Construction and Building Materials 79 (2015) 8389

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Initial moisture and mixing effects on higher quality recycled coarse

aggregate concrete
Alexander S. Brand a,, Jeffery R. Roesler b, Andres Salas c
a

Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, 205 N. Mathews Ave., B233 NCEL, Urbana, IL 61801, USA
Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, 205 N. Mathews Ave., 1211 NCEL, Urbana, IL 61801, USA
c
Department of Civil and Agricultural Engineering, Universidad Nacional de Colombia, Bogota D.C., Colombia
b

h i g h l i g h t s
 RCA performs best in concrete when in a partially-saturated moisture condition.
 Two-stage mixing could improve the concrete strength properties.
 Shrinkage of concrete with RCA was not affected by the mixing method.

a r t i c l e

i n f o

Article history:
Received 5 September 2014
Received in revised form 2 January 2015
Accepted 5 January 2015
Available online 21 January 2015
Keywords:
Recycled concrete aggregate
Two-stage mixing approach
Aggregate moisture condition
Recycled aggregate concrete
Strength
Shrinkage

a b s t r a c t
Higher quality recycled concrete aggregates (RCA) were used as coarse aggregates in a concrete mixture.
The initial aggregate moisture condition (oven dry, partially saturated, fully saturated) and mixing
method (normal, two-stage) were assessed in order to limit the differences between concrete containing
RCA and virgin aggregates. Concrete with partially-saturated RCA using two-stage mixing was benecial
for improving the compressive and split tensile strengths. For this material, the mixing method did not
affect the concrete free drying shrinkage at any age. Overall, the two-stage mixing method increased
the concrete workability and could potentially improve the concrete strength properties, particularly
when the RCA is partially saturated.
2015 Elsevier Ltd. All rights reserved.

1. Introduction
Under modern airport renovation plans, agencies are faced with
disposing signicant amounts of concrete waste because of reconguration and reconstruction of the aireld. Concrete pavement
waste can be crushed into an aggregate and placed as granular base
or subbase layer for asphalt or concrete pavements, which has
been done extensively over the years. The concrete waste can also
be used as an aggregate in new concrete materials to replace virgin
aggregates, as has been under investigation for the several decades,
primarily for highway rigid pavements [13]. There have been
fewer studies of using RCA for aireld rigid pavements, either as
a support material [4,5], which has been the most common, or as
a replacement of virgin aggregate in concrete [68]. In order to
minimize adverse impact to aircraft operation and performance,
Corresponding author.
E-mail addresses: abrand2@illinois.edu (A.S. Brand), jroesler@illinois.edu (J.R.
Roesler), asalasm@unal.edu.co (A. Salas).
http://dx.doi.org/10.1016/j.conbuildmat.2015.01.047
0950-0618/ 2015 Elsevier Ltd. All rights reserved.

one primary requirement of utilizing RCA in new aireld concrete

is that it would not signicantly affect the concrete strength,
shrinkage, and surface properties.
The moisture condition of the RCA has been found to affect the
workability and strength of the concrete [9], mainly because of the
higher absorption capacity of the RCA relative to virgin aggregate.
Researchers have published studies on RCA concrete with the initial moisture state of RCA in the oven dry [10], air dry [11], prewetted (soaked for 10 min) [1214], partially saturated [15,16],
fully saturated [17,18], or otherwise described as presoaked conditions without further details [1921]. Etxeberria et al. [15] cautioned that fully saturated RCA may result in the failure of the
new interfacial transition zone (ITZ) between the new cement
paste and the RCA. Poon et al. [22] explicitly studied the effects
of three RCA moisture conditions (oven dry, air dry, and saturated
surface dry) and found that the oven dry condition had the greatest
initial slump but also had the fastest slump loss while the air dry
condition (50% of the maximum absorption capacity) resulted in
the greatest compressive strength. Barra de Oliveira and Vazquez

84

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

[23] also considered the effects of three RCA moisture conditions

(oven dry, partially saturated at 8890% absorption capacity, and
saturated surface dry) and found that the partially saturated RCA
had slightly greater strengths and better freeze/thaw resistance
than the other two moisture states. A RILEM report [24] stated that
pre-soaking the RCA is benecial to prevent rapid loss of workability but there is not a signicant difference in compressive strength
between air dry and saturated surface dry RCA.
The thickness of the ITZ in concrete is on the order of 2040 lm,
which corresponds to approximately 2040% of the total volume of
the matrix between aggregates [25]. For many low slump concretes, the ITZ is the controlling factor for the concrete strength
because of two main factors: (1) the interface between the aggregate and the ITZ, i.e., the bond, and (2) the larger porosity present
in the ITZ [26]. Compared to virgin aggregate, the ITZ with RCA can
be morphologically different because of its higher porosity and
absorption capacity [27]. One study found that the strength of
the old ITZ between the aggregate and mortar in the RCA can affect
the strength of the new concrete [28]. Another study found that the
microhardness of the old ITZ is affected by its quality and not the
quantity of the adhered mortar [29]. In an attempt to improve
the ITZ between the new cement paste and the RCA, two similar
approaches have been proposed: the two-stage mixing approach
(TSMA) by Tam et al. [30] and the double mixing (DM) method
by Ryu [31]. In later studies, Tam et al. [32,35] modied the TSMA
to examine the effects of how the concrete is affected by different
mixing times and mixture components. In TSMA method, the RCA
is coated with a cement slurry in an attempt to ll the cracks and
pores in the RCA with hydrating cement. Subsequently, variations
of TSMA/DM have been utilized by other researchers
[7,8,21,33,34,3640]. Compared to conventional concrete mixing
approaches, TSMA/DM with RCA has been found to increase the
compressive [3033,35,3739], split tensile [31,35,38], exural
strengths [33,35]; increase the modulus of elasticity
[32,33,35,38]; potentially improve the ITZ [3037,39] while reducing ITZ thickness [31]; reduce the chloride ion penetrability [31,37]
(more notably at later ages [34] or earlier ages [38]); generally
reduce the water permeability [34]; and shrinkage [34,38] and
creep strains [34]. Overall, the literature has shown that there is
a benecial effect of TSMA on mean RCA concrete properties without explicitly demonstrating the statistical improvement.
A recent study on airport RCA materials showed that 100% volume replacement of the virgin coarse aggregate with RCA did not
statistically reduce its strength but did produce rapid loss in workability [7]. The initial RCA moisture condition at batching appeared
to be the main factor in the loss in workability, as noted by

previous researchers [22]. The objectives of this study were to

determine the statistical signicance in terms of workability,
strength, and shrinkage of replacing the virgin aggregate type with
high quality RCA given different mixing approaches (normal mixing
method versus TSMA) and initial coarse aggregate moisture contents (oven dry, partially-saturated, and saturated surface dry).
While another study has specically examined the effects of these
three moisture conditions on RCA concrete mechanical and durability properties [23], this research will verify if there is a statistical
improvement in the concrete strength and shrinkage properties of
100% high-quality coarse RCA relative to virgin aggregate concrete
with the TSMA compared to the normal mixing procedure (NMP).
2. Materials and methods
2.1. Cementitious materials
All concrete mixtures in this study contained a total cementitious content of
517 lb/yd3 (307 kg/m3), which was composed of 80% Type I portland cement and
20% Class C y ash. This total cementitious content was selected based on the minimizing total cementitious content in the paving mix for the airport.
2.2. Aggregate properties
The RCA in this study was sourced from a single stockpile of crushed concrete at
the OHare International Airport [41]. This RCA in particular is of higher quality and
less variability relative to conventional RCA or construction and demolition waste,
as there are stricter protocols as to what can and cannot be used in concrete for aireld rigid pavements. From X-ray diffraction the RCA was found to contain dolomite, quartz, and remnant hydrated cement phases.
Two virgin coarse aggregates were blended (i.e., sieved and recombined) in this
study, which were classied, according to the Illinois Department of Transportation
[42], as CA-7 and CA-16 (see Table 1). The RCA was sieved and recombined to match
the blended virgin aggregate gradation. The same ne aggregate was used in all
mixes, which was a natural sand classied as FA-02. The specic gravity (saturated
surface dry) and absorption capacity of the aggregates are also shown in Table 1. As
expected, the RCA was found to have a lower specic gravity and higher absorption
capacity relative to virgin aggregates.
The RCA was evaluated for the residual mortar content (RMC) using a method
outlined by Abbas et al. [43] in which the RCA was immersed in sodium sulfate
solution while undergoing freezing and thawing cycles. This method found that
the particles retained on the #4 (4.75 mm) sieve had an RMC of about 26% while
particles larger than 25 mm had an RMC of about 13%.
2.3. Aggregate moisture conditions
The three initial aggregate moisture conditions used in this study were oven dry
(OD), saturated surface dry (SSD), and partially-saturated (approximately 8085%
SSD). The OD condition was met by drying the aggregates in a 105 C oven for at
least 24 h prior to batching. The SSD condition was obtained by soaking the aggregates in water for at least 24 h prior to batching and then removing the excess surface moisture off of the aggregates until the SSD condition was reached. In order to

Table 1
Aggregate gradations and properties.
Gradation (percent cumulative passing)
Sieve

Virgin coarse 1

Virgin coarse 2

Combined grading of CA-7

and CA-16 for RCA

Natural sand
ne aggregate
FA-02

U.S.

mm

CA-16

CA-7

3/4 in
1/2 in
3/8 in
#4
#8
#16
#30
#50
#100
#200

19
12.5
9.5
4.75
2.36
1.18
0.6
0.3
0.15
0.075

100%
100%
95%
41%
6%
2%
1%
1%
1%

83%
20%
6%
1%
0%
0%
0%
0%
0%

93%
66%
59%
47%
31%
19%
10%
1%
0%

100%
100%
100%
92%
68%
42%
21%
2%
0%
0%

2.68
2.73%

2.67
1.90%

2.41
5.51%

2.57
2.43%

Aggregate properties
Specic gravity
Absorption capacity

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

obtain the partially-saturated state, saturated aggregates were sprayed with water
24 h prior to mixing and then placed on top of a plastic sheet in a controlled temperature room to let the excess water evaporate. Prior to mixing in the concrete, the
moisture condition of the aggregate was veried by drying a sample in an oven. All
ne aggregates used in all mixtures were in an air dry condition.

85

both mixtures had the same combined gradation. To ensure a constant w/cm ratio,
the moisture content of the aggregates was accounted for and the mix water
adjusted accordingly to provide sufcient water for the aggregates to be in SSD condition. To provide consistent workability between mixes, the only chemical admixture used for all mixes was a high range water reducer, the dosages of which are
shown in Table 3.

2.4. Batching methodologies

2.7. Statistical analysis
The two mixing methods used in this study were the normal mixing process
(NMP) per ASTM C192 [44] and the two-stage mixing approach (TSMA). For the
NMP, the coarse aggregate was added to the mixing pan with about one-third of
the mix water. The aggregate was then mixed with the water for about 30 s. Mixing
was then stopped and the cementitious materials and ne aggregate were added to
the mixing pan. Mixing resumed again with the remaining two-thirds of the mix
water slowly added. After addition of all the water, the batch was mixed for
3 min, allowed to rest for 3 min, and mixed a nal 2 min. Chemical admixtures were
added slowly during the rst 3 min of mixing when needed.
The TSMA used in this study followed similarly to the modied version summarized by Tam and Tam [35]. The mixing steps were as follows: (1) coarse aggregates
and cementitious materials were mixed for 1 min; (2) half of the water was added
and then mixed for 1 min; (3) virgin ne aggregate and remaining water is added
with chemical admixtures and then mixed for 2 min.
2.5. Concrete testing
After mixing was complete, the fresh concrete was evaluated for initial slump
[45] and then laboratory specimens were cast into molds. Cylindrical specimens
measuring 4 inches (100 mm) in diameter and 8 inches (200 mm) in height were
used to evaluate the compressive [46] and split tensile [47] strengths. Beams measuring 6 by 6 by 21 inches (150 by 150 by 530 mm) were used to evaluate the exural strength under third-point loading [48]. The free drying shrinkage was
determined using 3 by 3 by 11.25-inch (75 by 75 by 285 mm) prisms that were
placed in an environmentally-controlled room at 50% relative humidity and 23 C.
This shrinkage test method was a slightly modied ASTM C157 [49] test in which
the specimens were placed in the humidity room immediately after removing from
the molds rather than after 28 days of curing. This was done in order to compare the
early-age shrinkage properties of RCA and virgin aggregate concrete.
2.6. Mixture designs
All concrete mixtures contained a total cementitious content of 517 lb/yd3
(307 kg/m3) and had a water-to-cementitious (w/cm) ratio of 0.42. The two mix
design constituents and proportions are shown in Table 2. The volume of coarse
aggregate was kept constant between the virgin and RCA aggregate mixtures and

Table 2
Concrete mix designs in lb/yd3 (kg/m3).

Cement
Fly ash
Water
Virgin coarse aggregate (CA-7)
Virgin coarse aggregate (CA-16)
RCA coarse aggregate
Virgin ne aggregate (FA-02)

Virgin aggregate concrete

RCA concrete

414 (246)
103 (61)
217 (129)
1495 (887)
500 (297)

1279 (758)

1799 (1067)
1279 (758)

Table 3
Initial slump and admixture dosage for each mix.
Aggregate
type

Mixing
method

Moisture
condition

High range water

reducer dosage*

Initial slump,
inch (mm)

Virgin

NMP

OD
80% SSD
SSD
OD
80% SSD
SSD

250
250
230
250
250
230

(3.83)
(3.83)
(3.53)
(3.83)
(3.83)
(3.53)

2.5 (65)
2.25 (60)
2 (50)
7 (160)
7 (160)
8.5 (215)

OD
80% SSD
SSD
OD
80% SSD
SSD

235
260
270
235
260
270

(3.60)
(3.99)
(4.14)
(3.60)
(3.99)
(4.14)

6 (150)
2.5 (65)
2.5 (65)
9 (230)
8 (205)
9 (230)

TSMA

RCA

NMP

TSMA

In mL per 100 kg cementitious (. oz. per 100 lbs cementitious).

Analysis of variance (ANOVA) tests were conducted to determine if the means

were statistically different for the various strength tests. An additional statistical
test was conducted, Tukeys signicant difference (TSD) test, which utilizes a studentized-distribution and the mean squared error from ANOVA [50]. The TSD test
is performed piecewise to compare which means were statistically similar.

3. Results and discussion

3.1. Initial slump
The results for initial slump for each mix can be found in Table 3
along with the dosage of the high range water reducer. As can be
seen, the high range water reducer dosage was the same for each
moisture condition between the two mixing methods. The TSMA
produced a higher initial slump than NMP for all mixes. Therefore,
it can be concluded that TSMA resulted in an initially more workable fresh concrete state relative to the NMP. The effect of aggregate type for a xed mixing procedure did not produce a clear
trend in the concrete workability.
3.2. Concrete strength
The compressive strength was evaluated at three ages (3, 7, and
28 days) using an average of three cylinder specimens. The average
compressive strength values for the virgin aggregate and RCA concrete mixtures can be found in Fig. 1. The TSMA consistently produced higher mean compressive strengths versus the NMP for all
moisture conditions, which is in agreement with the literature
[3033,35,3739]. The virgin aggregate concrete mixes had higher
compressive strengths than the RCA concrete mixes (see Figs. 1 and
2), as anticipated from past studies, but not as drastic of strength
changes as have been reported when using lower quality crushed
concrete as aggregates; for example, Kou et al. [38] found 50
70% reductions in the compressive strength when using 50% fresh
concrete waste as a coarse aggregate replacement, and Poon et al.
[27] found compressive strength reductions of 21% and 7% at
28 days when using normal-strength and high-performance concrete, respectively, as 100% replacements of the coarse aggregate.
For the concrete tested in this experiment at 28 days, the largest
strength differences between the virgin and RCA concretes was for
the OD aggregate condition, where the RCA concrete compressive
strength was 24% lower than virgin aggregate concrete. At the
80% SSD moisture condition, the RCA concrete compressive
strength was only 1012% lower than the virgin aggregate concrete. In terms of the aggregate moisture condition, the highest
compressive strengths were achieved from the 80% SSD, followed
by SSD, and then the OD conditions at all ages for the virgin aggregate NMP and RCA NMP and TSMA concretes. For the virgin aggregate TSMA mixes, the SSD and 80% SSD compressive strengths
were similar, but the OD condition was still the weakest. These
trends were consistent over all ages tested.
ANOVA tests on the compressive strength results revealed that
not all means were statistically similar for all strength tests, so the
TSD test was performed for the 28-day strengths, as seen in Table 4.
The mixtures are grouped into categories in order to easily connect
which means are statistically similar. The virgin mixtures were statistically stronger than the RCA mixtures except for the Virgin OD
NMP mix. The strongest mixtures were with TSMA and partiallysaturated to SSD moisture conditions. For RCA, the greatest

86

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

(a)

Table 4
Groupings for 28-day compressive strength based on the TSD test results.
Tukey
grouping*

Mean
(psi)

Standard
deviation
(psi)

Concrete mixture

A
A

8020
7904
7544
7395
7304
7206
7064
6639
6518
6384
5897
5446

36.5
147.1
19.8
9.4
68.2
37.3
74.7
23.9
134.9
32.9
48.4
37.6

Virgin 80%SSD TSMA

Virgin SSD TSMA
Virgin OD TSMA
Virgin 80%SSD NMP
Virgin SSD NMP
Virgin OD NMP
RCA 80%SSD TSMA
RCA 80%SSD NMP
RCA SSD TSMA
RCA SSD NMP
RCA OD TSMA
RCA OD NMP

B
B
D
D
F
F

C
C
C
E
E
G
H

(b)

Fig. 1. Average compressive strength for the (a) virgin aggregate and (b) RCA
concrete mixes. Error bars indicate one standard deviation. NMP = normal mixing
procedure, TSMA = two-stage mixing approach, SSD = saturated surface dry,
OD = oven dry. 1 psi = 6.9 kPa.

Means with the same letter are not signicantly different.

In the concrete mixture design, the mix water was adjusted to

account for the water needed to bring the aggregates to SSD condition. Given the relatively high absorption capacity of this RCA
source (5.5%), the total mix water added to the mixture with OD
RCA was greater than the total mix water added to the mixture
with fully- or partially-saturated RCA. Since it can take up to
24 h for RCA to become fully saturated [9], which is longer than
the time for the concrete to set, it is likely that the OD RCA never
reached the SSD condition, despite the necessary water added. As
a result, the effective w/cm ratio for the OD RCA mixtures was
higher than the mixtures with fully- or partially-saturated RCA,
which explains why the mixtures with OD RCA resulted in the lowest compressive strengths. Additionally, Etxeberria et al. [15] cautioned that fully saturated RCA may result in the failure of the new
interfacial transition zone (ITZ) between the new cement paste and
the RCA, which explains why the mixtures with partially-saturated
RCA performed better than the SSD RCA mixtures. Other studies
have also indicated that concrete with partially-saturated RCA will
result in greater compressive strengths than concrete with OD or
SSD RCA [23].
The split tensile and exural strengths were evaluated at an age
of 28 days as the average of three specimens for exural strength
and two specimens for split tensile strength. As expected, the
RCA concrete resulted in lower mean split tensile and exural
strengths than the virgin aggregate concrete as seen in Figs. 2
and 3, respectively. The TSMA mixes resulted in higher mean
strengths than the NMP mixes. For exural strength, the general

Fig. 2. Average split tensile strength at 28 days for the virgin aggregate and RCA
concrete mixes. Error bars indicate one standard deviation. NMP = normal mixing
procedure, TSMA = two-stage mixing approach, SSD = saturated surface dry,
OD = oven dry. 1 psi = 6.9 kPa.

strength was achieved with TSMA and a partially saturated moisture condition, which was statistically similar to one of the virgin
aggregate concrete mixtures, while the OD RCA yielded the weakest strength irrespective of mixing method. For certain aggregate
types and moisture conditions, TSMA resulted in higher strengths
than NMP, often resulting in statistically higher strengths. Finally,
for almost all aggregate conditions, the TSMA had statistically
greater mean strength relative to NMP except for the RCA SSD mixture, where the means were statistically similar.

Fig. 3. Average exural strength at 28 days for the virgin aggregate and RCA
concrete mixes. Error bars indicate one standard deviation. NMP = normal mixing
procedure, TSMA = two-stage mixing approach, SSD = saturated surface dry,
OD = oven dry. 1 psi = 6.9 kPa.

87

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

Table 5
Groupings for 28-day split tensile and exural strengths based on the TSD test results.
Tukey grouping*

Mean (psi)

28-day split tensile strength

I
I
I
I
J
J
K
K
K
L
L
L

Standard deviation (psi)

Mixture

812
804
789
784
752
750
696
684
682
631
623
623

8.6
1.1
11.0
7.8
3.9
2.1
4.3
20.4
4.2
7.9
4.5
2.9

Virgin OD TSMA
Virgin 80%SSD TSMA
Virgin SSD TSMA
Virgin OD NMP
RCA 80%SSD TSMA
Virgin 80%SSD NMP
Virgin SSD NMP
RCA SSD TSMA
RCA OD TSMA
RCA SSD NMP
RCA 80%SSD NMP
RCA OD NMP

799
733
708
683
681
678
663
608
595
588
587
556

17.1
150.8
14.3
1.8
0.8
10.8
2.4
0.8
11.2
6.0
3.9
12.1

Virgin 80%SSD TSMA

RCA 80%SSD NMP
RCA 80%SSD TSMA
Virgin SSD TSMA
Virgin 80%SSD NMP
Virgin OD TSMA
Virgin OD NMP
RCA SSD TSMA
RCA OD TSMA
Virgin SSD NMP
RCA SSD NMP
RCA OD NMP

28-day exural strength

N
N
N
N
N
N
N

P
P
P
P
P
P
P
P
P

M
M
M
M
M
M

O
O
O
O
O
O
O
O
O

Means with the same letter are not signicantly different.

(a)

(b)

TSMA achieved greater split tensile strengths relative to the

NMP. The effect of initial aggregate moisture content did not show
as clear of trends as in the compressive strength specimens. Since
the strength of the ITZ affects the concrete tensile strength more
than the compressive strength [25], the lack of clear trend in tensile/exural strength results relative to compressive strength suggests that the moisture condition of this RCA source and TSMA do
not affect the ITZ as strongly as other studies have suggested.
The ANOVA test revealed that the not all means were statistically similar for the split tensile and exural strength results, so
the TSD test was performed (Table 5). For the split tensile results,
the virgin aggregate concrete was stronger for most conditions but
the aggregate moisture state and mixing procedure did not show
the same behavior as in the compressive strength results. For
RCA, the TSMA had higher split tensile strengths than NMP with
some of the RCA TSMA mixtures having statistically similar
strengths to the virgin aggregate concretes. For exural strengths,
there were no clear statistical trends, suggesting that the aggregate
type, initial moisture condition, mixing process and testing conguration did not distinguish similar trends seen in split tensile and
compressive strength tests.

3.3. Free drying shrinkage

Fig. 4. Drying shrinkage behavior of the (a) virgin aggregate and (b) RCA concrete
mixes. NMP = normal mixing procedure, TSMA = two-stage mixing approach,
SSD = saturated surface dry, OD = oven dry.

trend was that the greatest strength was achieved by the 80% SSD
condition. For the split tensile strength, virgin aggregates produced
a higher mean concrete strength than the RCA mixtures and the

The free drying shrinkage was determined as the average of

three specimens for the virgin aggregate concrete and the RCA concrete mixes, as shown in Fig. 4. The virgin aggregate concrete
resulted in greater variability between mixture shrinkage differences, ranging from 490 to 620 microstrain at 90 days, while
the RCA concrete had less variability between mixture shrinkage
differences, ranging from 560 to 630 microstrain at 90 days.
Overall, the magnitude of free shrinkage for either of the virgin
aggregate or RCA concretes is in line with typical concrete shrinkage values, which is around 200 to 800 microstrain [51]. In general, the shrinkage strains were lower for the TSMA versus NMP
mixes, but the difference was often less than 10% at later ages.
For the virgin aggregate concrete mixes at later ages and for either

88

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

Table 6
Groupings for drying shrinkage at 7 and 90 days based on the TSD test results.
Tukey grouping*
7-day shrinkage
Q
R
Q
R
Q
R

Mean (le)

S
S
S
S
S
S

T
T
T
T
90-day shrinkage
U
U
W
W

AA
AA
*

V
V
X
X

Z
Z

Y
Y
Y
Y
Y

Standard deviation (le)

Mix

190
197
213
240
243
247
247
247
293
293
297
320

10.0
20.8
20.8
0.0
5.8
23.1
5.8
5.8
15.3
5.8
15.3
26.5

RCA OD TSMA
RCA OD NMP
Virgin OD TSMA
RCA 80%SSD TSMA
RCA SSD TSMA
Virgin OD NMP
RCA SSD NMP
RCA 80%SSD NMP
Virgin SSD TSMA
Virgin 80%SSD TSMA
Virgin 80%SSD NMP
Virgin SSD NMP

493
510
533
553
560
580
587
590
593
600
620
630

5.8
0.0
5.8
15.3
0.0
0.0
5.8
0.0
5.8
0.0
17.3
0.0

Virgin OD TSMA
Virgin OD NMP
Virgin 80%SSD TSMA
Virgin 80%SSD NMP
RCA 80%SSD TSMA
RCA SSD TSMA
Virgin SSD TSMA
RCA 80%SSD NMP
RCA OD TSMA
RCA SSD NMP
Virgin SSD NMP
RCA OD NMP

Means with the same letter are not signicantly different.

NMP or TSMA, the OD condition resulted in the lowest shrinkage

strains, followed by 80% SSD, and nally SSD resulted in the largest
amount of shrinkage. Conversely, for the RCA mixes at later ages
for both NMP and TSMA, the 80% SSD condition had the least
amount of shrinkage, followed by SSD, and nally OD had the
greatest amount of shrinkage. For OD virgin aggregates, the free
mix moisture could be more easily drawn into the larger aggregate
pores and reduce the free drying shrinkage, whereas for RCA, it
would be more difcult for free moisture to move into smaller
reclaimed mortar pores and thus it stayed in the paste.
An ANOVA test determined that the free shrinkage means were
not statistically similar, so a TSD test was conducted for the data to
compare the strains at 7 and 90 days (see Table 6). The grouping
for free shrinkage revealed that at early ages (7 days), the drying
shrinkage means were very similar for many of the mixes but overall the RCA mixtures had less shrinkage than the virgin aggregate
concrete mixes. However, at 90 days, the mean shrinkage between
mixes was more statistically different. No clear trends between the
initial moisture condition or mixing method could be inferred from
the statistical analysis of the shrinkage at 90 days except that four
out of six virgin aggregate mixtures had the lowest shrinkage
magnitudes.
In conjunction with the previous argument that the RCA in concrete with OD RCA may not sufciently attain a fully-saturated
condition after mixing, the free drying shrinkage is greater for concrete with OD RCA since the effective w/cm ratio would be greater
than concrete with fully- or partially-saturated RCA. This hypothesis is veried in that the free drying shrinkage at 90 days could be
statistically lower for concrete with fully- or partially-saturated
RCA relative to concrete with OD RCA.
4. Conclusions
The two-stage mixing approach (TSMA) has been shown by previous researchers to improve the hardened properties of recycled
concrete aggregate (RCA) concrete, particularly for concrete

demolition waste. In this research, the effect of the initial aggregate

moisture condition and mixing method was studied to determine
its effect on the workability, strength, and shrinkage properties
of concrete containing a higher quality coarse recycled concrete
aggregate (RCA) relative to virgin coarse aggregate concrete. The
laboratory testing factorial considered two coarse aggregate types
(virgin, RCA), three initial aggregate moisture conditions (oven dry,
partially saturated, fully saturated), and two mixing methods (normal, TSMA). The results revealed that the strongest statistical difference in mean compressive strength was between virgin and
RCA concrete. The TSMA approach achieved higher compressive
and split tensile strengths compared with the NMP for both virgin
aggregate and RCA concrete. These trends were not noted in the
exural strength test results. Overall, the strength properties for
RCA concrete were greatest when the RCA was at least in the partially-saturated moisture state with TSMA. At early ages, the RCA
concrete had lower free shrinkage but at later ages (90 days) the
virgin aggregate concrete had the overall lower shrinkage. For high
quality RCA, TSMA can improve the compressive and tensile
strength by less than 10% while the initial aggregate moisture state
only impacted the compressive strength and shrinkage results. Relative to NMP, TSMA will also improve the initial workability of the
fresh concrete.

Acknowledgements
Acknowledgement is given to the support from the Center of
Excellence for Airport Technology (CEAT) at the University of Illinois at Urbana-Champaign (UIUC) provided by the OHare Modernization Program (OMP) and the City of Chicago. The authors would
also like to thank Rossi Contractors, Inc., for the crushed concrete
aggregates provided for the laboratory study and Mr. Ross Anderson of Bowman and Barrett, Inc., of Chicago, Illinois. Additional
acknowledgement is given to UIUC students Catalina Arboleda
and Ryan Smith.

A.S. Brand et al. / Construction and Building Materials 79 (2015) 8389

References
[1] Yrjanson W. Recycling of Portland cement concrete pavements, synthesis of
highway
practice
154,
national
cooperative
highway
research
program. Washington (DC): Transportation Research Board; 1989.
[2] Wade MJ, Cuttell GD, Vandenbossche JM, Yu HT, Smith K.D, Snyder MD,
Performance of concrete pavements containing recycled concrete aggregate,
report No. FHWA-RD-96-164, McLean, Virginia: University of Minnesota,
Federal Highway Administration; 1997.
[3] Sturtevant J. Performance of rigid pavements containing recycled concrete
aggregates [Master of Science thesis]. Durham, New Hampshire: University of
New Hampshire; 2007.
[4] Saeed A, Hammons MI, Runo Feldman D, Poole T. Evaluation, design, and
construction techniques for aireld concrete pavement used as recycled
material for base, nal report No. IPRF-01-G-002-03-5. Skokie, Illinois: Applied
Research Associates, Innovated Pavement Research Foundation; 2006.
[5] Saeed A, Hammons MI. Use of recycled concrete as unbound base aggregate in
aireld and highway pavements to enhance sustainability. In: Proceedings of
the 2008 aireld and highway pavements conference. Bellevue, Washington,
October 2518, 2008.
[6] Hironaka MC, Cline GD, Shoemaker NF. Recycling of Portland cement concrete
airport pavements An experimental investigation, nal report, Port
Hueneme, California: Naval Civil Engineering Laboratory; 1987.
[7] Salas A, Roesler JR, Lange D. Batching effects on properties of recycled concrete
aggregates for aireld rigid pavements. In: Proceedings of the 2010 FAA
worldwide airport technology transfer conference, Atlantic City, New Jersey,
April 2022, 2010.
[8] Jeong H. Processing and properties of recycled aggregate concrete [Master of
Science thesis], Urbana, Illinois: University of Illinois at Urbana-Champaign;
2011.
[9] Nealen A, Schenk S, The inuence of recycled aggregate core moisture on
freshly mixed and hardened concrete properties, Darmstadt Concrete: Annual
Journal on Concrete and Concrete Structures. 13, 1998. http://www.b-i-m.de/
public/tudmassiv/dacon13nealenschenk.htm.
[10] Latif Al-Mufti R, Fried AN. The early age non-destructive testing of
concrete made with recycled concrete aggregate. Constr Build Mater
2012;37:37986.
[11] Sri Ravindrarajah R, Loo YH, Tam CT. Recycled concrete as ne and coarse
aggregates in concrete. Mag Concr Res 1987;39:21420.
[12] Gonzlez-Fonteboa B, Martnez-Abella F, Eiras-Lpez J, Seara-Paz S. Effect of
recycled coarse aggregate on damage of recycled concrete. Mater Struct
2011;44:175971.
[13] Gonzlez-Fonteboa B, Martnez-Abella F, Carro Lpez D, Seara-Paz S. Stressstrain relationship in axial compression for concrete using recycled saturated
coarse aggregate. Constr Build Mater 2011;25:233542.
[14] Sagoe-Crentsil KK, Brown T, Taylor AH. Performance of concrete made with
commercially produce coarse recycled concrete aggregate. Cem Concr Res
2001;31:70712.
[15] Etxeberria M, Vzquez E, Mar A, Barra M. Inuence of amount of recycled
coarse aggregates and production process on properties of recycled aggregate
concrete. Cem Concr Res 2007;37:73542.
[16] Casuccio M, Torrijos MC, Giaccio G, Zerbino R. Failure mechanism of recycled
aggregate concrete. Constr Build Mater 2008;22:15006.
[17] Berndt ML. Properties of sustainable concrete containing y ash, slag and
recycled concrete aggregate. Constr Build Mater 2009;23:260613.
[18] Bordelon A, Cervantes V, Roesler JR. Fracture properties of concrete containing
recycled concrete aggregates. Mag Concr Res 2009;61:66570.
[19] Xiao J, Li J, Zhang Ch. Mechanical properties of recycled aggregate concrete
under uniaxial loading. Cem Concr Res 2005;35:118794.
[20] Fathifazl G, Ghani Razaqpur A, Burkan Isgor O, Abbas A, Fournier B, Foo S.
Creep and drying shrinkage characteristics of concrete produced with coarse
recycled concrete aggregate. Cem Concr Compos 2011;33:102637.
[21] Zhang W, Ingham JM. Using recycled concrete aggregates in New Zealand
ready-mix concrete production. J Mater Civ Eng 2010;22:44350.
[22] Poon CS, Shui ZH, Lam L, Fok H, Kou SC. Inuence of moisture states of natural
and recycled aggregates on the slump and compressive strength of concrete.
Cem Concr Res 2004;34:316.
[23] Barra de Oliveira M, Vazquez E. The inuence of retained moisture in
aggregates from recycling on the properties of new hardened concrete.
Waste Manage 1996;16:1137.
[24] Hansen TC, editor. Recycling of demolished concrete and masonry, RILEM
report 6. London: E & FN Spon; 1992.

89

[25] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. Upper Saddle River: Pearson
Education; 2003.
[26] Bentur A, Odler I. Development and nature of interfacial microstructure. In:
Maso JC, editor. Interfacial transition zone in concrete, RILEM Report 11.
London: E & FN Spon; 1996.
[27] Poon CS, Shui ZH, Lam L. Effect of microstructure of ITZ on compressive
strength of concrete prepared with recycled aggregates. Constr Build Mater
2004;18:4618.
[28] Ryu JS. An experimental study on the effect of recycled aggregate on concrete
properties. Mag Concr Res 2002;54:712.
[29] Otsuki N, Miyazato S, Yodsudjai W. Inuence of recycled aggregate on
interfacial transition zone, strength, chloride penetration, and carbonation of
concrete. J Mater Civ Eng 2003;15:44351.
[30] Tam VWY, Gao XF, Tam CM. Microstructural analysis of recycled aggregate
concrete produced from two-stage mixing approach. Cem Concr Res
2005;25:1195203.
[31] Ryu JS. Improvement on strength and impermeability of recycled concrete
made from crushed concrete coarse aggregate. J Mater Sci Lett
2002;21:15657.
[32] Tam VW-Y, Gao X-F, Tam CM. Comparing performance of modied two-stage
mixing approach for producing recycled aggregate concrete. Mag Concr Res
2006;58:47784.
[33] Tam VWY, Tam CM, Wang Y. Optimization on proportion for recycled
aggregate in concrete using two-stage mixing approach. Constr Build Mater
2007;21:192839.
[34] Tam VWY, Tam CM. Assessment of durability of recycled aggregate concrete
produced by two-stage mixing approach. J Mater Sci 2007;42:3592602.
[35] Tam VWY, Tam CM. Diversifying two-stage mixing approach (TSMA) for
recycled concrete aggregate: TSMAs and TSMAsc. Constr Build Mater
2008;22:206877.
[36] Tam VWY, Gao XF, Tam CM, Ng KM. Physio-chemical reactions in recycle
aggregate concrete. J Hazard Mater 2009;163:8238.
[37] Kong D, Lei T, Zheng J, Ma C, Jiang J, Jiang J. Effect of mechanism of surfacecoating pozzalanics materials around aggregate on properties and ITZ
microstructure of recycled aggregate concrete. Constr Build Mater
2010;24:7018.
[38] Kou SC, Zhan BJ, Poon CS. Feasibility study of using recycled fresh concrete
waste as coarse aggregates in concrete. Constr Build Mater 2012;28:54956.
[39] Li W, Xiao J, Sun Z, Kawashima S, Shah SP. Interfacial transition zones in
recycled aggregate concrete with different mixing approaches. Constr Build
Mater 2012;35:104555.
[40] Rupnow TD, Schaefer VR, Wang K, Hermanson BL. Improving portland cement
concrete mix consistency and production rate through two-stage mixing, nal
report No. IHRB Project TR-505, Ames, Iowa: Center for Transportation
Research and Education, Iowa State University, Iowa Department of
Transportation; 2007.
[41] Salas A, Brand AS, Roesler JR, Arboleda C, Lange DA. Properties of recycled
concrete aggregates for aireld rigid pavements, CEAT COE report No. 34,
Center of Excellence for Airport Technology, OHare Modernization Program,
2013.
[42] Standard specications for road and bridge construction. Illinois department
of transportation. Illinois: Springeld; 2007.
[43] Abbas A, Fathifazl G, Fournier B, Isgor OB, Zavadil R, Razaqpur AG, et al.
Quantication of the residual mortar content in recycled concrete aggregates
by image analysis. Mater Charact 2009;60:71629.
[44] ASTM C192. Standard practice for making and curing concrete test specimens
in the laboratory. West Conshohocken: ASTM International; 2007.
[45] ASTM C143. Standard test method for slump of hydraulic-cement
concrete. West Conshohocken: ASTM International; 2012.
[46] ASTM C39. Standard test method for compressive strength of cylindrical
concrete specimens. West Conshohocken: ASTM International; 2010.
[47] ASTM C496. Standard test method for splitting tensile strength of cylindrical
concrete specimens. West Conshohocken: ASTM International; 2004.
[48] ASTM C78. Standard test method for exural strength of concrete (using
simple beam with third-point loading). West Conshohocken: ASTM
International,; 2010.
[49] ASTM C157. Standard test method for length change of hardened hydrauliccement mortar and concrete. West Conshohocken: ASTM International; 2008.
[50] Mason RL, Gunst RF, Hess JL. Statistical design and analysis of experiments.
2nd ed. Hoboken: John Wiley & Sons; 2003.
[51] ACI 209.1R-05. Report on factors affecting shrinkage and creep of hardened
concrete. Farmington Hills: American Concrete Institute; 2005.