Construction and Building Materials 161 (2018) 587–597

Contents lists available at ScienceDirect

Construction and Building Materials

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

Flexural behavior of partially fiber-reinforced high-strength concrete

beams reinforced with FRP bars
Haitang Zhu a, Shengzhao Cheng a,⇑, Danying Gao a,b, Sheikh M. Neaz c, Chuanchuan Li a
a
School of Water Conservancy & Environment, Zhengzhou University, Zhengzhou 450001, China
b
Henan University of Engineering, ZhengZhou 451191, China
c
School of Civil, Mining and Environmental Engineering, University of Wollongong, Wollongong, NSW 2522, Australia

h i g h l i g h t s

The behavior of partially fiber-reinforced high-strength concrete beam was investigated.

The optimum thickness of FRHSC layer was 0.57 times of the total depth of the beam.

The addition of steel fibers only in the tension zone reduced the ductility of the beam.

A simplified calculation method for predicting the moment of inertia of FRHSC beams was developed.

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

Article history: In this paper, the flexural behavior of partially fiber-reinforced high-strength concrete (FRHSC) beams
Received 23 March 2017 reinforced with FRP bars was investigated. A total of 12 beams were tested under four-point bending.
Received in revised form 17 August 2017 The effects of the thickness of FRHSC layer, steel fiber volume fraction and FRP reinforcement ratio were
Accepted 1 December 2017
studied. The failure mode, flexural capacity, deflection, crack width and ductility of the tested beams
Available online 12 December 2017
were investigated. The results showed that adding steel fibers in tension zone is an effective way to over-
come the large deflection and large crack width of FRP bar reinforced concrete beams and reduce the cost,
Keywords:
however, the ductility of tested beams decreased with the addition of steel fiber into tension zone. The
Partially fiber reinforced concrete
BFRP bar
optimum thickness of FRHSC layer in FRP bar reinforced concrete beams was 0.57 times of the total depth
Steel fiber of the beam. Higher FRP reinforcement ratio provided better flexural performance including higher flex-
Flexural behavior ural capacity, post-cracking stiffness and ductility and smaller crack width. A simplified calculation
Numerical calculation method for predicting the effective moment of inertia of fully and partially FRHSC beams reinforced with
FRP bars was developed. The beam deflections calculated by the proposed method match well with the
experimental results.
Ó 2017 Published by Elsevier Ltd.

1. Introduction In recent years, Fiber Reinforced Polymer (FRP) bar is consid-

The durability of reinforced concrete structures is regarded as a
major concern due to the corrosion of steel reinforcement. Reduc-
tion of mechanical strength of reinforced concrete structures and
poor bond behavior between steel and concrete are often caused
by the corrosion of steel reinforcement. Significant efforts have
been devoted to in recent years to overcome the problems associ-
ated with the corrosion of steel reinforcement in reinforced con-
crete structures including the use of galvanized steel
reinforcement [1] or addition of rust inhibitor in concrete [2].
⇑ Corresponding author.
E-mail address: chengshengzhao@aliyun.com (S. Cheng).
ered an ideal alternative to steel reinforcing bar [3]. Compared to
the traditional steel reinforcing bars, FRP bars have the character-
istics of corrosion resistance, high specific strength and superior
manufacturability [4]. However, the stress-strain behavior of FRP
bars is linear without any obvious yield point. Also, FRP bar rein-
forced concrete members fail in a brittle manner. Furthermore,
compared to steel reinforcing bars, the modulus of elasticity of
the FRP bar is significantly low and the bond between FRP bar
and concrete is relatively weak. Thus, large deflection and large
crack width are often observed in FRP bar reinforced concrete
structures [5]. Due to the linear stress-strain behavior and low
modulus of elasticity of the FRP bar, the design of FRP bar rein-
forced concrete structures is usually controlled by the serviceabil-
ity limit state requirements [6]. Hence, the high strength of FRP

https://doi.org/10.1016/j.conbuildmat.2017.12.003
0950-0618/Ó 2017 Published by Elsevier Ltd.
588 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597
bars cannot be fully exploited and hence the engineering applica-
tions of FRP bars are still limited.
To mitigate the disadvantages of FRP bars, new FRP bar and FRP
reinforcement system were investigated in a few research studies.
For example, hybrid FRP bars were fabricated with different contin-
uous fibers. Herris et al. [7] reported that stress-strain behavior of
steel can be simulated by hybrid FRP bars which combined different
types of fibers. The ductility of hybrid FRP bar reinforced concrete
beams was found to be close to the corresponding steel bar rein-
forced concrete beams. However, the complex fabrication process
and high manufacturing cost of hybrid FRP bars have limited the
engineering applications of hybrid FRP bars in concrete structures.
A new reinforcement system consisted of FRP bars and steel bars
was also proposed in Qu et al. [8]. It was reported that reinforcing
concrete beams with both FRP bars and steel bars was an effective
way for improving the serviceability and ductility of FRP bar rein-
forced concrete beams. However, the corrosion of steel bars was
not fully mitigated in the proposed hybrid reinforcement system
in Qu et al. [8]. Hence, the corrosion of steel reinforcement still
remains a critical issue for the hybrid reinforcement system.
Recently, in a few research investigations, it was observed that
the deflection and crack width of FRP bar reinforced concrete
beams were restrained and the ductility was improved by the addi-
tion of randomly distributed fibers [9,10]. It is well known that the
influence of fibers on the deflection and crack is mainly achieved
by the fibers in the tension zone of concrete beams. The addition
of fibers in a partial depth (in the tensile zone) of concrete beams
can not only achieve the full effect of fibers but also can provide a
considerable cost saving [11,12]. Therefore, the addition of fibers in
a partial depth of FRP bar reinforced concrete beams may be a pos-
sible way to improve the flexural behavior and to reduce cost. The
positive effect of fibers on flexural behavior of fiber reinforced con-
crete (FRC) beam reinforced with FRP bar was investigated in the
literature. However, the flexural behavior of partially FRC beams
has not yet been investigated adequately. Thus, the flexural behav-
ior of partially FRC beams reinforced with FRP bars needs to be
extensively investigated.
In this paper, the flexural behavior of partially fiber-reinforced
high-strength concrete (FRHSC) beams reinforced with FRP bars
including flexural capacity, deflection response, cracking response
and ductility was investigated. In total 11 FRP bar reinforced par-
tially FRHSC beams with various thicknesses of FRHSC layer, steel
fiber volume fractions and FRP reinforcement ratios and one tradi-
tional steel bar reinforced FRHSC beams were cast and tested. The
tested beam deflections at service loads were compared with the
beam deflections calculated according to ACI 440.1R-15 for short
term service load deflections. Considering the positive effect of
steel fibers on beam deflection, a simplified method for beam
deflection has been developed which was based on appropriate
calculations of the effective moment of inertia of fully and partially
FRHSC beams reinforced with FRP bars.
2. Experimental program
2.1. Materials and mix proportion of concrete
River sand with a maximum particle size of 5 mm was used as
fine aggregate. Crushed limestone with the size ranging from 5 to
20 mm was used as coarse aggregate. Type Ⅰ Portland cement and
fly ash (FA) were used as cementitious materials. Polycarboxylic
acid water reducing agent was used to improve the workability
of concrete. Five types of reinforcing bars were used in this exper-
imental investigation: steel bars with diameters of 6 mm, 10 mm
and 14 mm and basalt FRP (BFRP) bars with diameters of 12 mm
and 14 mm. The surfaces of FRP bars were deformed to improve
the bond performance between FRP bars and concrete. The
mechanical properties of reinforcing bars are shown in Table 1.
The fibers used in FRHSC are hooked steel fibers. The properties
of the fibers are reported in Table 2 and shapes of steel fibers are
shown Fig. 1. In the experimental program, five concrete mix pro-
portions were designed and used, as reported in Table 3.
2.2. Test specimens
In this study, a total of 12 beam specimens were cast and tested
under four-point bending. The specimen dimensions and reinforce-
ment details are shown in Fig. 2 and Table 4. All tested beam spec-
imens were 2100 mm long with 150 mm  300 mm cross-section.
The clear span of the beam specimens under four-point bending
was 1800 mm. The concrete clear cover for the tension reinforce-
ment was 25 mm. Each beam specimen was reinforced laterally
with 10 mm diameter steel stirrup at a center-to-center spacing
of 75 mm in the shear region. Compression reinforcements and
stirrups were not used in the test zone (Fig. 2). It is noted that
the test zone is under pure bending moment without any shear
stress. Although the length of the steel fiber is greater than the
thickness of concrete cover, the distribution of steel fibers in the
pure bending region is not significantly affected by the concrete
cover. The steel fibers with 0–2% by volume were added to the
FRHSC (the depth is 90 mm–300 mm). Eleven beam specimens
were reinforced longitudinally with FRP bars having the diameter
of 12 mm and 14 mm and one beam specimen was reinforced lon-
gitudinally with steel bars having the diameter of 14 mm. The lon-
gitudinal reinforcement ratio of the beam specimens varied from
0.56% to1.65%. Meanwhile, the standard specimens with the size
of 150 mm  150 mm150 mm were used for compressive
strength test and splitting tensile strength test. Prisms with the
size of 150 mm  150 mm300 mm were cast and tested for mod-
ulus of elasticity. The specimens were made of two types of con-
crete, i.e. plain concrete and FRHSC with the target compressive
strength of concrete was equal to 60 MPa.
In the specimen identification in Table 4, the first character
indicates the longitudinal reinforcement type, where B represents
BFRP bar and S represents steel bar. The second character in the
specimen identification indicates the steel fiber volume fraction,
where 0 represents 0% by volume of steel fiber, 1 represents 0.5%
by volume of steel fiber, 2 represents 1% by volume of steel fiber,
3 represents 1.5% by volume of steel fiber and 4 represents 2% by
volume of steel fiber. The third character in the specimen identifi-
cation indicates the steel fiber reinforcement form, where A repre-
sents fully steel fiber reinforced and P represents partially steel
fiber reinforced. The forth character in the specimen identification
indicates the longitudinal reinforcement area, where 1 represents
2U12, 2 represents 2U14, 3 represents 3U14 and 4 represents
4U14. The fifth character in the specimen identification represents
the thickness of FRHSC layer in partially steel fiber-reinforced
beams, where 1 represents 90 mm, 2 represents 130 mm, and 3
represents 170 mm.
2.3. Test setup and measurements
All beams were simply supported and were tested under the
four-point bending, as shown in Fig. 3. The load was increased
monotonically until beams failed. Before the appearance of the first
crack, each load stage was 3 kN and after the appearance of the
first crack, each load stage was 10 kN. All the deflections, strains
in the reinforcement and concrete, crack widths and crack distribu-
tions were recorded after 3 mins of sustained loading. The deflec-
tions at mid-span, at two loading points, at two mid-points
between the center of the beam and the loading points, and at sup-
ports were measured by seven linear voltage differential trans-
 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597 589

Table 1
Mechanical properties of reinforcing bars.

Bar type Diameter (mm) Modulus of Elasticity (GPa) Yield strength (MPa) Tensile strength (MPa)
BFRP-1 12 43.26 — 1034.1
BFRP-2 14 41.79 — 1030.6
Steel-1 6 202 320 465
Steel-2 10 232.96 485 610
Steel-3 14 196 506.8 686.2

Table 2
Properties of steel fiber.

Type of fiber Tensile strength (MPa) Modulus of elasticity (GPa) Length (mm) Diameter (mm) Length-diameter ratio
Hooked steel fiber 1345 200 35 0.55 65

Fig. 1. Shape and size of steel fibers.
formers (LVDTs). The crack widths at the location of longitudinal
reinforcement were measured by crack width comparator. The
arrangement of LVDTs and schematic of the location of crack width
measuring point are shown in Fig. 4 and Fig. 5, respectively.
3. Experimental results and discussions
3.1. Cracking loads, failure modes and flexural capacities
Due to the inclusion of steel fibers, the stiffness of the speci-
mens was expected not to be decreased immediately at the
appearance of the first crack. Thus, the first crack point obtained
from load-deflection curve was observed critically. The cracking
load was defined as the load when the slope of load-deflection
curve suddenly decreased. The summary of test results is reported
in Table 5. The fibers in tension zone carried a portion of tensile
stress leading to a higher cracking load with the increase of fiber
volume fraction and thickness of FRHSC layer. The cracking loads
of the specimens with different reinforcement ratios did not vary
significantly. This is because the modulus of elasticity of FRP bars
(43.26GPa/41.79GPa) was similar to the modulus of elasticity of
concrete (41.6–43.9GPa). Hence, the FRP bar reinforcement ration
did not have a significant effect on the initial stiffness of the beam
specimens.
Due to the linear stress-strain behavior of FRP bars, two types of
failure modes were observed for the tested beam specimens: con-
crete crushing and FRP bar rupture. The results show that the fail-
ure mode of specimens was influenced by FRP reinforcement ratio
and the thickness of FRHSC layer. For beam specimens with differ-
ent FRP reinforcement ratios (B2P1-3, B2P2-3, B2P3-3, B2P4-3),
FRP bar rupture was observed in Specimen B2P1-3. When the load
reached to the ultimate load, a substantial portion of fibers in ten-
sion zone were pulled out. At that moment, as almost all of the ten-
sile stress was carried by FRP bars, the FRP bars failed when the
stress reached to ultimate tensile strength. With the increase of
FRP reinforcement ratio, the capacity of FRP bars improved and
hence concrete crushing occurred before the rupture of FRP bar.
For beams with different FRHSC layers (Specimens B0A2, B2P2-1,
B2P2-2, B2P2-3, B2A2), the failure mode due to concrete crushing
was observed in Specimens B0A2, B2P2-1, B2P2-2 and B2P2-3.
When steel fibers were added in the full cross section of the spec-
imen, compressive strength, peak compressive strain and ultimate
compressive strain improved due to the presence of steel fibers
[13]. At the ultimate load, FRP bars ruptured before concrete
crushing.
FRP reinforcement ratio, the thickness of FRHSC layer and steel
fiber volume fraction have different influences on flexural capacity
of the tested beam specimens. The test results showed that a
higher FRP reinforcement ratio provided a higher flexural capacity.
Compared to Specimen B2P1-3, the ultimate loads of Specimens

Table 3
Mix proportion of Concrete.

No. qsf (%) W/C Unit weight Unit weight Unit weight Unit weight Unit weight Unit weight Unit weight Fiber
W (kg/m3) C (kg/m3) FA (kg/m3) S (kg/m3) G (kg/m3) WRA (kg/m3) (kg/m3)
1 0 0.27 180 480 180 612 998 3.84 0
2 0.5 583 951 4.08 39.25
3 1.0 555 905 4.08 78.5
4 1.5 526 858 4.32 117.75
5 2.0 497 811 5.28 157

Note: qsf is the steel fiber volume fraction.

590 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597

Fig. 2. Dimensions and reinforcement details of a typical specimen tested.

Table 4
Details of test specimens.

Specimen Type of reinforcement Stirrups Af qsf (%) hsf (mm) Ec (GPa)

Plain concrete FRHSC
B0A2 BFRP C10@75 2U14 0 0 41.6 –
B2P2-1 BFRP C10@75 2U14 0/1.0 90 41.6 42.4
B2P2-2 BFRP C10@75 2U14 0/1.0 130 41.6 42.4
B2P2-3 BFRP C10@75 2U14 0/1.0 170 41.6 42.4
B2A2 BFRP C10@75 2U14 1.0 300 – 42.4
B1P2-3 BFRP C10@75 2U14 0/0.5 170 41.6 42.8
B3P2-3 BFRP C10@75 2U14 0/1.5 170 41.6 41.6
B4P2-3 BFRP C10@75 2U14 0/2.0 170 41.6 43.9
B2P1-3 BFRP C10@75 2U12 0/1.0 170 41.6 42.4
B2P3-3 BFRP C10@75 3U14 0/1.0 170 41.6 42.4
B2P4-3 BFRP C10@75 4U14 0/1.0 170 41.6 42.4
S2A2 Steel C10@75 2U14 1.0 300 – 42.4

Note: Af is the area of FRP reinforcement, hsf is the thickness of FRHSC layer, Ec is the modulus of elasticity of concrete.

Fig. 5. Location of measuring point.

Fig. 3. The test setup.

Fig. 4. Schematic of flexural test setup and locations of LVDTs.

 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597 591

Table 5
Summary of test results.

Specimen Fcr (kN) Fsd (kN) Fu (kN) Du/mm Fsd/Fu Failure mode DF
B0A2 25.04 78.68 215.27 37.09 0.37 Concrete crushing 8.58
B2P2-1 37.23 90.23 219.46 34.93 0.41 Concrete crushing 7.20
B2P2-2 44.77 104.36 225.11 32.47 0.46 Concrete crushing 6.16
B2P2-3 46.21 115.88 229.03 31.86 0.51 Concrete crushing 5.87
B2A2 47.03 117.29 264.13 43.83 0.44 FRP bar rupture 9.20
B1P2-3 44.12 95.16 218.33 37.04 0.44 Concrete crushing 7.45
B3P2-3 51.18 119.87 230.67 30.73 0.52 Concrete crushing 5.45
B4P2-3 53.54 128.47 237.29 31.83 0.54 Concrete crushing 5.79
B2P1-3 43.61 99.76 177.50 30.80 0.56 FRP bar rupture 5.07
B2P3-3 47.33 130.63 273.96 30.84 0.48 Concrete crushing 6.01
B2P4-3 46.31 141.93 290.21 29.21 0.49 Concrete crushing 5.72
S2A2 52.25 – 195.00 48.60 – Steel bar yield –

Note: Fcr is the cracking load, Fsd is the service load controlled by deflection, Fu is the ultimate load, Du is the ultimate deflection, DF is the deformation factor.

300 300
B2A2 B2A2
250 250
B2P2-3
200 S2A2 200

Load (kN)
Load (kN)

B2P2-1
150 150 B2P2-2
B2A0
100 100

50 50

0 0
0 10 20 30 40 50 0 10 20 30 40 50
Midspan deflection (mm) Midspan deflection (mm)
(a) Beams with different reinforcements types (b) Beams with different FRHSC layers
300 300
B4P2-3 B2P4-3
250 250 B2P3-3

B3P2-3 B2P2-3
200 200
B2P2-3
Load (kN)

Load (kN)

150 150 B2P1-3

B0A2
100 B1P2-3 100

50 50

0 0
0 10 20 30 40 50 0 10 20 30 40 50
Midspan deflection (mm) Midspan deflection (mm)
(c) Beams with differentsteel fiber volume fractions (d) Beams with different FRP reinforcement ratios
Fig. 6. Load versus mid-span deflection.

B2P2-3, B2P3-4 and B2P4-3 increased by 23.4% to 63.5% due to the
increase of FRP reinforcement ratio. It indicates that increasing FRP
reinforcement ratio is an effective way to improve the flexural
capacity of partially FRSHC beams reinforced with FRP bars. For
the tested beam specimens which were partially reinforced with
steel fibers, the ultimate loads slightly increased with the increase
of the thickness of FRHSC layer and steel fiber volume fraction. This
is because FRP bars carried the majority of tensile stress, although a
portion of tensile stress was carried by steel fibers. When fibers
were added in full cross section, the ultimate load was greatly
improved and the failure mode also changed from concrete crush-
ing to FRP bar rupture as the performance of concrete in the com-
pression zone improved by the addition of steel fibers.
3.2. Load-deflection response
The load-deflection curves of the tested beam specimens rein-
forced with steel bars and FRP bars are shown in Fig. 6(a). There
was a significant difference between the two load-deflection
curves. For the Specimen S2A2, the load-deflection curve showed
a tri-linear response. The first part of the load-deformation
response represented the stiffness of uncracked beam section.
The second part of the load-deformation response represented
the cracked stiffness of the beam section. When the first crack
appeared, the stiffness of the beam decreased. With the increase
of the load, the stress of steel bars increased until the yield strength
was reached. Afterward, the deflection of the beam specimen sig-
592 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597

nificantly increased representing a softening response. Since there
was no obvious yield point for FRP bars, the load-deflection curve
of Specimen B2A2 showed a linear response after the cracking load.
Due to the low modulus of elasticity of the FRP bars, the slope of
the load-deflection curve of S2A2 was significantly larger than that
of B2A2, as shown in Fig. Fig. 6(a). Before the steel bar in Specimen
S2A2 yielded, the mid-span deflections of Specimen S2A2 were
about 45% smaller than that of Specimen B2A2 at the same load.
Fig. 6(b) shows the load-deflection curves of the test beam spec-
imens with different FRHSC layers. Under the cracking load, the
slope of load-deflection curves did not show significant differences.
This result indicated that the effect of steel fibers on the initial
stiffness of the beam was not significant [14]. After cracking, the
stiffness of tested beam specimen decreased with the increase of
the number and width of cracks. Due to the presence of steel fiber,
the integrity of the specimens was improved and the degradation
of stiffness was restrained. When the thickness of FRHSC layer
increased, the positive effects of steel fibers on the post cracking
stiffness became significant. For Specimen B2P2-3, the thickness
of FRHSC layer was 170 mm (hsf = 0.57 h), steel fibers were dis-
tributed in the full tension zone, and hence the crack resistance
effect was fully achieved. At the same load, the mid-span deflection
of Specimen B2P2-3 was approximately equal to that of Specimen
B2A2 (the beam which was fully reinforced with steel fibers). For
Specimen B2P2-1, B2P2-2, B2P2-3 and B2A2, on average, the
deflections were 18.69%, 34.18%, 40.86%, 44.14%, respectively,
lower than the deflection of Specimen B0A2.
Fig. 6(c) shows that the stiffness of tested beam specimens was
positively influenced by the steel fiber volume fraction. Because of
the low modulus of elasticity of the FRP bars and large deflection of
FRP bar reinforced structures, the design of concrete beams rein-
forced with FRP bars is usually controlled by the serviceability
limit state requirements and hence the high flexural capacity of
concrete beams cannot be fully exploited. The load-deflection
curves showed that the reduction of the stiffness was restrained
by the inclusion of steel fibers. The loads at all deflection increased
significantly with the inclusion of steel fibers. For B1P2-3, B2P2-3,
B3P2-3 and B4P2-3, the mid-span deflections were about 23.5%,
42.93%, 48.22% and 52.33%, respectively, smaller than the mid-
span deflection of B0A2. The comparison between the service load
(Fsd) and ultimate load is shown in Table 5. The service load was
calculated based on the load at an assumed service mid-span
deflection limit of l/180. As expected, due to the low modulus of
elasticity of the FRP bars, the service load of B0A2 was just 0.365
times of the ultimate load. With the addition of steel fibers, the
ratio of the service load and ultimate load was improved. The ser-
vice load increased by 63.28% for Specimen B4A2-3.
The load-deflection curves of the tested beam specimens with
different FRP reinforcement ratios are shown in Fig. 6(d). Before
the appearance of the first crack, the load-deflection response of
the beam specimens were similar. This can be attributed to the mod-
ulus of elasticity of the FRP bars which was similar to that of con-
crete. The FRP reinforcement ratio had a significant influence on
the post cracking bending stiffness of the tested beam specimens.
The tensile stress at the cracked section was mainly carried by FRP
bars. Hence, a higher FRP reinforcement ratio can reduce the tensile
strain of FRP bars at the cracked section and provide a higher post
cracking bending stiffness. For B2P2-3, B2P3-3, B2P4-3, the mid-
span deflections were 23.96%, 38.94%, 40.86%, 47.34%, respectively,
lower than the mid-span deflection of Specimen B2P1-3.

350 350

300 300

250 250
Load (kN)
Load (kN)

200 200

150 150 B0A2

100 B2P2-1
100
B2P2-2
50 S2A2 B2P2-3
50
B2A2 B2A2
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Maximum crack width (mm) Maximum crack width (mm)
(a) Beams with different reinforcements types (b) Beams with different FRHSC layers

350 350

300 300

250 250
Load (kN)
Load (kN)

200 200

150 B0A0 150

B1P2-3 B2P1-3
100 100
B2P2-3 B2P2-3
50 B3P2-3 50 B2P3-3
B4P2-3 B2P4-3
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Maximum crack width (mm) Maximum crack width (mm)
(c) Beams with different steel fiber volume fractions (d) Beams with different FRP reinforcement ratios
Fig. 7. Load versus maximum crack width.
 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597 593

3.3. Cracking response
The load-maximum crack width curves are shown in Fig. 7. For
all beams, the development of cracks can be divided into 3 stages:
(i) after the appearance of the first crack (flexural crack), the num-
ber and depth of cracks increased rapidly with the increase of
applied load; (ii) the rate of increase of the number and depth of
cracks was decelerated and shear cracks appeared as the load con-
tinued to increase; (iii) the number of cracks remained the same
but the cracks continued to propagate slowly to the edge of com-
pression zone until the specimens failed.
For Specimens S2A2 and B2A2, the number of crack in a pure
bending section of S2A2 was more than that of B2A2. This can be
attributed to the higher modulus of elasticity of the steel bar and
the better bond performance between steel bar and concrete. As
shown in Fig. 7(a), the load-maximum crack width curve of S2A2
was significantly different from that of B2A2. Before steel bars
yielded, the crack width increased linearly with the increase of
the load. Afterward, the crack width increased rapidly with slight
increase in the load. Because of the linear stress-strain behavior
of FRP bar, for Specimen B2A2, the crack width increased linearly
until beam failed. Due to the higher modulus of elasticity of the
steel bar, before steel bar yield, at the same load, the crack width
of S2A2 was 66.67–81.17% smaller than that of B2A2.
For beams with different FRHSC layers, a larger number of
cracks were observed in the beam specimens with a higher depth
of FRHSC layer. Compared to the beams reinforced with steel fiber,
the depth of cracks in B0A2 developed rapidly and the number of
cracks was less. As shown in Fig. 7(b), after the first crack appeared,
for the increase of load, the crack width of steel fiber-reinforced
beams was significantly smaller than that of B0A2. However, for
B2P2-1, B2P2-2, B2P2-3, B2A2, no significant difference in the
crack width was observed. This was because, as the depths of
cracks were shallow and the widths of cracks were small, the crack
resistance effect of steel fibers at the bottom of the beams was fully
achieved while the effect of steel fibers in a higher position of the
beams was not implemented. When the load exceeded 100 kN,
cracks developed deeper and steel fibers restrained the develop-
ment of cracks in a higher position of the beams, hence, crack
width was smaller for beam specimens with a higher depth of
the FRHSC layer. When the load exceeded 160 kN, a considerable
portion of steel fibers at the bottom of the beams was pulled out.
At the ultimate load, the crack width of B2P2-1 was quite similar
to that of B0A2. For B2P2-3 and B2A2, the cracking behavior was
similar. This indicated that, for the FRP bar reinforced FRHSC
beams, the crack resistance effect of steel fiber was mainly
achieved by the steel fibers which were at the depth lower than
0.57 times of the total depth of the beams. For B2P2-1, B2P2-2,
B2P2-3 and B2A2, the maximum crack widths were 27.9%,
35.24%, 45.03%, 44.51%, respectively, smaller than the maximum
crack width of Specimen B0A2.
The addition of steel fiber led to a greater number of cracks in
the tested beam specimens. However, with the increase of steel
fiber volume fraction, the number of cracks did not vary signifi-
cantly. In Fig. 7(c), it can be seen that the cracks widths were effec-
tively restrained by steel fibers. Compared to B0A2, the maximum
crack widths for B1P2-3, B2P2-3, B3P2-3 and B4P2-3 were reduced
by 22.74%–55.97% by the addition of steel fibers with 0.5–2.0% vol-
ume fraction.
For beams with different FRP reinforcement ratios, a higher FRP
reinforcement ratio provided a greater number of cracks and lower
speed of the development of crack depth and width. In Fig. 7(d), at a
certain load, a smaller crack width was obtained when a higher FRP
reinforcement ratio was used. This indicates that increasing the FRP
reinforcement ratio is an effective way to reduce the crack width. In
addition, for the beam specimens with higher FRP reinforcement
ratio, more shear cracks appeared and propagated deeply. This
can be attributed to the larger shear stress corresponding to the lar-
ger ultimate load. However, due to the adequate shear resistance
provided by the stirrups, the widths of shear cracks were not signif-
icantly large. For Specimens B2P2-3, B2P3-3 and B2P4-3, the max-
imum crack widths were 30.2%, 51.46%, 63.15%, respectively,
smaller than the maximum crack width of Specimen B2P1-3.
Table 6 shows the average crack widths and standard deviation
for crack widths at the load of 80kN, 100kN and 120kN. With the
increase of applied load, the average crack widths and the standard
deviation of the crack widths increased, as reported in Table 6. Since
the modulus of elasticity of the steel bar is higher and the bond
behavior between steel fiber and concrete is better, the average
crack width of Specimen S2A2 was 66.77%–70.87% and standard
deviation of the crack widths was 67.87%–70.85% smaller than
those of Specimen B2A2, respectively. With the addition of steel
fiber, the integrity of specimens improved and the tensile stress
of concrete became more uniform. This resulted in a smaller aver-
age crack width and smaller standard deviation of the crack widths.
With the increase of steel fiber volume fraction and thickness of
FRHSC layer, the average crack width decreased by 13.89%–
86.26% and the deviation for crack widths decreased by 3.05%–
66.06%. Higher FRP reinforcement ratio led to a smaller average
crack width and smaller standard deviation of the crack widths.
For the specimens with the FRP reinforcement ratio of 0.77%–
1.65%, average crack widths were 8.70%–71.19% and deviations of
the crack widths were 19.47%–69.97% smaller than those of B2P1-3.
The bond performance between FRP bars and concrete was rel-
atively weak. For Specimen B0A2, the horizontal cracks, which
were caused due to bond failure, appeared at the location of longi-
tudinal reinforcement. With the increase of the load, the horizontal
cracks developed continuously. At the ultimate load, two adjacent

Table 6
Comparison between crack widths.

Specimen Average crack width (mm) Maximum crack width (mm) Standard Deviation of the crack width
F = 80kN F = 100kN F = 120kN F = 80kN F = 100kN F = 120kN F = 80kN F = 100kN F = 120kN
B0A2 0.684 0.820 0.900 1.100 1.150 1.400 0.221 0.243 0.295
B2P2-1 0.269 0.539 0.775 0.460 1.010 1.130 0.135 0.228 0.286
B2P2-2 0.227 0.348 0.503 0.470 0.720 1.260 0.125 0.124 0.249
B2P2-3 0.189 0.314 0.481 0.370 0.660 0.920 0.091 0.127 0.168
B2A2 0.201 0.270 0.339 0.360 0.700 0.820 0.093 0.121 0.161
B1P2-3 0.275 0.378 0.509 0.650 0.770 1.300 0.131 0.185 0.21
B3P2-3 0.151 0.233 0.307 0.450 0.580 0.820 0.113 0.129 0.171
B4P2-3 0.094 0.162 0.227 0.270 0.500 0.720 0.075 0.116 0.164
B2P1-3 0.207 0.368 0.545 0.800 1.300 1.600 0.113 0.243 0.353
B2P3-3 0.100 0.126 0.166 0.250 0.360 0.580 0.059 0.083 0.112
B2P4-3 0.080 0.130 0.157 0.180 0.300 0.400 0.049 0.075 0.106
S2A2 0.059 0.079 0.113 0.080 0.140 0.220 0.027 0.039 0.052
594 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597
vertical cracks were connected by a horizontal crack, the concrete
cover spalled off. For the beams reinforced with steel fibers, no hor-
izontal cracks and the spalling of the concrete cover was observed.
This indicates that the bond performance between FRP bars and
concrete was increased by the addition of steel fibers.
3.4. Ductility
Since there was no obvious yield point in FRP bars, the tradi-
tional theory of ductility cannot be applied to the structures rein-
forced with FRP bars. Vijay and Ganga [15] proposed a new
ductility index, deformation factor (DF), for concrete structures
reinforced with FRP bars. It was defined as the ratio of the energy
absorbed at the ultimate load to the energy absorbed at the service
load. The service load was taken as the load corresponding to mid-
span deflection limit of span/180. The ductility factor (DF) for the
tested specimens are shown in Table 5.
For beams with different FRHSC layers, the DFs of partially
fiber-reinforced beams were smaller than the DF of Specimen
B0A2. The DF decreased with the increase of the thickness of
FRHSC layer (hsf 170 mm). This can be attributed to that for the
beam specimens partially reinforced with steel fiber, the deflec-
tions were restrained by the steel fibers in the tension zone and
the ultimate deflections were smaller and service loads were larger
than those of Specimen B0A2, while the ultimate loads were not
significantly different. However, when hsf was equal to 300 mm,
the DF of Specimen B2A2 was larger than the DF of Specimen
B0A2. This is mainly because the compressive strength and ulti-
mate compressive strain of compressed concrete improved due
to the addition of steel fibers in the compression zone. Hence,
the ultimate load and ultimate deflection of Specimen B2A2 were
larger than those of the beams partially reinforced with steel fiber.
This indicates that the improvement of ductility of FRP bar rein-
forced concrete beams can be achieved by the addition of steel
fibers in the compression zone. Thus, the addition of steel fibers
in the full depth of structures is necessary for the structures with
higher ductility requirements.
For the beam specimens with different steel fiber volume frac-
tions, it can be observed that steel fiber volume fraction is negatively
correlated with the ductility of the beams which were partially rein-
forced with steel fiber. Hence the addition of steel fibers in the ten-
sion zone was not beneficial for the ductility of the beam.
For beams with different FRP reinforcement ratios, the FRP rein-
forcement ratio influenced the ductility of the beam specimens. It
can be observed that there was no significant difference in the ulti-
mate deflections of the beams with different FRP reinforcement
ratios. However, for Specimens B1P2-3, B2P2-3, B3P2-3, the ulti-
mate loads were significantly improved by the increase of FRP rein-
forcement ratio and a higher DF was obtained for a higher FRP
reinforcement ratio. Comparing Specimens B4P2-3 and B3P2-3,
the rate of increase of the ultimate load was not higher than the
rate of increase of the service load, which led the DF of Specimen
B4P2-3 slightly smaller than the DF of Specimen B3P2-3. This
was similar to the result reported in a previous study [16].
3.5. Discussion on full and partial steel fiber reinforcement
As mentioned in section 1, the deflection and crack width of
concrete beams reinforced with FRP bars are usually large and
the design is controlled by serviceability limit state. For the con-
crete beams reinforced with FRP bars, the enhancement of the ser-
vice load and stiffness are more important than the enhancement
of the ultimate strength. Thus, the main purpose of the addition
of steel fibers into FRP bars reinforced concrete beams is to
increase the service load and decrease the deflection and crack
width. The partial reinforcement with fiber did not increase the
ultimate strength. However, the deflection and crack width
decreased and service load increased significantly. In general, the
beneficial effect of partial steel fiber reinforcement in service load,
deflection and crack width is approximately similar to the benefi-
cial effect of full steel fiber reinforcement. In addition, the cost of
the partial steel fiber reinforcement is 40% lower (the optimum
thickness of FRHSC layer is 0.57 times of the total depth of the
beam) than the cost of the full steel fiber reinforcement which is
a considerable cost saving. Therefore, at a certain steel fiber vol-
ume fraction, partial steel fiber reinforcement is recommended
over full steel fiber reinforcement.
The amounts of steel fibers in B2A2 and B4P2-3 are approxi-
mately the same. Compared with B2A2, the service load of B4P2-3
increased by 9.53%, the deflection and maximum crack width of
B4P2-3 decreased by 15.07% and 24.88%, respectively. Although
the ductility of the specimens partially reinforced with steel fibers
is lower, the ductility indices DFs of the specimens partially rein-
forced with steel fibers exceeded the minimum requirement [13]
of 4. Therefore, at a certain amount of steel fiber, the combination
of high fiber volume and partial steel fiber reinforcement is recom-
mended over the combination of low fiber volume and full steel fiber
reinforcement. However, for the structures with the requirements of
higher ductility, full steel fiber reinforcement may be required.
4. Theoretical study on service deflection
Due to the low modulus of elasticity of the FRP bars, the design
of concrete beams reinforced with FRP bars is typically governed
by the serviceability limit state requirements [17]. Therefore, a cal-
culation method for predicting the deflection of concrete beams
reinforced with FRP bars is needed. The effective moment of inertia
method is used to predict the deformation of the bending member
(i.e. beams) in ACI 440.1R-15 [18]. ACI 440.1R-15 [18] suggests that
the effective moment of inertia can be calculated by Eq. (1).
Icr
Ie ¼
2
 6 Ig Where M a P M cr ð1Þ
1c M cr
Ma
1  IIcrg
where, Ie is the effective moment of inertia, Icr is the moment of
inertia of cracked section, Ig is the moment of inertia of gross sec-
tion, Mcr is the cracking moment, Ma is the applied moment, c is a
factor (c = 1.72–0.72(Mcr/Ma)) which is related to the loading and
boundary conditions.
ACI 440.1R-15 [18] assumes that before the cracking of con-
crete, the section is homogeneous and the contribution of rein-
forcement to the moment of inertia of the gross section is
negligible. The moment of inertia of the gross section can be calcu-
lated by Eq. (2).
3
bh
Ig ¼ ð2Þ
12
where, b is the width of the beam, h is the total depth of the beam.
Based on the assumption that the concrete contribution in the
tension zone can be neglected after the cracking of concrete, ACI
440.1R-15 [18] suggests that the moment of inertia of cracked sec-
tion can be calculated by Eqs. (3) and (4).
b 3 3 2 2
Icr ¼ d k þ nf Af d ð1  kÞ ð3Þ
3
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
k¼ 2qnf þ ðqnf Þ2  qnf ð4Þ
where, d is the distance from the extreme compression fiber to the
centroid of tensile reinforcement, nf is the modulus ratio between
the FRP bar Ef and the plain concrete Ec, Af is the area of tensile rein-
forcement, q is the reinforcement ratio.
 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597 595

For the conventional FRP bar reinforced concrete beams, the The influence of steel fiber on the property of concrete is depen-
concrete contribution in the tension zone is usually neglected after dent on the distribution and orientation of steel fibers. The distri-
cracking. However, in the case of FRHSC beams reinforced with FRP bution of steel fibers can be described by nonuniformity
bars, as the tensile stress can be carried by the steel fibers in coefficient of steel fiber distribution gv. The orientation of steel
cracked section, the contribution of steel fibers in the tensile zone fibers can be described by steel fiber orientation coefficient g0.
must be considered. Therefore, prediction of the service load Zhang [19] reported that the total area of steel fibers in the section
deflections of FRHSC beams reinforced with FRP bars by the equa- can be calculated by Eq. (7).
tions suggested in ACI 440.1R-15 [18] may not be appropriate.
Asf ¼ g0 gv bhqsf ¼ gbhqsf ð7Þ
4.1. Moment of inertia of fully FRHSC beams where, g is the effective coefficient of steel fiber and qsf is the steel
fiber volume fraction.
Since the distance of the centroid of each single steel fiber from Gao [20] reported that for hooked steel fiber, the effective coef-
the neutral axis of the beam is unknown, the area and moment of ficient of steel fiber is between 0.16–0.33. The effective coefficient
inertia of a single steel fiber cannot be calculated. To obtain the of steel fiber g is taken as 0.16 herein.
area and moment of inertia of steel fibers, the steel fibers can be The moment of inertia of gross section of fully FRHSC beams can
considered as a whole. In the fully FRHSC beams, the average dis- be calculated by Eq. (8).
bh 3 i
tance of the steel fibers in compression zone from the neutral axis
3 2
is considered 0.5 times of the depth of compression zone and the Ig ¼ x0 þ ðh  x0 Þ þ ðnf  1ÞAf ðd  x0 Þ
3
average distance of steel fibers in tension zone from the neutral
nsf  1 h 3 3
i
axis is considered 0.5 times of the depth of tension zone. The con- þ Asf x0 þ ðh  x0 Þ ð8Þ
ceptual diagram of gross and cracked sections of FRHSC beams is 3h
shown in Fig. 8. The cracked and transformed cracked sections of fully FRHSC
The gross section and transformed uncracked section of fully beams are shown in Fig.8(e) and (f). There is a significant difference
FRHSC beams are shown in Fig.8(a) and (b). According to the con- between conventional and steel fiber reinforced concrete beams.
dition that the moments of area of compression zone and tension According to the condition that the moments of areas of compres-
zone are equal, Eqs. (5) and (6) can be derived. sion and tension zone are equal, Eqs (9) and (10) can be derived.
2
1 2 nsf  1 x20 1 2 1 2 nsf  1 x2cr ðh  xcr Þ
bx þ Asf ¼ bðh  x0 Þ þ ðnf  1Þ Af ðd  x0 Þ bx þ Asf ¼ nf Af ðd  xcr Þ þ nsf Asf ð9Þ
2 0 2 h 2 2 cr 2 h 2h
2
ðh  x0 Þ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi


þ ðnsf  1Þ Asf ð5Þ
2h ðnf Af þ nsf Asf Þ þ ðnf Af þ nsf Asf Þ2 þ 2 b  Ahsf ðnf Af d þ n2sf Asf hÞ
xcr ¼
1
bhþ ðnf  1ÞAf d þ nsf21 Asf h
2
b  Ahsf
x0 ¼ 2
ð6Þ
bh þ ðnf  1ÞAf þ ðnsf  1ÞAsf ð10Þ
where, x0 is the depth of compression zone in gross section, Asf is where, xcr is the depth of compression zone in cracked section.
the total area of steel fibers in section, and nsf is the modulus ratio The moment of inertia of cracked section of fully FRHSC beams
between the steel fiber Ef and the plain concrete Ec. can be calculated by Eq. (11).

(a) Gross section (b) Transformed uncracked (c) Gross section (d) Transformed uncracked
of fully FRSHC section of fully FRSHC of partially FRSHC section of partially FRSHC

(e) Cracked section (f) Transformed cracked (g) Cracked section (h) Transformed cracked
of fully FRSHC section of fully FRSHC of partially FRSHC section of partially FRSHC
Fig. 8. Calculations for gross and cracked sections.
596 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597

Table 7
Comparison of experimental and theoretical deflections at service load.

Specimen Fsc (kN) De (mm) DACI (mm) Dt (mm) De/DACI De /Dt

B2P2-1 88.66 9.78 11.49 8.94 0.86 1.10
B2P2-2 98.40 8.95 12.17 9.04 0.74 0.99
B2P2-3 103.08 7.98 11.93 8.64 0.64 0.88
B2A2 100.00 7.20 16.10 11.63 0.60 0.84
B1P2-3 88.38 8.70 10.22 8.60 0.86 1.03
B3P2-3 110.00 8.19 12.71 8.12 0.65 1.01
B4P2-3 118.18 9.16 14.64 8.35 0.63 1.11
B2P1-3 94.29 8.73 13.33 9.25 0.75 1.08
B2P3-3 146.15 12.12 14.80 11.88 0.84 1.05
B2P4-3 160.00 11.93 13.83 11.53 0.87 1.04
Average 0.76 1.01
Standard deviation 0.12 0.09
Coefficient of variation 0.15 0.09

Note: De is the experimental service deflection, DACI is the theoretical service deflection calculated by ACI model, Dt is the theoretical service deflection calculated by the
model suggested by this paper.

b 3 2 nsf  1 nsf 3 The moment of inertia of cracked section of fully FRHSC beams
Icr ¼ x þ nf Af ðd  xcr Þ þ Asf x3cr þ Asf ðh  xcr Þ ð11Þ
3 cr 3h 3h can be calculated by Eq. (18).
 2
b 3 2 hsf nsf 2
4.2. Moment of inertia of partially FRHSC beams Icr ¼ xcr þ nf Af ðd  xcr Þ þ nsf Asf h  xcr  þ Asf hsf ð18Þ
3 2 12
Since there is no steel fiber in the compression zone of partially
FRHSC beams, the calculation method of moment of inertia and 4.3. Verification
depth of compression zone of partially FRHSC beams will be differ-
ent from the calculation method of fully FRHSC beams. The gross To verify the calculation method developed in this paper, the
section and transformed uncracked section of partially FRHSC service load deflections of tested beam specimens were calculated
beams are shown in Fig.8(c) and (d). According to the condition by Eq. (19) considering the second moments of area calculated by
that moments of areas of compression zone and tension zone are the recommendations in ACI 440.1R-15 [18] and the method devel-
equal, Eq. (12) and Eq. (13) can be derived. oped herein.
 
1 2 1 hsf PS
2
bx0 ¼ bðh  x0 Þ þ ðnf  1ÞAf ðd  x0 Þ þ ðnsf  1ÞAsf h  x0  D¼ ð3L2  4S2 Þ ð19Þ
2 2 2 48Ec Ie
ð12Þ where, D is the deflection, P is the applied load, L is the span of the

 beam and S is the shear span of the beam.
2
1
2
bh þ ðnf  1ÞAf d þ ðnsf  1ÞAsf h  h2sf The maximum allowable crack width of FRP bar reinforced con-
x0 ¼ ð13Þ crete members, ACI 440.1R-15 and CAN/CSA S806 recommend 0.5
bh þ ðnf  1ÞAf þ ðnsf  1ÞAsf
mm for exterior exposure and 0.7 mm for interior exposure. There-
where, hsf is the thickness of FRHSC layer. fore, the service load (Fsc) was defined as the load at the crack
For partially FRHSC beams, the total area of steel fibers in the width limit of 0.7 mm. The experimental and theoretical deflec-
section can be calculated by Eq. (14). tions at service load are shown in Table 7. It can be seen that the
effect of steel fiber was not considered in ACI 440.1R-15 [18] and
Asf ¼ gbhsf qsf ð14Þ
hence the deflections were significantly overestimated. The mid-
The moment of inertia of the gross section of partially FRHSC span deflections calculated by the method developed in this paper
beams can be calculated by Eq. (15). agree very well with the experimental results.
b 3 3 2
Ig ¼ ½x þ ðh  x0 Þ  þ ðnf  1ÞAf ðd  x0 Þ þ ðnsf 5. Conclusion
3 0
 2
hsf nsf  1 2
 1ÞAsf h  x0  þ Asf hsf ð15Þ This study experimentally investigated the influences of the
2 12 thickness of fiber-reinforced high-strength concrete (FRHSC) layer,
The cracked and transformed cracked section of partially FRHSC steel fiber volume fraction and FRP reinforcement ratio on the flex-
beams are shown in Fig. 8(g) and (h). It can be seen in Fig. 8 ural behavior of partially FRHSC beams reinforced with FRP bars.
(g) and (h) that in the cracked section, the concrete in tension zone Also, in order to predict the service load deflection of FRHSC beams
failed and the tensile stress is carried by steel fibers in FRHSC layer reinforced with FRP bars, a new calculation method for the
and FRP bars. According to the condition that moments of areas of moment of inertia was developed. Based on the experimental
compression zone and tension zone are equal, Eqs. (16) and (17) and theoretical results, the followings conclusions can be drawn.
can be derived. FRP reinforcement ratio significantly influenced the flexural
  capacity and failure mode of tested beam specimens. Increasing
1 2 hsf
bxcr ¼ nf Af ðd  xcr Þ þ nsf Asf h  xcr  ð16Þ FRP reinforcement ratio was found to be an effective way to
2 2 improve the flexural capacity of partially FRHSC beams reinforced
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi with FRP bars. The effects of the thickness of FRHSC layer and steel
h
iffi
ðnf Af þ nsf Asf Þ þ ðnf Af þ nsf Asf Þ2 þ 2b nf Af d þ nsf Asf h  h2sf fiber volume fraction on the flexural capacity were not highly sig-
xcr ¼ nificant. However, for fully FRHSC beams reinforced with FRP bars,
b the flexural capacity was significantly greater than that of partially
ð17Þ FRHSC beams. Also, the failure mode changed from concrete crush-
 H. Zhu et al. / Construction and Building Materials 161 (2018) 587–597
ing to FRP bar rupture in fully FRHSC beams. Increasing the thick-
ness of FRHSC layer, steel fiber volume fraction and FRP reinforce-
ment ratio resulted in a proportional decrease in the deflection,
crack width and crack deviation. For partially FRHSC beams rein-
forced with FRP bars, when the thickness of FRHSC layer was
0.57 times of the total depth of the beams, the steel fiber had the
same effect on restraining deflection and cracks as fully FRHSC
beams.
The addition of steel fibers in tension zone was effective and
economical for overcoming the large deflection and crack width
of FRP bar reinforced concrete beams.
Ductility of partially FRHSC beams reinforced with FRP bars
reduced with the increase of the thickness of FRHSC layer and steel
fiber volume fraction. It is necessary to add steel fibers in the full
depth of structures for the structures with high ductility
requirements.
The service load deflections of fully and partially FRHSC beams
reinforced with FRP bars were significantly overestimated by ACI
440.1R-15. A new calculation method for predicting the effective
moment of inertia of fully and partially FRHSC beams reinforced
with FRP bars was developed in this paper. The mid-span deflec-
tions calculated by the developed method agree well with the
experimental mid-span deflections.
Acknowledgements
The research presented in this paper was supported by The
National Natural Science Foundation of China (No. 51178433).
References
[1] C. Soriano, A. Alfantazi, Corrosion behavior of galvanized steel due to typical
soil organics, Constr. Build. Mater. 102 (2016) 904–912.
[2] N.T. Duong, T.T.X. Huang, A. Nicolay, Y. Paint, M. Olivier, Corrosion protection
of carbon steel by solvent free epoxy coating containing hydrotalcites
intercalated with different organic corrosion inhibitors, Process. Organic
Coat. 101 (2016) 331–341.
[3] S.H. Alsayed, Y.A. Al-Salloum, T.H. Almusallam, Performance of glass fiber
reinforced plastic bars as a reinforcing material for concrete structures,
Compos. Part B-Eng. 31 (6–7) (2000) 555–567.
597
[4] S. El-Gamal, B. Benmokrane, E. El-Salakawy, P. Cousin, A. Wiseman, Durability
and structural performance of carbon fibre reinforced polymer-reinforced
concrete parking garage slabs, Can. J. Civ. Eng. 36 (4) (2009) 617–627.
[5] A. Nanni, Flexural behavior and design of RC members using FRP
reinforcement, J. Struct. Eng. 119 (11) (1993) 3344–3359.
[6] D.Y. Yoo, N. Banthia, Y.S. Yoon, Predicting service deflection of ultra-high-
performance fiber-reinforced concrete beams reinforced with GFRP bars,
Compos. Part B-Eng. 99 (2016) 381–397.
[7] H.G. Harris, W. Somboonsong, F.K. Ko, New ductile hybrid FRP reinforcing bar
for concrete structures, ASCE J. Compos. Constr. 2 (1) (1998) 28–37.
[8] W. Qu, X. Zhang, H. Huang, Flexural behavior of concrete beams reinforced
with hybrid (GFRP and steel) bars, ASCE J. Compos. Constr. 13 (5) (2009) 350–
359.
[9] S.H. Alsayed, A.M. Alhozaimy, Ductility of concrete beams reinforced with FRP
bars and steel fiber, J. Compos. Mater. 33 (19) (1999) 1792–1806.
[10] W.K. Lee, D.C. Jansen, K.B. Berlin, L.E. Cohen, Flexural cracks in fiber-reinforced
concrete beams with fiber-reinforced polymer reinforcing bars, ACI Struct. J.
(2010) 321–329.
[11] S.K. Padmarajaiah, A. Ramaswamy, Flexural strength predictions of steel fiber
reinforced high-strength concrete in fully-partially prestressed beam
specimens, Cem. Concr. Compos. 26 (2004) 275–290.
[12] S.K. Padmarajaiah, A. Ramaswamy, Comparative study on flexural response of
full and partial depth fiber-reinforced high-strength concrete, J. Mater. Civ.
Eng. 14 (2) (2002) 130–136.
[13] H. Wang, A. Belarbi, Ductility characteristics of riber-reinforced-concrete
beams reinforced with FRP rebars, Constr. Build Mater. 25 (5) (2011) 2391–
2401.
[14] J.M. Yang, K.H. Min, H.O. Shin, S.Y. Yoon, Effect of steel and synthetic fibers on
flexural behavior of high-strength concrete beams reinforced with FRP bars,
Compos. Part B-Eng. 43 (2012) 1077–1086.
[15] P. Vijay, Rao V. Ganga, Bending behavior and deformability of glass fiber-
reinforced polymer reinforced concrete members, ACI Struct. J. 98 (6) (2001)
834–842.
[16] I.M. Metwally, Evaluation of existing model for predicting of flexural behavior
of GFRP-reinforced concrete members, HBRCJ, Hous Build Res Cent 5 (1) (2009)
46–58.
[17] M.S. Issa, I.M. Metwally, S.M. Elzeiny, Influence of fibers on flexural behavior
and ductility of concrete beams reinforced with GFRP rebars, Eng. Struct. 33
(5) (2011) 1754–1763.
[18] ACI Committee 440.1R-15.Guide for the design and structural of concrete
reinforced with Fiber-reinforced polymer (FRP) bars. Farmington Hills(MI):
American Concrete Institute, 2015 pp. 83.
[19] Zhang Ming, Fatigue Performance and Calculation Method of Steel Fiber
Reinforced High-Strength Concrete Beams [D], Zhengzhou University,
Zhengzhou.
[20] Gao Danying, Basic Theory of Steel Fibre Reinforced Concrete [M], Scientific
and Technical Documentation Press, Beijing, 1994.