TECHNICAL NOTE

TECH NOTE NO: 34

TITLE: Fracture and Drying Shrinkage Properties of Concrete Containing
Recycled Concrete Aggregate
AUTHORS: Victor Cervantes, Dr. Jeffery Roesler, and Amanda Bordelon
CONTACT: University of Illinois, Dept. of Civil and Environmental
Engineering, 1211 NCEL, MC-250 Urbana, IL. 61801
Ph: (217) 265 – 0218, e-mail: jroesler@uiuc.edu
DATE: January 9, 2007

1. EXECUTIVE SUMMARY
The objective of this study was to show that using recycled concrete as a coarse
aggregate with and without the addition of synthetic fibers can produce a paving concrete
with similar fracture properties as those made with virgin coarse aggregate. Six different
concrete mixtures were produced. Two mixtures were made completely with virgin
coarse aggregate, two mixtures only used recycled coarse aggregate (RCA), and the last
two were a blend of virgin and recycled coarse aggregate. For each of the mixtures three
point bending, shrinkage, compression, and tensile splitting tests were conducted.

Testing results demonstrated that the use of RCA alone reduced the peak load
capacity of the concrete beam specimen, the total fracture energy of the concrete, and
created greater drying shrinkage. However, when 0.2% volume fraction of fibers were
added the total fracture energy of all the mixtures became similar. The blending of RCA
and virgin coarse aggregate at a 50-50 volume percentage also produced a pavement
concrete with similar fracture and shrinkage properties to that of the virgin coarse
aggregate concrete.
2. INTRODUCTION
Recycling concrete is a viable option to decrease the demand on high quality
natural resources and to limit the amount of waste that is disposed in landfills. Recycled
concrete has been primarily used as a unbound material in embankments, bases, and sub-
bases. Engineers have also used recycled concrete as an aggregate in the construction of
new structures such as concrete pavements but with limited frequency. The use of
recycled concrete in load bearing structures has not gained wide acceptance probably
because of the lack of accessible information on the subject, such as expected fresh and
hardened material properties.

Concrete is not the only recycled material that has been used in previous
construction applications. Recycled asphalt, fly ash, and slag have been used in past
projects [9]. Recycled materials contribute to material sustainability, reduce
environmental impact of demolished materials, and can have positive financial
implications for certain projects. The cost of a project could decrease if concrete does not
have to be hauled and dumped, and instead be used to replace a portion of virgin
aggregate in the new concrete structure. The main objective of this study is to quantify
the fracture and shrinkage properties of one type of recycled coarse aggregate concrete
(RCAC) for rigid pavements. Secondly, increase the fracture properties of the specific
RCAC to obtain equal or greater fracture properties than virgin aggregate concrete
(VAC) containing similar virgin materials.

3. BACKGROUND
3.1 General Information
There is a common reluctance to use recycled concrete as an aggregate in new
concrete due to the limited information on the topic. One of the major issues with the use
of recycled concrete has been the loss of strength which may be attributed to concrete
mixture constituents, RCA blending percentage, water-cement ratios, and aggregate
gradation [6]. A number of different studies have been conducted to analyze each of
these factors and in general the results show a decrease in compression and tensile
strength is expected with RCAC. Other studies have also documented that RCA has a
higher water absorption capacity which causes a higher water demand and leads to issues
like greater drying shrinkage values [15].

A 2006 study focused on the chemical-mineralogical characteristics of RCAC on

the cement hydration for several sources of RCA [10]. A 30% direct replacement of
virgin aggregate with recycled aggregate had no influence on SiO2, Al2O3, or CaO
content. Once the replacement level was above 30%, a marginal decrease in SiO2 and a
marginal increase in Al2O3 and CaO levels were realized.

Another recent study [18] addressed the surface permeability of recycled

concrete, which used both fine and coarse recycled aggregates. The concrete made with
the recycled aggregate had an increased porosity resulting in lower densities and reduced
mechanical properties. Results showed that both the water and air permeability were two
times greater for the concrete made with recycled aggregates as opposed to the virgin
aggregate. This study reported that the main problems of durability come from the use of
recycled fine aggregate.

3.2 Application of RCA Concrete

The use of recycled concrete can be traced back to post-World War II Europe. At
that time, there was a great need for the Europeans to rebuild their countries. Rubble from
damaged buildings and structures was used as an aggregate in new concrete structures
[17]. Once the demand for materials no longer existed, conventional materials for new
construction were utilized for a period of time. In modern day Europe, there has been an
increase in the use of recycled concrete. The government has taken an active role in the
disposal of building materials and has created incentives that encourage the use of
materials like recycled concrete [5].

The state of Illinois also has used recycled concrete in several transportation
projects. In 1986 and 1987, two interstate projects, sponsored by the Illinois Department
of Transportation (IDOT), were constructed with recycled concrete. Each of the projects
was approximately eight centerline miles long and involved rehabilitating an existing
stretch of the interstate. The first project was a continuously reinforced concrete
pavement (CRCP) inlay on I-57 near Effingham, Illinois and the second project was an
asphalt concrete pavement inlay south of Ullin, Illinois also on I-57.

The CRCP project was done to replace a faulted jointed reinforced concrete
pavement that contained high quality aggregates. The concrete was crushed using a jaw
and roll crusher. The new CRCP required a 10-inch concrete surface and a 7-inch cement
stabilized base. The slab geometry also included an 18 inch widened lane. In this project,
both recycled coarse and fine aggregates were used in the concrete mixture. Fly ash and
natural sands were also utilized in order to improve the concrete workability. Periodic
friction tests, ride quality tests, and condition surveys were conducted. A six year
evaluation of the I-57 CRCP inlay found no major problems [13]. The CRCP inlay is
currently 20 years old and still provides a very smooth riding surface. However, this
project is beginning to show some signs of deterioration due to settlement cracking which
has been attributed to the tube feeding process used during construction [12].

3.3 Economy
Approximately 1 million tons of concrete were disposed of in California landfills
in 2003 [3]. Some part of that concrete had the potential of becoming RCA and being
used in new construction projects. The state and federal government and construction
companies would have saved money on hauling and dumping costs. Cost saving would
also have occurred in not having to purchase virgin coarse aggregate for new concrete
construction.

4. RESEARCH OBJECTIVES
The purpose of this research was to determine the fracture properties of a specific
recycled concrete aggregate concrete (RCAC) as compared to concrete with virgin
aggregate concrete (VAC) with similar mixture constituents and properties. The second
objective of the research was to develop a RCAC that had similar or greater fracture
properties than the VAC. Finally, the drying shrinkage characteristics of RCAC were also
measured and documented relative to the plain concrete mixture.

5. METHODOLOGY
5.1 Materials
The first phase of this study looked at the effect recycled coarse aggregate with a
high volume fraction of synthetic fibers would have on the fracture properties of paving
concrete [6]. In this phase, the recycled concrete was crushed in the University of Illinois
laboratory using a jack hammer and a small lab crusher. For this phase only three-point
bend (TPB) specimens were cast. The second phase of the study utilized recycled
concrete that was crushed by a local construction company, University Construction. The
main purpose of the second phase testing was to produce RCAC that was economically
feasible while still obtaining fracture properties that exceed the VAC. For the second
phase, TPB, compression, splitting tension, and drying shrinkage specimens were cast.

Figures 1a and 1b are photos of the concrete slabs that were crushed. These slabs
had been cast and tested for a concrete fatigue study and as such, the fresh and hardened
properties of this concrete were known [19]. The concrete had a compressive strength of
58.3 MPa, a split tensile strength of 4.5 MPa, and a modulus of elasticity of 32.0 GPa at
an age of one year. Figure 1c is an image of the recycled coarse aggregate that was
produced after crushing the concrete slabs with the Hazemag 1515 Impact crusher at
University Construction. The RCA was not completely free of debris as seen in Figure
1d. The RCA contained small amounts of asphalt, brick, and wood. Since this company
accepts and crushes all types of urban building materials, some contamination was
expected.

(a) (b)

(c) (d)

Figure 1: (a) and (b) photos of the concrete slabs prior to being crushed;
(c) Image of the crushed RCA; (d) debris in the RCA
5.2 Coarse Aggregate Gradation
Prior to the mix design being generated, the recycled coarse aggregate had to be
sieved and graded to ensure a quality mixture. The grading process took place during
each of the project phases after the old concrete slabs were crushed. Previous publications
suggest that the gradation of RCA be the same as that of the virgin aggregate [16].
During the first phase, the grading curve that was achieved with the laboratory crusher is
seen in Figure 2. In order to reduce the RCA’s water demand, all particles less than the
#4 sieve were not used. For the second phase, the RCA was sieved so that it would be
within the minimum and maximum limits described in two specifications pertinent to the
study, the FAA 25mm Coarse Aggregate requirement (P-501) and the IDOT CA-7
requirement (see Figures 3a and 3b respectively). The construction company’s operation
produced a coarser aggregate gradation for the RCA that was not ideal for coarse
aggregate to be used in concrete mixtures. However, in future work a more optimal
combined aggregate (RCA and virgin aggregate) gradation could be developed and
produced by the same crusher, if necessary.

Gradation - Maximum size 25 mm

Cumulative Amount Passing

100

80

60 RCA
(%)

Min
40 Max
20

0
1 10 100
Sieve Size (mm)

Figure 2: Recycled coarse aggregate gradation curve for Phase 1 testing

FAA 25mm Coarse Aggregate

100
Cumulative Amount

80
Passing (%)

RCA
60
Min
40
Max
20

0
1 10 100
Sieve Size (mm)

Figure 3a: Recycled coarse aggregate gradation curve for Phase 2 (FAA limits for 25mm
nominal maximum size aggregate)
 IDOT- CA 7

Cumulative Amount Passing

100

80

60 RCA
(%)
Min
40 Max
20

0
1 10 100
Sieve Size (mm)

Figure 3b: Recycled coarse aggregate gradation curve for Phase 2 (IDOT CA-7 limits for
25mm nominal maximum size aggregate)

5.3 Design Mixtures

In the first phase of the research, four different concrete mixtures were cast. The
first concrete mixture had 100% virgin coarse aggregate. The second mixture included
the 100% virgin coarse aggregate with the addition of 12.1 lb/yd3 of synthetic fibers. The
third mixture replaced the virgin coarse aggregate with 100% recycled coarse aggregate.
The final mixture had 100% recycled coarse aggregate and 12.1 lb/yd3 of synthetic fibers.
Table 1 lists the mix proportions for the Phase 1 mixtures [6]. Note, the mixture
proportions in Phase 1 were not adjusted for the different coarse aggregate specific
gravities. Table 2 shows the physical properties of the fine and coarse aggregates. The
bulk specific gravity at SSD for the recycled coarse aggregate was lower than that of the
virgin aggregate. The absorption capacity for the RCA was approximately 2.5 times
greater than that of the virgin aggregate.

Table 1: Concrete mixture design for Phase 1 testing

Synthetic Fiber
Plain Concrete Reinforced
Concrete
Mix ID PCC FRC
3 3 3
Material kg/m lb/yd kg/m Lb/yd3
Water 183 308 183 308
Type I Cement 360 607 360 607
Coarse aggregate 976 1645 976 1645
Fine aggregate 807 1360 80 1360
Synthetic Fibers --- --- 7.2 12.1
 Table 2: Fine and coarse aggregate properties
Absorption
BSGssd
Capacity
RCA - coarse 2.42 5.27%
Virgin Coarse
Aggregate 2.64 2.01%
Virgin Fine
Aggregate 2.51 1.79%

The second phase of testing focused on creating a cost effective RCAC mixture
that would achieve similar fracture properties to concrete with virgin aggregate. Table 3
shows all six concrete mixtures. The first two mixtures are with virgin aggregate. These
mixtures remained the same as in Phase 1 except that 3 lbs/yd3 of synthetic fibers were
used instead of 12.1 lbs/yd3 in the FRC mixture, which equals to a fiber amount of 0.19%
and 0.80% by volume respectively. For the third and fourth mixtures, RCA proportions
were adjusted to account for the lower bulk specific gravity of the RCA relative to the
virgin aggregate mixtures. The fiber content was also 3 lbs/yd3, which is the minimum
manufacturer recommendation for this fiber type. The last two mixtures are with a
blended coarse aggregate mixture (by volume), i.e. 50% virgin coarse aggregate and 50%
RCA with fibers (50-50 Blend FRC), and without fibers (50-50 Blend). In order to
reduce variability in water demand and concrete workability, all mixtures containing
RCA were presoaked over night to ensure saturation during batching [4]. Excess water
above SSD condition, typically 1 to 2 percent, was measured and accounted for in the
batch weights.

Table 3: Concrete mixture designs for Phase 2 testing (lb/yd3)

VAC VAC RCAC RCAC 50-50 50-50
Mixture FRC FRC FRC
Water 308 308 308 308 308 308
Type I Cement 607 607 607 607 607 607
RCA 0 0 1508 1508 754 754
Coarse
aggregate
(Virgin) 1645 1645 0 0 823 823
Fine aggregate 1360 1360 1360 1360 1360 1360
Synthetic
Macrofibers 0 3 0 3 0 3

Two concrete beams, six cylinders and three shrinkage prisms were cast for each
mix designs to determine the concrete’s fracture properties, split tensile strength,
compressive strength, and free shrinkage. After 24 hours, the specimens were de-molded.
Cylinders and beams were placed in a lime water bath for seven days before they were
tested. Shrinkage samples, 75mm x 75mm x 285mm, were measured and held in a
climate controlled room with 23ºC and 50% relative humidity. The free shrinkage tests
were measured using a modified ASTM C157 [22] procedure where specimens are only
cured one day prior to drying. The compression and tensile splitting tests were conducted
in accordance to ASTM C 39 [1] and ASTM C 496 [2], respectively, using cylindrical
samples with a four inch diameter and a height of eight inches.

5.4 Fracture Properties

The strength of concrete can be a good indicator of the concrete’s quality and
potential performance. However, measuring only the compressive and tensile strength of
concrete will not guarantee the performance of the concrete structure due to the
interaction of the material behavior with the geometry of the structure. For example,
concrete of various qualities can be made to have a similar tensile strength but produce
varying degrees of structural performance. The fracture properties of the concrete can
provide more description of the maximum load capacity of the material and the crack
propagation resistance or toughness of the material. The critical stress intensity factor
(KIC) and critical crack tip opening displacement (CTODc) describe the initial cracking
properties of the concrete while the total fracture energy (GF) characterizes the crack
propagation and load carrying capacity of the partial cracked concrete.

The three-point bend (TPB) samples were used in conjunction with the two
parameter fracture model [8] to obtain the concretes’ initial fracture properties. Figure 4
presents the TPB sample geometry and loading configuration that was used for
calculating the two parameter fracture model parameters.

b = 150 mm

a0 =
50mm
H0

CMOD
t=
80 mm
S = 600 mm

L = 700 mm

Figure 4: Schematic representation of the three-point bend experiment

In order to calculate the fracture properties of the concrete, the load (P) versus
crack mouth open displacement (CMOD) must be measured and plotted. Details on the
testing procedure can be found in [8]. The specims were loaded and unloaded to
propagate the crack and to determine the fracture properties. Figure 5 is a schematic
representation of a P versus CMOD for one load-unload curve.
 P

Pc

Ci
Cu
1 Pc
2

CMODPc CMODec CMOD

Figure 5: Schematic representation of the two-parameter method

The P-CMOD plot was then used to determine the peak load (Pc) as well as computing
the loading and unloading compliances (Ci and Cu, respectively). The compliances are
can then be used in equations 1 and 2 to estimate the elastic modulus of the material, E1
and E2.

6 Sa 0 g 2 (α 0 )
E1 = (1)
Ci b 2 t

6Sac g 2 (α c )
E2 = (2)
CU b 2 t
where
(a + H 0)
α0 = 0 (3)
(b + H 0)

(ac + H 0)
αc = (4)
(b + H 0)
0.66
g 2 (α ) = 0.76 − 2.28α + 3.87α 2 − 2.04α 3 + (5)
(1 − α ) 2

Since the elastic modulus of the bulk concrete does not change, equations 1 and 2 can be
set equal to each other to calculate the critical elastic crack length, ac. The critical crack
length is then used to determine the KIC, from equation 6 below:

S πa c g1 (a c / b )
K IC = 3(Pc × 0.5Wh ) (6)
2b 2 t
where S, b, HO, a0 and t are shown in Figure 4, Pc is the peak load, Wh is equal to the
beam weight multiplied by S/L, and g1(ac/b) is the geometric correction factor determined
by equation 7.

g1 ⎜ c ⎟ =
[
⎛ a ⎞ 1.99 − (aC / b )(1 − a c / b ) 2.15 − 3.93ac / b + 2.70(ac / b )
2
] (7)
⎝ b⎠ π (1 + 2a / b)(1 − ac / b )3 / 2

The CTODc was calculated using equation 8, where β0 is equal to a0/ac [13].

6(Pc + 0.5Wh )Sa c g 2 (a c / b) ⎡ ⎤

CTODC = 2
⎛ a ⎞
( )
. ⎢(1 − β 0 ) + ⎜1.081 − 1.49 c ⎟ β 0 − β 02 ⎥ (8)
2

Eb t ⎣ ⎝ b ⎠ ⎦

The initial fracture energy release rate GIC was calculated using equation 9, where E is
concrete’s elastic modulus, which was determined by the beam’s geometry and the
compliance in equations 1 and 2.

G IC =
(K )
IC
2

(9)
E

The total fracture energy, GF, was calculated using equation 10.

W0 + mgδ 0
GF = (10)
(b − a0 )t
where W0 is the area under the P versus CMOD curve, mg is the self weight of the beam,
δ0 is the value of deflection when the descending branch of the softening curve goes to
zero [8].

6. RESULTS AND DISCUSSION

6.1 Phase 1
Three beams for each mixture were cast. Figure 6 is the comparison of the
average softening curves for the concrete containing virgin coarse aggregate relative to
the RCA. Table 4 shows the fracture parameters collected for this case. The peak load of
the virgin aggregate mixture was 38 percent greater than the RCA mixture. The CTODc
at failure for the virgin aggregate mixture was less than the RCA mixture. The total
fracture energy (GF) for the virgin aggregate beams was approximately two times greater
than the RCA beams while the fracture toughness was only 20 percent greater. The
elastic modulus of the RCA mixture was lower than the virgin aggregate mixture.

Table 4: Fracture properties plain concrete with virgin or RCA (Phase 1)

Peak Load E KIC CTODc GF

(kN) (GPa) (MPa-m1/2) (mm) (N/m)
Beam 1 3.76 27.6 1.17 0.0152 127
Beam 2 3.95 32.1 1.52 0.0267 113
VAC
Beam 3 3.42 27.4 1.07 0.0138 115
Average 3.71 29.0 1.25 0.0186 119

Beam 1 3.05 26.6 1.13 0.0212 64

Beam 2 2.75 22.0 1.13 0.0289 46
RCAC
Beam 3 2.91 11.0 0.94 0.0300 74
Average 2.90 19.9 1.07 0.0267 61

4
3.5
3 Virgin
Load (kN)

2.5 RCA
2
1.5
1
0.5
0
0 0.2 0.4 0.6 0.8 1
CMOD (mm)

Figure 6: Comparison of TPB softening curves for concrete mixtures containing virgin or
RCA (Phase 1)

As part of the first phase, 12.1 lb/yd3 were added to each of the concrete mixtures.
Figure 7 shows the P versus CMOD curves of the virgin aggregate and the RCA mixtures
with fibers. Table 5 presents the fracture parameters based on the FRC mixtures of Phase
1. As expected, the addition of fibers increased the total fracture energy for both the
virgin aggregate and RCA mixtures. The virgin aggregate mixture with fibers had a total
fracture energy up to 2mm CMOD that was approximately 36 percent larger than the
RCA mixture with fibers. The peak load was 38 percent greater for the virgin aggregate
mixture with fibers, which was the main contributor in the total fracture energy difference
between the two mixtures. The residual load capacity, defined as the load when the
CMOD equals 2mm, was almost the same for the VAC with fibers and RCAC with
fibers. The fracture toughness for the two mixtures with fibers was also very similar. The
calculated elastic modulus for the recycled coarse aggregate FRC is lower and the
CTODc is higher than the virgin aggregate FRC mixture.
 Table 5: Fracture properties of Phase 1 virgin aggregate beams
1
Peak Load E KIC CTODc GF
(kN) (GPa) (MPa-m1/2) (mm) (N/m)
Beam 1 3.79 28.6 1.33 0.0223 357
VAC Beam 2 4.42 28.4 1.38 0.0188 460
FRC Beam 3 3.28 27.7 1.05 0.0140 385
Average 3.83 28.3 1.25 0.0184 401

Beam 1 2.63 27.2 1.23 0.0405 236

RCAC Beam 2 2.63 24.7 1.33 0.0501 303
FRC Beam 3 3.01 22.3 1.33 0.0434 345
Average 2.76 24.7 1.30 0.0447 295
1
up to 2mm CMOD

4
3.5
3
Virgin FRC
Load (kN)

2.5 RCA FRC

2
1.5
1
0.5
0
0 0.5 1 1.5 2
CMOD (mm)

Figure 7: Load versus CMOD curves of virgin aggregate and RCA mixtures with fibers
(Phase 1)

6.2 Phase 2
6.2.1 Compressive and Tensile Splitting Results
Compressive and splitting tensile tests were conducted for all mixtures in Phase 2.
Three compression and three tensile splitting samples were cast for each of the six
mixtures. The average 7-day strengths are reported in Table 6. The compressive strengths
ranged from 22.9 MPa for the 50-50 Blend to 31.2 MPa for the virgin coarse aggregate
concrete. There were no consistent trends between the split tensile and compressive
strength for the virgin aggregate, RCA, and blended aggregate concrete mixtures with or
without fibers.
 Table 6: Average strength results of Phase 2 samples
Mixture Compressive Strength Tensile Strength
Type (MPa) (psi) (MPa) (psi)
VAC 31.2 4528 2.61 378
VAC FRC 30.3 4396 2.93 425
RCA 27.8 4030 2.45 356
RCA FRC 23.8 3450 2.86 415
50-50 Blend 22.9 3328 2.84 412
50-50 Blend FRC 24.4 3539 2.63 382

6.2.2 Fracture Properties Results

The Phase 2 fracture testing used the TPB specimen to determine the initial
fracture parameters and the total fracture energy. Figure 8a shows the experimental set
up of TPB specimen for Phase 2 tests. Figure 8b shows how the synthetic fibers bridge
the crack at high levels of deformation. Two beams were cast for each mixture. Figure 9
shows the average load-deformation curve between the virgin aggregate samples with 3
lbs/yd3 of fibers and without fiber reinforcement.

Figure 8a and 8b: Experimental set up of three point bending test and synthetic
fibers bridging the crack

As expected, the peak load between the VAC with and without fibers are very
similar. Table 7 lists the fracture properties of the virgin aggregate beams. The total
fracture energy up to 4mm for the beams with the fibers is about 3.2 times greater that
that of the beams without fibers for at a CMOD of 4mm. The GF at a CMOD of 2mm of
the VAC beams with 3 lbs/yd3 is 64% less than the beams with 12.1 lbs/yd3 of fibers.
Even with 3 lbs/yd3, the fiber-reinforced VAC was able to support loads at crack mouth
openings greater than 1mm.
 Table 7: Fracture properties of virgin coarse aggregate beams (Phase 2)
1 2
Mixture Peak Load E KIC CTODc GF GF
Type (kN) (GPa) (MPa-m1/2) (mm) (N/m) (N/m)
Beam 1 2.92 27.2 1.06 0.0182 - 63
VAC Beam 2 3.57 24.7 1.18 0.0195 - 83
Average 3.25 26.0 1.12 0.0189 - 73

Beam 1 3.68 26.81 1.35 0.0262 153 254

VAC
FRC Beam 2 3.00 25.25 1.24 0.0292 133 217
Average 3.34 26.03 1.30 0.0277 143 236
1
up to 2mm CMOD; 2up to 4mm CMOD

3.5

3
2.5 No FRC
Load (kN)

FRC
2
1.5
1
0.5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
CMOD (mm)

Figure 9: Load versus CMOD curves for virgin coarse aggregate concrete with 3 lbs/yd3
of fibers and without fibers (Phase 2).

Figure 10 shows the load versus CMOD of the RCAC mixtures with and without
fibers. Similar to the virgin coarse aggregate mixtures, there was little difference in the
peak load capacity of RCAC with and without fibers. There was an even larger difference
in the total fracture energy between the beams with and without fiber reinforcement as
seen in Table 8. The RCAC with fibers had 5 times greater total fracture energy at 4mm
than RCAC without fibers. Furthermore, the total fracture energy between the RCAC and
VAC with fibers was approximately despite a 38% reduction in the fracture energy from
the VAC to RCAC without fibers (73 to 45 N/m).
 Table 8: Fracture properties of RCAC (Phase 2)
1 2
Mixture Peak Load E KIC CTODc GF GF
Type (kN) (GPa) (MPa-m1/2) (mm) (N/m) (N/m)
Beam 1 2.95 30.1 1.13 0.0196 - 40
RCAC Beam 2 3.01 25.8 1.06 0.0186 - 49
Average 2.98 28.0 1.09 0.0191 - 45

Beam 1 3.06 28.4 1.12 0.0193 165 278

RCAC
FRC Beam 2 3.20 28.0 1.13 0.0192 118 165
Average 3.13 28.2 1.12 0.0192 142 222
1
up to 2mm CMOD; 2up to 4mm CMOD

3.5

2.5
No FRC
2
Load (kN)

FRC
1.5

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4
-0.5
CMOD (m m )

Figure 10: RCAC beams fracture behavior with and without 3 lbs/yd3 of fiber
reinforcement (Phase 2)

The third case in Phase 2 involved concrete mixtures with blended coarse
aggregate, i.e., 50% virgin coarse aggregate and 50% RCA, with and without fibers. Two
beams of each mixture were cast. As seen in Figure 11, the peak load was approximately
the same for the plain and FRC blended mixtures. Table 9 shows the fracture properties
for the blended concrete specimens. The blended FRC beams had a total fracture energy
at 4 mm that was 3.3 times greater than plain 50-50 blended aggregate beams. The
blended coarse aggregate mixture without fibers had a similar fracture energy to the VAC
without fibers. The total fracture energy at 4 mm CMOD of the blended FRC was
approximately the same as the RCAC and VAC with fibers (3 lbs/yd3).
 Table 9: Fracture properties of the 50-50 blend of virgin coarse aggregate and RCA
1 2
Peak Load E KIC CTODc GF GF
(kN) (GPa) (MPa-m1/2) (mm) (N/m) (N/m)
Beam 1 2.68 27.2 0.97 0.0162 - 76
50-50
Beam 2 3.09 22.3 1.09 0.0226 - 63
Blend
Average 2.89 24.7 1.03 0.0194 - 69

50-50 Beam 1 2.78 27.2 0.84 0.0093 109 154

Blend
Beam 2 2.84 24.7 0.95 0.0179 119 302
FRC
Average 2.81 26.0 0.90 0.0136 114 228
1
up to 2mm CMOD; 2up to 4mm CMOD

3.5

2.5 No FRC
Load (kN)

FRC
2

1.5

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4
CMOD (m m )

Figure 11: Load-deformation of concrete mixture with blended coarse aggregate, 50-50
blend, with 3lbs/yd3 and without fibers.

6.2.3 Discussion of Fracture Property Results

The results of Phase 2 testing are summarized in Table 10 to allow for fracture
property comparison in terms of coarse aggregate type and use of fiber reinforcement.
Figure 12 is a summary of Load-CMOD results for the virgin aggregate, RCA, and
blended coarse aggregate mixtures. For the plain concrete (unreinforced), the virgin
coarse aggregate produced the highest peak loads. (Note, when averaging Load-CMOD
curves between specimens, the average plot does not show the average peak load
calculated from Table 10 since the peak loads from different specimens occur at different
CMOD levels.) The total fracture energy is higher for the concretes containing virgin
coarse aggregate and the blended coarse aggregate relative to the RCA. This behavioral
difference can also be visualized in Figure 12 as the RCAC has a lower post-peak curve.
For this coarse aggregate type and RCA properties, the fracture properties of a concrete
mixture with RCA and no fibers can be significantly improved by blending RCA with a
virgin aggregate (e.g. 50-50 blend per this study).
 Table 10: Summary of concrete fracture properties of virgin, RCA, and blended coarse
aggregate mixtures without fibers
2
Peak Load E KIC CTODc GF
(kN) (GPa) (MPa-m1/2) (mm) (N/m)
Beam 1 2.92 27.2 1.06 0.0182 63
VAC Beam 2 3.57 24.7 1.18 0.0195 83
Average 3.25 26.0 1.12 0.0189 73

Beam 1 2.95 30.1 1.13 0.0196 40

RCAC Beam 2 3.01 25.8 1.06 0.0186 49
Average 2.98 28.0 1.09 0.0191 45

Beam 1 2.68 27.2 0.97 0.0162 76

50-50
Blend Beam 2 3.09 22.3 1.09 0.0226 63
Average 2.89 24.7 1.03 0.0194 69
2
up to 4mm CMOD

3.5

3
VAC
2.5
RCAC
2
Load (kN)

50-50 Blend
1.5

1
0.5

0
-0.5 0 0.2 0.4 0.6 0.8 1

CMOD (m m )

Figure 12: Average load-deformation curves for plain concrete mixtures containing virgin
aggregate, RCA, and blended RCA-virgin coarse aggregates

Phases 1 and 2 testing results for 12 lb/yd3 and 3 lb/yd3 respectively are displayed
in Table 11 for comparison of fiber volume and coarse aggregate type. As seen in Table
11 and Figure 13, the fracture behavior of RCAC can be made similar to virgin coarse
aggregate or a 50-50 blended mixture if only 3 lbs/yd3 of synthetic fibers are added. The
peak load capacity of the virgin concrete and blended aggregate mixtures were higher
than the RCA with fibers but all show similar post-peak softening curves. One other point
is that the fracture properties of RCAC with fibers exceed the fracture properties of non-
reinforced concrete with virgin aggregate, RCA, or blended coarse aggregate mixtures.
Table 11: Summary of fracture properties of virgin and RCAC with 12 lbs/yd3 (Phase 1)
and 3 lbs/yd3 (Phase 2) of fiber reinforcement
Phase 1 Phase 2 Phase 2
1 1 2
Peak Load E KIC CTODc GF GF GF
(kN) (GPa) (MPa-me) (mm) (N/m) (N/m) (N/m)
Beam 1 3.68 26.8 1.35 0.0262 357 153 254
VAC Beam 2 3.00 25.2 1.24 0.0292 460 133 217
FRC Beam3 - - - - 385 - -
Ave. 3.34 26.0 1.30 0.0277 401 143 236

Beam 1 3.06 28.4 1.12 0.0193 236 165 278

RCAC Beam 2 3.20 28.0 1.13 0.0192 303 118 165
FRC Beam 3 - - - - 345 - -
Ave. 3.13 28.2 1.12 0.0192 295 142 222

50-50 Beam 1 2.78 27.2 0.84 0.0093 - 109 154

Blend Beam 2 2.84 24.7 0.95 0.0179 - 119 302
FRC Ave 2.81 26.0 0.90 0.0136 - 114 228
1
up to 2mm CMOD; 2up to 4mm CMOD

3.5
3 VAC
2.5 RCAC
Load (kN)

2 50-50 Blend
1.5
1
0.5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
CMOD (mm)

Figure 13: Effect of RCA, virgin, and blended coarse aggregate on the fracture behavior of
3 lb/yd3 fiber reinforced concrete mixtures

Due to the ductility and bridging capabilities of fibers, the 4 mm range of the
extensometer used for the CMOD measurement required modification. A second device to
measure the CMOD to larger openings, called a yoyo gauge, is shown in Figure 8. This
gage allowed calculation of the concrete fracture energy up to 20 mm range. This was able
to capture the total energy of the FRC mixtures when the post-peak curve was close to zero
load capacity. Figure 14 shows the softening curve of one of the 50-50 Blend FRC beams.
This particular beam was able to hold load of 0.15 kN at a 16 mm CMOD. This increased
the total fracture energy of the concrete beam from 228 N/m at 4mm CMOD to 517 N/m at
16 mm CMOD.
50-50 Blend FRC Beam

2.5

2
Load (kN)

1.5

0.5

0
0 2 4 6 8 10 12 14 16
CMOD (mm)

Figure 14: Fracture behavior of concrete mixture with blended RCA-virgin coarse
aggregate up to 16mm CMOD.

6.2.4 Drying Shrinkage Results

Drying shrinkage and mass loss of the Phase 2 concrete mixtures was examined
according to ASTM C157 [22] in a 23°C and 50% relative humidity chamber. One
modification from this test standard was the initial drying began one day after casting.
Significantly higher drying shrinkage of RCAC (10 to 50%) has been reported in the
literature [7, 11, 14]. Table 12 and Figure 15 present the average free shrinkage strains
based on three shrinkage prisms (75x75x285mm). During the first 7 days, the free drying
shrinkage was approximately the same for all samples. By 28 days, the RCA samples
had the highest shrinkage magnitude. The free shrinkage of the virgin and blended
coarse aggregate was very similar at 28 days. The RCAC had approximately 28 percent
greater free shrinkage at 28 days than the VAC. The addition of fibers did not
significantly alter the shrinkage values found in the plain concrete mixtures. A recent
study by Roesler et al. [20] on airport concrete paving mixtures found the drying
shrinkage of four different concrete mixtures varied from 335 to 417 microstrain at 28
days. The main difference between this study and the referenced study was the water
cement ratio and coarse aggregate types used.
 DRYING SHRINKAGE
75x75x285 mm specimen
800

700

Average Dry Shrinkage (microstrain)

600

500

400

300
Virgin FRC Virgin Agg.
200
RCA FRC RCA
100
50% Vir 50% RCA 50% Vir 50% RCA FRC

0
0 5 10 15 20 25 30 35 40 45 50
Concrete Age (days)

Figure 8: Free drying shrinkage concrete mixtures containing virgin, RCA, and blended
coarse aggregate with and without fiber reinforcement

Table 12: Free drying shrinkage of concrete (Phase 2)

Day 1 3 7 14 28 43
Average
Microstrain 0.00 210 325 427 505 562
VAC
Weight loss (%) 0.00 2.25 3.23 3.46 3.68 3.87
Weight loss (g) 0.00 89.6 128.5 138 147 153
Average
VAC Microstrain 0.00 185 321.7 430 523 555
FRC Weight loss (%) 0.00 2.35 3.25 3.56 3.79 3.97
Weight loss (g) 0.00 91.6 126 139 148 155
Average
Microstrain 0.00 170 337 452 645 720
RCAC
Weight loss (%) 0.00 2.41 2.93 3.26 3.73 3.90
Weight loss (g) 0.00 90.6 110 122 140 147
Average
RCAC Microstrain 0.00 117 277 403 608 648
FRC Weight loss (%) 0.00 2.75 3.27 3.59 4.05 4.22
Weight loss (g) 0.00 100 119 131 148 154
Average
50% Virgin Microstrain 0.00 128 268 352 518 552
50% RCA Weight loss (%) 0.00 3.17 3.84 4.03 4.42 4.53
Weight loss (g) 0.00 121 147 154 169 174
Average
50% Virgin Microstrain 0.00 113 300 378 535 580
50% RCA
Weight loss (%) 0.00 2.86 3.47 3.66 4.03 4.17
FRC
Weight loss (g) 0.00 112 136 143 157 163
7. CONCLUSION
The incorporation of recycled concrete as a coarse aggregate in concrete mixtures
has been successfully used in the state of Illinois and can offer economic benefits for new
concrete pavement construction. During the first phase of this research, the fracture
properties of concrete containing 100% recycled coarse aggregate was shown to be less
than virgin aggregate concrete (VAC), but with the addition of a high volume of fibers
(0.8%), the fracture properties of recycled concrete aggregate concrete (RCAC) exceeded
the plain concrete with virgin aggregates and was similar to the fiber reinforced concrete
with virgin aggregates. The second phase of this research demonstrated that a 50-50 blend
of virgin and recycled coarse aggregate could give similar fracture properties to 100%
virgin coarse aggregate. Furthermore, a cost effective addition of fibers (0.2%) to RCAC
could also exceed the fracture properties of virgin and blended coarse aggregate concrete.
The peak loads of VAC were always greater than RCAC and blended coarse aggregate
concrete even with fibers. Overall, the RCA with fiber-reinforcement showed similar
fracture properties as the virgin aggregate and blended aggregate fiber-reinforcement
mixtures.

Measurement of the free drying shrinkage of prism samples showed that the RCAC
without the fibers had the highest shrinkage strain at 28 and 43 days followed by the RCA
concrete with the fibers and finally the virgin aggregate concrete mixtures. The drying
shrinkage values obtained from this testing were higher than previous concrete paving
mixtures due to the increased cement content, higher water to cement ratio and RCA. The
increased shrinkage of the RCA may be compensated in future mixture designs by reducing
the water to cement ratio.

8. REFERENCES
[1] ASTM C 39. Standard Test Method for Compressive Strength of Cylindrical
Concrete Specimens. ASTM International (2001).
[2] ASTM C 496. Standard Test Method for Splitting Tensile Strength of Cylindrical
Concrete Specimens. ASTM International (1996).
[3] Contractor’s Report to the Board, Statewide Waste Characterization Study, California
December 2004. pg. 14

[4] ENV 206:1992. Concrete. Performance, production, placing and compliance criteria.

[5] Schimmoller, V. E., Holtz, K., Dr. Eighmy, T. T., Wiles, C., l Smith, M.., Malasheskie,
M., Rohrbach, G.J., Schaftlein, S., Helms, G., Campbell, R. D., Van, C. H., Deusen,, B. F.,
Almborg J. A., “Recycled Materials in European Highway Environments: Uses,
Technologies, and Policies” FHWA International Practices Report, October 2000

[6] Figueroa, H. D., Fracture Energy of Plain and Reinforced Recycled Concrete Pavement,
Pavement Design, II Class Project, 2006
[7] Hansen, T. C., and E. Boegh. Elasticity and Drying Shrinkage of Recycled-Aggregate
Concrete, ACI Journal, Vol. 82, No. 5, 1985, pp. 648-652.

[8] Jenq, Y. and S.P. Shah. Two Parameter Model for Concrete. J. Engng Mech. 111 (10),
1227-1241 (1985)

[9] Mindess, S. Young, F. and Darwin, D., Concrete Second Edition, Upper Saddle River,
NJ, Prentice Hall, 2003

[10] Limbachiya, M.C., Marrocchino, E., Koulouris A., “Chemical–mineralogical

characterisation of coarse recycled concrete aggregate” Waste Management, 2006, pp 1-
8.

[11] Park, S., Sim. J., “Fundamental Properties of Concrete using Recycled Concrete
Aggregate produced through Advanced Recycling Process” TRD 2006 Annual Meeting,
2006

[12] Roesler et al. 2005-Asce, Journal of Performance of Constructed Facilities

[13] Schutzbach, A. M., Recycling Old Portland Concrete Cement – Performance

Evaluation of FAI 57 Inlays, Illinois Department of Transportation. February 1993

[14] Shah, S.P., Ouyang, C., Swarts, S.E., Fracture Mechanics of Concrete, John Wiley
and Sons, New York, 1995

[15] Shayan, A., Xu, A., “Performance and Properties of Structural Concrete Made with
Recycled Concrete Aggregate” ACI Materials Journal, September/October 2003 pp. 371-
380

[16] Tam, V. W. Y., Tam, C.M., “Crushed aggregate production from centralized
combined and individual waste sources in Hong Kong” Construction and Building
Materials, 2006 pp. 1-8

[17] Yranjson, W. A., “Recycling Portland Concrete Cement” American Concrete

Pavement Association, 1981 pp. 431-444

[18] Zaharieva, R., Buyle-Bodin, F., Skoczylas, F., Wirquin, E., “Assessment of the surface
permeation properties of recycled aggregate concrete” Cement and Concrete Composites.
2003 pp. 223-232

[19] Roesler, J.R., Littleton, P.C., Hiller, J.E., Long, G.E., Effect of Stress State on
Concrete Slab Fatigue Resistance, Draft Final Report for Federal Aviation
Administration, University of Illinois, Urbana, IL, October 2004, 227 pp.
[20] Roesler, J., Gaedicke, C., Lange, Villalobos, S., Rodden, R., and Grasley, Z. (2006),
“Mechanical Properties of Concrete Pavement Mixtures with Larger Size Coarse
Aggregate,” ASCE 2006 Airfield and Highway Pavement Conference, Atlanta, GA, 11
pp.

[21] Cuttell, G. D., M. B. Snyder, J. M. Vandenbossche and M. Wade, "Performance of

Concrete Pavements Containing Recycled Concrete Aggregate," Transportation Research
Journal 1574, Transportation Research Board, Washington, DC, 1997.

[22] ASTM C 157. Standard Test Method for Length Change of Hardened Hydraulic-
Cement Mortar and Concrete ASTM International, (1999).