assumed that
vinyl asbestos floor tile contains approximately 10 percent asbestos
(Jenkins 1985), although it may contain as much as 20 percent (Dolan
1985). Consumer exposure to asbestos in the floor tiles can result from
two mechanisms: (1) the removal of the floor tile by the consumer and
(2) normal wearing of the tile that frees asbestos fibers. Inhalation
exposure assessments for these two mechanisms follow and the results are
presented in Table 4-1.
4.1 Removal of the Floor Tile by the Consumer
The physical requirements of sanding, scraping, or chipping asbestos
floor tiles can result in significant concentrations of fibers in the
3
air. Concentrations as high as 25,109 ng/m have been reported for a
single sample collected over 2 1/2 hours of post-removal air monitoring
(Jenkins 1985). It is expected that the consumer would only remove floor
tile containing asbestos once in his lifetime and would take four to
eight hours to complete the task (Anderson et al. 1982). Because of the
1 ight activity involved in removing the tiles, an average adult
3
inhalation rate of 1.3 i /hr is assumed (Versar 1986a).
4.2 Floor Tile Wear
Normal wear of vinyl floor tiles containing asbestos can result in
3
asbestos air concentrations as high as 50 ng/m (Dufour 1984). For the
3
two following inhalation exposure estimates, concentrations of 25 ng/m
3
and 50 ng/m were used for best estimate and worst case scenarios,
respectively. It is assumed that the duration of each exposure event
averages 8 hours/day over 350 days/year. For an average adult, an
3
inhalation rate of 1.1 m /hr was used to reflect the breathing rate of
a typical office worker. Based on these conditions, the inhalation
exposure to asbestos was estimated.
4-1
-------
4-1 I nhj )dt ton Exposure to Aiiiustos from Viny 1 Tile
bcerur 10
Best est im.it e
Worst case
Source
Jrrnvn : Li I ! Me i : It* t>> t hi^
JOn.ijiwi'
d
«
Aiiie-itos concent rat ion
26,1 Ch* rig/m0
Zfc, 109 ruj/nf*
Jerk ' ns
. (IStaS)
Durjt iOr- ot exposure
4 tWurs
8 flours
Anderson et a 1.
riuinner o* exposures
i exposure/lifetime
! enposure ' 1 ifet 'me
Versar
est irrate
j nrij l.u iCr, rote
1.3 in1', ftr
I j ns'.'hr
Versar
i1aeGd)
tie'.to-. ng.m'
•>0 rsg? m *
DufOur
f'.904)
Dur.it icn o* exposure
« hours/Cldy
W hours/day
Versar
estimatec
liuintie; o! exposure;.
?5Q tidy/year.
50 day/year.
Versar
estimatec
40 years.' 1 ifet lme
JO ypars/1 ;fet irne
Versar
est iirwie0
; nr:n l.it ion rate
1 1 rrr'/nr
1 J nr\>rir
Versar
(19d6a1
(3 962)
J> 1 (¦' type i'. rnr,',ot i le .inri '<9 f>G percent of trie f tuer-, wore greater tnan S mn.
''lie mean finer length was 0 63 urn witn a standard deviation of 0.20; a" asbestos was chrysolite
W,,KV ion U- liat.ee! on a typical oft ice worker.
4-2
-------
5. CONSUMER EXPOSURE DURING BRAKE REPAIR
This section discusses consumer exposure to asbestos that occurs
during do-it-yourself (01Y) brake repair. It is organized into three
sections; (1) exposed population, (2) consumer vs. occupational
exposure, and (3) exposure calculation.
5.1 Exposed Population
Although exposure to consumers during do-it-yourself brake repair has
not been carefully analyzed, a surprisingly 1arge population instal1s its
own brakes. According to SMRB (1982), 13,186,000 consumers are estimated
to instal1 brakes in cars or small trucks (9,132,000 people who actually
purchased the brakes pi us 4,054,000 people who are members of the
purchaser's household). This survey was based on a time frame of one
year (1981), and thus it only presents a partial population of consumers
who instal1 their own brakes. In reality, brakes in cars and small
trucks are estimated to require repair (replacement) approximately once
every three years (this assumes that brakes are typically changed when
vehicles travel between 25,000 and 30,000 miles ; USD0C (1985)
estimates that the average car travels between 9,830 and 9,976 miles per
year). Therefore, the population cited in this survey would not be
expected to purchase or instal1 brakes in the two subsequent years (1982
and 1983); however, it can be assumed that two more populations, each of
which is equal in size to the 1981 estimate, would instal1 brakes in both
1982 and 1983. In 1984, the same consumers identified in the survey
would be expected to replace the brakes that they installed in 1981.
Consequently, the total number of consumers who instal1 brakes is
estimated to be 39,558,000 (13,186,000 different consumers each year, for
three years); their frequency of exposure is once every three years.
Personal communication between CI ay Carpenter, Versar Inc., and Dan
Turner of the American Automobile Association on January 28, 1987.
5-1
-------
These data have been confirmed by an independent source. The 1983
home and auto DIY study, which is a survey conducted by Home and Auto
Magazi ne. reported that 64.9 percent of the survey population performed
drum brake overhauls and that 61,2 percent of the survey population
performed disc brake overhaul s (farnsworth and McKittrick 1984). This
survey was designed to estimate how many people routinely did a certain
event (e.g., instal1 their own brakes), rather than estimating how many
people did something in a given time slice (e.g., in the last year). The
survey population comprised customers of automotive and home repair
stores and auto department and discount store chains. The automobile do-
it-yourself population is estimated to be between 60 and 65 million
*
people. Assuming that approximately 63 percent of this survey
population does their own brake repair, this corresponds to between
37,800,000 and 40,950,000 people. Again, it is assumed that this
population has a frequency of exposure of once every three years.
5.2 Consumer vs. Occupational Exposure
Consumer exposure during brake repair would differ from occupational
brake repair because of three important factors: (1) consumers do not
have a standard pi ace to perform auto repairs; in some cases, the repairs
would occur in an enclosed garage, but in most cases they would occur
outdoors, (2) consumers would not have available the professional
equipment found in repair shops, (3) consumers have 1imited experience,
and thus the time required to make repairs would be longer, and
(4) consumers are unlikely to perform more than one brake job per year.
More information on these factors and recommendations for developing
exposure estimates follow:
Personal communication between Jim Konz of Versar Inc. and Richard
Weinberg, editor of Home and Auto Magazine, on June 27, 1985.
-------
1. Consumers would perform brake repair either in a garage or
outdoors. According to 1983 data from the Bureau of the Census,
38.821,000 of the 54,724.000 owner-occupied housing units in the
U.S. have garages or carports. This distinction between
garages and carports is important because carports are not
enclosed and thus would not contain the released asbestos fibers
in the consumer's work area. No data are available that
differentiate between the number of garages and carports. An
estimate can be made, however, based on the year the structure
was built. The data and assumptions used to estimate the number
of owner-occupied housing units with garages are presented in
Table 5-1. It was estimated that approximately 54 percent of the
owner-occupied housing has a garage avai1 able. No data could be
found on the number of renter-occupied housing units that have
garages or carports. However, data on garages can be estimated
based on the total number of renter-occupied structures in the
U.S. (29,914.000 units). The estimated number of rental units
with garages is derived in Table 5-2. It was estimated that only
20 percent of renter-occupied housing will have garages, compared
to 54 percent of owner-occupied housing. Of the 91,675,000
housing units in the country, 35,534.000 or 39 percent are
estimated to have garages available for brake repair.
2. A typical consumer would not have access to the professional
equipment used in a repair shop. This professional equipment can
lower exposures (e.g., high-efficiency particulate air vacuum
system) or increase brake repair exposures (e.g., compressed
air). It is expected that consumers would only have the
mechanical tools necessary to remove the wheel casing, brake
components, etc. and equipment such as a brush or a household
vacuum cleaner to remove loose brake dust.
The removal of the worn brake pad and loose dust will result in
the highest levels of exposure. fiber air sample results for
brake assembly clear-off methods have been reported in Roberts
and Zumwalde (1982). For dry brush cleaning, the asbestos
concentration varies from 0.61 f/cc (21-minute sampling time) to
0.81 f/cc (12-minute sampling time). It is not expected that the
consumer would wet the brush before cleaning. Another source
(Rodelsperger et al. 1986) reported fiber concentrations for dry
brushing of 0.09 f/cc. This was based on personal sampling for
30 minutes or more and it only included fibers with lengths
Personal communication between Clay Carpenter, Versar Inc., and Dick
Cranston, Bureau of the Census, on January 13, 1987.
5-3
-------
6730H
Table 5-1. Estimated Number of Garages in Owner-Occupied Housing
Estimated
number of
Estimated
units with
number of
Nunber of
garages or
units with
total units8
carports'*
garages6
Date structure buiIt
(mi 11 ions)
(mi 11 ions)
(millions)
Apri1 1970 or later
14,534
10,300
5,200
1955 to March 1970
5.929
4,200
2,100
1960 to 1964
5,493
3.900
1,900
1950 to 1959
10.243
7,300
7,300
1940 to 1949
4,918
3,500
3,500
1939 or earlier
13,607
9.700
9.7.00
TOTAL
54,724
38.900s
29,700
aBureau of Census (1985).
bThis assures an even distribution, over time, of the nunber of units
with garages or carports.
cJt is assumed that structures buiIt before 1960, which are identified
as having a garage or carport, actually have garages. Structures built
in 1960 or later were assured to have 50 percent carports and 50 percent
garages.
dThe actual nmber of housing units with garages or carports in 1983
was 38,821,000; the number in the table is higher because of rounding.
5-4
-------
6730H
Table 5-2. Estirated Number of Garages Associated
with Renter-Occupied Housing
Number of
total units®
(mil lions J
Est mated
percent
with
garages
Estimated
nimber
with
garages
Units in structure
(2)
(mi 11 ions)
1, detached
1, attached
Z to 4
5 to 9
10 to 19
20 to 49
50 or more
Mobile hone or trailer
8,127
1.441
7,797
3,636
3,165
2,173
2,811
763
4,400
400
1,000
200
0
0
0
0
TOTAL
29,913
6,000
'Bureau of Census (1905).
^Assumed the same percent as owner-occupied, since the vast majority of
owner-occupied units are detached.
cAssumed one-half of the estimated percent of garages for i, detached.
dAssumed one-half of the estimated percent of garages for 1, attached.
eAssumed one-half of the estimated percent of garages for 2 to 4 units
in structure.
'Assumed that essentially no garages would be associated with
structures with a large nutter of units or with mobile hosnes or trailers.
5-5
-------
greater than 5 urn. Finally, a study conducted in Finland found
concentrations ranging from 0.3 to 0.6 f/cc for the opening and
brushing of drum brakes. This was based on four samples from two
workplaces with sampling times ranging from 2 to 13 minutes
(Kauppinen and Korhonen 1987).
For vacuum cleaning, Roberts and Zumwalde (1982) report
concentrations varying from 0.00 f/cc to 0.07 f/cc (sample time -
2 minutes). Another study, using wet and dry shop vacuums, found
asbestos concentrations ranging from 0.0 to 2.1 f/cc (sample time
- 1 minute), with most concentrations 0.2 f/cc or less (Lemmons
n.d.). These data, however, are not directly applicable to
consumer brake repair because the typical filter in a house vacuum
cleaner cannot collect the fine asbestos fibers found in brake
dust {USEPA 1986). Therefore, it is 1ikely that concentrations
resulting from consumer vacuum cleaning will approach, if not
surpass, those found with dry brushing. It is assumed that the
peak concentrations found during brake repair from assembly clean
off are 0.61 to 0.81 f/cc, with an average of 0.71 f/cc.
3. According to information reported in Bragg (1986), a ful1-time
professional brake mechanic can perform a replacement and repair
job in about 1.6 hours. It is not realistic to expect that a
consumer, who does brake repair approximately once every three
years, would be able to do this task as quickly. It is assumed
that a consumer doing brake repairs may take up to twice as long
as a professional brake mechanic or 3.T hours.
5. 3 Estimated Exposure
The only exposure pathway expected during brake repair is
inhalation. Exposure is expected to vary depending on the different
phases of the brake work that is being done by the consumer, and the
exposure estimates were organized accordingly. Two key phases are
expected: (1) the high, short duration, exposure levels that occur
during cleaning of the residual brake dust and (2) 1ower levels of a
1onger duration that occur when asbestos emissions mix with the
surrounding air. Both of these exposure components are discussed below.
(1) Dust Cleaning Exposure - Loose asbestos dust accumulates in the
brake housing and the consumer will probably remove it by dry brushing.
Exposure estimates for this operation are presented in Table 5-3. These
5-6
-------
5730H
Table 5-3. Estimated Consumer Exposure from Brake Dust Brushing
Parameter
Data
Source
Concentrat ion
0,71 f/cc
Roberts and Zumalde (1982)
F iber type
Chrysot ile
Versar (1984)
fiber size
>5 in
Roberts and Zwrwalde (1982)
Breathing rate
1.3 m3/hr
Versar (1986a)
Hours/day
0.3s
Versar estimate
Days per three years
-1
Versar estimate
Number of persons exposed
39.SS8.000
SMRB (1982)
Exposures/1ifet irae
10b
Versar estimate
aThis corresponds to approximately 17 minutes, which is the average sampling
time for dry brushing reported in Roberts and Zutmalde (1982).
blt is assumed exposure will occur once every three years over 30 years
(ages 20 to 50). therefore, the actual lifetime exposure frequency is 10.
5-7
-------
estimates correspond to only 9 percent (0.3 hours) of the entire exposure
period. This timeframe was selected because it was the average sampling
period reported in Roberts and Zumwalde (1982) for dry brushing.
(2) Low Level Exposure - A consumer doing brake repair will be
exposed to 1ow {secondary) 1evels of asbestos after the releases from
dust cleaning mix with the surrounding air. The exposure level will be
highly dependent on where the brake repai r takes pi ace. If the repair
occurs in a garage, the asbestos concentrations could approach those
found in a commercial brake repair shop; however, if the repair occurs
outdoors, secondary levels of asbestos should approach ambient levels.
It was assumed that people who had a garage available would use it for
repairing brakes; al1 other consumers are assumed to perform this work
outdoors. The population that repair brakes was assumed to be evenly
distributed across the entire U.S. population with respect to garage
ownership. Secondary exposure levels for brake repair in garages are
expected to occur over a 2.4 hour period. This is assuming that the
first one-half hour results in no asbestos releases (i.e., during the
removal of the tire, mounting bolts, etc.) and that the high levels from
brushing occur in the next 17 minutes (0.3 hours). Based on a total work
time of 3,2 hours, this leaves 2.4 hours for secondary exposure to
3
occur. Secondary exposure levels were estimated to be 0.035 f/cm (the
derivation of the estimated secondary exposure 1evels is presented in
Appendix B). The estimated exposure to secondary levels of asbestos
likely to occur outdoors during brake repair was estimated to be
negligible compared to the time-weighted average (TWA) for dust
cleaning. Therefore, only dust cleaning is assumed to contribute to
asbestos exposure during outdoor brake repair. A summary of the
estimated consumer exposures from brake repair is presented in Table 5-4.
5-8
-------
6730H
Table 5-4, Estimated Consumer Exposure During Brake Repair
locat ton
Concentration
(f/cc)
Hours/
day
Breathing
rate
(ir3/hr)
Days
per
three
years
Ninter of Exposures/
persons exposed lifetime^
Brake repair »n a garage
Exposure from brake dusting 0.71 0,3 1.3
Secondary exposure 0.035 2.4 1.3
15,428,000
15,428,000
10
10
Brake repair outdoors
Exposure from brake dusting 0.71 0.3 1.3
Secondary exposure -0 2.4 1.3
24,130.000
24,130,000
10
10
All asbestos is ehrysot i le and the fiber size is greater than 5 pni.
bit is assumed that exposure will occur once every three years over 30 years; therefore, the
actual lifetime exposure frequency is 10.
5-9
-------
6. OTHER CONSUMER EXPOSURES
Tzanetos (1987) presents a list of consumer products containing
asbestos. The major product categories sti11 thought to contain asbestos
are listed in Table 6-1.
Some of the product categories 1isted in Table 6-1 have already been
covered in this report. Exposure estimates for two other product
categories, mi 11 board and cement sheet, were presented in USCPSC (1986).
Estimated asbestos exposures for mi 11 board being installed with hand and
power tools ranged from 0.044 to 8.0 f/cc-hr averaged over an entire
house and from 0.29 to 51 f/cc-hr in a closed room where the millboard
was being instal led. For cement sheet, exposures were estimated assuming
that the sheet was installed using of an abrasive wheel saw and a drill
or a toothed saw and a drill. Exposures ranged from 2.0 to 9.6 f/cc-hr
averaged over an entire house and 13 to 62 f/cc-hr in a closed room where
the sheet was being installed. Populations exposed to mi 11 board during
installation were not estimated because 1imited information was available
and because it could not be confirmed if consumers can still purchase
this product.
Of the remaining product categories listed in Table 6-1, roofing felt
was considered to be one of the most likely categories to result in
significant consumer exposure. This is because it may release asbestos
when cut and because a large number of consumers are involved in roof
replacements and repairs. Additional information on the other product
categories could not be found. A discussion of the potential for
exposure to roofing felt follows.
Roofing felt, or tar paper, is a sheet material that is generally
supplied in three-foot wide rolls. It can be used as a base material for
several types of common roofs or it can be used as a 1ayering material in
built-up roofs (built-up roofs are generally composed of three to five
1ayers of tar paper and bitumen). Roofing felt is rolled onto the
6-1
-------
7059H
Table 6*1. Consmer Product Categories Thought to
Contain Asbestos
Asbestos product
category
Examples
Mi 1Iboard
Mallboard, garage paneling
Beater-add gaskets
Putips, lawn mowers
High-grade electrical paper
Electrical insulation
Roofing felt
Roofing felt
Asbestos-cement flat sheet
Ovens, safes/vaults
Asbestos-cement shingle
House siding
Braking materials
Disc brake pads, brake blocks
Asbestos cloth
Wood burning stove door seals
Asbestos thread, yarn, etc.
Wicks, mantels, stove seals
Sheet gaskets
Boilers, internal-combustion
eng ines
Packing
Internal-combustion engines, pumps
Roof coatings
Roof coating
Non-roofing adhesives.
Driveway sealant, epoxy adhesive.
sealants, and coatings
glazing compound
Asbestos-reinforced plastic
Skillets, toasters, automobiles
Sealant tape
Automobiles
Source: Tzanetos (1981).
6-2
-------
surface of the roof, cut, and nailed or stapled into place. Roofing
cement is used in some cases to seal the edges or areas adjacent to ducts
or vents.
No asbestos-containing roofing felt was found during a shelf survey,
only fiber glass-reinforced products. Furthermore, if asbestos is
contained in some roofing felt, the asbestos would be tightly held in the
tar matrix, and this material is always covered with some other
material. The most likely time for exposure to occur would be when
roofing felt is cut to fit the roof. No data were available to estimate
this exposure level.
6-3
-------
7. REFERENCES
Anderson PH, Grant MA, Hclnnes RG, Farino WJ. 1982. GCA Corporation.
Analysis of fiber release from certain asbestos products. Draft final
report. Washington, DC: Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. Contract No. 68-01-5960.
Anderson et al. 1973. Asbestos emissions from brake dynamometer tests.
Presented at the SAE Automotive Engineering Meeting, Detroit, MI. As
reported in RTI (1986).
Bingham TH, Anderson DW, Cooley PC. 1983. A benefit-cost analysis of
asbestos product bans. Draft report. U.S. Environmental Protection
Agency. As reported in .RTI (1986).
Bragg G. 1986. Exposure to asbestos: An analysis of the technical
aspects of the Environmental Protection Agency proposal to ban and phase-
out asbestos. Opening written comments of the Asbestos Inforrnation
Association/North America and the Asbestos Insti tute. Volume 11. June
29, 1986.
Bureau of Census. 1985. Annual housing survey: 1983, urban and
regional housing characteristics, Part E. Washington, DC: U.S.
Government Printing Office. Series H-150-83.
Cha S, Carter P. 1982. Estimation of urban automobile brake emissions
through simulation of wear dynamics. Research Triangle Park, NC: U.S.
Environmental Protect ion Agency. Contract No. 68-02-2566. TR-82-08.
Dolan CW. 1985. Charles W. Dolan, 6220 Agnes Road, N.E., Tacoma,
Washington 95422. Letter to Neilson Bros & Sons, Inc., P.O. Box 70395,
Seattle, Washington 99107. Dated March 12, 1985.
Dufour G. 1984. Indoor asbestos air pollution from floor tiles.
Proceedings of the workshop on the monitoring and evaluation of airborne
asbestos levels fol1 owing an abatement program held at National Bureau of
Standards. March 12-13, 1984, sponsored by USEPA and NBS.
Farnsworth BT, McKittrick RS. 1984. The DIY consumer market, 1984
reference guide. Indianapolis, IN: DIY Research Institute.
GSC. 1986. General Sciences Corp. Graphical Exposure Modeling system
(GEMS) user's guide. Volume 2. modeling. Appendix A. User's guide for
GEMS atmospheric modeling system (GAMS) Version 1.1. Landover, MD: U.S.
Environmental Protection Agency. Office of Pesticides and Toxic
Substances. GSC-TR8658. Contract No. 68-02-3970.
7-1
-------
HomePro Systems, Inc. 1985. The homebook. Falls Church, VA: HomePro
Systems, Inc.
ICF. 1986. Asbestos exposure review. Washington, DC: Economics and
Technology Division, Office of Toxic Substances, U.S. Environmental
Protection Agency.
Jacko MG, DuCharme RT, Sotners JH. 1973. Brake and clutch emissions
generated during vehicle operation. SAE Automobile Engineering Meeting.
Detroit, MI; May 14-18.
Jenkins J. 1985. Asbestos fibers release from VA tile. NAC Journal.
Summer Issue.
Jennings PD et al. 1985. Methods for assessing consumer exposure to
chemical substances. Volume 7. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency. EPA 560/5-85-007.
Kauppinen T, Korhonen K. 1987. Exposure to asbestos during brake
maintenance of automotive vehicles by different methods. Am. Ind. Hyg.
Assoc. J. 48(5):499-504.
Lemmons T. n.d. Method of clearing debris from brake drums, study of
asbestos exposures. Report to the Office of Toxic Substances, Office of
Chemical Control. Research Triangle, NC: U.S. Environmental Protection
Agency.
MVMA. 1986, Motor Vehicle Manufacturers Association. MVMA motor
vehicle facts and figures '86. Detroit, MI: MVMA.
RTI. 1986. Research Triangle Institute. Regulatory impact analysis of
controls on asbestos and asbestos products. Appendix N. Prepared for
the U.S. Environmental Protection Agency.
Roberts DR, Zumwalde RD. 1982. Industrial hygiene summary report of
asbestos exposure for brake mechanics. Cincinnati, OH: Division of
Surveillance, Hazard Evaluations and Field Studies, National Institute of
Occupational Safety and Health. As reported in ICF (1986).
Rodelsperger K, John H, Bruckel B, et al. 1986. Asbestos dust exposure
during brake repair. American Journal of Industrial Medicine 10:63-72.
Rohl AN, Langer AM, Wolff MS, and Weisman I. 1976. Asbestos exposure
during brake lining maintenance and repair. Environmental Research.
12:110-128.
7-2
-------
SMRB. 1982. Simmons Market Research Bureau. Simmons media and market
report. New York, NY: Simmons Market Research Bureau, Inc.
Sebastien P. 1982. Indoor airborne asbestos pollution from the ceiling
and floor. Science Vol. 216. June 1982. pp. 1410-1412.
Tzanetos P. 1987. List of consumer and industrial products containing
asbestos. Memorandum from Peter Tzanetos, ICF Corp. 1850 K Street,
Washington, DC 20006, to Christine Augustyniak, U.S. Environmental
Protection Agency. Dated January 30, 1987.
USCPSC. 1986. U.S. Consumer Product Safety Commission. Asbestos in
consumer products: options. Washington, DC: U.S. Consumer Product
Safety Commission.
USDOC. 1983. U.S. Department of Commerce. County and city data book.
1983. Bureau of the Census.
USDOC. 1984. U.S. Department of Commerce. Population and land area of
urbanized areas for the United States and Puerto Rico: 1980 and 1970.
1980 Census of Population. Bureau of the Census.
USDOC. 1985a. U.S. Department of Commerce. Statistical abstract of the
United States 1986. Washington, DC: Bureau of the Census.
USDOC. 1985b. U.S. Department of Commerce. 1982 Census of
transportation. Truck inventory and use survey. Washington, DC: Bureau
of the Census.
USD0T. 1986. U.S. Department of Transportation. National
transportation statistics. Annual report 1986. Washington, DC: USD0T.
Research and Special Programs Administration.
USEPA. 1980. U.S. Environmental Protection Agency. NEDS National
emissions data system information. Research Triangle Park, NC: Office
of Air Quality Planning and Standards. EPA - 450/9 - 80 - 013d.
USEPA. 1984. U.S. Environmental Protection Agency. User's guide to
M0BILE3 (mobi1e source emissions model). Ann Arbor, MI: Office of Air
and Radiation.
USEPA. 1986. Guidance for preventing asbestos disease among auto
mechanics. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency.
7-3
-------
Versar. 1984. Exposure assessment for asbestos. Washington, DC: U.S.
Environmental Protection Agency. EPA Contract No. 68-01-6271.
Versar. 1985. Consumer exposure to adhesives. Draft final report.
Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency. EPA Contract No. 69-02-3968.
Versar. 1986a. Users' guide to the computerized consumer exposure
models. Washington, DC: Exposure Evaluation Division, U.S.
Environmental Protection Agency. Contract No. 68-02-3968.
Versar. 1986b. Standard scenarios for estimating exposure to chemical
substances during use of consumer products: do-it-yourself products.
Volume 11. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency. EPA Contract No. 68-02-3968.
Williams RL, Muhlbaier JL. 1980. Contribution of brake lining wear to
urban airborne asbestos. General Motors Research Laboratories. 6MR-3422.
Williams RL, Muhlbaier JL. 1982. Asbestos brake emissions.
Environmental Research 29:70-82.
Ziskind et al. 1982. Study of cancer mortality in chrysotile asbestos
mining counties. Research Triangle Park, NC: U.S. Environmental
Protection Agency. Contract No. SAI-068-82-531. As reported in RTI
(1986).
7-4
-------
APPENDIX A
Determination of Ambient Asbestos Concentrations
from Vehicle Brakes
-------
NOTES TO APPENDIX A
Appendix A contains the worksheet that was developed by Versar to
estimate the ambient asbestos concentration in 24 U.S. cities resulting
from vehicle brakes. This worksheet was developed on a personal computer
equipped with spreadsheet software.
The method used to arrive at the estimated concentrations is
described, in general, in the text and, in detail, in the footnotes to
each column in the worksheet. Bases for the input data 1isted in the
table in the lower right-hand corner of the first worksheet page are
discussed in these notes.
Fractions of gas used by cars and trucks: The total quantity of gas used
by light-duty gasoline-powered vehicles (from NEDS) was apportioned
between cars and 1 ight trucks based on the number of vehicles registered
in each class as shown in Table 2-4.
Mi leage: The mileage values that are used by USEPA in estimating
emissions included in, and reported by, NEDS (USEPA 1980) were used
here. It is expected that current figures would be higher than those
used here because of recent trends toward higher mileage vehicles;
however, because the reported values are still used in NEDS, they are
also used in these calculations.
Total distance travelled bv each group: For cars and light trucks, this
value was determined by multiplying the number of vehicles in each class
by the average distance traveled per vehicle per year as reported in
Table 2-4. However, this method was not used for the cateogry of heavy
trucks because of the 1arge differences in the average distance travelled
by heavy trucks as reported in various sources (see footnote (b) to Table
2-4). To lessen the chance of propagating any errors in this value, the
A-1
-------
total annual distance travelled by heavy trucks was taken directly from a
report by the Motor Vehicle Manufacturers Association (MVMA 1986). Note
that the value taken directly from MVMA (1986) corresponds well with that
11
which would be derived from data in Table 2-4 (1.18 x 10 km from MVMA
(1986), 1.22 x 1011 km derived from Table 2-4), indicating a good
possibility that the figure reported in Table 2-4 is correct.
Average national emissions: Total airborne asbestos emissions from
vehicular brakes (in kg/yr) for cars, 1ight trucks, and heavy trucks from
3 studies are reported in Table 2-3. These values were averaged over the
three studies, and the average value converted to units of g/sec. The
average values that were used in this worksheet were based on actual
values obtained before any rounding took pi ace (values reported in Table
2-3 have been rounded to two significant figures). The averages that
were used are:
Cars: 2,800 kg/yr
Light trucks: 1,753 kg/yr
Heavy trucks: 2,655 kg/yr
For example, for light trucks:
(1753 kg/yr) x (103 g/kg) x (1 yr/365 days) x (1 day/24 hr)
x (1 hr/3600 s) = 0.05559 g/s.
A 2
-------
Ijblr fi-l. Anbirnl fHlwsto* Irvrh HrtuHinf Irzm Vehicular Shrikes - Worksheet
lesijiulor til
ASA
A B
c#*i« suubce im «w sim
Ui9M2l BIKN tkiMZI
C
cm pop
(21
D E
taw* GA5 USED 8< L Iffll
pup ill bis vehicles i«i>i*i
F
EflS IGE0 8f MRS
lift 1 (5)
S
wis iho m ii.
TRUCKS IGai(6l
M
BIS 1MB B1 II (W
EftS IM1X5 IB111*1
1
run use it it aw
DIESEL inns (HIIUI
J
RUES 1Y CARS
IHIUSI171
I l«lt liort, M>
ftllliaor* Co
208
1*
mm
mm
*.3IW{*06
3.*I2*I«*
9.016**0?
3.*26U*07
7.8568f»0?
S. 711*09
2 Oiiltinoofa, IN
H»ilt«n
m
16
169565
25*H9
l,®J55E*06
».»?!£«?
2.2«7«'07
9.93WI*06
3.6163*07
l.*3£«09
1 Dnwir, CO
Ornvpr
m
IT
*92365
*92365
»!»((
2.*i6lf*06
6. *3681*0?
2. ?*71E *07
5.2 WE *07
*.I0t*99
* Oflroit, HI
Hiyr*
351
19
imm
J2825W
6.997lf*06
5.5J*7£.06
1.*6211*08
6.I05«>0?
1.191**06
5 Houston, H
Hgrrii
1**0
36
1595136
2315609
I.19J6EMJ9
1.MI«*06
2. W45f *0#
l.015X*06
2. J0*X«06
1.596*10
1 Ln Rngrltf, CA
Im Bn|*ln
1204
35
2*6150
2916850
2.9652EMJ9
2. 3*551*09
6. 19731*06
2.l7*tE«06
t.0ri)E*06
3.W*I0
I Kim), fl
Bide
U.l
9,*
3*6665
I6W59
6. WW *06
5.173tt*06
l.36?M*«
*. 9?06E*07
7.32351*0?
6.72t*W
S toctirstcr, Iff
Hhitm
66.5
9.*
2*17*1
imm
2.26«»06
1.79551*0#
*.7»«lf*07
l.*««.0?
1.939tE*0?
3.03E»M
9 St. 1ouii, »
SI. Lcxiii
155
II
*53065
~51065
1.35*X *06
I.07I3E*06
2.6Jt«5£*07
6.151«*06
5.09Bt*07
1.B1EW9
10 Uictuti, K5
Stdgntick
262
tt
imii
mm
l.3*Wt«06
1. (fine KM
2.6003EK)?
7.9**0(*06
3.5*12f *07
l.)*»09
II Boil*, It
Ml
102
10
102*51
13*6*6
. 6.037lf*07
*. 775Jf*#7
I.26IMHJ?
1.2953*0?
1.03*06
12 Oilrlnlon, WV
Ktiwhi
)2.S
a. 5
tl'M
151*51
?,533«*6?
5.956TE*0J
1.57*51*07
2.I232EM)?
2.8J8«*(»
I.OOE'M
|] Malt), m
St. lain
in
13
92111
102AM
6.526*E*0?
i.nmm
1.762*1*07
6.225«*06*
1.11121*91
l.l«*OT
1* Eui}»r*, OS
lire
St
5.2
105625
162*95
I.|92«*06
9.*322E>07
2.*9221*0?
6.»5*«*06
2.67*21*0?
1.5* K»
15 lirtoln, HE
Urciitrr
m
12
111912
W3550
6.*39(jC*07
5.09321.07
1.3*561*07
5.6W.06
1.27561 K)7
«.5BE*(«
It Plwil|M*rjf, H
Rant (wry
332
tl
wmi
n mi
6.»37E*0?
*.9B6?E*07
1.3I7«*07
9. WK)t»06
I.*790£t07
6. *tt*06
I? QtfWii, II
Icier
n
6.7
9M2J
100951
*.9I21E»07
3. BBSS to?
l.0266£*07
». >06
1. 756V *07
6.53*08
It Portlird, m.
Cuitxrltrd
60.6
ht
61572
107099
»,09*3£*«
J. 1916M?
l.900?l*0?
5.?WS*«6
I.«5«t0?
1.216*09
11 (lain, NY
Oifwirig
Njpj
HA
*.3
3532?
mil
3.805* t0?
3.0105E*0?
?.95*K»«
2.3»EK<6
I.2910E*6*«»06
5, *9106*06
«.?«*«
23 lochnlrr, M
OlnlNd
«
7
57B90
60*73
2.9258*0?
?. ?m*o6
?.619C(*06
1.308ft *06
*.93EK»6
2* SinUfr, m
Smtl Ft
11.5
»
*6953
52042
3.966«»07
J. 152* *07
6.33071*06
1. 71601 *06
*.37606 tflfc
5.311*06
ID*
nt
m
It
mm
mm
« for this (Mrr, th* misioos mtrt 4tttr«i'«* *1 » *Htrn»lt «tllw*f. S*« Srcttmt ,or ln •»»'«A*,lon of
mo sua fmssims mi#
USB IMOOOUT Wis
(11
Ihi coin" «wil"»tor 11 "
ti to itffntilr thr colwiM mtMn the iqnilion »«< (or thu 4tt»r»in»lion.
tmotn
(2)
USSUC. 1963.
FKC1I0N a GAS 1I5EI If HWS;
0.791
1010L BISIfKI Tfismif®
»»iftDi sow', »ilimit
(11
IK DOC. 196*.
fftXIIW Of EMS IU0 EH IBJCHS:
0.209
UWSIMO rtmi* IfME 2 *11
LI IMJCKSIIVO ((BUB IftHE 2-*H
2.0**12
5.i*«H
HI
wss.
HUimi IU9PR I960)
tfflwY IRUCKSinO WflW 1966li
1. IMM!
CMS lltCli
16.65
RLllMUl
2.7JE*»2
151
E . 0.791
nail irwks men
II m 60S IIMM IUBS;
12.81
*,»1
Mm Miiim. imssiws is/stci
161
E • 0.201
Ifflvt DIESEL 1 BUCKS IWC.I:
5.11
(Mi
0.06611
H mcK:
O.tftUl
III
f > 16.65
11 MY IdCK;
«L;
».04ia>
-------
libit fr-l. (Conlirtuvd)
mgMtor til
K
I
N
N
0
P
e
*
S
1
MEA
mus «ii.
MILES BY HEW* MS
MIES IT WW* 61ESEI
T0IB. HUES
TOTAL SISTMX
ESI «M OltSS. tSl MEfl WISS 11.
1ST fiSEfl fltlSS
esi MEfl aiss mc based on win
mens (musim
1HJCXS (HUES) (9)
mas mueshioi
miiESiiiu
<101)1121
CMS (|/wtlU3>
1IMXS i|/i)(UI
Vim IKS t|/il 1151
flll|/llll6>
[KISS (u|/kJ)1171
1 hllim, 1®
i.iaws
l.UC'H
4.54t*0»
J.iJEKW
1.211*10
3.93E <«
I.I5SHH
7.l« «
l,aE-03
I.62E-0I
2 OiilliKWf*, W
£.M£'04
4.811*0?
2.0**06
2,0tt»09
6. ME 05
1.02M8
1.9«-0<
2.ME-04
I.1S-0I
1 ten**-, CO
M#S«1S
3.0Irt8
5. 341*09
1.6CE*S9
2.III-W
l.32£-«
i.7S-0(
I.I7E-04
1. WE-01
4 Detroit, HI
1.171 *09
2.3* *04
i, us tea
i. mm
I.OIMO
3,3HE-«
3.9K-M
I.W-03
I.20E-O2
3 Houston, H
3.20E»09
!.Xi»09
1.461*10
2.J5SUO
?.««
I.52M4
I.WE-M
2.7J1-03
2.21-02
% L<* ftnjcln, Efl
i.mm
I.OtE'W
2.3tErtfl
5.0**10
1.1**10
2,JOE-01
1.271-03
3.«-03
I.90E-O3
3.27i-02
? Mini, a
i.mm
2.411*04
4.2»t0«
i.mm
3, Ml *09
1.2*-04
fcO€-05
1.64E 04
3.53E-04
14JI 01
1 luctuntrr, W
fc.ttrtl
I.12t»0i
i.mm
2.4St<8
I.Ht-05
1.UEHXS
I.27E-05
2.041 04
2.52E-OI
1 SI. Louit, »
1.UE.M
iKifl
i.mm
(,03l«09
I.2IE-W
5,Ki-<6
J.W-M
5.6M 04
1.63-01
10 iictiili, US
3.59E»0i
3.MMl
2.0S*OI
2.I»E*«9
3.5IE*09
i. iaw
5. lit -05
».!*-«
9.74E-02
11 lam, 11
i.mm
i.mm
J. HE>07
I.IZEtOt
1.JIE*09
MK-05
l.%f-05
I.9SE-0S
1.511-04
2. 04
l.3W«09
3.991'-05
i.m -05
i.m-04
2.0Ef-04
3.5IE-0I
IS ftjrtlMd, PC
i. iot»oe
9.lt£«Ot
MiWS
k.m-m
2.2«H6
9. HI-05
1.611-04
3.971-01
IS Eluri, wr
I.02£t08
1.17E *07
l,«t«>7
3.3IIM*
13**0#
i.we-is
».«-«
4.47E-05
I.44EHX)
20 M
9.K3t»07
9. JCtHS
1. 42£t07
S-lItCH
1. Ill-05
i.saE-oa
2.3S-0S
6.45E-OS
6.34E-OI
21 &«!#, fl
2.I9E»0J
4.3BEM)?
9.65EH«
l.5SE«09
S.22E-C9
2.«-05
5.52E-05
I.32E-04
4.7M-01
22 Pin* II,M, M
I.5JH07
I.JK*?
3.171*07
i-OK.04
i.09E«<*
2.6JE-05
l.2«H»
4.S9E-05
I.5U-0J
5.42E-01
21 fbcfunttr, HH
9,**««
J.28MJ
2.49M7
6.0*. OS
9.70f.«
3.23E-05
1,521-05
4.14C-05
l.«-65
3-Mt-Ol
24 &inU Fl, HI
1. <67E ~cm
1.1* >06
2,53£t07
t.3a<«
l.02£«O9
3.*a-«
1. tIt -05
3,t4£-«
».ti£ 06
J. 411-01
Itllinrt, HH
l.50t-M
l.nt-05
5. ME-05
2.43S-04
1.63-01
• for thu tntrr, tb* nution *r# «rt»»»inrt by in ilttrMU ¦•Ihod. Sw Stctlan 2.2 for « «i»l«njllon of M» •thai,
ill G ¦ 12.11 *>ln'|>l
IfI » « i.OT
110) I i 3.71 ¦)ln/|il
till IJ i I • L ' II i tC/OI
»2l H • I.609 M/«il*
uii to. own «I > i.6o» i « ic/ii
(HI t®. 05559 i * i I.6M i 5.5**11! »
(IS) t».0#42 ¦ (I * HI » 1.609 I I.IKMH • IC/B!
Oil 1.609 ¦ ICfOl . til » 0.08»Htf.«»W • t« « 0.t«i3/i.S*t|tl ~ III « *1 . 0.06*2/1. IK»IIJI
till SnulH o« (OtKS *c4f
-------
l«M* ft~l. (Continue)
tarp
Miu
SmII
<11
u
V
u
I
*
1
M
08
ftf5.fi
CMXMtmillH
rac. nan
COC. ttfMf
awBiKD cr»c
CMBWMtlW
owe. nan
coc. mm
CMIKS D3NC.
CflRS U|/i3IH8l
IREXS ti|fi3)(19) 1 RUCKS
lu|/a3M2tl
CARS lrt|/«3N221
TRUCKS (M/fcJ) I23J 1FUXS Infill (241
1«1>*311251
1 l*lli«or», m
UK-05
2.W-05
1.lit-04
2.0*-04
0.0&3I
0.0299
0.1156
0.209?
t D>«IUnoofi, IN
I.3IE-06
3.47E06
2. ?JE <15
3.3IE-05
0.0014
0.00J5
0.0223
0.0331
3 Bfttvtr, £0
J.CSE-05
I.45E 05
5.&HJ5
f.Itf-05
0.0309
0.0145
o,05a
0.0976
4 Drtrail, HI
2.45E-OS
1, IS 05
4, 33E-05
I.SS-05
0.0245
0.0115
0.0435
o.om
5 Houston, II
1.(IE-OS
I.83E-06
3.6* 05
i.«-os
0.0161
0.0W1
0.0363
0.0W9
& Lot CA
8.8*-05
~.IS -05
i.m-m
2.58E-04
0.0M3
0.0415
0.1284
0.2583
rmmt, a
~.~It-05
2. ore-os
5.641-05
I.2IC-04
0.0411
0,0201"
0.0564
0.1212
1 locfcfitw, Iff
2.0K-05
1.JS-06
2.08E-05
5.IW-05
0.020#
0.HO98
0.0208
0.0514
1 SI. louii, M>
J.0CH!
1.40E-06
I.2K-05
1,161-05
0,0200
0.0094
0.%22
0.0116
10 Uictiili, KS
i.ttE-a
5.IIH*
2.4*-05
4,0*-05
e,oio»
0.0051
0,0249
0,0408
1) Bom, IB
i.re-oe
~.«-<*
I.8K-05
3. HE-05
0.0081
0.0041
0.0186
0,0314
It Our 1 tit on, IN
1.#11-05
USE-05
1.04E-04
0.01II
0.008$
0.0715
0,1042
13 Sslulh, m
1,011-05
~.?*-»
l.4lt 05
2.1*-05
0.0101
0.004J
0.03*1
0.0281
14 lu|n», OS
1.2 *-05
1.05-05
4.1(t-05
1.9* 05
0,0223
0.0105
0.0411
O.OI99
IS lircoln, M
I.40C-06
6.59E 07
2.IIE-06
4.8*1 06
0.0014
O.OOOI
0.0021
0.0041
16 Nonl|ai«rj, flt
6.3*-06
Z.UE-M
uw-os
2.6* 05
o.eca
0.0030
0.0161
0,0263
IT OdtKi, 11
I.W-05
6.59E-06
5.1* 05
I.25E-05
0.0140
O.OOii
0.0519
0.0725
ilPailind, K
l.ME-45
l,HE-Q»
I.Ot-05
1.411-05
o.om
0.0069
0.0363
0.0641
11 Elain, in
3.30C-C5
1,53-05
3.361-05'
1,221-05
0.0330
0.0155
O.OJ36
0.0822
20 10pi, Cfl
I.JK-05
8.36E-06
I.4JI 05
4.0*-05
0.01II
0.0084
0,014?
0.0409
21 OcjU, a
2.49E-05
1.lit-05
2.64E-05
C.Jtt-05
0.0241
0.0111
o.
-------
APPENDIX B
Secondary Concentrations of Asbestos in a Garage
Occurring from Brake Repair
-------
Appendix B
Secondary Concentrations of Asbestos in a Garage
Occurring from Brake Repair
The purpose of this appendix is to estimate the asbestos
concentration in i garage resulting from dust cleaning. From the brake
dust cleaning scenario in the text, dust cleaning was assumed to take
place for 0,3 hours, resulting in an asbestos concentration of 0,71 f/cc
near this operation. If it is assumed that this concentration is
3
contained in a 5 m volume around the consumer, the number of released
asbestos fibers in the garage after 0.3 hours is as follows;
0.71 f/cm3 x 1 x 106 cm3/m3 x 5 m3 = 3.6 x 105 f
3
Based on a typical one-car garage volume of 60 m (Jennings et al.
1985), the average concentration throughout the entire garage is
3,6 x 10® f/60 m3 or 5.9 x 104 f/m3. The average concentration
would decrease at this time since no additional releases are expected and
because fresh air wi11 exchange with the contaminated air in the garage.
The following equation was used to estimate the average asbestos
concentration in the garage for the remaining 2.4 hours that the consumer
will require to finish the brake job:
-CoV -m(Q/V)t
-avg (tl - t0) = -jjjjp e
t
to
where
Cayq = Average concentration over the period
tj = Length of exposure =2.4 hours
t0 = Assume start of exposure =0
C0 - Initial concentration = 5.9 x ID4 /nr
m = Mixing factor - 0.5 (Jennings et al. 1985)
Q = Ventilation flow rate = 60 nr/hr
(For 1 air exchange/hour, which is typical from
Jennings et al. 1985)
V - Garage volume = 60 m3 (Jennings et al. 1985)
B-1
-------
This equation is from Jennings et al. (1985) and it is for estimation
of gaseous contaminant concentration in an indoor environment. It is
assumed that the asbestos particles (<5 um), which will be released
during brake dusting, are so fine that they will behave in a manner
similar to gases. Substituting into the above equation:
5.9 x 10^ f/m3x 60 m3
-0.5 60 x 2.4
Cavg (2.4 - 0) - -
60 m3
0.5 x 60 m3
5,9 x 104 f/m3 x 60 m3
0.5 x 60 m3
-0.5 60 m3/hr x 0.0
+
60 m3
2.4 - 0
( -1.2 x 105 x 0.30) + (1.2 x 1Q5 x 1.0)
2.4
= 3.5 x 104 f/m3 or 0.035 f/cm3
B-2
------- |