sensors

Article
Investigation of Bonding Behavior of FRP and Steel
Bars in Self-Compacting Concrete Structures Using
Acoustic Emission Method
Bo Di 1 , Jingkai Wang 2 , Haotian Li 1,3 , Jinhang Zheng 1 , Yu Zheng 1, * and Gangbing Song 4
1 School of Environment and Civil Engineering, Dongguan University of Technology, Dongguan 523808,
China; dibo@dgut.edu.cn (B.D.); 2172332355@email.szu.edu.cn (H.L.); 201641201307@dgut.edu.cn (J.Z.)
2 State Key Laboratory of Coastal and Offshore Engineering, Dalian University of Technology,
Dalian 116024, China; wjk@mail.dlut.edu.cn
3 College of Civil Engineering Shenzhen University, Shenzhen University, Shenzhen 518060, China
4 Department of Mechanical Engineering, University of Houston, Houston, TX 77204-4006, USA;
gsong@uh.edu
* Correspondence: zhengy@dgut.edu.cn; Tel.: +86-158-9966-2977




Received: 8 December 2018; Accepted: 24 December 2018; Published: 4 January 2019


Abstract: To extend understanding of the bonding behavior of fiber reinforced polymer (FRP) and
steel bars in self-compacting concrete (SCC), an experimental series consisting of 36 direct pull-out
tests monitored by acoustic emission (AE) were performed in this paper. The test variables involved
rebar type, bar diameter, embedded length, and polypropylene (PP) fiber volume content. For
each test, the pull-out force and free end slip were continuously measured and compared with the
corresponding AE signals. It was found that the proposed AE method was effective in detecting the
debonding process between the FRP/steel bars and the hosting concrete. The AE signal strength
exhibited a good correlation with the actual bond stress-slip relationship measured in each specimen.
Based on the AE location technique, the invisible non-uniform distribution of bonding stress along
the bar was further revealed, the initial location of damage and the debonding process were captured.
Additionally, the contribution of bar-to-concrete load-bearing mechanism (chemical adhesion, friction,
and mechanical interlocking) to sustain the pull-out force was effectively clarified by studying the
collected signals in the frequency domain of AE methods. The experimental results demonstrate that
the proposed AE method has potential to detect the debonding damage of FRP/steel bar reinforced
SCC structures accurately.

Keywords: acoustic emission; FRP; self-compacting concrete; bonding; pull-out test

1. Introduction
High-performance and environment-friendly materials are currently attracting a widespread
attention in civil engineering. High strength, light weight, long-term durability, and low maintenance
costs are some remarkable advantages of these new materials compared to traditional ones.
Self-compacting concrete (SCC) is a new type of concrete material, which owes excellent workability
and high resistance to segregation. Due to its high degree of flowability, no vibration equipment
is needed for the compacting procedure. This material is a technically viable substitute for normal
concrete (NC) and has gained wide use in different applications [1,2]. Currently, fiber reinforced
polymer (FRP) has been recognized as a construction material [3–6] and can be used as an alternative
reinforcement for concrete structures due to its high strength, light weight, and high corrosion
resistance [7,8]. With the advantages of FRP and SCC, combining them would provide a promising
solution in the construction projects [9]. However, the application of FRP bars as internal reinforcement

Sensors 2019, 19, 159; doi:10.3390/s19010159 www.mdpi.com/journal/sensors

Sensors 2019, 19, 159 2 of 14

in SCC structures is yet in its early stages, and the mechanical properties should be further studied.
A relevant behavior problem is the bond between FRP composites and SCC, since it is a critical factor
that influences the structural performance under both service and ultimate conditions [10]. In addition,
with the recent emphasis on structural health monitoring [11–13] and damage detection [14–16], the
monitoring of bonding behavior is of great importance to enhance the safety level and to widen the
application of FRP reinforced SCC structures [17–20].
Many studies have been conducted to investigate the bonding behavior between FRP/steel bars
and normal concrete [21–31]. The key factors affecting bonding performance, such as rebar type, bar
diameter, embedment length, confinement pressure, and concrete strength, have been studied based
on either direct pull-out test or beam test [21–33]. Baena et al. [21] carried out an experimental program
consisting of 88 pull-out specimens by using carbon fiber reinforced polymer (CFRP) and glass fiber
reinforced polymer (GFRP) bars, and these bars presented a bonding strength lower than the steel
bars. Based on the results from previous studies, the bonding strengths of specimens with FRP bars
were typically lower than conventional steel bars. On the contrary, the slip of FRP bars relative to
the surrounding concrete was usually greater than that of steel bars [21–23]. Moreover, a trend of
decreasing bonding strength with increasing bar diameters can be observed in the literature [21,24–26].
The major reasons for this trend could be the Poisson effect [21], the shear lag effect [24,25], and
the size effect [26]. By installing a strain probe inside the pre-drilled FRP bar, Alzahrani et al. [27]
observed the nonlinear distribution of bond stress along the embedded length. It was found that
doubling the embedment length reduced the average bonding strength by 25%. The dependence of
bond strength on embedment length could be explained by the nonlinear stress distribution, since
the nonlinear stress distribution is more evident in the case of the larger embedment length, which
results in the lower average bonding strength [28,29]. Furthermore, some experimental and theoretical
investigations have been carried out on the bond behavior of FRP/steel bars in SCC structures [32–35].
Mazaheripour et al. [32] conducted beam tests to study the bonding performance between GFRP
bars and steel fiber reinforced self-compacting concrete. It was concluded that concrete cover and
bonding length played an important role on the bonding strength of GFRP bars. The high fiber contents
contributed to a better bonding behavior by reducing crack width in the concrete cover, which resulted
in an increase in the average residual bond stress. It should be mentioned that the most commonly used
bond stress-slip curves acquired by direct pull-out tests or bending tests are the macro-reflection of the
bond between reinforcement bars and concrete. The advanced measuring methods should be adopted
to further reveal the bar-to-concrete load-bearing mechanism and the debonding damage process.
Piezoceramic transducers have the advantages of both sensing and actuation capacities [36,37],
high bandwidth [38,39], and low cost, and have been actively studied to monitor debonding problems
in various structures [40–43], including rebar and concrete debonding [44–46]. In addition, in the form
of a smart aggregate, a piezoceramic transducer can be easily and reliably embedded in a concrete
structure [47,48], or in the form of a patch, a piezoceramic transducer can be surface-bonded on
a structure of interest [49,50], including a steel rebar or a composite rebar. Jiang et al. proposed
an innovative method using piezoceramic transducers and wavelet packet analysis to detect the
debonding between an FRP bar and the concrete structure [18]. Xu et al. developed a novel
piezoceramic-based active sensing approach to monitor the debonding between a GFRP bar and the
concrete structure, and experimental results demonstrated the effectively of the proposed approach [44].
With the help of piezoceramic transducers and the hierarchical clustering analysis, Sevillano et al.
identified interfacial crack-induced debonding in FRP reinforced concrete beams [46].
Utilizing the high bandwidth of piezoceramic transducers, the acoustic emission (AE) technique,
a passive monitoring method, is often used for damage detection, including debonding monitoring.
Compared to other sensing methods, the AE technique collects the transient elastic waves caused
by the rapid release of energy in the process of materials fracture. By analyzing these transient
elastic waves, the damage degree of materials can be investigated and evaluated. This technique has
been increasingly used in civil engineering [51,52], mechanical engineering [53,54], and others [55,56].
Sensors 2019, 19, x FOR PEER REVIEW 3 of 14
Sensors 2019, 19, 159 3 of 14

and others [55,56]. Bunnori et al. [57] applied the AE technique to detect the early cracks in concrete
beams and
Bunnori found
et al. [57] that the the
applied AE AEparameters
techniquewere sensitive
to detect the to thecracks
early initiation and the beams
in concrete growthand
of cracks.
found
Aldahdooh et al. [58] classified the types of cracks (flexural or shear cracks) of RC beams subjected
that the AE parameters were sensitive to the initiation and the growth of cracks. Aldahdooh et al. [58] to
four-point bending by AE technique. Moreover, some researches applied the AE technique
classified the types of cracks (flexural or shear cracks) of RC beams subjected to four-point bending by
to
monitor the debonding process of steel bars in concrete structures. Balazs et al. [59] applied the AE
AE technique. Moreover, some researches applied the AE technique to monitor the debonding process
technique to monitor the damage accumulation on deformed steel bar to concrete interaction and
of steel bars in concrete structures. Balazs et al. [59] applied the AE technique to monitor the damage
confirmed that the AE parameters were consistent with the test phenomena. Gallego et al. [60]
accumulation on deformed steel bar to concrete interaction and confirmed that the AE parameters
utilized the AE signals obtained from the pull-out tests to compare the performance of black steel
were consistent with the test phenomena. Gallego et al. [60] utilized the AE signals obtained from the
and hot-dip galvanized steel. It was revealed that it was possible to identify the transitional points of
pull-out tests to compare the performance of black steel and hot-dip galvanized steel. It was revealed
pullout force–slip curves by measuring the AE activity. Abouhussien et al. [61,62] found that the
that it was possible to identify the transitional points of pullout force–slip curves by measuring the AE
cumulative number of hits and cumulative signal strength were in a good relationship with the
activity. Abouhussien et al. [61,62] found that the cumulative number of hits and cumulative signal
different damage degree of debonding process, and proposed the developed intensity classification
strength were in a good relationship with the different damage degree of debonding process, and
charts based on AE intensity analysis. Furthermore, Wang et al. [63] conducted a series of pull-out
proposed the developed intensity classification charts based on AE intensity analysis. Furthermore,
tests to study the bond behavior between corroded steel bars and concrete based on the AE method.
Wang et al. [63] conducted a series of pull-out tests to study the bond behavior between corroded steel
The test results showed that the AE location technique could be used to detect the actual crack
bars and concrete based on the AE method. The test results showed that the AE location technique
development, and the characteristics of the AE signal reflected the bonding behavior of the pull-out
could be used to detect the actual crack development, and the characteristics of the AE signal reflected
specimens.
the bonding behavior of the pull-out specimens.
Nevertheless, the AE signals obtained from pull-out tests have not been utilized to evaluate the
Nevertheless, the AE signals obtained from pull-out tests have not been utilized to evaluate the
bonding behavior of FRP bars in SCC structures. This paper presents an experimental investigation
bonding behavior of FRP bars in SCC structures. This paper presents an experimental investigation
aimed at clarifying the load-bearing mechanism and debonding damage process between FRP/steel
aimed at clarifying the load-bearing mechanism and debonding damage process between FRP/steel
bars and SCC. The pull-out specimens with variable rebar types, bar diameters, embedded lengths,
bars and SCC. The pull-out specimens with variable rebar types, bar diameters, embedded lengths,
and concrete mixtures were designed and fabricated. The bonding behavior is identified in terms of
and concrete mixtures were designed and fabricated. The bonding behavior is identified in terms of
the stress-slip curves and the AE signals measured by two AE sensors attached at the end faces of
the stress-slip curves and the AE signals measured by two AE sensors attached at the end faces of the
the bar. Test results indicate that the proposed AE method can be used to detect the debonding
bar. Test results indicate that the proposed AE method can be used to detect the debonding process of
process of FRP/steel bars in SCC structures.
FRP/steel bars in SCC structures.

2. Experimental
2. Experimental Program
Program

2.1. Material Properties

2.1.1. Reinforcement Bars

Three
Three types
typesofof
reinforcing
reinforcingmaterials werewere
materials used used
in the in
study,
the which
study,included basalt fiberbasalt
which included reinforced
fiber
polymer
reinforced polymer (BFRP), GFRP, and steel bars. The surface treatment and characteristics of are
(BFRP), GFRP, and steel bars. The surface treatment and characteristics of the bars used the
shown in Figure
bars used 1. BFRP
are shown bars with
in Figure a textured
1. BFRP bars withsurface and GFRP
a textured barsand
surface with a helically
GFRP wrapped
bars with and
a helically
sand-coated
wrapped andsurface were adopted,
sand-coated respectively.
surface were adopted,On the other hand,
respectively. On the ribbed
othersteel
hand,bars weresteel
ribbed usedbars
for
comparison purposes. As shown in Table 1, two diameters, 12 and 20 mm,
were used for comparison purposes. As shown in Table 1, two diameters, 12 and 20 mm, werewere considered for BFRP
bars, while only
considered one diameter
for BFRP bars, while(12only
mm)onewasdiameter
adopted(12 formm)
GFRP andadopted
was steel bars. The geometrical
for GFRP and
and steel bars.
mechanical properties of the reinforcement bars are listed in Table 1.
The geometrical and mechanical properties of the reinforcement bars are listed in Table 1.

(a) (b) (c)

Figure 1. Reinforcement bars: (a) BFRP bar; (b) GFRP bar; (c) Steel bar.
bar.
Sensors 2019, 19, 159 4 of 14

Table 1. Properties of the reinforcement bars (Mean values).

Diameter Yield Strength Ultimate Strength Elastic Modulus Strain at Ultimate

Material Type
(mm) (N/mm2 ) (N/mm2 ) (N/mm2 ) Strength (%)
BFRP 12 / 1032 48 2.2
BFRP 20 / 900 45 2.1
GFRP 12 / 1153 52 2.0
Steel 12 487 589 210 10.0

2.1.2. Self-Compacting Concrete

In this investigation, SCC was used in the direct pull-out specimens. To examine the influence of
fiber content on the bonding behavior of FRP/steel bars, three different fiber volume contents, 0.0%,
0.3%, and 0.6%, were adopted. The concrete used for the pull-out specimens was prepared in the
laboratory and its compositions are shown in Table 2. Ordinary Portland cement labeled as 32.5R,
fine river sand and crushed gravel aggregate with maximum size of 20 mm were used. Fly ash with
fineness of 250 meshes was adopted to replace cement for up to 50% by weight to produce the SCC.
Polypropylene fibers (PP fibers) of 12 mm length and 18 um diameter were used. The SCC mixture
showed good workability and cohesion, and the total spread measured in the slump-flow tests ranged
from 720 and 800 mm, without deposition and segregation.

Table 2. Composition of the concrete mixture.

Quantity 1 (kg)
Components
SCC-0.0% SCC-0.3% SCC-0.6%
Cement 32.5R 13.68 15.12 18.00
Fly ash 18.24 20.16 24.00
Limestone powder 4.56 5.04 6.00
Fine river sand 55.68 55.68 55.68
Crushed granite 40.32 40.32 40.32
Water 12.77 13.55 15.60
Superplasticizer 0.073 0.081 0.096
Polypropylene fibers 0.00 0.169 0.352
1 SCC = self-compacting concrete; xx% = volume content of polypropylene fibers.

For each batch of concrete mix, three standard cube samples (150 × 150 × 150 mm) and two
cylinder samples (300 × 150 mm) were cast and cured for 28 days under the same conditions as
the specimens. The concrete compressive strength and tensile strength were determined based on
the standard cube samples. Additionally, the cylinder samples were used to determine the Young’s
modulus of the concrete. The mechanical properties of the SCC used for the pull-out specimens are
listed in Table 3.

Table 3. Mechanical properties of the self-compacting concrete (SCC) batches (Mean values).

Compressive Strength Tensile Strength Elastic Modulus

Batch Designation
(N/mm2 ) (N/mm2 ) (N/mm2 )
SCC-0.0% 54.4 3.7 3.1 × 104
SCC-0.3% 49.5 3.7 3.1 × 104
SCC-0.6% 48.5 3.5 2.8 × 104

2.2. Details of Test Specimens

A series of experimental test consisting of 36 direct pull-out tests using AE were performed to
investigate the bonding behavior of FRP/steel bars in SCC. Concrete cubes sized 200 × 200 × 200 mm
with different types of bars, bar diameters, embedded lengths, and fiber contents, were designed and
Sensors 2019, 19, 159 5 of 14
Sensors 2019, 19, x FOR PEER REVIEW 5 of 14

Sensors 2019, 19, x FOR PEER REVIEW 5 of 14

vertically inSome
fabricated. the moulds before casting,
metal moulds were used lying
forin the central
casting portioncubes.
the concrete of the The
specimens,
bars wereas shown
installed in
Figure 2a.inTo
vertically theavoid
moulds fiberbefore
damage at the
casting, gripping
lying in the zone, a special
central portion protection
of the system consisting
specimens, as shown of
in
vertically in the moulds before casting, lying in the central portion of the specimens, as shown in
steel
Figure sleeve and expansive cement was adopted at the loaded end of rebars. Moreover, plastic tubes
Figure 2a.
2a. To
Toavoid
avoidfiber
fiberdamage
damage at at
thethe
gripping
gripping zone, a special
zone, protection
a special system
protection consisting
system of steel
consisting of
without
sleeve andbonding
expansiverequisites
cement to cement-based
was adopted at materials
the loaded wereofadopted
end rebars. to cover the
Moreover, bars tubes
plastic for providing
without
steel sleeve and expansive cement was adopted at the loaded end of rebars. Moreover, plastic tubes
non-contact
bonding areas between
requisites concretematerials
to cement-based and bars. Theadopted
bonded toareas were positioned in the middle of the
without bonding requisites to cement-based were materials were cover
adopted thetobars for the
cover providing
bars fornon-contact
providing
concrete
areas cubes
between for all
concrete specimens,
andconcrete and
bars. The the embedded length L was changed from 40 mm to 120 mm,
andbonded areas wereareas
positioned in the middle
in theofmiddle
the concrete
e
non-contact areas between bars. The bonded were positioned of the
as shown
cubes in specimens,
forcubes
all Figure 2c. and the embedded length Le was changed from 40 mm to 120 mm, as shown
concrete for all specimens, and the embedded length Le was changed from 40 mm to 120 mm,
in Figure 2c.
as shown in Figure 2c.

130
Steel sleeve

125 130
Steel sleeve
GFRP bar

125
GFRP bar
Plastic tube

60

200
80
Plastic tube

60
Concrete

200
60
80
Concrete

100
60

100
(a) (b) (c)
(a) (a) Metal mould; (b) Final (b)
Figure 2. Pull-out specimen: specimen; (c) Specimen(c)
dimensions (mm).

Figure 2. Pull-out specimen: (a) Metal mould; (b) Final specimen; (c) Specimen dimensions (mm).
After 24 h of casting, all the specimens were demoulded and then cured in a natural indoor
environment
After 24 hatof a temperature of 25
casting, all the ± 3 °C and
specimens were a humidity
demouldedof aboutthen
95%cured
for 28d.
in aFigure 2bindoor
natural shows
and
the final specimen reinforced with GFRP
◦ bar of 12 mm diameter (Φ12) and with an embedment
environment
environment at temperatureofof2525±±33 C
ataatemperature °Cand
anda ahumidity
humidityof of
about 95%
about 95%forfor
28d. Figure
28d. 2b shows
Figure the
2b shows
length
final of 80
specimenmm reinforced withwith
GFRPGFRP
bar ofbar
12 of
mm
the final specimen reinforced 12diameter (Φ12) and
mm diameter (Φ12) with
andanwith
embedment length
an embedment
of 80 mm
length of 80 mm
2.3. Test Setup and Loading Procedure
2.3. Test
TheSetup
2.3. Test test and
Setup and Loading
setup Loading Procedure
Procedureis shown in Figure 3a, where a hydraulic driving anchor puller,
configuration
withThe
a maximum load capacity ofis 100 kN inand a maximum stroke of 60 mm, was used to puller,
provide the
The test
test setup
setupconfiguration
configuration isshown
shown Figure
in Figure3a,3a,
where
wherea hydraulic
a hydraulicdriving anchor
driving anchor with
puller,
pull-out
awith
maximum force.
loadThe specimens
capacity of 100were
kN100 loaded
and under an
a maximum incrementally
stroke of 60 mm, increasing
was used to load
providecondition until
a maximum load capacity of kN and a maximum stroke of 60 mm, was used to the pull-out
provide the
bonding
force. Thefailure,
specimensthe increasement
were loaded of the
under load
an was taken
incrementally as 0.5 kN.
increasing The free
load end slip
condition of the
until bar
bondingwas
pull-out force. The specimens were loaded under an incrementally increasing load condition until
measured using
failure, one linear of variable differential transformer (LVDT) mounted at bar
the bottom of each
bondingthe increasement
failure, the load
the increasement ofwas takenwas
the load as 0.5 kN.as
taken The
0.5free
kN.end
Theslip
freeofend
the slip was
of themeasured
bar was
specimen.
using one One load
linear cell with
variable capacitytransformer
differential of 200 kN was utilized
(LVDT) to determine
mounted at the the magnitude
bottom of each of loading,
specimen.
measured using one linear variable differential transformer (LVDT) mounted at the bottom of each
and load
One the cell
corresponding free 200
end
kNslip of each tospecimen was continuously recordedand viathea
specimen. Onewithloadcapacity
cell withofcapacity was
of 200utilized determine
kN was utilized the magnitude
to determine of loading,
the magnitude of loading,
data-acquisition
corresponding system.
free end slipfree
of each
and the corresponding endspecimen
slip ofwas eachcontinuously
specimen recorded via a data-acquisition
was continuously recorded system.
via a
data-acquisition system.

(a) (b) (c)

Figure3.3.The
Figure The (a)direct
direct pull-out test system:
pull-out test system: (b)
(a) Loading device and
(a) Loading displacement
device (c)
measurement
and displacement equipment;
measurement
(b) AE acquisition
equipment; (b) AEsystem setup;system
acquisition (c) Locations of Locations
setup; (c) the AE sensors.
of the AE sensors.
Figure 3. The direct pull-out test system: (a) Loading device and displacement measurement
equipment; (b) AE acquisition system setup; (c) Locations of the AE sensors.
The AE signals emitted from the debonding process of FRP bar with the hosting concrete cube
were collected by a Micro-II digital system (manufactured by Physical Acoustics Corporation,
The AE signals emitted from the debonding process of FRP bar with the hosting concrete cube
were collected by a Micro-II digital system (manufactured by Physical Acoustics Corporation,
Sensors 2019, 19, 159 6 of 14

The AE signals emitted from the debonding process of FRP bar with the hosting concrete cube were
collected by a Micro-II digital system (manufactured by Physical Acoustics Corporation, Princeton,
NJ, USA), as shown in Figure 3b. PAC R6a sensors, with an effective operation frequency ranging
from 35–100 kHz, were adopted in this test. The R6a sensor has been widely used on metal and FRP
structures due to its high sensitivity and low resonance frequency properties. The peak sensitivity
(Ref V/(m/s)) of R6a sensors is 75 dB. Before the experiment, the pencil lead break testing will be done
to ensure that the R6a sensors have high accuracy. The sampling rate is 1MHz. To shield the noise
signals, the threshold was set as 45 dB. To improve the accuracy of linear location, PDT (peak definition
time: ensures correct identification of the signal peak for risetime and peak amplitude measurements),
HDT (hit definition time: ensures that each AE signal from the structure is reported as one and only
one hit), and HLT (hit lockout time: inhibits the measurement of signals after the hit stored to avoid
measuring reflection) were set as 300, 600, and 1000 µs respectively to reduce reflected waves. Two AE
sensors were attached to the end faces of the bar for linear locating (see Figure 3c).

3. Detection Principles
The AE technique is used to collect and analyze the transient elastic wave caused by the rapid
release of energy of materials, due to the process of deformation or fracture of the materials under
stress [64,65]. As this phenomenon occurs inside materials, the elastic wave caused by fracture
contained the fracture information of materials. Therefore, it is feasible to identify the damage process
of the material by analyzing the elastic wave. In this paper, the AE signal strength and the frequency
components were used to describe the debonding process. The signal strength is mathematically
defined as the integral of the rectified voltage signal over the duration of the AE waveform packet.
Compared to the other AE parameters (such as energy, amplitude), it was not only more sensitive but
also had larger measuring ranges.
Furthermore, the linear location technique is applied in this investigation. It is evident that the
debonding process between the reinforcement bars and SCC is available to release a lot of elastic waves.
Those waves can propagate along the bars and is collected by AE sensors which are located at the
two ends of the bars, as shown in Figure 3c. The arriving time ttop and tbottom is defined as the time of
the elastic wave reaching the top AE sensor and the bottom AE sensor, respectively. Compared to the
arriving time of two AE sensors, the location of damage can be determined, as shown in Equation (1):

Ltop = [(ttop − tbottom ) S + L]/2, (1)

where Ltop is the distance from the AE sensor which is mounted on the top of the bars; S is the
elastic wave speed; and L is the total length of the bars. As a result, the damage characteristics in the
debonding process are investigated through studying the damage location, the signal strength and the
damage frequency. It is found that the distribution of the signal strength and the frequency through
the reinforcing bars can be clearly presented in the AE method. In addition, the relationship of the
location of signal strengths and the frequency of the damage signals between the testing time is also
adopted to study the failure mechanism of the bonding behavior of FRP–SCC effectively.

4. Discussion of Test Results

4.1. Mechanical Properties

The samples were designated according to rebar type (BFRP, GFRP, and steel), bar diameter
(12 mm and 20 mm), embedded length (40 mm, 80 mm and 120 mm), and fiber volume content
(0.0%, 0.3% and 0.6%). For instance, the BFRP reinforced SCC sample cast with fiber content of 0.3%,
bar diameter of 12 mm, and embedded length of 80 mm is designated as SCC-0.3%-BFRP-d12-80.
A uniform distribution of bond stress within the embedded length is assumed, and the average bond
stress is defined as:
τ = P/(πdl) (2)
Sensors 2019, 19, x FOR PEER REVIEW 7 of 14

where P is the pull-out force (kN), d is the nominal bar diameter (mm), and l is the embedded length
(mm). The19,experimental
Sensors 2019, 159 results obtained from the direct pull-out tests, as well as the mode 7 of of
14
failure, are summarized in Table 4. In this table, Pmax is the maximum pull-out force, τmax is the
maximum average bond strength, sfp is the slip value at the bond strength for the free end, and τre is
where
the P is the
residual pull-out
bond stressforce (kN),
in the postd ispeak
the nominal
phase. For barall
diameter (mm), and lthe
tested specimens, is the embedded
average bond length
stress
(mm). The experimental results obtained from the direct pull-out tests,
defined by Equation (2) and the corresponding slip at free end were evaluated, as shown as well as the mode of failure,
in the
are summarized
following section.in Table 4. In this table, Pmax is the maximum pull-out force, τ max is the maximum
average bond strength, sfp is the slip value at the bond strength for the free end, and τ re is the residual
bond stress in the post peak phase. Table For all tested
4. Bond specimens,
results of the average bond stress defined by
the specimens.
Equation (2) and the corresponding slip at free end were evaluated, as shown in the following section.
Pmax τmax sfp τre Failure
Specimen designation τre/τmax
(kN) (N/mm 2) (mm)
Table 4. Bond results of the specimens. (N/mm 2) Mode 1
SCC-0.0%-BFRP-d12-40 21.76 14.43 3.61 7.36 51.0% PO
Pmax τ max sfp τ re Failure
SCC-0.0%-BFRP-d12-80
Specimen Designation 31.79
(kN) 10.54
(N/mm ) 2 3.51
(mm) (N/mm )4.56
2 43.3%
τ re /τ max
Mode PO
1

SCC-0.0%-BFRP-d12-120
SCC-0.0%-BFRP-d12-40 28.73
21.76 6.35
14.43 4.47
3.61 7.361.93 51.0% 30.5% PO PO
SCC-0.0%-BFRP-d20-80
SCC-0.0%-BFRP-d12-80 44.48
31.79 8.85
10.54 1.83
3.51 4.563.41 43.3%38.5% PO PO
SCC-0.0%-BFRP-d12-120 28.73 6.35 4.47 1.93
SCC-0.3%-BFRP-d12-80 26.81 8.89 3.81 3.15 30.5%35.5% PO PO
SCC-0.0%-BFRP-d20-80 44.48 8.85 1.83 3.41 38.5% PO
SCC-0.6%-BFRP-d12-80
SCC-0.3%-BFRP-d12-80 24.22
26.81 8.03
8.89 4.41
3.81 3.152.72 35.5%33.9% PO PO
SCC-0.6%-BFRP-d12-80
SCC-0.0%-GFRP-d12-80 24.22
30.46 8.03
10.10 4.41
3.41 2.727.19 33.9%71.2% PO PO
SCC-0.0%-GFRP-d12-80 30.46 10.10 3.41 7.19 71.2% PO
SCC-0.3%-GFRP-d12-80
SCC-0.3%-GFRP-d12-80 18.67
18.67 6.19
6.19 7.15
7.15 2.762.76 44.6%44.6% PO PO
SCC-0.6%-GFRP-d12-80
SCC-0.6%-GFRP-d12-80 11.88
11.88 3.94
3.94 8.23
8.23 2.602.60 65.8%65.8% PO PO
SCC-0.0%-Steel-d12-80 62.85 20.84 0.76 7.25 34.8% PO
SCC-0.0%-Steel-d12-80
SCC-0.3%-Steel-d12-80
62.85
51.42
20.84
17.05
0.76
1.16 5.05
7.25 29.6%34.8% PO PO
SCC-0.3%-Steel-d12-80
SCC-0.6%-Steel-d12-80 51.42
28.68 17.05
9.51 1.16
1.24 2.965.05 31.1%29.6% PO PO
SCC-0.6%-Steel-d12-80 28.68 19.51
PO = pull-out.1.24 2.96 31.1% PO
1 PO = pull-out.

4.1.1. Effect
4.1.1. Effect of
of Rebar
Rebar Type
Type on
on Bonding
Bonding Behavior
Behavior
Figure 44 shows
Figure shows thethe relationship
relationship between
between bond
bond stress
stress and
and the
the slip
slip obtained,
obtained, respectively,
respectively, forfor
SCC-0.0%-BFRP-d12-80, SCC-0.0%-GFRP-d12-80 and SCC-0.0%-Steel-d12-80,
SCC-0.0%-BFRP-d12-80, SCC-0.0%-GFRP-d12-80 and SCC-0.0%-Steel-d12-80, which differed in which differed in rebar
type only.
rebar As shown
type only. in Figure
As shown 4, the
in Figure 4, BFRP reinforced
the BFRP reinforcedspecimen
specimen hadhadsimilar bonding
similar bondingbehavior
behaviorto
that of the GFRP one. They provide, however, some 50% of bond strength
to that of the GFRP one. They provide, however, some 50% of bond strength compared with the compared with the steel
reinforced
steel specimen
reinforced but with
specimen muchmuch
but with greater slip values.
greater ThisThis
slip values. phenomenon
phenomenon is mainly caused
is mainly by the
caused by
fact fact
the thatthat
the elastic modulus
the elastic modulusof BFRP and and
of BFRP GFRP bars bars
GFRP are lower than than
are lower that of theofsteel
that the bar
steel(Table 1), as
bar (Table
alsoassuggested
1), by Caro
also suggested byetCaro
al. [29]
et and other
al. [29] scholars
and [21–23]. Moreover,
other scholars there is no
[21–23]. Moreover, evident
there is nodropping
evident
stage for specimens
dropping stage for reinforced
specimens with FRP bars,
reinforced withespecially
FRP bars, forespecially
the GFRP for specimen.
the GFRP Thespecimen.
residual bondThe
stress ratio of the sand-coated GFRP bar used in this study was about 71.2%,
residual bond stress ratio of the sand-coated GFRP bar used in this study was about 71.2%, approximately two times
higher than thattwo
approximately of ribbed steel bar
times higher (34.8%).
than that of ribbed steel bar (34.8%).

Figure
Figure 4. Bond stress
4. Bond stress versus
versus slip
slip at
at free
free end of specimens with different rebar types.

4.1.2. Effect of Bar Diameter on Bonding Behavior

4.1.2. Effect of Bar Diameter on Bonding Behavior
The influence of bar diameter on bonding behavior can be inferred from specimens
SCC-0.0%-BFRP-d12-80 and SCC-0.0%-BFRP-d20-80 (Table 4). The two specimens were almost identical
except for bar diameters. The comparison of the bond stress-slip curves is plotted in Figure 5. As shown
Sensors
Sensors 2019,
2019, 19,
19, xx FOR
FOR PEER
PEER REVIEW
REVIEW 88 of
of 14
14

The influence of of bar

bar diameter
diameter on on bonding
bonding behavior
behavior can can bebe inferred
inferred fromfrom specimens
SensorsThe
2019, influence
19, 159 specimens
8 of 14
SCC-0.0%-BFRP-d12-80
SCC-0.0%-BFRP-d12-80 and SCC-0.0%-BFRP-d20-80 (Table 4). The two specimens were almost
and SCC-0.0%-BFRP-d20-80 (Table 4). The two specimens were almost
identical
identical except
except forfor bar
bar diameters.
diameters. The The comparison
comparison of of the
the bond
bond stress-slip
stress-slip curves
curves isis plotted
plotted in
in Figure
Figure
5. As
in As
5. shown
Figure in
5, the
shown Figure
in increase 5, the
Figure 5,inthe increase
barincrease in bar
diameterinfrom diameter from
12 mm to from
bar diameter 20 mm 12 mm to
12resulted
mm to 20 20 mm
in the resulted
mmdecrease in the
the decrease
resultedofinbond strength
decrease
of
by bond
of aboutstrength
bond strength by
by about
16%. Meanwhile, about the16%.
16%. Meanwhile,
corresponding
Meanwhile, slipthe
the corresponding
value decreasedslip
corresponding value
by 48%.
slip valueThedecreased by
by 48%.
experimental
decreased 48%. The
results
The
experimental
experimental results
confirm the tendency confirm
results confirm the
of larger the tendency
bar tendency of
diametersof larger
tolarger bar
develop diameters
barlower to develop
bond strength.
diameters to develop lower
During
lowerthebond
bond strength.
pull-out test,
strength.
During
During the pull-out test, the larger the rebar diameter, the more serious the Poisson effect and the
the the
larger thepull-out
rebar test, the
diameter, larger
the the
more rebar
serious diameter,
the Poissonthe more
effect serious
and the the
more Poisson
obvious effect
the and
shear lag
the
more obvious
phenomenon, the
which shear lag
results phenomenon,
in the lower which
bonding results
strength. in the
The lower
reasons bonding
for
more obvious the shear lag phenomenon, which results in the lower bonding strength. The reasonsthis strength.
phenomenon The reasons
can also
for this
this phenomenon
be found
for can
in the literature
phenomenon also
also be
be found
can [21,24–26]. found inin the
the literature
literature [21,24–26].
[21,24–26].

Figure
Figure 5.
5. Bond
Bond stress
Bond stress versus
stress versus slip
slip at
at free
free end
end of
of specimens
specimens with
with different
different bar
bar diameters.
diameters.

4.1.3.
4.1.3. Effect
Effect ofof Embedded
Embedded Length on Bonding
Length on Bonding Behavior
Behavior
The bond stress-slip relationship for the BFRP bar
The bond stress-slip relationship for the BFRP bar specimens
specimens withwith different
different embedded
embedded lengths
lengths
(40
(40 mm,
mm, 80 80 mm, and 120
mm, and 120 mm)
mm) are
are shown
shown in in Figure
Figure 6. Clearly, there
6. Clearly,
Clearly, there is
there is aa decreasing
decreasing trend
trend ofof the
the
maximum
maximum averageaverage bond stress with the increase in embedded length. During the pull-out test, the
average bond stress with the increase in embedded length. During the pull-out test, the
bond stress
stress transfers
transfers gradually
gradually from
from the
the loaded
loaded end
end toto the
thefree
freeend,
end, having
having a non-linear
a distribution
non-linear
bond stress transfers gradually from the loaded end to the free end, having a non-linear distribution of
distribution
bond
of bond stress along
stress the
along embedment
the embedment length. The
length. nonlinear
The nonlinear stress distribution
stress is
distribution more
of bond stress along the embedment length. The nonlinear stress distribution is more evident in the is evident
more in
evidenttheincase
the
of larger
case
case of embedment
of larger
larger length,length,
embedment
embedment which results
length, whichinresults
which a lower
results inbond
in strength
aa lower
lower bond[27–29].
bond strength
strengthThe[27–29].
nonlinear
[27–29]. Thedistribution
The nonlinear
nonlinear
of stress
distributionalongof the bar
stress is
alongfurther
the confirmed
bar is furtherby AE method
confirmed by in
AEthis study.
method
distribution of stress along the bar is further confirmed by AE method in this study. in this study.

Figure
Figure 6.
Figure 6. Bond
Bond stress
Bond stress versus
stress versus slip
versus slip at
at free
free end
end of
of specimens
specimens with
with different
different embedded
embedded lengths.
lengths.
4.1.4. Effect of
4.1.4. of Fiber Volume
Volume Content on Bonding Behavior
4.1.4. Effect
Effect of Fiber
Fiber Volume Content
Content on on Bonding
Bonding Behavior
Behavior
Figure
Figure 7 illustrates the influence of PP fiber volume contents on the bonding behavior of
Figure 77 illustrates
illustrates the
the influence
influence ofof PP
PP fiber
fiber volume
volume contents
contents on
on the
the bonding
bonding behavior
behavior of of
specimens reinforced
specimens reinforced with
with different
different types
types of
of bars
bars (BFRP,
(BFRP, GFRP
GFRP and steel
and steel bar).
bar). As
As shown
shown in in Figure
Figure 7,7,
specimens reinforced with different types of bars (BFRP, GFRP and steel bar). As shown in Figure 7,
higher
higher PP fiber volume contents unexpectedly developed lower bonding strengths. Compared
higher PPPP fiber
fiber volume
volume contents
contents unexpectedly
unexpectedly developed
developed lower
lower bonding
bonding strengths.
strengths. Compared
Compared with with
with
the the SCC-0.0%-BFRP-d12-80, SCC-0.0%-GFRP-d12-80, and SCC-0.0%-Steel-d12-80 specimens, the
the SCC-0.0%-BFRP-d12-80,
SCC-0.0%-BFRP-d12-80, SCC-0.0%-GFRP-d12-80,
SCC-0.0%-GFRP-d12-80, and and SCC-0.0%-Steel-d12-80
SCC-0.0%-Steel-d12-80 specimens,
specimens, the the
bonding strengths
bonding strengths ofofSCC-0.6%-BFRP-d12-80,
SCC-0.6%-BFRP-d12-80, SCC-0.6%-GFRP-d12-80,
SCC-0.6%-GFRP-d12-80, andand
SCC-0.6%-Steel-d12-80
SCC-0.6%-Steel-d12-80 are
bonding strengths of SCC-0.6%-BFRP-d12-80, SCC-0.6%-GFRP-d12-80, and SCC-0.6%-Steel-d12-80
decreased
are by 23.8%, 61.0%, and 54.4%, respectively. The non-uniform distribution of PP fibers in
are decreased
decreased by by 23.8%,
23.8%, 61.0%,
61.0%, and
and 54.4%,
54.4%, respectively.
respectively. The
The non-uniform
non-uniform distribution
distribution of
of PP
PP fibers
fibers in
in
mortar increases the damage in the matrix and at the interface during the pull-out test, resulting in
poorer friction resistance and a lower bonding strength.
Sensors 2019, 19, x FOR PEER REVIEW 9 of 14

mortar increases
Sensors the damage in the matrix and at the interface during the pull-out test, resulting
2019, 19, 159 9 of 14 in
poorer friction resistance and a lower bonding strength.

(a) (b)

(c)
Figure
Figure 7. Bond
7. Bond stress
stress versus
versus slipslip at free
at free endend
of of specimenswith
specimens withdifferent
different fiber
fiber volume
volume contents:
contents: (a)
(a) Specimen reinforced with BFRP bars; (b) Specimen reinforced with GFRP
Specimen reinforced with BFRP bars; (b) Specimen reinforced with GFRP bars; (c) Specimenbars; (c) Specimen
reinforced
reinforced with
with steel
steel bars.
bars.
4.2. AE Characteristic Parameters
4.2. AE Characteristic Parameters
Figure 8 shows a comparison of the AE signal strength and the bond stress. It can be found
Figure
that the 8bond
shows a comparison
stress and the signalof the AE signal
strength reached strength and thevalue
the maximum bondalmost
stress.atIt the
cansame
be found
time. that
the bond stressthe
Meanwhile, and the signal
maximum strength
value of signalreached
strengththe maximum
in the value
steel bar was muchalmost
largeratthan
the those
sameoftime.
the BFRPthe
Meanwhile, andmaximum
GFRP bars.value
Furthermore,
of signalthere are some
strength in thefluctuations in themuch
steel bar was AE signal
largerstrength curves.
than those of the
These fluctuations represent some damage which release lower energy. However,
BFRP and GFRP bars. Furthermore, there are some fluctuations in the AE signal strength curves. the stress-slip curves
acquired
These by traditional
fluctuations testing
represent methods
some cannot
damage reflectrelease
which these damages clearly. In
lower energy. addition, AE
However, the signals
stress-slip
include the information of frequency and damage location, so it can reflect the debonding process
curves acquired by traditional testing methods cannot reflect these damages clearly. In addition, AE
more comprehensively and accurately.
signals include the information of frequency and damage location, so it can reflect the debonding
Figure 9 illustrates the variation tendency of damage location with time. The Y-axis represents the
process more comprehensively and accurately.
location of the damage along the embedment length: every dot represents one AE event, and the color
of dots represents the value of the signal strength. In the steel bar, as shown in Figure 9c, a few dots
distributed in the loaded end and middle of the embedment length at the beginning of the test, and the
signal strength of these dots was low. After debonding, the dots gradually spread to the free end of the
steel bar and the dots with a larger signal strength occurred. This variation tendency is in agreement
with the expectation that the debonding damage initiates at the loaded end, then propagates to the
free end of the bars. By contrast, the damage location of the GFRP reinforced concrete specimens
always distributed in the middle and free end of the bar, as shown in Figure 9b. In the BFRP reinforced
concrete model, the dots also distributed in the middle and free end of the bar, but the dots with larger
signal strength occurred in the middle of the bar firstly, and gradually spread to the free end of the bar,
as shown in Figure 9a.

(a) (b)
Meanwhile, the maximum value of signal strength in the steel bar was much larger than those of the
BFRP and GFRP bars. Furthermore, there are some fluctuations in the AE signal strength curves.
These fluctuations represent some damage which release lower energy. However, the stress-slip
curves acquired by traditional testing methods cannot reflect these damages clearly. In addition, AE
signals include
Sensors 2019, 19, 159the information of frequency and damage location, so it can reflect the debonding
10 of 14
process
Sensors more comprehensively
2019, 19, x FOR PEER REVIEWand accurately. 10 of 14

Sensors 2019, 19, x FOR PEER REVIEW (c) 10 of 14

(a) (b)
Figure 8. Corresponding relationship between signal strength and bond stress: (a)
SCC-0.0%-BFRP-d12-80; (b) SCC-0.0%-GFRP-d12-80; (c) SCC-0.0%-Steel-d12-80.

Figure 9 illustrates the variation tendency of damage location with time. The Y-axis represents
the location of the damage along the embedment length: every dot represents one AE event, and the
color of dots represents the value of the signal strength. In the steel bar, as shown in Figure 9c, a few
dots distributed in the loaded end and middle of the embedment length at the beginning of the test,
and the signal strength of these dots was low. After debonding, the dots gradually spread to the free
end of the steel bar and the dots with a larger signal strength occurred. This variation tendency is in
agreement with the expectation that the debonding damage initiates at the loaded end, then
propagates to the free end of the bars. By contrast, the damage location of the GFRP reinforced
concrete specimens always distributed in the middle and free end of the bar, as shown in Figure 9b.
In the BFRP reinforced concrete model, the dots(c) also distributed in the middle and free end of the
bar, but the
Figure dots with larger
8.8.Corresponding
Corresponding signal strength
relationship
relationship occurred
betweenbetween insignal
the middle
signal strength of the
strength
and bond bar firstly,
and
stress: bond
(a) and gradually
stress: (a)
SCC-0.0%-BFRP-
spread to the free end
SCC-0.0%-BFRP-d12-80; of the bar, as shown in Figure
(b) SCC-0.0%-GFRP-d12-80;
d12-80; (b) SCC-0.0%-GFRP-d12-80; 9a.
(c) SCC-0.0%-Steel-d12-80.
(c) SCC-0.0%-Steel-d12-80.

Figure 9 illustrates the variation tendency of damage location with time. The Y-axis represents
the location of the damage along the embedment length: every dot represents one AE event, and the
color of dots represents the value of the signal strength. In the steel bar, as shown in Figure 9c, a few
dots distributed in the loaded end and middle of the embedment length at the beginning of the test,
and the signal strength of these dots was low. After debonding, the dots gradually spread to the free
end of the steel bar and the dots with a larger signal strength occurred. This variation tendency is in
agreement with the expectation that the debonding damage initiates at the loaded end, then
propagates to the free end of the bars. By contrast, the damage location of the GFRP reinforced
concrete specimens always distributed in the middle and free end of the bar, as shown in Figure 9b.
In the BFRP reinforced concrete model, the dots also distributed in the middle and free end of the
bar, but the dots with larger signal strength occurred in the middle of the bar firstly, and gradually
spread to the free(a)
end of the bar, as shown in Figure
(b) 9a. (c)
Figure9. 9. Variation
Figure Variation tendency
tendencyofofdamage
damagelocation with
location time:
with (a) SCC-0.0%-BFRP-d12-80;
time: (b)
(a) SCC-0.0%-BFRP-d12-80;
(b)SCC-0.0%-GFRP-d12-80;
SCC-0.0%-GFRP-d12-80; (c)(c)
SCC-0.0%-Steel-d12-80.
SCC-0.0%-Steel-d12-80.

Figure1010illustrates
Figure illustratesthethe frequency
frequency distribution
distribution of
of AE
AE signals
signalsininthe thedebonding
debondingprocess of of
process
specimens reinforced with different types of bars (BFRP, GFRP, and steel bar). As
specimens reinforced with different types of bars (BFRP, GFRP, and steel bar). As shown in Figure 10, shown in Figure
the10,frequency
the frequency of debonding
of debonding process
process in in
thethe BFRPbar
BFRP barand
andthe
the GFRP
GFRP bar barwas
wasconcentrated
concentrated around
around
100kHz;
100kHz; however,
however, thatininthe
that thesteel
steelbar
bardistributed
distributed in
in the
the around
around100100kHZ,
kHZ,140–155
140–155kHz,kHz,170–185
170–185 kHzkHz
and 260–295 kHz. It’s obvious that the frequency domain in the steel bar is wider
and 260–295 kHz. It’s obvious that the frequency domain in the steel bar is wider than that in the FRP than that in the
FRP
bar, bar, particularly
particularly in the highin the high frequency
frequency range.
range. This canThis can be attributed
be attributed to the factto that
the the
factmechanical
that the
mechanical interlocking between the steel ribs and concrete paste is much stronger than that
interlocking between the steel ribs and concrete paste is much stronger than that between the FRP bar

(a) (b) (c)

Figure 9. Variation tendency of damage location with time: (a) SCC-0.0%-BFRP-d12-80; (b)
SCC-0.0%-GFRP-d12-80; (c) SCC-0.0%-Steel-d12-80.
Sensors 2019, 19, 159 11 of 14
Sensors 2019, 19, x FOR PEER REVIEW 11 of 14

and concrete
between the paste.
FRP barAs aand
result, the debonding
concrete paste. Asprocess generates
a result, more high-frequency
the debonding signal in
process generates the
more
steel bar than in the FRP bar.
high-frequency signal in the steel bar than in the FRP bar.

(a) (b) (c)

Figure 10.
10. Frequency
Frequency distribution
distribution ofofAE
AEsignal
signal in the
in the debonding
debonding process
process of specimens:
of specimens: (a)
(a) SCC-
0.0%-BFRP-d12-80; (b) SCC-0.0%-GFRP-d12-80;
SCC-0.0%-BFRP-d12-80; (b) SCC-0.0%-GFRP-d12-80; (c) SCC-0.0%-Steel-d12-80.
(c) SCC-0.0%-Steel-d12-80.

5. Conclusions
5. Conclusions
The AE
The AE method
method was was utilized
utilized to to investigate
investigate the the bond
bond behavior
behavior between
between FRP/steel
FRP/steel bars bars andand
self-compacting concrete
self-compacting concreteininthis thisstudy.
study. AA total of 36
total of pull-out
36 pull-outteststests
withwith
different bar types
different (BFRP,(BFRP,
bar types GFRP,
GFRP, and steel), bar diameters (12 mm and 20 mm), embedded lengths (40 mm, 80 mm, and and
and steel), bar diameters (12 mm and 20 mm), embedded lengths (40 mm, 80 mm, and 120 mm), 120
PP fiber
mm), andcontents
PP fiber(0.0%,
contents0.3%, and 0.6%)
(0.0%, 0.3%, andwere0.6%)
performed. For each test,
were performed. Forthe
each average
test, thebond stress-slip
average bond
relationshiprelationship
stress-slip and the AE and due to thetheAE debonding
due to the process were continuously
debonding process were measured.
continuouslyBased measured.
on the test
results, the following conclusions are drawn.
Based on the test results, the following conclusions are drawn.
The type
The type of of pull-out
pull-out failure
failure was was observed
observed for for all
all the
the specimens
specimens in in this
this study.
study. DueDue to to the
the low
low
elastic modulus
elastic modulus and andthe
thedifferent
different surface
surface treatment,
treatment, thethe
FRPFRPreinforced
reinforcedspecimens
specimensshowshow a lower bond
a lower
strength and a greater corresponding slip value than the steel reinforced
bond strength and a greater corresponding slip value than the steel reinforced specimens. The bond specimens. The bond strength
between between
strength bars and the barshosting
and theSCC decreases
hosting SCC with the increase
decreases with the in increase
the bar diameter
in the bar due to the Poisson
diameter due to
effect and the shear lag phenomenon. In general, the specimens
the Poisson effect and the shear lag phenomenon. In general, the specimens with a shorter with a shorter bar embedded length
bar
can achieve a higher level of average bond strength because of the
embedded length can achieve a higher level of average bond strength because of the non-linear non-linear distribution of stress
along the bar.ofThe
distribution non-uniform
stress along the bar. distribution of PP fibers
The non-uniform may lead aof
distribution poorer friction
PP fibers may resistance
lead a poorerat the
interface during the pull-out test, thus resulting a lower bond strength.
friction resistance at the interface during the pull-out test, thus resulting a lower bond strength.
The AE
The AE signal
signal strength
strength was was consistent
consistent withwith the
the bond
bond stress
stress measured
measured in in each
each specimen,
specimen, which which
can be used as a basis to monitor the bonding strength of FRP/steel
can be used as a basis to monitor the bonding strength of FRP/steel bars in SCC. Meanwhile, the bars in SCC. Meanwhile, the AE
AE
location technique revealed the non-linear distribution of bond stress
location technique revealed the non-linear distribution of bond stress along the bar, the initial along the bar, the initial location
of damage
location of and
damagethe debonding process ofprocess
and the debonding the pull-out
of thetests can betests
pull-out captured
can bebycaptured
the proposedby the AE location
proposed
technique. Moreover, it is revealed that the frequency components
AE location technique. Moreover, it is revealed that the frequency components of the AE signalsof the AE signals measured in the
FRP and steel reinforced test specimens were significantly different.
measured in the FRP and steel reinforced test specimens were significantly different. This This phenomenon indicates that
failure mode of
phenomenon the mechanical
indicates that failureinterlocking
mode ofinthe the mechanical
case of steel interlocking
is stronger than the case
in the failureofmode
steel ofis
stronger than the failure mode of interfacial friction in the FRP reinforced concrete models.inAsthe
interfacial friction in the FRP reinforced concrete models. As a result, the frequency released a
steel bar is generally higher than that in FRP reinforcing materials.
result, the frequency released in the steel bar is generally higher than that in FRP reinforcing
materials.
Author Contributions: G.S. and Y.Z. conceived the idea; B.D. and J.W. designed the experiments; J.W., H.L. and
J.Z. performed the experiments; B.D. and J.W. analyzed the data; B.D. and J.W. wrote the paper; G.S. and Y.Z.
Author Contributions:
made critical revision to G.S. and Y.Z. conceived the idea; B.D. and J.W. designed the experiments; J.W., H.L.
the paper.
and J.Z. performed the experiments;
Funding: The research was supported B.D.byand
the J.W. analyzed
National the data;
Natural B.D.
Science and J.W. wrote
Foundation the (No.
of China paper;51678149),
G.S. and
Y.Z. made critical revision to the paper.
the Guangdong Science and Technology Planning (No. 2016A010103045), the Research Start-up Funds of
DGUT (No. GC300502-31) and the Student’s Platform for Innovation and Entrepreneurship Training Program
Funding: The research was supported by the National Natural Science Foundation of China (No. 51678149), the
(No. 201811819034).
Guangdong Science and Technology Planning (No. 2016A010103045), the Research Start-up Funds of DGUT
(No. GC300502-31) and the Student’s Platform for Innovation and Entrepreneurship Training Program (No.
201811819034).
Sensors 2019, 19, 159 12 of 14

Acknowledgments: The authors greatly appreciate the constructive comments from the anonymous reviewers,
whose comments helped to improve the quality of the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Domone, P.L. Self-compacting concrete: An analysis of 11 years of case studies. Cem. Concr. Compos. 2006, 28,
197–208. [CrossRef]
2. Domone, P.L. A review of the hardened mechanical properties of self-compacting concrete. Cem. Concr.
Compos. 2007, 29, 1–12. [CrossRef]
3. Miller, T.C.; Chajes, M.J.; Mertz, D.R.; Hastings, J.N. Strengthening of a steel bridge girder using CFRP plates.
J. Bridge Eng. 2001, 6, 514–522. [CrossRef]
4. Zheng, Y.; Li, C.; Yang, J.; Sun, C. Influence of arching action on shear behavior of laterally restrained concrete
slabs reinforced with GFRP bars. Compos. Struct. 2015, 132, 20–34. [CrossRef]
5. Xia, L.; Zheng, Y. Deep Embedment (DE) FRP Shear Strengthening of Concrete Bridge Slabs under Loads
Close to Supports. Appl. Sci. 2018, 8, 721. [CrossRef]
6. Zheng, Y.; Sun, C.; Deng, T.; Yang, J.; Lu, Z. Arching action contribution to punching failure of
GFRP-reinforced concrete bridge deck slabs. Arab. J. Sci. Eng. 2014, 39, 8609–8625. [CrossRef]
7. Pendhari, S.S.; Kant, T.; Desai, Y.M. Application of polymer composites in civil construction: A general
review. Compos. Struct. 2008, 84, 114–124. [CrossRef]
8. Nanni, A.; De, L.A.; Zadeh, H.J. Reinforced Concrete with FRP Bars: Mechanics and Design; CRC Press:
Boca Raton, FL, USA, 2014.
9. Dhonde, H.B.; Mo, Y.L.; Hsu, T.T.; Vogel, J. Fresh and hardened properties of self-consolidating
fiber-reinforced concrete. ACI Mater. J. 2007, 104, 491.
10. Yan, F.; Lin, Z.; Yang, M. Bond mechanism and bond strength of GFRP bars to concrete: A review. Compos.
Part B Eng. 2016, 98, 56–69. [CrossRef]
11. Song, G.; Wang, C.; Wang, B. (Eds.) Structural Health Monitoring (SHM) of Civil Structures; MDPI: Basel,
Switzerland, 2018; ISBN1 978-3-03842-783-4. ISBN2 978-3-03842-784-1.
12. Balageas, D.; Fritzen, C.P.; Güemes, A. (Eds.) Structural Health Monitoring; John Wiley & Sons: Hoboken, NJ,
USA, 2010; Volume 90.
13. Sun, M.; Staszewski, W.J.; Swamy, R.N. Smart sensing technologies for structural health monitoring of civil
engineering structures. Adv. Civ. Eng. 2010, 724962. [CrossRef]
14. Li, F.; Murayama, H.; Kageyama, K.; Shirai, T. Guided wave and damage detection in composite laminates
using different fiber optic sensors. Sensors 2009, 9, 4005–4021. [CrossRef] [PubMed]
15. Li, D.; Ho, S.C.M.; Song, G.; Ren, L.; Li, H. A review of damage detection methods for wind turbine blades.
Smart Mater. Struct. 2015, 24, 033001. [CrossRef]
16. Messina, A.; Williams, E.J.; Contursi, T. Structural damage detection by a sensitivity and statistical-based
method. J. Sound Vib. 1998, 216, 791–808. [CrossRef]
17. Long, W.J.; Khayat, K.H.; Lemieux, G.; Hwang, S.D.; Xing, F. Pull-out strength and bond behavior of
prestressing strands in prestressed self-consolidating concrete. Materials 2014, 7, 6930–6946. [CrossRef]
18. Jiang, T.; Kong, Q.; Patil, D.; Luo, Z.; Huo, L.; Song, G. Detection of debonding between fiber reinforced
polymer bar and concrete structure using piezoceramic transducers and wavelet packet analysis. IEEE Sens.
J. 2017, 17, 1992–1998. [CrossRef]
19. Luo, M.; Li, W.; Hei, C.; Song, G. Concrete infill monitoring in concrete-filled FRP tubes using a PZT-based
ultrasonic time-of-flight method. Sensors 2016, 16, 2083. [CrossRef]
20. Li, W.; Fan, S.; Ho, S.C.M.; Wu, J.; Song, G. Interfacial debonding detection in fiber-reinforced polymer
rebar–reinforced concrete using electro-mechanical impedance technique. Struct. Health Monit. 2018, 17,
461–471. [CrossRef]
21. Baena, M.; Lluís, T.; Turon, A.; Barris, C. Experimental study of bond behaviour between concrete and FRP
bars using a pull-out test. Compos. Part B Eng. 2009, 40, 784–797. [CrossRef]
22. Tighiouart, B.; Benmokrane, B.; Gao, D. Investigation of bond in concrete member with fiber reinforced
polymer (FRP) bars. Constr. Build. Mater. 1998, 12, 453–462. [CrossRef]
Sensors 2019, 19, 159 13 of 14

23. Bank, L.C.; Puterman, M.; Katz, A. The effect of material degradation on bond properties of fiber reinforced
plastic reinforcing bars in concrete. ACI Mater. J. 1998, 95, 232–243.
24. Gao, D.Y.; Brahim, B. Influential factors of bond properties between fiber reinforced polymer (FRP) rebars
and concrete. Ind. Constr. 2001, 31, 9–14.
25. Li, C.C.; Gao, D.Y.; Wang, Y.L.; Tang, J.Y. Effect of high temperature on the bond performance between basalt
fibre reinforced polymer (BFRP) bars and concrete. Constr. Build. Mater. 2017, 141, 44–51. [CrossRef]
26. Bazant, Z.P.; Sener, S. Size effect in pullout tests. ACI Mater. J. 1988, 85, 347–351.
27. Alzahrani, M.M.; Aldulaijan, S.U.; Nanni, A.; Bakis, C.E.; Boothby, T. Evaluation of bond using FRP rods
with axisymmetric deformations. Constr. Build. Mater. 1999, 13, 299–309. [CrossRef]
28. Gu, X.; Yu, B.; Wu, M. Experimental study of the bond performance and mechanical response of GFRP
reinforced concrete. Constr. Build. Mater. 2016, 114, 407–415. [CrossRef]
29. Caro, M.; Jemaa, Y.; Dirar, S.; Quinn, A. Bond performance of deep embedment FRP bars epoxy-bonded into
concrete. Eng. Struct. 2017, 147, 448–457. [CrossRef]
30. Jiang, C.; Wu, Y.; Dai, M. Degradation of steel-to-concrete bond due to corrosion. Constr. Build. Mater. 2018,
158, 1073–1080. [CrossRef]
31. Zhou, A.; Chow, C.L.; Lau, D. Structural behavior of GFRP reinforced concrete columns under the influence
of chloride at casting and service stages. Compos. Part B Eng. 2018, 136, 1–9. [CrossRef]
32. Mazaheripour, H.; Barros, J.A.O.; Sena-Cruz, J.M.; Pepe, M.; Martinelli, E. Experimental study on bond
performance of GFRP bars in self-compacting steel fiber reinforced concrete. Compos. Struct. 2013, 95,
202–212. [CrossRef]
33. Mazaheripour, H.; Barros, J.A.O.; Sena-Cruz, J.; Soltanzadeh, F. Analytical bond model for GFRP bars to
steel fiber reinforced self-compacting concrete. J. Compos. Constr. 2013, 17, 04013009. [CrossRef]
34. Golafshani, E.M.; Rahai, A.; Seb, M.H. Bond behavior of steel and GFRP bars in self-compacting concrete.
Constr. Build. Mater. 2014, 61, 230–240. [CrossRef]
35. Mousavi, S.S.; Dehestani, M.; Mousavi, S.M. Bond strength and development length of glass fiber-reinforced
polymer bar in unconfined self-consolidating concrete. J. Reinf. Plast. Compos. 2016, 35, 924–941. [CrossRef]
36. Sethi, V.; Song, G. Multimodal vibration control of a flexible structure using piezoceramic sensor and actuator.
J. Intell. Mater. Syst. Struct. 2008, 19, 573–582. [CrossRef]
37. Wang, F.; Huo, L.; Song, G. A piezoelectric active sensing method for quantitative monitoring of bolt
loosening using energy dissipation caused by tangential damping based on the fractal contact theory.
Smart Mater. Struct. 2017, 27, 015023. [CrossRef]
38. Xu, Y.; Luo, M.; Hei, C.; Song, G. Quantitative evaluation of compactness of concrete-filled fiber-reinforced
polymer tubes using piezoceramic transducers and time difference of arrival. Smart Mater. Struct. 2018,
27, 035023. [CrossRef]
39. Ji, Q.; Ho, M.; Zheng, R.; Ding, Z.; Song, G. An exploratory study of stress wave communication in concrete
structures. Smart Struct. Syst. 2015, 15, 135–150. [CrossRef]
40. Qin, F.; Kong, Q.; Li, M.; Mo, Y.L.; Song, G.; Fan, F. Bond slip detection of steel plate and concrete beams
using smart aggregates. Smart Mater. Struct. 2015, 24, 115039. [CrossRef]
41. Hong, X.; Liu, Y.; Liufu, Y.; Lin, P. Debonding Detection in Hidden Frame Supported Glass Curtain Walls
Using the Nonlinear Ultrasonic Modulation Method with Piezoceramic Transducers. Sensors 2018, 18, 2094.
[CrossRef]
42. Xu, B.; Li, B.; Song, G. Active debonding detection for large rectangular CFSTs based on wavelet packet
energy spectrum with piezoceramics. J. Struct. Eng. 2012, 139, 1435–1443. [CrossRef]
43. Yan, B.; Zou, Q.; Dong, Y.; Shao, X. Application of PZT Technology and Clustering Algorithm for Debonding
Detection of Steel-UHPC Composite Slabs. Sensors 2018, 18, 2953. [CrossRef] [PubMed]
44. Xu, K.; Ren, C.; Deng, Q.; Jin, Q.; Chen, X. Real-time monitoring of bond slip between GFRP bar and concrete
structure using piezoceramic transducer-enabled active sensing. Sensors 2018, 18, 2653. [CrossRef] [PubMed]
45. Giurgiutiu, V.; Harries, K.; Petrou, M.; Bost, J.; Quattlebaum, J.B. Disbond detection with piezoelectric wafer
active sensors in RC structures strengthened with FRP composite overlays. Earthq. Eng. Eng. Vib. 2003, 2,
213–223. [CrossRef]
46. Sevillano, E.; Sun, R.; Gil, A.; Perera, R. Interfacial crack-induced debonding identification in
FRP-strengthened RC beams from PZT signatures using hierarchical clustering analysis. Compos. Part B Eng.
2016, 87, 322–335. [CrossRef]
Sensors 2019, 19, 159 14 of 14

47. Kong, Q.; Robert, R.H.; Silva, P.; Mo, Y.L. Cyclic crack monitoring of a reinforced concrete column under
simulated pseudo-dynamic loading using piezoceramic-based smart aggregates. Appl. Sci. 2016, 6, 341.
[CrossRef]
48. Feng, Q.; Kong, Q.; Huo, L.; Song, G. Crack detection and leakage monitoring on reinforced concrete pipe.
Smart Mater. Struct. 2015, 24, 115020. [CrossRef]
49. Du, G.; Kong, Q.; Zhou, H.; Gu, H. Multiple cracks detection in pipeline using damage index matrix based
on piezoceramic transducer-enabled stress wave propagation. Sensors 2017, 17, 1812. [CrossRef]
50. Xu, J.; Wang, C.; Li, H.; Zhang, C.; Hao, J.; Fan, S. Health Monitoring of Bolted Spherical Joint Connection
Based on Active Sensing Technique Using Piezoceramic Transducers. Sensors 2018, 18, 1727. [CrossRef]
51. Li, W.; Ho, S.C.M.; Patil, D.; Song, G. Acoustic emission monitoring and finite element analysis of debonding
in fiber-reinforced polymer rebar reinforced concrete. Struct. Health Monit. Int. J. 2017, 16, 674–681. [CrossRef]
52. Li, W.; Xu, C.; Ho, S.C.M.; Wang, B.; Song, G. Monitoring concrete deterioration due to reinforcement
corrosion by integrating acoustic emission and FBG strain measurements. Sensors 2017, 17, 657. [CrossRef]
53. Kulandaivelu, P.; Kumar, P.S.; Sundaram, S. Wear monitoring of single point cutting tool using acoustic
emission techniques. Sadhana-Acad. Proceed. Eng. Sci. 2013, 38, 211–234. [CrossRef]
54. Behrens, B.-A.; Huebner, S.; Woelki, K. Acoustic emission a promising and challenging technique for process
monitoring in sheet metal forming. J. Manuf. Process. 2017, 29, 281–288. [CrossRef]
55. Alhashan, T.; Addali, A.; Teixeira, J.A.; Elhashan, S. Identifying bubble occurrence during pool boiling
employing acoustic emission technique. Appl. Acoust. 2018, 132, 191–201. [CrossRef]
56. Alhashan, T.; Elforjani, M.; Addali, A.; Teixeira, J. Monitoring of bubble formation during the boiling process
using acoustic emission signals. Int. J. Eng. Res. Sci. 2016, 2, 66–72.
57. Bunnori, N.M.; Lark, R.J.; Holford, K.M. The use of acoustic emission for the early detection of cracking in
concrete structures. Mag. Concr. Res. 2011, 63, 683–688. [CrossRef]
58. Aldandooh, M.A.A.; Bunnori, N.M. Crack classification in reinforced concrete beams with varying
thicknesses by mean of acoustic emission signal features. Constr. Build. Mater. 2013, 45, 282–288. [CrossRef]
59. Balazs, G.L.; Grosse, C.U.; Koch, R.; Reinhardt, H.W. Damage accumulation on deformed steel bar to concrete
interaction detected by acoustic emission technique. Mag. Concr. Res. 1996, 48, 311–320. [CrossRef]
60. Gallego, A.; Benavent-Climent, A.; Suarez, E. Concrete-galvanized steel pull-out bond assessed by acoustic
emission. J. Mater. Civ. Eng. 2016, 28, 04015109. [CrossRef]
61. Abouhussien, A.A.; Hassan, A.A.A. Acoustic emission monitoring for bond integrity evaluation of reinforced
concrete under pull-out tests. Adv. Struct. Eng. 2017, 20, 1390–1405. [CrossRef]
62. Abouhussien, A.A.; Hassan, A.A.A. Acoustic emission-based analysis of bond behavior of corroded
reinforcement in existing concrete structures. Struct. Control. Health Monit. 2017, 24, e1893. [CrossRef]
63. Wang, L.; Yi, J.; Xia, H.; Fan, L. Experimental study of a pull-out test of corroded steel and concrete using the
acoustic emission monitoring method. Constr. Build. Mater. 2016, 122, 163–170. [CrossRef]
64. Liu, P.F.; Chu, J.K.; Liu, Y.L.; Zheng, J.Y. A study on the failure mechanisms of carbon fiber/epoxy composite
laminates using acoustic emission. Mater. Des. 2012, 37, 228–235. [CrossRef]
65. Geng, J.S.; Sun, Q.; Zhang, Y.C.; Cao, L.W.; Zhang, W.Q. Studying the dynamic damage failure of concrete
based on acoustic emission. Constr. Build. Mater. 2017, 149, 9–16. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).