Construction and Building Materials 419 (2024) 135540

Contents lists available at ScienceDirect

Construction and Building Materials

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

Study on fracture characteristics and fracture mechanism of fully recycled

aggregate concrete using AE and DIC techniques
Yingjie Xu a, b, Hongniao Chen a, b, c, d, *, Yihui Liang a, b, Jie Shen a, b, Huaxiang Yang a
a
Research Center of Space Structures, Guizhou University, Guiyang 550025, China
b
College of Civil Engineering, Guizhou University, Guiyang 550025, China
c
Key Laboratory of Structural Engineering of Guizhou Province, Guiyang 550025, China
d
Key Laboratory of Rock and Soil Mechanics and Engineering Safety of Guizhou Province, Guiyang 550025, China

A R T I C L E I N F O A B S T R A C T

Keywords: Fracture behaviors of manufactured sand-recycled coarse aggregate concrete (MSRCAC), recycled fine aggregate
Fully recycled aggregate concrete concrete (RFAC), and fully recycled aggregate concrete (FRAC) were investigated by performing three-point
Recycled coarse aggregate bending tests on single-edge notched beams using acoustic emission (AE) and digital image correlation (DIC)
Recycled fine aggregate
techniques. The effects of recycled fine aggregate (RFA), recycled coarse aggregate (RCA), and their interaction
Fracture parameter
Fracture process zone
on fracture parameters, AE characteristics and fracture process zone (FPZ) of concrete were comparatively
AE characteristics analyzed. The results indicated that it was mainly transgranular failure and intergranular failure for RFAC, while
transgranular failure for MSRCAC and FRAC. The RFA and RCA reduced the initial fracture toughness, unstable
fracture toughness and fracture energy of concrete, and their interaction resulted in poor fracture properties of
FRAC. In addition, the RFA and RCA significantly increased the AE activities, the proportion of tensile cracks,
and the high-frequency amplitude and peak frequency of concrete. The toughening effect of FPZ was weakened
by the RFA and RCA, resulting in an obvious increase in FPZ length and damage propagation of FRAC. Mean­
while, the width of FPZ was reduced, the maximum FPZ width of concrete with different combinations of
recycled aggregates was approximately 1.8–3.1 times the maximum aggregate size, which could be explained as
the RFA and RCA reduced the aggregate interlocking effect of concrete, while the RCA had a greater effect on the
damage mechanism and FPZ of concrete than the RFA.

causes a significant waste of resources and potential environmental

pollution[4]. The depletion of natural aggregates and the accumulation
1. Introduction of waste concrete will affect the sustainable development of the con­
struction industry, it is urgent to find a method to reduce the con­
With the rapid development of the construction industry, a large sumption of natural aggregates and recycle waste concrete[7].
number of commercial buildings and public facilities have been con­ The use of recycled aggregates and manufactured sand as building
structed, urbanization and industrialization have led to the consumption materials to prepare recycled aggregate concrete (RAC) can effectively
of large quantities of natural aggregates, especially river sand[1]. About alleviate the depletion of natural aggregates and the accumulation of
2.6 billion cubic meters of river sand was mined every year in China[2]. waste concrete[8], which is critical to the sustainability of the con­
Currently, many countries are already facing the depletion of river sand, struction industry[9]. However, this measure has many problems.
while excessive exploitation of river sand will cause significant social Compared to river sand, manufactured sand has sharper edges and more
and environmental impacts[3]. Meanwhile, the urban expansion, de­ powder content[10], while manufactured sand can be substituted for
molition of old houses, and expiration of building lifespan have gener­ river sand through mixed design[11,12]. Furthermore, the mechanical
ated a large amount of abandoned concrete. In 2021, the construction properties, durability, and fracture properties of RAC are lower than
waste reached 3.2 billion cubic meters in China[4]. In fact, waste con­ those of normal concrete due to the old mortar on the surface of recycled
crete accounts for approximately 40% of the total waste in the world[5], aggregates[13–15]. Recently, the mechanical properties and durability
while the recovery rate of waste concrete is very low. In many countries, of RAC have been studied extensively[16–18], while little attention has
construction waste is still stored outdoors or directly buried[6], which

* Corresponding author at: Research Center of Space Structures, Guizhou University, Guiyang 550025, China.
E-mail address: [email protected] (H. Chen).

https://doi.org/10.1016/j.conbuildmat.2024.135540
Received 15 December 2023; Received in revised form 31 January 2024; Accepted 19 February 2024
Available online 28 February 2024
0950-0618/© 2024 Elsevier Ltd. All rights reserved.
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Nomenclature R rise time

A amplitude
fcu cube compressive strength Dt duration time
fts splitting tensile strength εt peak tensile strain
L length of beam w0 stress-free crack opening displacement
D height of beam lFPZ length of FPZ
t thickness of beam Δa crack propagation length
a0 notch length WFPZ width of FPZ
δ mid-span deflection WFPZ,max maximum WFPZ
S span of beam MS manufactured sand
kini cracking nominal stiffness RFA recycled fine aggregate
Pini crack initiation load RCA recycled coarse aggregate
δini δ at Pini MSNAC MS normal aggregate concrete
kc critical nominal stiffness MSRCAC MS concrete with 100% RCA replacement rate
δc δ at Pc RFAC concrete with 100% RFA replacement rate
E elastic modulus FRAC concrete with 100% RFA and RCA replacement rates
CMODc crack mouth opening displacement at peak load RAC recycled aggregate concrete
CTODc crack tip opening displacement at peak load ITZ interfacial transition zone
ac critical effective crack length RCAC recycled coarse aggregate concrete
CMODd CMOD considering the effect of additional central load FPZ fracture process zone
Pd additional load DIC digital image correlation
Kini
IC initial fracture toughness AE acoustic emission
KunIC unstable fracture toughness CMOD crack mouth opening displacement
GF fracture energy CTOD crack tip opening displacement
ft uniaxial tensile strength COD crack opening displacement

been paid to its fracture properties, especially fully recycled aggregate of FRAC.
concrete (FRAC), which is prepared by completely replacing natural It should be noted that RAC has regional characteristics[34]. Due to
aggregates with recycled fine and coarse aggregates. After macroscopic the different natural aggregate sources, material mix proportion, age,
cracks appear in concrete, the traditional strength theory is no longer and service environment, the quality of original concrete will affect the
suitable for evaluating the reliability of concrete structures[19]. The fracture properties of RAC[7,14,35]. In addition, the above studies
fracture properties are critical for the durability and reliability evalua­ almost used river sand as the raw material for concrete and few re­
tion of concrete structures[15,20]. Therefore, it is necessary to study the searchers focused on manufactured sand-recycled coarse aggregate
fracture behaviors of RAC. concrete (MSRCAC). The manufactured sand with different lithologies
The recycled coarse aggregate (RCA) and recycled fine aggregate may affect the fracture properties of manufactured sand concrete[1,36].
(RFA) complicate the interfacial transition zone (ITZ) and pore structure In general, the use of MSRCAC, RFAC and especially FRAC will bring
of RAC[21,22]. Wu et al.[23] indicated that the fracture energy and huge environmental and economic benefits, and play an important role
fracture toughness of recycled coarse aggregate concrete (RCAC) in the realization of “carbon peak” and “carbon neutralization”. How­
decreased with the increase of RCA replacement rate. Guo et al.[24] ever, the fracture mechanism of RAC is still vague, and the effect of the
noted that the RCA accelerated the rate of crack propagation in RCAC. interaction of RFA and RCA on the fracture behaviors of concrete is not
Furthermore, the use of RCA discretized the shape of the RCAC softening clear. These uncertainties limit the engineering applications of
curve[25]. Kazemian et al.[26] also indicated that RCAC exhibited MSRCAC, RFAC and FRAC.
worse nonlinear fracture behaviors than normal concrete. The second­ Digital image correlation (DIC) and acoustic emission (AE) tech­
ary cracks and damage width of RCAC decreased with increasing RCA niques are well applied to characterize surface deformation and damage
replacement rate[27]. The fracture behaviors of concrete are closely evolution in materials[37,38]. Currently, a number of researchers have
related to the damage zone of the crack front (fracture process zone, investigated the fracture behaviors of concrete using DIC or AE tech­
FPZ)[28]. Moreover, the FPZ may be the intrinsic factor of the size effect niques[32,39,40]. However, there are fewer reports on the simultaneous
of fracture parameters of concrete[29,30]. Currently, little attention has use of both techniques to evaluate the fracture characteristics of con­
been paid to the FPZ of RAC. Xu et al.[27] indicated that RCA reduced crete, especially for RAC[24]. It is noteworthy that the DIC technique
the aggregate interlocking effect of RCAC, and the toughening effect of can only provide the measurement of surface deformation while the AE
FPZ of concrete decreased with the increase of RCA replacement rate. technique can monitor internal damages, thus the use of both techniques
The local fracture mechanism of concrete was changed by RCA[24]. can realize mutual corroboration of the experimental observations,
Compared to RCAC, there is a paucity of studies on the fracture which is helpful to reveal the fracture mechanisms of MSRCAC, RFAC
properties of recycled fine aggregate concrete (RFAC) and FRAC. The and FRAC.
use of RFA increases the risk of cracking in concrete[31]. Tang et al.[32] To investigate the effect of the RFA, RCA, and interaction of RFA and
indicated that the fracture parameters of RFAC decreased with RCA on the fracture properties, FPZ, and fracture mechanism of con­
increasing RFA replacement rate. Akono et al.[33] also noted that RFA crete, the fracture parameters, AE characteristics, and FPZ characteris­
reduced the crack propagation resistance of concrete. The microstruc­ tics of RFAC, MSRCAC and FRAC were studied by DIC and AE
ture of RFAC was the fundamental factor causing the complexity of its techniques. First, the failure characteristics, load-displacement curves,
fracture properties[21]. The fracture mechanism of FRAC under the critical parameters, fracture toughness, and fracture energy of concrete
interaction of RFA and RCA will be more complicated. Currently, there with different combinations of recycled aggregates were compared and
are few studies on the fracture properties of FRAC, and only Xiao et al. analyzed. Then, the AE characteristic parameters such as hit count, ring
[15] conducted a preliminary investigation on the fracture parameters count, RA-AF, amplitude and peak frequency were evaluated. Last, the

2
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

characteristics of FPZ, including FPZ evolution, FPZ length and FPZ 2.2. Three-point bending test
width, were qualitatively and quantitatively analyzed using AE and DIC
techniques. Fig. 3 shows the three-point bending test setup. The size of the single-
edge notched beam was 515 mm × 100 mm × 100 mm (L×D×t), and
2. Experimental program the span (S) was 400 mm. The notch depth (a0) was 30 mm and the
width was about 0.5 mm, which was made by a line-cutting machine.
2.1. Materials The loading device was an MTS testing machine with 100 kN and the
loading rate was 0.05 mm/min. A clip gauge was employed to monitor
P•O 42.5 Ordinary Portland Cement was used as the binder. The li­ the crack mouth opening displacement (CMOD), and two linear
thology of natural coarse aggregate (NCA, 5–10 mm) and manufactured displacement sensors (LVDTs) were used to determine the mid-span
sand (MS, <5 mm) was dolomite in Guizhou. The source of waste con­ deflection (δ), and the data obtained were recorded using a TST col­
crete was pavement concrete in Guiyang urban area, it was crushed by a lector. To reduce the gap between the specimen and the testing machine,
jaw crusher to obtain RFA (<5 mm) and RCA (5–10 mm). Since the the high-strength gypsum was placed between the specimen and the
micro-powder content of RFA was not conducive to the mechanical support, and a preload of 100 N was applied.
properties of RAC[41], the RFA was subjected to secondary screening. The DIC technique was used to characterize the full-field displace­
The four types of aggregates are shown in Fig. 1. ment of concrete surfaces[52]. As shown in Fig. 3(c), the size of the
To understand the differences between natural aggregates and monitoring area was 100 mm × 100 mm, and the crack tip opening
recycled aggregates, the physical properties tests were conducted on displacement (CTOD) was monitored by DIC technique. The frequency
aggregates according to GB/T 25176–2010[42], GB/T 25177–2010 of image acquisition was 1 picture/s and the VIC-3D software was
[43], GB/T 14684–2022[44], and GB/T 14685–2022[45]. The strength employed to analyze the collected images. The AE technique was used to
of waste concrete was determined according to JGJ/T 384–2016[46]. evaluate the damage development of concrete[37]. Four AE sensors
All the test results are shown in Table 1. It can be seen that the physical were set up, as shown in Fig. 3(d). According to the AE operation manual
properties of RFA and RCA are lower than those of MS and NCA due to and lead breaking results, the preamplifier gain was 34 dB, the filter was
the old mortar on the aggregate surface[47], while the particle size 25–300 kHz, the threshold was 35 dB, the peak discrimination time was
distribution of RFA after secondary screening was similar to that of MS, 0 us, the hit duration time was 200 us, and the hit locking time was 400
as shown in Fig. 2. us.
The concrete mix is shown in Table 2. Due to the high water ab­
sorption rate of the recycled aggregates, water replenishment measures 3. Results and discussions
were taken to avoid affecting the water-cement ratio of concrete. The
RCA was immersed in water until the saturated surface was dry[48] 3.1. Failure mode
before concrete was mixed. Additional water was added to supplement
water for RFA[49], and the water replenishment rate was 5.06%. Each The fracture mechanism of RAC may be altered due to the large
group of three beams (515 mm×100 mm×100 mm) and six cubes differences between recycled and natural aggregates, the evaluation of
(150 mm×150 mm×150 mm) were prepared for the three-point failure modes is helpful in understanding the failure mechanism of RAC
bending test, cube compression test, and splitting tensile test respec­ [15,25]. The failure modes were obtained by observing the aggregate
tively. The two-stage mixing method was used to mix concrete[50] and boundary of the actual left and right fracture surfaces. For ease of pre­
the test specimens were placed in the concrete curing box for 28 days sentation, the transgranular failure is circled in red and the intergranular
after pouring. failure is circled in blue, as shown in Fig. 4. It can be seen that RFAC and
According to GB50081–2019[51], the compressive and splitting MSNAC have similar failure modes, which are intergranular failure and
tensile tests were performed on the cubes. The test results are shown in transgranular failure. The microporosity of the mortar increases and the
Table 2. It can be seen that RFA and RCA reduced the cube compressive strength decreases due to RFA[33]. The microcracks in the recycled
strength (fcu) and splitting tensile strength (fts) of concrete. Compared mortar of RFAC will propagate first under load[53], while the cracks
with MSNAC, the fcu of RFAC, MSRCAC and FRAC decreased by about that cannot penetrate the NCA will propagate towards the ITZ and
8.2%, 8.3% and 9.4%, respectively, and their fts reduced by about 3.0%, recycled mortar, leading to intergranular failure.
8.3% and 11.0%, respectively. The effect of RCA on fcu and fts is greater However, the fracture surfaces of MSRCAC and FRAC were mainly
than that of RFA. As a concrete skeleton, the coarse aggregate is the main transgranular failure, especially FRAC, as shown in Fig. 4(c) and (d). Due
transfer path of stress, while the mortar only fills the gap in the skeleton to the old mortar on the aggregate surface, the strength of the RCA is
[21]. Therefore, the RCA may have a greater influence on the basic lower than that of the NCA, which makes RCA unable to resist crack
mechanical properties of concrete than RFA, which is consistent with the propagation, resulting in transgranular failure. The RCA, as the skeleton
results of Xiao et al.[9]. of concrete, is a critical factor in resisting crack propagation[25]. When

Fig. 1. Natural and recycled aggregates: (a) NCA; (b) RCA; (c) MS; (d) RFA.

3
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Table 1
Physical properties of aggregates.
Water Water Crush Apparent density Loose bulk Tight bulk density Fineness MB Strength of waste
absorption (%) content (%) index (%) (kg/m3) density (kg/m3) (kg/m3) modulus value concrete (MPa)

NCA 0.38 - 11.34 2708 1490 1648 - - -

RCA 5.58 - 15.47 2660 1325 1417 - - 57.78
MS 2.12 0.96 16.83 2702 1708 1938 2.89 1.10 -
RFA 7.77 2.71 23.01 2649 1465 1709 2.87 1.04 57.78

3.3. Critical parameters

3.3.1. Nominal stiffness

The nominal stiffness is the force required for the unit displacement
of the material, and it is an important parameter for evaluating the
flexural capacity of concrete beams. The crack initiation and peak load
are the critical moments to characterize the fracture of concrete. To
quantify the influence of RFA and RCA on the bearing capacity of con­
crete, the cracking nominal stiffness and critical nominal stiffness of
RAC were compared and analyzed. The cracking nominal stiffness (kini)
is as follows:
Pini
kini = (1)
δini
where Pini is the crack initiation load, and δini is the deflection at Pini.
Due to the lack of data from some strain gauges, this paper uses the AE
Fig. 2. Particle size distribution of MS and RFA.
technique to determine the Pini. The moment of the first mutation of AE
event is defined as the crack initiation[57], as shown in Fig. 6. The
critical nominal stiffness (kc) is as follows:
the crack front is RCA, the crack propagation will be hindered[54],
while due to the aggregate defects, the RCA fractured during the Pc
kc = (2)
microcracks diffusion around it. The RCA changes the fracture mode of δc
concrete, which may affect the fracture mechanism of RAC.
where Pc is the peak load and δc is the deflection at Pc.
The Pini and kini of the RAC are shown in Fig. 7 and Table 3. It can be
3.2. Load-displacement curve
seen that the RFA and RCA reduced the Pini and kini of concrete.
Compared with MSNAC, the Pini of RFAC, MSRCAC and FRAC decreased
Fig. 5 shows the average load (P)-displacement curves of the RAC. It
by about 42.5%, 16.3% and 48.1%, respectively. The Pini of RFAC is
can be seen that P-δ, P-CMOD and P-CTOD curves of concrete with
lower than that of MSRCAC, while it is not significantly different from
different combinations of recycled aggregates show similar trends.
that of FRAC. Compared to coarse aggregates, the influence of the
However, compared with MSNAC, the slope of the rising section of the P-
strength of the cement matrix on Pini is significantly increased, which is
displacement curve of RAC decreased and the slope of the descending
consistent with the research of Xiao et al.[15].
section increased. The use of RFA and RCA reduced the initial stiffness,
In addition, compared to MSNAC, the kini of RFAC, MSRCAC and
bearing capacity, and ductility of concrete, and increased plastic
FRAC decreased by about 3.0%, 11.2% and 12.5%, respectively. It
deformation. Due to the mortar on the surface of recycled aggregate, the
should be noted that the RFAC with lower Pini has higher kini than the
microstructure of aggregates, ITZs and mortars of RAC become
MSRCAC, which implies that the MSRCAC has a larger crack initiation
extremely complex[55,56], resulting in RAC having lower crack resis­
deflection than the RFAC. However, the larger the deformation of con­
tance and faster softening rate than normal concrete[24,27,33]. It
crete, the lower the modulus of elasticity[21], which directly affects the
should be noted that RCA has a greater influence on the initial stiffness,
stiffness of concrete. The sand ratio of this work was 35.7% and the
bearing capacity and plastic deformation of concrete than RFA.
proportion of coarse aggregates was 64.3%, and the coarse aggregate as
Furthermore, FRAC exhibits more obvious nonlinear behavior under the
the skeleton is critical to concrete stiffness. Therefore, the RCA has a
interaction of RFA and RCA.
greater effect on the kini of concrete than the RFA.
Fig. 8 and Table 3 show the Pc and kc of the RAC. Compared with
MSNAC, the Pc of RFAC, MSRCAC and FRAC decreased by about 10.3%,
12.8% and 21.8%, respectively, and the kc reduced by about 20.2%,
22.3% and 26.3%, respectively. The effect of RCA on Pc and kc of

Table 2
Concrete mix proportions (kg/m3).
Groups Cement Water NCA MS RCA RFA Added water fcu (MPa) fts (MPa)

Avg SD Avg SD

MSNAC 443 210 1091 605 - - - 54.50 0.33 3.36 0.20

RFAC 443 210 1091 - - 605 31 50.05 1.79 3.26 0.21
MSRCAC 443 210 - 605 1091 - - 49.98 0.80 3.08 0.03
FRAC 443 210 - 0 1091 605 31 49.36 0.68 2.99 0.08

Notes: MSNAC was a normal MS concrete; RFAC was a concrete with 100% RFA replacement rate; MSRCAC was a MS concrete with 100% RCA replacement rate; FRAC
was a concrete with 100% RFA and RCA replacement rates.

4
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 3. Three-point bending test device: (a) front view; (b) rear view; (c) schematic diagram of front view; (d) schematic diagram of rear view.

Fig. 4. Fracture surfaces of concrete: (a) MSNAC; (b) RFAC; (c) MSRCAC; (d) FRAC. Notes: Red circle: aggregate was broken (transgranular failure); blue circle:
aggregate was pulled out (intergranular failure).

5
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 5. P-displacement curves of RAC: (a) P-δ; (b) P-CMOD; (c) P-CTOD.

Fig. 6. Determination of crack initiation load.

Fig. 7. Pini and kini of RAC: (a) Pini; (b) kini.

concrete is greater than that of RFA. It is noteworthy that the kc of RFAC, and micropores make the microcracks in the RAC extremely active
MSRCAC and FRAC is more significant than the attenuation of their kini, under load[24,58], the rapid coalescence and localization of micro­
this indicates that recycled aggregates increase the nonlinear fracture cracks lead to the increase of nonlinear deformation of RAC after crack
behaviors of RAC after crack initiation. A large number of initial defects initiation.

6
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Table 3
Critical parameters of RAC.
Groups Pini (kN) Pc (kN) kini (kN/mm) kc (kN/mm) δc (mm) CMODc (mm) CTODc (mm) E (GPa) ac (mm)

Avg SD Avg SD Avg SD Avg SD Avg SD Avg SD Avg SD Avg SD Avg SD

MSNAC 2.71 0.14 3.99 0.23 41.3 0.6 37.7 0.1 0.103 0.004 0.022 0.003 0.0091 0.0001 35.8 1.0 45.6 2.6
RFAC 1.56 0.05 3.58 0.13 40.1 0.7 30.0 0.5 0.117 0.002 0.034 0.001 0.0152 0.0011 30.7 0.6 44.3 3.7
MSRCAC 2.27 0.14 3.48 0.10 36.7 1.1 29.3 0.4 0.112 0.008 0.032 0.005 0.0148 0.0033 28.3 0.7 44.4 3.0
FRAC 1.41 0.10 3.12 0.04 36.1 2.2 27.8 0.6 0.123 0.011 0.039 0.005 0.0179 0.0028 25.4 0.5 45.2 0.5

Fig. 8. Pc and kc of RAC: (a) Pc; (b) kc.

3.3.2. Elastic modulus than RFA is also an important factor. On the other hand, the E is posi­
The elastic modulus is the stress required for the unit strain of the tively correlated with the flexural capacity of concrete, the variation of E
material, which can explain the variation of the nominal stiffness of can well explain that the nominal stiffness of RFAC is higher than that of
RAC. According to Xu et al.[59], the elastic modulus (E) of concrete can MSRCAC.
be calculated by a three-point bending test. The E is as follows:
3.3.3. δc, CTODc and CMODc
1 [ (π a ) ]
(3) The RFA and RCA increase the nonlinear behavior of concrete, which
0
E= 3.70 + 32.60tan2
tCi 2 D
affects the peak deformation of RAC, such as CMOD and CTOD at peak
where Ci is the flexibility of the elastic phase of the P-CMOD curve. load. However, as the characteristic parameters of the two-parameter
Fig. 9 shows the E of concrete with different combinations of recy­ fracture model, they are important for evaluating the fracture proper­
cled aggregates. It can be seen that the RFA and RCA reduced the E of ties of concrete[61]. Therefore, the deflection (δc), crack mouth opening
concrete. Compared with MSNAC, the E of RFAC, MSRCAC and FRAC displacement (CMODc), and crack tip opening displacement (CTODc) at
decreased by about 14.2%, 20.8% and 28.9%, respectively. The E of peak load of RAC were compared and analyzed.
FRAC decreased significantly under the interaction of RFA and RCA. It Fig. 10 shows the δc, CMODc and CTODc of RAC. It can be seen that
should be noted that the E of MSRCAC was lower than that of RFAC. δc, CMODc and CTODc have similar trends. The detailed peak de­
Wang et al.[60] indicated that the hydration of unhydrated cement on formations are shown in Table 3. Compared to MSNAC, the δc increased
the surface of FRA improved the adverse effects of old mortar. by about 13.1%, 8.5% and 19.3% for RFAC, MSRCAC, and FRAC,
Furthermore, the RCA is the skeleton of concrete, its content is greater respectively, the CMODc increased by about 53.6%, 41.6% and 72.0%,
respectively, and the CTODc increased by about 66.5%, 62.0% and
96.1%, respectively. Luo et al.[62] concluded that the large deformation
of RAC was caused by active microcracks, and the multiphase ITZs and
fragile aggregates resulted in a large deformation of RAC[6]. From a
macro perspective, the recycled aggregates reduced the E of concrete,
and the low E led to a large critical deformation. It should be noted that
the RFA caused larger δc, CMODc and CTODc of concrete than RCA. Xiao
et al.[15] considered that the deformation of concrete was mainly
influenced by microcracks before the peak load. However, the under­
lying causes require further investigation.

3.3.4. Critical effective crack length

The critical effective crack length (ac) is the effective crack length at
peak load, which is an important parameter for calculating fracture
toughness. According to Xu et al.[59], the ac is as follows:
( )1/2
2D tE(CMODd )
ac = arctan − 0.1135 (4)
π 32.6Pc
Fig. 9. Elastic modulus of RAC.

7
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 10. Critical displacement of RAC: (a) δc; (b) CMODc; (c) CTODc.

where E is the elastic modulus, as shown in Eq. (3) and CMODd is the [59], the Kini
IC is as follows:
CMOD considering the effect of additional central load:
3(Pini + Pd )S √̅̅̅̅̅ (a0 )
CMODd = CMODc + Pd Ci (5) K ini
IC = a0 F (7)
2D2 t D
where Pd is the additional load. ( )[ ( )2 ]
a0 a0 a0 a0
(a ) 1.99- D 1 − D 2.15-3.93 D + 2.7 D
4(M 1 − M 2 )
(6) (8)
0
Pd = F = ( )( )3/2
S D
1 + 2 aD0 1 − aD0
where M1 is the center bending moment caused by the weight of
beam and other factors, and M2 is the center bending moment where Pd is the additional load, as shown in Eq. (6).
compensated by the support moment. The calculation formula of Kun IC is as follows:
Fig. 11 shows the ac of RAC. It can be seen that there is no significant
difference in ac between concrete with different combinations of recy­ 3(Pc + Pd )S √̅̅̅̅̅ (ac )
K un
IC = ac F (9)
cled aggregates. The ac of RAC was about 44.3–45.6 mm, as shown in 2D2 t D
Table 3. Compared with MSNAC, the ac of RFAC, MSRCAC and FRAC ( )[ ( )2 ]
changed by − 2.9%, 2.6% and − 0.9%, respectively. Tang et al.[32] ac ac ac
(a ) 1.99- D 1 − D 2.15-3.93 D + 2.7 D
ac

indicated that the ac of RFAC with different RFA replacement rates was F
c
= ( )( )3/2 (10)
D
similar. Furthermore, Chen et al.[63] also found that there was no sig­ 1 + 2 aDc 1 − aDc
nificant difference in the ac of RAC with different RCA replacement
rates. Therefore, the ac may not be sensitive to the change in recycled where ac is the critical equivalent crack length, as shown in Eq. (4).
aggregates. Fig. 12 (a) and Table 4 show the Kini IC of RAC. It can be seen that the
use of RFA and RCA reduced the Kini ini
IC of concrete, and the KIC of concrete
3.4. Fracture toughness with different combinations of recycled aggregates were about
0.28–0.51 MPa•m1/2. Compared with MSNAC, the Kini IC of RFAC,
The stress intensity factor corresponding to the unstable crack MSRCAC and FRAC decreased by about 37.7%, 13.7% and 45.8%,
propagation is called fracture toughness, which is an important
respectively. It should be noted that the Kini IC of RFAC and FRAC is
parameter to evaluate the crack resistance of concrete. To distinguish
significantly reduced, which is consistent with the results of Pini.
the fracture process of stable crack propagation and unstable crack
Compared to RFA, the RCA as a skeleton has less effect on the Kini IC of
propagation, the double-K fracture model proposed by Xu and Reinhardt
[64] was used to evaluate the fracture toughness of RAC. The initial concrete. The strength of cement matrix is the key factor in controlling
cracking of concrete[15,32], the RFA weakens the cement matrix.
fracture toughness (Kini un
IC ) and unstable fracture toughness (KIC ) are the
The Kun
IC of RAC is shown in Fig. 12 (b) and Table 4. The KIC of
un
main parameters of the double-K fracture model. According to Xu et al.
concrete with different combinations of recycled aggregates were about
0.9–1.09 MPa•m1/2. Compared with MSNAC, the Kun IC of RFAC, MSRCAC
and FRAC decreased by 6.2%, 11.5% and 17.7%, respectively. The Kun IC
reflects the ability of concrete to resist the unstable propagation of
cracks, the interaction of RFA and RCA significantly reduces the crack
resistance of FRAC. As the coarse aggregates form the skeleton of con­
crete and the fine aggregates only fill the voids in the skeleton, thus the
effect of RCA on the unstable fracture toughness of concrete is greater
than that of RFA. In addition, the relationship between Kini un
IC and KIC was
compared and analyzed, as shown in Fig. 12 (b). The Kini un
IC /KIC of RAC
was about 0.310–0.467. Compared with MSNAC, there was no signifi­
cant difference in Kini un ini un
IC /KIC of MSRCAC, the KIC /KIC of RFAC and FRAC
decreased by about 34%. This indicates that the KIC /Kun
ini
IC is not sensitive
to RCA, while as the use of RFA increases the risk of cracking in concrete,
its effect on the Kini un
IC /KIC is significant.

Fig. 11. Critical effective crack length of RAC.

8
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 12. Kini un ini un

IC and KIC of RAC: (a) KIC ; (b) K IC .

due to the increase in porosity. Due to the old mortar on the surface of
Table 4
the aggregates, the RCA reduces the aggregate strength of the concrete
Fracture parameters of RAC.
and introduces multiple ITZs[55], and RFA reduces the cement matrix
1/2
Groups Kini
IC (MPa•m
1/2
) Kun
IC (MPa•m ) Kini un
IC /KIC
GF (N/m) strength[33]. These factors lead to a decrease in the tensile strength, the
Avg SD Avg SD - Avg SD number of secondary cracks and the damage width of RAC[27,32], thus
the use of RFA and RCA reduces the dissipation capacity of RAC.
MSNAC 0.511 0.029 1.094 0.032 0.467 104.6 6.9
RFAC 0.319 0.010 1.026 0.091 0.310 80.7 4.3 It should be noted that the RCA has a greater effect on the GF of
MSRCAC 0.441 0.031 0.968 0.064 0.455 77.7 4.2 concrete than the RFA, which may be related to the failure mode of RAC.
FRAC 0.277 0.019 0.900 0.003 0.308 75.8 2.2 The failure mode of RFAC had two forms, including intergranular failure
and transgranular failure, while the MSRCAC was mainly transgranular
failure, as shown in Section 3.1. The intergranular failure may dissipate
3.5. Fracture energy
more energy than the transgranular failure[15], therefore the GF of
RFAC is larger than that of MSRCAC. In addition, compared with Kun IC ,
Fracture energy (GF) is the energy required for the material to form a
the GF attenuation of RFAC, MSRCAC and FRAC was more obvious. This
fracture surface per unit area[65], which is important for evaluating the
is due to the fact that only one moment is considered for the calculation
energy dissipation of concrete, the size of the FPZ, and the shape of the
of Kun
IC , while the calculation of GF includes the whole load history.
tensile softening curve. According to the suggestion of Petersson[66],
the GF is as follows:
A0 3.6. AE results
GF = (11)
(D − a0 )t
Acoustic emission (AE) is an elastic wave phenomenon generated
where A0 is the envelope area of the P-δ curve. when the material is broken, and the characteristics of elastic waves can
The GF of RAC is shown in Fig. 13 and Table 4. It can be seen that the be characterized by AE parameters. The variation of AE parameters is
GF of RFAC, MSRCAC and FRAC were significantly reduced, and the GF closely related to the fracture pattern of materials[57] and can be used
of concrete with different combinations of recycled aggregates was to evaluate the damage of materials[68]. The AE waveform and AE
about 75.8–104.6 N/m. Compared to MSNAC, the GF of RFAC, MSRCAC parameters are shown in Fig. 14. The ring count is the number of events
and FRAC were reduced by about 22.9%, 25.7% and 27.5%, respec­ that the waveform exceeds the threshold. The amplitude is the
tively. Ghorbel et al.[67] concluded that the decrease in GF of RAC was maximum amplitude of the waveform. The rise time is the time interval

Fig. 13. Fracture energy of RAC. Fig. 14. Diagram of AE waveform and AE parameters.

9
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

between the first time the waveform exceeds the threshold and the phenomenon has been found in four-point bending tests of reinforced
maximum magnitude. The duration time is the time interval between recycled aggregate concrete beams[71]. In addition, the accumulative
the first time the waveform exceeds the threshold and the last time it hit count is smaller and the accumulative ring count is larger for
falls below the threshold. A complete waveform is called a hit. MSRCAC compared to RFAC, which may be due to the fact that the RCA
The sources of AE signal of concrete fracture mainly include the fracture has a longer wavelength of the elastic wave than the recycled
initiation and propagation of microcracks, the fracture of ITZs, mortar mortar fracture.
and aggregates, and friction between fracture surfaces[54]. It should be The hit count and ring count can only reflect the AE activity. To
noted that the use of RFA and RCA makes the microstructure of concrete distinguish the fracture strength of concrete with different combinations
porous and fragile[33,55], meanwhile the microcracks propagation and of recycled aggregates, the distribution of AE amplitude was analyzed.
damage mechanism of concrete under the interaction of RFA and RCA is Besides, to avoid high-frequency features being covered by low-
more complicated than that of single recycled aggregates[9]. The frequency signals, events greater than 45 dB were counted.
microscopic damage inside the concrete under load can be detected by The amplitude distribution of RAC is shown in Fig. 17. It can be seen
the AE technique[69], which is positive for understanding the interac­ that the low-medium frequency signals (50–55 dB) of concrete with
tion of RFA and RCA. To investigate the damage evolution and fracture different combinations of recycled aggregates are similar. However,
mechanism of RFAC, MSRCAC and FRAC, their AE characteristics (e.g., compared with MSNAC, the low-frequency signals (45–50 dB) of RFAC,
hit, ring count, amplitude, RA-AF and peak frequency) were compara­ MSRCAC and FRAC were decreased by about 10.0%, 12.8% and 12.4%,
tively analyzed. respectively. Meanwhile, the medium-frequency (55–60 dB), medium-
high frequency (60–65 dB), and high-frequency (>65 dB) signals of
3.6.1. Hit, ring count and amplitude RFAC (about 9.9%, 18.8% and 21.4%), MSRCAC (about 38.1%, 47.0%
Fig. 15 shows the hit count and ring count of concrete with different and 25.9%), and FRAC (about 36.2%, 35.2% and 32.4%) were signifi­
combinations of recycled aggregates. It can be seen that the AE signal is cantly increased. In general, it is the fracture of the mortars and ag­
inactive before the peak load, this indicates that the inside of RAC is gregates that causes the medium-high and high frequencies of the AE
mainly microcrack propagation and microcrack bridging before unsta­ signals[54], indicating that the RFAC, MSRCAC and FRAC have more
ble crack propagation. However, the hit and ring counts of RFAC, mortar and aggregate fractures than the MSNAC. In addition, compared
MSRCAC and FRAC increased significantly after the peak load. The RFA with RFAC, the 45–50 dB, 55–60 dB, 60–65 dB and >65 dB signal var­
and RCA reduced the strength of cement matrix and aggregates in iations of MSRCAC were − 3.1%, 8.1%, 6.4% and 0.8%, respectively. The
concrete[33,70], and a large number of recycled aggregates and recy­ RCA caused a higher amplitude distribution than RFA, this indicates that
cled mortars fractured under load, leading to active AE signals[54]. the fracture process of the MSRCAC is more intense than that of the
To further evaluate the AE activity of concrete with different com­ RFAC.
binations of recycled aggregates, the accumulative hit counts and
accumulative ring counts were quantitatively analyzed. As the speci­ 3.6.2. RA-AF
mens stopped loading at different times, the AE signals before 0.5 kN Although the three-point bending test is a mode I fracture test, the
(post-peak) were compared, as shown in Fig. 16. It can be seen that the random distribution of aggregates may cause cracks to be deflected and
accumulative hit count and accumulative ring count of RAC increased fracture surfaces to be tilted. The ratio of tensile to shear cracks varies
significantly, especially the FRAC. Compared to MSNAC, the accumu­ with the deflection angle of the crack, and the crack mechanism of
lative hit count of RFAC, MSRCAC and FRAC increased by approxi­ concrete is the coexistence of shear and tensile cracks[72]. The shear
mately 102.6%, 87.0% and 134.8%, respectively, and the accumulative cracks release shear waves and the tensile cracks release expansion
ring count increased by about 67.9%, 81.6% and 103.2%, respectively. waves[73], the RA and AF parameters can better distinguish between
The RFA and RCA reduce the elastic modulus of concrete, while concrete these two waveforms. In general, the higher AF and lower RA indicate a
with lower elastic modulus has higher AE activity[58], and a similar higher percentage of tensile cracks, while the opposite is true for shear

Fig. 15. Hit count and ring count of RAC: (a) hit count; (b) ring count.

10
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 16. Accumulative hit count and accumulative ring count of RAC: (a) accumulative hit count; (b) accumulative ring count.

where CNTS is the ring count and Dt is the duration time, as shown in
Fig. 14.
Fig. 19 shows the distribution of RA-AF of concrete with different
combinations of recycled aggregates. It can be seen that RFA and RCA
increased the number of high AF and low RA, and the crack path of RAC
was straighter than that of MSNAC, especially FRAC. The deflection
angle of the crack is negatively correlated with the proportion of tensile
cracks[40]. This indicates that the RFAC, MSRCAC, and FRAC have
more tensile cracks than MSNAC. However, the distribution of AF-RA
can only provide a qualitative description of the ratio of tensile to
shear cracks in concrete. To further evaluate the crack mechanism of
RAC, the ratio of tensile to shear cracks was quantitatively analyzed, as
shown in Fig. 20.
It should be noted that the ratio of AF to RA is controversial and it is
difficult to be reliably determined, to quantify the ratio of tensile to
shear cracks in concrete with different combinations of recycled ag­
Fig. 17. AE amplitude of RAC. gregates, 0.1 has been used as the ratio of AF and RA[74]. It can be seen
from Fig. 20 that the ratio of tensile to shear cracks in RAC is approxi­
cracks, as shown in Fig. 18. To evaluate the crack mechanism of RFAC, mately 2.4–3.3. Compared with MSNAC, the ratio of tensile to shear
MSRCAC and FRAC, the AF and RA were studied. The RA and AF are cracks in RFAC, MSRCAC and FRAC increased by about 10.5%, 34.7%
shown in Eqs. (12) and (13). and 40.2%, respectively, indicating that RFA and RCA increased the
R number of tensile cracks in concrete, especially in MSRCAC and FRAC.
RA = (12) Due to the old mortar on the surface of the recycled aggregates, the
A
strength of the recycled aggregates is low, resulting in cracks passing
where R is the rise time and A is the amplitude, as shown in Fig. 14. directly through the recycled aggregates. Conversely, cracks will bypass
the natural aggregates due to their high strength. The large number of
CNTS
AF = (13) fractures of the recycled aggregates reduces the deflection angle of the
Dt
cracks, leading to an increase in the proportion of tensile cracks[40]. In
addition, the coarse aggregates as the concrete skeleton play a key role
in resisting crack propagation, while the fine aggregates filling the
skeleton gap have little effect on the crack[21]. Therefore, compared
with MSNAC, the ratio of tensile to shear cracks of MSRCAC and FRAC
increased significantly, while there was no significant difference in the
ratio of tensile to shear cracks of RFAC. In general, the increase in the
proportion of tensile cracks indicates that the crack resistance and en­
ergy dissipation capacity of RAC are reduced, which is consistent with
the results of the Kun
IC and GF of RAC.

3.6.3. Peak frequency

Peak frequency is an important AE parameter to characterize the
damage evolution and crack propagation of concrete, and the evaluation
of the peak frequency of concrete with different combinations of recy­
cled aggregates is helpful in understanding the fracture mechanism of
RAC. The AE waveform was converted to the frequency domain by Fast
Fourier Transform (FFT), and the maximum frequency in the frequency
Fig. 18. Schematic diagram of AF-RA.

11
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 19. AF-RA of RAC: (a) MSNAC; (b) RFAC; (c) MSRCAC; (d) FRAC.

frequency), 100–150 kHz (medium-high frequency), and > 150 kHz

(high frequency) were quantified. Due to the different stop times of the
specimens, the peak frequencies before 0.5 kN (post-peak) were
analyzed, as shown in Fig. 22. It can be seen that there was no significant
difference in the peak frequency of 50–100 kHz of concrete with
different combinations of recycled aggregates. However, compared with
MSNAC, the peak frequencies of 0–50 kHz of RFAC, MSRCAC and FRAC
decreased by about 18.5%, 36.7% and 58.2%, respectively, and the peak
frequencies of 100–150 kHz and >150 kHz increased by about 28.6%,
45.7%, 82.4% and 160.9%, 67.3%, 130.7%, respectively. The results
indicate that the fracture mechanism of RAC was changed by RCA and
RFA. In addition, compared to RFAC, the peak frequencies of 0–50 kHz
and 100–150 kHz of MSRCAC changed by − 22.4% and 13.3%, respec­
tively, indicating that the fracture process of MSRCAC is more intense
than that of RFAC, which is consistent with the conclusion of AE
amplitude.
Fig. 20. Proportion of tensile and shear cracks in RAC.

3.7. Characteristics of FPZ

domain was called the peak frequency[75].
The peak frequency of RAC is shown in Fig. 21. It can be seen that the
The region of microcracks at the tip of the traction-free crack is
peak frequencies of concrete with different combinations of recycled
known as the fracture process zone (FPZ)[29]. Due to the FPZ, the
aggregates were mainly distributed at 25–200 kHz. The number of peak
frequencies of RFAC, MSRCAC and FRAC was more than that of MSNAC
before peak load, while the peak frequency was positively correlated
with the damage of concrete[76]. This indicates that RFAC, MSRCAC
and FRAC have higher damage than MSNAC before unstable crack
propagation.
It should be noted that the difference in peak frequency between
concrete with different combinations of recycled aggregates is mainly
after the peak load. The RFA and RCA significantly increased the num­
ber of peak frequencies of concrete, especially MSRCAC and FRAC. Due
to the defects of recycled aggregates, the microcracks in RAC are more
active under load than in MSNAC[33], while the small-scale cracks
formed by the localization of microcracks cause a large number of AE
signals[76]. In addition, as previously mentioned, the RFAC, MSRCAC
and FRAC have more mortar and aggregate fractures than MSNAC, and
their fractures lead to high peak frequencies[77].
To quantitatively evaluate the peak frequency of RAC, the peak
frequencies of 0–50 kHz (low frequency), 50–100 kHz (medium Fig. 22. Distribution of peak frequency of RAC.

Fig. 21. Peak frequency of RAC: (a) MSNAC; (b) RFAC; (c) MSRCAC; (d) FRAC.

12
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

fracture behavior of concrete is nonlinear. Furthermore, as a toughening strain matrix, and then the boundary points of the filtered horizontal
mechanism, the FPZ is critical to the crack resistance of concrete[39], strain matrix were extracted. Finally, the boundary points were taken as
and the FPZ may be an intrinsic factor to the size effect of the fracture the target points of the coordinate matrix, and the horizontal distance
parameters of concrete[29,30]. To investigate the effect of the interac­ between the target points was considered as the FPZ width. The average
tion of RFA and RCA on the FPZ of concrete, a systematic comparative of the maximum horizontal strain at the FPZ tip at multiple moments
analysis of the FPZ evolution, FPZ length, and FPZ width of concrete was used as the filter value of the horizontal strain matrix. The param­
with different combinations of recycled aggregates was conducted using eters used to determine the length and width of FPZ are shown in
AE and DIC techniques. Table 5.

3.7.1. Determination of the length and width of the FPZ 3.7.2. Evolution of FPZ
To understand the length of FPZ, the tip and tail of the FPZ need to be To evaluate the evolution of the FPZ of concrete with different
determined. Ideally, the position where the crack opening displacement combinations of recycled aggregates, a comparative analysis of the
(COD) is equal to 0 can be considered as the tip of FPZ[29]. The contours of horizontal strain for four moments, 0.8Pc,FRAC, Pc,FRAC, 0.5Pc,
threshold crack identification method (TCIM) proposed by Xu et al.[27] FRAC (post-peak) and 0.3Pc,FRAC (post-peak), was performed, as shown in
was used to determine the tip of FPZ. In this method, the peak tensile Fig. 24, the Pc,FRAC is the peak load of FRAC. It can be seen that the
strain was used as the filter value of the horizontal strain matrix, and damage zone of concrete increases with load, the strain localization
then the boundary points of the filtered horizontal strain matrix were zones of RFAC, MSRCAC and FRAC were significantly larger than those
extracted. The boundary points were used as the target points of the of MSNAC before the Pc,FRAC, especially FRAC. The RFA and RCA reduce
horizontal displacement matrix, and the COD was obtained by the dif­ the strength of the cement matrix and aggregate[33,79], leading to
ference of horizontal displacement of the target points. The peak tensile faster damage propagation in RAC. It should be noted that the RFAC had
strain (εt) is calculated as follows: more bridging cracks than other concretes at Pc,FRAC. Due to RFA, the
microcracks in the cement matrix are active[33], while NCA as a skel­
ft
εt = (14) eton restricts the propagation of microcracks, resulting in discontinuous
E
cracks in RFAC. However, the strain localization zones of concrete with
where E is the elastic modulus, as shown in Eq. (3), ft is the uniaxial different combinations of recycled aggregates at 0.5Pc,FRAC (post-peak)
tensile strength, which is calculated as follows according to CEB-90[78]: and 0.3Pc,FRAC (post-peak) were similar. Concrete, as a quasi-brittle
material, its macroscopic cracks will rapidly propagate after unstable
ft = 0.9fts (15) crack propagation[27], the brittleness may mask the effect of RFA and
RCA on its damage area.
where fts is the splitting tensile strength, as shown in Table 2. The
To further understand the effect of RAC and RFA on the damage
average value of the εt was used as the filter strain[27], as shown in
evolution of concrete, the amplitude distributions of RAC at peak load
Table 5.
and stop load were comparatively analyzed, as shown in Fig. 25. The
It should be noted that the COD of the tip of FPZ determined using
amplitude distributions were obtained by entering the measured con­
the TCIM method may not be equal to 0 due to the slight elastic defor­
crete sound velocity and AE sensor coordinates into the AE post-
mation. According to the suggestion of Xu et al.[27], COD≤0.5 μm was
processing software (Vallen Visual AE). In addition, since voids and
considered as the tip of FPZ, as the calculation accuracy of the DIC
cracks in the concrete may cause echoes, this will increase the number of
technique is about 0.5 μm[28]. In addition, when the COD at the tail of
low-frequency amplitudes and cause positioning errors, only the am­
FPZ was less than the stress-free crack opening displacement (w0), the
plitudes above 50 dB were considered. According to the signal strength,
notch tip was regarded as the tail of FPZ. However, the FPZ was shifted
the amplitudes were divided into medium frequency (50–60 dB),
forward as a whole when the COD was larger than w0, and the tail of FPZ
medium-high frequency (60–70 dB) and high frequency (>70 dB).
needed to be redefined, as shown in Fig. 23. In general, the position
As shown in Fig. 25 (a), there was no significant AE phenomenon
where COD was equal to w0 was considered to be the tail of FPZ[29].
(>50 dB) in concrete with different combinations of recycled aggregates
According to the suggestion of Petersson[66], w0 is calculated as follows:
at peak load, indicating that the AE phenomenon at this stage is the
3.6GF bridging and localization of microcracks, and the damage propagation of
w0 = (16)
ft RAC is stable. However, the RFAC, MSRCAC and FRAC had more active
AE signals than MSNAC, especially FRAC. The use of RFA and RCA in­
where GF is the fracture energy, as shown in Eq. (11). creases the micro-voids in concrete, resulting in microcracks becoming
The reliable FPZ width can be determined by the AE technique[54], active under load[33]. In addition, the effect of RFA on the activity of
and the determined FPZ width is only a maximum FPZ width, while the microcracks in concrete may be greater than that of RCA.
FPZ determined by DIC technique is continuous. Therefore, the AE and Fig. 25 (b) shows the two-dimensional localization of the amplitude
DIC techniques were used to determine the FPZ width. In the AE tech­ at stop load. It can be seen that a large amount of damage was produced
nique, the area where the AE event was greater than or equal to 20% of in the concrete during the unstable propagation of the crack to stop load,
the maximum number of events was considered as the FPZ width[69]. In which may be caused by the rapid coalescence and propagation of small-
the DIC technique, the FPZ width was determined according to the scale cracks[76]. Compared with MSNAC, the amplitude of
method proposed by Xu et al.[27]. In this method, the maximum hori­ medium-high and high frequencies of RFAC, MSRCAC, and FRAC were
zontal strain of the FPZ tip was taken as the filter value of the horizontal significantly increased, indicating that the use of RFA and RCA

Table 5
FPZ identification parameters.
Groups εt Average of εt Strain of crack tip Average of strain of crack tip w0

Avg SD

MSNAC 0.000085 0.000096 0.000148 0.000025 0.000150 0.1243

RFAC 0.000096 0.000151 0.000027 0.0989
MSRCAC 0.000098 0.000152 0.000029 0.1009
FRAC 0.000107 0.000149 0.000026 0.1013

13
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 23. Schematic diagram of FPZ: (a) CTOD≤w0; (b) CTOD>w0.

significantly increases the damage of concrete. In addition, the MSRCAC microscopic fracture response of the FPZ through the recycled mortar
and FRAC were more severely damaged than the RFAC because the and RCA. The macroscopic effects of RFA and RCA on the FPZ may be
aggregate fracture releases more energy than mortar fracture[76]. concealed due to the high brittleness of the concrete during the insta­
Although the fracture surfaces of MSNAC and RFAC had similar failure bility phase.
modes, RFAC had more damage than MSNAC, this indicates that the However, the fully developed lFPZ of concrete decreased with the
number of mortar fractures in RFAC is higher than in MSNAC due to the addition of RFA and RCA. As shown in Fig. 26 (b) and (d), the fully
fragile cement matrix. developed lFPZ of RAC were about 59.4–67.2 mm, accounting for
In summary, the RFA and RCA have a large influence on the damage 0.85–0.96 of ligament height. Compared with MSNAC, the fully devel­
development in concrete. However, the strain contours and two- oped lFPZ of RFAC, MSRCAC and FRAC decreased by about 7.8%, 7.7%
dimensional localization can only provide a qualitative assessment of and 11.6%, respectively. Recycled aggregates increased the opening
the damage of the RAC. To investigate the effect of RCA and RFA on the deformation of concrete[32,63], and the decrease in tensile strength and
damage mechanism of concrete, it is necessary to quantify the FPZ. dissipation capacity led to the decrease in the stress-free crack opening
of RAC. Therefore, the fully developed lFPZ of RFAC, MSRCAC and FRAC
3.7.3. Length of FPZ was reduced. It should be noted that the effects of RFA and RCA on the
The distribution of cohesive stress can be reflected by the length of fully developed FPZ of concrete may be similar.
FPZ (lFPZ), which is critical in evaluating the crack propagation resis­ In phase III, the lFPZ of RAC decreased after the fully developed FPZ
tance of concrete[29]. Fig. 26 shows the relationship between the lFPZ due to the boundary effect[39,80]. As shown in Fig. 26 and Fig. 27, the
and the load and crack propagation length (Δa) of concrete with use of RFA and RCA increased the reduction rate of FPZ, especially
different combinations of recycled aggregates. It can be seen that the FRAC, while the FPZ reduction rates of RFAC and MSRCAC were similar,
development trend of FPZ of four types of concrete was similar. Due to indicating that there is no significant difference between the effects of
the aggregate interlocking of concrete[54], the development of FPZ was RFA and RCA on the fully developed FPZ. In addition, with the reduction
nonlinear. According to the FPZ development characteristics of RAC, it of FPZ, the development of microcracks was limited and the AE accu­
was divided into the stable development stage (phase I), the unstable mulative ring counts gradually stabilized.
development stage (phase II), and the retraction stage (phase III), which
was consistent with the research results of Xu et al.[27]. 3.7.4. Width of FPZ
In phase I, the lFPZ increased with the addition of RFA and RCA. As Fig. 28 shows the WFPZ of the RAC determined using the AE and DIC
shown in Fig. 26 (b) and (c), the lFPZ of RAC at peak load were about techniques. It can be seen that the use of RFA and RCA reduced the WFPZ
20.2–34.1 mm, accounting for about 0.29–0.49 of the ligament heights. of concrete. As shown in Fig. 28 (a) and Fig. 29, the maximum WFPZ
Compared with MSNAC, the lFPZ of RFAC, MSRCAC and FRAC increased (WFPZ,max) of concrete with different combinations of recycled aggre­
by about 48.7%, 35.9% and 68.7%, respectively. In addition, the AE gates was about 17.7–31.3 mm (about 1.8–3.1 times the maximum
activity of the four concrete types was low before the peak load, as aggregate size). Compared with MSNAC, the WFPZ,max of RFAC, MSRCAC
shown in Fig. 27. The development of the FPZ did not induce a signifi­ and FRAC decreased by about 33.0%, 38.2% and 43.6%, respectively.
cant AE signal, indicating that the FPZ is mainly extended microcracks The decrease in strength of the cement matrix and coarse aggregate
and bridged microcracks at this stage. However, the activity of micro­ leads to the weakening of the aggregate interlocking effect[27,32],
cracks in concrete is increased by RFA, the lFPZ of MSRCAC at peak load which is detrimental to WFPZ. The reduction of the interlocking effect
was about 8.6% lower than that of RFAC. Compared to RCA, RFA has a may cause a series of problems, including the shortening of secondary
greater effect on the FPZ of concrete before FPZ unstable development. cracks, the reduction of dissipation capacity, and the narrowing of
In phase II, the lFPZ of concrete developed rapidly, as shown in Fig. 26 damage width. It should be noted that the WFPZ,max of RFAC was about
(a) and Fig. 27. The lFPZ of concrete with different combinations of 8.6% higher than that of MSRCAC. Guo et al.[24] concluded that
recycled aggregates were similar. However, there were significant dif­ transgranular failure limited the propagation of microcracks to other
ferences in the AE characteristics at this stage, which is due to the large parts, which led to a decrease in WFPZ. However, the fracture surface of

14
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 24. Horizontal strain contours of RAC at different times: (a) MSNAC; (b) RFAC; (c) MSRCAC; (d) FRAC.

MSRCAC had more transgranular failure than that of RFAC. This in­ shown in Fig. 28 (b) and Fig. 29. It can be seen that the trend of WFPZ of
dicates that RCA has a greater influence on the aggregate interlocking concrete with different combinations of recycled aggregates were
effect of concrete than RFA. In addition, FRAC showed the lowest WFPZ, similar. Due to the extended and bridged microcracks, the WFPZ gradu­
the aggregate interlocking effect of FRAC was significantly reduced ally increased before the peak load, and the WFPZ,max of the four types of
under the interaction of RFA and RCA. concrete occurred after the unsteadily developed FPZ. However, the
The DIC technique can only monitor the deformation of the surface, WFPZ of concrete with different combinations of recycled aggregates
while the AE technique can monitor the internal damage, thus the WFPZ, were significantly different, the use of RFA and RCA reduced the WFPZ of
max determined by the AE technique is more accurate than that of the concrete. It should be noted that the continuous opening and closing of
DIC technique. To ensure the reliability of the WFPZ determined by the the secondary cracks caused the variation of the WFPZ[27], the activity
DIC technique, the WFPZ,max determined by the AE technique and the of secondary cracks can be reflected by WFPZ. Compared with MSNAC,
DIC technique were compared, as shown in Fig. 29. The WFPZ,max of RAC the activity of secondary cracks in RFAC, MSRCAC and FRAC decreased.
determined by the DIC technique was about 19.5–30.4 mm, and the Although microcracks are more active in RAC than in normal concrete
maximum difference between the WFPZ,max determined by the DIC and [33], the fragile recycled mortar and RCA have fractured before the
AE techniques were not more than 10.6%. Therefore, the WFPZ deter­ microcracks extend outward.
mined by the DIC technique is reliable. With the propagation of macroscopic cracks, a large amount of
To obtain continuous WFPZ, the WFPZ was further investigated using fracture of the coarse aggregates and mortars led to the decrease of the
the DIC technique. The averaged FPZ width (WFPZ,avg) and WFPZ,max are WFPZ. After the FPZ fully developed, the difference between WFPZ,avg and

15
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 25. Two-dimensional localization of AE amplitude at different moments: (a) peak load; (b) stop load.

Fig. 26. lFPZ of RAC: (a) P/Pc-lFPZ/(D-a0); (b) lFPZ-Δa/(D-a0); (c) the lFPZ at peak load; (d) the lFPZ at fully developed FPZ.

WFPZ,max gradually decreased and the secondary cracks gradually closed. 4. Conclusions
The lower external loads cannot drive new secondary cracks in concrete
due to the boundary conditions, the damaged width of concrete with Fracture characteristics of the MSNAC, RFAC, MSRCAC and FRAC
different combinations of recycled aggregates gradually stabilized. were investigated using AE and DIC techniques and the following

16
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Fig. 27. lFPZ and accumulative ring count of RAC: (a) MSNAC; (b) RFAC; (c) MSRCAC; (d) FRAC.

Fig. 28. WFPZ of RAC:(a) AE technique; (b) DIC technique.

conclusions were drawn: (Pini, Pc and E) and nominal stiffness (kini and kc) of concrete but
(1) The failure modes of the RFAC included intergranular and increased the critical displacement (δc, CMODc and CTODc). The RFA
transgranular fractures, while the MSRCAC and FRAC were mainly had a greater effect on the Pini and critical displacement than that of the
transgranular fractures. The RFA and RCA reduced the bearing capacity RCA, while the kini and E were more sensitive to the RCA.

17
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Data availability

Data will be made available on request.

Acknowledgements

The authors gratefully acknowledge the financial support of the

National Natural Science Foundation of China (No. 52278251,
51768011) and Guizhou Provincial Sciences and Technology Projects
(ZK[2022] Key 007).

References

[1] S.K. Kirthika, S.K. Singh, A. Chourasia, Alternative fine aggregates in production of
sustainable concrete- a review, J. Clean. Prod. 268 (2020) 27, https://doi.org/
10.1016/j.jclepro.2020.122089.
Fig. 29. WFPZ,max of the RAC determined using AE and DIC techniques. [2] H. Zhang, H. Zhao, F.Q. Liu, H. Yang, Y. Geng, K.A. Pan, Models for uniaxial stress-
strain relationship and thermal properties of fine recycled aggregate concrete
exposed to elevated temperatures, J. Build. Eng. 66 (2023) 22, https://doi.org/
(2) The fracture resistance and fracture energy of concrete were 10.1016/j.jobe.2023.105869.
reduced by the RFA and RCA. The Kini un
IC , KIC and GF of concrete with [3] A. Srivastava, S.K. Singh, Utilization of alternative sand for preparation of
different combinations of recycled aggregates were about sustainable mortar: a review, J. Clean. Prod. 253 (2020) 16, https://doi.org/
10.1016/j.jclepro.2019.119706.
0.28–0.51 MPa•m1/2, 0.9–1.09 MPa•m1/2 and 75.8–104.6 N/m, [4] D.Y. Wang, C.X. Lu, Z.M. Zhu, Z.Y. Zhang, S.Y. Liu, Y.C. Ji, Z.Q. Xing, Mechanical
respectively. Compared with MSNAC, the Kini un
IC , KIC and GF of RFAC,
performance of recycled aggregate concrete in green civil engineering: review,
Case Stud. Constr. Mater. 19 (2023) 17, https://doi.org/10.1016/j.cscm.2023.
MSRCAC and FRAC decreased by about 6.2%− 45.8%. The Kini IC and e02384.
Kini un un
IC /KIC were sensitive to the RFA, while the KIC and GF were sensitive
[5] A.H. Hossein, H.A. Jafari, R. Khoshnazar, The role of performance metrics in
comparative LCA of concrete mixtures incorporating solid wastes: a critical review
to the RCA.
and guideline proposal, Waste Manag. 140 (2022) 40–54, https://doi.org/
(3) The damage of RAC was dominated by the propagation and 10.1016/j.wasman.2022.01.010.
localization of microcracks before the peak load, and the difference in [6] F.D. Salgado, F.D. Silva, Recycled aggregates from construction and demolition
waste towards an application on structural concrete: a review, J. Build. Eng. 52
AE behaviors was mainly concentrated after unstable crack propagation.
(2022) 20, https://doi.org/10.1016/j.jobe.2022.104452.
The RFA and RCA significantly increased the AE activity, high frequency [7] X.L. Yang, Y.S. Liu, J.L. Liang, Y.J. Meng, H.L. Rong, D.J. Li, Y.F. Chen, J.L. Lv, Y.
amplitude and high peak frequency in concrete, while the RCA caused a X. Jiang, Y. Liu, Straightening methods for RCA and RAC-a review, Cem. Concr.
more intense fracture response in concrete than the RFA. Using the RFA Compos. 141 (2023) 33, https://doi.org/10.1016/j.cemconcomp.2023.105145.
[8] M.S. Ganesh, P. Jagadeesh, Assessment of usage of manufactured sand and recycled
and RCA, the ratios of tensile to shear cracks of RFAC, MSRCAC and aggregate as sustainable concrete: A review. 2nd International Conference on
FRAC increased by about 10.5%, 34.7% and 40.2%, respectively. Recent Trends in Construction Materials and Structures (ICON), Elsevier, Electr
(4) The RFA and RCA accelerated the growth of the FPZ and weak­ Network, 2021, pp. 1029–1034.
[9] J. Xiao, X. Ma, Q. Liu, H. Zhang, Z. Duan, Evolvement and research progress of
ened its toughening effect. The lFPZ of RAC was about 20.2–34.1 mm at concept for full recycled concrete, J. Archit. Civ. Eng. 38 (02) (2021) 1–15, https://
peak load, accounting for approximately 0.29–0.49 of the ligament doi.org/10.19815/j.jace.2020.11036.
height, and the lFPZ of the RFAC, MSRCAC and FRAC increased by about [10] W. Shen, Z. Yang, L. Cao, L. Cao, Y. Liu, H. Yang, Z. Lu, J. Bai, Characterization of
manufactured sand: particle shape, surface texture and behavior in concrete,
48.7%, 35.9% and 68.7% than that of the MSNAC. However, the fully Constr. Build. Mater. 114 (2016) 595–601, https://doi.org/10.1016/j.
developed lFPZ decreased with the increase of RFA and RCA, the fully conbuildmat.2016.03.201.
developed lFPZ of RAC were about 59.4–67.2 mm, representing [11] L. Lin, The research course of pitsand and pitsand concrete, Sichuan Building,
Science 36 (03) (2010) 188–190 [in Chinese].
0.85–0.96 of the ligament height, and the fully developed lFPZ of RFAC,
[12] W.G. Shen, J.L. Wu, X.J. Du, Z.T. Li, D.L. Wu, J.T. Sun, Z.W. Wang, X.J. Huo, D.
MSRCAC and FRAC decreased by about 7.8%, 7.7% and 11.6%. Q. Zhao, Cleaner production of high-quality manufactured sand and ecological
(5) The RFA and RCA reduced the WFPZ of concrete, and the WFPZ,max utilization of recycled stone powder in concrete, J. Clean. Prod. 375 (2022) 14,
https://doi.org/10.1016/j.jclepro.2022.134146.
of RAC was approximately 17.7–31.3 mm, which was 1.8–3.1 times the
[13] J. Xiao, W. Li, Y. Fan, X. Huang, An overview of study on recycled aggregate
maximum aggregate size. Compared with MSNAC, the WFPZ,max of concrete in China (1996–2011), Constr. Build. Mater. 31 (2012) 364–383, https://
RFAC, MSRCAC and FRAC decreased by about 33.0%, 38.2% and doi.org/10.1016/j.conbuildmat.2011.12.074.
43.6%, respectively. The RFA and RCA reduced the aggregate inter­ [14] J. Kim, Influence of quality of recycled aggregates on the mechanical properties of
recycled aggregate concretes: an overview, Constr. Build. Mater. 328 (2022) 12,
locking effect, resulting in a reduction in secondary crack activity and https://doi.org/10.1016/j.conbuildmat.2022.127071.
damage width in concrete, while the effect of the RCA on the damage [15] J.Z. Xiao, Y.X. Tang, H.N. Chen, H.H. Zhang, B. Xia, Effects of recycled aggregate
mechanism of concrete was greater than that of the RFA. combinations and recycled powder contents on fracture behavior of fully recycled
aggregate concrete, J. Clean. Prod. 366 (2022) 14, https://doi.org/10.1016/j.
jclepro.2022.132895.
CRediT authorship contribution statement [16] K. Ouyang, J.H. Liu, S.H. Liu, B.X. Song, H. Guo, G.S. Li, C.J. Shi, Influence of pre-
treatment methods for recycled concrete aggregate on the performance of recycled
concrete: a review, Resour. Conserv. Recycl. 188 (2023) 18, https://doi.org/
Yingjie Xu: Writing – original draft, Methodology, Investigation, 10.1016/j.resconrec.2022.106717.
Conceptualization. Hongniao Chen: Writing – review & editing, Su­ [17] M. Ibrahim, W. Alimi, R. Assaggaf, B.A. Salami, E.A. Oladapo, An overview of
factors influencing the properties of concrete incorporating construction and
pervision, Funding acquisition. Yihui Liang: Writing – review & editing.
demolition wastes, Constr. Build. Mater. 367 (2023) 30, https://doi.org/10.1016/j.
Jie Shen: Writing – review & editing. Huaxiang Yang: Writing – review conbuildmat.2023.130307.
& editing. [18] H. Guo, C.J. Shi, X.M. Guan, J.P. Zhu, Y.H. Ding, T.C. Ling, H.B. Zhang, Y.L. Wang,
Durability of recycled aggregate concrete - a review, Cem. Concr. Compos. 89
(2018) 251–259, https://doi.org/10.1016/j.cemconcomp.2018.03.008.
[19] S. Xu, Fracture mechanics of concrete, Science Press, Beijing, 2011.
Declaration of Competing Interest [20] S. Xu, L. Dong, B. Wang, Q. Li, Development of Concrete Fracture Mechanics 30
Years in China, J. Hydraul. Eng. 45 (S1) (2014) 1–9, http://doi.org/10.13243/j.
The authors declare that they have no known competing financial cnki.slxb.2014.S1.001. [in Chinese].
[21] C. Wang, J. Xiao, Material properties and structural behaviors of recycled fine
interests or personal relationships that could have appeared to influence aggregate concrete: an overview, China Civ. Eng. J. 55 (05) (2022) 37–53, https://
the work reported in this paper. doi.org/10.15951/j.tmgcxb.2022.05.008.

18
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

[22] L.H. Zhu, Q.J. Ning, W. Han, L.L. Bai, Compressive strength and microstructural [49] M.E. Sosa, Y.A.V. Zaccardi, C.J. Zega, A critical review of the resulting effective
analysis of recycled coarse aggregate concrete treated with silica fume, Constr. water-to-cement ratio of fine recycled aggregate concrete, Constr. Build. Mater.
Build. Mater. 334 (2022) 14, https://doi.org/10.1016/j. 313 (2021) 7, https://doi.org/10.1016/j.conbuildmat.2021.125536.
conbuildmat.2022.127453. [50] A.S. Brand, J.R. Roesler, A. Salas, Initial moisture and mixing effects on higher
[23] K. Wu, S. Luo, J. Zheng, Fracture propertie analysis of recycled aggregate concrete quality recycled coarse aggregate concrete, Constr. Build. Mater. 79 (2015) 83–89,
based on digital image correlation technique, Eng. Mech. 39 (03) (2022) 147–157, https://doi.org/10.1016/j.conbuildmat.2015.01.047.
https://doi.org/10.6052/j.issn.1000-4750.2021.01.0067. [51] Standard for test methods of concrete physical and mechanical properties: GB/T
[24] M.H. Guo, S.Y. Alam, A.Z. Bendimerad, F. Grondin, E. Roziere, A. Loukili, Fracture 50081-2019, China Architecture & Building Press, Beijing, 2019. [in Chinese].
process zone characteristics and identification of the micro-fracture phases in [52] B. Pan, Digital image correlation for surface deformation measurement: historical
recycled concrete, Eng. Fract. Mech. 181 (2017) 101–115, https://doi.org/ developments, recent advances and future goals, Meas. Sci. Technol. 29 (8) (2018)
10.1016/j.engfracmech.2017.07.004. 32, https://doi.org/10.1088/1361-6501/aac55b.
[25] T. Li, J.Z. Xiao, Y.M. Zhang, B.C. Chen, Fracture behavior of recycled aggregate [53] J. Sainz-Aja, I. Carrascal, J.A. Polanco, C. Thomas, Fatigue failure
concrete under three-point bending, Cem. Concr. Compos. 104 (2019) 11, https:// micromechanisms in recycled aggregate mortar by mu CT analysis, J. Build. Eng.
doi.org/10.1016/j.cemconcomp.2019.103353. 28 (2020) 11, https://doi.org/10.1016/j.jobe.2019.101027.
[26] F. Kazemian, H. Rooholamini, A. Hassani, Mechanical and fracture properties of [54] S.T. Li, X.Q. Fan, X.D. Chen, S.S. Liu, Y.Z. Guo, Development of fracture process
concrete containing treated and untreated recycled concrete aggregates, Constr. zone in full-graded dam concrete under three-point bending by DIC and acoustic
Build. Mater. 209 (2019) 690–700, https://doi.org/10.1016/j. emission, Eng. Fract. Mech. 230 (2020) 14, https://doi.org/10.1016/j.
conbuildmat.2019.03.179. engfracmech.2020.106972.
[27] Y.J. Xu, H.N. Chen, Y.X. Tang, Study on fracture parameters and fracture process [55] A. Djerbi, Effect of recycled coarse aggregate on the new interfacial transition zone
zone of manufactured-sand recycled aggregate concrete, Constr. Build. Mater. 361 concrete, Constr. Build. Mater. 190 (2018) 1023–1033, https://doi.org/10.1016/j.
(2022) 16, https://doi.org/10.1016/j.conbuildmat.2022.129613. conbuildmat.2018.09.180.
[28] S. Bhowmik, S. Ray, An experimental approach for characterization of fracture [56] F.H. Wang, D.Y. Gao, Z. Xu, T. Zhang, Comparative study on axial compressive
process zone in concrete, Eng. Fract. Mech. 211 (2019) 401–419, https://doi.org/ properties of steel fiber-reinforced recycled-fine-aggregate concrete tested by prism
10.1016/j.engfracmech.2019.02.026. and cylinder specimens, Constr. Build. Mater. 367 (2023) 24, https://doi.org/
[29] Z. Wu, H. Rong, J. Zheng, F. Xu, W. Dong, An experimental investigation on the 10.1016/j.conbuildmat.2023.130311.
FPZ properties in concrete using digital image correlation technique, Eng. Fract. [57] S.W. Hu, J. Lu, F.P. Xiao, Evaluation of concrete fracture procedure based on
Mech. 78 (17) (2011) 2978–2990, https://doi.org/10.1016/j. acoustic emission parameters, Constr. Build. Mater. 47 (2013) 1249–1256, https://
engfracmech.2011.08.016. doi.org/10.1016/j.conbuildmat.2013.06.034.
[30] R.X. Zhou, H.M. Chen, Mesoscopic investigation of size effect in notched concrete [58] D. Kocab, L. Topolar, B. Kucharczykova, R. Halamova, The analysis of acoustic
beams: The role of fracture process zone, Eng. Fract. Mech. 212 (2019) 136–152, emission signals detected during the loading of cement-based materials, Eng. Fail.
https://doi.org/10.1016/j.engfracmech.2019.03.028. Anal. 99 (2019) 18–25, https://doi.org/10.1016/j.engfailanal.2019.01.002.
[31] C.C. Che, D. Li, F.D. Chen, Study on the cracking sensitivity of recycled fine [59] S.L. Xu, Q.H. Li, Y. Wu, L.X. Dong, Y. Lyu, H.W. Reinhardt, C.K.Y. Leung, G. Ruiz,
aggregate concrete based on orthogonal experiment. 2nd International Conference S. Kumar, S.W. Hu, RILEM Standard: testing methods for determination of the
on Structures and Building Materials (ICSBM), Trans Tech Publications Ltd, double-K criterion for crack propagation in concrete using wedge-splitting tests
Hangzhou, PEOPLES R CHINA, 2012, pp. 641–645. and three-point bending beam tests, recommendation of RILEM TC265-TDK,
[32] Y.X. Tang, J.Z. Xiao, D.C. Wang, M.Z. Zhang, Effect of carbonation treatment on Mater. Struct. 54 (6) (2021) 11, https://doi.org/10.1617/s11527-021-01786-8.
fracture behavior of low-carbon mortar with recycled sand and recycled powder, [60] Y.Y. Wang, H. Zhang, Y. Geng, Q.H. Wang, S.M. Zhang, Prediction of the elastic
Cem. Concr. Compos. 142 (2023) 15, https://doi.org/10.1016/j. modulus and the splitting tensile strength of concrete incorporating both fine and
cemconcomp.2023.105178. coarse recycled aggregate, Constr. Build. Mater. 215 (2019) 332–346, https://doi.
[33] A.T. Akono, J.X. Chen, M.M. Zhan, S.P. Shah, Basic creep and fracture response of org/10.1016/j.conbuildmat.2019.04.212.
fine recycled aggregate concrete, Constr. Build. Mater. 266 (2021) 10, https://doi. [61] Y. Jenq, S.P. Shah, Two Parameter Fracture Model for Concrete, J. Eng. Mech. 111
org/10.1016/j.conbuildmat.2020.121107. (10) (1985) 1227–1241.
[34] M. Bravo, J. de Brito, J. Pontes, L. Evangelista, Mechanical performance of concrete [62] S.R. Luo, S.C. Ye, J.Z. Xiao, J.L. Zheng, Y.T. Zhu, Carbonated recycled coarse
made with aggregates from construction and demolition waste recycling plants, aggregate and uniaxial compressive stress-strain relation of recycled aggregate
J. Clean. Prod. 99 (2015) 59–74, https://doi.org/10.1016/j.jclepro.2015.03.012. concrete, Constr. Build. Mater. 188 (2018) 956–965, https://doi.org/10.1016/j.
[35] Z.H. Duan, C.S. Poon, Properties of recycled aggregate concrete made with conbuildmat.2018.08.159.
recycled aggregates with different amounts of old adhered mortars, Mater. Des. 58 [63] W. Chen, H.F. Yang, Fracture performance of concrete incorporating different
(2014) 19–29, https://doi.org/10.1016/j.matdes.2014.01.044. levels of recycled coarse aggregate, Struct. Concr. 10 (2020), https://doi.org/
[36] H. Li, Z. Wang, F. Huang, Z. Yi, Y. Xie, D. Sun, R. Sun, Impact of different 10.1002/suco.201900389.
lithological manufactured sands on high-speed railway box girder concrete, Constr. [64] S. Xu, H.W. Reinhardt, A simplified method for determining double-K fracture
Build. Mater. 230 (2020), https://doi.org/10.1016/j.conbuildmat.2019.116943. parameters for three-point bending tests, Int. J. Fract. 104 (2) (2000) 181–209.
[37] S. Gholizadeh, Z. Leman, B. Baharudin, A review of the application of acoustic [65] S. Khalilpour, E. BaniAsad, M. Dehestani, A review on concrete fracture energy and
emission technique in engineering, Struct. Eng. Mech. 54 (6) (2015) 1075–1095, effective parameters, Cem. Concr. Res. 120 (2019) 294–321, https://doi.org/
https://doi.org/10.12989/sem.2015.54.6.1075. 10.1016/j.cemconres.2019.03.013.
[38] X.D. He, R. Zhou, Z.Y. Liu, S.L. Yang, K. Chen, L. Li, Review of research progress [66] P.E. Petersson, Crack growth and development of fracture zones in plain concrete
and development trend of digital image correlation, Multidiscip. Model. Mater. and similar materials, Report TVBM-1006 (1981).
Struct. 34 (2023), https://doi.org/10.1108/mmms-07-2023-0242. [67] E. Ghorbel, G. Wardeh, Influence of recycled coarse aggregates incorporation on
[39] W. Dong, H. Rong, Q. Wu, J. Li, Investigations on the FPZ evolution of concrete the fracture properties of concrete, Constr. Build. Mater. 154 (2017) 51–60,
after sustained loading by means of the DIC technique, Constr. Build. Mater. 188 https://doi.org/10.1016/j.conbuildmat.2017.07.183.
(2018) 49–57, https://doi.org/10.1016/j.conbuildmat.2018.08.077. [68] J.J. Men, J.C. Wang, L.Y. Guo, K. Wang, Acoustic emission behavior and damage
[40] Y.Z. Guo, X.D. Chen, Z. Wang, Y.J. Ning, L.H. Bai, Identification of mixed mode evaluation of recycled aggregate concrete under compression, Struct. Control.
damage types on rock-concrete interface under cyclic loading, Int. J. Fatigue 166 Health Monit. 27 (10) (2020) 21, https://doi.org/10.1002/stc.2612.
(2023) 15, https://doi.org/10.1016/j.ijfatigue.2022.107273. [69] K. Haidar, G. Pijaudier-Cabot, J.F. Dube, A. Loukili, Correlation between the
[41] Z.M. Ma, J.X. Shen, C.Q. Wang, H.X. Wu, Characterization of sustainable mortar internal length, the fracture process zone and size effect in model materials,
containing high-quality recycled manufactured sand crushed from recycled coarse Materials & Structures 38 (276) (2005) 201–210.
aggregate, Cem. Concr. Compos. 132 (2022) 18, https://doi.org/10.1016/j. [70] K.Y. Wu, S.R. Luo, J.L. Zheng, J.T. Yan, J.Z. Xiao, Influence of carbonation
cemconcomp.2022.104629. treatment on the properties of multiple interface transition zones and recycled
[42] Recycled fine aggregate for concrete and mortar: GB/T 25176-2010, Standards aggregate concrete, Cem. Concr. Compos. 127 (2022) 17, https://doi.org/
Press of China, Beijing, 2010. [in Chinese]. 10.1016/j.cemconcomp.2021.104402.
[43] Recycled coarse aggregate for concrete: GB/T 25177-2010, Standards Press of [71] J. Yu, X. LV, Y. Qin, S. Jiang, Acoustic emission characteristics of recycled concrete
China, Beijing, 2010. [in Chinese]. beams with different substitution rates during bending, Concrete 08 (2020) 27–31,
[44] Sand for construction: GB/T 14684-2022, Standards Press of China, Beijng, 2022. https://doi.org/10.3969/j.issn.1002-3550.2020.08.007.
[in Chinese]. [72] G. Ma, Y.J. Xie, G.C. Long, Z. Tang, C. Tang, H. Wang, Y.Y. Wei, J.T. Li,
[45] Pebble and crushed stone for construction: GB/T 14685-2022, Standards Press of Experimental study on acoustic emission and surface morphology characteristics of
China, Beijing, 2022. [in Chinese]. concrete under different fracture modes, Theor. Appl. Fract. Mech. 123 (2023) 15,
[46] Technical specification for testing concrete strength with drilled core method: JGJ/ https://doi.org/10.1016/j.tafmec.2022.103702.
T 384-2016, China Architecture & Building Press, Beijing, 2011. [in Chinese]. [73] C. Chen, X.Q. Fan, X.D. Chen, Experimental investigation of concrete fracture
[47] Q. Feng, B.Q. Liu, Y.P. Zhang, C. Zhang, D. Wang, Multi-scale grading utilization behavior with different loading rates based on acoustic emission, Constr. Build.
based on the characteristics of recycled concrete aggregates: a review, Front. Mater. 237 (2020) 10, https://doi.org/10.1016/j.conbuildmat.2019.117472.
Mater. 10 (2023) 13, https://doi.org/10.3389/fmats.2023.1219075. [74] N.N. Kencanawati, S. Iizasa, M. Shigeishi, Fracture process and reliability of
[48] A.A.M. Amer, K. Ezziane, A. Bougara, M. Adjoudj, Rheological and mechanical concrete made from high grade recycled aggregate using acoustic emission
behavior of concrete made with pre-saturated and dried recycled concrete technique under compression, Mater. Struct. 46 (9) (2013) 1441–1448, https://
aggregates, Constr. Build. Mater. 123 (2016) 300–308, https://doi.org/10.1016/j. doi.org/10.1617/s11527-012-9986-z.
conbuildmat.2016.06.107. [75] R. Janeliukstis, A. Clark, M. Papaelias, S. Kaewunruen, Flexural cracking-induced
acoustic emission peak frequency shift in railway prestressed concrete sleepers,

19
Y. Xu et al. Construction and Building Materials 419 (2024) 135540

Eng. Struct. 178 (2019) 493–505, https://doi.org/10.1016/j. [78] CEB-90-1991, Committee Euro-International du Beton, 1991.
engstruct.2018.10.058. [79] B. Lu, C.J. Shi, Z.J. Cao, M.Z. Guo, J.L. Zheng, Effect of carbonated coarse recycled
[76] L. Wang, Y. Zhao, J. Guo, J. Zhang, Acoustic emission characteristics of concrete concrete aggregate on the properties and microstructure of recycled concrete,
and its constituent materials under uniaxial failure, J. Civ. Environ. Eng. (2022) J. Clean. Prod. 233 (2019) 421–428, https://doi.org/10.1016/j.
1–9, http://doi.org/10. 11835/j. issn. 2096-6717. 2023. 006. jclepro.2019.05.350.
[77] J. Li, J. Yan, J. Niu, Effect of freeze-thaw action on acoustic emission [80] Y.X. Tang, H.N. Chen, J.Z. Xiao, Size effects on the characteristics of fracture
characteristics of self-compacting lightweight aggregate concrete, Eng. Mech. 39 process zone of plain concrete under three-point bending, Constr. Build. Mater. 315
(09) (2022) 133–140+169, https://doi.org/10.6052/j.issn.1000- (2022) 13, https://doi.org/10.1016/j.conbuildmat.2021.125725.
4750.2021.05.0373.

20