Journal of Cleaner Production 278 (2021) 123248

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Stress-strain behaviour of steel-fibre-reinforced recycled aggregate

concrete under axial tension
Shuai Zong, Zhenzhen Liu, Shan Li, Yiyan Lu*, Aohan Zheng
School of Civil Engineering, Wuhan University, Wuhan City, 430072, China

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

Article history: The drastic increase in the generation of construction and demolition waste (CDW) is causing serious
Received 21 February 2020 pollution to the natural environment, such as space reduction, heavy metal contamination, and resource
Received in revised form shortage. In this research, steel fibre and recycled coarse aggregate (RCA) are used to form a steel-fibre-
6 July 2020
reinforced recycled aggregate concrete (SFRAC) to improve the recycling and reuse rate of CDW. Different
Accepted 7 July 2020
quality replacement rates of the recycled aggregate (RAR) and volume percentage of the steel fibres (Vf)
Available online 11 August 2020
are used to explore the influence on the tensile behaviour of the concrete. When the Vf increased, the
^ as de
Handling editor. Cecilia Maria Villas Bo tensile strength and peak strain of SFRAC increased by 10.41e55.90% and 9.70e48.03%, respectively,
Almeida compared to those of the recycled aggregate concrete (RAC). SFRAC has an energy absorption capacity
5.71e13.31 times higher than that of RAC. The experimental results demonstrate that the internal defects
Keywords: of the RAC on the tensile property can be compensated by the bridging and anchoring effect of the steel
Recycled aggregate concrete fibres. In addition, mathematical formulas are established for predicting the tensile strength, peak strain,
Steel fibre and stress-strain curves of the SFRAC. Moreover, SFRAC is recyclable and can be reused at least twice. The
Axial tensile behaviour
method for recycling SFRAC is feasible and cost-effective from an industrial point of view.
Stress-strain curves
© 2020 Elsevier Ltd. All rights reserved.
Energy absorption

1. Introduction imminent and of significant concern to society. The main approach

to alleviating the above problem is the utilisation of recycled
Rampant urbanisation, building ageing, and uncontrollable aggregate concrete (RAC) and its derivatives. As an alternative to
natural disasters inevitably produce significant construction and natural coarse aggregate (NCA), construction waste is fabricated
demolition waste (CDW). In 2016, the European Union produced into a recycled aggregate (RCA) to form RAC through a series of
932.7 million tons of CDW, which has become the largest waste operations (i.e. sorting, crushing, cleaning, and sieving). RAC not
stream (Eurostat, 2018). In 2017, the United States generated 569.4 only offers a promising solution for construction waste but can also
million tons of CDW (EPA, 2017). China, one of the biggest pro- contribute to reducing the dependence on natural gravel sources.
ducers of CDW globally, is expected to generate CDWs that exceed However, RAC is limited by its poor mechanical properties, and
1.8 billion tons per year (Zhang, 2017). Despite such a large amount therefore, is merely used in laying road surfaces, low-rise buildings
of CDW, only 20e30% of CDW is recovered globally (Lo  pez Ruiz with simple structures, and low-stress structural components in
et al., 2020). The recycling and reuse rate of CDW, particularly in practical engineering (Malesev et al., 2010). Compared to natural
China, is less than 5% (Huang et al., 2018). Based on the 90% recy- aggregate concrete (NAC), the compressive strength and split ten-
cling rate of CDW in Japan, at least 1.97 billion tons of CDW should sile strength of RAC are decreased by 13e32% (Hansen, 1986; Nixon,
be recycled globally. Comparatively, even though the construction 1978; Wang et al., 2019; Xiao et al., 2005) and 10e24%
industry is experiencing a shortage of natural gravel resources, the (Chakradhara Rao et al., 2011; Etxeberria et al., 2007; Yang et al.,
waste recycling process is inefficient in most cases, resulting in 2008), respectively. Its flexural strength (Padmini et al., 2009;
large volumes of CDW being disposed of in a landfill or even ille- Xiao, 2018), frost resistance (Bogas et al., 2016; Omary et al., 2016;
gally dumped. The exploitation and upcycling of CDW are Zaharieva et al., 2004), durability (Evangelista and de Brito, 2010;
Guo et al., 2018; Thomas et al., 2013), and modulus of elasticity
(Corinaldesi, 2010; Xiao et al., 2005; Zhou and Chen, 2017) also
present a decreasing trend with an increasing proportion of RCA.
* Corresponding author. Moreover, RAC has salient drying shrinkage (Sagoe-Crentsil et al.,
E-mail address: [email protected] (Y. Lu).

https://doi.org/10.1016/j.jclepro.2020.123248
0959-6526/© 2020 Elsevier Ltd. All rights reserved.
2 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

2001; Silva et al., 2015; Yildirim et al., 2015) and creep behaviour during the tensile process of concrete, and provide the necessary
characteristics (Domingo-Cabo et al., 2009; Fathifazl et al., 2011; constitutive relation for complete elasto-plastic mechanics anal-
Tam et al., 2015). ysis, cross-sectional stress analysis on ultimate state, mechanism
These drawbacks are mostly as a result of the preparation pro- analysis on damage and fracture, and seismic performance
cess of RCA. RCA mostly presents irregular edges and corners or a evaluation.
flaky grain-shape owing to repeated crushing. Concurrently, Therefore, in this study, axial tension experiments were con-
massive micro-cracks are generated and propagated in RCA by ducted on 48 concrete specimens to address the shortcomings in
mechanical extrusion during the crushing process. Furthermore, the existing literature, which aims to expand the application scope
the residual cement mortar adheres to RCA, decreasing the density of RAC, improve the recycling rate of CDW, and provide practical
and water absorption (Jamkar and Rao, 2004; Nagataki et al., 2004). solutions and theoretical bases for the design methodology of
The mechanical properties of the interfacial transition zones (ITZs) SFRAC structures. The influences of the quality replacement rate of
in the microstructure affect the overall mechanical properties of the the recycled aggregate (RAR) and the steel fibre volume (Vf) on the
concrete. Particularly in RAC, the ITZs between the RCA and the old axial tensile behaviour of the RAC are investigated. The failure
or new cement mortar are complex and weak (Poon et al., 2004; mode, stress-strain response, axial tension strength, peak strain,
ez del Bosque et al., 2017; Wang et al., 2013).
Sa and energy absorption are analysed, which can be used in practical
The addition of steel fibres into RAC is an innovative method for engineering design of RAC and SFRAC structures. Moreover, a
compensating for the internal micro-defects and making steel- stress-strain prediction model and the recycling process for SFRAC
fibre-reinforced RAC (SFRAC) feasible as a suitable material in are proposed.
practical engineering. Carneiro et al. (2014) found that the addition
of steel fibres also modified the fracture process of the RAC. 2. Experimental setup
Concurrently, the failure process of the tested specimens gradually
presented outstanding ductile failure characteristics with an in- 2.1. Test parameters and test specimens
crease in the steel fibre volume fraction. Mohseni et al. (2017)
attempted to modify the microstructure of the RAC by employing In this study, the effects of the parameters, RAR and Vf, on the
steel fibres and polypropylene fibres. Conducting scanning electron SFRAC mechanical properties under axial tension are investigated
microscopy revealed that during the hydration process, the fibres by testing 48 specimens. All the experimental specimens have the
did not chemically react with the cement mortar and that the C-S-H same design size, 150 mm
150 mm
550 mm. The different Vf are
products were dense and homogenous around the fibres. These 0%, 0.6%, 1.2%, and 1.8%, and the RARs are designed to be 0%, 50%,
indicated that the fibre and cement paste function together in the 75%, and 100%. Three specimens are fabricated and tested for each
ITZs. Senaratne et al. (2016) considered the SFRAC to be an inno- case to ensure the precision of the test. The naming rule, as listed in
vative and promising structural material and proposed that 30% Table 1, includes the RAR, Vf, and the serial number of the test
RCA with 0.6% volume fraction of steel fibres was a satisfactory specimens with identical parameters. For example, specimen
combination. Liu et al., 2019 adopted the SFRAC as a filling to form R50%-St0.6%-1 is the first specimen with 50% RAR and 0.6% Vf.
concrete-filled steel tube columns. The steel fibres could alleviate
the friability and fissility of the RAC, and the steel tube caused the 2.2. Materials
SFRAC to be in a state of triaxial stress, leading to the compressive
performance of SFRAC. 2.2.1. Coarse aggregates
Guo et al. (2014) and Xie et al. (2015) observed that the addition In the present study, two types of coarse aggregates were
of rubber, forming a rubberised SFRAC, can also increase the adopted. The RCA was sourced from the laboratory waste concrete,
compressive strength, impact resistance, and energy dissipation which has an original strength of C30. The waste concrete was
capacity of concrete. Rubber provides a high strain capacity, and crushed by a jaw crusher and sieved to 5e25 mm, same as the NCA
steel fibre provides significant residual post-peak strength. How- size. Compared to the NCA, the RCA is in contact with more hard-
ever, the slump flow may be challenging to meet the requirement ened cement mortar on its surface and also has numerous cavities
in the field due to the increased rough surface and surface fiction of and micro-cracks. Additionally, the RCA is mostly irregular angular
rubber particles (Si et al., 2018). Xie et al. (2018a, 2018b) verified shaped. Fig. 1 presents the distribution curves of the particle sizes
that both the filler effect and the pozzolanic effect of silica fume of the NCA and RCA.
enhance the mechanical behaviour of concrete, especially its
compressive strength. Compared to silica fume, fly ash is mainly
used to increase the workability and durability of concrete. The Table 1
Details of the specimens.
pore structure is improved, and particularly the volume of macro-
pores is reduced. Furthermore, the use of fly ash is more cost- Group RAR (%) Vf (%)
effective than silica fume for concrete production (Corinaldesi R0%-St0.0% 0 0.0
and Moriconi, 2009). R0%-St0.6% 0 0.6
The use of SFRAC as a sustainable and eco-friendly composite R0%-St1.2% 0 1.2
material, is therefore proposed. Presently, more studies are centred R0%-St1.8% 0 1.8
R50%-St0.0% 50 0.0
on the mechanical properties of RAC than on those of SFRAC. In R50%-St0.6% 50 0.6
addition, these studies mainly focus on the split tensile strength R50%-St1.2% 50 1.2
and flexural strength, whereas researches on the stress-strain R50%-St1.8% 50 1.8
behaviour of RAC and SFRAC under axial tension are scarce. Uni- R75%-St0.0% 75 0.0
R75%-St0.6% 75 0.6
axial tensile testing is not recommended for the standard testing of
R75%-St1.2% 75 1.2
new mixtures because tensile tests are difficult to carry out and R75%-St1.8% 75 1.8
interpret (FIB, 2013). Notably, the tensile behaviour influences the R100%-St0.0% 100 0.0
cracking, bonding, and durability of concrete members. Further- R100%-St0.6% 100 0.6
more, the stress-strain curve is a macroscopic reflection of the R100%-St1.2% 100 1.2
R100%-St1.8% 100 1.8
extension and damage accumulation of internal micro-cracks
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 3

fibre can provide mechanical interlock due to frictional effect,

thereby achieving the desired bond strength with the matrix
(Beglarigale and Yazıcı, 2015; Wille and Naaman, 2012). Zhang et al.
(2013) asserted that steel fibre, possessing large aspect ratio, was
beneficial to tensile behaviour of SFRAC. Le Hoang and Fehling
(2017) found that the short steel fibres were more effective than
long steel fibres for resisting the micro-cracks. Therefore, hooked-
end steel fibres with short length and large aspect ratio were
chosen in this study, whose properties are listed in Table 3.
Based on the reference (Zhao, 1999), adding steel fibre volume in
excess of 2% could lead to a significant decrease in concrete flow-
ability. However, steel fibre content of less than 0.5% would fail to
affect the mechanical properties of concrete. Therefore, Vf was set
as 0.6%, 1.2%, and 1.8% to strengthen the tensile behaviour of the
RAC in this study.

2.2.4. Cementitious materials

Fig. 1. Distribution curves of the particle sizes of the fine and coarse aggregates.
The cementitious materials used in the test were ordinary
Portland cement 42.5 (P$O 42.5) and fly ash (FA). The ordinary
With reference to the Chinese codes Pebble and crushed stone for Portland cement was the main binder, and fly ash was chosen as an
construction (GB/T 14685-2011, 2011) and Recycled coarse aggregate auxiliary cementitious material. Certainly, the mechanical proper-
for concrete (GB/T 25177-2010, 2010), the measured data of the ties of ITZs between the cement paste and coarse aggregate is a
physical and mechanical properties of the coarse aggregates are crux, particularly in the RAC. Nevertheless, the cementitious C-S-H
given in Table 2. The measured physical parameters broadly are the formed by pozzolanic reaction between fly ash and inherent
apparent density, water absorption, and crushing index, and the Ca(OH)2 can strengthen the ITZs. The compositions of the cemen-
mechanical property analysis is mainly focused on the crushing titious materials are provided in Table 4.
index. As recorded in Table 2, the apparent density of the RCA is
only the 89.34% that of the NCA, whereas the water absorption of 2.3. Mixture proportions
the RCA, which occurs in 30 min, is 2.74 times higher than that of
the NCA. In addition, the crushing indices of the NCA and RCA are Concrete mixing water consists of two parts: original mixing
7.71% and 14.70%, respectively. The crushing index affects the water and additional mixing water. The dosage of the original
compressive strength. A large crushing index generally is related to mixing water depends on the designed water-binder ratio. All the
low compressive strength. The experimental results demonstrate mixtures had identical water-binder ratios of 0.326. Considering
that although the RCA has a lower apparent density and the higher water absorption of the RCA, additional mixing water
compressive strength than that of NCA, it seems to have more was required during the test to ensure satisfactory workability of
tendency to absorb water. the mixture. The dosage of additional mixing water is determined
by the usage amount of the RCA and its water absorption. The
proportions for each sample of this study are summarised in
2.2.2. Fine aggregates
Table 5. It is noteworthy that the fluidity is low when Vf is 1.8%.
River sand with a particle size of 2.5 mm was selected as a fine
Thus, it is necessary to extend the stirring time appropriately to
aggregate. The moisture content of the river sand was precisely
prevent the steel fibres from accumulating in the mixture. The
measured prior to casting concrete to ensure no interference of the
compressive strength (fcu) and serviceability performance of the
waterebinder ratio. The fineness modulus of the river sand was
concrete samples were measured complying with the Chinese
2.46, and its grading curve is depicted in Fig. 1. The upper and lower
codes Standard for test method of performance on ordinary concrete
limits of fine aggregate usage were according to the Chinese code
(GB/T 50080-2016, 2016) and Standard for test method of mechanical
Pebble and crushed stone for construction (GB/T 14685-2011, 2011).
properties on ordinary (GB/T 50081-2016, 2016). Table 6 records fcu
The fine aggregate satisfied the requirement for construction, as
values with various RARs and Vf.
expected.

2.4. Specimen preparation

2.2.3. Steel fibre
In concrete specimens, steel fibre failure modes are broadly Based on the Chinese code Standard test methods for fibre rein-
classified into debonding failure and rupture failure during the forced concrete (CECS13:2009, 2009), the specimens were sized as
tensile process; debonding failure is the main failure mode because 150 mm
150 mm
550 mm, and it was pre-embedded with an
steel fibres possess favourable tensile properties. Therefore, for axial steel bar. Fresh concrete was cast into a customised wooden
improving the tensile properties of concrete, it is vital and indis- mould inserted with a deformed bar with a diameter of 20 mm.
pensable to enhance the bond characteristics between the matrix Pre-rust cleaning of the bars was conducted to ensure excellent
and the fibres. The fibre material, type, and shape affect the bond bonding with the concrete. Subsequent to pouring the concrete, the
behaviour. Compared to a straight steel fibre, a hooked-end steel test specimens were covered with a polyethene film to prevent
moisture loss. During a 28-day curing stage in the laboratory, grid
Table 2
lines were laid on all sides of the specimen, for the observation of
Physical and mechanical properties of coarse aggregates. the fracture morphology during the test process.
Type Apparent density (kg/m3) Water absorption (%) Crush index (%)
2.5. Test set-up
NCA 2721 1.08 7.71
RCA 2431 2.96 14.70
The test programme consists of both direct tensile tests and
4 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Table 3
Performance indices of the hooked-end steel fibres.

Parameters Measured value Picture

Length (lf) 30 mm±1 mm

Diameter (df) 0.52 mm ± 0.02 mm
Length-to-diameter ratio 57.6
Density 7.85 g cm3
Tensile strength 1060 MPa
Schematic diagram

comprehensive tests. The direct tensile tests were conducted on a

servo-hydraulic closed-loop universal testing machine with a



ε  ε2

e ¼

1

100% (1)
ε1 þ ε2

600 kN capacity. The experiment devices and the measurement
points in the experiment are depicted in Fig. 2. In the set-up, hy-
draulic chucks were placed at the upper or lower sides of the where ε1 and ε2 are the strain value of both sides of the specimen.
testing machine to clamp the pre-embedded steel bar during the Testing devices were connected to a German IMC real-time data
test. Two linear variable differential transducers (LVDTs) were fixed acquisition system to record the history of the load, strain, and axial
on the prefabricated steel hoops, located at both ends of the deformation. In the formal loading, force-control loading was
specimen, to record the axial deformation. Two strain gauges were
attached to the axis of both symmetrical surfaces. Before the formal
loading, a pre-loading test within 15e20% of the ultimate load was Table 6
conducted. The pre-loading test was applied to control the eccen- Compressive strengths of specimens.
tricity of each specimen within 15% by continually adjusting the Specimen ID fcu (MPa) Mean value (MPa) RSD (%)
location and ensuring proper operation of the testing machine and
No. 1 No. 2 No. 3
data acquisition system. The eccentricity was calculated using Eq.
R0%-St0.0% 45.2 44.7 43.7 44.5 1.72
(1).
R0%-St0.6% 46.3 49.0 46.1 47.1 3.44
R0%-St1.2% 51.6 49.8 49.6 50.3 2.19
R0%-St1.8% 53.5 54.3 49.8 52.5 4.57
Table 4 R50%-St0.0% 42.6 40.9 41.4 41.6 2.10
Chemical compositions of P$O 42.5 and FA. R50%-St0.6% 44.5 43.6 46.5 44.9 3.31
R50%-St1.2% 48.9 47.3 51.0 49.1 3.78
Mineral compositions P$O 42.5 FA
R50%-St1.8% 52.6 50.5 49.2 50.8 3.38
3
Specific gravity 3.16 kg/m 1.51 kg/m3 R75%-St0.0% 37.3 37.9 38.6 37.9 1.72
Al2O3 4.4% 39.5% R75%-St0.6% 42.1 42.3 43.1 42.5 1.25
CaO 63.2% 13.7% R75%-St1.2% 44.0 47.4 45.5 45.6 3.73
SiO2 20.5% 15.1% R75%-St1.8% 47.2 48.5 50.0 48.6 2.89
SO3 2.7% 1.6% R100%-St0.0% 36.9 37.1 36.4 36.8 0.98
Water e 0.1% R100%-St0.6% 39.4 41.6 43.9 41.6 5.40
a
LO.I 2.3% 3.8% R100%-St1.2% 44.8 44.8 46.4 45.3 2.04
a R100%-St1.8% 45.8 48.0 47.7 47.2 2.53
LO. I ¼ Loss on ignition.

Table 5
Concrete mixture proportions.

Specimen ID Content (kg/m3) Plasticizer (%) Vf (%) D (mm)

P$O 42.5 NCA RCA Sand Water FA

R0%-St0.0% 382.0 810.0 0.0 788.0 178.0 164.0 0.8 0.0 661
R0%-St0.6% 382.0 810.0 0.0 788.0 178.0 164.0 0.8 0.6 655
R0%-St1.2% 382.0 810.0 0.0 788.0 178.0 164.0 0.8 1.2 612
R0%-St1.8% 382.0 810.0 0.0 788.0 178.0 164.0 0.8 1.8 572
R50%-St0.0% 382.0 405.0 405.0 788.0 185.6 164.0 0.8 0.0 647
R50%-St0.6% 382.0 405.0 405.0 788.0 185.6 164.0 0.8 0.6 639
R50%-St1.2% 382.0 405.0 405.0 788.0 185.6 164.0 0.8 1.2 599
R50%-St1.8% 382.0 405.0 405.0 788.0 185.6 164.0 0.8 1.8 549
R75%-St0.0% 382.0 202.5 607.5 788.0 189.4 164.0 0.8 0.0 622
R75%-St0.6% 382.0 202.5 607.5 788.0 189.4 164.0 0.8 0.6 617
R75%-St1.2% 382.0 202.5 607.5 788.0 189.4 164.0 0.8 1.2 575
R75%-St1.8% 382.0 202.5 607.5 788.0 189.4 164.0 0.8 1.8 527
R100%-St0.0% 382.0 0.0 810.0 788.0 193.2 164.0 0.8 0.0 613
R100%-St0.6% 382.0 0.0 810.0 788.0 193.2 164.0 0.8 0.6 608
R100%-St1.2% 382.0 0.0 810.0 788.0 193.2 164.0 0.8 1.2 551
R100%-St1.8% 382.0 0.0 810.0 788.0 193.2 164.0 0.8 1.8 503
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 5

3. Experimental results and analyses

3.1. Experimental observation and specimen failure pattern

Fig. 3 exhibits the crack morphologies of the concrete. The

failure processes of the NAC and RAC are notably similar. A few
observable cracks are found on the specimen surface in the early
stage of the test loading. Subsequently, a penetrating crack is
rapidly formed in the specimen’s middle area as the tensile loading
increases. Concurrently, it is accompanied by a loud sound, and the
concrete specimen breaks immediately. The crack morphologies of
the NAC and RAC present an inclination angle approaching hori-
zontality. In general, it is difficult to observe the crack expansion
and extension process owing to brittle failure.
The failure form of the SFRAC significantly differs from that of
the NAC and RAC. However, its failure process is almost similar to
that of the steel-fibre-reinforced NAC (SFNAC). Initially, there is no
crack on the specimen surface; however, micro-cracks appear in
the middle of the specimen as the tensile loading increases. As the
test load further increases, the micro-cracks expand, accompanied
by a slight and noticeable cracking sound. As the peak load ap-
proaches, the micro-cracks gradually coalesce into a macro-crack,
with the concrete peeling off. The close-up photographs of the
crack propagation at different time phases are exhibited in Fig. 4.
After the peak load is reached, the tensile load drops until the test is
completed, and the slope of the decreasing phase becomes steep as
RAR increases but flat as Vf increases. Finally, the macro-crack is
transformed into an approximately horizontal penetrating crack.
Fig. 2. Testing machine and measurement points. Unlike the RAC cases, the failure processes of the SFRAC usually
show ductile failure behaviour.
The fracture surfaces of the RAC and SFRAC specimens are dis-
adopted with a rate of 0.4 MPa/min. When the peak load was played in Fig. 5. The fracture surface of the RAC specimen is found to
reached, the loading mode was automatically altered to the be clear, flat, and relatively symmetrical. Notably, the fracture po-
displacement-control loading mode with a rate of 0.1e0.2 mm/min. sition is mainly located at the old cement mortar, which contains
Notably, when the specimen was close to failure, the valve of the several cavities. Contrastingly, the fracture surface of the SFRAC
testing machine was stopped for adjustment until the specimen specimen is rough with a large amount of cement mortar scattered
was broken. The maximum load was recorded and fixed to be in the section, and the pulled-out or broken steel fibres are exposed
0.01 MPa. During the formal loading, the load history was obtained in the failure section.
from the load transducer of the testing machine, while the defor- The failure modes of the steel fibres were observed and statis-
mation was computed by the average reading of the two LVDTs. The tically analysed. h, calculated using Eq. (2), is defined as the ratio of
tensile stress was calculated by dividing the measured tensile load the number of broken steel fibres to the number of pulled-out steel
by the cross-section area. All the test methods follow the Chinese fibres. Furthermore, a relatively large h value is associated with
code Standard test methods for fibre reinforced concrete fully utilised steel fibres. This study stipulates that if any of the
(CECS13:2009, 2009).

Fig. 3. Typical failure modes of the concrete.

6 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Fig. 4. Close-up photographs of the crack propagation at different time phases.

Fig. 5. Fracture surfaces of RAC and SFRAC.

following conditions are met, the steel fibres can be determined as number of pulled-out steel fibres in the two failure surfaces.
having been pulled out. The criteria for the determination are as Moreover, c21 and c22 are the number of broken steel fibres in the
follows: two failure surfaces. The statistical results are listed in Table 7.
From Fig. 6, it is noted that when Vf ¼ 1.2%, h exhibits a
 Steel fibres are completely presented in the failure section. decreasing trend as RAR and Vf increasing. This can be attributed to
 A part of the steel fibres is exposed; however, the integrity of the the insufficient compact of the recycled concrete matrix to
shapes of the hook-end steel fibres is maintained, and there is no providing a good anchorage for steel fibres. Moreover, steel fibres
notable fracture section. with increasing volume content are expected to agglomerate more;
 Steel fibres are largely deformed, and there is a noticeable therefore, their effect cannot be fully exerted.
looseness, which can be felt by hand.

Steel fibres not satisfying any of the above conditions are 3.2. Axial tensile stress-strain curves
determined to be broken. The criteria for counting are as follows:
As can be seen from Figs. 7 and 8, the stress-strain curves ob-
 The steel fibres that are pulled out or broken are counted in the tained from the experiments can be divided into two characteristic
two failure surfaces of a particular specimen. parts: the ascending stage and the descending branch. The stress-
 The number of steel fibres that are pulled out from the two strain curves of the NAC and RAC show only an ascending stage.
failure surfaces is added, and the result is calculated using Eq. However, the SFNAC and SFRAC stress-strain curves have a signif-
(3). icant descending branch. This is because both the NAC and RAC
 The average of the broken steel fibres in the two failure surfaces undergo brittle failure. As the bearing capacity is lost instantly, the
is obtained, and the result yields the number of steel fibres in protective device of the testing machine is activated, and simulta-
the failure section, as calculated using Eq. (4): neously, the data collection is stopped. In contrast, the SFNAC and
SFRAC undergo a ductile failure. Depending on the bridging effect
h ¼ C2 =C1 (2) between the steel fibre and the concrete matrix, the steel fibres can
provide considerable tensile capacity, which retards the micro-
crack expansion in the concrete.
C1 ¼ c11 þ c21 (3) Fig. 7 displays the influence of the RAR on the stress-strain
curves. The peak stress (st) and the peak strain (εt) decrease as
C2 ¼ ðc11 þ c12 Þ=2 (4) RAR increases, and the overall slopes of the curves in the ascending
stage also exhibit a decreasing trend. A high RAR content is related
Where C1 is the number of pulled-out steel fibres, and C2 is the to extra RCA usage. In the tension process, the more the old cement
number of broken steels fibres. Further, c11 and c21 represent the mortar attached to the coarse aggregates, the greater the
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 7

Table 7
Statistical results.

Specimen ID No. 1 No. 2 Mean value h (%)

Broken Pulled-out Broken Pulled-out Broken Pulled-out

R50%-St0.6% 22 86 26 81 24 84 28.57
R0%-St1.2% 43 173 47 199 45 186 24.19
R50%-St1.2% 45 188 40 205 43 197 21.83
R75%-St1.2% 32 208 41 197 37 203 18.23
R100%-St1.2% 37 201 29 186 33 194 17.01
R50%-St1.8% 50 263 55 281 53 272 19.49

compared to those of NAC. The macro mechanical properties are

hindered by the high RAR content, which increases the internal
defects in the concrete. However, the ft values of SFRAC increases by
10.41e43.87%, 17.97e49.77%, and 26.67e55.90% compared to those
of the RAC when RAR is 50%, 75%, and 100%, respectively. An in-
crease in the Vf decreases the spacing between fibres, which makes
more steel fibres to sustain the tensile load. Conversely, the steel
fibres can provide considerable tensile capacity depending on their
outstanding tensile properties and the friction and anchorage force
between the fibres and the concrete matrix. In addition, it should be
noted that the improvement of steel fibres on the ft is more sig-
nificant as the RAR increases. Accordingly, an empirical formula for
predicting the ft is proposed:
   
ft ¼ 1 þ n , lf
k , rg þ 0:24 ,f 2=3
cu (5)

.
Fig. 6. Statistical result. lf ¼ Vf ,lf df (6)

Where rg represents the RAR of the concrete, lf is regarded as the

deformation. In addition, numerous micro-cracks, which can easily
steel fibre parameter, k ¼ 0.059, and n ¼ 0.428. Comparisons of
extend under the applied tensile load, are accumulated in the RCA
the present experimental results, other published results (Han
during mechanical crushing. Thus, the elastic modulus of the RAC
et al., 2006; Okeh et al., 2019; Xiao and Lan, 2006), and the pre-
exhibits a decreasing trend. This is also reflected in the SFRAC.
dicted values of ft are shown in Fig. 10(a), which shows a satisfac-
However, the decreasing trend is more alleviated as the RAR
tory result.
increases.
Fig. 8 exhibits the influence of the Vf on the stress-strain curves.
Notably, the stress-strain curves of the SFRAC include a significant
descending branch unlike those of the RAC. As the Vf increases, the 3.4. Peak strain
st and εt of the SFRAC increases. Concurrently, the stress-strain
curves of the SFRAC with high Vf tend to be complete, and the Table 10 records the measured values of the peak strain (εt) of
descending branch becomes relatively flat. Although the RAR and Vf the concrete with different RARs and Vf, as well as the average
have a remarkable impact on the stress-strain curves, their shapes values and RSD, of each test group. The requirements for the ho-
for all the RACs are similar to those of the NAC. Furthermore, the mogeneity of variance are satisfied among the test groups, such
shapes of the stress-strain curves for all the SFRACs are similar to that the two-way analysis of ANOVA can be used, as shown in
those of the SFNACs. Table 11. The F value illustrates that Vf has a more prominent effect
on εt than RAR.
3.3. Tensile strength As can be seen from Fig. 9(b), as the RAR increases, the εt of RAC
shows a tendency to increase and subsequently decrease.
Table 8 lists the measured values of the axial tensile strength (ft) Compared to the εt value of NAC, the εt values of RAC increase by
of the concrete with different RARs and Vf, as well as the average 13.45% when RAR is 50%, whereas it decreases by 4.78% and 9.31%
values and relative standard deviation (RSD) of each test group. The when RAR is 75% and 100%, respectively. Although RCA is prone to
low RSD indicates that the experimental results are accurate, deform during tension process, the internal defects in the concrete
repeatable, and reproducible. In addition, the requirements for the accumulate due to the increased usage of RCA, making it easy for
homogeneity of variance are satisfied among the test groups, such internal micro-cracks to develop into penetrating cracks. However,
that the two-way analysis of variance (ANOVA) can be used, as the εt values of SFRAC increase by 9.70e39.38%, 17.26e40.57%, and
shown in Table 9. From the F value, it can be seen that both RAR and 20.50e48.03% compared to those of RAC when RAR is 50%, 75%, and
Vf have a significant impact on ft, and the effect of Vf on ft is higher 100%, respectively. The εt value of RAC increases as Vf increases
than that of RAR, as depicted in Fig. 9(a). because the steel fibres distributed in the concrete play a pivotal
Generally, the ft of RAC exhibits a decreasing trend as RAR in- role in retarding the formation and propagation of macro-cracks.
creases and an increasing trend as Vf increases. Without the addi- Hence, the SFRAC undergoes a larger deformation under the peak
tion of steel fibres, when RAR is 50%, 75%, and 100%, the ft values of load than the RAC. Based on the experimental data, the following
RAC decreases by 12.66%, 29.55%, and 36.69% respectively empirical formula is proposed for predicting εt of the SFRAC:
8 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Fig. 7. The influence of RAR on the stress-strain curves. (a) Vf ¼ 0.0%, (b) Vf ¼ 0.6%, (c) Vf ¼ 1.2%, and (d) Vf ¼ 1.8%.

     2
εt ¼ 1 þ m , lf
b , rg þ c ,ft0:54
106 ; (7) 0:312
 0:238 rg þ 2:582
at ¼ ; and (10)
1 þ q,lf
where the parameters, b ¼ 3.6, c ¼ 63, and m ¼ 0.2, are obtained


using a data regression analysis. The theoretical calculation results dy

E0
a¼ ¼ ; (11)
are presented in Fig. 10(b), which agree well with the test data and dx
x ¼ 0 Ec
other published results (Han et al., 2006; Okeh et al., 2019; Xiao and
Lan, 2006). where E0 and Ec are the initial tangential elastic modulus and the
peak secant modulus of the concrete, respectively. In addition, the
3.5. Predicted tensile stress-strain relations dimensionless parameter, a, is the initial tangent slope of the
stress-strain curve, whose numerical value corresponds to the
To further research the mechanical properties of this new ma- value of the initial elastic modulus of the test specimen. A relatively
terial or for its application in practical engineering, it is necessary to low a value is associated with the plastic strain (εp) accounting for a
analyse the expression for the axial tensile stress-strain curve of the relatively small proportion of εt. The a value is also related to the
SFRAC. In this study, based on Guo’s analysis (Guo, 2004), a elastic modulus. A large a value is associated with large elastic
piecewise analytical expression is used for predicting the stress- modulus changes in the tension loading process of the concrete.
strain curves of the SFRAC. This expression is derived based on Furthermore, the parameter, at, represents the downtrend in the
the normal concrete, and it introduces the RAR and the Vf param- descending portion of the stress-strain curve. A high value of at is
eters. The normalised axial tensile stress-strain relation of the correlated with a steep descending branch of the curve and poor
SFRAC can be approximated using the following equation: ductility of the tested concrete specimen. By performing regression
8 analysis on the measured data, the parameters, a and at, can be
6
>
< a,x  ða  1Þ,x ð0 < x  1Þ determined using the following formulas.
y¼ x ; (8)  
>
: ðx > 1 Þ
at ðx  1Þ1:7 þ x a ¼ 1:2
0:06 rg  0:047lf þ 1 (12)

ε s  
x ¼ y¼ ; (9) q ¼ exp 0:877lf þ 1:325 rg  5:726 (13)
εt st
Figs. 11e14 are a comparison of the predicted and experimental
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 9

Fig. 8. The influence of Vf on the stress-strain curves. (a) RAR ¼ 0%, (b) RAR ¼ 50%, (c) RAR ¼ 75%, and (d) RAR ¼ 100%.

curves, and the predicted curves match well with the curves from
Table 8 the practical data.
Tensile strengths of all the concrete specimens.

Specimen ID ft (MPa) Mean value (MPa) RSD (%) 3.6. Energy absorption analysis
No. 1 No. 2 No. 3
The toughness of a material reflects its ability to absorb energy
R0%-St0.0% 3.12 3.14 2.99 3.08 2.64
during deformation and fracture, such that its value is equal to the
R0%-St0.6% 3.77 3.49 3.50 3.59 4.43
R0%-St1.2% 4.10 4.09 4.06 4.08 0.51 deformation energy per unit volume. The better the toughness of
R0%-St1.8% 4.36 4.33 4.34 4.34 0.35 the material, the less prone it is to brittle fracture. By integrating
R50%-St0.0% 2.63 2.70 2.73 2.69 1.91 the stress-strain curves of the test specimens, the areas below the
R50%-St0.6% 3.01 2.99 2.92 2.97 1.59 curves were used to express the energy absorption when they were
R50%-St1.2% 3.47 3.71 3.30 3.49 5.90
broken. Subsequently, Eb, Ef, and Ej represent the pre-peak absorbed
R50%-St1.8% 3.76 3.96 3.88 3.87 2.60
R75%-St0.0% 2.21 2.14 2.17 2.17 1.62 energy, post-peak absorbed energy, and total absorbed energy,
R75%-St0.6% 2.45 2.68 2.55 2.56 4.50 respectively, as listed in Table 12.
R75%-St1.2% 3.11 3.18 3.41 3.23 4.85 Fig. 15(a) reveals that without the addition of steel fibres, Eb
R75%-St1.8% 3.23 3.13 3.39 3.25 4.04
decreases by 6.32%, 35.33%, and 49.53% as RAR increases. Notably, in
R100%-St0.0% 1.99 1.83 2.04 1.95 5.62
R100%-St0.6% 2.41 2.53 2.46 2.47 2.44 Fig. 15(b), SFRAC has considerable Ef which is not possessed by RAC.
R100%-St1.2% 2.93 3.12 3.01 3.02 3.16 The deformed steel fibres can absorb more energy during the
R100%-St1.8% 3.00 3.16 2.96 3.04 3.48 tension process, particularly the steel fibres that are broken
through the concrete crack. With an increase in the Vf, the Ej values
of SFRAC are 7.16e13.31 times, 6.04e9.39 times, and 5.71e8.39
times higher than those of RAC when RAR is 50%, 75%, and 100%,
Table 9 respectively. From Fig. 15(c), the Ej values of SFRAC are dramatically
Tests of between-subjects effects. higher than those of RAC, which also reflect that the toughness of
Source Degree of freedom Mean Square F value Sig. SFRAC is superior to that of RAC.
In summary, the mechanical properties of RAC are limited,
RAR 3 3.183 279.357 0.000
Vf 3 3.343 293.400 0.000 mainly due to its internal defects caused by the use of RCA. The low
compressive strength and tensile strength of RAC, particularly in
the brittle failure mode, make structure members much vulnerable
under the complicated stress environment, which significantly
10 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Table 10
Peak strains of all the concrete specimens.

Specimen ID εt (mε) Mean value (mε) RSD (%)

No. 1 No. 2 No. 3

R0%-St0.0% 103.20 96.11 101.65 100.32 3.72

R0%-St0.6% 143.40 151.34 144.13 146.29 3.00
R0%-St1.2% 164.89 166.05 157.59 162.84 2.82
R0%-St1.8% 172.85 181.08 174.24 176.06 2.50
R50%-St0.0% 121.80 111.18 108.46 113.81 6.19
R50%-St0.6% 124.61 121.80 128.13 124.85 2.54
R50%-St1.2% 157.12 145.24 150.15 150.84 3.96
R50%-St1.8% 167.69 151.60 156.64 158.63 5.19
R75%-St0.0% 98.27 95.96 92.33 95.52 3.13
R75%-St0.6% 120.27 105.62 110.13 112.01 6.70
R75%-St1.2% 139.75 132.46 130.61 134.27 3.60
R75%-St1.8% 132.23 139.52 129.45 133.73 3.89
R100%-St0.0% 87.53 93.31 92.09 90.98 3.35
R100%-St0.6% 111.31 114.56 103.03 109.63 5.42
R100%-St1.2% 122.32 133.76 118.31 124.80 6.42
R100%-St1.8% 127.36 136.02 140.64 134.68 5.01

Table 11
Tests of between-subjects effects.

Source Degree of freedom Mean Square F value Sig.

RAR 3 2655.573 83.850 0.000

Vf 3 6162.692 194.588 0.000

3.7. Feasibility analysis

However, the multi-recycling of concrete is an important

problem, particularly as these recycled concrete structures ought to
Fig. 9. The influence of different RARs and Vf on the tensile strength and peak strain.
be demolished after their service life has ended. Thus, the recy-
(a) Experimental values of the tensile strength and (b) experimental values of the peak clability of SFRAC is an essential factor that affects its scope of
strain. application in practical engineering. Therefore, some preliminary
attempts have been made by several researchers to address this
crux. Although the RCA properties gradually deteriorate as recy-
affects the safety of the engineering structure. Compared with RAC, cling cycles increase, third-generation RACs can reach the target
SFRAC can be used as a load-bearing structural material, and it strength (40 MPa) at 28 d and satisfy the requirement for at least
reduces the cross-sectional area of load-bearing members due to its 500 freeze-thaw cycles. Based on the test results of chloride ion
relatively high compressive strength. Moreover, SFRAC combines permeability coefficients and carbonation depth of RAC, the dura-
the outstanding properties of ductility and toughness, which can bility of the third-generation RACs could fulfil the designed life
strengthen the anti-impact and anti-seismic capacity of the struc- requirements for at least 50 years under harsh and cold environ-
ture and reduce the damage caused by the worst-load. Further- mental conditions (Zhu et al., 2019). Additionally, the physical and
more, steel fibres retard the expansion of cracks in members of the mechanical properties of RCAs tend to stabilise with the number of
structure, reducing external erosion and enhancing the durability cycles. The maximum loss that will be registered for concrete made
of the whole structure. Thus, the safety and service life of the RAC with recycled aggregates that originate from concretes with the
structure will be effectively improved. same composition as the one being produced is 12.9% and 19.4% for

Fig. 10. Comparisons of the predicted and experimental values. (a) Axial tensile strength and (b) peak strain.
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 11

Fig. 11. Comparison of the predicted and experimental curves when Vf ¼ 0.0%. (a) RAR ¼ 0%, (b) RAR ¼ 50%, (c) RAR ¼ 75%, and (d) RAR ¼ 100%.

Fig. 12. Comparison of the predicted and experimental curves when Vf ¼ 0.6%. (a) RAR ¼ 0%, (b) RAR ¼ 50%, (c) RAR ¼ 75%, and (d) RAR ¼ 100%.
12 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Fig. 13. Comparison of the predicted and experimental curves when Vf ¼ 1.2%. (a) RAR ¼ 0%, (b) RAR ¼ 50%, (c) RAR ¼ 75%, and (d) RAR ¼ 100%.

Fig. 14. Comparison of the predicted and experimental curves when Vf ¼ 1.8%. (a) RAR ¼ 0%, (b) RAR ¼ 50%, (c) RAR ¼ 75%, and (d) RAR ¼ 100%.
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 13

Table 12
Absorbed energy of concrete specimens.

Specimens ID Eb (kJ/m3) Ef (kJ/m3) Ej (kJ/m3)

No. 1 No. 2 No. 3 No. 1 No. 2 No. 3 No. 1 No. 2 No. 3

R0%-St0.0% 187.9 173.8 179.1 e e e 187.9 173.8 179.1

R0%-St0.6% 308.0 301.8 281.6 1245.1 1079.1 1120.6 1553.1 1380.9 1402.2
R0%-St1.2% 388.9 400.6 367.0 1398.$3 1491.6 1647.5 1787.2 1892.2 2014.5
R0%-St1.8% 427.5 458.5 379.5 2343.4 2426.1 2281.3 2770.9 2884.6 2660.8
R50%-St0.0% 162.5 178.6 165.5 e e e 162.5 178.6 165.5
R50%-St0.6% 210.5 212.5 210.2 1081.1 991.4 920.0 1291.6 1203.9 1130.2
R50%-St1.2% 335.3 334.2 324.7 1460.2 1701.1 1379.9 1795.5 2035.3 1704.6
R50%-St1.8% 345.3 328.2 334.4 1835.9 2114.1 1783.6 2181.2 2442.3 2118.0
R75%-St0.0% 122.4 112.7 114.6 e e e 122.4 112.7 114.6
R75%-St0.6% 139.9 128.5 133.2 853.0 782.6 1022.4 992.9 911.1 1155.6
R75%-St1.2% 268.6 230.3 252.6 1176.8 1001.9 1125.9 1445.4 1232.2 1378.5
R75%-St1.8% 249.3 252.2 258.0 1328.8 1132.9 1533.9 1578.1 1385.1 1791.9
R100%-St0.0% 87.5 92.1 93.3 e e e 87.5 92.1 93.3
R100%-St0.6% 151.1 161.0 141.8 718.9 786.8 935.6 870.0 947.8 1077.4
R100%-St1.2% 206.8 224.4 195.6 944.8 1427.6 1250.7 1151.6 1652.0 1446.3
R100%-St1.8% 210.5 234.6 247.5 1044.6 936.0 1098.3 1255.1 1170.6 1345.8

Fig. 15. Energy consumption of the test specimens. (a) Pre-peak absorbed energy, (b) post-peak absorbed energy, and (c) total absorbed energy.

compressive and tensile strengths, respectively (Abreu et al., 2018). through the interview and compared with Construction machinery
The structural use of multi-recycled RCA from precast rejected fee quota in Hubei (Hubei Provincial Construction Engineering
pieces have produced good outcomes, and its negative aspect can Standard Quota Management General Station, 2018) and Prices of
be counteracted by a reduction in the density of the concrete raw materials in the Hubei Province (Huber Provincial Development
(Salesa et al., 2017). Furthermore, in-depth interviews were con- and Reform Commission, 2020) to verify the authenticity.
ducted with the staff of two CDW-recycling companies’s. From the Taking the output of 5000 tons of RCA as an example, Table 13
interview, it was found that it is feasible to recycle SFRAC by sorting, lists the main process and the corresponding detailed cost data.
mechanical crushing, screening, and magnetic separation and that In the final production stage, finished products of 5e35 mm ag-
the total mass of RCAs, which is 55e70% the weight of the waste gregates are produced at a price of approximately ¥47.24 t-1 and
concrete, can be reused at least twice. This is related to the sand rate sold at a price of ¥65 t1, while the market price of NCA is ¥123 t1.
of the original concrete and physical properties of coarse aggre- Unquestionably, using RCA produced by recycling methods is more
gates. The smaller the sand rate, the larger the particle size. The economical in the long-term than using NCA. Additionally, the
higher the hardness, the greater the RCAs obtained from the economic feasibility of the proposed recycling process is analysed
recycling process. Furthermore, some valuable data were obtained from an industrial point of view, as shown in Tables 14 and 15. In
14 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

Table 13
Production method and cost analysis for 5000 tons of RCAs.

Process Cost (CNY)

Sorting process
Working capitala 2380.0
Fuel 248.3
Labour 2840.0
Crushing processb
Working capitala 3920.0
Fuel 2880.0
Labour 1420.0
Screening
Working capitala 8339.0
Fuel 562.5
Labour 7100.0
Magnetic separation
Working capitala 2352.0
Fuel 2760.0
Labour 1420.0
Disposing residues 25000.0
Others
The charge of dumping, stockpiling, transportation, etc. 175000.0
Total 236221.8
a
Working capital includes depreciation cost, overhaul cost, maintenance cost, and dismantling cost.
b
Crushing process includes primary crushing and secondary crushing.

Table 14
Cost analysis for the proposed recycling method.

Project Amount (CNY)

Equipment
Excavator equipment 2,600,000
Crusher equipment 4,500,000
Magnetic separator equipment 1,200,000
Screener equipment 2,150,000
Land 1,000,000
Construction engineering 2,000,000
Installation engineering 1,000,000
Electromechanical engineering 700,000
Circulating capital 1,500,000
Others 300,000
Total 16,950,000

Table 15 favourable tensile properties. This study explores the stress-strain

Cost and benefit analysis for the CDW-recycling company. behaviour of the concrete under axial tensile loading, including
Categories Amount (CNY) Note the effects of RAR and the Vf. Moreover, the economic realizability
and feasibility of the proposed recycling process were analysed
Sales 58,500,000 About 900,000 t/year of ¥65.00 t-1
Cost 42,516,000 About 900,000 t/year of ¥47.24 t-1
from an industrial point of view. Based on the extensive test data
Enterprise management 1,755,000 3% of sales per year and analyses, the following conclusions can be drawn:
Staff salaries 6,435,000 11% of sales per year
Miscellaneous expenses 1,170,000 2% of sales per year 1. Adding steel fibres into RAC to form SFRAC is convenient and
Tax 4,095,000 7% of sales per year
effective. Compared to RAC, the mechanical properties of SFRAC
Profits 2,529,000
satisfy the requirements of practical engineering, and the safety
and service life of the RAC structure is effectively improved,
which would consequently improve the recycling and reuse rate
general, the annual profit margin of the CDW-recycling company is of CDW.
14.92%, and the investment recovery period is 6.7 years. Consid- 2. Steel fibres in the concrete matrix play a pivotal role in the
ering the period of engineering construction, equipment installa- bridging and anchoring effect to retard crack propagation,
tion, and trial production, the actual investment recovery period is which changes the brittle failure mode of RAC. Therefore, the
approximately 7.3 years. The cost expended for concrete recycling stress-strain curve of SFRAC has a visible descending section,
can be compensated by the high demand for RCA. Therefore, from showing favourable ductility characteristics. Additionally, steel
an industrial point of view, the proposed recycling process is a fibres are mainly pulled out rather than broken in the concrete
feasible and cost-effective method for recycling SFRAC, which also during the tensile process.
helps to protect the environment and achieve construction 3. Both RAR and Vf have a significant impact on the mechanical
sustainability. properties of the concrete, and the internal defect of the con-
crete due to the use of RCA can be compensated by adding steel
4. Conclusions fibres. As the Vf increases from 0% to 1.8%, the tensile strength
and peak strain increases by 10.41e55.90% and 9.70e48.03%
SFRAC is a sustainable and eco-friendly composite material with
 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248 15

compared to those of RAC, respectively. Furthermore, SFRAC has coarse aggregate on properties of concrete. Mater. Struct. Constr. 44, 205e220.
https://doi.org/10.1617/s11527-010-9620-x.
an energy absorption capacity that is 5.71e13.31 times higher
Corinaldesi, V., 2010. Mechanical and elastic behaviour of concretes made of
than that of RAC. recycled-concrete coarse aggregates. Construct. Build. Mater. 24, 1616e1620.
4. The analytical expression formulas for the introduced parame- https://doi.org/10.1016/j.conbuildmat.2010.02.031.
ters, RAR and Vf, of ordinary concrete, are extended to for SFRAC Corinaldesi, V., Moriconi, G., 2009. Influence of mineral additions on the perfor-
mance of 100% recycled aggregate concrete. Construct. Build. Mater. 23,
aimed at predicting its tensile strength, peak strain, and stress- 2869e2876. https://doi.org/10.1016/j.conbuildmat.2009.02.004.
strain curves. The predictions agree well with the experimental Domingo-Cabo, A., La zaro, C., Lo  pez-Gayarre, F., Serrano-Lo  pez, M.A., Serna, P.,
results, suggesting that the SFRAC can be used directly in Castan ~ o-Tabares, J.O., 2009. Creep and shrinkage of recycled aggregate concrete.
Construct. Build. Mater. 23, 2545e2553. https://doi.org/10.1016/
theoretical research and practical engineering. j.conbuildmat.2009.02.018.
5. SFRAC is recyclable, and the total mass of RCAs, which is the EPA, 2017. Construction and demolition: material-specific data | facts and figures
55e70% weight of the waste concrete, can be reused at least about materials, waste and recycling | US EPA [WWW Document]. URL. https://
www.epa.gov/facts-and-figures-about-materials-waste-and-recycling/
twice. From an industrial point of view, the proposed method for construction-and-demolition-material-specific. accessed 4.2.20.
recycling SFRAC is feasible and cost-effective, and the invest- Etxeberria, M., V azquez, E., Marí, A., Barra, M., 2007. Influence of amount of recycled
ment recovery period is approximately 7.3 years. coarse aggregates and production process on properties of recycled aggregate
concrete. Cement Concr. Res. 37, 735e742. https://doi.org/10.1016/
j.cemconres.2007.02.002.
This study selected the concrete strength and fibre type that are Eurostat, 2018. Waste statistics. WWW Document. https://ec.europa.eu/eurostat/
commonly used in practical engineering. Thus, the influence of tgm/refreshTableAction.do?tab¼table&amp;plugin¼1&amp;
pcode¼ten00106&amp;language¼en. accessed 4.2.20.
fibre material, type, and shape on tensile behaviour of SFRAC needs
Evangelista, L., de Brito, J., 2010. Durability performance of concrete made with fine
to be solved quantitatively based on more solid databases. In the recycled concrete aggregates. Cement Concr. Compos. 32, 9e14. https://doi.org/
feasibility analysis section, the local supportive policies for CDW- 10.1016/j.cemconcomp.2009.09.005.
recycling company were not described in detail, which could Fathifazl, G., Ghani Razaqpur, A., Burkan Isgor, O., Abbas, A., Fournier, B., Foo, S.,
2011. Creep and drying shrinkage characteristics of concrete produced with
shorten the investment recovery period by reducing the cost of the coarse recycled concrete aggregate. Cement Concr. Compos. 33, 1026e1037.
recycling process. Additionally, more experimental works and https://doi.org/10.1016/j.cemconcomp.2011.08.004.
further numerical analyses are underway to explore the stress- FIB, 2013. Materials. In: Fib Model Code for Concrete Structures 2010, Wiley Online
Books. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, pp. 74e150.
strain response of SFRAC under complex stress state, such as https://doi.org/10.1002/9783433604090.ch5.
biaxial or triaxial stress state. Expanding the application scope of GB/T 14685-2011, 2011. Pebble and Crushed Stone for Construction.
SFRAC from a structural perspective is worth attempting in the GB/T 25177-2010, 2010. Recycled Coarse Aggregate for Concrete.
GB/T 50080-2016, 2016. Standard for Test Method of Performance on Ordinary
future. Concrete.
GB/T 50081-2016, 2016. Standard for Test Method of Mechanical Properties on
Funding sources Ordinary Concrete.
Guo, H., Shi, C., Guan, X., Zhu, J., Ding, Y., Ling, T.C., Zhang, H., Wang, Y., 2018.
Durability of recycled aggregate concrete e a review. Cement Concr. Compos.
Funding: This work was supported by the Special Project on 89, 251e259. https://doi.org/10.1016/j.cemconcomp.2018.03.008.
Technical Innovation of Hubei [Grant Nos. 2019ACA142]; the Guo, Y., Zhang, J., Chen, G., Xie, Z., 2014. Compressive behaviour of concrete
structures incorporating recycled concrete aggregates, rubber crumb and
[Special Project on Application Fundamental Frontier Research of reinforced with steel fibre, subjected to elevated temperatures. J. Clean. Prod.
Wuhan [Grant Nos. 2019020701011465]; and the Project funded by 72, 193e203. https://doi.org/10.1016/j.jclepro.2014.02.036.
China Postdoctoral Science FoundationGrant Nos. 2019M662712]. Guo, Z., 2004. The Strength and Constitutive Relationship of Concrete: Principle and
Application. Tsinghua University Press, Beijing in Chinese).
The authors greatly appreciate their financial support. Han, R., Zhao, S., Qu, F., 2006. Experimental study on the tensile performance of
steel flber reinforced concrete. China Civ. Eng. J. 39, 63e67. https://doi.org/
10.15951/j.tmgcxb.2006.11.010.
CRediT authorship contribution statement
Hansen, T.C., 1986. Recycled aggregates and recycled aggregate concrete second
state-of-the-art report developments 1945-1985. Mater. Struct. 19, 201e246.
Shuai Zong: Writing - original draft, preparation, Formal anal- https://doi.org/10.1007/BF02472036.
ysis, Investigation. Zhenzhen Liu: Writing - review & editing, Huang, B., Wang, X., Kua, H., Geng, Y., Bleischwitz, R., Ren, J., 2018. Construction and
demolition waste management in China through the 3R principle. Resour.
Methodology. Shan Li: Methodology. Yiyan Lu: Conceptualization, Conserv. Recycl. 129, 36e44. https://doi.org/10.1016/j.resconrec.2017.09.029.
Supervision. Aohan Zheng: Software, Proofreading. Hubei Provincial Construction Engineering Standard Quota Management General
Station, 2018. Construction machinery fee quota in Hubei Province. WWW
Document. http://www.hbcic.net.cn/wcm.files/upload/CMShbcic/201905/
Declaration of competing interest 201905301146058.pdf. accessed 4.22.20.
Huber Provincial Development and Reform Commission, 2020. Prices of raw ma-
terials in the Hubei Province. WWW Document. http://fgw.hubei.gov.cn/fbjd/
The authors declare that they have no known competing
xxgkml/jgzn/zsdw/jgjczx/sssnjg/202001/t20200117_1951147.shtml. accessed
financial interests or personal relationships that could have 4.22.20.
appeared to influence the work reported in this paper. Jamkar, S.S., Rao, C.B.K., 2004. Index of Aggregate Particle Shape and Texture of
coarse aggregate as a parameter for concrete mix proportioning. Cement Concr.
Res. 34, 2021e2027. https://doi.org/10.1016/j.cemconres.2004.03.010.
References Le Hoang, A., Fehling, E., 2017. Influence of steel fiber content and aspect ratio on
the uniaxial tensile and compressive behavior of ultra high performance con-
Abreu, V., Evangelista, L., de Brito, J., 2018. The effect of multi-recycling on the crete. Construct. Build. Mater. 153, 790e806. https://doi.org/10.1016/
mechanical performance of coarse recycled aggregates concrete. Construct. j.conbuildmat.2017.07.130.
Build. Mater. 188, 480e489. https://doi.org/10.1016/j.conbuildmat.2018.07.178. Liu, Zhenzhen, Lu, Yiyan, Liao, Jiancong, 2019. Axial behavior of slender steel tube
Beglarigale, A., Yazıcı, H., 2015. Pull-out behavior of steel fiber embedded in flow- filled with steel-fiber-reinforced recycled aggregate concrete. Journal of
able RPC and ordinary mortar. Construct. Build. Mater. 75, 255e265. https:// Constructional Steel Research 162. https://doi.org/10.1016/j.jcsr.2019.105748.
doi.org/10.1016/j.conbuildmat.2014.11.037. pez Ruiz, L.A., Roca Ramo
Lo n, X., Gasso Domingo, S., 2020. The circular economy in
Bogas, J.A., De Brito, J., Ramos, D., 2016. Freeze-thaw resistance of concrete pro- the construction and demolition waste sector e a review and an integrative
duced with fine recycled concrete aggregates. J. Clean. Prod. 115, 294e306. model approach. J. Clean. Prod. 248 https://doi.org/10.1016/
https://doi.org/10.1016/j.jclepro.2015.12.065. j.jclepro.2019.119238.
Carneiro, J.A., Lima, P.R.L., Leite, M.B., Toledo Filho, R.D., 2014. Compressive stress- Malesev, M., Radonjanin, V., Marinkovi c, S., 2010. Recycled concrete as aggregate for
strain behavior of steel fiber reinforced-recycled aggregate concrete. Cement structural concrete production. Sustainability 2, 1204e1225. https://doi.org/
Concr. Compos. 46, 65e72. https://doi.org/10.1016/j.cemconcomp.2013.11.006. 10.3390/su2051204.
CECS13:2009, 2009. Standard Test Methods for Fiber Reinforced Concrete. Chinese Mohseni, E., Saadati, R., Kordbacheh, N., Parpinchi, Z.S., Tang, W., 2017. Engineering
Planning Press. and microstructural assessment of fibre-reinforced self-compacting concrete
Chakradhara Rao, M., Bhattacharyya, S.K., Barai, S.V., 2011. Influence of field recycled containing recycled coarse aggregate. J. Clean. Prod. 168, 605e613. https://
16 S. Zong et al. / Journal of Cleaner Production 278 (2021) 123248

doi.org/10.1016/j.jclepro.2017.09.070. Xie, J., Huang, L., Guo, Y., Li, Z., Fang, C., Li, L., Wang, J., 2018b. Experimental study on
Nagataki, S., Gokce, A., Saeki, T., Hisada, M., 2004. Assessment of recycling process the compressive and flexural behaviour of recycled aggregate concrete modi-
induced damage sensitivity of recycled concrete aggregates. Cement Concr. Res. fied with silica fume and fibres. Construct. Build. Mater. 178, 612e623. https://
34, 965e971. https://doi.org/10.1016/j.cemconres.2003.11.008. doi.org/10.1016/j.conbuildmat.2018.05.136.
Nixon, P.J., 1978. Recycled concrete as an aggregate for concrete-a review. Mater. Xie, J.H., Guo, Y.C., Liu, L.S., Xie, Z.H., 2015. Compressive and flexural behaviours of a
Construccio n 11, 371e378. https://doi.org/10.1007/BF02473878. new steel-fibre-reinforced recycled aggregate concrete with crumb rubber.
Okeh, C.A.O., Begg, D.W., Barnett, S.J., Nanos, N., 2019. Behaviour of hybrid steel fibre Construct. Build. Mater. 79, 263e272. https://doi.org/10.1016/
reinforced self compacting concrete using innovative hooked-end steel fibres j.conbuildmat.2015.01.036.
under tensile stress. Construct. Build. Mater. 202, 753e761. https://doi.org/ Yang, K.H., Chung, H.S., Ashour, A.F., 2008. Influence of type and replacement level
10.1016/j.conbuildmat.2018.12.067. of recycled aggregates on concrete properties. ACI Mater. J. 105, 289e296.
Omary, S., Ghorbel, E., Wardeh, G., 2016. Relationships between recycled concrete https://doi.org/10.14359/19826.
aggregates characteristics and recycled aggregates concretes properties. Yildirim, S.T., Meyer, C., Herfellner, S., 2015. Effects of internal curing on the
Construct. Build. Mater. 108, 163e174. https://doi.org/10.1016/ strength, drying shrinkage and freeze-thaw resistance of concrete containing
j.conbuildmat.2016.01.042. recycled concrete aggregates. Construct. Build. Mater. 91, 288e296. https://
Padmini, A.K., Ramamurthy, K., Mathews, M.S., 2009. Influence of parent concrete doi.org/10.1016/j.conbuildmat.2015.05.045.
on the properties of recycled aggregate concrete. Construct. Build. Mater. 23, Zaharieva, R., Buyle-Bodin, F., Wirquin, E., 2004. Frost resistance of recycled
829e836. https://doi.org/10.1016/j.conbuildmat.2008.03.006. aggregate concrete. Cement Concr. Res. 34, 1927e1932. https://doi.org/10.1016/
Poon, C.S., Shui, Z.H., Lam, L., Fok, H., Kou, S.C., 2004. Influence of moisture states of j.cemconres.2004.02.025.
natural and recycled aggregates on the slump and compressive strength of Zhang, D., 2017. Report on the inspection of the implementation of the" law of the
concrete. Cement Concr. Res. 34, 31e36. https://doi.org/10.1016/S0008- people’s Republic of China on the prevention and control of environmental
8846(03)00186-8. pollution by solid wastes. WWW Document. http://npc.people.com.cn/n1/2017/
S
aez del Bosque, I.F., Zhu, W., Howind, T., Matías, A., S anchez de Rojas, M.I., 1102/c14576-29622406.html. accessed 4.18.20.
Medina, C., 2017. Properties of interfacial transition zones (ITZs) in concrete Zhang, L., Gao, D., Zhu, H., Lou, Z., 2013. Experimental research of the splitting
containing recycled mixed aggregate. Cement Concr. Compos. 81, 25e34. tensile strength of steel fiber reinforced recycled concrete. J. North China Inst.
https://doi.org/10.1016/j.cemconcomp.2017.04.011. Water Conserv. Hydroelectr. Power 34, 27e31. https://doi.org/10.3969/
Sagoe-Crentsil, K.K., Brown, T., Taylor, A.H., 2001. Performance of concrete made j.issn.1002-5634.2013.01.007.
with commercially produced coarse recycled concrete aggregate. Cement Concr. Zhao, G., 1999. Steel Fiber Reinforced Concrete Structure. China Architecture &
Res. 31, 707e712. https://doi.org/10.1016/S0008-8846(00)00476-2. Building Press, Beijing.
 Pe
Salesa, A.,  Esteban, L.M., Vicente-Vas, R., Orna-Carmona, M.,
rez-Benedicto, J.A., Zhou, C., Chen, Z., 2017. Mechanical properties of recycled concrete made with
2017. Physico-mechanical properties of multi-recycled self-compacting con- different types of coarse aggregate. Construct. Build. Mater. https://doi.org/
crete prepared with precast concrete rejects. Construct. Build. Mater. 153, 10.1016/j.conbuildmat.2016.12.163.
364e373. https://doi.org/10.1016/j.conbuildmat.2017.07.087. Zhu, P., Hao, Y., Liu, H., Wei, D., Liu, S., Gu, L., 2019. Durability evaluation of three
Senaratne, S., Gerace, D., Mirza, O., Tam, V.W.Y., Kang, W.H., 2016. The costs and generations of 100% repeatedly recycled coarse aggregate concrete. Construct.
benefits of combining recycled aggregate with steel fibres as a sustainable, Build. Mater. 210, 442e450. https://doi.org/10.1016/j.conbuildmat.2019.03.203.
structural material. J. Clean. Prod. 112, 2318e2327. https://doi.org/10.1016/
j.jclepro.2015.10.041.
Si, R., Wang, J., Guo, S., Dai, Q., Han, S., 2018. Evaluation of laboratory performance of
Glossary
self-consolidating concrete with recycled tire rubber. J. Clean. Prod. 180,
823e831. https://doi.org/10.1016/j.jclepro.2018.01.180. NAC: Natural coarse aggregate concrete
Silva, R.V., De Brito, J., Dhir, R.K., 2015. Prediction of the shrinkage behavior of RAC: Recycled coarse aggregate concrete
recycled aggregate concrete: a review. Construct. Build. Mater. 77, 327e339. NCA: Natural coarse aggregate
https://doi.org/10.1016/j.conbuildmat.2014.12.102. RCA: Recycled coarse aggregate
Tam, V.W.Y., Kotrayothar, D., Xiao, J., 2015. Long-term deformation behaviour of SFNAC: Steel-fibre-reinforced NAC
recycled aggregate concrete. Construct. Build. Mater. 100, 262e272. https:// SFRAC: Steel-fibre-reinforced RAC
doi.org/10.1016/j.conbuildmat.2015.10.013. CDW: Construction and demolition waste
Thomas, C., Setie n, J., Polanco, J.A., Alaejos, P., Sa
nchez De Juan, M., 2013. Durability ITZs: Interfacial transition zones
of recycled aggregate concrete. Construct. Build. Mater. 40, 1054e1065. https:// P·O: Ordinary Portland cement
doi.org/10.1016/j.conbuildmat.2012.11.106. FA: Fly ash
Wang, W.L., Kou, S.C., Xing, F., 2013. Deformation properties and direct shear of LVDT: Linear variable differential transformer
medium strength concrete prepared with 100% recycled coarse aggregates. RSD: Relative standard deviation
Construct. Build. Mater. 48, 187e193. https://doi.org/10.1016/ RAR: Replacement rates of RCA
j.conbuildmat.2013.06.065. Vf: Volume percentage of the steel fibre
Wang, Y., Hughes, P., Niu, H., Fan, Y., 2019. A new method to improve the properties lf: Characteristic value of steel fibre
of recycled aggregate concrete: composite addition of basalt fiber and nano- lf: Steel fibre length
silica. J. Clean. Prod. 236, 117602 https://doi.org/10.1016/j.jclepro.2019.07.077. df: Steel fibre diameter
Wille, K., Naaman, A.E., 2012. Pullout behavior of high-strength steel fibers fcu: Compressive strength of concrete
embedded in ultra-high-performance concrete. ACI Mater. J. 109, 479e487. ft: Tensile strength of concrete
Xiao, J., 2018. Strength of recycled aggregate concrete. In: Springer Tracts in Civil e: Eccentricity
Engineering. Springer, pp. 143e165. https://doi.org/10.1007/978-3-662-53987- s: Stress
3_6. st: Peak stress
Xiao, J., Li, J., Zhang, C., 2005. Mechanical properties of recycled aggregate concrete ε: Strain
under uniaxial loading. Cement Concr. Res. 35, 1187e1194. https://doi.org/ εt: Peak strain
10.1016/j.cemconres.2004.09.020. εp: Plastic strain
Xiao, J.Z., Lan, Y., 2006. Investigation on the tensile behavior of recycled aggregate Eb: Pre-peak absorbed energy
concrete. J. Build. Mater. 9, 154e158. Ef: Post-peak absorbed energy
Xie, J., Fang, C., Lu, Z., Li, Z., Li, L., 2018a. Effects of the addition of silica fume and Ej: Total absorbed energy
rubber particles on the compressive behaviour of recycled aggregate concrete E0: Initial tangential elastic modulus
with steel fibres. J. Clean. Prod. 197, 656e667. https://doi.org/10.1016/ Ec: Peak secant modulus
j.jclepro.2018.06.237.