Color profile: Generic CMYK printer profile

Composite Default screen

117

Tire-reinforced earthfill. Part 3: Environmental

assessment
Vince O’Shaughnessy and Vinod K. Garga

Abstract: A reinforced earthfill was constructed using approximately 10 000 scrap tires. Water samples were collected
from a drainage system installed below the tire-reinforced earthfill and analyzed for chemical quality. Additional tests
on water quality were performed in laboratory test columns in which tire chips were embedded in sand or clay to
provide a conservative estimate of the potential release of toxic compounds. Field monitoring of the effluent indicated
that no significant adverse effects on groundwater quality had occurred over a period of 2 years. Laboratory batch tests
performed on tire chips embedded in sand provided evidence of an increase in solution of certain metal elements
which in some cases exceeds their respective drinking-water standards. This increase was attributed to the exposed steel
reinforcement found in the tire chips. The amount of organic compounds leached from the tire chips decreased with the
number of exposure periods or pore volumes flushed through the soil.

Key words: whole scrap tires, water quality, tire chips, field investigation, laboratory lysimeter tests.

Résumé : L’on a construit un remblai armé en utilisant environ 10 000 pneus usés. Des échantillons d’eau prélevés
dans un système de drainage installé sous le remblai armé avec des pneus ont été analysés du point de vue de la
qualité chimique. Des essais additionnels sur la qualité de l’eau ont été réalisés dans des colonnes d’essai en
laboratoire dans lesquelles des grenailles de pneus ont été enfouies dans le sable ou l’argile pour fournir une évaluation
sécuritaire de la décharge potentielle de composés toxiques. Les mesures dans l’effluent sur le terrain ont indiqué qu’il
n’y a eu aucun effet indésirable sur la qualité de l’eau sur une période de 2 ans. Des essais en laboratoire réalisés sur
des lots de grenailles de pneus enfouies dans le sable ont fourni l’évidence qu’il y a eu accroissement dans la solution
de certains éléments métalliques dont certains ont dépassés leur norme respective pour l’eau potable. Cet accroissement
a été attribué à l’armature exposée des grenailles de pneus. La quantité de composés chimiques lessivés dans les
grenailles de pneus a diminué avec le nombre de périodes d’exposition ou de volumes des pores injectés dans le sol.

Mots clés : pneus usés complets, qualité d’eau, grenailles de pneus, étude sur le terrain, essais lysimétriques en
laboratoire.

[Traduit par la Rédaction] O’Shaughnessy and Garga 131

Introduction test fill (Fig. 1). Complementary laboratory lysimeter testing

The disposal of scrap tires has become a major environ-
mental concern. It is estimated that the United States alone
discards approximately 250 million scrap tires annually of
which almost 80% are diverted into three major markets:
tire-derived fuel (TDF), civil engineering applications, and
products made from scrap rubber (Blumenthal 1998). If lay-
ers of tires, side by side, are filled with soil and tied together
to make a reinforcing mat, the resulting structure can be
used as a retaining wall or reinforced fill and therefore can
provide a practical alternative for the use of this waste mate-
rial. To investigate this potential, a test fill approximately
57 m × 17 m in plan, using 10 000 whole tires or tires with
one sidewall removed, embedded in sand and a cohesive
soil, was constructed (Garga and O’Shaughnessy 2000). Wa-
ter-quality sampling was undertaken on the effluent from the
Received March 10, 1998. Accepted June 17, 1999.
V. O’Shaughnessy1 and V.K. Garga. Department of Civil
Engineering, University of Ottawa, Ottawa, ON K1N 6N5,
Canada.
1 rubber by cross-linking the polymer chains within the rubber
Present address: Inter-Tech Engineering Inc., 251 Bank St.,
Suite 608, Ottawa, ON K2P 1X3, Canada.
on tire chips embedded in a laboratory-grade kaolin and
clean silica sand were also conducted.
One of the major concerns in using shredded or whole
tires in civil engineering applications is the potential for
harmful substances to leach from the tires. To date, only a
few reports specifically address this concern. Although data
from embankments constructed with tire chips are reported,
there are no published data on water quality from whole tires
or tires with one sidewall removed embedded in soil.
Tires are fabricated with vulcanized rubber that contains
reinforcing textile cords, high strength steel or fabric belts,
and a high strength steel wire reinforcing bead. The most
commonly used tire rubber is styrene–butadiene copolymer
(SBR) containing about 25% styrene. Other tire rubbers used
are natural rubber (cis-polyisoprene), synthetic cis-
polyisoprene, and cis-polybutadiene. A typical chemical
composition for tire rubber is provided in Table 1. Carbon
black is added to strengthen the rubber and increase abrasion
resistance. The extender oil used in fabrication is a mixture
of aromatic hydrocarbons that acts to soften the rubber and
increase its workability. Sulphur is employed to harden the
which prevents high-temperature deformation. The zinc ox-

Can. Geotech. J. 37: 117–131 (2000) © 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:36:55 PM
Color profile: Generic CMYK printer profile
Composite Default screen

118 Can. Geotech. J. Vol. 37, 2000

Fig. 1. Schematic diagram of the field effluent collection system and a typical cross section of the test embankment.

Table 1. Typical composition of tire rubber (Williams et al. 1990). States Environmental Protection Agency (U.S. EPA) while
background samples did not. Reported concentrations of so-
Compound Weight (kg) Weight (%) dium and alkalinity exceeded the Ontario drinking water ob-
Rubber polymer (SBR) 6.09 62.1 jectives (ODWO, Ontario Ministry of the Environment and
Carbon black 3.04 31 Energy 1994). The ODWO are more stringent: elements
Extender oil 0.19 1.9 such as barium, chromium, lead, selenium, copper, and alu-
Zinc oxide 0.19 1.9 minum all have lower acceptable concentrations limits when
Stearic acid 0.12 1.2 compared with U.S. EPA standards (Table 3).
Sulphur 0.11 1.1 Investigations by Humphrey et al. (1997) have revealed no
Accelerator 0.07 0.7 significant adverse effects (after 2.5 years) on the water
Total 9.81 9.99 quality from two tire chips fills placed above the groundwa-
ter table. Results of the chemical analysis indicated that tire
chips did not increase the concentration levels of inorganic
ide, stearic acid, and an organo-sulphur accelerator are used elements having a primary drinking water standard (RALs),
to aid in the vulcanization process and enhance the physical such as barium (Ba), cadmium (Cd), chromium (Cr), lead
properties of rubber (Williams et al. 1990). Antioxidant and (Pb), and selenium (Se). Some evidence suggested that iron
other additives are also added to prevent deterioration of the (Fe) and manganese (Mn), both having an aesthetic drinking
rubber complex. water standard, may under certain conditions exceed their
respective concentration limits. Any organic compounds
leached from the tire chips were below their detection limits.
Summary of previous investigations
Lerner et al. (1993) investigated the potential use of tire
The Minnesota Pollution Control Agency (MPCA) (Twin chips as a drainage bed for domestic septic tanks. Labora-
City Testing Corp. 1990) conducted laboratory leaching tests tory water quality tests, in which tire chips were soaked in
on tires chips subjected to different environmental condi- different environments (various pH and ionic strengths), in-
tions. These tire chips were not embedded in soil. The study dicated that typical rubber compounds such as zinc and
reported that barium, cadmium, chromium, lead, selenium, benzothiazoles were leached from the tire chips regardless of
and zinc are constituents of concern in acidic environments environmental condition. Water samples taken from a pilot
(pH 3.5 to 5), and certain types of hydrocarbons (polynuclear septic drainage bed revealed fewer contaminants. It was rec-
aromatic hydrocarbons, PAHs) may be released under basic ommended to soak the tires in open lagoons for a period of
conditions (pH 8.0). 6–12 months, possibly removing many potential toxic com-
Eldin and Senouci (1992) examined the potential environ- pounds. This would render them more acceptable for envi-
mental impact of a test embankment constructed using tire ronmental uses.
chips in Wisconsin by periodic analysis of water specimens Abernethy (1994) conducted toxicity tests on water sam-
collected from two lysimeters placed beneath the fill. The ples in which an automobile tire had been submerged in
chemical analyses of inorganic compounds only indicated 300 L of water. The “tire water” caused 100% mortality of
that chloride, iron, and manganese were occasionally in ex- rainbow trout fry, usually within 48 h. Lethality was not
cess of the regulatory allowable limits (RALs) of the United observed for three other test species (Daphnia magna,
© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:36:58 PM
Color profile: Generic CMYK printer profile
Composite Default screen
O’Shaughnessy and Garga
Ceriodaphnia, and fathead minnows). The contaminated
water samples were analyzed for 143 target compounds.
Only zinc was found, and its concentration was well below
the lethal level. Gas chromatography – mass spectrometry
(GC–MS) analyses were performed to identify nontarget
chemicals. Up to 62 organic compounds were detected, but
less than half were identified (a known compound). Bench-
top treatments, such as aeration, and the additions of acid,
base, an antioxidant, and a metal-chelating agent did not re-
duce the toxicity of the tire water. However, the addition of
activated carbon completely removed the lethal effects. The
chemical cause(s) of the toxicity to trout remained uncon-
firmed.
A more recent study by Abernethy et al. (1996) revealed
that tires placed in a tank of flowing water were nonlethal to
trout so long as the flow rate was greater than 1.5 L/min per
600 L of water volume. The rate of chemical release from
tires decreased with each subsequent submersion period and
was attributed to a continuous leaching process which de-
pleted the tires of leachable chemical substances. Chemical
release may also be inhibited by bacterial growth and any
compositional changes occurring at the surface of the tire.
The study also revealed that tires collected from an artificial
reef in Lake Erie were less toxic than scrap tires which were
never exposed to an aquatic environment. It is of interest to
note that the amount of chemical release from the reef tires
was much less than that from recently discarded tires. A re-
duction in contaminant levels in water surrounding an
aquatic tire structure may occur by such environmental pro-
cesses as biodegradation, photolysis, and particle binding.
Reported exothermic reactions in tire shred fills
One important environmental concern with the use of rub-
ber products as fills is the potential for exothermic reaction.
An exothermic reaction is one where heat is released as a re-
sult of chemical or biochemical reaction. In 1995, there were
three cases of tire chips fills in the United States which were
reported to be smouldering after construction (Humphrey
1996). The three projects are SR 100 in Ilwaco, Washington,
Falling Springs Road fill in Garfield County, Washington,
and a retaining-wall project alongside Highway I-70 in Col-
orado. Importantly, in all three cases, the fills were con-
structed entirely of tire chips, where a high proportion of
steel belt was exposed in a matrix where crumb rubber was
sometimes encountered. The upper surfaces of these fills
were provided with a course of soil cover. Hot spots have
also been reported in four stockpiles of tire chips in the
United States.
For ignition to occur, it is essential that an initial exother-
mic reaction occurs which can raise the temperature above
the ignition point. According to Humphrey (1996), the po-
tential causes of initial exothermic reaction are oxidation of
exposed steel belts and wires, oxidation of rubber, microbes
consuming exposed steel wires or generating acidic condi-
tions, and microbes consuming liquid petroleum products.
Humphrey found that in all three cases where an exothermic
reaction was observed, appropriate conditions for oxidation
existed. In some cases, it was the lack of adequate cover
which minimized availability of free oxygen, while in other
cases an exothermic reaction occurred in areas exposed to
fertilizer-rich soil or where crumb rubber was found. Guide-
I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:36:58 PM
119
lines are now available to minimize exothermal reactions in
tire shred fills (Whetten et al. 1997).
It should be noted that the University of Ottawa prototype
test fills and retaining walls have been constructed in a
markedly different manner from the cases with tire chips
where an exothermic reaction has occurred. The use of
whole tires, or tires with one sidewall removed buried in a
mass of compacted cohesive soil or sand, significantly re-
duces, if not eliminates, the supply of free oxygen. Impor-
tantly, no crumb rubber is generated in the slicing process.
The prototype embankment has been hydroseeded, with no
sign of any exothermic activity.
Testing program
Laboratory and field tests were carried out to identify any
potential toxic substances leaching from buried tires. The
purpose of the investigation was to gather information on the
short-term (with respect to the service life of these struc-
tures) chemical leaching trends as a function of exposure
time, soil type, and environmental conditions and evaluate in
which medium (if any) they are produced, promoted, or re-
moved.
Field monitoring
The tire-reinforced earthfill was composed of three inde-
pendent 10 m long sections, each employing a gravity retain-
ing wall structure and a reinforced slope. A detailed
description is provided by Garga and O’Shaughnessy (2000).
The three sections were as follows: section A comprised cut
tires (one sidewall removed) filled with sand, section B
whole tires filled with sand, and section C cut tires filled
with a cohesive soil. A 76 mm diameter perforated pipe was
installed in the sand drainage blanket, under each section, to
collect leachate samples as rain water percolates down
through the tire-reinforced fill. The drain pipe was wrapped
in a geotextile to prevent clogging and was graded to flow
towards the retaining-wall section. The effluent was col-
lected in three separate wells consisting of a vertical PVC
cylindrical container 150 mm in diameter and approximately
1 m in depth. The top of the well was fitted with a protective
cap that could be easily removed during sampling (Fig. 1).
Epoxy was used to assemble all pipes and fittings.
Water samples were collected, periodically, using a hand
pump. After each sampling period, all three wells were
bailed dry, and therefore samples collected in each well were
composed of water accumulated since the previous sampling
period. The water samples were stored in 1 L HDPE bottles
and refrigerated to minimize degradation. It should be noted
that the wells remained dry during the summer months, and
therefore most water samples could be collected only during
the spring and fall months.
Laboratory leaching tests
Laboratory testing of lysimeter columns with whole or
sliced tires is clearly impractical. The use of tire chips in
lysimeters provides conservative test results compared with
those for whole tires because of the larger surface area and,
in particular, the exposed metal reinforcement. Also, the
lysimeters permit a controlled testing environment, in
contrast to field scenarios. Tests can be conducted at vari-
© 2000 NRC Canada
Color profile: Generic CMYK printer profile
Composite Default screen

120 Can. Geotech. J. Vol. 37, 2000

Fig. 2. Schematic diagram of the laboratory lysimeter column tests.

ous hydraulic and chemical conditions. In particular, the

Fig. 3. Photograph of tire chips placed within the lysimeter lysimeter columns permit the simulation of seepage over
column. long periods of time, especially in clayey soils. Laboratory
testing consisted of 12 large lysimeter columns 150 mm in
diameter and 400 mm long that were filled with various
mixtures of tire chips and soil, as illustrated in Fig. 2. Soil
and tire chip mixtures were divided into two main groups:
sand and tire chips, and pure kaolinitic clay and tire chips.
Twelve tires collected from an automobile parts recycling
site were cut into 50 mm transverse sections with a band
saw and each of these sections was further cut into four sep-
arate pieces. Typical chip size was 50 mm by 50 mm. The
tire chip content varied typically from 25 to 30% by volume
of the lysimeter chamber (Fig. 3). An outline of the testing
program is given in Table 2. The tire chips were randomly
distributed and hand compacted with a tamping rod. The
weights of soil and tire chips used in each lysimeter are indi-
cated in Table 2.

Sand and tire chips

Four lysimeters filled with 100% quartz sand and tire
chips were exposed to three different initial environmental
conditions: acidic, neutral, and basic. These tests represent
cases where relatively stagnant groundwater (ponding of rain
or snowmelt in surface aquitards) is exposed to automobile
tires for long periods of time. The acidic condition was pro-
vided by a 0.1% solution of sulphuric acid (pH ≈ 3.5) to sim-
ulate the effects of acid rain. The neutral and basic conditions
were provided by distilled water (initial pH of 6.5) and a
0.05% solution of sodium hydroxide (pH ≈ 9.5), respec-
tively. Tire chips and sand were initially washed with dis-
tilled water, and then the tire chips and quartz sand were
flushed with the appropriate leaching medium. The water
was continually circulated by chemical feed pumps to assure
proper mixing. Continuous circulation of the effluent is con-
servative, since it represents batch test conditions and does
not represent boundary conditions in tire-reinforced struc-
tures constructed above the groundwater table. Water sam-
ples were collected in 1 L HDPE bottles at the termination
of the test and were refrigerated to minimize degradation of
the sample. The exposure time, referred to as a leaching cy-
cle, was varied from 90 to 180 days. The testing procedure
was repeated for each leaching cycle using the same tire
chips.
© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:37:59 PM
Color profile: Generic CMYK printer profile
Composite Default screen

O’Shaughnessy and Garga 121

Table 2. Outline of the lysimeter testing program.

Test Soil mass Leaching medium Tire chips Tire chips Initial leaching
series Lysimeter cell composition (kg) mass (kg) mass (kg) volume (%) medium
1 Sand + tire chips 7.87 1.98 2.44 30 Distilled water,
pH≈6.5
2 Sand + tire chips (sidewalls only) 8.43 2.12 1.87 25 Acidic, pH≈3.5
(H2SO4)
3 Sand + tire chips 7.86 1.97 2.47 30 Basic, pH≈9.5
(NaOH)
4 Sand + tire chips 7.87 1.98 2.43 30 Acidic, pH≈3.5
(H2SO4)
5 Kaolinite (w = 38.5%, control sample) 9.37 3.53 0.00 0 Distilled water,
pH≈6.5
6 Kaolinite + tire chips (w = 37%) 7.50 2.82 1.64 20 Distilled water,
pH≈6.5
7 Kaolinite (w = 38.5%, control sample) 9.37 3.53 0.00 0 Acidic, pH≈3.5
(H2SO4)
8 Kaolinite + tire chips (w = 37.8%) 7.02 2.65 2.01 25 Acidic, pH≈3.5
(H2SO4)
9 Kaolinite + tire chips (w = 37.8%) 7.03 2.63 2.05 25 Acidic, pH≈3.5
(H2SO4)
10 Kaolinite (w = 37.5%, control sample) 9.12 3.44 0.00 0 Basic, pH≈9.5
(NaOH)
11 Kaolinite + tire chips (w = 37.5%) 6.38 2.41 2.41 30 Basic, pH≈9.5
(NaOH)
12 Kaolinite + tire chips (w = 37.5%) 6.84 2.58 2.00 25 Basic, pH≈9.5
(NaOH)

Clay and tire chips
The eight remaining lysimeters are divided into three
groups (Table 2). Each group was composed of a standard
test with no tire chips and one or two columns filled with
clay and tire chips. Again, each group was subjected to the
different leaching mediums (acidic, neutral, basic). The tire
chips at known weight per volume of soil were randomly
placed within the lysimeter cylinder with a mixture of pure
kaolin clay (obtained from Fisher Company Ltd., Colorado).
The kaolin sample was fabricated by mixing dry clay with
distilled water. Flushing of the pore water (in the pore space
between the clay particles and tire chips which is occupied
by water) was achieved by placing a source reservoir over
the top of the soil – tire chip column and applying a pneu-
matic pressure of 200 kPa. The source reservoirs were filled
with the appropriate initial environmental solution and were
as follows: two (numbers in each group include the stan-
dard) with distilled water; three with a 0.1% solution of
sulphuric solution; and three with a 0.05% solution of so-
dium hydroxide. The flushed pore water was collected into
1 L graduated cylinders at the base of the lysimeter cell.
Samples were analyzed after passing of one pore volume
(total volume of voids in the soil – tire chip sample). If pos-
sible, a maximum of five pore volumes for each environmen-
tal cell were tested (five tire water samples were collected
from each environmental cell).
Chemical compounds of interest
Water samples from the field and those collected from
lysimeters were analyzed for selected inorganic elements to
compare them with the requirements of the Ontario drinking
water objectives. Based upon published environmental as-
sessments on tire chip fills, particular attention was given to
iron (Fe), manganese (Mn), and zinc (Zn). The inorganic
samples were acidified and filtered. Cation concentrations
were measured using a Thermo Jarwell Ash Atomic Scan in-
ductively coupled plasma – atomic emission spectrometer
(ICP–AES). Anion concentrations were measured using a
Waters 880 high-pressure liquid chromatograph with con-
ductivity detector. The organic chemical compounds leached
from scrap tires can be classified as arylamines or phenols
and alylated forms. Organic analysis of tire water performed
by Abernethy (1994) identified four compounds which were
found in all samples: aniline, 4-(1-methyl-1-phenylethyl)-
phenol, benzothiazole, and 4-(2-benzothiazolythio)-morpholine.
The most prevalent compounds were reported to be 2(3H)-
benzothiazolone and benzothiazole, which are the probable
stable end products of mercaptobenzothiazole (MBT)
formed by methylation, photolysis, and oxidation (Brownlee
et al. 1992), and 4-(2-benzothiazolythio)-morpholine, which
is a common compound used as a delayed-action accelerator
in rubber processing (Taylor and Son 1982). Therefore,
these constituents were selected as target compounds to es-
tablish any chemical leaching trends in this investigation. A
list of the chemical compounds and parameters of interest
for both field and laboratory tests is given in Table 3.
Results
Field monitoring
Regrettably, no independent control structure was built
due to the financial constraints. Hence continuous monitor-
ing of background concentration levels of the targeted ele-
ments could not be performed. However, samples of backfill
© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:00 PM
Color profile: Generic CMYK printer profile
Composite Default screen

122 Can. Geotech. J. Vol. 37, 2000

Table 3. A list of the chemical compounds of interest and drinking water regulations.
ODWO concentration U.S. EPA regulatory
Compound ODWO typea limits (mg/L) allowable limits (mg/L)
Inorganic
Alkalinity (as CaCO3) OG 500 na
Aluminum (Al) OG 0.1 0.2
Barium (Ba) MAC 1 2
Cadmium (Cd) MAC 0.005 0.005
Calcium (Ca) na na
Chromium (Cr) MAC 0.05 0.1
Chloride (C1) AO 250 250
Conductivity (as µS) na na
Copper (Cu) AO 1 1.3
Fluoride (F) MAC 1.5 4
Iron (Fe) AO 0.3 0.3
Lead (Pb) MAC 0.01 0.015
Magnesium (Mg) na na
Manganese (Mn) AO 0.05 0.05
pH AO 6.5–8.5 (no units)
Potassium (K) na na
Selenium (Se) MAC 0.01 0.05
Sodium (Na) AO 200 na
Sulphate (SO4) AO 500 250
Zinc (Zn) AO 5 5
Organic
Benzothiazole nl nl
(1,1-dimethylethyl)-2-methoxyphenol nl nl
2,5-dibutylthiophene nl nl
4-(2,2,4-trimethylpentyl)phenol nl nl
2(3H)-Benzothiazolone nl nl
4-(2-benzothiazolythio)morpholine nl nl
Note: na, not applicable; nl, parameter not listed.
a
MAC, maximum acceptable concentration; AO, aesthetic objective; OG, operational guidelines.

material were tested in the laboratory to measure the back-
ground concentrations which could potentially be leached.
The results are presented in Table 4. Several samples of both
soil types were taken from their respective stockpiles at vari-
ous times during the construction of the prototype embank-
ment, and 100 mL of distilled water was added to 50 g of
dry soil. The mixture was vigorously stirred for a period of
24 h. The supernatant was filtered and analyzed for the ap-
propriate chemical compounds. There are certain limitations
to this procedure: it cannot duplicate field environmental
conditions such as temperature and biological activity, and
dilution may result in certain compounds being below their
detection limits. Furthermore, the procedure does not moni-
tor the progressive leaching of constituents with time and
(or) dilution by infiltration. Nonetheless, results do indicate
that selenium, with concentrations ranging between 0.012
and 0.014 mg/L, was the only compound of interest which
exceeded the ODWO. All others were either below detection
limits or below their respective allowable concentration lim-
its.
Results of the chemical analysis performed on water sam-
ples collected from each well at the test fill site are given in
Tables 5–7. The well located to collect effluent from the co-
2
Concentrations above this level have been known to, or are suspected to, produce adverse health effects.
hesive soil section C was dry most of the time. Indeed, all
wells remained dry during the summer months. Also, the
presence of a final 2 m surcharge of clay fill on the tire-
reinforced earthfill acts as a hydraulic barrier and therefore
limits infiltration into the structure. The amount of seepage
through the test fill was not monitored. It was not the intent
of this investigation to monitor any seepage due to rainfall
on a continuous basis. Rather, testing was carried out to as-
certain if any toxic compounds were leached out. The results
indicate that barium (Ba), chromium (Cr), selenium (Se),
and fluoride (F) elements, having maximum acceptable con-
centration (MAC2) limits, were present in trace amounts or
were below their detection limits. The results were similar to
those reported in the background analysis (Table 4). The de-
tection limits of lead (Pb), chromium (Cr), and cadmium
(Cd) in section A of the test fill and lead and chromium in
section B were above their respective drinking water con-
centration limits for the water samples collected 3 months
after completion of the test embankment. Due to the lack of
a control structure, it is difficult to ascertain if this single
peak event originated from compounds having been leached
from the tires or from the fill, or from the construction site
itself, or if it is simply a testing error. However, the high

© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:01 PM
Color profile: Generic CMYK printer profile
Composite Default screen

O’Shaughnessy and Garga 123

Table 4. Background concentrations of the pore water for both sand and cohesive soil.
Inorganic elements and organic Detection limits Sand Cohesive soil
compounds (mg/L) (mg/L) (mg/L)
Barium (Ba) 0.001 0.021 0.009
Cadmium (Cd) 0.0005 <0.0005 <0.0005
Chromium (Cr) 0.004 <0.004 <0.004
Lead (Pb) 0.004 <0.004 <0.004
Selenium (Se) 0.01 0.012 0.014
Fluoride (F) 0.5 <0.5 <0.5
Manganese (Mn) 0.004 0.054 0.014
Iron (Fe) 0.003 <0.003 <0.003
Zinc (Zn) 0.04 <0.4 <0.4
Copper (Cu) 0.001 <0.001 <0.001
Sodium (Na) 0.02 10.5 13.6
Chloride (CI) 1.5 2.1
Sulphate (SO4) 14.3 18.8
pH (no units) nd nd
Aluminum (Al) 0.002 0.064 0.049
Calcium (Ca) 0.005 18.7 31.8
Magnesium (Mg) 0.002 3.2 15.7
Potassium (K) 0.02 3.3 4.1
Alkalinity (as CaCO3) nd nd
Conductivity (as µS) nd nd
Benzothiazole 0.001 zd zd
(1,1-dimethylethyl)-2-methoxyphenol 0.001 zd zd
2,5-dibutylthiophene 0.001 zd zd
4-(2,2,4-trimethylpentyl)phenol 0.001 zd zd
2(3H)-Benzothiazolone 0.001 zd zd
4-(2-benzothiazolythio)morpholine 0.001 zd zd
Note: nd, not determined; zd, zero detected.

concentrations of lead, cadmium, and chromium suggest that
these are anomalous results and are believed to be caused by
testing inaccuracies, since lead, chromium, and cadmium
were below their detection limits in all the other samples
tested. Also, investigations by Eldin and Senouci (1992) and
Humphrey et al. (1997) on the water quality from tire chip
fills placed above the groundwater table indicated that none
of the contaminants having a MAC designation exceeded
their regulatory limits. The concentrations of these elements
in solution were generally found to be below their respective
detection limits.
Inorganic elements having an aesthetic objective (AO), in-
cluding iron (Fe), manganese (Mn), zinc (Zn), copper (Cu),
sodium (Na), chloride (Cl), and sulphate (SO4), were all be-
low the applicable limits except for iron, manganese, and
aluminum concentrations for one sampling period 3 months
after the beginning of field monitoring in both sand sections.
These peak concentration levels are possibly due to testing
inaccuracies, as stated earlier. None of the elements sur-
passed the ODWO in the limited number of samples ob-
tained from the cohesive fill section.
Previous research shows that tire chips in road embank-
ments can account for the increase in iron, manganese, and
aluminum concentrations, exceeding the ODWO limits in
the surrounding groundwater (Eldin and Senouci 1992;
Humphrey et al. 1997). The increase of such metals is attrib-
uted to the large amounts of exposed steel reinforcements as
a result of tire shredding. In contrast, engineered fills using
whole or cut tires would severely limit the amount of
exposed steel reinforcement. Also, the tire rubber would
protect the mild steel reinforcement from corroding (AB-
Malek and Stevenson 1986). This is evident by the low con-
centration levels of metals (with the exception of observa-
tion in one collection period) which were similar to or below
the measured background levels for both sand and cohesive
fills. It is interesting to note that higher levels of manganese
and aluminum concentrations were released from both back-
fill soils used in the field experiment prior to being exposed
to tires (Table 4). Thus, for future investigations, it may be
critical to establish control wells to distinguish background
levels from those obtained from leachate from buried tire
structures. Evidence collected so far indicates that the
amount of inorganic elements capable of being leached out
of engineered soil structures using whole or cut tires will be
insignificant.
The concentration levels of the targeted organic compounds
from all field monitoring wells were below the test method
detection limits with the exception of one compound, 4-(2-
benzothiazolythio)morpholine, which was detected twice in
the well located in section B (whole tires in sand). However,
these concentrations were significantly below levels mea-
sured in the laboratory lysimeter tests. The presence of 4-(2-
benzothiazolythio)morpholine indicates that leaching of
some organic compounds from tires placed above the water
table may occur. The presence of this compound could have
resulted from the exposed tire face of the retaining wall

© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:02 PM
Thursday, March 09, 2000 1:38:03 PM
I:\cgj\Cgj37\Cgj02\T99-086.vp

Composite Default screen

Color profile: Generic CMYK printer profile
124
Table 5. Water-quality results of the effluent collected from section A.

Detection Regulatory Sampling date

Inorganic elements and organic limits limits, ODWO 1994– 1994– 1995– 1995– 1995– 1996– 1996– 1996–
compounds (mg/L) (mg/L) 09–16 11–20 06–08 10–03 11–06 05–04 06–17 08–14
Barium (Ba) 0.001 1 0.041 0.086 0.028 0.055 0.037 0.018 0.015 0.035
Cadmium (Cd) 0.0005 0.005 <0.0005 0.02 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005
Chromium (Cr) 0.004 0.05 nd 0.13 <0.004 <0.004 <0.004 <0.004 <0.004 <0.004
Lead (Pb) 0.004 0.01 <0.004 1.3 <0.004 <0.004 <0.004 <0.004 <0.004 <0.004
Selenium (Se) 0.01 0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.01 0.01
Fluoride (F) 0.5 1.5 0.177 nd 0.287 <0.5 <0.5 <0.5 <0.5 <0.5
Manganese (Mn) 0.004 0.04 <0.004 0.18 <0.004 <0.004 <0.004 <0.004 <0.004 <0.004
Iron (Fe) 0.003 0.3 <0.003 0.93 <0.003 <0.003 <0.003 0.05 0.044 0.067
Zinc (Zn) 0.04 5 0.013 0.44 0.002 0.086 <0.04 0.13 0.14 0.14
Copper (Cu) 0.001 1 0.003 0.34 0.007 0.11 0.052 0.003 <0.001 <0.001
Sodium (Na) 0.02 200 13 24 8.4 7.8 6.6 5 1.4 <0.02
Chloride (Cl) 250 3.1 nd 1.8 nr nr 9.3 1.9 3.3
Sulphate (SO4) 500 60.9 nd 16.3 nr nr 4.8 4.5 11.2
pH (no units) 6.5–8.5 7.1 6.7 6.3 6.8 7.1 7.4 7.2 7.2
Aluminum (Al) 0.002 0.1 0.007 0.76 0.002 0.021 0.03 0.005 0.063 0.025
Calcium (Ca) 0.005 na 87 82 53 81 62 53 53 63
Magnesium (Mg) 0.002 na 18 16 18 20 15 10 9 11
Potassium (K) 0.02 na 2.4 2.7 1.9 2.9 1.1 7 7.2 8.9
Alkalinity (as CaCO3) 500 246 323 246 340 262 154 354 92
Conductivity (as µS) na 460 600 480 626 340 258 340 390
Benzothiazole 0.001 nl zd zd zd zd zd zd zd zd
(1,1-dimethylethyl)-2-methoxyphenol 0.001 nl zd zd zd zd zd zd zd zd
2,5-dibutylthiophene 0.001 nl zd zd zd zd zd zd zd zd
4-(2,2,4-trimethylpentyl)phenol 0.001 nl zd zd zd zd zd zd zd zd
2(3H)-Benzothiazolone 0.001 nl zd zd zd zd zd zd zd zd
4-(2-benzothiazolythio)morpholine 0.001 nl zd zd zd zd zd zd zd zd

Can. Geotech. J. Vol. 37, 2000

Note: Concentrations are in mg/L unless stated otherwise. na, not applicable; nd, not determined; nl, parameter not listed; nr, not reported; zd, zero detected.
© 2000 NRC Canada
Thursday, March 09, 2000 1:38:05 PM
I:\cgj\Cgj37\Cgj02\T99-086.vp

Composite Default screen

Color profile: Generic CMYK printer profile
O’Shaughnessy and Garga
Table 6. Water-quality results of the effluent collected from section B.

Detection Regulatory Sampling date

Inorganic elements and organic limits limits, ODWO 1994– 1995– 1995– 1995– 1995– 1996– 1996– 1996–
compounds (mg/L) (mg/L) 09–16 11–20 06–08 10–03 11–06 05–04 06–17a 08–14
Barium (Ba) 0.001 1 0.019 0.057 0.022 0.032 0.023 0.017 0.021
Cadmium (Cd) 0.0005 0.005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005 <0.0005
Chromium (Cr) 0.004 0.05 nd 0.012 <0.004 <0.004 <0.004 <0.004 <0.004
Lead (Pb) 0.004 0.01 <0.004 0.022 <0.004 <0.004 <0.004 <0.004 <0.004
Selenium (Se) 0.01 0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.019
Fluoride (F) 0.5 1.5 0.84 nd 0.287 <0.5 <0.5 0.57 <0.5
Manganese (Mn) 0.004 0.05 0.004 0.078 <0.004 <0.004 <0.004 <0.004 <0.004
Iron (Fe) 0.003 0.3 <0.003 2.7 <0.003 <0.003 <0.003 <0.003 0.037
Zinc (Zn) 0.04 5 0.008 0.15 <0.04 <0.04 <0.04 0.036 0.19
Copper (Cu) 0.001 1 0.003 0.11 0.005 0.012 0.023 0.014 0.038
Sodium (Na) 0.02 200 17 32 26 23 17 15 13
Chloride (Cl) 250 3.2 nd 2.5 nr nr 5.5 3.1
Sulphate (SO4) 500 100 nd 94.5 nr nr 22.1 13.9
pH (no units) 6.5–8.5 6.8 7.4 6.8 7.2 7.4 7.6 7.4
Aluminum (Al) 0.002 0.1 0.019 1.6 0.002 0.017 0.026 0.013 0.015
Calcium (Ca) 0.005 na 63 75 52 79 60 79 81
Magnesium (Mg) 0.002 na 24 36 45 35 28 34 34
Potassium (K) 0.02 na 1.9 3.7 2.2 2.9 1.3 7 7
Alkalinity (as CaCO3) 500 184 308 354 400 08 262 400
Conductivity (as µS) na 450 740 775 711 498 758 588
Benzothiazole 0.001 nl zd zd zd zd zd zd zd
(1,1-dimethylethyl)-2-methoxyphenol 0.001 nl zd zd zd zd zd zd zd
2,5-dibutylthiophene 0.001 nl zd zd zd zd zd zd zd
4-(2,2,4-trimethylpentyl)phenol 0.001 nl zd zd zd zd zd zd zd
2(3H)-Benzothiazolone 0.001 nl zd zd zd zd zd zd zd
4-(2-benzothiazolythio)morpholine 0.001 nl zd zd zd 0.017 zd 0.002 zd
Note: Concentrations are in mg/L unless stated otherwise. Abbreviations as in Table 5.
a
Dry well.
© 2000 NRC Canada

125
Thursday, March 09, 2000 1:38:07 PM
I:\cgj\Cgj37\Cgj02\T99-086.vp

Composite Default screen

Color profile: Generic CMYK printer profile
126
Table 7. Water-quality results of the effluent collected from section C.
Sampling date
Detection Regulatory
Inorganic elements and organic limits limits, ODWO 1994– 1994– 1995– 1995– 1995– 1996–05– 1996– 1996–
compounds (mg/L) (mg/L) 09–16 11–20a 06–08 10–03a 11–06a 04 06–17a 08–14a
Barium (Ba) 0.001 1 0.031 0.086 0.024
Cadmium (Cd) 0.0005 0.005 <0.0005 <0.0005 <0.0005
Chromium (Cr) 0.004 0.05 nd <0.004 <0.004
Lead (Pb) 0.004 0.01 <0.004 <0.004 <0.004
Selenium (Se) 0.01 0.01 <0.01 <0.01 <0.01
Fluoride (F) 0.5 1.5 0.64 nd 0.59
Manganese (Mn) 0.004 0.05 >0.2 0.036 <0.004
Iron (Fe) 0.003 0.3 <0.003 <0.003 <0.003
Zinc (Zn) 0.04 5 0.038 <0.04 0.045
Copper (Cu) 0.001 1 0.004 0.005 0.016
Sodium (Na) 0.02 200 33 32 31
Chloride (Cl) 250 22.6 nd 8.4
Sulphate (SO4) 500 106 nd 47
pH (no units) 6.5–8.5 6.6 6 7.6
Aluminum (Al) 0.002 0.1 0.007 0.002 0.002
Calcium (Ca) 0.005 na 78 71 80
Magnesium (Mg) 0.002 na 24 36 35
Potassium (K) 0.02 na 2.3 6.3 7
Alkalinity (as CaCO3) 500 292 231 310
Conductivity (as µS) na 550 220 540
Benzothiazole 0.001 nl zd zd zd
(1,1-dimethylethyl)-2-methoxyphenol 0.001 nl zd zd zd
2,5-dibutylthiophene 0.001 nl zd zd zd
4-(2,2,4-trimethylpentyl)phenol 0.001 nl zd zd zd
2(3H)-Benzothiazolone 0.001 nl zd zd zd
4-(2-benzothiazolythio)morpholine 0.001 nl zd zd zd

Can. Geotech. J. Vol. 37, 2000

Note: Concentrations are in mg/L unless stated otherwise. Abbreviations as in Table 5.
a
Dry well.
© 2000 NRC Canada
Color profile: Generic CMYK printer profile
Composite Default screen
O’Shaughnessy and Garga
Table 8. Chemical composition of Unimin sand.
Chemical compound Mean percent by weight
Silicon dioxide (SiO2) 99.710
Iron oxide (Fe2O3) 0.050
Aluminum oxide (Al2O3) 0.072
Calcium oxide (CaO) 0.010
Titanium oxide (TiO2) 0.006
Magnesium oxide (MgO) 0.019
Potassium oxide (K2O) 0.030
Sodium oxide (Na2O) 0.006
Loss on ignition 0.096
Fig. 4. The variation in iron (Fe) concentration as a function of
the leaching cycle and environment: quartz sand series.
which was subject to ultraviolet breakdown. The placement
of a protective facing, such as shotcrete or panel facing,
could eliminate this potential problem. No organic com-
pounds were detected in the other two wells. However, it is
possible that the measured levels of 4-(2-benzothiazoly-
thio)morpholine may also have originated from contami-
nated hydrocarbons at the construction site. The test
embankment was built on land currently being used for auto-
mobile parts recycling. Humphrey et al. (1997) also reported
that no measurable levels of organic compounds were found
in tire chip fills placed above the groundwater table. These
fills were monitored for a period of over 2 years. The impact
of leachable organics from buried tires on the environmental
is still unknown.
Laboratory tests on tire chips embedded in sand
The tire chips were embedded in a Unimin No. 2075 sand
manufactured by Unimin Canada Inc. This sand is composed
of 99.71% quartz (SiO2), which is considered inert (except
when exposed to hydrofluoric acid). Results of chemical
analysis on this sand are shown in Table 8. In view of these
results, the determination of leachate characteristics from
sand alone was not deemed to be necessary. Thus, the only
source of the targeted compounds is the tire chips them-
selves. All targeted elements having a MAC designation (Ba,
Cd, Cr, F, Pb, Se) were below their respective detection lim-
its or their concentration levels were insignificant. The varia-
tion in iron concentration with exposure time (note that each
I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:14 PM
127
Fig. 5. The variation in aluminum (Al) concentration as a func-
tion of the leaching cycle and environment: quartz sand series.
OG, operational guideline (Ontario drinking water objectives).
exposure time represents a single leaching cycle) for the
four different simulated environments and testing conditions
is shown in Fig. 4. The iron concentration generally ex-
ceeded the drinking standard limits. However, Fe levels fell
below the designated AO limit for all test environments after
a cumulated exposure time of 400 days. The decrease in sol-
uble iron concentration levels is associated with the forma-
tion of rust (Fe2O3) which precipitated out of solution. The
acidic environment promoted the oxidation of the exposed
steel elements, thus decreasing the available iron for each
subsequent leaching cycle. This may explain why soluble
iron was only found in the basic environment after the sec-
ond leaching cycle (130 days). It should be noted that in
each consecutive leaching cycle the leaching medium was
totally replaced, thereby removing all soluble elements in
that test. In addition, cutting of a whole tire by a band saw
instead of a shredder produced tire chips having a smaller
number of exposed steel reinforcement wires. A large
amount of rust aggregate, a mixture of rust and sand parti-
cles, was observed to have been formed around the exposed
steel reinforcing elements in the tire chips and seems to have
slowed the leaching process. This rust aggregate was re-
moved after the second leaching cycle and may explain the
increase in measurable Fe after the end of the third leaching
cycle.
A trend similar to that for iron concentration was also ob-
served for aluminum levels (Fig. 5). After the second leach-
ing cycle (an exposure time of 130 days), the tire chips were
cleaned using distilled water and the rust aggregates were re-
moved. This cleaning process exposed fresh steel reinforce-
ments which may have produced the increase in Al
concentrations in the neutral and acidic conditions for the
final batch test of 180 days. Manganese concentration levels
were generally above the acceptable concentration limits
except for tire chips placed in the basic environment
(Fig. 6), indicating that manganese is more readily leached
from the steel reinforcement under neutral or acidic condi-
tions. Unlike iron, manganese being more soluble did not
form a visible precipitate. The concentration levels of zinc in
all environments were below the AO designated limit. Zinc
was easier to leach out of the tire chips under a basic condition
© 2000 NRC Canada
Color profile: Generic CMYK printer profile
Composite Default screen
128
Fig. 6. The variation in manganese (Mn) concentration as a func-
tion of the leaching cycle and environment: quartz sand series.
Fig. 7. The variation in zinc (Zn) concentration as a function of
the leaching cycle and environment: quartz sand series.
(Fig. 7). The long-term potential for zinc in leaching out of
the tire rubber is a concern, since zinc oxide is used in the
vulcanization process to produce tire rubber and is responsi-
ble for approximately 2% of the total weight of the tire (Ta-
ble 1). Lysimeter test results indicate that it may be
beneficial to mix the tire chips and the soil together to form
a composite material. The presence of a soil within the tire
chip matrix may reduce leaching of inorganic elements.
The tire chips used in earth fills, under appropriate condi-
tions, would increase the concentration of iron, aluminum,
zinc, and manganese in the surrounding groundwater. This
finding is consistent with studies of water-quality monitoring
of tire chip embankments placed above the groundwater ta-
ble (Eldin and Senouci 1992; Humphrey et al. 1997). As in-
dicated in the reported field study, this increase is generally
associated with the high number of exposed steel reinforce-
ment wires. However, engineered soil structures using whole
tires would have a lower impact on the water quality, since
the steel reinforcement is not exposed and the rubber layer
would protect the mild steel reinforcement from corroding
(AB-Malek and Stevenson 1986). This analogy is consistent
with results obtained from the field monitoring program.
I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:23 PM
Can. Geotech. J. Vol. 37, 2000
Fig. 8. The variation in benzothiazole concentration as a function
of the leaching cycle and environment: quartz sand series.
Fig. 9. The variation in 2(3H)-benzothiazolone concentration as a
function of the leaching cycle and environment: quartz sand series.
Also, the investigations carried out in the present research
indicate that the rubber tended to neutralize the leaching me-
dium. The pH measurements after the completion of a leach-
ing cycle in the lysimeters indicated that both the initial
acidic medium and the basic medium were neutralized.
Benzothiazole concentration decreased with the number
of washes and exposure time except for test series 2 in
which chips obtained from tire sidewalls only were placed in
an acidic environment (Fig. 8). Minimum leaching of benzo-
thiazole occurred in the neutral environment. The basic envi-
ronment promoted the release of this organic compound;
however, there was a marked decrease in concentration after
the second leaching cycle. The 2(3H)-benzothiazolone and
phenol compounds disappeared after the second leaching cy-
cle, as shown in Figs. 9 and 10, respectively. Again, the
release of the phenol compound was promoted under basic
conditions. The 2(3H)-benzothiazolone was not leached for
the tire chips subjected to a neutral environment. 4-(2-Ben-
zothiazolythio)morpholine was the only target compound to
show a steady increase in concentration levels, in all four
test series, with increasing exposure time (Fig. 11). This
trend may indicate that 4-(2-benzothiazolythio)morpholine is
© 2000 NRC Canada
Color profile: Generic CMYK printer profile
Composite Default screen
O’Shaughnessy and Garga
Fig. 10. The variation in (1,1-diethylethyl)-2-methoxyphenol con-
centration as a function of the leaching cycle and environment:
quartz sand series.
Fig. 11. The variation in 4-(2-benzothiazolythio)-morpholine con-
centration as a function of the leaching cycle and environment:
quartz sand series.
a more stable end byproduct of the leaching process. This
may also explain why it was the only organic compound to
be detected during field monitoring.
The total amount of chemicals released from the tire
chips, independent of the leaching environment, decreased
with the number of leaching cycles and increased exposure
time, with the possible exception of 4-(2-benzothiazoly-
thio)morpholine compound. Abernethy et al. (1996) also re-
ported that the rate of chemical release decreased during
each tire submersion period. The decrease in the rate of
chemical release could be attributed to several factors. There
is only a limited amount of leachable material found at the
surface of the tire rubber. Leaching of the contaminants is
probably limited to the tire surface, since water (the leaching
medium) is essentially unable to penetrate the impervious
tire rubber, and therefore depletion of the available com-
pounds would occur only at the initial stages and over a
short time period. AB-Malek and Stevenson (1986) studied
the physical condition of vulcanized natural rubber sub-
I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:29 PM
129
Fig. 12. The variation in Al and Mn concentrations as a function
of leaching environments and pore-volume flushes: kaolin clay
series.
merged in 24 m of seawater for a period of 42 years. The
amount of water absorbed into the tire rubber was minimal,
only 4.7%. The limited amount of water adsorption was at-
tributed to the formation of a thin surface layer (0.05 mm) of
an iron base material. This thin layer may inhibit the release
of organic compounds. Also, the growth of bacteria on the
surface of submerged tires may also reduce the amount of
chemicals released (Arbernethy et al. 1996). Alternatively,
the chemical composition at the tire surface can be altered
over time which can modify the chemical release of organic
compounds.
Laboratory tests on tire chips embedded in clay
The chemical composition of the kaolin clay is given in
Table 9. All test series demonstrated that the inorganic target
elements were either below their respective concentration
limits or below the concentration levels found in the control
samples. For example, variations in Al concentrations,
leached under acidic conditions, and Mn concentrations,
leached under basic conditions, are provided in Fig. 12. The
variation in the organic compound concentrations for tire
chips embedded in the kaolin clay subject to a neutral envi-
ronment is shown in Fig. 12. The release of benzothiazole
and 2(3H)-benzothiazolone was minimal and fell to zero af-
ter two pore volumes of flushing. The 4-(2-
benzothiazolythio)morpholine had the highest peak concen-
tration after two pore volumes but disappeared after three
pore volumes. The phenol compound showed an increase in
concentration with an increase in the number of pore vol-
umes of flushing. However, the low concentrations of tar-
geted compounds make it difficult to determine if the
variation in concentrations with each pore volume passed is
a true leaching pattern or an artifact of testing variability.
The amount of chemical release was much less than that
associated with tire chips embedded in sand. The chemical-
release trends for tire chips embedded in clay under neutral,
acidic, and basic conditions are presented in Figs. 13, 14,
and 15, respectively. In all tests, a consistent chemical trend
was observed. The concentration of the target compound
would reach a maximum level and then would begin to
© 2000 NRC Canada
Color profile: Generic CMYK printer profile
Composite Default screen

130 Can. Geotech. J. Vol. 37, 2000

Table 9. Fisher kaolinite components. Fig. 14. The variation in organic compound concentrations
leached from tire chips embedded in a kaolin clay subjected to
Element (ppm) an acidic environment: test series 9.
Ba 49
Ce 92
Co 4
Cr 183
La 62
Nb 28
Nd 48
Ni 20
Pb 30
Rb 12
Sr 59
Th 28
V 158
Y 16
Zn 788
Zr 110
Compound (wt.%)
Silicon dioxide (SiO2) 44.92 Fig. 15. The variation in organic compound concentrations
Aluminum oxide (Al2O3) 38.25 leached from tire chips embedded in a kaolin clay subjected to a
Titanium oxide (TiO2) 1.71 basic environment: test series 11.
Iron oxide (Fe2O3) 0.54
Phosphorous oxide (P2O5) 0.29
Sodium oxide (Na2O) 0.25
Potassium oxide (K2O) 0.06
Calcium oxide (CaO) 0.01

Fig. 13. The variation in organic compound concentrations

leached from tire chips embedded in a kaolin clay subjected to a
neutral environment: test series 6.

decrease with increasing number of pore volumes, or would
disappear altogether. The amount of chemicals leached from
tire chips embedded in clay was generally less than that as-
sociated with tire chips embedded in sand. This leaching
trend could be the result of the clay soil covering the tire
chip surface which would limit the amount of surface area
exposed to the leaching medium. Also, these chemical com-
pounds could be absorbed by the clay particles, restricting
their movement. The release of organic compounds from a
saturated clay medium will be strongly time dependent.
Conclusions
Field monitoring of the prototype test embankment con-
structed with tires above the water table indicates that insig-
nificant adverse effects on groundwater quality had occurred
over a period of 2 years. Targeted inorganic elements having
an MAC designation, including barium (Ba), cadmium (Cd),
chromium (Cr), lead (Pb), selenium (Se), and fluoride (F),
were all well below their respective concentration limits.
The analyses for elements having a designated AO, includ-
ing manganese (Mn), copper (Cu), iron (Fe), zinc (Zn),
sodium (Na), chloride (Cl), and sulphate (SO4), were also
well below their respective concentration limits. In this
investigation, only one target organic compound, namely 4-
(2-benzothiazolythio)morpholine, was detected in one of
three test sections and in two of seven sampling events.
Laboratory lysimeter tests performed on tire chips embedded
in inert Unimin quartz sand provided evidence of an increase
in aluminum (Al), iron (Fe), zinc (Zn), and manganese (Mn)

© 2000 NRC Canada

I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:35 PM
Color profile: Generic CMYK printer profile
Composite Default screen
O’Shaughnessy and Garga
concentrations which, with the exception of zinc, exceed
their respective drinking water standards. The increase in
aluminum, iron, and manganese was attributed to the ex-
posed steel reinforcements in the tire chips. The use of tire
chips provides a conservative estimate for fills reinforced
with whole tires, since the latter case provides limited expo-
sure of steel reinforcement and affords additional protection
by the rubberized cover. All inorganic target elements were
below detection limits or background levels for tire chips
embedded in a commercially processed kaolin clay.
The amount of organic compounds leached from the tire
chips decreased with the number of exposure periods or pore
volumes flushed through the soil, with the exception of one
compound in one test performed on leachate collected from
tire chips embedded in sand. Consequently, the probability
of release of any toxicant is higher when using newly dis-
carded tires. Tire-reinforced earth structures should be built
using scrap tires which have been stockpiled or have been
used for some time and should ideally be placed above the
permanent groundwater table.
Environmental considerations would indicate that the use
of whole tires or tires with one sidewall removed embedded
in soil may be preferable to the use of tire chips because the
release of metals is inhibited by the protective rubber cover.
There is no evidence of an exothermal reaction having taken
place in the prototype tire-reinforced embankment.
Acknowledgements
The authors acknowledge the Waste Reduction Branch,
Ontario Ministry of the Environment and Energy, for finan-
cial support through a research agreement with the second
author to study the use of scrap tires in embankments and re-
taining walls.
References
Abernethy, S. 1994. The acute lethality to rainbow trout of water
contaminated by an automobile tire. Report ISBN 0-7778-2381-
0, Aquatic Toxicology Section, Standards Development Branch,
Ontario Ministry of the Environment and Energy, Toronto.
Abernethy, S.G., Montemayor, B.P., and Penders, J.W. 1996. The
aquatic toxicity of scrap automobile tires. Report ISBN 0-7778-
4835-X, Aquatic Toxicology Section, Standards Development
I:\cgj\Cgj37\Cgj02\T99-086.vp
Thursday, March 09, 2000 1:38:36 PM
131
Branch, Ontario Ministry of the Environment and Energy, To-
ronto.
AB-Malek, K., and Stevenson, A. 1986. The effects of 42 years
immersion in sea water on natural rubber. Journal of Material
Sciences, 21: 147–154.
Blumenthal, M. 1998. Scrap tire market development: the impact
of state programs. Resource Recycling, Vol. 17, No. 3, pp. 15–
19.
Brownlee, B.G., Carey, J.H., MacInnis, G.A., and Pellizzari, I.T.
1992. Aquatic environmental chemistry of 2-(thiocyano-
methylthio) benzothiazole and related benzothiazoles. Environ-
mental Toxicology and Chemistry, 11: 1153–1168.
Eldin, N.N., and Senouci, A.B. 1992. Use of scrap tires in road
construction. Journal of Construction Engineering and Manage-
ment, ASCE, 118(3): 561–576.
Garga, V.K., and O’Shaughnessy, V. 2000. Tire-reinforced earthfill.
Part 1: Construction of a test fill, performance, and retaining
wall design. Canadian Geotechnical Journal, 37: 75–96.
Humphrey, D.N. 1996. Investigation of exothermic reaction in the
shred fill located on SR 100 in Ilwaco, Washington. Federal
Highway Administration, U.S. Department of Transportation,
Washington, D.C.
Humphrey, D.N., Katz, E.L., and Blumenthal, M. 1997. Water
quality effects of tire chips fills placed above the groundwater
table. In Testing soil mixed with waste or recycled materials.
Edited by M.A. Wasemiller and K.B. Hoddinott. American Soci-
ety for Testing and Materials, Special Technical Report 1275,
pp. 299–313.
Lerner, A., Naugle, A., LaForest, J., and Loomis, W. 1993. Study
of waste tire leachability in potential disposal and usage envi-
ronments. Amended volume 1: final report. Department of Envi-
ronmental Engineering Sciences, College of Engineering,
University of Florida, Gainesville.
Ontario Ministry of the Environment and Energy. 1994. Ontario
drinking water objectives. Revised edition. Ontario Ministry of
the Environment and Energy, Report ISBN 0-7743-8985-0.
Taylor, R., and Son, P.N. 1982. Rubber chemicals. In Kirk-Othmer
encyclopedia of environmental chemistry. Vol. 20. 3rd ed.. John
Wiley and Sons Inc., Toronto, pp. 337–364.
Twin City Testing Corp. 1990. Environmental study of the use of
shredded waste tires for roadway sub-grade support. Report for
the Groundwater and Solid Waste Division, Minnesota Pollution
Control Agency by Twin City Testing Corp., St. Paul, Minn.
Whetten, N., Weaver, J., Humphrey, D., and Sandford, T. 1997.
Rubber meets the road in Maine. Civil Engineering, 67(9): 60–
63.
Williams, P.T., Besler, S., and Taylor, D.T. 1990. The pyrolysis of
scrap automotive tyres: the influence of temperature and heating
rate on product composition. Fuel, 69: 1474–1482.
© 2000 NRC Canada