Construction and Building Materials 142 (2017) 42–50

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effects of nano-particles on failure process and microstructural

properties of recycled aggregate concrete
Wengui Li a,f,⇑, Chu Long b, Vivian W.Y. Tam c,d, Chi-Sun Poon e, Wen Hui Duan f
a
Centre for Built Infrastructure Research, School of Civil & Environmental Engineering, University of Technology Sydney, Sydney, NSW 2007, Australia
b
College of Civil Engineering, Hunan University, Changsha, Hunan 410082, PR China
c
School of Computing, Engineering and Mathematics, Western Sydney University, Penrith, NSW 2751, Australia
d
College of Civil Engineering, Shenzhen University, PR China
e
Department of Civil and Environmental Engineering, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China
f
Department of Civil Engineering, Monash University, Clayton, Victoria 3800, Australia

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 Failure process of RAC was influenced

by the relative strength between new
and old cement mortars.
 Colloidal NS could enter into ITZ
regions in RAC to increased hydration
products by secondary chemical
reaction.
 Powdered NL could not enhance the
compressive strength of RAC due to
the particles agglomeration.
 Colloidal NS could effectively reduced
porosity and water absorption of RAC.

a r t i c l e i n f o a b s t r a c t

Article history: The effects of nano-particles including nano-silica (NS) and nano-limestone (NL) on the crack propagation
Received 23 November 2016 and microstructure properties of recycled aggregate concrete (RAC) were experimentally investigated in
Received in revised form 13 February 2017 this study. The crack initiation and propagation of nano-particles modified RAC with different nano-
Accepted 8 March 2017
particle modification were evaluated using digital image correlation technique (DIC). The microstructures
and porosity of interfacial transition zones (ITZ) in nano-modified RAC were also examined using scan-
ning electron microscopy (SEM) and mercury intrusion porosimetry (MIP). It was found that the
Keywords:
micro-cracks were typically derived from relatively weak ITZs in RAC, and then progressively propagated
Recycled aggregate concrete (RAC)
Nano-particles
along the compressive loading direction. The meso-crack developments eventually led to final splitting
Micro-cracks failure. The results indicated that compared to NL, NS was more effective in improving the microstructure
Interfacial transition zone (ITZ) properties and enhance the mechanical strength of RAC. The porosity and water absorption of RAC were
Water absorption obviously reduced by the NS incorporation. However, due to particles agglomeration, NL could not effec-
Microstructural properties tively improve the microstructure of RAC for further enhancing the RAC mechanical properties.
Furthermore, in terms of severe particles agglomeration, NL was even detrimental to the mechanical
strength of RAC especially at the late-age.
Ó 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Centre for Built Infrastructure Research, School of Civil & Environmental Engineering, University of Technology Sydney, Sydney, NSW 2007,
Australia.
E-mail address: wengui.li@uts.edu.au (W. Li).

http://dx.doi.org/10.1016/j.conbuildmat.2017.03.051
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 W. Li et al. / Construction and Building Materials 142 (2017) 42–50
1. Introduction
The study on recycled aggregate concrete (RAC) have become a
hot topic in the field of construction material. A lot of researchers
realized that RAC can effectively recover waste into useful
resources. However, compared to ordinary concrete, there are still
some shortcomings of RAC for practical applications in terms of
mechanical properties [1–5], which is one of the major factors lim-
iting RAC in structural applications. To address this concern, a ser-
ies of research studies had been conducted to develop methods to
improve the RAC mechanical performance. Because the properties
of the recycled aggregate to a large extent determines the perfor-
mances of the RAC [6,7], increasing number of studies were carried
out to improve the quality of recycled aggregate. These methods
include microwave heating [8], chemical immersion [9], biological
precipitation [10,11], carbonization [12,13], and surface treatment
of recycled aggregate [14–17]. However, more effective methods
are still needed to achieve better mechanical properties for RAC.
For the failure mechanism of RAC, previous studies [18–21] have
confirmed that the mechanical properties of RAC is closely related
to the properties of the interfacial transition zone (ITZ) between
new cement mortar and recycled aggregate [22–24].
Nano-particles can act as nuclei for cement phases, further pro-
moting cement hydration process due to their high reactivity as fil-
ler and densifying the microstructure, thereby, and leading to
reduce porosity number of nano-particle has been studied for
incorporating in cement and concrete including nano-silica (NS)
[25–27], nano-limestone (NL) [28–29], nano-metallic oxide [30–
32], graphene oxide [33–35], carbon nano-tubes [36–37]. Two
nano-particle types, NS and NL, are the most commonly used in
cementitious material applications. Currently, there are typically
two methods which have been reported for producing nanosized
silica and limestone particles: (i) high energy milling of raw mate-
rial (top-down approach); and (ii) chemical synthesis (bottom-up
approach) [38,39]. Previous studies usually showed that they can
act as nuclei for promoting cement hydration process. These
nano-particles can also act as physical fillers for densifying the
microstructure of the cement mortar and ITZs in RAC. Nano-
particle has advantages in improving the mechanical properties
of cementitious composites. The modifications of NS and NL have
been demonstrated in NAC [26,28,29]. NS exhibits strong poz-
zolanic effect, which can promote the secondary hydration of
cement, i.e. pozzolanic reaction, generating further calcium silicate
hydrates (C-S-H), and fills the pore structures. On the other hand,
NL has a strong effect of crystal nucleus and physical filling effects
which can largely improve the concrete microstructure. RAC which
is more porous than natural aggregate concrete is more likely to be
improved by nano-particles. But further studies on the failure
mechanism of nano-particles modified RAC are still necessary.
In this study, NS and NL at different dosages were used to
enhance the properties of RAC. The effects of nano-particle on
the mechanical strength and crack propagation of RAC were evalu-
ated with incorporation dosages and mixing methods. Scanning
electron microscopy (SEM) and mercury intrusion porosimetry
(MIP) techniques were used to analyze the failure mechanism
and microstructure properties of the nano-modified RAC with NS
and NL. The related results can provide an insight into and enhanc-
ing RAC properties and promoting structural applications.
Table 1
Chemical composition of ordinary Portland cement.
SiO2 Al2O3 Fe2O3
Composition (%) 21.00 5.40 2.20
43
2. Material and method
2.1. Sample preparation
Ordinary Portland cement (OPC 42.5) was used in the experi-
ments, and the chemical compositions are shown in Table 1. The
fine aggregate used was common natural river sands, and the fine-
ness modulus was 2.6. Natural gravel with continuous gradation
from 5 to 26.5 mm was chosen as natural coarse aggregate. It
had water absorption of 0.96% and bulk density of 1,395 kg/m3.
The recycled coarse aggregate (RCA) was obtained from the demo-
lition of wasted concrete at the Shanghai Port wharf, China. The
RCA had a water absorption value of 6.30%, and a bulk density of
1290 kg/m3. The powdered NL was purchased from Xuan Cheng
Jing Rui New Material Co., Ltd, China. The colloidal NS was pur-
chased from Zhejiang Yuda Chemical Co., LTD, China. The physio-
chemical properties of the nano-particles (powdered NL and col-
loidal NS) used are shown in Tables 2 and 3, respectively. A naph-
thalene based superplasticizer was used to improve the RAC
workability and serve as the dispersant of NL powder in solution.
After NL powder was mixed with a small amount of water. Then
NL solution was subjected to ultrasonication to improve the dis-
persion by a Sonics Vibra-Cell vcx-500 ultrasonic processor
(Vibra-Cell-Sonics & Materials, Inc.) with an energy intensity of
40,000 J with amplitude of 80% and pulse of 4 s.
2.2. Experimental program
The RCA was used to replace the natural coarse aggregate by
weight at 30%. According to NS dosage (1.0 and 2.0% by weight
of cement) and two different mixing methods, NS-RAC was divided
into S1, S2, S3 and S4. According to the different NL dosages (1.0
and 2.0% by weight of cement) and superplasticizer addition
(1.0% by weight of cement), NL-RAC was divided into L1, L2, L3
and L4. The compressive strength of the RAC was tested after 7-
day and 28-day standard curing at 20 ± 2 °C and relative humidity
of 90 ± 5%. Concrete cubes of 150  150  150 mm were cast using
different dosages of nano-particles (powdered NL and colloidal
NS), as shown in Table 4. Three cubes were tested to measure
the compressive strength. To investigate the failure process of
RAC, some cubes were cut into slices of 150  150  30 mm for
uniaxial compression test. To study the effect of nano-particles
on the RAC, the natural aggregate concrete (NAC), RAC, nano-
silica modified RAC (NS-RAC) and nano-limestone modified RAC
(NL-RAC) were prepared by an effective water to cement ratio of
0.5 for comparison.
Due to the high water absorption of RCA, an additional amount
of water was added to assure the same effective water to cement
ratio for all the nano-particles modified RAC specimens. The addi-
tional water was calculated by the water absorbed from air-dry
condition to saturated-surface-dried (SSD) condition [5,23]. There
were two mixing methods (Method 1 and Method 2) for the RAC
preparation, as shown in Fig. 1. For the first mixing method with
the insight from the two-stage mixing method developed by Tam
et al. [5], RCA and nano-particles were firstly mixed for 60 s, and
subsequently mixed for another 30 s with half of the mixing water
added. Then, cement, sand and the remaining water were added
and mixed for a further 60 s. For the second mixing method, RCA,
SO3 CaO MgO K2O Na2O Total
2.00 65.40 3.40 – – 99.40
44 W. Li et al. / Construction and Building Materials 142 (2017) 42–50

Table 2
Physiochemical properties of nanolimestone.

Composition Appearance Density (g/cm3) Size (nm) PH Content Character

Indicators White powder 2.5–2.6 15–40 8–9 98.5% Hydrophilic

Table 3
Physiochemical properties of colloidal nano-silica.

Index property Appearance Content Size (nm) PH Density Viscosity Specific surface
Indicators Liquid 30 ± 1% 8–15 8–9.5 1.19–1.21 g/ml 35cP 320–400 m2/g

Table 4
Mix proportions of RAC (kg/m3).

Specimens Effective w/c ratio Nano-particles (NS and NL) Cement Sand Coarse Tap water Superplasticizer (%) Mixing method
aggregate
NCA RCA Mixing Additional
NS-RAC S1 3.8 (NS) 376.2 622 845 363 190 15 0 Method 1
S2 7.6 (NS) 372.4 622 845 363 190 15 1.0
S3 3.8 (NS) 376.2 622 845 363 190 15 0 Method 2
S4 7.6 (NS) 372.4 622 845 363 190 15 1.0
NL-RAC L1 3.8 (NL) 376.2 622 845 363 190 15 0 Method 1
L2 7.6 (NL) 372.4 622 845 363 190 15 1.0
L3 3.8 (NL) 316.2 622 845 363 160 15 0 Method 2
L4 7.6 (NL) 312.4 622 845 363 160 15 1.0
RAC 0 380.0 622 845 363 190 15 0 Method 2
NAC 0 380.0 622 1208 0 190 0 0 Method 2

Each RAC mixture has three duplicate specimens. Nano-limestone is ultrasonically dispersed with water.

Fig. 1. Two different mixing methods for the nano-particles modified RACs.

cement and sand were firstly mixed for 30 s, and subsequently
mixed for another 60 s with half of the mixing water. Secondly,
nano-particles and the remaining water were added and mixed
for 60 s.
The compressive experiment on nano-particles modified RAC
was conducted the in Key Laboratory of Building Safety and Energy
Efficiency, Hunan University, Ministry of Education of China. The
current experimental system consists of three parts: a digitally
controlled servo-hydraulic loading system with a capacity of
6000 kN, a computer vision DIC system and image analysis soft-
ware. The loading was applied with a displacement control and
the loading rate was controlled at 0.04 mm/min. The image records
of the whole front surface of RAC specimen was captured with a
camera of the DIC during testing. Two Teflon sheets of 0.2 mm
thickness were inserted between the specimen and the machine
loading plates for reducing the frictional shear constraints. Strain
contour maps were produced to analyze the crack initiation and
propagation during compressive loading. The SEM (FEI Quanta
200 FEG) equipped with an energy dispersive X-ray spectroscopy
was used to examine the microstructure of ITZ regions. The sam-
ples are small fractured pieces of hardened RAC samples obtained
after mechanical strength testing and soaked in acetone to stop the
 W. Li et al. / Construction and Building Materials 142 (2017) 42–50
hydration (old and new ITZs in RAC). The MIP is based on the prin-
ciple that mercury, a typical non-wetting liquid, can only intrude a
porous material if a certain pressure is applied on the measured
samples. MIP samples around the new ITZ region of each type of
concrete were broken into several small particles with diameters
of 3.5–5.0 mm and soaked in acetone to stop further hydration.
Then they were dried at 60 °C in oven for 24 h before MIP exami-
nation. The experiments were carried out under low pressure of
0.28 MPa and high pressure of 413.70 MPa respectively. A glass
tube with the sample and mercury in was subsequently placed in
a low and high pressure analysis port [40]. For the water absorp-
tion measurement, the three replicated samples of each concrete
type were cured in standard curing condition for 28 days. After
this, they were taken out from the curing room and allowed to
dry superficially for about 2 h in the laboratory. Their cube surface
was subsequently coated with epoxy adhesive paint and the next
day they were put in a well-ventilated oven at 105 °C for 2 days.
At the end of this period, the samples were taken out from the
oven, were wrapped with thin plastic foil and stored in the labora-
tory for 2 days [41,42]. Water capillary absorption measurements
were started at when the samples cooled from oven. The samples
were put in a water tank with the bottom-as-cast surface
immersed up to 2 mm in the water for a period of 6 h. The water
level in the tank was kept constant during this period. The samples
were taken out from the water tank at intervals of 10 min, 30 min,
1, 2, 3, 4, 5 and 6 h were weighed on a 0.01 g balance, after being
wiped with a dry paper towel.
3. Results and discussions
3.1. Crack propagation
The crack pattern of nano-modified RAC was observed at differ-
ent stages of compressive loading. The RAC specimens of
150  150  30 mm were tested to investigate the real-time crack
propagation under compressive loading using digital image corre-
lation (DIC) technique [43,44]. Under uniaxial compression, RAC
specimen can be considered as a state of plane stress, the 2D-
(a) 30 % of peak load (b) 60 % of peak load
Fig. 2. Crack propagation of RAC using grayscale map of strain by DIC.
45
digital image correlation was capable of examining the two-
dimensional cracking propagation. The horizontal strain distribu-
tion on the surface of the RAC was presented as a series of gray
scale maps. Based on the real-time record of crack propagations
captured by DIC, it was found that the failure processes of the
RAC specimens were similar. But compared to the crack propaga-
tions, the crack evolutions in each of the stages were different for
the RAC with different nano-particle modifications. During the first
stage, when the compressive load was 30% of the peak load, the
stress level was relatively low. Micro-crack initiations surrounding
the recycled coarse aggregate (RCA) were not very obvious, as
shown in Fig. 2. Only a few micro-cracks appeared because the ten-
sile stress was greater than the bonding strength between the
cement mortar and RCA. But the micro-cracks were in a stable
state. During the second stage, when the load level was greater
than 30%, but less than 70% of the peak load, the micro-cracks
around RCA were produced and gradually developed, as shown in
Fig. 3. Among these cracks, the more longitudinal cracks were
found than transverse cracks. During the third stage, when the load
level was greater than 70%, but less than 90% of the peak load, the
width and the amount of cracks dramatically increased as shown
in Fig. 4. The cracks gradually propagated from the ITZs to the
cement mortar areas in RAC. Finally, instantaneous crushing failure
was observed when the cracks developed vertically through the
RAC specimens generally along the loading direction. Overall,
according to the crack propagation patterns with the RAC speci-
mens prepared with different nano-particles (NS and NL), it was
found that the micro-crack initiation was obviously affected by
the relative strength of the new and old cement mortars. The crack
propagation and development pattern information of the nano-
modified RAC at different loading levels during compression was
captured with the DIC strain grayscale images. The image analysis
software (Optical Fringe Pattern Analysis) was then used to calcu-
late the displacements and strain distribution over the specimen
surface. The results were found consistent with the previous
results [19,45]. The micro-crack initiation and development were
captured much earlier by the DIC than by the naked human eye.
As for the RAC, most cracks propagated through both the ITZ
(c) 80 % of peak load
46 W. Li et al. / Construction and Building Materials 142 (2017) 42–50

(a) 30 % of peak load (b) 60 % of peak load (c) 80 % of peak load

Fig. 3. Crack propagation of NL-RAC using grayscale map of strain by DIC.

(a) 30 % of peak load (b) 60 % of peak load (c) 80 % of peak load

Fig. 4. Crack propagation of NS-RAC using grayscale map of strain by DIC.

regions and old cement mortar due to the comparable strength
between the new and the old cement mortars, as shown in Fig. 2.
With respect to NL-RAC, most cracks propagated through the ITZ
regions and the old cement mortar because the agglomeration
effect made the ITZs even weaker when NL was added, as shown
in Fig. 3. For NS-RAC, the RCA was obviously crushed and the cracks
developed through old cement mortar, which could be attributed
to the much stronger bonding strength between the new and old
cement mortars when NS was added, as shown in Fig. 4. Therefore,
it can be concluded that the crack initiation and propagation pat-
terns of RAC under compression was dependent on the relative
strengths of new and old ITZs, which were affected by the nano-
particles modification.
3.2. Nano-particles modification
From Fig. 5, it was found that the NS-modified RAC (S1, S2, S3
and S4) exhibited higher 7-day compressive strength compared
to both RAC and NAC. Compared to the RAC without nano-
particles, the improvement of compressive strength varied from
20.7% to 24.9%. As shown in Fig. 6, the 28-day compressive
strength also increased by 14.6% to 21.6% when NS was added.
 W. Li et al. / Construction and Building Materials 142 (2017) 42–50 47

45 dispersion of NL, which usually causes particles agglomeration in

NS-RAC RAC. The result also implied that the superplasticizer can some-
40 NL-RAC what reduce the NL agglomeration. As the NL was not able to
Compressive strength (MPa)

35 enhance the strength of RAC, NL actually played a negative role

NAC
RAC
30

25

20

15

10

0
NAC RAC S1 S2 S3 S4 L1 L2 L3 L4
Nano-particles modified RAC

Fig. 5. Compressive strength of nano-modified RAC after 7-day curing.

60
55 NS-RAC
Compressive strength (MPa)

50 (a) RAC
NL-RAC
45 NAC RAC
40
35
30
25
20
15
10
5
0
NAC RAC S1 S2 S3 S4 L1 L2 L3 L4
Nano-particles modified RAC

Fig. 6. Compressive strength of nano-modified RAC after 28-day curing.

The compressive strengths of NS-RAC cured after both 7-day and

28-day were also evidently higher than those NAC. The results
showed that NS was more effective to improve the early-age (b) NL-RAC
strength of RAC than the late-age strength. The result also con-
firmed that the NS significantly enhanced the compressive
strength by promoting secondary cement hydration. Therefore, it
implies that the NS increased the C-S-H gel formation as a result
of the reaction between NS and calcium hydroxide (CH) and filled
the spaces in the C-S-H crystal lattice as physical nano-fillers
[46,47]. By comparing the 7-day and 28-day compressive strengths
of RAC, it was found that at the same NS dosage, there was no dis-
tinct difference between the mixing Method 1 and 2. This means
that the mixing methods did not obviously affect the NS modifica-
tion of RAC, because Method 1 could not evidently densify the
microstructure and increased the compressive strength by premix-
ing the RCA and nano-particles. This indicates that NS modification
of RAC was mainly due to the secondary chemical reaction, while
just slightly due to the physical filling effect [5].
However, for the NL modification, it was found that the 7-day
compressive strengths of L1 and L2 were higher than that of RAC,
but the 7-day compressive strengths of L3 and L4 were slightly
lower than that of RAC. On the other hand, the 28-day compressive (c) NS-RAC
strengths of the NL-modified RAC were relatively lower than that
of RAC without nano-particles. This may be attributed to the poor Fig. 7. Microstructure of new and old ITZs in nano-particles modified RAC.
48 W. Li et al. / Construction and Building Materials 142 (2017) 42–50

5.0
4.5
4.0

Water absorption (%)

3.5
3.0
2.5
2.0
RAC
1.5
NL-RAC
1.0 NS-RAC
0.5
0.0
0 1 2 3 4 5 6
(a) NL-RAC Time (hours)

Fig. 9. Water absorption of nano-particles modified RAC after different curing time.

new ITZ of NL-modified RAC than the NS-modified RAC. The results
showed that for NS-RAC, both the new and old cement mortar
appeared denser. But for NL-RAC, there were still some micro-
cracks in the cement mortar. NS was more likely to enter into ITZs
to promote the chemical reaction and accelerate the hydration pro-
cess. As a result, more hydration products such as C-S-H gels were
found and both the new and old ITZs became significantly denser.
However, for NL, due to the poor dispersion, it was more difficult to
the NL particle to enter into the ITZs to accelerate chemical
reactions.

3.4. Water absorption

(b) NS-RAC
Fig. 8. Highlighted ITZs of nano-particles modified RAC.
in the RAC modification because of the poor dispersion of NL pow-
der. Therefore, the physical filling effect and effect of crystal
nucleus of NL seemed not significant because of the nano-
particles agglomeration involved. Although superplasticizer
seemed to improve the NL dispersion in RAC to a certain degree,
alternative effective methods are still needed to be explored to
improve the dispersion of NL powder for the modification.
3.3. Microstructure properties
The microstructures of new and old ITZs in RAC were character-
ized by SEM as shown in Fig. 7. It was found that compared to RAC,
both the NS- and NL-modified RAC had denser microstructures and
less micro-cracks in ITZs, especially in the NS-RAC. The results
showed that NS rendered a better modification effect compared
to NL in terms of strength improvement and porosity reduction.
From Fig. 8, it was found that there are more micro-cracks in the
The transport properties and capillary absorption were greatly
influenced by the microstructures and porosities of the concrete
specimens. The water absorption values for nano-particle modified
RAC are shown in Fig. 9. It is found that the water absorption val-
ues of RAC, NL-RAC and NS-RAC after 6 h were 4.52%, 4.00% and
3.62%, respectively. The RAC exhibited the highest water absorp-
tion, while NS-RAC had the lowest water absorption. This sug-
gested that under the same conditions, both NS and NL
incorporations can reduce the water absorption by improving
microstructures and decreasing porosity in RAC. It should be noted
that the reduction in the water absorption of the NS-modified RAC
was more significant when compared to the NL-modified RAC.
Although, the agglomeration significantly reduced the bonding
strength of ITZ, and eventually improvement in compressive
strength of NL-RAC, NL was still effective in densify the microstruc-
tures as physical fillers.
3.5. Porosity and pore size distribution
It showed that the total porosities of ITZs regions in RAC mod-
ified with NS and NL became obviously less than that of RAC. From
Table 5, the total porosity of RAC was 12.00%, while NL-RAC and

Table 5
Pore size distribution of nano-particle modified RAC.

Specimens Total porosity (%) Averagepore diameter (nm) Pore size distribution (%)
0–10 nm 10–50 nm 50–1000 nm >1000 nm
RAC 12.00 48.34 7.2 9.2 43.2 40.4
NL-RAC 8.82 71.65 2.8 10.8 37.6 48.8
NS-RAC 9.06 52.64 7.7 4.8 60.4 27.1
 W. Li et al. / Construction and Building Materials 142 (2017) 42–50 49

14 to that of NL-RAC, but the pore size distribution was very different.
It was found that the NS-RAC had a very narrow porosity distribu-
12 tion peak and the lowest critical pore width value. The threshold
RAC and critical pore width can provide a useful indicator of mechanical
NL-RAC strengths because they played an important influence on the
10
NS-RAC microstructure characteristics of the RAC. It was observed that
Porosity (%)

8 the most probable pore for RAC, NL-RAC and NS-RAC were
28,070 nm, 188,600 nm and 132.5 nm, respectively. The results
showed that the pores of ITZ were mainly composed of capillary
6
pore (from 50 nm to 1.0 lm) and macro-pores (larger than
1.0 lm) in RAC and NL-RAC, while in NS-RAC, capillary-pores (from
4
50 nm to 1.0 lm) accounted for most of the total pores. Therefore,
the NS was effective in improving the microstructure and signifi-
2 cantly reduced the macro-scale pores in RAC, but somewhat
increased the capillary-pores. However, although the NL partially
0 reduced the capillary-pores by the physical filling effect, but com-
1 10 100 1000 10000 100000
pared to the RAC without nano-particles, it could not significantly
Pore diameter (nm) reduce the macro-pores. Therefore, NL was not able to reduce the
(a) Cumulative pore distribution after 28-day curing total porosity of RAC and increase the compressive strength, which
was consistent with the compressive strength results.
0.07

0.06
RAC 4. Conclusions
NL-RAC
0.05 NS-RAC The effects of nano-particles on the crack propagation and
microstructural properties of RAC were experimentally investi-
dV/(lgd)(ml/g)

0.04 gated in this study. Based on results, the following conclusion

0.03
0.02
0.01
0.00
10 100 1000 10000
Size (nm)
(b) Differential pore size distribution after 28-day curing
Fig. 10. Porosity and pore size distribution of nano-particles modified RAC.
NS-RAC exhibited very similar results of around 8.82% and 9.06%,
respectively. This result can be used to explain why RAC, NL-RAC
and NS-RAC exhibited increasing trend in the water absorption.
The higher porosity usually results in higher water absorption or
transport properties. Furthermore, the porosity with diameter
greater than 1000 nm was 40.4% in RAC, while the porosity of
NL-RAC and NS-RAC were 48.8% and 27.1%, respectively. Compare
to the RAC without nano-particles, the amount of pores greater
than 1000 nm were reduced by 13.3% for NS-RAC, but were
increased by 8.4% for NL-RAC. This indicates that the NS can
improve the microstructures and reduce the porosity of NS-RAC,
while NL seems not as effective as in reducing the large-size pores
NL-RAC. However, the NL significantly reduced the pores with size
from 50 to 1000 nm by 13% in NL-RAC, while the NS adversely
increased the pores in range of 50–1000 nm by 40% in NS-RAC. It
implies that the NL is more effective in reducing the medium-
size pores in the ITZs than the NS in nano-particle modified RAC.
The measured pores can be divided into five size ranges, includ-
ing gel micro-pores with diameter less than 10 nm, meso-pores of
10–50 nm, capillary pores of 50 nm to 1.0 lm and macro-pore lar-
ger than 1.0 lm [48,49]. The RAC, NL-RAC and NS-RAC exhibited
different ranges in the pores size distribution and cumulative
porosity, as shown in Fig. 10. The porosity of NS-RAC was very close
can be drawn:
(1) The micro-crack typically initiated in the weak ITZs of RAC
with different nano-particle modifications. The initiation
location was affected by the relative bonding strength
between new and old cement mortar. The micro-cracks
propagated along the loading direction, finally leading to
wholly splitting failure.
100000 (2) The micro-crack initiation and propagation provided funda-
mental information for the failure mechanism of RAC. When
the strength of new ITZ was inferior to that of old ITZ, the
first observable crack usually appeared at the new ITZ
region. But when the strength of the new ITZ was higher
than that of old ITZ, the first micro-crack mostly occurred
at the old ITZ.
(3) Colloidal NS could easily enter into the ITZs of RAC to accel-
erate hydration process and increase the hydration products
by further chemical reactions. Thus, the further product
growth filled the pores and voids with physical filling effect,
then significantly improved the microstructure properties
and enhanced the RAC compressive strength. The compres-
sive strength of NS-RAC was also evidently higher than that
NAC.
(4) The powdered NL was not effective in improve the bonding
strength between new and old cement mortars in RAC to
enhance the compressive strength because of the particle
agglomeration effect. Furthermore, NL even decreased the
compressive strength at the late age due to the poor disper-
sion of NL. Therefore, alternative effective dispersion meth-
ods are needed to improve the dispersion and reduce the
particle agglomeration.
(5) Colloidal NS was more effective in reducing the micro-scale
pores and the water absorption of NS-modified RAC com-
pared to powdered NL-modified RAC. Although the NL could
reduce the capillary-pores by physical filling effect, it was
still unable to obviously reduce the macro-pores, and to
increase the compressive strength of NL-modified RAC.
50 W. Li et al. / Construction and Building Materials 142 (2017) 42–50
Acknowledgements
The authors gratefully acknowledge the financial support of the
Australian Research Council (DE150101751) and Australian
Research Council Research Hub for Nanoscience Based Construc-
tion Materials Manufacturing (IH150100006). The authors are also
grateful for the financial supports of the National Natural Science
Foundation of China (51408210) and University of Technology Syd-
ney Early Career Researcher Grant, Australia. The constructive sug-
gestions and comments from Professor Surendra P. Shah at
Northwestern University, USA are greatly appreciated.
References
[1] W.G. Li, J. Xiao, C. Shi, C.S. Poon, Structural behavior of composite members
with recycled aggregate concrete – an overview, Adv. Struct. Eng. 18 (6) (2015)
919–938.
[2] S. Senaratne, D. Gerace, O. Mirza, V.W.Y. Tam, W.H. Kang, The costs and
benefits of combining recycled aggregate with steel fibres as a sustainable,
structural material, J. Clean. Prod. 112 (4) (2016) 2318–2327.
[3] J. Xiao, W.G. Li, Y. Fan, X. Huang, An overview of study on recycled aggregate
concrete in China (1996–2011), Constr. Build. Mater. 31 (2012) 364–383.
[4] S.C. Kou, C.S. Poon, M. Etxeberria, Influence of recycled aggregates on long term
mechanical properties and pore size distribution of concrete, Cem. Concr.
Compos. 33 (2011) 286–291.
[5] V.W. Tam, X. Gao, C. Tam, Microstructural analysis of recycled aggregate
concrete produced from two-stage mixing approach, Cem. Concr. Res. 35 (6)
(2005) 1195–1203.
[6] M.C. Rao, S.K. Bhattacharyya, S.V. Barai, Influence of field recycled coarse
aggregate on properties of concrete, Mater. Struct. 44 (2011) 205–220.
[7] N.A. Abdulla, Effect of recycled coarse aggregate type on concrete, J. Mater. Civ.
Eng. 27 (10) (2015) 1–9.
[8] A. Buttress, A. Jones, S. Kingman, Microwave processing of cement and
concrete materials – towards an industrial reality?, Cem Concr. Res. 68 (2015)
112–123.
[9] S.C. Kou, C.S. Poon, Properties of concrete prepared with PVA-impregnated
recycled concrete aggregates, Cem. Concr. Compos. 32 (8) (2010) 649–654.
[10] Anna M. Grabiec, Justyna Klama, Daniel Zawal, Daria Krupa, Modification of
recycled concrete aggregate by calcium carbonate biodeposition, Constr. Build.
Mater. 34 (2012) 145–150.
[11] A.A. Elhakam, A.E. Mohamed, E. Awad, Influence of self-healing, mixing
method and adding silica fume on mechanical properties of recycled
aggregates concrete, Constr. Build. Mater. 35 (2012) 421–427.
[12] C. Shi, Y. Li, J. Zhang, W.G. Li, L. Chong, Z. Xie, Performance enhancement of
recycled concrete aggregate – a review, J. Clean. Prod. 112 (2016) 466–472.
[13] J.S. Qiu, D. Tng, E.H. Yang, Surface treatment of recycled concrete aggregates
through microbial carbonate precipitation, Constr. Build. Mater. 57 (2014)
144–150.
[14] H. Zhang, Y. Zhao, T. Meng, S.P. Shah, Surface treatment on recycled coarse
aggregates with nanomaterials, J. Mater. Civ. Eng. 28 (2) (2016) 1–11.
[15] S. Ismail, M. Ramli, Mechanical strength and drying shrinkage properties of
concrete containing treated coarse recycled concrete aggregates, Constr. Build.
Mater. 68 (2014) 726–739.
[16] J. Ryou, Y. Lee, Characterization of recycled coarse aggregate via a surface
coating method, Int. J. Concr. Struct. Mater. 8 (2) (2014) 165–172.
[17] Y.L. Liang, Z.M. Ye, F. Vernerey, Development of processing methods to
improve strength of concrete with 100% recycled coarse aggregate, J. Mater.
Civ. Eng. 27 (5) (2015) 1–9.
[18] M. Casuccio, M.C. Torrijos, G. Giaccio, R. Zerbino, Failure mechanism of
recycled aggregate concrete, Constr. Build. Mater. 22 (2008) 1500–1506.
[19] J. Xiao, W.G. Li, Z.H. Sun, S.P. Shah, Crack propagation in recycled aggregate
concrete under uniaxial compressive loading, ACI Mater. J. 109 (4) (2012) 451–
461.
[20] J. Xiao, W.G. Li, D.J. Corr, S.P. Shah, Effects of interfacial transition zones on the
stress-strain behavior of modeled recycled aggregate concrete, Cem. Concr.
Res. 52 (2013) 82–99.
[21] J. Xiao, W.G. Li, Z.H. Sun, D.A. Lange, S.P. Shah, Properties of interfacial
transition zones in recycled aggregate concrete tested by nanoindentation,
Cem. Concr. Compos. 37 (2013) 276–292.
[22] A. Sidorovaa, E. Vazquez-Ramonichb, M. Barra-Bizinotto, J.J. Roa-Rovira, E.
Jimenez-Pique, Study of the recycled aggregates nature’s influence on the
aggregate-cement paste interface and ITZ, Constr. Build. Mater. 68 (3) (2014)
677–684.
[23] W.G. Li, J. Xiao, Z. Sun, S. Kawashima, S.P. Shah, Interfacial transition zones in
recycled aggregate concrete with different mixing approaches, Constr. Build.
Mater. 35 (2012) 1045–1055.
[24] C.S. Poon, Z.H. Shui, L. Lam, Effect of microstructure of ITZ on compressive
strength of concrete prepared with recycled aggregates, Constr. Build. Mater.
18 (6) (2004) 461–468.
[25] W.G. Li, Z. Luo, C. Long, C. Wu, W.H.P. Duan, S.P. Shah, Effects of nanoparticle
on the dynamic behaviors of recycled aggregate concrete under impact
loading, Mater. Des. 112 (2016) 58–66.
[26] D. Kong, D.J. Corr, P. Hou, Y. Yang, S.P. Shah, Influence of colloidal silica sol on
fresh properties of cement paste as compared to nano-silica powder with
agglomerates in micron-scale, Cem. Concr. Compos. 63 (2015) 30–41.
[27] S.P. Shah, P.K. Hou, S. Maria, M.S. Konsta-Gdoutos, Nano-modification of
cementitious material: toward a stronger and durable concrete, J. Sustain.
Cem. Based Mater. 5 (1–2) (2016) 1–22.
[28] W.G. Li, Z. Huang, F. Cao, Z. Sun, S.P. Shah, Effects of nano-silica and nano-
limestone on flow ability and mechanical properties of ultra-high-
performance concrete matrix, Constr. Build. Mater. 95 (1) (2015) 366–374.
[29] W.G. Li, Z. Huang, T. Zu, C. Shi, W. Duan, S.P. Shah, Influence of nanolimestone
on the hydration, mechanical strength, and autogenous shrinkage of ultrahigh-
performance concrete, J. Mater. Civ. Eng. 28 (1) (2015) 315–323.
[30] N. Farzadnia, A.A. Ali, R. Demirboga, Characterization of high strength mortars
with nano alumina at elevated temperatures, Cem. Concr. Res. 54 (2013) 43–
54.
[31] R. Polat, R. Demirboga, W.H. Khushefati, Effect of nano and micro size of CaO
and MgO, nano-clay and expand perlite aggregate on the autogenous
shrinkage of mortar, Constr. Build. Mater. 81 (2015) 268–275.
[32] R. Zhang, X. Cheng, P. Hou, Z.M. Ye, Influences of nano-TiO2 on the properties
of cement-based materials: hydration and drying shrinkage, Constr. Build.
Mater. 81 (2015) 35–41.
[33] Z. Pan, L. He, L. Qiu, A.H. Korayem, G. Li, J.W. Zhu, F. Collins, D. Li, W.H. Duan, M.
C. Wang, Mechanical properties and microstructure of a graphene oxide-
cement composite, Cem. Concr. Compos. 58 (2015) 140–147.
[34] K. Gong, Z. Pan, A. Korayem, L. Qiu, D. Li, F. Collins, C. Wang, W.H. Duan,
Reinforcing effects of graphene oxide on Portland cement paste, J. Mater. Civ.
Eng. A4014010 (2014).
[35] C.Q. Lin, W. Wei, Y.H. Hu, Catalytic behavior of graphene oxide for cement
hydration process, J. Phys. Chem. Solids 89 (2016) 12–133.
[36] B.M. Tyson, R.A. Al-Rub, A. Yazdanbakhsh, Carbon nanotubes and carbon
nanofibers for enhancing the mechanical properties of nanocomposite
cementitious materials, J. Mater. Civ. Eng. 23 (7) (2011) 1028–1035.
[37] Z.S. Metaxa, M.S. SeoKonsta-Gdoutos, M.C. Hersam, S.P. Shah, Highly
concentrated carbon nanotube admiture for nano fiber reinforced
cementitious materials, Cem. Concr. Compos. 34 (2012) 612–617.
[38] K. Sobolev, M. Ferrada-Gutiérrez, How nanotechnology can change the
concrete world: part 2, Am. Ceram. Soc. Bull. 84 (11) (2005) 16–19.
[39] S.J. Lee, W.M. Kriven, Synthesis and hydration study of Portland cement
components prepared by the organic steric entrapment method, Mater. Struct.
38 (1) (2005) 87–92.
[40] Z.M. Wu, C.J. Shi, K.H. Khayat, Influence of silica fume content on
microstructure development and bond to steel fiber in ultra-high strength
cement-based materials (UHSC), Cem. Concr. Compos. 71 (2016) 97–109.
[41] T.C. Rilem, 116-PCD: permeability of concrete as a criterion of its durability.
Final report. Concrete durability – an approach towards performance testing,
Mater. Struct. 32 (1999) 163–173.
[42] S. Kolias, C. Georgiou, The effect of paste volume and of water content on the
strength and water absorption of concrete, Cem. Concr. Compos. 27 (2005)
211–216.
[43] S. Choi, S.P. Shah, Propagation of microcracks in concrete studied with
subregion scanning computer vision (SSCV), ACI Mater. 96 (2) (1999) 255–260.
[44] D. Corr, M. Accardi, L. Graham-Brady, S. Shah, Digital image correlation
analysis of interfacial debonding properties and fracture behavior in concrete,
Eng. Fract. Mech. 74 (1–2) (2007) 109–121.
[45] W.G. Li, J. Xiao, Z.H. Sun, S.P. Shah, Failure processes of modeled recycled
aggregate concrete under uniaxial compression, Cem. Concr. Compos. 34 (10)
(2012) 1149–1158.
[46] P.K. Hou, S. Kawashima, K.J. Wang, D.J. Corr, J.S. Qian, S.P. Shah, Effects of
colloidal nanosilica on rheological and mechanical properties of fly ash-
cement mortar, Cem. Concr. Compos. 35 (1) (2013) 12–22.
[47] S. Kawashima, J.T. Seo, D. Corr, M. Hersam, S.P. Shah, Dispersion of CaCO3
nanoparticles by sonication and surfactant treatment for application in fly ash-
cement systems, Mater. Struct. 47 (6) (2013) 1011–1023.
[48] Q. Zeng, K.F. Li, T.F. Chong, P. Dangla, Pore structure characterization of cement
pastes blended with high-volume fly-ash, Cem. Concr. Res. 42 (1) (2012) 194–
204.
[49] P.K. Metha, P.J.M. Monterio, Concrete, Microstructure, Properties and
Materials, McGraw-Hill, London, 2006.