Construction and Building Materials 183 (2018) 605–617

Contents lists available at ScienceDirect

Construction and Building Materials

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

Analytical investigation on the load-moment characteristics of GFRP bar

reinforced circular NSC and HSC columns
Hayder Alaa Hasan, M. Neaz Sheikh, Muhammad N.S. Hadi ⇑
School of Civil, Mining and Environmental Engineering, University of Wollongong, Australia

h i g h l i g h t s

Procedures of establishing analytical P  M interaction diagrams are proposed.

Behaviour of GFRP bar reinforced NSC (GFRP-NSC) columns is investigated.

Behaviour of GFRP bar reinforced HSC (GFRP-HSC) columns is investigated.

Axial load-bending moment interactions of GFRP-NSC columns are presented.

Axial load-bending moment interactions of GFRP-HSC columns are presented.

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

Article history: In this study, the efficiency of Glass Fibre Reinforced-Polymer (GFRP) bar reinforced normal strength
Received 4 December 2017 concrete (NSC) and high strength concrete (HSC) columns in sustaining axial and flexural loads was
Received in revised form 15 May 2018 analytically investigated. Experimental data from available literature were used as benchmarks for the
Accepted 5 June 2018
analytical investigations conducted in this study. In addition, a comprehensive parametric study was
carried out to investigate the effect of different parameters (i.e., compressive strength of concrete,
mechanical properties and reinforcement ratio of GFRP bars, and slenderness ratio of the columns) on
Keywords:
the performance of concrete columns reinforced with GFRP bars. It was observed that under concentric
Normal strength concrete (NSC)
High strength concrete (HSC)
axial load, the improvements in the axial load carrying capacity due to increasing GFRP longitudinal rein-
Columns forcement ratio were more pronounced in GFRP bar reinforced NSC columns than in GFRP bar reinforced
Glass Fibre-Reinforced Polymer (GFRP) bars HSC columns. It was also observed that HSC columns reinforced longitudinally with GFRP bars with small
Interaction diagrams longitudinal reinforcement ratio or low tensile modulus of elasticity might experience a tensile failure of
the GFRP bars located on the tension side of the column cross-sections, especially if the columns are
subjected to a high level of axial load eccentricity.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction bars are anisotropic and their compressive strength is compara-

The corrosion of steel reinforcement of concrete structures
located in harsh and aggressive environments is one of the main
causes of deterioration (loss in strength and serviceability) of the
reinforced concrete (RC) structures. Fibre-Reinforced Polymer
(FRP) reinforcement is a corrosion-resistant material. Hence, FRP
bars are considered as one of the viable alternatives to conven-
tional steel bars as reinforcement for RC members [1]. In addition
to the corrosion resistance, FRP bars possess other attractive
characteristics, such as high tensile strength-to-weight ratio and
non-electrical and non-magnetic conductivity [2]. However, FRP
⇑ Corresponding author.
E-mail addresses: hah966@uowmail.edu.au (H.A. Hasan), msheikh@uow.edu.au
(M.N. Sheikh), mhadi@uow.edu.au (M.N.S. Hadi).
tively smaller than their tensile strength [3]. Therefore, steel bars
cannot be directly replaced with FRP bars in RC members due to
the differences in the mechanical properties of the steel and the
FRP bars [4].
Several research studies were conducted in recent years to
investigate the effect of replacing steel bars with FRP bars on the
behaviour of concentrically and eccentrically loaded circular and
square normal strength concrete (NSC) columns [5–9]. It was
reported that the compressive strength of the FRP bar was about
30–77% of the tensile strength [10–12]. It was also reported that
FRP bar reinforced concrete columns sustained about 5–13% lower
axial load than concrete columns reinforced with the same amount
of steel bars [13,14]. Furthermore, it was reported that the longitu-
dinal FRP bars contributed to about 5–15% of the total axial load
carrying capacity of the FRP bar reinforced concrete columns, while

https://doi.org/10.1016/j.conbuildmat.2018.06.042
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
606 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617

Nomenclature

Ac area of the concrete in the compression region of the Pn;CCS30 the axial load carrying capacity of GFRP bar reinforced
column specimen cross-section 30 MPa concrete columns
Afi area of the GFRP longitudinal bars Pn;CCSi axial load carrying capacity of GFRP bar reinforced con-
Ag gross area (the area of the reinforced concrete core plus crete columns with concrete compressive strength of
the area of concrete cover) of the column specimen 40, 50, 60, 70 or 80 MPa
b1 parameter defines the height of the equivalent rectan- DP n;MPLR the increase in the axial load carrying capacity of GFRP
gular stress block (ERSB) bar reinforced concrete columns due to increasing ten-
bsi width of each concrete strip sile elastic modules of FRP bars
c neutral axis depth PHV axial load carrying capacity of concrete columns rein-
Cc concrete compression force in the compression region forced longitudinally with GFRP bars having tensile
of the column specimen cross-section elastic modules of 70 GPa
C ci concrete compression force in each individual concrete Pn;LV axial load carrying capacity of concrete columns rein-
strip located in the compression zone of the column forced longitudinally with GFRP bars having tensile
specimen cross-section elastic modules of 35 GPa
C cfi concrete compression force of the concrete area dis- DP n;LRR the increase in the axial load carrying capacity of GFRP
placed by the ith layer of GFRP longitudinal reinforce- bar reinforced concrete columns due to increasing the
ment GFRP longitudinal reinforcement
dci distance between the mid-height of the ith concrete Pn;LRR1% axial load carrying capacity of concrete columns with
strip to the extreme concrete compression fibre that GFRP longitudinal reinforcement ratio of 1%
has the ultimate concrete compressive strain Pn;LRRi the axial load carrying capacity of concrete columns
dfi distance between the extreme concrete compression with GFRP longitudinal reinforcement ratio of 2%, 3%
fibre of the column cross-section to the centre of the or 4%
ith layer of GFRP longitudinal reinforcement q concrete stress–strain curve fitting factor
e eccentricity of the applied axial load r radius of gyration of the column specimen
Ec elastic modulus of the concrete rc radius of the column specimen cross-section
Ef tensile elastic modulus of FRP bars t si depth of each concrete strip

fc concrete axial stress y distance between the centroid of the column specimen
f0 c compressive strength of concrete obtained from testing to the centroid of concrete in the compression region
concrete cylinders at age of 28 days column specimen cross-section
f ci concrete stress in each concrete strip a1 parameter defines the width of the equivalent rectangu-
f cfi concrete axial stress in the concrete area displaced by lar stress block (ERSB)
ith layer of GFRP longitudinal reinforcement af reduction factor that accounts for the difference be-
f0 co unconfined concrete strength tween the compressive and the tensile strengths of
F fi force in the ith layer of GFRP longitudinal reinforcement FRP bars
f fi stress in the ith layer of GFRP longitudinal reinforce- ec concrete axial strain
ment eci average strain in the ith concrete strip
f fu ultimate tensile strength of FRP bars ecfi concrete strain of the concrete area displaced by the ith
h diameter of the column specimen layer of GFRP longitudinal reinforcement
j factor that controls the slope of the ascending and the eco unconfined concrete strain
descending branches of the concrete stress–strain curve ecu ultimate concrete compressive strain
l length of the column specimen efi strain in the ith layer of GFRP longitudinal reinforce-
kl the effective length of the column specimen ment
m number of GFRP longitudinal bars h an angle used in expressing the area of the concrete in
M bending moment the compression region of the column cross-section
M ci bending moment of each individual concrete strip £mid the curvature at mid-height of the column
M cfi bending moment for the concrete area displaced by the Dmid the deflection at mid-height of the column
ith layer of GFRP longitudinal reinforcement qf GFRP longitudinal reinforcement ratio
P axial load qs steel longitudinal reinforcement ratio
DP n;CCS the increase in the axial load carrying capacity of GFRP
bar reinforced concrete columns due to increasing the
concrete compressive strength

the contribution of the same amount of longitudinal steel bars ran-
ged between 12 and 16% of the total axial load carrying capacity of
steel bar reinforced concrete columns [5,6,15]. In addition, it was
observed that under eccentric axial loads, GFRP bar reinforced
NSC columns generally exhibited a slightly lower stiffness in the
ascending part of the axial load-axial deformation curves than
the steel bar reinforced NSC columns due to the lower modulus
of elasticity of the GFRP bars compared to the steel bars [16,17].
It was also observed that the eccentrically loaded GFRP bar rein-
forced NSC columns failed due to the crushing of concrete in the
compression side [16,17]. The spacing of the horizontal cracks in
the tension side of the eccentrically loaded GFRP bar reinforced
NSC columns depends on the pitch of the GFRP helices [16,17].
Research studies on the behaviour of FRP bar reinforced high
strength concrete (HSC) columns (especially columns subjected
to eccentric axial loads) are limited. Hales et al. [18] observed that
the failure of GFRP bar reinforced HSC columns tested under eccen-
tric axial loads with small eccentricities was due to the crushing of
the concrete accompanied by the compressive rupture of longitu-
dinal GFRP bars and the tensile rupture of the GFRP helices. It
was also observed that the hoop strain in the GFRP helices at
failure was significantly lower in HSC columns than in NSC
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617
columns. Hadhood et al. [19] found that GFRP longitudinal bars
located in the compression region of GFRP bar reinforced HSC col-
umns tested under eccentric axial load with large eccentricities
were able to sustain the applied eccentric axial load even in the
post-peak stage of the axial load-axial deformation behaviour. It
was also observed that increasing the GFRP longitudinal reinforce-
ment ratio from 2.2% to 3.3% did not significantly influence the
axial load-axial deformation behaviour of the HSC columns tested
under concentric axial load and slightly improved the axial load-
axial deformation behaviour of the HSC columns tested under
eccentric axial load. Hadi et al. [20] observed that the direct
replacement of the longitudinal and transverse steel reinforcement
with the same amount of GFRP reinforcement resulted in a slight
reduction in the axial load carrying capacity of the concentrically
loaded HSC columns. However, Hasan et al. [21] reported that
GFRP bar reinforced HSC columns tested under 25 and 50 mm
eccentric axial loads experienced about 10% and 12% lower axial
load carrying capacity compared to the axial load carrying capacity
of HSC columns reinforced longitudinally with similar amount of
steel bars.
The behaviour of GFRP bar reinforced concrete columns in gen-
eral, and HSC columns, in particular, have not been extensively
investigated in the available literature. Also, the effects of different
parameters (i.e., reinforcement ratio and the mechanical properties
of GFRP bars, compressive strength of the concrete and the slen-
derness ratio of the column) on the performance of GFRP bar rein-
forced NSC and HSC columns need further investigations. In
addition, there is a lack of design guidelines for FRP bar reinforced
NSC and HSC columns. The CSA S806-12 [22] and ACI 440.1R-15
[23] have not provided design recommendations for the design of
GFRP bar reinforced concrete columns. In this study, the analytical
methods and design recommendations adopted for conventional
steel bar reinforced concrete columns were used to assess the
behaviour of GFRP bar reinforced circular NSC and HSC columns
under axial and flexural loads. Also, a detailed parametric study
was conducted to further investigate the effect of different param-
eters (i.e., compressive strength of concrete, mechanical properties
Axial load
Bending moment
Fig. 1. Schematic of the analytical axial load-bending moment interaction diagram.
607
and reinforcement ratio of GFRP bars, and slenderness ratio of the
columns) on the behaviour of GFRP bar reinforced circular NSC and
HSC columns. Experimental data from the available literature were
used as a benchmark for the analytical investigations carried out in
this study. The findings of the analytical investigations will con-
tribute to the development of guidelines for the analysis and
design of GFRP bar reinforced concrete columns under different
loading conditions.
2. Analytical considerations
The equivalent rectangular stress block (ERSB) method and the
layer-by-layer integration method were commonly used to estab-
lish axial load-bending moment (P  M) interaction diagrams for
steel bar reinforced concrete columns. Given the lack of the exper-
imental research on FRP bar reinforced concrete columns, the fea-
sibility of the ERSB method and the layer-by-layer integration
method in investigating the axial and the flexural response of GFRP
bar reinforced concrete columns has not been adequately assessed.
In this study, the two analytical methods (ERSB method and the
layer-by-layer integration method) were used to predict the
P  M interaction diagrams for GFRP bar reinforced circular NSC
and HSC columns. In both methods, the analytical P  M interac-
tion diagrams were established based on eight points (Point A to
Point H), as shown in Fig. 1.
To analytically compute the axial load and the corresponding
bending moment capacities at Points A-H, eight different neutral
axis depths were considered based on the axial load eccentricities
expressed as ratios of the diameter of the columns (e=h ¼ 0, 0.06,
0.12, 0.25, 0.5, 0.75, 1 and 1, where e is the axial load eccentricity
and h is the diameter of the column). Point A (e=h ¼ 0Þ represents
the axial load carrying capacity of columns subjected to concentric
axial load with zero eccentricity. Points B-G represent the axial
load and the corresponding bending moment capacities of columns
subjected to combined axial load and bending moment. Point H
(e=h ¼ 1Þ represents the pure bending moment capacity of the
column.
608 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617

0

The analytical P  M interaction diagrams of the GFRP bar rein- f c  30 0
forced concrete columns were developed based on several assump-
b1 ¼ 0:85  0:05 P 0:65 for f c > 30 MPa ð1Þ
7
tions: 1) The strength of the concrete in tension is ignored, 2) The
For an assumed neutral axis depth (assumed e/h ratio), the
contribution of the GFRP longitudinal bars in compression is con-
strain in the ith layer of GFRP longitudinal reinforcement, efi , can
sidered, 3) Plane sections orthogonal to the bending axis remain
be computed using Eq. (2).
plane after bending, 4) The axial strain in the concrete, ec , and
 
the axial strain in GFRP reinforcing bars, ef , are equal at any point c  dfi
and 5) The GFRP bars are fully bonded to the surrounding concrete.
efi ¼ ecu ð2Þ
c
Hence, the bond-slip response between the GFRP bars and the con-
crete was ignored. The bond-slip response between the GFRP bars where ecu is the ultimate concrete compressive strain and dfi is the
and the concrete is part of future research studies by the authors. distance between the extreme compression fibre of the column
The subsequent sections describe the analytical methods adopted cross-section to the centre of the ith layer of GFRP longitudinal rein-
to derive the P  M interaction diagrams of GFRP bar reinforced forcement (Fig. 2a). In ACI 318-14 [24], the ultimate concrete com-
concrete columns. pressive strain, ecu , is recommended to be considered as 0.003.
The force in the ith layer of GFRP longitudinal reinforcement Ffi
and the concrete compression force in the compression region of
2.1. Equivalent rectangular stress block (ERSB) method
the column cross-section C c can be computed using Eqs. (3)–(6).
The equivalent rectangular stress block (ERSB) method is pre- F fi ¼ efi Ef Afi ð3Þ
sented in different building design codes, including ACI 318-14
0
[24]. The ERSB of ACI 318-14 [24] was developed based on data C c ¼ a1 f c Ac ð4Þ
obtained from testing NSC columns [25]. It was reported that the
ERSB of ACI 318 is not applicable for HSC columns, especially for
Ac ¼ ðh  sin h cos hÞr 2c ð5Þ
HSC columns subjected to concentric axial load [26–28]. This is
because the ERSB of ACI 318 may overestimate the strength of  
r c  b1 c
HSC columns and may lead to an unsafe design. These observa- h ¼ cos1 ð6Þ
rc
tions, however, were reported for HSC columns reinforced with
conventional steel bars. In this study, the feasibility of using the where Afi is the area of the GFRP longitudinal bars; Ac is the area of
ERSB of ACI 318-14 [24] in designing NSC and HSC columns rein- the concrete in the compression region of the column cross-section;
forced with GFRP bars was investigated. and rc is the radius of the column cross-section (equal to h=2), as
Two parameters a1 and b1 are usually used to define the ERSB in shown in Fig. 2a. It is worth mentioning that the tensile strains,
the ACI 318-14 [24]. The parameter a1 is recommended to have a stresses and forces are considered as negative and the compressive
0
constant value of 0.85 for any f c . The parameter b1 is recommended strains, stresses and forces are considered as positive.
0 0
to have a value of 0.85 for f c less than or equal to 30 MPa. For f c The axial load carrying capacity, Pn, represents the summation
greater than 30 MPa, the value of b1 is calculated using Eq. (1). of the axial forces in the concrete and GFRP bars (Eq. (7)). The

Fig. 2. Strain and stress profiles of GFRP bar reinforced concrete column: (a) Equivalent rectangular stress block method and (b) layer-by layer integration method.
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617 609

qffiffiffiffi
bending moment capacity, Mn, is computed by taking moment of 0
Ec ¼ 3320 f c þ 6900 ðMPaÞ ð15Þ
the internal forces at the centroid of the column (Eq. (8)).
The cross-section of the column was divided into n number of
X
m
small concrete strips. In this study, the depth of each concrete strip,
Pn ¼ C c þ F fi ð7Þ
tsi , was taken as 1 mm, which is small enough to obtain reasonably
i¼1
accurate results. The average width of each concrete strip, bsi , was
determined using Eq. (16).
X
m

 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
þ
Mn ¼ C c y F fi r c  dfi ð8Þ  
2
1
i¼1 bsi ¼ 2 r2c  r c  i  t si ð16Þ
2
where m is the number of the GFRP longitudinal bars. The y repre-
sents the distance between the centroid of the column specimen to For an assumed neutral axis depth, c (assumed e/h ratio), the
the centroid of concrete in the compression region. The y can be average strain in the ith concrete strip, eci ; was determined as:
computed using Eq. (9).  
c  dci
eci ¼ ecu ð17Þ
3
c
2r c ðsin hÞ
¼
y ð9Þ  
3ðh  sin h cos hÞ 1
dci ¼ i t si ð18Þ
2
2.2. Layer-by-layer integration method
where dci is the distance between the mid-height of the ith concrete
strip to the extreme compression fibre. For a given eci , the concrete
In the layer-by-layer integration method, the contribution of
stress in each concrete strip, f ci , can then be calculated by substitut-
the individual constituents of GFRP bar reinforced circular concrete
ing the values of eci into Eq. (11). Eventually, the concrete compres-
columns (concrete, GFRP bars and the area of concrete replaced by
sive force in each individual concrete strip located in the
the GFRP bars) were computed separately as follow:
compression zone of the column cross-section, C ci , can be computed
using Eq. (19):
2.2.1. Compressive force of concrete
The unconfined concrete stress-strain model proposed by C ci ¼ f ci bsi t si ð19Þ
Thorenfeldt et al. [29] was used to determine the axial stress in
In addition, the bending moment of each individual concrete
the concrete. Based on the unconfined concrete stress-strain model
strip, M ci , can be determined by taking moment about the centre-
of Thorenfeldt et al. [29], the stresses in concrete are computed
line of the column cross-section using Eq. (20).
using Eq. (10).
 

  1
0 ec Mci ¼ C ci rc  i  tsi ð20Þ
f co q
eco 2
fc ¼  jq ð10Þ
ec
q1þ eco
2.2.2. Reinforcement tension and compression forces
For assumed neutral axis depth, c (e=h ratio) in the section
where f c is the axial stress in the concrete at any axial strain, ec . The
above, the strain in the ith layer of GFRP longitudinal reinforce-
f0 co represents the unconfined concrete strength of the concrete
ment, efi , can be determined using Eq. (2). Once the strain in the
which is equal to 85% of the compressive strength of the concrete
GFRP longitudinal reinforcement is obtained, the stress in the ith
at age of 28 days, f0 c. The eco is the unconfined concrete strain corre-
layer of GFRP longitudinal reinforcement can then be calculated
sponding to the unconfined compressive strength of concrete. The q
using Eq. (21).
is the concrete stress-strain curve fitting factor, while k is a factor
that controls the slope of the ascending and the descending f fi ¼ efi Ef ð21Þ
branches of the concrete stress-strain curve. The values of eco ; q
and j were determined using Eqs. (11)–(13) according to Collins Eventually, the tension or compression forces and the corre-
and Mitchell [30]: sponding bending moment of the ith layer of GFRP longitudinal
reinforcement at the centreline of the column cross-section can
0  
fc q be determined using Eq. (22) and Eq. (23), respectively.
eco ¼ ð11Þ
Ec q  1 F fi ¼ efi Ef Afi ð22Þ
 0 

fc Mfi ¼ F fi r c  dfi ð23Þ
q ¼ 0:8 þ ð12Þ
17
For ðec =eco Þ 6 1.0, 2.2.3. Compressive force of concrete replaced by GFRP longitudinal
j ¼ 1:0 ð13aÞ reinforcement
In order to avoid any overestimation in the axial load carrying
For ðec =eco Þ > 1.0,
capacity and the corresponding bending moment of the GFRP bar
 0 
fc reinforced concrete columns, the contribution of the concrete area
j ¼ 0:67 þ P 1:0 ð13bÞ replaced by the longitudinal reinforcement must be subtracted,
62
especially when a large amount of longitudinal reinforcement is
The Ec in Eq. (11) represents the elastic modulus of the concrete, used. Since the axial strain in the concrete, ec , and the axial strain
which was computed using Eq. (14) for normal strength concrete in GFRP reinforcing bars, ef , were assumed to be equal at any point,
[24] and Eq. (15) for high strength concrete [31]. the concrete strain, ecfi , of the concrete area replaced by the ith
qffiffiffiffi
0 layer of GFRP longitudinal reinforcement can be computed using
Ec ¼ 4730 f c ðMPaÞ ð14Þ
Eq. (24):
610 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617

   
X
c  dfi X
n
1 m

 X m


ecfi ¼ efi ¼ ecu ð24Þ Mn ¼ C ci r c  i  t si þ F fi rc  dfi  C cfi r c  dfi
c 2
i¼1 i¼1 i¼1

For a given ecfi , the concrete axial stress, fcfi in the concrete area
replaced by the ith layer of GFRP longitudinal reinforcement can
then be calculated by substituting ecfi into Eq. (11). The compres-
sive force of the concrete, C cfi , and the corresponding bending
moment, M cfi , for the concrete area replaced by the ith layer of
GFRP longitudinal reinforcement can then be determined using
Eqs. (25) and (26), respectively.
C cfi ¼ f ci Afi


M cfi ¼ C cfi r c  dfi
It should be noted that the compressive force of the concrete
and the corresponding bending moment of the concrete area
replaced by the ith layer of GFRP longitudinal reinforcement were
accounted for concrete in the compression zone only. It should be
also noted that the tensile strains, stresses and forces are consid-
ered as negative and the compressive strains, stresses and forces
are considered as positive.
Finally, the resultant axial load carrying capacity, Pn, and the
bending moment, Mn, of the entire GFRP bar reinforced circular con-
crete column can be calculated by the summation of the forces and
the moment of the individual component of the columns (concrete,
longitudinal reinforcement and the concrete area replaced by the
longitudinal reinforcement) as expressed in Eqs. (27) and (28),
respectively.
X
n X
m X
m
Pn ¼ C ci þ F fi  C cfi
i¼1 i¼1 i¼1
ð28Þ
3. Experimental benchmark
The validity and the reliability of the analytical methods
adopted in this study to investigate the axial and the flexural
response of GFRP bar reinforced concrete columns were verified
ð25Þ with experimental data obtained from research studies found in
the available literature, which are described below.
ð26Þ
(i) Hadi et al. [16] and Karim et al. [17] experimentally tested
GFRP bar reinforced circular NSC specimens under concen-
tric axial load, eccentric axial load with different eccentric-
ities and four-point bending. The specimens were 205 mm
in diameter and 800 mm in height and were cast with NSC
having a compressive strength of 37 MPa. The specimens
tested in Hadi et al. [16] and Karim et al. [17] were rein-
forced longitudinally with six #4 (nominal diameter = 12.7
mm) GFRP bars. The specimens tested in Hadi et al. [16]
were reinforced transversely with #3 (nominal diameter
= 9.5 mm) GFRP helices having a pitch of 60 mm, whereas
the specimens tested in Karim et al. [17] were reinforced
transversely with #3 GFRP helices having a pitch of 30
mm.
(ii) Hadhood et al. [19] experimentally tested GFRP bar rein-
forced circular HSC specimens under concentric axial load
ð27Þ and eccentric axial load with different eccentricities. The
specimens tested in Hadhood et al. [19] were 305 mm in

3000 3000
Hadi et al. [16] Karim et al. [17]
2500 ERSB method (ACI 318-14 [24]) 2500 ERSB method (ACI 318-14 [24])
Layer-by-layer integration method Layer-by-layer integration method
2000 2000
Axial load (kN)
Axial load (kN)

1500 1500

1000 1000

500 500

a 0
b
0
0 25 50 75 100 0 25 50 75 100
Bending moment (kN.m) Bending moment (kN.m)

6000 6000
Hadhood et al. [19] Hasan et al. [21]
5000 ERSB method (ACI 318-14 [24]) 5000 ERSB method (ACI 318-14 [24])
Layer-by-layer integration method Layer-by-layer integration method
4000 4000
Axial load (kN)

Axial load (kN)

3000 3000

2000 2000

1000 1000
c d
0 0
0 50 100 150 200 0 50 100 150 200
Bending moment (kN.m) Bending moment (kN.m)

Fig. 3. Analytical P  M interaction diagrams versus experimental P  M interaction diagrams of GFRP bar reinforced concrete column specimens tested in: (a) Hadi et al. [16],
(b) Karim et al. [17], (c) Hadhood et al. [19] and (d) Hasan et al. [21].
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617
diameter and 1500 mm in height and were cast with HSC
having a compressive strength of 70.2 MPa. The specimens
tested in Hadhood et al. [19] were reinforced longitudinally
with eight #5 (nominal diameter = 15.9 mm) GFRP bars and
transversely with #3 GFRP helices having a pitch of 80 mm.
(iii) Hasan et al. [21] experimentally tested GFRP bar reinforced
circular HSC specimens under concentric axial load, eccen-
tric axial load with different eccentricities and under four-
point bending. The specimens tested in Hasan et al. [21]
were 210 mm in diameter and 800 mm in height and were
cast with HSC having a compressive strength of 85 MPa.
The specimens tested in Hasan et al. [21] were reinforced
longitudinally with six #4 GFRP bars and transversely with
#3 GFRP helices having a pitch of 60 mm.
Fig. 3 compares the experiment P–M interaction diagrams of
the specimens tested in Hadi et al. [16], Karim et al. [17], Had-
hood et al. [19] and Hasan et al. [21] with the analytical P–M
interaction diagrams generated using the ERSB method and the
layer-by-layer integration method described in the previous sec-
tions. The comparison between the analytical interaction dia-
grams and the experimental results revealed that ACI 318-14
[24] stress block parameters provided reasonable predictions for
the strength of the GFRP bar reinforced NSC specimens under dif-
ferent loading conditions. However, it was observed that the P–M
interaction diagrams obtained using the ACI 318-14 [24] stress
block parameters slightly overestimated the strength of GFRP
bar reinforced HSC specimens in the part of diagram that repre-
sented specimens subjected to axial load with small eccentricities
(Fig. 3c and d). Similar observation was reported in Hadhood et al.
[19]. On the other hand, the P–M interaction diagrams plotted
using the layer-by-layer integration method either coincided with
the experimental results or were on the safe side for GFRP bar
reinforced NSC and HSC specimens. It was also observed that
under pure flexural load, the differences between the interaction
diagrams plotted using the ACI 318-14 [24] equivalent rectangu-
lar stress block method and the layer-by-layer integration method
are negligible.
Since the layer-by-layer integration method provided safer pre-
diction for the axial load and the bending moment capacities of
GFRP bar reinforced columns compared to the ACI 318-14 [24]
equivalent rectangular stress block method, the layer-by-layer
integration method was used in investigating the effect of different
parameters on the performance of GFRP bar reinforced concrete
columns under axial and flexural loads. The following section dis-
cusses the outcomes of the parametric study in detail.
4. Parametric study
A parametric study was conducted to investigate the effect of
different parameters on the performance of GFRP bar reinforced
NSC and HSC columns under axial and flexural loads. The parame-
ters considered in the parametric study are the compressive
strength of concrete, the reinforcement ratio and the mechanical
properties of GFRP longitudinal bars and the slenderness ratio of
the columns.
Two reference columns were considered in the parametric
study: GFRP bar reinforced NSC and GFRP bar reinforced HSC col-
umns. Also, two compressive strengths were considered: normal
0
strength concrete with f c of 40 MPa for the reference GFRP bar
0
reinforced NSC column and high strength concrete with f c of 80
MPa for the reference GFRP bar reinforced HSC column. The refer-
ence columns were considered to be reinforced longitudinally with
GFRP bars with longitudinal reinforcement ratio of 2% and trans-
611
versely with GFRP helices with transverse reinforcement ratio of
3%. The tensile modulus of elasticity of the longitudinal and trans-
verse GFRP reinforcement was assumed to be equal to 67.8 GPa
and 76.8 GPa, respectively. Each analytical investigation performed
in this parametric study represents a modification of either the
compressive strength of the concrete, the material properties of
the GFRP reinforcement or the column dimensions from those in
the reference columns. Noting that for each new column reported
in the parametric study, only one parameter was changed from
those used in the reference columns. To investigate the effect of
the compressive strength of the concrete, the axial load Pn and
the corresponding bending moment M n presented in the paramet-
ric study were normalised as:
Pn
P ¼ ð29Þ
Ag
Mn
M ¼ ð30Þ
Ag h
where Ag is the gross cross-sectional area (the area of the reinforced
concrete core plus the area of concrete cover) of the column and h is
the diameter of the column which is equal to 2r c .
4.1. Effect of GFRP longitudinal bars in the compression zone of the
columns
Since the behaviour of the FRP bars under concentric axial load
is complicated, the CSA S806-12 [22] recommended neglecting the
contribution of the FRP bars when used as longitudinal reinforce-
ment in concrete columns. Also, the ACI 440.1R-15 [23] provided
no guidelines for using FRP bars as longitudinal reinforcement in
concrete columns. In order to investigate the effect of the GFRP
longitudinal bars on the strength capacity of the GFRP bar
reinforced circular concrete columns, three analytical P   M 
interaction diagrams were generated and compared with the
experimental P  M  integration diagrams of the GFRP bar rein-
forced concrete column specimens tested in Hadi et al. [16], Karim
et al. [17], Hadhood et al. [19] and Hasan et al. [21], as shown in
Fig. 4.
In the first analytical P   M  interaction diagram (Analytical 1),
the contribution of the GFRP longitudinal bars located in the com-
pression region of the column cross-section was included. In the
second analytical P   M  interaction diagram (Analytical 2), the
contribution of the GFRP longitudinal bars located in the compres-
sion region of the column cross-section was neglected and the area
of the GFRP bars was substituted by an equivalent area of concrete,
as suggested in Zadeh and Nanni [32]. Replacing the GFRP longitu-
dinal bars by the same amount of concrete area might underesti-
mate the exact load sustained by the GFRP longitudinal bars
especially for columns tested under concentric axial load and
under eccentric axial load with small eccentricities. In order to
clarify this point, another P  M  interaction diagram (Analytical
3) was drawn, in which the GFRP bars located in the compression
region of the column cross-section were ignored and were not sub-
stituted by concrete area.
In comparison with the GFRP bar reinforced NSC column spec-
imens tested in Hadi et al. [16] and Karim et al. [17] and the GFRP
bar reinforced HSC column specimens tested in Hadhood et al. [19]
and Hasan et al. [21], it was observed that neglecting the contribu-
tion of the GFRP bars located in the compression region of the GFRP
bar reinforced NSC and HSC columns, as recommended in the CSA
S806-12 [22] and ACI 440.1R-15 [23], resulted in overly conserva-
tive predictions. It was also found that rational predictions were
612 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617

100 100
Hadi et al. [16] Karim et al. [17]
Analytical 1 Analytical 1
80 Analytical 2 80 Analytical 2
Analytical 3 Analytical 3

P* (N/mm2)
P* (N/mm2)

60 60

40 40

20 20

a b
0 0
0 2 4 6 8 10 0 2 4 6 8 10
M* (N/mm2) M* (N/mm2)

200 200
Hadhood et al. [19] Hasan et al. [21]
Analytical 1 Analytical 1
160 160 Analytical 2
Analytical 2 Analytical 3
Analytical 3
P* (N/mm2)

P* (N/mm2)
120 120

80 80

40 40

c d
0 0
0 5 10 15 20 0 5 10 15 20
M* (N/mm2) M* (N/mm2)

Fig. 4. Effect of GFRP longitudinal bar in the compression zone of GFRP bar reinforced concrete column specimens tested in: (a) Hadi et al. [16], (b) Karim et al. [17], (c)
Hadhood et al. [19] and (d) Hasan et al. [21].

obtained when the contribution of the GFRP bars located in the
compression region of the columns was taken into consideration.
4.2. Effect of the GFRP longitudinal reinforcement ratio
According to the ACI 318-14 [24], the longitudinal reinforce-
ment ratio in RC columns should not be less than 1% and should
not exceed 8%. However, a maximum longitudinal reinforcement
ratio of 4% was recommended in case the longitudinal bars of the
column are needed to be lap spliced, as the lap splice region will
have two times the longitudinal reinforcement ratio if all lap
splices are set at the same location [24]. Therefore, the effect of
four GFRP longitudinal reinforcement ratios (qf ¼ 1%, 2%, 3% and
4%) on the P   M  interaction diagrams of GFRP bar reinforced cir-
cular NSC and HSC columns were investigated in this study.
Fig. 5a and 5b present the effects of increasing qf on the P   M 
0
interaction diagrams of GFRP bar reinforced NSC (f c ¼ 40 MPa) and
0
HSC (f c ¼ 80 MPa) columns. Fig. 6a and b show the relationship
between the e=h ratio and the increase in the axial load carrying
capacity of GFRP bar reinforced NSC and HSC columns due to
increasing qf which was calculated as:
Pn;LRRi  Pn;LRR1%
DPn;LRR ð%Þ ¼
Pn;LRR1%
where P n;LRR1% represents the axial load carrying capacity of NSC and
HSC columns with GFRP longitudinal reinforcement ratio of 1% for
different e=h ratios and P n;LRRi represents the axial load carrying
capacity of NSC and HSC columns with GFRP longitudinal reinforce-
ment ratio of 2%, 3% or 4% for different e=h ratios.
Increasing qf led to an increase in the axial load carrying
capacity of both GFRP bar reinforced NSC and HSC columns
under different loading conditions. It was found that under con-
centric axial load ðe=h ¼ 0), the improvements in the axial load
carrying capacity due to increasing qf were more pronounced
in GFRP bar reinforced NSC columns. Increasing qf from 1% to
4% resulted in an increase of 10% and 4.5% in the axial load car-
rying capacity of the GFRP bar reinforced NSC and HSC columns,
respectively. This is because the contribution of the GFRP longi-
tudinal bars with respect to the maximum axial load sustained
by the NSC columns was greater than the contribution of the
GFRP longitudinal bars with respect to the maximum axial load
sustained by the HSC columns. This is attributed to the fact that
HSC columns sustained much greater axial loads than NSC col-
umns. However, as the load eccentricity increased (e=h
increased) the improvement in the axial load carrying capacity
due to increasing qf became approximately the same in both
GFRP bar reinforced NSC and HSC columns as the contribution
of GFRP longitudinal bars in HSC columns became close to the
contribution of GFRP longitudinal bars in NSC columns.
It was also observed that under a high level of loading eccentric-
ity (e/h > 0:8), reducing qf leads to high tensile strains in the GFRP
ð31Þ
longitudinal bars located on the tension side of the GFRP bar rein-
forced NSC and HSC columns. Hence, insufficient qf of less than 1%
might cause a tensile failure in the GFRP bars before the P  M 
interaction diagrams reach the pure flexural condition. A similar
observation was reported in Choo et al. [2] for GFRP bar reinforced
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617 613

100 50
= 40 MPa = 1% = 40 MPa a
= 2%
= 3%
80 40
= 4%

60 30
(N/mm2)

(%
40 20

= 2%
20 10
= 3%

a = 4%
0 0
0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 1.2

(N/mm2)

100 50
= 80 MPa = 1% = 80 MPa b
= 2%
= 3% 40
80
= 4%

60 30
(N/mm2)

(%

40 20

= 2%
20 10
= 3%

= 4%
0
b 0
0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 1.2
(N/mm2)

Fig. 5. Effect of GFRP longitudinal reinforcement ratio on the P   M interaction Fig. 6. The effect of increasing GFRP longitudinal reinforcement ratio qf on the
diagrams of GFRP bar reinforced concrete columns: (a) NSC columns and (b) HSC relationship between the e=h ratio and the increase in the nominal axial load
columns. carrying capacity of GFRP bar reinforced concrete columns: (a) NSC columns and (b)
HSC columns.

square NSC columns and in Hadhood et al. [33] for GFRP bar rein-
forced circular HSC columns
4.3. Effect of the mechanical properties of GFRP longitudinal bars
Based on the mechanical properties of the GFRP bars used in
calculating the contribution of the GFRP longitudinal bars in the
axial load carrying capacity of the GFRP bar reinforced concrete
columns, two approaches were adopted in the previous research
studies. In the first approach, the axial load sustained by the GFRP
longitudinal bars is computed using the ultimate tensile strength
of the GFRP bars, af f fu Af . In the second approach, which was con-
sidered in this study, the axial load sustained by the GFRP longitu-
dinal bars is computed using the tensile elastic modulus of the
GFRP bars, ef Ef Af .
According to ACI 440.1R-15 [23], the tensile modulus of elastic-
ity of GFRP bars ranges between 35 GPa and 51 GPa. However, the
specimens tested by Hasan et al. [21] were reinforced longitudi-
nally with GFRP bars having a tensile modulus of elasticity of
67.8 GPa. Therefore, the effect of the tensile modulus of elasticity
of GFRP bars was investigated in this study considering the lowest
value (LV) of tensile modulus of elasticity of 35 GPa reported in ACI
440.1R-15 [23] and a highest value (HV) of tensile modulus of elas-
ticity of 70 GPa, which is slightly greater than the value reported in
Hasan et al. [21]. The lowest and the highest values of Ef consid-
ered in this study may cover almost all the commercially available
GFRP bars.
Fig. 7a and b show the effects of changing the value of Ef on the
0
P  M  interaction diagrams of GFRP bar reinforced NSC (f c ¼ 40
0
MPa) and HSC (f c ¼ 80 MPa) columns, respectively. Fig. 8 shows
the relationship between the e=h ratio and the increase in the nom-
inal axial load carrying capacity of GFRP bar reinforced NSC and
HSC columns due to increasing Ef which was calculated using Eq.
(32):
Pn;HV  Pn;LV
DPn;MPLR ð%Þ ¼ ð32Þ
Pn;LV
where P n;LV represents the axial load carrying capacity of NSC and
HSC columns reinforced longitudinally with GFRP bars having Ef =
35 GPa for different e=h ratios and P HV represents the axial load car-
rying capacity of NSC and HSC columns reinforced longitudinally
with GFRP bars having Ef ¼ 70 GPa for different e=h ratios.
It was found that under concentric axial load, increasing Ef of
the longitudinal GFRP bars from 35 GPa to 70 GPa led to an
increase of about 5.4% and 3% in the total axial load carrying capac-
ity of GFRP bar reinforced NSC and HSC columns, respectively. It
was also found that under moderate levels of loading eccentricity
(0.3 < e/h < 0.7), increasing Ef of the GFRP longitudinal bars
resulted in an approximately similar increase (about 22%) in the
614 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617

100 100
= 40 MPa Ef = 35 GPa f'c =30 MPa
f'c =40 MPa
Ef = 70 GPa
80 80 f'c =50 MPa
f'c =60 MPa
f'c =70 MPa
(N/mm2)

(N/mm2)
60 60 f'c =80 MPa

40 40

20 20

a
0 0
0 2 4 6 8 10
0 2 4 6 8 10
(N/mm2) (N/mm2)

100 Fig. 9. Effect of the concrete compressive strength on the P   M  interaction

= 80 MPa diagrams of GFRP bar reinforced concrete columns.
Ef = 35 GPa

Ef = 70 GPa
80
that the impact of the modulus of elasticity of GFRP bars on the
axial load carrying capacity of the columns was more pronounced
in columns subjected to a high level of axial load eccentricity.
(N/mm2)

60
No bar rupture was observed in NSC and HSC columns rein-
forced longitudinally with GFRP bars having Ef of 35 GPa or 70
40
GPa and subjected to a high level of loading eccentricity
(e=h > 0:7). However, it was found that in the HSC specimen rein-
forced longitudinally with GFRP bars having Ef = 35 GPa and sub-
20 jected to pure flexural loads, the tensile strain in the outermost
tensile GFRP reinforcement was about 75% of the rupture strain.
Therefore, it is recommended not to reinforce HSC concrete col-
0
b umns longitudinally with GFRP bars having a low tensile modulus
0 2 4 6 8 10 of elasticity to avoid the tensile failure of the GFRP bars, especially
(N/mm2) if the columns are expected to be subjected to a high level of axial
load eccentricity.
Fig. 7. Effect of the mechanical properties of GFRP longitudinal bars on the P   M 
interaction diagrams of GFRP bar reinforced concrete columns: (a) NSC columns and
(b) HSC columns. 4.4. Effect of the concrete compressive strength

Fig. 9 shows the effects of increasing the compressive strength

50 0
of concrete f c on the P  M  interaction diagrams of GFRP bar rein-
f'c = 40 MPa
forced circular concrete columns. As initially expected, increasing
0
40
f'c = 80 MPa the f c led to an increase in the axial load carrying capacities and
the corresponding bending moments of the GFRP bar reinforced
concrete columns under different loading conditions (different
30 e=h ratios). Fig. 10 shows the relationship between the e=h ratio
(%

and the increase in the nominal axial load carrying capacity of

0
GFRP bar reinforced concrete columns due to increasing f c , which
20 was calculated using Eq. (33):

Pn;CCSi  Pn;CCS30
DPn;CCS ð%Þ ¼ ð33Þ
10 Pn;CCS30

0 reinforced 30 MPa concrete columns for different e=h ratios and
0 0.2 0.4 0.6 0.8 1 1.2
Fig. 8. The relationship between the e=h ratio and the increase in the nominal axial
load carrying capacity of GFRP bar reinforced NSC and HSC columns due to
increasing modulus of elasticity Ef of the longitudinal GFRP bars.
axial load carrying capacity of GFRP bar reinforced NSC and HSC
columns (Fig. 8). The aforementioned observations were consistent
with the observations reported in Hadhood et al. [33] who found
where P n;CCS30 represents the axial load carrying capacity of GFRP bar
Pn;CCSi represents the axial load carrying capacity of GFRP bar rein-
forced concrete columns with compressive strength concrete of
40, 50, 60, 70 or 80 MPa for different e=h ratios.
The increase in the axial load carrying capacity of the GFRP bar
0
reinforced concrete columns due to increasing f c was more pro-
nounced in columns subjected to a low level of axial load eccen-
tricity (e=h < 0:2Þ, where the compressive strength of the
concrete mostly dominates the total axial load carrying capacity
of the columns. A similar observation was reported in Hadhood
et al. [33]. It was also found that under pure concentric axial load
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617 615

250
= 40 MPa

= 50 MPa
200
= 60 MPa

= 70 MPa
150
= 80 MPa
(%
,

100
Δ

Deformed
50
shape Section A-A

Non-deformed
0 shape
0 0.2 0.4 0.6 0.8 1 1.2

Fig. 10. The relationship between the e=h ratio and the increase in the nominal
axial load carrying capacity of GFRP bar reinforced concrete columns due to
0
increasing concrete compressive strength, f c

0
(e=h < 0Þ, increasing f c from 30 MPa to 50 MPa resulted in an
increase of about 52% in the maximum axial load sustained by
0
the GFRP bar reinforced concrete columns. Increasing f c from 30
MPa to 80 MPa resulted in an increase of about 143% in the maxi-
mum axial load sustained by the GFRP bar reinforced concrete col-
0
umns. On the other hand, increasing f c from 30 MPa to 80 MPa Fig. 11. Schematic of a typical eccentrically loaded, pin ended reinforced concrete
resulted in an increase of about 61% in the maximum load sus- column.

tained by the GFRP bar reinforced concrete columns subjected to

high level of loading eccentricity (e=h ¼ 1Þ.
The compressive strength of concrete was found to have a smal-
ler effect on the pure bending moment capacity of the GFRP bar
reinforced concrete specimens compared to the effect of compres-
sive strength of concrete on the concentric axial load carrying
capacity of the columns, especially for GFRP bar reinforced HSC
Axial load

(Fig. 9). This is because the final failure of GFRP bar reinforced con-
crete specimens under pure flexural loads is governed by the ulti-
mate concrete compressive strain ecu at the extreme concrete
compression fibre of the specimen cross-section.

4.5. Effect of the column slenderness

Short
The slenderness ratio, kl/r is usually used in classifying various Column
Slender
columns into short, intermediate and slender columns and it rep- Column
resents the ratio between the effective length of the column, kl,
O
to the radius of gyration of the column, r. In order to investigate Bending moment
the effect of the column slenderness, it is often convenient to
establish a family of slender column interaction diagrams. Fig. 12. Typical P  M interaction diagrams for GFRP bar reinforced short and
slender concrete columns.
Fig. 11 illustrates a typical eccentrically loaded, pin-ended rein-
forced concrete column deformed laterally and sustained an addi-
tional bending moment due to the lateral deformation,D. For short and shown in Fig. 11. Hence, the deflection at the mid-height of
columns, the effect of the lateral deformation is usually neglected. the columns, Dmid , can be calculated as:
Therefore, the relationship between the applied axial load and the


corresponding bending moment experienced by the column is Dmid ¼ l=p2 £mid ð34Þ
almost linear ðM ¼ P:eÞ, as shown in Line O  A in Fig. 12. However,
as the length of the column increases, the lateral deformation £mid ¼ ecu =c ð35Þ
exhibited by the column increases nonlinearly resulting in a non-
linear amplified bending moment ðM ¼ P:e þ P:DÞ, refer to Line where £mid represents the curvature at mid-height of the column.
O  B1 in Fig. 12. The actual failure of a slender column occurs when the applied
In order to take the effect of the deflection of slender columns axial load and the corresponding amplified bending moment line
(second-order effect) into account, slender columns were consid- of the column intersect the P  M interaction diagram (Point B1
ered to bend in a single curvature. Accordingly, the deformed in Fig. 12). However, the design engineers might be interested in
shape of the slender columns can be assumed to be a half-sine expressing the failure of a reinforced concrete column in terms of
wave, as explained in Bazant et al. [34] and Jiang and Teng [35] the applied axial load and the corresponding bending moment at
616 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617
the ends of the column (Point A1 ). Changing the eccentricity of the
axial load applied at the ends of the slender column results in dif-
ferent points on the interaction diagram (A1 and A2 ), as shown in
Fig. 12. Consequently, the interaction diagram of a slender column
is presented by the curved line passing through Points A1 and A2 .
Such curves illustrate the axial loads and the maximum end bend-
ing moments causing the failure of the slender column.
Broms and Viest [36] and Pfrang and Siess [37] comprehen-
sively investigated the effects of several parameters on the axial
load carrying capacity of slender steel bar reinforced concrete col-
umns. Broms and Viest [36] and Pfrang and Siess [37] observed
that the most significant parameters affecting the axial load carry-
ing capacity and the behaviour of steel bar reinforced hinged slen-
der concrete columns were the slenderness ratio ðkl=rÞ; the
eccentricity of axial loads applied at the ends of column e1 and
e2 ; the ratio of the end loading eccentricities ðe1 =e2 Þ; the compres-
0
sive strength of the concrete f c ; and steel longitudinal reinforce-
ment ratio ðqs Þ. In this study, the behaviour of GFRP bar
reinforced hinged slender concrete columns was analytically
investigated considering the GFRP longitudinal reinforcement ratio
 
qf of 2%. Also, in this study, the loading eccentricities at the ends
of the column specimen e1 and e2 were assumed to be equal
(e1 =e2 ¼ 1Þ. The influence of the slenderness ratio ðkl=rÞ, concrete
0
compressive strength f c and the eccentricity of axial loads applied
at the ends of the column are presented graphically in Fig. 13. The
interaction diagrams in Fig. 13 are plotted in terms of the maxi-
100
= 40 MPa
80
P* (N/mm2)
60
40
20
0
0 2 4 6
M* (N/mm2)
100
= 80 MPa
80
P* (N/mm2)
60
40
20
0
0 2 4 6
M* (N/mm2)
Fig. 13. Effect of slenderness ratio kl=r on P  M interaction diagrams of GFRP bar
reinforced concrete columns: (a) NSC columns and (b) HSC columns.
mum axial loads and corresponding bending moments that can
be applied at the ends of GFRP bar reinforced concrete columns
with different slenderness ratios.
As expected, increasing the kl=r ratio led to a significant
reduction in the axial load carrying capacity and the correspond-
ing bending moments of the GFRP bar reinforced both NSC and
HSC columns under concentric and eccentric axial loads. Broms
and Viest [36] observed that steel bar reinforced concrete col-
umns with high concrete compressive strength were found to
be more affected by slenderness ratio of the column. In this
study, however, it was observed that GFRP bar reinforced NSC
0 0
(f c 40 MPa) and HSC (f c = 80 MPa) columns experienced an
approximately similar reduction in the maximum axial load sus-
tained by the columns due to increasing the slenderness ratio of
the columns. For example, under pure axial compression load,
increasing the kl=r ratio of the columns from 16 to 80 resulted
in a reduction of about 41% and 39% in the maximum axial load
of GFRP bar reinforced NSC and HSC columns, respectively. Fur-
ther experimental investigations on the behaviour of GFRP bar
reinforced slender concrete columns with different slenderness
ratios are needed to assess the analytical observations reported
in this study.
5. Conclusions
In this study, the behaviour of GFRP bar reinforced NSC and HSC
columns were analytically investigated. Experimental data from
available literature were used as benchmarks for the analytical
investigations performed in this study. In addition, the effect of dif-
kL/r = 16 ferent parameters (i.e., the compressive strength of concrete,
mechanical properties and the reinforcement ratio of GFRP bars
kL/r = 32
and the slenderness ratio of the column) on the axial and flexural
kL/r = 48 response of the GFRP bar reinforced NSC and HSC columns were
analytically examined. Based on the analytical investigations car-
kL/r = 64
ried out in this study, the following conclusions are drawn:
kL/r = 80
(1) Under concentric axial load, the improvements in the axial
load carrying capacity of GFRP bar reinforced concrete col-
umns due to increasing GFRP longitudinal reinforcement
ratio were more pronounced in NSC columns than in HSC
columns. However, as the axial load eccentricity increased,
the improvement in the axial load carrying capacity due to
a increasing GFRP longitudinal reinforcement ratio became
approximately the same in both GFRP bar reinforced NSC
8 10
and HSC columns.
(2) The NSC and HSC columns reinforced longitudinally with
GFRP bars having a small reinforcement ratio (1%) and a
kL/r = 16
low tensile modulus of elasticity (35 GPa) might experience
kL/r = 32 a tensile failure of the GFRP bars located on the tension side
kL/r = 48
of the columns, especially if the columns are subjected to a
high level of axial load eccentricity.
kL/r = 64
(3) Under concentric axial load, increasing the compressive
kL/r = 80 strength of the concrete from 30 MPa to 80 MPa led to an
increase of about 143% in the maximum axial load sustained
by the GFRP bar reinforced concrete columns. However, the
improvement in the axial load carrying capacity of the GFRP
bar reinforced concrete columns significantly decreased
when columns were under high level of axial load
eccentricities.
(4) The GFRP bar reinforced NSC and HSC columns experienced
b an approximately similar reduction in the maximum axial
8 10
load sustained by the columns due to increasing the slender-
ness ratio, kl=r, of the columns.
(5) The analytical investigations reported in this study reveals
that GFRP bars can be used as longitudinal reinforcements
 H.A. Hasan et al. / Construction and Building Materials 183 (2018) 605–617
to enhance the performance of concrete columns under axial
and flexural loads. In addition, ignoring the contribution of
the GFRP longitudinal reinforcement may lead to an overly
conservative estimation for the maximum axial load sus-
tained by GFRP bar reinforced concrete columns.
Conflict of interest
Authors declare that there is no conflict of interest.
Acknowledgments
All authors acknowledge the University of Wollongong, Aus-
tralia for providing the research facilities. The first author would
like to acknowledge the Iraqi Government for the support of his
full Ph.D. scholarship.
References
[1] L.C. Bank, Composites for Construction: Structural Design with FRP Materials,
Wiley, Hoboken, NJ, 2006.
[2] C.C. Choo, I.E. Harik, G. Hans, Strength of rectangular concrete columns
reinforced with fiber-reinforced polymer bars, ACI Struct. J. 103 (3) (2006)
452–459.
[3] B. Benmokrane, O. Chaallal, R. Masmoudi, Glass fibre reinforced plastic (GFRP)
rebars for concrete structures, Constr. Build. Mater. 9 (6) (1995) 353–364.
[4] ISIS, Reinforcing concrete structures with fibre reinforced polymers, 2007.
<http://www.vrodaustralia.com.au/wpcontent/themes/vrod/pdf/
engineering/codes-and-guides/01.pdf>.
[5] A. De Luca, F. Matta, A. Nanni, Behavior of full-scale glass fiber-reinforced
polymer reinforced concrete columns under axial load, ACI Struct. J. 107 (5)
(2010) 589–596.
[6] H. Tobbi, A.S. Farghaly, B. Benmokrane, Concrete columns reinforced
longitudinally and transversally with glass fiber-reinforced polymer bars, ACI
Struct. J. 109 (4) (2012) 551–558.
[7] H. Mohamed, M. Afifi, B. Benmokrane, Performance evaluation of concrete
columns reinforced longitudinally with FRP bars and confined with FRP hoops
and spirals under axial load, J. Bridge Eng. 19 (7) (2014). 04014020.
[8] M.Z. Afifi, H.M. Mohamed, B. Benmokrane, Theoretical stress–strain model for
circular concrete columns confined by GFRP spirals and hoops, Eng. Struct. 102
(2015) 202–213.
[9] A. Amer, M. Arockiasamy, M. Shahawy, Ultimate strength of eccentrically
loaded concrete columns reinforced with CFRP bars, Proc., 2nd Int. Conf. on
Advanced Composite Materials in Bridge and Structures (ACMBS-II), Canadian
Society for Civil Engineering, Montreal, 1996.
[10] O. Chaallal, B. Benmokrane, Physical and mechanical performance of an
innovative glass-fiber-reinforced plastic rod for concrete and grouted
anchorages, Can. J. Civ. Eng. 20 (2) (1993) 254–268.
[11] K. Kobayashi, T. Fujisaki, Compressive behavior of FRP reinforcement in non-
prestressed concrete members, in: Proc., 2nd Int. RILEM Symp. on Non-
Metallic (FRP) Reinforcement for Concrete Structures, E & FN Spon, London,
1995, pp. 267–274.
[12] D.H. Deitz, I.E. Harik, H. Gesund, Physical properties of glass fiber reinforced
polymer rebars in compression, J. Compos. Constr. 7 (4) (2003) 363–366.
[13] S.H. Alsayed, Y.A. Al-Salloum, T.H. Almusallam, M.A. Amjad, Concrete columns
reinforced by GFRP rods. 4th Int. Symp. on Fiber-Reinforced Polymer
Reinforcement for Reinforced Concrete Structures SP-188, C.W. Dolan,
S.H. Rizkalla, A. Nanni, (eds.), American Concrete Institute, Farmington Hills,
MI, 1999, 103–112.
617
[14] M.Z. Afifi, H.M. Mohamed, B. Benmokrane, Strength and axial behavior of
circular concrete columns reinforced with CFRP bars and spirals, J. Compos.
Constr. 18 (2) (2013). 04013035.
[15] M. Afifi, H. Mohamed, B. Benmokrane, Axial capacity of circular concrete
columns reinforced with GFRP bars and spirals, J. Compos. Constr. 18 (1)
(2013). 04013017.
[16] M.N.S. Hadi, H. Karim, M.N. Sheikh, Experimental investigations on circular
concrete columns reinforced with GFRP bars and helices under different
loading conditions, J. Compos. Constr. 20 (4) (2016). 04016009.
[17] H. Karim, M.N. Sheikh, M.N.S. Hadi, Load and moment interaction diagram for
circular concrete columns reinforced with GFRP bars and GFRP helices, J.
Compos. Constr. 21 (1) (2017). 04016076.
[18] T.A. Hales, C.P. Pantelides, L.D. Reaveley, Experimental evaluation of slender
high-strength concrete columns with GFRP and hybrid reinforcement,, J.
Compos. Constr. 20 (6) (2016). 04016050.
[19] A. Hadhood, H.M. Mohamed, B. Benmokrane, Experimental study of circular
high-strength concrete columns reinforced with GFRP bars and spirals under
concentric and eccentric loading, J. Compos. Constr. 21 (2) (2016).
04016078.
[20] M.N.S. Hadi, H.A. Hasan, M.N. Sheikh, Experimental investigations of circular
high strength concrete columns reinforced with Glass fiber-reinforced
polymer bars and helices under different loading conditions, J. Compos.
Constr. 21 (4) (2017). 04017005.
[21] H.A. Hasan, M.N. Sheikh, M.N.S. Hadi, Performance evaluation of high strength
concrete and steel fibre high strength concrete columns reinforced with GFRP
bars and helices, Constr. Build. Mater. 134 (2017) 297–310.
[22] CSA (Canadian Standards Association), Design and construction of building
components with fiber reinforced polymers, CAN/CSA S806-12, Rexdale, ON,
Canada, 2012.
[23] ACI (American Concrete Institute), Guide for the design and construction of
structural concrete reinforced with FRP bars, ACI: 440.1R-15, Farmington Hills,
MI, 2015.
[24] ACI (American Concrete Institute), Building code requirements for structural
concrete, ACI: 318M-14, Farmington Hills, MI, 2014.
[25] T.-H. Tan, N.-B. Nguyen, Flexural behavior of confined high-strength concrete
columns, ACI Struct. J. 102 (2) (2005) 198–205.
[26] H.H. Ibrahim, J.G. MacGregor, Modification of the ACI rectangular stress block
for high-strength concrete, ACI Struct. J. 94 (1) (1997) 40–48.
[27] T. Ozbakkaloglu, M. Saatcioglu, Rectangular stress block for high-strength
concrete, ACI Struct. J. 101 (4) (2004) 475–483.
[28] E. Canbay, G. Ozcebe, U. Ersoy, High-strength concrete columns under
eccentric load, J. Struct. Eng. 132 (7) (2006) 1052–1060.
[29] E. Thorenfeldt, A. Tomaszewicz, J.J. Jensen, Mechanical properties of high
strength concrete and application to design, in: Proceedings of the
Symposium: Utilization of High-Strength Concrete, Stavanger, Norway, 1987,
pp. 149–159.
[30] M.P. Collins, D. Mitchell, Prestressed Concrete Structures, Prentice Hall,
Englewood Cliffs, 1991, p. 766.
[31] ACI (American Concrete Institute), Report on high strength concrete, ACI 363R-
10, Farmington Hills, MI, 2010.
[32] H. Zadeh, A. Nanni, Design of RC columns using glass FRP reinforcement, J.
Compos. Constr. 17 (3) (2013) 294–304.
[33] A. Hadhood, H.M. Mohamed, F. Ghrib, B. Benmokrane, Efficiency of glass-
fiber reinforced-polymer (GFRP) discrete hoops and bars in concrete
columns under combined axial and flexural loads, Compos. B Eng. 114
(2017) 223–236.
[34] Z.P. Bazant, L. Cedolin, M.R. Tabbara, New method of analysis for slender
columns, ACI Struct. J. 88 (4) (1991) 391–401.
[35] T. Jiang, J.G. Teng, Behavior and design of slender FRP confined circular RC
columns, J. Compos. Constr. 17 (4) (2013) 443–453.
[36] B. Broms, I.M. Viest, Long reinforced concrete columns – a symposium, Trans.
ASCE 126 (2) (1961) 308–400.
[37] E.O. Pfrang, C.P. Siess, Behavior of restrained reinforced concrete columns, J.
Struct. Div. 90 (ST5) (1964) 231.