recycling
Article
Properties of Shredded Roof Membrane–Sand
Mixture and Its Application as Retaining Wall
Backfill under Static and Earthquake Loads
Bennett Livingston and Nadarajah Ravichandran *
Glenn Department of Civil Engineering, Clemson University, 202 Lowry Hall, Clemson, SC 29634, USA;
[email protected]
* Correspondence: [email protected]; Tel.:+1-864-656-2818
Academic Editor: Michele Rosano
Received: 6 December 2016; Accepted: 30 March 2017; Published: 5 April 2017
Abstract: About 20 billion square feet of Ethylene Propylene Diene Monomer (EPDM) rubber is
installed on roofs in the United States and most of them will be reaching the end of their lifespan soon.
The purpose of this study is to investigate potential reuses of this rubber in Civil Engineering projects
rather than disposing it into landfills. First, laboratory tests were performed on various shredded
rubber-sand mixtures to quantify the basic geotechnical engineering properties. The laboratory test
results show that the shredded rubber-sand mixture is lightweight with good drainage properties
and has shear strength parameters comparable to sand. This indicates that the rubber-sand mixture
has potential to be used for retaining wall backfill and many other projects. To assess the economic
advantage of using shredded rubber-sand mixtures as a lightweight backfill for retaining walls
subjected to static and earthquake loadings, geotechnical designs of a 6 m tall gravity cantilever
retaining wall were performed. The computed volume of concrete to build the structural components
and volume of backfill material were compared with those of conventional sand backfill. Results show
significant reductions in the volume of concrete and backfill material in both static and earthquake
loading conditions when the portion of shredded rubber increased in the mixture.
Keywords: shredded EPDM rubber; roof membrane; lightweight fill; retaining wall backfill
1. Introduction
The growing human population and consumption of raw materials to meet our needs are not
only depleting our resources, but also increasing the amount of waste materials that may be disposed
in landfills or simply thrown into open lands. To combat this problem, Civil Engineers have been
investigating various environmentally sound recycling practices to salvage and reuse construction
materials such as concrete, rubber, glass, metal slag, asphalt pavement and other waste materials
such as waste tires and roof membranes. One area of interest within geotechnical engineering is the
use of waste material as an alternative to coarse grained soil. Replacing heavy soil with lightweight
recycled material such as rubber products having comparable engineering properties can not only
solve a number of geotechnical engineering design and maintenance issues, but also help to reduce
the consumption of natural resources for a sustainable future. Today, most of this rubber waste
is derived from the hundreds of millions of scrap tires generated annually. Significant amounts
of research have been performed in the last decade to investigate the applicability of waste scrap
tires as an alternative to soils in geotechnical engineering [1–4]. Another rubber product which is
removed in large quantities, but hasn’t received much attention from the geotechnical engineering
community is Ethylene Propylene Diene Monomer (EPDM) roofing membrane. A pilot recycling
program launched in 2006 by the EPDM Roofing Association (ERA) has estimated that between 2006
Recycling 2017, 2, 8; doi:10.3390/recycling2020008
www.mdpi.com/journal/recycling
Recycling 2017,
2017, 2,
2, 88
Recycling
2 of 14
estimated that between 2006 and 2010 around 6 million square feet of EPDM rubber was dumped
and
around
million square
of EPDM
rubber
dumped
into landfills
or available
to be
into 2010
landfills
or 6available
to be feet
recycled.
Similar
to was
waste
tires, these
roof membranes
can
be
recycled.
Similar
to
waste
tires,
these
roof
membranes
can
be
shredded
and
used
as
a
viable
lightweight
shredded and used as a viable lightweight alternative to heavy granular soil in geotechnical
alternative
to These
heavy applications
granular soil may
in geotechnical
engineering.
These applications
include
backfill
engineering.
include backfill
for retaining
walls, cover may
systems
in landfills
for
retaining
walls,
cover
systems
in
landfills
[5],
subgrade
layers
for
roadways
[6],
embankment
[5], subgrade layers for roadways [6], embankment fill [7,8], and Mechanically Stabilized Earth
fill
[7,8],
andfill
Mechanically
Stabilized Earth (MSE) wall fill [9].
(MSE)
wall
[9].
Roofing
is already
Protection Agency
Agency (EPA)
(EPA)
Roofing waste
waste material
material is
already aa problem
problem as
as the
the Environmental
Environmental Protection
reported
that
approximately
40%
of
landfill
waste
comes
from
construction
and
demolition
debris,
reported that approximately 40% of landfill waste comes from construction and demolition debris,
with
with aa quarter
quarter of
of that
that generated
generated from
from roofing
roofing materials
materials [10].
[10]. Over
Over the
the last
last 40
40 years,
years, 20
20 billion-plus
billion-plus
square
feet
of
rubber
roof
membrane
has
been
installed
on
roofs
in
the
United
States
[10].
square feet of rubber roof membrane has been installed on roofs in the United States [10]. With
With roof
roof
membrane
having
a
lifespan
ranging
from
30
to
60
years,
large
quantities
of
roof
membrane
will soon
membrane having a lifespan ranging from 30 to 60 years, large quantities of roof membrane
will
be
dumped
into landfills
if alternative
uses aren’t
found.found.
Luckily,
recycling
programs
launched
by the
soon
be dumped
into landfills
if alternative
uses aren’t
Luckily,
recycling
programs
launched
ERA
have
that recycling
roof membrane
is an easy
process
but needs
studies
by the
ERAshown
have shown
that recycling
roof membrane
is an
easy process
butcomprehensive
needs comprehensive
to
properly
use
in
Civil
Engineering
applications.
The
process
starts
by
first
removing
the
ballast
studies to properly use in Civil Engineering applications. The process starts by first removing
the
covering
the
roof
membrane,
cutting
it
into
smaller
widths
while
avoiding
roof
fasteners
and
areas
ballast covering the roof membrane, cutting it into smaller widths while avoiding roof fasteners and
with
rollingrolling
it up, and
shipping
to the recycling
facility to
grindto
it into
[11].
Figure
areasadhesive,
with adhesive,
it up,
and shipping
to the recycling
facility
grindchips
it into
chips
[11].1
illustrates
this process
by process
showingby
theshowing
roof membrane
and after
shredding.
Furthermore,
Figure 1 illustrates
this
the roofbefore
membrane
before
and after
shredding.
storage
of
EPDM
rubber
is
not
an
issue
as
it
is
with
shredded
tires
since
roof
membrane
rubber
not
Furthermore, storage of EPDM rubber is not an issue as it is with shredded tires sinceisroof
◦ C).
susceptible
to
fire
due
to
its
higher
fire
resistance
(flash
point
>
200
membrane rubber is not susceptible to fire due to its higher fire resistance (flash point > 200 °C).
Figure 1. Roof membrane rubber cuts (photo courtesy of EPDMroofs.org) and shreds.
Figure 1. Roof membrane rubber cuts (photo courtesy of EPDMroofs.org) and shreds.
The objectives of the present study are to (a) determine the geotechnical engineering properties
The objectives of the present study are to (a) determine the geotechnical engineering properties
of the composite material (shredded roof membrane mixed with sand) from laboratory testing and
of the composite material (shredded roof membrane mixed with sand) from laboratory testing and
(b) perform geotechnical design and evaluate the economic advantage of using the composite as an
(b) perform geotechnical design and evaluate the economic advantage of using the composite as an
alternative to conventional backfill for retaining walls under static and earthquake loading
alternative to conventional backfill for retaining walls under static and earthquake loading conditions.
conditions. The mixtures that were tested have ratios of shredded rubber and sand ranging from
The mixtures that were tested have ratios of shredded rubber and sand ranging from 100:0, 75:25,
100:0, 75:25, 50:50, 25:75, and 0:100%. To assess the economic advantage of using shredded
50:50, 25:75, and 0:100%. To assess the economic advantage of using shredded rubber-sand mixtures
rubber-sand mixtures as a lightweight backfill for retaining walls subjected to static and earthquake
as a lightweight backfill for retaining walls subjected to static and earthquake shaking, geotechnical
shaking, geotechnical design (length of toe and heel of base slab/footing) of a 6 m tall (stem height
design (length of toe and heel of base slab/footing) of a 6 m tall (stem height above base slab) gravity
above base slab) gravity cantilever retaining wall was performed. The volume of concrete to build
cantilever retaining wall was performed. The volume of concrete to build the structural components
the structural components and volume of backfill material were computed based on the design
and volume of backfill material were computed based on the design results for each rubber-sand
results for each rubber-sand mixture and compared with that of conventional sand backfill.
mixture and compared with that of conventional sand backfill.
2. Summary of Geotechnical Engineering Properties of Shredded Roof Membrane-Sand Mixture
2. Summary of Geotechnical Engineering Properties of Shredded Roof Membrane-Sand Mixture
and Potential
Potential Applications
Applications in
in Civil
Civil Engineering
Engineering
and
Since most
most of
of the
the focus
focus has
has been
been on
on the
the laboratory
laboratory testing
testing and
and performance
performance evaluation
evaluation of
of waste
waste
Since
tires, there
thereisisa gap
a gap
in research
the research
for using
other rubber
waste materials
such roof
as EPDM
roof
tires,
in the
for using
other rubber
waste materials
such as EPDM
membrane
membrane
that
may
behave
similarly.
Evaluating
the
properties
and
performance
of
such
new
that may behave similarly. Evaluating the properties and performance of such new materials may lead
materials may lead to cost effective, environmentally friendly, and sustainable recycling
Recycling 2017, 2, 8
Recycling 2017, 2, 8
3 of 14
3 of 14
applications.
In this
section, the results
of key
are summarized
forIn
various
shredded
to
cost effective,
environmentally
friendly,
andlaboratory
sustainabletests
recycling
applications.
this section,
the
rubber-sand
(by
weight).
results
of keymixtures
laboratory
tests
are summarized for various shredded rubber-sand mixtures (by weight).
The tests
teststhat
thatwere
were
performed
on shredded
the shredded
rubber-sand
mixtures
aretosimilar
to the
performed
on the
rubber-sand
mixtures
are similar
the laboratory
laboratory
testing on
performed
shredded
tires
by[1],
Cecich
et al. [1],
Moo-Young
et al. et
[12],
and
testing
performed
shreddedon
tires
by Cecich
et al.
Moo-Young
et al.
[12], and Warith
al. [13].
Warithtests
et al.included
[13]. These
testsanalyses,
includeddensity
sieve analyses,
density
tests,
direct
shear
tests, and
hydraulic
These
sieve
tests, direct
shear
tests,
and
hydraulic
conductivity
conductivity
teststhe
tograin
measure
grain sizedry
distribution,
dryfriction
unit weight,
friction angle,
cohesion,
tests
to measure
size the
distribution,
unit weight,
angle, cohesion,
and hydraulic
and hydraulicofconductivity
of various
shreddedmixtures.
rubber–sand
mixtures. rubber
The shredded
used in
conductivity
various shredded
rubber–sand
The shredded
used inrubber
the laboratory
the
laboratory
testing
was
provided
by
RK
HydroVac
Inc.
and
was
shredded
to
an
average
size of
testing was provided by RK HydroVac Inc. and was shredded to an average size of 1.27 mm. A medium
1.27
mm. A medium
fine-grained
sand
was used
for preparing
mixtures
to
fine-grained
sand to
obtained
locally
wasobtained
used forlocally
preparing
different
mixtures different
by mixing
it with
by mixingroof
it with
shreddedDue
rooftomembrane.
Due in
tosize,
the difference
size, shape
weightrubber
of the
shredded
membrane.
the difference
shape andinweight
of theand
shredded
shredded
rubber
and
sand,
preparing
uniform
mixtures
was
challenging.
Each
of
the
three
and sand, preparing uniform mixtures was challenging. Each of the three aforementioned mixtures
aforementioned
mixtures
containing
shredded
rubbermixed
and sand
werewith
thoroughly
hand
containing
shredded
rubber
and sand were
thoroughly
by hand
extra caremixed
beingby
taken
to
with extra
care beingoftaken
to prevent
segregation
of the
sand and
shredded
rubber
particles.
Each
prevent
segregation
the sand
and shredded
rubber
particles.
Each
of the lab
tests for
the various
of the labwas
testsrepeated
for the various
samples
wasresults
repeated
two times
andare
thethe
results
presented
are
samples
two times
and the
presented
below
average
of the below
two tests.
the averagedetails
of the two
tests.
Additional
can be found[14].
in Kyser and Ravichandran [14].
Additional
can be
found
in Kyserdetails
and Ravichandran
2.1. Particle Size Distributions of Constituent Materials
Sieve analyses were conducted to obtain the particle size distribution which provided insight
into the hydraulic conductivity, compressibility, and
and shear
shear strength
strength of
of the
the material.
material. The tests were
conducted following the ASTM D422 procedure [15]. The results for each material are presented in
Figure 2.
100
Rubber
Sand
Percent passing
80
60
40
20
0
100.00
10.00
1.00
0.10
0.01
Particle diameter (mm)
Figure 2. Particle size distribution of shredded rubber and sand.
Figure 2. Particle size distribution of shredded rubber and sand.
By examining the distribution curves in Figure 2, it can be seen that the overall size of the
By
examining
distributionlarger
curvesthan
in Figure
seen that
that the
overall
size ofwith.
the shredded
shredded
rubber the
is somewhat
that2,ofit can
the besand
it was
mixed
Further
rubber
is somewhat
larger than curves
that of shows
the sand
that
was mixedrubber
with. is
Further
of the
examination
of the distribution
that
theit shredded
poorlyexamination
graded, while
the
distribution
curves
shows
that
the
shredded
rubber
is
poorly
graded,
while
the
sand
is
more
uniformly
sand is more uniformly graded. Using Figure 2, the diameter corresponding to 50% passing (D50) of
graded.
Using rubber
Figure 2,
thefound
diameter
corresponding
to the
50%effective
passing (D
shredded
rubber
was
50 ) of the
the shredded
was
to be
3.0 mm with
diameter
which
is the
diameter
found
to
be
3.0
mm
with
the
effective
diameter
which
is
the
diameter
corresponding
to
10%
passing
corresponding to 10% passing (D10) being 2.1 mm. With these results the uniformity coefficient (Cu)
(D
being 2.1 mm.
With these
the uniformity
(Cu )respectively.
and coefficient
ofsand
curvature
(Cc )
10 )coefficient
and
of curvature
(Cc)results
were calculated
to becoefficient
1.52 and 0.93,
The
however,
were
calculated
to sized
be 1.52
and 0.93,
respectively.
The
sand
has
much
has much
smaller
particles
with
a D50 of 0.45
mm
andhowever,
D10 of 0.24
mm.
Thesmaller
Cu andsized
Cc of particles
the sand
with
a
D
of
0.45
mm
and
D
of
0.24
mm.
The
C
and
C
of
the
sand
were
calculated
to smaller
be 2.29
u
c
50
were calculated
to be 2.29 and100.93, respectively. The fact that the shredded rubber chips are
and
0.93,
respectively.
The
fact
that
the
shredded
rubber
chips
are
smaller
than
shredded
tires
than shredded tires points out a clear advantage for the shredded rubber chips while mixingpoints
with
out
theof
shredded
rubberrubber
chips while
withfor
sand
sincemixing
the smaller
size of
sanda clear
since advantage
the smallerfor
size
the shredded
chips mixing
will allow
easier
procedures.
When mixing the rubber chips and sand, the large size difference does not allow for uniformity in
the mixtures. The rubber chips and sand particles that are similar in size may be easier to mix in the
Recycling 2017, 2, 8
4 of 14
the shredded rubber chips will allow for easier mixing procedures. When mixing the rubber chips
size difference does not allow for uniformity in the mixtures. The rubber 4chips
of 14
and sand particles that are similar in size may be easier to mix in the field, allowing for easy grading
field,
allowing
for easypractices
grading can
practices.
Easy grading
practices
can beasadvantageous
forand
project
practices.
Easy grading
be advantageous
for project
budgets,
it will save time
cost
budgets,
as it will
time and
cost in crew labor and equipment maintenance.
in crew labor
and save
equipment
maintenance.
and sand,
Recycling
2017,the
2, 8large
2.2. Dry
Dry Unit
Unit Weight
Weight
2.2.
Determining the
the dry
dry unit
unit weight
weight of
of the
the material
material is
is important
important as
as the
the self-weight
self-weight affects
affects the
the in
in
Determining
situ vertical
vertical and
and horizontal
horizontal stresses,
stresses, and
and the
the settlement
settlement of
of underlying
underlying soil
soil and
and structures.
structures. The
The use
use of
of
situ
this new
new composite
composite material
material may
may vary
vary depending
depending upon
upon the
the application
application discussed
discussed in
in later
later sections,
sections, as
as
this
change in
in the
the degree
degree of
of compaction
compaction will
will change
change the
the unit
unit weight.
weight. In
Inthis
thisstudy,
study, unit
unit weight
weight of
of the
the
aa change
material in its
60%
Standard
Proctor
compaction
energy
was determined
following
ASTM
material
itsdry
drystate
stateat at
60%
Standard
Proctor
compaction
energy
was determined
following
D698 [16]
and
as and
suggested
in Moo-Young
et al. [12]
shredded
tires; a material
similar to
the new
ASTM
D698
[16]
as suggested
in Moo-Young
et for
al. [12]
for shredded
tires; a material
similar
to
composite
material
used
in
this
study.
It
should
be
noted
that
using
the
vibration
method
to
compact
the new composite material used in this study. It should be noted that using the vibration method to
the sample
easily
segregate
materialsmaterials
and therefore
was not was
used.not
In used.
addition,
the unit the
weights
compact
thewill
sample
will
easily segregate
and therefore
In addition,
unit
in their uncompacted
states were
measured
for usingfor
it as
loose
fill,loose
drainage
and filtration
element.
weights
in their uncompacted
states
were measured
using
it as
fill, drainage
and filtration
The results
different
shreddedshredded
roof membrane-sand
mixtures mixtures
are presented
in Figure in
3. Figure 3.
element.
Thefor
results
for different
roof membrane-sand
are presented
20
Dry unit weight (kN/m3)
Compacted (60% Standard Proctor)
Uncompacted
16
12
8
4
0
0
20
40
60
80
100
Percentage of roof membrane
Figure 3. Variation of dry unit weight of the composite material with percentage of shredded roof
Figure 3. Variation of dry unit weight of the composite material with percentage of shredded
membrane.
roof membrane.
The unit weights of shredded rubber were found to be similar to that of shredded tires reported
The unit
weights
of shredded
rubber
be seen
similar
that of3 shredded
reported in
by
by Cecich
et al.
[1] and
Moo-Young
et al.were
[12].found
It cantobe
in to
Figure
that withtires
the increase
Cecich
et
al.
[1]
and
Moo-Young
et
al.
[12].
It
can
be
seen
in
Figure
3
that
with
the
increase
in
percentage
percentage of shredded rubber in the mixture, the uncompacted and compacted unit weights
of shreddeddecrease
rubber infrom
the mixture,
the3 uncompacted
and
compacted unit
weights
decrease
3, respectively
sustainably
14.4 kN/m
and 16.2 kN/m
for the
100% sustainably
sand mixture
to 5.0
3 and 16.2 kN/m3 , respectively for the 100% sand mixture to 5.0 kN/m3 and 6.2 kN/m3 ,
from
14.4
kN/m
3
3
kN/m and 6.2 kN/m , respectively for the 100% rubber mixture. The decrease in unit weight verifies
respectively
for the
rubber
mixture.
The decrease
unithas
weight
verifiesestablished
that this material
can
that
this material
can100%
behave
similarly
to shredded
tires, in
which
reasonably
literature.
behave similarly to shredded tires, which has reasonably established literature.
2.3. Effective Friction Angle
2.3. Effective Friction Angle
The friction angle is another important parameter used in the geotechnical application. The
The friction angle is another important parameter used in the geotechnical application. The shear
shear strength of the composite coarse material is a function of its friction angle and overburden
strength of the composite coarse material is a function of its friction angle and overburden pressure.
pressure. Direct shear and triaxial apparatuses are widely used for measuring the drained and
Direct shear and triaxial apparatuses are widely used for measuring the drained and undrained friction
undrained friction angle of frictional material in the laboratory. Between these two apparatuses, the
angle of frictional material in the laboratory. Between these two apparatuses, the direct shear test is an
direct shear test is an easy to use apparatus that gives reasonably accurate results. The various
easy to use apparatus that gives reasonably accurate results. The various mixtures of shredded rubber
mixtures of shredded rubber and sand were tested in their loose state in a geotechnical laboratory
using a direct shear apparatus following the standard procedures outlined in ASTM D 3080 [17]. The
tests were conducted at four different normal (vertical) stresses including the reference pressure (1
atm.). For each normal stress, two tests were conducted and the results were averaged to ensure the
accuracy of the results.
Recycling 2017, 2, 8
5 of 14
and sand were tested in their loose state in a geotechnical laboratory using a direct shear apparatus
following the standard procedures outlined in ASTM D 3080 [17]. The tests were conducted at5 four
Recycling 2017, 2, 8
of 14
different normal (vertical) stresses including the reference pressure (1 atm.). For each normal stress,
two tests
were
conducted
andtesting
the results
averaged
to ensure
the accuracy
of the percentages
results.
It was
observed
during
thatwere
highly
compressible
mixtures
with large
of
It
was
observed
during
testing
that
highly
compressible
mixtures
with
large
percentages
of
shredded rubber did not display a peak in the shear stress vs. shear displacement plot. According to
shredded rubber
did
not shear
display
a peak
theconsidered
shear stresstovs.
shear
displacement
plot.
According
AASHTO
T 236-86
[18],
failure
is in
also
occur
when
the sample
experiences
a
to
AASHTO
T
236-86
[18],
shear
failure
is
also
considered
to
occur
when
the
sample
experiences
horizontal displacement equal to 10% of the sample diameter. For these particular samples, the sheara
horizontal
displacement
equal
to 10%
the sample diameter.
particular
samples,
the shear
stress
was considered
to be
at the
shearofdisplacement
of 10% ofFor
thethese
sample
diameter.
Additionally,
it
stress
was considered
to be
thefit
shear
10%stress
of thevs.
sample
diameter.
Additionally,
was
noticed
that the lines
ofat
best
useddisplacement
to create the of
shear
normal
stress plots
to obtain
it was noticed
that thealines
of cohesion
best fit used
to create
shear
vs. normal
stress
plots
to obtain
friction
angle yielded
small
intercept
forthe
each
test,stress
including
the 100%
sand
mixture.
It
frictionbe
angle
yielded
a small cohesion
intercept
for each
including
the 100%
sand
should
noted
that researchers
have reported
cohesion
fortest,
sandy
soils in direct
shear
testsmixture.
results
It should
be noted
researchers
reported
for sandy
soils in direct failure
shear tests
results
and
attributed
it that
to be
apparenthave
cohesion
andcohesion
a nonlinear
Mohr-Coulomb
envelope.
and
attributed
it
to
be
apparent
cohesion
and
a
nonlinear
Mohr-Coulomb
failure
envelope.
Therefore,
Therefore, for simplicity, all cohesion values were set to zero and the best fit lines were drawn
for simplicity,
all cohesion
values
were set
toare
zero
and the best
fit lines
were4,
drawn
though
the origin.
though
the origin.
The results
of these
tests
presented
below
in Figure
and are
the average
of
The
results
of
these
tests
are
presented
below
in
Figure
4,
and
are
the
average
of
each
of
the
test
results.
each of the test results.
Friction angle (deg.)
40
38
36
34
Experiment
32
Best fit line (R2 = 0.65)
30
0
20
40
60
80
100
Percentage of roof membrane
Figure 4. Variation of effective friction angle with percentage of shredded roof membrane.
Figure 4. Variation of effective friction angle with percentage of shredded roof membrane.
The best fit line for the variation of friction angle with the percentage of roof membrane
The best
fit line 4forshows
the variation
frictionthe
angle
with angle
the percentage
of roof membrane
presented
presented
in Figure
that, in of
general,
friction
of the composite
material increased
in
Figure
4
shows
that,
in
general,
the
friction
angle
of
the
composite
material
increased
from
from approximately 32° to 38° when the amount of roof membrane increased from 0 to 100%,
◦
◦
approximately
to 38understood
when the amount
roof membrane
0 to 100%,
respectively.
It 32
is well
that the of
friction
angle of aincreased
material from
is a function
of respectively.
its mineral
It
is
well
understood
that
the
friction
angle
of
a
material
is
a
function
of
its
mineral
content,
geometric
content, geometric properties, and physical properties. In the case of the shredded rubber-sand
properties,
and
physical
properties.
In
the
case
of
the
shredded
rubber-sand
mixtures,
materials
mixtures, materials with different mineral content, geometry and physical properties
were
with
different
mineral
content,
geometry
and
physical
properties
were
interacting
during
shearing.
interacting during shearing. Therefore, it is difficult to provide a single reason for the observed
Therefore,In
it is
difficultit to
provide
single
reason forinteraction
the observed
behavior.
In general,
can be
due to
behavior.
general,
can
be duea to
the complex
of rubber
particles
with itsand
particles
the
complex
interaction
of
rubber
particles
with
sand
particles
which
increased
the
resistance
against
which increased the resistance against shearing. When the rubber particles deformed, the interaction
shearing.complex
When the
rubber
particles deformed,
interaction
becomes
complexWhen
and the
becomes
and
the interpretation
of the the
results
becomes
very complex.
theinterpretation
results from
of
the
results
becomes
very
complex.
When
the
results
from
Figure
4
were
compared
to al.
the[1],
friction
Figure 4 were compared to the friction angle of shredded tires reported by Cecich et
and
angle
of
shredded
tires
reported
by
Cecich
et
al.
[1],
and
Moo-Young
et
al.
[12],
the
friction
angles
of
Moo-Young et al. [12], the friction angles of shredded rubber-and mixtures were found to be notably
shredded
rubber-and
mixtures
were
found
to
be
notably
higher.
This
means
that
the
various
mixtures
higher. This means that the various mixtures can provide resistance against shear that is greater than
can provide
resistance
againstfor
shear
greater stress
than that
of shredded
alone for thefailure
same
that
of shredded
tires alone
the that
sameisnormal
as explained
by tires
Mohr-Coulomb
normal stress
as explained
by Mohr-Coulomb
equation.
Therefore,
not only be
willreduced,
the amount
equation.
Therefore,
not only
will the amountfailure
of waste
materials
sent to landfills
but
of
waste
materials
sent
to
landfills
be
reduced,
but
feasible
engineering
designs
will
also
be possible
feasible engineering designs will also be possible without compromising on strength.
without compromising on strength.
2.4. Hydraulic Conductivity
Another important engineering characteristic of the shredded rubber-sand mixtures that was
determined in the laboratory was its hydraulic conductivity. Based on the particle size distribution
and friction angle, it can be used as a retaining wall backfill in certain areas where there is no
Recycling 2017, 2, 8
6 of 14
2.4. Hydraulic Conductivity
Another important engineering characteristic of the shredded rubber-sand mixtures that was
determined
and
Recycling 2017, in
2, 8the laboratory was its hydraulic conductivity. Based on the particle size distribution
6 of
14
friction angle, it can be used as a retaining wall backfill in certain areas where there is no surcharge load
expected. For
such
application,
material
should have
significantly
higher
hydraulic
conductivity
to
surcharge
load
expected.
For the
such
application,
the material
should
have
significantly
higher
ensure thatconductivity
there is no pore
waterthat
pressure
wall. The
build-up
of pore
pressures
hydraulic
to ensure
there build-up
is no porebehind
water the
pressure
build-up
behind
the wall.
The
can induce
stress, can
demanding
larger structures
decrease
of the
shear capacity
build-up
of additional
pore pressures
induce additional
stress, and/or
demanding
larger
structures
and/or
of the composite
material.
Additionally,
due material.
to the much
smaller sand
being
able to
be
decrease
of the shear
capacity
of the composite
Additionally,
dueparticles
to the much
smaller
sand
placed
in-between
the
larger
rubber
particles,
it
is
important
to
understand
how
water
flows
through
particles being able to be placed in-between the larger rubber particles, it is important to understand
the mixture.
how
water flows through the mixture.
Past studies
studies on
on shredded
shredded tires
tires by
by Warith
Warith et al. [13] used the constant head test to determine
determine
Past
hydraulic conductivity,
performed
in this
study,
sincesince
the sand
particles
and the
larger
hydraulic
conductivity,which
whichwas
wasalso
also
performed
in this
study,
the sand
particles
and
the
rubberrubber
particles
wouldwould
allowallow
for high
conductivity
to betoobserved.
When
preparing
the
larger
particles
for hydraulic
high hydraulic
conductivity
be observed.
When
preparing
samples
for testing,
compaction
energyenergy
was not
applied
while filling
thefilling
specimen
tube. Therefore,
the
samples
for testing,
compaction
was
not applied
while
the specimen
tube.
the mixtures
testedwere
in their
loose
state.loose
Additionally,
special carespecial
was taken
to create
uniform
Therefore,
thewere
mixtures
tested
in their
state. Additionally,
care was
takena to
create
at the
top ofat
thethe
specimen
using minimal
by leveling
it withby
a straight
Once
asurface
uniform
surface
top of tube
the specimen
tube vibration
using minimal
vibration
levelingedge.
it with
a
the
samples
were
prepared,
they
were
then
tightly
secured
to
the
reservoir
tube
creating
a
vacuum
seal.
straight edge. Once the samples were prepared, they were then tightly secured to the reservoir tube
Using a laboratory
permeameter,
the tests were
performedthe
and
thewere
required
data sets
were
creating
a vacuum seal.
Using a laboratory
permeameter,
tests
performed
and
the recorded.
required
Oncesets
the testing
was complete,
recorded and
in Figure
5. It should
be noted
data
were recorded.
Once the results
testing were
was complete,
the plotted
results were
recorded
and plotted
in
that during
testing
the 100%
rubber
mixture,
hydraulic
conductivity
was so large
Figure
5. It the
should
beofnoted
that shredded
during the
testing
of thethe
100%
shredded
rubber mixture,
the
that the head
had to be reduced
order
to increase
experiment
time, thus
reducing
the effectthe
of
hydraulic
conductivity
was so in
large
that
the headthe
had
to be reduced
in order
to increase
experiment
time, thus
reducing
effect of error in recording the time of the experiment.
error in recording
the time
of thethe
experiment.
Hydraulic Conductivity (cm/sec)
2.5
2.0
1.5
1.0
0.5
0.0
0
20
40
60
80
100
Percentage of roof membrane
Figure 5. Variation of hydraulic conductivity with percentage of shredded roof membrane.
Figure 5. Variation of hydraulic conductivity with percentage of shredded roof membrane.
The hydraulic conductivity of the various shredded rubber-sand mixtures shown in Figure 5
Thean
hydraulic
conductivity
of the
shredded
mixtures
shown
in Figure
display
interesting
trend. It can
bevarious
seen that
addingrubber-sand
25% shredded
rubber
did not
have a5
display
an
interesting
trend.
It
can
be
seen
that
adding
25%
shredded
rubber
did
not
have
a
significant
significant effect on the hydraulic conductivity. Once 50% shredded rubber was added, the
effect on the
hydraulic conductivity.
Once 50% shredded
was to
added,
the hydraulic
conductivity
hydraulic
conductivity
increased dramatically.
When rubber
compared
shredded
tires, the
hydraulic
increased
dramatically.
When
compared
to
shredded
tires,
the
hydraulic
conductivity
at
50%
shredded
conductivity at 50% shredded rubber and greater matches the results of research conducted
by
rubber
and
greater
matches
the
results
of
research
conducted
by
Warith
et
al.
[13].
The
higher
hydraulic
Warith et al. [13]. The higher hydraulic conductivity for the composite material is a desirable
conductivity
for the
composite material
is a applications
desirable property
forallow
many
engineering
property
for many
geotechnical
engineering
and will
forgeotechnical
free drainage
with little
applications
and
will
allow
for
free
drainage
with
little
to
no
build-up
of
pore
pressures.
to no build-up of pore pressures.
2.5. Potential Applications of Shredded Roof Membrane-Sand Mixture in Civil Engineering
By using shredded rubber-sand mixtures in civil engineering applications, engineers can save
20 billon square feet of dilapidated rubber roof membrane from being disposed of in landfills. The
summary of experimental results presented in the previous sections shows that the new composite
material has many desirable properties for using it as a replacement for heavy granular soil.
Recycling 2017, 2, 8
7 of 14
2.5. Potential Applications of Shredded Roof Membrane-Sand Mixture in Civil Engineering
By using shredded rubber-sand mixtures in civil engineering applications, engineers can save
20 billon square feet of dilapidated rubber roof membrane from being disposed of in landfills. The
summary of experimental results presented in the previous sections shows that the new composite
material has many desirable properties for using it as a replacement for heavy granular soil.
The results show that the shredded rubber-sand mixtures have a low dry unit weight, a high
friction angle, and can be extremely porous once at least 50% shredded rubber is added to the mixture.
All of these properties in combination with each other can lead to civil engineering designs that are
just as sustainable as those with other lightweight materials such as shredded tires. Since the rubber
is naturally lighter than soil, its dry unit weight in shredded form is lower than that of soil fill. The
lowered dry unit weight will decrease the stresses induced on structures leading to designs that can be
economical. Moreover, since the roof membrane rubber is lighter, fill mixtures with large rubber ratios
can be placed on soils that are soft and compressible and not induce large settlements. Additionally,
the direct shear testing of the roof membrane rubber with and without sand has shown that it has
a high friction angle. Results show, in general, that the shredded rubber has higher shear resistance
than that of sand. This indicates that the composite material can be used in situations where shearing
failure planes may develop in embankments or roadway subgrades. Lastly, the pore distribution
of the particles with at least 50% rubber becomes large enough for water to flow freely through it.
While the shredded rubber-sand mixture has proven to behave well in the laboratory, it is currently an
unused material that has many potential uses. To help fill the void left in this field of study, a potential
application of shredded rubber-sand mixtures is discussed below.
3. Geotechnical Design and Analysis of Retaining Wall Backfilled with Recycled Shredded Roof
Membrane-Sand Mixture
While there are many potential applications of the shredded rubber-sand mixtures as lightweight
fill, its use as retaining wall backfill is of particular interest. Due to the large amounts of material
needed for backfill volume, shredded rubber mixed with sand has the potential to drastically reduce
landfill volumes by diverting the roof membrane for reuse. Additionally, much smaller retaining
walls can be built when lightweight backfill is used, leading to cost savings in concrete volume and
excavation volume as shown by the research on shredded tires in retaining walls by Cecich et al. [1]
and Ravichandran and Huggins [4]. Since there is no research quantifying the benefits of shredded
rubber-sand backfills, the aim of this example problem is to do so by determining the cost savings
when using it as retaining wall backfill. A total of five backfill mixtures were studied, which included
the 100:00, 75:25, 50:50, 25:75, and 0:100 shredded rubber:sand mixtures previously discussed. Using
the results obtained from laboratory testing, the retaining wall dimensions and cost for these four
mixtures were compared to the dimensions and cost of a retaining wall backfilled with conventional
sand fill. These analyses were conducted for the static condition with no earthquake loading and the
dynamic condition where earthquake loading was applied to determine the applicability of shredded
rubber in seismically active areas.
3.1. Problem Definition and Methodology
This problem includes a 6 m high cantilever gravity wall with a backfill inclination of 4H:1V
as illustrated in Figure 6. The toe of the retaining wall is buried 1 m below the in situ soil which is
assumed to contain some clayey soil. The slope of the backfill material is considered to be 1.5H:1V,
which is deemed as being a practical inclination value. Groundwater will not be part of the retaining
wall designs and is assumed to be deep below the footing of the retaining wall.
3.1. Problem Definition and Methodology
This problem includes a 6 m high cantilever gravity wall with a backfill inclination of 4H:1V as
illustrated in Figure 6. The toe of the retaining wall is buried 1 m below the in situ soil which is
assumed to contain some clayey soil. The slope of the backfill material is considered to be 1.5H:1V,
Recycling 2017, 2, 8
which is deemed as being a practical inclination value. Groundwater will not be part of the retaining
wall designs and is assumed to be deep below the footing of the retaining wall.
8 of 14
Figure 6. Design model for various backfill mixtures.
As is the typical method among engineers today, the Allowable Stress Design (ASD) method [19]
was used along with Coulomb’s earth pressure theory when calculating the retaining wall
dimensions, considering the passive pressure acting on the toe to be negligible. The Mononobe
Okabe method [20,21] was used for the dynamic condition to determine the seismic active earth
pressure coefficient (Kae ). Factors of safety against overturning failure, sliding failure, and bearing
capacity failure were set to the minimum values for the static and dynamic condition discussed below.
The earthquake used for dynamic case was the 1940 El Centro Earthquake that struck Imperial Valley,
California, which registered a magnitude of 7.0 on the Richter scale and had a peak ground acceleration
(PGA) of 0.33 [22]. As per standard practice, a horizontal seismic coefficient (kh ) of one-half the PGA
was used in the analysis while the vertical seismic coefficient (kv ) was set to be zero.
3.2. Properties of the Composite Material (Sand-Shredded Rubber Mixture) for Design
Values needed for the designing process of the various retaining walls were gathered from the
laboratory tests that were conducted and are presented in Table 1 below. The properties of the in
situ soil are assumed values that are typically used in design, while the values of the 100% sand
mixture are assumed values of sand backfill typically used when compacted to 95% or higher of the
dry unit weight.
Table 1. Properties of in situ and backfill materials used in design.
Material
Unit Weight (kN/m3 )
Friction Angle (deg.)
Cohesion (kPa)
In situ Soil
Standard backfill
Rubber:sand ratio = 100:0
Rubber:sand ratio = 75:25
Rubber:sand ratio = 50:50
Rubber:sand ratio = 25:75
Rubber:sand ratio = 0:100
18.07
18.85
6.39
8.12
10.15
13.38
16.17
28.00
34.00
37.31
38.64
36.33
36.42
31.66
30
0
0
0
0
0
0
3.3. Geotechnical Design of Retaining Wall
The lateral active earth pressures behind the wall were determined by using the dry unit weight
and friction angle results for the various shredded rubber-sand mixtures derived from lab testing
along with the Coulomb lateral earth pressure theory for the static condition and the Mononobe
Okabe method [20,21] for the dynamic condition. Once the earth pressures were established, the ASD
method [19] was used to determine the dimensions of the retaining walls with the various backfill
ratios that could provide the minimum factor of safeties for overturning, sliding, and bearing capacity
failure. This included the minimum factor of safety of 2.0 for the static design with clay type in situ
soils and a minimum factor of safety of 1.10 for the dynamic design as is typically required by engineers
today. The retaining wall dimensions for both the static and dynamic conditions were continually
changed until they met the minimum factors of safety for overturning, sliding, and bearing capacity
Recycling 2017, 2, 8
9 of 14
failures. As presented in Tables 2 and 3 below, the retaining wall dimensions were able to be reduced
to a very small size when more shredded rubber was added to the backfill, while still meeting the
minimum factors of safety presented in Table 4. Furthermore, schematics of the retaining walls for the
static condition illustrated in Figure 7 and for the dynamic condition illustrated in Figure 8, show the
drastic reduction in the dimensions when more shredded rubber was added to the backfill material.
Table 2. Design outcomes for static loading condition.
Material (Rubber:Sand Ratio)
Toe Length (m)
Heel Length (m)
100:0
25:75
50:50
75:25
100:0
3.90
2.33
1.73
1.25
1.15
3.25
1.60
0.91
0.40
0.00
Recycling 2017, 2, 8
9 of 14
Table 2. Design outcomes for static loading condition.
Table
3. Design outcomes
for earthquake
condition.
Material
(Rubber:Sand
Ratio)
Toe Length loading
(m)
Heel Length (m)
100:0
Material 25:75
(Rubber:Sand Ratio)
3.90
Toe Length
2.33 (m)
3.25
Heel Length
(m)
1.60
50:50
1.73
0:100
4.60
1.00 0.91
75:25
1.25
25:75
3.26
0.00 0.40
50:50
2.75
0.00 0.00
100:0
1.15
75:25
2.13
0.00
100:0
1.89 loading condition.
0.00
Table
3. Design outcomes for earthquake
Material (Rubber:Sand Ratio)
Toe Length (m)
Heel Length (m)
of safeties.
0:100 Table 4. Computed factor4.60
1.00
25:75
3.26
0.00
50:50
2.75 of
0.00 of
Material
Factor
Factor
Mode of Failure
(Rubber:Sand Ratio)75:25
Safety-Static
Safety-Earthquake
Load
2.13 Load
0.00
100:0 Overturning
1.89
0.00
3.48
1.89
Sliding
2.00
1.10
0:100
Table
4. Computed factor11.00
of safeties.
Bearing
capacity
6.46
Material (Rubber:Sand Ratio)
25:75
0:100
25:75
50:50
50:50
75:25
75:25
100:0
100:0
(a)
Mode
of Failure
Factor of Safety-Static
Overturning
2.55 Load
Overturning
3.48
Sliding
2.00
Sliding
2.00
Bearing
capacity
9.11
Bearing capacity
11.00
Overturning
2.55
Overturning
2.17
Sliding
2.00
Sliding
2.00
Bearing capacity
9.11
Bearing
capacity
8.15
Overturning
2.17
Overturning
2.00
Sliding
2.00
Bearing
capacity
8.15
Sliding
2.14
Overturning
2.00
Bearing
capacity
7.83
Sliding
2.14
Overturning
2.01
Bearing
capacity
7.83
Sliding
2.13
Overturning
2.01
Sliding
2.13
Bearing
capacity
8.74
Bearing capacity
8.74
(b)
Factor of Safety-Earthquake
Load
1.66
1.89
1.17
1.10
5.36
6.46
1.66
1.67
1.17
1.35
5.36
5.88
1.67
1.67
1.35
5.88
1.59
1.67
6.47
1.59
1.68
6.47
1.75
1.68
1.75
6.93
6.93
(c)
Figure 7. Design outcomes for static condition (a) 100% Sand (b) 50% Rubber: 50% Sand (c) 100%
Figure 7. Rubber
Design(thickness
outcomes
for static condition (a) 100% Sand (b) 50% Rubber: 50% Sand (c) 100%
of base slab and stem = 0.5 m).
Rubber (thickness of base slab and stem = 0.5 m).
Recycling 2017, 2, 8
Recycling 2017, 2, 8
10 of 14
(a)
(b)
10 of 14
(c)
Figure 8. Design outcomes for earthquake loading condition (a) 100% Sand (b) 50% Rubber: 50%
Figure Sand
8. Design
outcomes for earthquake loading condition (a) 100% Sand (b) 50% Rubber: 50% Sand
(c) 100% Rubber (thickness of base slab and stem = 0.5 m).
(c) 100% Rubber (thickness of base slab and stem = 0.5 m).
3.4. Results of the Retaining Wall Designs in the Static and Dynamic Loading Conditions
retaining
wallin
dimensions
both
conditions,
it can be
seen in Figures 7
3.4. Results When
of thecomparing
Retainingthe
Wall
Designs
the Staticfor
and
Dynamic
Loading
Conditions
and 8 that the size of the retaining walls and the excavation volumes decreased as more shredded
When
comparing
dimensions
both in
conditions,
it can
seen in Figures 7
rubber
was added tothe
the retaining
sand. This iswall
the direct
result of thefor
decrease
dry unit weight
andbe
increase
in friction
angle
more
rubber was
mixed
as illustrated
in Figures
3 and 4 as
in more
the
and 8 that
the size
ofasthe
retaining
walls
andwith
thesand,
excavation
volumes
decreased
shredded
testing
discussion.
In the
Coulomb
Theory
and Mononobe
method and increase
rubber laboratory
was added
to the
sand. This
is the
directPressure
result of
the decrease
in dryOkabe
unit weight
[20,21], as the unit weight decreased and the friction angle increased, the lateral active earth
in friction
angle as more rubber was mixed with sand, as illustrated in Figures 3 and 4 in the laboratory
pressures against the wall decreased. Therefore, the lowest earth pressure observed was in the 100%
testing shredded
discussion.
In backfill,
the Coulomb
Theory
Mononobe
Okabe method
rubber
resultingPressure
in the smallest
walland
dimensions
and excavation
volume. [20,21],
When as the unit
to the
wall for
the 100%
sand backfill,
the reduction
of earth
earth pressure
allowed
weight compared
decreased
andretaining
the friction
angle
increased,
the lateral
active
pressures
against the wall
for
the
reinforced
concrete
volume
to
be
largely
reduced.
Additionally,
the
lowered
earth
pressures
decreased. Therefore, the lowest earth pressure observed was in the 100% shredded rubber backfill,
allowed for smaller heel sizes in the retaining walls, resulting in smaller excavation volumes. These
resulting
in the smallest wall dimensions and excavation volume. When compared to the retaining
reductions can be clearly seen in Tables 5 and 6 below when comparing the volumes of reinforced
wall forconcrete
the 100%
sand backfill,
the reduction
ofretaining
earth pressure
allowed
for the
concrete
and backfill
per unit length
for the various
walls. It was
also observed
thatreinforced
for the
condition
once shredded
rubber wasthe
added
to the backfill,
a design where
no heel
used
volumedynamic
to be largely
reduced.
Additionally,
lowered
earth pressures
allowed
foris smaller
heel sizes
becomes possible.
By eliminating
the heel,excavation
the fill volumes
become These
the same
for all fourcan
retaining
in the retaining
walls, resulting
in smaller
volumes.
reductions
be clearly seen in
walls in the dynamic condition using shredded rubber in the backfill. However, when comparing
Tables 5cut
and
6 below when comparing the volumes of reinforced concrete and backfill per unit length
volumes presented in Table 6, it can be seen that they experienced a non-linear decrease from
for the that
various
walls.
It was This
alsoisobserved
thatoffor
dynamic
condition
once shredded
of theretaining
100% backfill
cut volume.
a direct result
thethe
various
shredded
rubber-sand
having
shrinkagea factors,
caused
cutisvolume
fluctuate. possible. By eliminating
rubber mixtures
was added
tovarious
the backfill,
designwhich
where
no the
heel
used to
becomes
the heel, the fill volumes become the same for all four retaining walls in the dynamic condition using
Table 5. Project costs and material volumes for static loading condition.
shredded rubber in the backfill. However, when comparing cut volumes presented in Table 6, it can be
Volume of
Cost of
Amount of Backfill
Cost of Backfill
Cost of Materials
Material
seen that they
experienced
a non-linear
decrease
from
backfill
cutLength
volume. This is a
Concrete
Per Unit
Concrete
Per
Per Unitthat
Lengthof the
Per 100%
Unit Length
Per Unit
(Rubber:Sand Ratio)
Length
(m
)
Unit
Length
($)
(m
)
($)
($)
direct result of the various shredded rubber-sand mixtures having various shrinkage factors, which
0:100
6.83
$839.12
93.08
$1460.95
$2300.07
caused the cut
volume to fluctuate.
25:75
5.22
$641.17
69.97
$3180.36
$3821.53
3
50:50
75:25
100:0
3
4.57
$561.87
62.99
$4737.35
$5299.22
4.08
$501.01
56.74
$6456.14
Table 5. Project
costs and
material volumes
for static $5955.12
loading condition.
3.83
$470.28
54.48
$7339.16
$7809.44
Material
Volume
of costsCost
of Concrete
ofloading Cost
of Backfill
Cost of
Table
6. Project
and material
volumes for Amount
earthquake
condition.
(Rubber:Sand Concrete Per Unit
Per Unit Length
Backfill Per Unit
Per Unit Length
Materials Per
Cost of
Amount of Backfill 3 Cost of Backfill
Ratio) Material Length Volume
(m3 ) of
($)
Length (m )
($) Cost of Materials
Unit Length ($)
(Rubber:Sand Ratio)
0:100
25:75
50:50
75:25
100:0
0:100
25:75
50:50
75:25
100:0
Concrete Per Unit
6.83Length (m3)
5.22 6.05
4.57 4.88
4.63
4.08
4.32
3.83 4.20
Concrete Per
Unit
Length ($)
$839.12
$743.84
$641.17
$599.99
$561.87
$568.64
$501.01
$530.52
$470.28
$515.77
Per Unit Length
(m3)93.08
65.6969.97
50.7062.99
51.92
56.74
51.92
54.48
54.48
Per Unit Length
Per Unit Length
($) $1460.95
($)
$1003.31$3180.36 $1747.14
$2304.29$4737.35 $2904.28
$3904.75
$4473.39
$5955.12
$5449.68
$5980.20
$7339.16
$7339.16
$7854.93
$2300.07
$3821.53
$5299.22
$6456.14
$7809.44
Table 6. Project costs and material volumes for earthquake loading condition.
Material
(Rubber:Sand
Ratio)
Volume of
Concrete Per Unit
Length (m3 )
Cost of Concrete
Per Unit Length
($)
Amount of
Backfill Per Unit
Length (m3 )
Cost of Backfill
Per Unit Length
($)
Cost of
Materials Per
Unit Length ($)
0:100
25:75
50:50
75:25
100:0
6.05
4.88
4.63
4.32
4.20
$743.84
$599.99
$568.64
$530.52
$515.77
65.69
50.70
51.92
51.92
54.48
$1003.31
$2304.29
$3904.75
$5449.68
$7339.16
$1747.14
$2904.28
$4473.39
$5980.20
$7854.93
Recycling 2017, 2, 8
11 of 14
3.5. Discussion on Retaining Walls Backfilled with the Composite Material
From reviewing the retaining wall dimensions for the static and dynamic conditions, it can be
seen that as more EPDM rubber is used, the footing sizes can be reduced while still meeting the
minimum factors of safety. Eventually the heel can be completely taken out to create an “L” shape
retaining wall. However, due to the large stress acting solely on the toe of the retaining wall, larger
quantities of reinforcing steel will be needed. When calculating the retaining wall dimensions, it
was observed that the primary advantage of using this material was that the low unit weight, high
friction angle, and high hydraulic conductivity of the shredded rubber-sand mixtures reduced the
active earth pressure acting behind the wall. As the unit weight decreased, the effective horizontal
stress acting on the wall decreased. Furthermore, the higher friction angle of the shredded rubber
provided larger shear resistance. Since the shredded rubber lowered the horizontal stress acting on
the wall and can handle large shearing forces, the amount of force that was transferred to the wall
in the form of the active earth pressure coefficients Ka and Kae for the static and dynamic condition,
respectively, were lowered. With less pressure on the wall with the increasing amount of EPDM rubber,
smaller dimensions and a reduction in the amount of materials needed to construct the wall can still
provide safe designs. Furthermore, while the hydraulic conductivity of the backfill is not considered
directly in the calculations, it should be noted that with the water being allowed to freely drain, pore
pressures behind the wall cannot build-up. If these pore pressures are allowed to build-up they will
induce unwanted stress on the wall, requiring larger designs.
It was also noticed in the designs for both the static and dynamic condition that as more EPDM
rubber was added to the backfill, the chances for overturning failure increased. This is evident in the
factors of safety presented in Table 4. In the case of the static loading condition, the critical factor of
safety that controls the design changed from sliding to overturning in the 75% rubber: 25% sand backfill
and in the case of dynamic loading condition, the change occurred for backfill with 100% rubber. The
reduced pressures on the wall caused by the shredded rubber equates to a smaller vertical component
of the resultant force (Pa and Pae ) acting on the wall, which decreased resistance to overturning.
The high chance of overturning also resulted in these designs having issues meeting the eccentricity
requirement in the bearing capacity analysis. It was found that the resultant force from the soil acting
on the bottom of the footing was outside the middle third of the footing (eccentricity ≥ width/6) and
was acting too far into the heel. This issue was particularly noticeable for the dynamic condition since
the seismic earth pressure was larger and acted at a higher location on the wall than the lateral active
earth pressure in the static condition. This resulted in a much larger overturning moment than the
static condition. The most economical way to correct these issues for both the static and dynamic
conditions was to extend the toe to the minimum eccentricity requirement (eccentricity = width/6).
It was noticed that even once the eccentricity was set to the lowest possible value, the factors of safety
illustrated in Table 4 for the retaining walls with shredded rubber in the dynamic condition couldn’t
be reduced below the minimum 1.10, which results in overdesign of the retaining walls under the
dynamic loading condition.
4. Special Considerations for Field Application
It should also be noted that if mixtures of shredded rubber and sand were to be used in the field,
care should to be taken to ensure that the material is properly mixed and that high volume change
upon loading does not occur. Segregation of the material was an issue during lab testing that had to
be addressed due to the small sized sand particles falling between the larger sized rubber particles.
A study by Yoon et al. [8] where field studies were conducted for an embankment made of shredded
tires and sand showed that proper mixing of these materials can be done with a front end loader,
which can also be done with mixtures of shredded rubber and sand. Additionally, a study by Edil
and Bosscher [23] has shown that preloading the embankment using a soil cap at least one meter in
thickness can reduce compressibility and potential settlement of the rubber particles from shredded
tires when loaded. The similarities of shredded tires and shredded rubber mean that this measure will
Recycling 2017, 2, 8
12 of 14
be effective for shredded rubber as well. An alternative method for scenarios where they will be mixed
and have a load applied is to have the shredded rubber and sand premixed and placed in bags. If the
sand and shredded rubber are mixed and bagged properly, the segregation and compressibility issues
may be reduced.
5. Cost Analysis
Since smaller lateral active earth pressures occur when shredded rubber is added to the sand in
both the static and dynamic condition, smaller dimensions for the retaining walls are possible that
still meet the design requirements. This will lead to cost savings in the amount of reinforced concrete
needed to build the retaining wall. Based on the cost of concrete per unit length presented in Tables 5
and 6, cost savings can range from 23% to 44% under static loading condition and 19% to 30% under
dynamic loading condition when using shredded rubber-sand mixture compared to conventional
granular soil backfill. Furthermore, the smaller walls will decrease labor costs and increase productivity
saving additional money.
After speaking with recycling facility managers in various cities for the price of shredded
EPDM rubber, it was discovered that rubber shred to the 3 mm size used in this study can range
between $0.10/lbs. to $0.15/lbs. This equates to an average cost of $135m3 ($0.12/lbs.) based on the
un-compacted dry unit weight of EPDM rubber from Figure 3. Compared to the cost of granular fill
of $15.70/m3 that was obtained from local grading contractors, recycled EPDM rubber is expensive
when used as fill. It was also discovered that the recycling facilities can shred the EPDM rubber to
a cheaper price of $0.08/ lbs. ($90/m3 ) if shred to a larger 1–2 in (2.5–5.0 cm) nominal size, which is
still expensive compared to sand. A cost analysis for all five retaining walls based on the volume of
concrete and backfill needed for static and dynamic case are presented in Tables 5 and 6, respectively.
Using the prices above, a local cost of reinforced concrete of $122.95/m3 , and weighted prices for the
three backfill mixtures with EPDM rubber and sand, the final cost per unit length for all five walls
was determined. It can be seen from Tables 5 and 6, that despite the wall size and excavation volume
shrinking considerable, the current cost of recycled EPDM rubber is too large for cost savings to occur.
Prior research by others on shredded tires has shown that using shredded roof membrane as fill
is economical, but that is not the case for shredded rubber. Currently, the cost of shredding tires to a
1–2 inches nominal size is approximately $0.04/lbs., which is half the cost of shredded rubber at that
size. However, the cost of shredded rubber at a 3 mm size does match the cost to shred tires to a 3 mm
size, due to the challenge associated with grinding the rubber to a very small size.
Scrap tires are currently economical, due to the environmental and public health concern
associated, which has caused the government to set-up a “tipping fee” to motivate people to recycle
them. This tipping fee is derived from a tax put on new tires when purchased and is used to pay
whoever is responsible for removing tires from landfills. This covers some of the cost to handle the
tires, thereby decreasing the cost for recycling facilities to buy and sell scrap tires. Presently no tipping
fee exists for roof membrane, which is driving up the price to buy shredded rubber. However, there
are several methods that could be put into practice to lower the price of shredded rubber. Primarily,
the government needs to catch up to the issue of 20 billon square feet of roof membrane coming to
the end of its lifespan and establish a tipping fee to handle this material. Secondly, the technology to
shred roof membrane needs to match the need to dispose of it in large quantities. Since a fair portion
of the cost to recycle roof membrane comes from the tearing off and transportation of the rubber to the
recycling facilities, a mobile unit could be devised to allow for shredding on-site during the roofing
tear-off projects. Applying these methods in practice will allow for the use of shredded rubber in
infrastructure, and will drive the market for recycling this material. While these methods will lower
the cost of shredded rubber, the unit rate costs can still vary. They should be evaluated for each project
specific condition, since availability, local markets, and distance to the recycling facility will affect cost.
Recycling 2017, 2, 8
13 of 14
6. Conclusions
Since past research on shredded tires has shown that using shredded rubber as lightweight fill
can show promising results in decreasing landfill volume and increasing infrastructure sustainability,
other sources of shredded rubber should be studied. By conducting a laboratory testing regiment
on recycled shredded roof membrane, its engineering properties were quantified and a void left in
this area of research was filled. From the test results, it was determined that shredded rubber mixed
with and without sand has similar properties to that of shredded tires. By using these properties to
design retaining walls backfilled with the various ratios of shredded rubber and sand while varying
the loading to include earthquake loading, it is proven that the size of the retaining walls and backfill
volumes can be greatly reduced. Furthermore, they still provide sufficient factor of safety values in
areas that are highly seismic as well as those with low seismic activity. Despite the smaller reinforced
concrete and excavation volumes, the current price of shredded rubber is too large to experience cost
savings when compared to granular fill. The roof membrane lacks of a tipping fee that currently allows
shredded tires to be economical as lightweight fill as well as recycling technology that can process
it cheaply.
As we are approaching the time when 20 billion square feet of roof membrane will reach the end
of its lifespan, there is a need to repurpose this material. The success seen of roof membrane reducing
the size of retaining walls also means that it has the potential to be used in MSE walls. Furthermore,
lab testing results show the potential for use in embankment fills with and without geosynthetics,
in subgrade layers for roadways, and in drainage layers for landfill cover systems. Although since no
such research has been done to test the suitability of shredded rubber-sand mixtures in these systems,
future work including analytical design, full scale or small scale modeling, or simulation using finite
element analysis is needed.
Acknowledgments: The authors would like to thank the reviewers and the editorial team for their constructive
comments and suggestions to improve the quality of the manuscript.
Author Contributions: Bennett Livingston conducted the research presented in this manuscript under the
supervision of Nadarajah Ravichandran.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
Cecich, V.; Gonzales, L.; Hoisaeter, A.; Williams, J.; Reddy, K. Use of shredded tires as lightweight backfill
material for retaining structures. Waste Manag. Res. 1996, 14, 433–451. [CrossRef]
Grimes, B.H.; Steinbeck, S.; Amoozegar, A. Analysis of tire chips as a substitute for stone aggregate in
nitrification trenches of onsite septic systems: Status and notes on the comparative microbiology of tire chip
versus stone aggregate trenches. North Carol. OnSite Wastewater Sect. Dep. Environ. Health (DEH-OSWS) 2003,
4, 18–21.
Hoppe, E.J.; Mullen, W.G. Final Report: Field Study of A Shredded-Tire Embankment in Virginia; Virginia
Transportation Research Council: Charlottesville, VA, USA, 2004.
Ravichandran, N.; Huggins, L. Applicability of shredded tire chips as a lightweight retaining wall backfill
in seismic regions. In Proceedings of the Geo-Congress 2014: Geo-characterization and Modeling for
Sustainability, Atlanta, GA, USA, 23–26 February 2014; Volume 234, pp. 3496–3505.
Reddy, K.R.; Stark, T.D.; Marella, A. Beneficial use of shredded tires as drainage material in cover systems
for abandoned landfills. Pract. Period. Hazard. Toxic Radioact. Waste Manag. 2010, 14, 47–60. [CrossRef]
Kang, J.M.; Lee, J.; Kim, Y.S.; Hong, S.S. Frost heave protection using soil-shredded tire mixture.
In Proceedings of the Fifth Biot Conference on Poromechanics, Vienna, Austria, 10–12 July 2013;
pp. 1314–1319.
Humphrey, D.N. Effectiveness of design guidelines for use of tire derived aggregates as lightweight
embankment fill. In Proceedings of the Recycled materials in geotechnics, Baltimore, MD, USA,
19–21 October 2004; pp. 61–74.
Recycling 2017, 2, 8
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
14 of 14
Yoon, S.; Prezzi, M.; Siddiki, N.Z.; Kim, B. Construction of a test embankment using a sand–tire shred
mixture as fill material. Waste Manag. 2006, 26, 1033–1044. [CrossRef] [PubMed]
Tatlisoz, N.; Edil, T.B.; Benson, C.H. Interaction between reinforcing geosynthetics and soil-tire chip mixtures.
J. Geotech. Geoenviron. Eng. 1998, 124, 1109–1119. [CrossRef]
Russo, M.; Hutchinson, M. EPDM A Roof System for Every Climate. Available online: http://www.
epdmroofs.org/epdm-todays-choice/epdm-roof-for-every-climate (accessed on 11 October 2016).
EPDM Roofing Association. Available online: http://www.epdmroofs.org/epdm-todays-choice/recycling#
sthash.7rZDJn2u.dpufv\T1\textgreater{} (accessed on 14 August 2016).
Moo-Young, H.; Sellasie, K.; Zeroka, D.; Sabnis, G. Physical and chemical properties of recycled tire shreds
for use in construction. J. Environ. Eng. 2003, 129, 921–929. [CrossRef]
Warith, M.A.; Evgin, E.; Benson, P.A.S. Suitability of shredded tires for use in landfill leachate collection
systems mixture as fill material. Waste Manag. 2004, 26, 1033–1044.
Kyser, D.; Ravichandran, N. Properties of chipped rubber roofing membrane and sand mixtures for civil
engineering applications. ASCE J. Build. Eng. 2016, 7, 103–113.
ASTM D422-63.
Standard Test Method for Particle-Size Analysis of Soils; ASTM International:
West Conshohocken, PA, USA, 2007.
ASTM D698. Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Standard Effort; ASTM
International: West Conshohocken, PA, USA, 2012.
ASTM D3080/D3080M-11. Standard Test Method for Direct Shear Test of Soils under Consolidated Drained
Conditions; ASTM International: West Conshohocken, PA, USA, 2011.
AASHTO T 236. Standard Method of Test for Direct Shear Test of Soils under Consolidated Drained Shear Conditions;
American Association of State Highway and Transportation Officials System: Washington, DC, USA, 2014.
Das, B.M. Principles of Foundation Engineering; Global Engineering: Boston, MA, USA, 2014.
Mononobe, N.; Matsuo, H. On the determination of earth pressures during earthquakes. In Proceedings of
the World Engineering Congress, Tokyo, Japan, 30 October–7 November 1929; Volume 9, pp. 179–187.
Okabe, S. General theory of earth pressures. J. Jpn. Soc. Civ. Eng. 1926, 12, 311.
Hays, W.W. Procedures for estimating earthquake ground motions. United States Government Printing
Office: Washington, DC, USA, 1980.
Edil, T.B.; Bosscher, P.J. Engineering proper-ties of tire chips and soil mixtures. Geotech. Test. J. 1994, 17,
453–464.
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).