United States       Office of Air Quality        EPA-450/4-79-011
Environmental Protection   Planning and Standards      June 1979
Agency          Research Triangle Park NC 27711

Air



Evaluation of Particulate


Emission Factors


for Vehicle Tire  Wear

-------
                                       EPA-450/4-79-011
Evaluation of Particulate  Emission
    Factors for Vehicle Tire  Wear
                          by

                      Joseph Carvitti

                  PEDCo Environmental, Inc.
                    2420 Pershing Road
                  Kansas City, Missouri 64108
                   Contract No. 68-02-2585
                     Assignment No. 9
               EPA Project Officer: Charles C. Masser
                       Prepared for

            U.S. ENVIRONMENTAL PROTECTION AGENCY
                Office of Air, Noise, and Radiation
             Office of Air Quality Planning and Standards
            Research Triangle Park, North Carolina 27711

                        June 1979

-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees,  and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35), U.S.  Environmental  Protection Agency,
Research Triangle Park,  North Carolina 27711; or for a nominal fee,
from the National Technical Information Service, 5285 Port Royal Road,
Springfield,Virginia 22I6I.
This report was furnished to the Environmental Protection Agency by
PEDCo Environmental,  Inc., 2420 Pershing Road, Kansas City, Missouri
64I08,  in fulfillment of Contract No.  68-02-2585, Assignment No. 9. The
contents of this report are reproduced herein as received from PEDCo
Environmental Inc.  The opinions, findings,  and conclusions expressed
are those of the author  and not necessarily those of the Environmental
Protection Agency.  Mention of company or product names is not to be
considered as an endorsement by the Environmental Protection Agency.
                    Publication No. EPA-450/4-79-011
                                n

-------
                            CONTENTS
Figures                                                        iv
Tables                                                         iv

1.   INTRODUCTION                                               1
2.   METHODS OF EVALUATING TIRE TREAD WEAR                      4
3.   CALCULATION OF TIRE WEAR ESTIMATES                        11
          Mechanics of Tire Wear Deposition                    16
4.   SUMMARY AND CONCLUSIONS                                   21

REFERENCES                                                     22
                                ill

-------
                             FIGURES
Number
  1

  2
Simplified Schematic of Tire that is Generating
  Cornering Force
Mechanics of Material Loss and Emissions from
  Tire Wear
Page

   2

  18
                             TABLES
Number
  1
  2
  3
Methods of Evaluating Tire Tread Wear
Emission Estimates from Tire Tread Wear
Rank of Emission Factors
   5
  13
  17
                                IV

-------
                            SECTION 1

                          INTRODUCTION


     Vehicle-related emissions are a major contributor to urban
particulate concentrations.   One of the vehicle-related emission
sources is tire wear.  It is estimated that 3.2xl08 to 7.2xl08 kg
of tire wear debris are generated each year in the United
States.1'2'3'4  Particles from tire wear have been identified by
microscopic analysis as a significant component of samples taken
from roadside, high volume,  particulate samplers.5'6

     The currently available emission estimate for tire wear,
presented in Compilation of Air Pollutant Emission Factors (
AP-42), is 0.20 g/veh-mi for a four-wheeled, light-duty vehicle.
This emission factor is widely used in regional particulate
analyses to estimate emissions from tire wear; however, this
factor needs to be reviewed, based on studies and test data re-
leased since its publication.

     Particulate from tire wear is generated as a result of
forces existing between the tire and the road surface.  These
forces are transmitted by the tire to the road surface during
acceleration, deceleration,  and directional changes.  Owing to
tire elasticity, tire deformity occurs with differences in the
magnitude or direction of travel that occur as the operator
controls the vehicle.

     The deformity of a tire negotiating a curve results in a
kidney-shaped contact area  (shown in Figure 1).  The front part
of the tire adheres to the road by friction.  Elastic forces
arising from distortions in the rear part of the tire overcome
friction, causing the tire to react horizontally and to slide
over the road.  Adhesion in the front part of the tire is instru-
mental in achieving a change in direction, while abrasion is
confined to the rear part of the contact area.  Similar action
occurs during braking and accelerating, but the direction of the
strain and sliding is then circumferential.

     Several factors influencing the magnitude of abrasion in the
rear part of the contact area have been studied.  These factors
are both external (tire forces, pavement characteristics, and
environmental conditions) and internal  (tread composition, tread
design, and other tire features).  Tire forces have been shown  to

-------
  WHEEL PLANE
CORNERING FORCE
   LATERAL  FORCE
                           CAR PATH
                               INSTANTANEOUS
                            DIRECTION OF MOTION
                               SUCCESSIVE POSITIONS
                                 OF TREAD ELEMENT

                                  CONTACT AREA OR PATH
 6 SLIP ANGLE

©ADHESION ZONE

® SLIDING ZONE
 Figure 1.   Simplified schematic  of tire that is  generating
                      cornering force.
 Source:   Reference 7

-------
be the major source of tire wear.7'8  These forces include longi-
tudinal forces of driving and braking and,  most significantly,
lateral or cornering forces.  Pavement type has also been shown
to influence the amount of tire wear.^  Studies have also mea-
sured tire wear as a function of tire tread temperature.8'10
Many studies have been performed to estimate wear as a function
of tread composition, tread design, or tire type.  Most of these
studies were done to evaluate performance ratings, relative to a
control tread stock, rather than to evaluate the rate of wear per
mile of travel.  Several experiments have defined what is known
as wear test severity.  From an analytical and experimental
standpoint, there are three wear severity classifications:  tire
force severity (intensity and distribution of tire forces),
abrasive surface severity (type and condition of pavement), and
weather temperature severity (tire and ambient temperature).7

-------
                            SECTION 2

              METHODS OF EVALUATING TIRE TREAD WEAR


     Testing of tires in the past has been done for two reasons:
control tests on samples selected at random from current pro-
duction; development tests designed to evaluate the effects of
experimental features of design, of compounding, or of changes in
processing procedures.10  More recently, the concern over partic-
ulate and gaseous emissions resulting from tread wear has prompted
experiments dealing with tread loss and airborne tire debris.

     Direct measurements of tire wear are made by two general
methods, indoor and outdoor testing.  Laboratory abrasion tests
can only be justified if they correlate with actual service
performance.  These indoor tests are usually run on a machine
that consists of a steel roller against which single or multiple
tires are pressed at controlled loads.  Outdoor tests are per-
formed using specially built trailers, single test vehicles, or a
fleet of test vehicles selected at random or contracted for.
Tread wear rates are determined either by weight loss or tread-
depth loss of the test tire.  In addition, indirect estimates of
tread wear are obtained through ambient air sampling in a tunnel
or near a roadway.

     The methods used in these tire wear studies are listed in
Table 1, with advantages and disadvantages of each method out-
lined therein.  References for data derived by each method are
also included.

     Of the seven test methods identified in Table 1, indoor
testing on a chassis dynamometer and outdoor vehicle testing
(either single vehicle or a fleet) are the methods determined to
be most representative of actual tire wear.  The indoor testing
provides a method for the determination of particle size distri-
butions, while the outdoor testing provides a measure of the wear
rate of a tire in actual use.

     Because both test methods directly measure tire miles trav-
eled and tire tread wear, the test results are easily compared
with one another.  Factors such as tire inflation pressure, road
surface type,  tire tread temperature, and tire force can be
closely controlled to provide reliable results in both cases.
The absolute wear rate may not be equal for both tests, but the

-------
                             TABLE 1.   METHODS OF EVALUATING TIRE  TREAD  WEAR
Technique
Advantages
Disadvantages
                                                                                               References
Indoor testing,
 on chassis
 dynamometer
 or tire
 abrasion mat
Outdoor testing
 specially
 built trailer
Low test cost
Relatively rapid procedure
Possible ±10% precision
Can test easily under varied
 conditions of load,  speed,
 tire inflation, etc.
No interference from adverse
 weather conditions
Tire surface temperature can be
 measured with relative ease
Particle collection method is
 simple
Particle size determination is
 possible
Accelerated procedure provides
 short testing periods
Precise control of test
 conditions possible
Typical of normal tire use
Cornering force can be varied
Forces compare well with actual
 forces encountered
Tire comparisons can be made
 under exactly equal tire forces
Independent study of three major
 external tire forces is possible
Accurate record of operating
 variable is obtained
 a   No  original data presented.

 (continued)
No wear from directional control
Tire in contact with the same
 surface during entire test, pro-
 ducing elevated tire surface
 temperature
Roller surface is not precisely
 comparable to a road surface
Different forces associated with
 rubber tire in contact with a
 roller or mat than on pavement
Tire acquires a "sticky" appear-
 ance not typical of on-road
 operation
Steel roller does not have same
 effect as a road surface on
 carcass performance
Fugitive dust may interfere with
 results

Testing costs can be high
Accelerated condition may cause
 abnormal tire wear
Restricted conditions of road
 surface and climatic conditions
Tire static load and velocity have
 only minor contributions; both
 are major forces encountered in
 actual use
Debris  is not collected for par-
 ticle  size determination
2 4 9 10
C- 1^ ) J t ' U 5 -
 n,i2,i3c
7,!0,a13a

-------
      TABLE  1  (continued).
     Technique
Advantages
Disadvantages
References
     Outdoor  testing,
       single  wheel
CTi
     Outdoor testing,
      single vehicle
      or fleet of
      vehicles
Catch plate mounted behind tire
 provides convenient sampling
Normal driving conditions
 provide wear abrasion
Particle size determination is
 possible
Accurate record of operation is
 obtained
Tread wear measured by weight
 or tread depth loss
Closely controlled test cond-
 itions sometimes possible
Random sampling possible
Wear measured under normal use
Accelerated tests can be conducted
Sectional tires can be used to
 determine an average wear rate
Results can be compared with indoor
 testing results
Accurate and detailed operation
 records may be available
Catch plate sometimes  lost
 during test
Small particles are swept away
 by airflow and neglected from
 analysis
Fraction of particles  caught
 cannot be quantified
Collection efficiency  can be as
 low as 4.5%
Material balance cannot be
 established
Lower limit of detection is 5 u
Below 5 u, dirt and rubber
 particles look similar

Long periods of testing required
Measurements difficult to obtain
 for a random fleet sample
Road surface and climatic effects
 difficult to evaluate
Maintenance conditions may not be
 under control of experimenter
Tire sizes and types are limited
Service records may be incomplete
Assumptions required on the den-
 sity of the average tire
Measurements may be affected by
 gaseous breakdown of tire
No particle size determination
 possible
2,7,9,10,a
 13,a14,15
        No original data presented.
     (continued)

-------
TABLE 1 (continued).
Technique
Advantages
Disadvantages
References
Rubber abrasion
 equipment (e.g.,
 PICO abrader)
Tunnel air
 sampling
Roadside high
 volume sampling
Low cost
Rapid test
Particles easily collected
Weight measurements provide
 precise analysis
Difference in selected compound
 (e.g., SBR) can be attributed
 to tire wear
Can measure submilligram amounts
 of SBR
Relatively simple sampling pro-
 cedure
Particle size distribution may
 be possible
Testing done under completely
 different conditions than
 normal tire use
Rubber curing is an important
 factor to be considered
Mechanics of wear are not the
 same as actual tire wear

Analytical method has possible
 interference from diesel fuel
 particulate
Removal of interfering sub-
 stances also removes SBR
Large air sample required to
 extract detectable amounts of
 SBR, composing only 0.1% of
  the sample
Deviation in the range of ±20%
 can be expected
Only 2 to 7% of SBR is
 airborne
Only 65 to 70% of SBR can be
 collected
Not all treads contain SBR

Only rough estimates of tread
 wear can be made
All of the disadvantages for
 tunnel air sampling also
 apply
16C
6,17,a18,a
   No original data presented.

-------
fractions occurring as airborne or sedimentary in the outdoor
tests can be assumed based on the fractions occurring in the
indoor tests.

     Outdoor trailer testing is not considered in the emission
factor determinations.  The advantages listed show that precise
control of conditions is possible and forces compare well with
those encountered in actual use, but the disadvantage of accel-
erated wear is considered more important.  Abnormal tire wear
under these accelerated conditions can easily result in erro-
neous, although seemingly correct, estimates.  Outdoor single
wheel testing is also eliminated from wear rate considerations.
The major reason for this is that the collection method is inade-
quate for quantifying particulate emissions from tire wear.  This
is also the main reason for eliminating the tunnel air sampling
and roadside high volume sampling methods.  The Pico method
abrades only a sample of rubber in a completely different manner
than actual tire abrasion.  This method is used primarily to
compare different rubber compounds and does not appear acceptable
for determination of tire wear.

     As indicated in Table 1,  there are five studies using indoor
test facilities that produced original data.2'4'9'11'l*
The indoor facilities included a chassis dynamometer, particulate
collection devices, and various control and monitoring devices.
In one experiment, gas samples were also collected.  The chassis
dynamometer is an instrument that allows an actual car or tire to
be driven on rollers, while control is maintained over speed,
load, and sometimes slip angle.  In four of the tests, the con-
tact rollers were constructed of steel; in one, the roller was a
36-mesh abrasive wheel with 120-mesh carborundum grit.  The
latter roller was constructed to simulate an actual road surface.
Most of the experiments had no control over slip angle, so no
wear from directional control was produced.

     The particulate collection system usually consisted of a box
completely enclosing the tire being tested.  Airborne particles
were measured in a known volume of air drawn through the box to
the measurement device.  In some cases, particle size deter-
minations were made using an Anderson sampler or a similar device.
Sedimentary particles settled to the bottom of the box and were
collected separately for weighing.  Speed, load, and miles trav-
eled were monitored in all tests.  Measurements of tread depth
and tire weight were sometimes made, in addition to particle
collection, in order to arrive at appropriate weight loss conver-
sion factors when comparing laboratory and outdoor road tests.

     The main advantages of indoor testing are that test condi-
tions are easily controllable, particle collection is simple, and
particle size determinations are possible.  The main disadvantage
associated with this type of testing is that the tire is in
                               8

-------
contact with the same surface during the entire test, which can
lead to elevated tire temperatures.  Also, wear from directional
control is sometimes neglected, different tire forces exist on
the roller than on a pavement, and the steel roller produces a
"sticky" tire appearance quite unlike the appearance of a tire
driven on pavement.

     Two of the indoor testers also performed outdoor vehicle
tests for direct comparison.2'9  Three other sources of original
outdoor vehicle test data are also available.7'14'15  Two of the
five testers used single vehicles and three used fleets for
determination of average wear rates.  Some of the outdoor tests
also employed trailer testing systems but, as stated earlier,
these data are not considered.

     The source of the best test data is believed to be average
fleet values.  In three tests, the vehicles were tested under
controlled conditions of tread composition, inflation pressure,
speed, loading, road surface, and tire rotation order.  One study
used owner-operated vehicles.  These volunteers maintained the
tires in whatever fashion was typical for them, and thereby
provided actual user data.  Most of the tire wear values are
expressed as a wear index, or the ratio of the test tire to a
control tire times 100.  It was, therefore, necessary to have
knowledge of the wear rate of the control tire.  The tire wear
rates were obtained by tire tread depth measurements or tire
weight measurements.  Tests that measure wear rate by both meth-
ods provide conversion factors for emission factor calculations.
Several tests were conducted to determine tire wear as a function
of a particular factor such as temperature or road surface.

     As apparent in Table 1, the main advantages of outdoor
vehicle testing are that wear measurements are made under actual
road conditions and that random sampling is possible.  In the
case of fleet tests, conditions can be closely controlled and
accurate operating data are available.  The main disadvantages
associated with outdoor vehicle tests are the long testing periods
required,  the difficulty in controlling maintenance conditions
for random fleets, and the lack of particle size distribution
data.  In addition, if weight measurements are made, conditions
of tire inflation and humidity must be equalized before valid
measurements can be obtained.

     All of these limitations must be considered before, during,
and after tire wear testing.  There are two means of minimizing
the test disadvantages.  The first necessitates close control of
the following conditions:  tire inflation pressure and humidity,
vehicle speed, tire loading, surface texture, tire rotation
order, vehicle maintenance, length of testing, and tire forces.

-------
The second demands an awareness of important testing require-
ments:  wear rate measurements, particulate collection devices,
and tire tread composition.
                               10

-------
                            SECTION 3

               CALCULATION OF TIRE WEAR ESTIMATES


     Several units are commonly used to measure tire wear and
particulate emissions from tire wear.  The following units have
been used in one or more of the tests to determine tread depth
and weight loss:
     °    mil/mi; mil/1000 mi (1 mil = 0.001 inch)

     0    mg/km

     0    g/veh-mi; mg/mi; mg/1000 mi

     0    cm3/1000 mi

     0    cmVlOOO km
     The current particulate emission factor from AP-42 is in
g/veh-mi for a four-wheeled, light-duty vehicle.  This factor can
be easily used with activity data (vehicle miles traveled) to
compute an emission inventory.  Therefore, it is desirable to
express all test data in these units.  Example conversion calcu-
lations are shown here to demonstrate the process and the assump-
tions necessary for conversion.

     1.   mil/mi to g/veh-mi -

          (mi I/mi) (5 g/mil)a = 5 g/veh-mi per mil/mi

                             or

    la.   (mil/mi) (.001 in. /mil ) (29.3 in. )b(n ) (4.33 in.)c(59 lb/ft3 )'
          (ft3/1728 in.3) (454 g/lb)  =6.28 g/veh-mi per mil/mi

     2 .   mg/km to g/veh-mi  -

          (mg/km) (1.609 km/mi) (g/1000 mg)
          = 1.609xlO~3 g/veh-mi per  mg/km
   Reference 7
   Reference 2,  average measured tire diameter
   Reference 2,  average measured tread width
   Reference 19,  assumed density of rubber
                                11

-------
3.   cm3/1000 mi to g/veh-mi -

     (cmVlOOO mi) (59 Ib/ft3)d (454 g/lb)(ft3/28,360 cm3)
     = 0.946xlO~3 g/veh-mi per cm3/1000 mi000

4.   cm3/1000 km to g/veh-mi -

     (cm3/1000 km)(0.946x!03 g/veh-mi per cm3/1000 mi)
     (1.609 km^mi)
     = 1.52x10 3 g/veh-mi per cm3/1000 km

5.   wear index to g/veh-mi -

     some measure of tire wear for a control tire is required

     wear of test tire = fc"^?^1 tire)(100|(aPPro-

     priate conversion factor)

     These conversion factors were applied to data from eight
tire wear tests that used either an indoor test facility or
outdoor vehicle testing to arrive at the emission factors pre-
sented in Table 2.

     Examination of Table 2 reveals a range of emission factors
from 0.12 to 0.51 g/veh-mi.  These values represent total emis-
sions (including gaseous emissions) and therefore refer to weight
lost in grams per vehicle mile traveled (VMT).  The first factor
reported from Reference 2, C.409 g/veh-mi,  was derived from
measurements taken from 17 owner-operated vehicles that accumu-
lated a total of 84,590 vehicle miles.  Tread depth measurements
were made on each tire in 18 different locations around the tire,
and odometer readings were recorded.  An average annual tire
tread loss was calculated by weighting winter and summer wear
rates by their respective mileage accruals, and by converting to
g/veh-mi using emission calculation method la.  Four control
vehicles were also operated.  Mileage accruals on two of these
vehicles were 2960 and 5988 VMT and calculated wear rates were
0.260 and 0.340 g/veh-mi.  The second factor extracted from
Reference 2, 0.183 g/veh-mi, was determined through indoor testing
methods.  Two vehicles were operated on a dynamometer for a total
of 100 miles.  The mode of operation was chosen to simulate a
normal driving cycle.  Particle size distribution measurements
were conducted and results indicated that 50 percent of the emis-
sions are in the respirable particulate size range.  This dis-
tribution was then applied to the fleet value to obtain an esti-
mate of the respirable particulate emitted by the fleet vehicles.
Reference 2 and Table 1 indicate, however,  that several disadvan-
tages are associated with this testing facility.
                                12

-------
                              TABLE 2.   EMISSION RATE ESTIMATES FROM TIRE TREAD WEARC

                                                    (g/veh-mi)
Total
emissions
0.409
0.183
b
0.34

0.51


0.44


0.248

0.14

0.12



Emission
range
0.360-0.794
0.178-0.185
b
0.27-0.40

0.22-0.87


0.40-0.50


0.161-0.342

0.075-0.200

0.09-0.18



Airborne
participate,
<30(j
b
b
0.008
b

b


b


b

0.003-0.008

b



Respirable
particulate,
<3.5u
0.204C
50%
b
b

b


b


b

b

b



Test type
Outdoor fleet
Indoor composite
Typical indoor
Outdoor fleet,
control car
Outdoor fleet,
cornering force
tests
Outdoor, as a
function of
tire temperature
Outdoor fleet,
in two states
Indoor,
combination
Outdoor fleet,
as a function
of tire temper-
ature
Ref.
No.
2
2
4
7

7


7


9

11

14



u>
     ?  Emission  rates are  for a four wheeled light duty vehicle.
     °  No data.
        Estimated value based on indoor chassis dynamometer tests.

-------
     The most representative emission estimates of total tire
wear are those presented in Reference 7.   This study was under-
taken to determine the effect of the more important elements of
tire wear, these being tire force,  pavement texture, and tire
surface temperature.  Tire force and pavement texture measure-
ments were made under accelerated conditions and most values
reported are the results of trailer testing.  Control vehicles
were also operated for the tire force studies; tire force was
shown to be the major factor of the three elements studied.   All
wear rates were determined by measurements of the tread depth
loss and by conversion to g/veh-mi using emission calculation
method 1.  The first value obtained from Reference 7, 0.34 g/veh-mi,
is determined to be the average wear rate of the control fleet.
The wear rates obtained from the control fleet were also compared
with the trailer wear rates.  Correlation coefficients showed
good correlation between fleet tests and low-severity trailer
tests (r = 0.95 at 150 Ib cornering force), and equally good
correlation at high-severity trailer tests (r = 0.97 at 400 Ib
cornering force).  The average value of these cornering force
tests is 0.51 g/veh-mi, with a range of 0.22 to 0.87, as shown in
Table 2.  This value, however, was obtained in accelerated tests
that may have produced abnormal wear.  The last factor extracted
from Reference /, 0.44 g/veh-mi, was determined as a function of
tire temperature studies on fleet vehicles.  The wear rates
showed good linerarity, when plotted against tire surface tem-
perature, suggesting an accurate determination of tire surface
temperature and tread loss.  The wear rate determined as a func-
tion of pavement texture is not shown in Table 2.  This study was
done with trailer tests under accelerated conditions, and also
considered slip angle as one of the study elements.  Because slip
angle on trailers is produced in a manner different than slip
angle produced under actual vehicle operation, the values obtained
have been omitted from the emission factor determinations.  The
average value obtained from these tests,  however, agreed well
with the others reported and was determined to be 0.36 g/veh-mi.

     Over 100,000 vehicle miles of testing were performed to
arrive at the emission factors presented in Reference 9.  The
average of all values is 0.248 g/veh-mi.  Fleet testing was
performed in Texas and California by two testing companies and in
the laboratory.  These tests were undertaken to determine the
relationship between specific abrasion of different  tire tread
stocks and the equivalency of road and laboratory testing methods.
Laboratory abrasion was performed using a 36-mesh abrasive wheel,
with 120-mesh carborundum grit flow rate adjusted to the rate  of
abrasion.  Tire wear measurements were made by the  tread depth
loss and tire weight loss methods.  Values were reported as
cm3/1000 mi; conversion to g/veh-mi was performed using emission
calculation method 3.  Reference 9 presents six different wear
rates representing tests of three different tread stocks in  the
two states.  In some tests, the same tires were used in both
                                14

-------
 states.   The average of all  reported values  is  0.248  g/veh-mi,
 with a range of 0.161 to 0.342 as shown in Table  2.   It was  found
 that the laboratory abrader  conditions  exaggerated losses  at
 lower severities for some tread stocks,  but  showed reasonably
 good agreement with road abrasion indexes  for others.   The dis-
 crepancy at lower severities is believed to  result from rubber
 oxidation effects.   Some differences were  also  observed between
 the results obtained from the two states,  and these differences
 were believed to be valid because of the surface  texture of
 different pavements on the two test routes.

      The value reported in Reference 11, 0.14 g/veh-mi,  was
 obtained at an indoor test facility using  a  steel roller appara-
 tus similar to the  dynamometer discussed in  Reference  2.   Five
 testing  modes were  used;  the value reported  is  the result  of a
 combination of all  of these  test modes.  The testing was per-
 formed over 150 tire miles;  all wear products were collected and
 analyzed for sedimentary and airborne particulate,  as  well as
 gaseous  products.   Material  collection  efficiency was  checked
 periodically by weighing the tire before and after each test but,
 once a material balance was  made,  no further weight loss measure-
 ments were performed.   Measurement of airborne  particulate revealed
 that only 1 to 10 percent of the total  wear  products occurred as
 airborne particulate;  the majority were sedimentary.   Further-
 more,  it was observed that this percentage is dependent on tire
 wear rate,  with higher percentages of airborne  particulate pro-
 duced at lower wear severities.   These  observations conflict with
 those reported in Reference  2,  where it was  concluded  that 50
 percent  of all emissions  from tire wear  are  in  the respirable
 particulate size range.   The upper value of  airborne particulate
 emissions  in Reference 2  is  comparable  to that  in Reference  4,
 which reported a typical  value of 0.008  g/veh-mi  using an  indoor
 testing  facility.

      The tests reported in Reference 14 were  performed to  deter-
 mine the effect of  tire surface temperature  on  the  amount  of
 tread wear.   Trailers  and a  fleet  of vehicles were used  for  these
 experiments.   All wear measurements  were made by  weighing  the
 tires  every 200  miles,  which also  provided for  a  regular tire
 rotation scheme.  The  values  are  reported as  wear index  (as  a
 function of temperature),  with the  tire wear  of the control
 vehicle  expressed in cm3/lOOO  km.  Emission calculation methods 4
 and  5 were  used  for conversion to  g/veh-mi.   In order  to calcu-
 late  an  emission factor from  the data presented,  it was necessary
 to use distribution plots  of  tire  surface temperatures.  Wear
 indexes  were determined for  12  different temperature ranges,  each
 5°C, either long distance or  commuter travel.  These indexes were
multiplied by  the percent of vehicle operation time when the tire
 surface  temperature was in the  specified temperature range.  This
yielded  a wear rate for long distance and commuter travel.   A
weighted average of these two values yielded the  emission  factor,
                                15

-------
0.12 g/veh-mi,  which appears in Table 2.   The calculation process
was cumbersome and reflects temperature variation readings taken
from one test vehicle over a period of 13 months.  Other inves-
tigators have commented on the difficulty of measuring tire
surface temperature and, for that reason, it appears that these
calculations are subject to a large degree of error.

     The eight emission factors presented in Table 2,  each the
result of separate testing, were ranked according to test approach
and sample size.  These values are listed in Table 3 with a brief
rationale for the ranking decision; they are listed in a descend-
ing order of apparent accuracy.

     The average of the eight emission factors is 0.30 g/veh-mi.
As mentioned several times previously, however, indoor tests
produce wear on the test tire that is quite different from wear
produced in normal operation.  Removing the indoor values yields
an average emission factor of 0.34 g/veh-mi.  This value repre-
sents a more accurate estimate of emissions from tire wear,
although the factors ranked 5 and 6 also have inaccuracies asso-
ciated either with the testing or with the calculations used.
Therefore, these t^o factors are also omitted from calculation of
the average value.  This yields an average of 0.36 g/veh-mi.
Thus, the most representative emission factor for tire wear is in
the range of 0.30 to 0.36 g/veh-mi, depending upon which test
results are included in the determination.

     It is more desirable to choose an emission  factor from the
literature that best represents the investigation efforts and
conclusions reached by all research studies, rather than to adopt
a value that represents an average of all work.  This provides a
value determined to be most representative  for future use.
Reviewing the list of emission factors presented in Table 3, it
is apparent that there is one value within  the range determined
to best represent all work done in this  area.  That value is 0.34
g/veh-mi.  This is also the value that is ranked highest in
accuracy, as determined by testing methods, reproducibility, and
data base.

MECHANICS OF TIRE WEAR DEPOSITION

     The preceding discussion  suggests the  value of 0.34  g/veh-mi
for use in estimating tire wear emissions.  Because this  emission
factor  is based on tread  depth loss  during  outdoor  vehicle  testing,
it represents total  tread loss  (converted to weight loss)  per
mile of operation.   It  then  is necessary to discover where  this
material  is  lost.  Many studies have  been undertaken  to  make  this
assessment.

     Figure  2 presents  a  schematic of possible methods  of weight
loss and  particulate emission  during the tire  wear  process.   As
                                 16

-------
                                   TABLE 3.   RANK OF EMISSION FACTORS
Rank
 Factor,
g/veh-mi
                                                        Reason
                                                               Reference
  7

  8
                 0.34
                 0.44
                 0.248
                 0.409
                 0.12
  0.51



  0.183

  0.14
This value  is the average wear  rate of a  fleet.  All
other tests performed  in this study support this
value.

From the same reference as the  above value, this
factor  is the average  value of  wear as a  function
of temperature tests.  Temperature appears to be a
major factor in the wear process.

This value  is the average of fleet tests  performed
in two  states.  Over 100,000 miles of testing
were driven.

This value  is the average wear  rate of a  randomly
selected fleet operated by vehicle owners in actual
service over a period  of 12 months.

Derived from wear as a function of temperature
studies, this value requires assumptions as to the
annual distribution of tire tread temperatures,
thereby allowing for analysis errors.

The result of cornering force tests,  this value was
arrived at in accelerated tests possibly with unusual
forces associated with testing.

These values were determined in indoor testing facil-
ities and should be considered  less accurate than out-
door testing results.
                                                                                                   14
 2

11

-------
presented in this figure, there are four fractions of emissions
from tire wear that can account for the measured weight loss.
These fractions are E.  (tire wear directly transferred to the
road surface), E  (sedimentary tire wear particulate), E  (air-
borne tire wear particulate), and E  (gaseous tire wear product)
Very little work has been done to determine the amount of tire
wear products which are gaseous in nature, but Myslinski con-
cluded that less than 0.064 percent of the wear products are
volatilized.12  It can be concluded that the gaseous fraction is
an insignificant portion of the 0.34 g/veh-mi factor.
      Figure 2.  Mechanics of material loss and emissions from tire wear.

     The most significant portion of tire wear emissions occurs
as sedimentary particles and as particles transferred directly to
the road surface.  It is estimated that these fractions account
for up to 90 percent of the material lost during the tire wear
process.2'3'4'6'11'18  Just as tire skid marks seem to disappear
with time, so these fractions of tire particulates are removed
from the road surface.  This removal process occurs in one of
several ways.  Some of these particles are removed by the washing
effect of rainfall or by the action of a street  sweeper.  Dis-
placement of this material occurs when passing vehicles pick  it
up, either by the action of the tires or by the  vortex produced
as they pass.  This material is then displaced to the roadside
and is usually found in measurable concentrations, decreasing
with distance from the roadway up to 5 meters away.  Further
removal occurs as a result of oxidation processes on the rubber,
which degrades the material into gaseous products.4'11  Another
action of passing vehicles is the grinding of large particles
into smaller particles.  Some of these particles are then able to
become airborne, and make up the fraction labeled E  in Figure 2.
E  is the portion of reentrained particulate  from paved streets
tfiat is attributable to tire particulates.
                                 18

-------
      An analysis of the deposition processes  leading to  the
 accumulation of particulate matter on paved streets  suggests  that
 10 Ib of tire rubber are deposited on a typical  curb-mile  of  road
 each day.20   This amount constitutes  4.2 percent of  the  total
 particulate  loading on a typical  street, and  is  defined  in Figure
 2 as E.  and  E .   This study also  determined that approximately  50
 percent of this  deposited material later becomes reentrained
 (E ).   This  material then becomes airborne  and can be transported
 long distances before settling again.

      The final fraction that accounts for tire weight loss is E ,
 the fraction of  particulate from  tire wear  that  is initially    a
 airborne.  This  fraction is usually estimated from indoor  test
 results to be 10 percent of the total weight  loss.2'4'11'21   This
 fraction of  the  tire wear emissions,  together with E (the reen-
 trained portion),  is what eventually  reaches  high vofume sam-
 plers.

      Some  indirect quantification of  E  + E  is  possible in
 studies using high volume samplers to Determine  the  impact of
 tire rubber  on air quality readings.   One of  these studies col-
 lected  tire  rubber particulate on glass  fiber filters.  The tire
 rubber  accounted for 2.5  percent  of the  total reading near a
 roadway.18   These  results were duplicated in  another study of
 particulate  calculated  near roadways,  where 2.5  percent of the
 particulate  was  of vehicle tire origin.5  These  two  studies
 confirm the  conclusions reached in Reference  20:  4.2 percent of
 the  total particulate loading  on  streets is tire  rubber, and  50
 percent of this  material  is  eventually reentrained and becomes
 airborne.  This  can be  seen in the following  calculations:
            Tire rubber
   Reentrained material
Reentrained tire rubber
                                  4.2% of total street loading
                            =     50% of deposited material
                                  (0.50M4.2) = 2.1% of total
                                  reentrained material

     This 2.1 represents the percent of reentrained particulate
collected at high volume samplers near roadways that is of tire
rubber origin.

     The initially airborne fraction of particulate is also
collected on samplers.  In order to calculate the total fraction
of particulate contributed by tire wear,  the following equation
can be applied:
Tire wear fraction =  2.1 +
                              Etotal -Ea)(0.50)
                                19

-------
         where 2.1 = percent of total reentrained material con-
                     tributed by tire wear

            Etotal = total tire weight loss (100%)

                E  = percent of E.  . ,, that is initially airborne
                 a   (10%)       totai

              0.50 = fraction of deposited material that becomes
                     reentrained

     Substituting these values yields a total of 2.6 percent of
the high volume sample contributed by tire wear particulate.
This agrees closely with References 5 and 18.
                                20

-------
                            SECTION 4

                     SUMMARY AND CONCLUSIONS
     It has been stated that tire wear is the result of fric-
tipnal forces imparted by the tire to the road in order to main-
tain control of a moving vehicle.  The wear process is a function
of external and internal tire forces, as well as speed, load, and
environmental conditions.  Cornering forces, pavement type, and
temperature have all been studied as independent variables af-
fecting the rate of tire wear and these studies have produced
comparable emission factors.

     The identification of particulate emissions from tire wear
is very complicated.  As discussed in this report, several meth-
ods are currently used to evaluate the weight lost from tires
during operation.  It has been determined that the most accurate
emission factor is 0.34 g/veh-mi, representing total weight loss.
The airborne fraction of this factor is 10 percent, or 0.034
g/veh-mi, indicating that 0.306 g/veh-mi are either directly
transferred to the roadway or emitted as sedimentary particulate.
If it is then assumed that 50 percent of this amount is reen-
trained, a total of 0.153 g/veh-mi is added to that fraction
which is initially airborne.  Summing the airborne and reentrained
fractions yields an overall emission factor for airborne partic-
ulate from tire wear of 0.19 g/veh-mi.  The following is a sum-
mary of particulate emissions from a four-wheeled, light-duty
vehicle.
Fraction
Total weight loss
Initially airborne particulate
Sedimentary and direct transfer to
road surface particulate
Reentrained particulate
Total airborne particulate
Percent of total
100
10
90
0.5(90) = 45
10 + 45 = 55
Emission factor,
g/veh-mi
0.34
0.03
0.31
0.16
0.19
                                21

-------
                           REFERENCES


1.   Marchesani, V. J.,  T. Towers,  and H.  C.  Wahlers.   Minor
     Sources of Air Pollutant Emissions.   J.  Air Poll.  Control
     Assoc.  20(l):19-22.  1970.

2.   Subramani, J. P.  Particulate Air Pollution from Automobile
     Tire Tread Wear.  Doctoral dissertation.  University of
     Cincinnati, Cincinnati,  Ohio.   1971.

3.   Dannis, J. L.  Rubber Dust from the Normal Wear of Tires.
     Rubber Chem. Technol.  47:1011-1037.   1974.

4.   Cadle, S. H., and R. L.  Williams.  Gas and Particle Emis-
     sions from Automobile Tires  in Laboratory and Field Studies.
     J. Air Poll. Control Assoc.   28(5):502-507.  1978.

5.   Cardina, J. A.  Particle Size Determination of Tire-Tread
     Rubber in Atmospheric Dusts.   Rubber Chem. Technol.  47:1005-
     1010.  1974.

6.   Pierson, W. R.,  and W. W. Brachaczek.  Airborne Particulate
     Debris from Rubber Tires.  Rubber Chem.  Technol.   47(5);1275-
     1299.  1974.

7.   Veith, A. G.  Accelerated Tire Wear under Controlled Con-
     ditions.  I:  Description of the Test System.  II:  Some
     Factors that Influence Tire  Wear.  Rubber Chem. Technol.
     46:801-841.  1973.

8.   Gough, V. E.  Tyre-to-Ground Contact Stresses.  Wear.
     2:107-126.  1958-1959.

9.   Davison, S., et al.  Laboratory Testing of Tread Stock-
     Abrasion Resistance.  Rubber World.   151:81-92.  1965.

10.  Whitby, G. S.  Synthetic Rubber.  New York, New York, John
     Wiley & Sons, Inc.  and London, Chapman and Hall,  Limited,
     1954.

11.  Williams, R.(L., and S.  H. Cadle.  Characterization of Tire
     Emissions Using an Indoor Testing Facility.  Rubber Chem.
     Technol.  51:7-25.   1978.

12.  Myslinski, T. E.  Exploration of Major Gaseous Products  from
     Rubber Tire Tread Wear.   Master's thesis.  University of
     Cincinnati, Cincinnati,  Ohio.  1971.

13.  Schallamach, A.  Recent Advances in Knowledge  of Rubber
     Friction and Tire Wear.   Rubber Chem. Technol.  41:209-244.
     1968.
                               22

-------
14.
15.
16.



17.



18.


19.



20.




21.
Grosch, K. A.  The Effect of Tyre Surface Temperature on the
Wear Rating of Tread Compounds.  Journal of Institute of
Rubber Industry.  1:35-39.  1967.

Cahill, T. A.  Contribution of Freeway Traffic to Airborne
Particulate Matter.  University of California, Davis, Cali-
fornia.  Prepared for California Air Resources Board.
Contract ARB-502, PB-231 617.  April 1974.

Standard Test Method for Rubber Property-Abrasion Resistance
(with the Pico Abrader).  ASTM Designation D 2228-69 (Re-
places D 2228-63T).

Brachaczek, W.,  and W. R. Pierson.  Analytical Method for
Measuring SBR Tire Debris in the Environment.  Rubber Chem.
Technol.  47:150-160.  1974.

Cardina, J. A.  The Determination of Rubber in Atmospheric
Dusts.  Rubber Chem. Technol.  46:232-241.  1977.

Perry, R. H., C. H. Chilton, and S.  D. Kirkpatrick.  Chem-
ical Engineers'  Handbook.  Fourth Edition.  New York, New
York, McGraw Hill Book Company, 1963.

Control of Reentrained Dust from Paved Streets.  U.S. Envi-
ronmental Protection Agency, Kansas  City,  Missouri.  Publi-
cation Number EPA 907/9-77-007.  August 1977.

Personal communication with Professor T. Cahill,  University
of California, Davis, California.  January 12, 1979.
                               23

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
 REPORT NO.
EPA-450/4-79-011
                            2.
TITLE AND SUBTITLE
Evaluation  of  Particulate Emission  Factor* for Vehicle
Tire Wear
            5. REPORT DATE

              June 1979
            6. PERFORMING ORGANIZATION CODE
                                                           !. RECIPIENT'S ACCESSION NO.
 AUTHOR(S)
Joseph  Corvitti
                                                          8. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING ORGANIZATION NAME AND ADDRESS
PEDCo Environmental, Inc.
2420 Pershing Road
Kansas  City,  MO  64108
                                                           10. PROGRAM ELEMENT NO.
            11. CONTRACT/GRANT NO.
               68-02-2585
               Task Number 9
2. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Monitoring and Data Analysis Division (MD-14)
Research  Triangle Park, NC  27711
             13. TYPE OF REPORT AND PERIOD COVERED

               Final
             14. SPONSORING AGENCY CODE
               200/04
5. SUPPLEMENTARY NOTES
EPA  Task Officer - Charles  C.  Masser
6. ABSTRACT
      Vehicle-related emissions  are a major contributor to urban particulate
 concentrations.  One of  the  vehicle-related emission sources is tire wear.   It is
 estimated that 3.2xl02 to  7.2xl08 kg of tire wear  debris are generated  each year
 in the United States.  Particles from tire wear  have been identified by microscopic
 analysis as a significant  component of samples taken from roadside  high volume
 particulate samplers.

      The currently available emission estimate for tire wear, presented in Compilation
 of Air Pollutant Emission  Factors (AP-42), is 0.20 g/veh-mi for a four-wheeled
 light duty vehicle.  This  emission factor is widely used in regional particulate
 analyses to estimate emissions  from tire wear; however, this factor needs to be
 reviewed, based on studies and  test data released  since its publication.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                                                                            COSATl Field/Group
 Air Pollution
 Particulate
 Emissions
 Emission Factor
  Emission Factor
  Particulate emissions
  Tire Wear
18. DISTRIBUTION STATEMENT
  Unlimited
19. SECURITY CLASS (ThisReport)
   Unclassified
                                                                          !1. NO. OF PAGES
28
                                              20 SECURITY CLASS (Thispage)

                                                 Unclassified
                                                                          22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE

-------