Structures 41 (2022) 1369–1389

Contents lists available at ScienceDirect

Structures
journal homepage: www.elsevier.com/locate/structures

Behavior of CFRP-strengthened RC beams with circular web openings in

shear zones: Numerical study
Yasin Onuralp Özkılıç a, *, Ceyhun Aksoylu b, Şakir yazman c, Lokman Gemi d,
Musa Hakan Arslan b
a
Necmettin Erbakan University, Faculty of Engineering, Department of Civil Engineering, Konya 42000, Turkey
b
Konya Technical University, Faculty of Engineering and Natural Sciences, Department of Civil Engineering, Konya 42075, Turkey
c
Selcuk University, Ilgın Vocational School, Konya 42615, Turkey
d
Meram Vocational School, Necmettin Erbakan University, Konya 42000, Turkey

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

Keywords: In practice, especially the basement floor beams are drilled and damaged by the users. In some cases, this damage
CFRP to the beams can be significant for the load-bearing element and the whole structure. In this study, the behavior
Circular opening of reinforced concrete beams with circular openings and the failure types resulting from strengthening these
Failure modes
beams with CFRP are parametrically investigated. The diameter of the opening/beam height ratio (D/H), con­
Reinforced concrete beam
Shear damage
crete compressive strength, stirrup spacing, the position of the opening to the beam support, the type of CFRP
application, CFRP ply orientation, and the number of CFRP layers were selected as parameters. Numerical
models were verified using 9 specimens having different circular openings with/without CFRP strengthening and
the analyses of 95 numerical models with the selected parameters were carried out utilizing the finite element
program, ABAQUS. The ultimate load capacity, ductility, stiffness, energy dissipation capacity and failure modes
of the beams were determined. As a result of the study, it was observed that there was no significant loss in
ductility for the beams with D/H ≤ 0.3 and the number of CFRP layer and type of application did not have a
significant effect on D/H ≤ 0.44. However, for the beams with D/H > 0.64, the CFRP application that completely
surrounds openings should be preferred instead of partial CFRP strengthening. In addition, the concrete strength
is an effective parameter for the beams with D/H ≤ 0.44. The effect of the stirrup spacings in the beam on the
ductile behavior was also limited with the increase in the hole diameter. The number of CFRP layers should
theoretically be 4 for an effective strengthening in beams with D/H > 0.44. Finally, U wrapping is recommended
instead of using full wrapping. It has been seen that the location and diameter of the hole are very important
parameters in the failure type of the beam.

1. Introduction using a core drilling machine. These openings, which were not foreseen
during the design of the building, cause potential loss of capacity and
Reinforced concrete (RC) buildings lose their strength over time due subsequent sectional damage under the influence of load. In practice, it
to various reasons [1–3]. Durability problems in the concrete due to is possible to see many examples of beams with openings in this way,
environmental conditions, damage to the reinforcement due to corro­ especially in basement floor beams (Fig. 1). Moreover, the cracks
sion, cracks due to deflection caused by permanent deformation in long- observed in practice due to these openings can be seen in [4].
term loads are some of these reasons. Apart from these reasons, the The effect of these openings drilled close to the support area on the
damages caused by the users during the installation of various instal­ shear stress distribution is extremely important. The possible increase in
lation pipes, especially in the basement and ground floors of reinforced shear stress in the beam supports causes the stresses around the hole to
concrete buildings, may cause a loss of capacity. Installation openings accumulate and increase the intensity of the principal tensile stresses,
are often opened close to the beam supports. The openings can be especially when considered together with the bending effect. In addi­
formed circularly using a core drilling machine or randomly without tion, the changing stresses cause the cracks to intensify around the hole

* Corresponding author.
E-mail address: yozkilic@erbakan.edu.tr (Y.O. Özkılıç).

https://doi.org/10.1016/j.istruc.2022.05.061
Received 23 January 2022; Received in revised form 14 April 2022; Accepted 17 May 2022
2352-0124/© 2022 Institution of Structural Engineers. Published by Elsevier Ltd. All rights reserved.
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 1. Web openings in RC frame beam (Anonymous) .

Fig. 2. The cracks intensify around the hole.

and the crack directions to be changed (Fig. 2). If these holes opened in reason, the studies in the literature are generally based on the circular
the beams are planned beforehand, necessary measures can be taken by section and especially the distance of the openings to the support, the
adding additional reinforcements around them. However, if the holes size of the openings (diameter/beam height ratio if it is circular), the
are drilled without prior planning, the strength and stiffness losses of effect of the geometry of the openings on the behavior are mentioned.
these beams under load must be controlled. From the studies, it is seen that the effect of the openings remains at a
In the literature, the shear and bending behavior of RC beams that very limited level until a certain diameter/beam height ratio in circular
have been drilled before or after their construction has been investigated openings [19,23]. In these studies, it has been reported that the effect of
by many researchers [5–25]. In these studies, the openings were the opening is greater when the opening approaches the support [26].
generally chosen as rectangular and circular. Considering that the holes Some strengthening measures should be taken in order to regain the
are drilled with core drilling machines and the installation openings are original capacity or to increase the capacity of the beams with weakened
circular, it can be said that circular holes are more common. For this cross-sections. Examples such as external steel reinforcement and

1370
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 3. Typical beam layout and reinforcement details.

Fig. 4. Test setup.

reinforced concrete jacketing often create application difficulties for finite element (FEM) programs. Experiment results can be predicted
beams [27]. One of the easiest and most effective strengthening methods with a great success rate by calibrating many model parameters that
for beams in reinforced concrete buildings, which are found to be affect the result, especially the stress–strain relationship of reinforced
inadequate in terms of shear, is a fiber-reinforced polymer (FRP) concrete and CFRP material, the connection acceptance of CFRP-
[28–31]. Primarily, it is known that FRP is used in many different sit­ concrete surface (perfect bond, cohesive layer, and etc.). For example
uations due to its superior mechanical properties [32–35] and especially Khalaf et al. [53], Mahmoud [54], Morkhade et al. [55], Almusallam
for strengthening beams [36–48]. Apart from these studies, the experi­ et al. [56], Elsanadedy et al. [49], El-Basiouny et al. [57], Sayed [58],
mental studies on the strengthening of rectangular openings in beam Salih et al. [59] and Mansour [60] conducted numerical studies on FRP
bodies with FRP of hollow beams are available in the literatüre [49–52]. strengthening the beams with openings.
Torunbalci [26] and Aksoylu et al. [19] investigated the effectiveness of In the literature, FE modeling is now recognized as a powerful
CFRP in strengthening holes of different diameters. The studies showed method and an economical alternative to examining the effect of
that CFRP in hollow beams brings significant gains in capacity and important parameters instead of experimental study [61,62]. The
ductility. It has been observed that the effectiveness of CFRP varies ac­ limited parametric variables on experiments of especially beams with
cording to other mechanical and geometric properties based on the opening motivated the authors to calibrate the existing test data with FE
experiments. modeling and subsequently to test important parameters of interest
In particular, studies in recent years show that the behavior of CFRP numerically. In this study, it is aimed to investigate the structural per­
reinforced beams can be predicted quite successfully with advanced formance of RC beams with circular openings which are strengthened by

1371
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 5. References and strengthened beams configuration.

1372
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Table 1
Test specimen properties.
Number of specimens Specimen names CFRP strengthening type around hole Hole diameter/beam height Specimen Sketches
(D/H)

1 RB-D75 — 0.30

2 RB-D110 — 0.44

3 RB-D160 — 0.64

4 CFRP-1-D75 Strips 0.30

5 CFRP-1-D110 0.44

6 CFRP-1-D160 0.64

7 CFRP-2-D75 Fully wrapped 0.30

8 CFRP-2-D110 0.44

9 CFRP-2-D160 0.64

RB: Represent reference beams, D: Represent diameter of openings, CFRP: Represent the strengthening by CFRP composite.

CFRP composite near the support regions with the use of numerical 125 × 250 mm. The beam geometries are ½ scaled, the stirrup diameter
models. The ultimate load capacity, ductility, stiffness, energy dissipa­ and spacing were selected as ϕ6 /250 mm and ϕ6 /300 mm at the shear
tion and failure modes of the beams were estimated using finite element and bending span, respectively. The longitudinal reinforcements were
modeling. utilized according to TS 500 and TBEC 2019 in the beams are B420c type
This study is a continuation of the existing experimental research by 2Ф14 + 1Ф12 tensile reinforcements and 2Ф8 compression re­
Aksoylu et al. [19]. Nine beams with different openings size and CFRP inforcements. The longitudinal reinforcement yield strength and ratio
configurations were tested. In the study, the experimental study of (ρ) were determined for the tension zone as 440 MPa and 0.0149,
Aksoylu et al. [19] was simulated and the numerical model was cali­ respectively. The reinforcements were designed as an under-balanced
brated based on the experimental findings. For this purpose, the finite mechanism (ρ = 0.0149 < ρb = 0.0209). RC beams with circular open­
element analysis software ABAQUS was utilized to create the finite ings were strengthened with different full wrapping applications using a
element model (FEM). The findings obtained from this study are very single uniaxial layer 400 gr/m2 CFRP in 90◦ configuration. The speci­
important for both researchers and engineers. First of all, the verified mens were tested under symmetrical two-point loading. In order to
numerical model results led to the production of a comprensive para­ obtain shear damage, the av/d ratio was selected as 3.2. Loading was
metric study, resulting in a more comfortable interpretation of the re­ measured using a load-cell (300 kN capacity), while the displacements
sults in terms of various parameters. The parameters selected in the were measured using two LVDTs. The measured data were obtained
study are the diameter of the circular openings/beam height ratio, the using a data acquisition system. The reinforcement layout of the speci­
concrete compressive strength of the beam, the spacing of the stirrups in mens, test setup, and also reference and wrapping configuration of CFRP
the beam and the location of the circular openings, CFRP configuration, strengthening at beams are shown in Fig. 3, Fig. 4, and Fig. 5, respec­
number of CFRP layer and CFRP ply orientation. The results of a total of tively. The specimens classified according to diameter/height (D/H) are
95 reinforced concrete beams modeled in this study created a very presented in Table 1. Beams were not strengthened by CFRP represents
comprehensive data set. as “RB”, diameter of the openings represents as “D” and strengthened by
composite materials represents as “CFRP” in Table 1. CFRP-1 represents
2. Experimental study strip strengthening while CFRP-2 represents full wrapping around the
hole. For instance, CFRP-1-D110 represents the beams with a diameter
In this study, the results of specimens whose experiments were car­ of 110 mm (D110) that are partially wrapped with CFRP strips (1). For
ried out by the author’s previous study [19] were used. The experi­ instance, CFRP-2-D160 represents the beams with a diameter of 160 mm
mental investigation involves nine circular opening beams strengthened (D160) that are fully wrapped around the hole (2).
with CFRP composites tested under four-point loading at the same shear There are three common types of shear failure in RC beams with web
span-to-depth ratios (av/d). The compression strength of production of openings: beam-type shear failure, frame-type shear failure, and web
the beams was determined as 35 MPa. In practice, the holes drilled in the compression failure [63]. In beam-type shear failure, the diagonal shear
beams are usually drilled according to the diameters of the installation cracks propagate along with the entire beam height (Fig. 6a). The frame-
pipes. In this study, the diameter of the openings was selected as 75 mm, type shear failure occurs when two distinct cracks form in the top and
110 mm, and 160 mm in order to be both suitable for these diameters bottom chords, and one of the chords fails due to this cracking (Fig. 6b).
and to be meaningful for comparison. As can be seen from the previous The web compression failure originates from the inadequate compres­
study of the authors [19], there is a significant decrease in the load- sion capacity of the diagonal strut in an RC beam. In other words, web
carrying capacity of the holes drilled in the regions close to the sup­ compression failure (Fig. 6c) is caused by the crushing of concrete be­
ports compared to the beams without holes. All beams are simply sup­ tween the diagonal cracks.
ported with a total length of 2500 mm, and a cross-section of the beam As shown in Fig. 7, the damage observed in three beams with circular

1373
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 6. Common types of shear failure near the hole.

holes is presented. All specimens (RB-D75, RB-D110, and RB-D160) were strengthened with four CFRP strips with complete wrappings near
exhibited similar behavior and the specimens failed due to the shear each circular hole opening. In other words, the CFRP strips adhered
damage before the beam reached its bending capacity. In other words, discretely over each opening zone in a full-wrapping configuration,
significant damage was observed around the holes, and the beam un­ where all strips were 100 mm wide (wf) and applied at 200 mm spacing
derwent shear-compression failure. While the hole diameter increased in (sf) (center-to-center of strips). Specimen CFRP-1-D75 reached its flex­
the first group of the test specimens, there was a significant decrease in ural capacity thanks to the effects of CFRP strengthening at the shear
the load-bearing capacity. However, it can be understood that the hole zone (Fig. 8a). As a result, the load-carrying capacity of specimen CFRP-
diameters in the beams should be limited to D/H ≤ 0.3 to obtain more 1-D110 is 50% more than specimen RB-D110. Besides, smiliar load-
ductile behavior in the beams. In terms of load-carrying capacities, it carrying capacity of CFRP-1-D110 is observed to that of specimen RB-
was observed that RB-D110 and RB-D160 decreased 33.5% and 66.7%, D75 and CFRP-1-D75. However, CFRP-1-D110 did not reach its flex­
respectively, compared to RB-D75. ural capacity due to the shear failure observed around the openings
In the experimental study conducted for the reference group, it was (Fig. 8b). Specimen CFRP-1-D160 reached its load-carrying capacity
observed that the shear damage around the hole was a frame-type failure 75% higher than RB-D160, but shear damage around the hole was
in all beams. In the second group of specimens (CFRP-1), the specimens observed similarly to RB-D160. The decrease in load-carrying capacity

1374
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 9. Strengthening specimens using CFRP-2.

Fig. 7. Reference specimens, RB-D75, RB-D110 and RB-D160.

affected the load capacity quite significantly. Besides, this failure mode
indicates that the beams with large openings are prone to diagonal
compression failure due to the loss of material in the web. It can be said
that premature failure of this kind cannot be prevented with the help of
CFRP full-wrapping application in the beams. As a result, when the D/H
ratio in beams is less than or equal to 0.3, flexural failure is observed in
beams strengthened with a type of CFRP-1. Otherwise, Although the
load-carrying capacity increases, the beam collapses with shear failure.
In the third group of experimental specimens (CFRP-2), fully CFRP
was applied to the surrounded of the hole. In other words, the third
group of specimen details is similar to the second group except in that
one full CFRP wrap around the hole. Specimen CFRP-2-D75 exhibited
approximately similar load-bearing capacity and damage mechanism to
specimen CFRP-1-D75. These strengthened applications were ensured
ductile behavior and flexural capacity. Besides the CFRP-2-D75 spec­
imen contributing to the ductility, it was observed that there was a
significant decrease in micro-damages occurring under CFRP applica­
tion and changing the damaged area (Fig. 9a). On the other hand, the
CFRP-2-D110 specimen exhibited quite different behavior from that of
specimen CFRP-1-D110 (Fig. 9b). Unlike CFRP-1-D110, the CFRP-2
technique applied on the CFRP-2-D110 specimen was observed to be
effective on ductility and load-bearing capacity. It is suggested that the
beam with a 110 mm circular opening should be strengthened by using
this technique. Finally, the CFRP-2-D160 specimen with a 160 mm hole
diameter was strengthened with different CFRP configuration tech­
niques similar to CFRP-2-D110. In examining the CFRP-2-D160 spec­
imen, failure occurred due to brittle and shear damage, similar to the
CFRP-1-D160 specimen (Fig. 9c). When the D/H ratio considerably in­
Fig. 8. Strengthening specimens using CFRP-1.
creases, the cross-section of the beam decrease. Therefore, the load-
bearing capacity of the beam is also decreased. It was clearly observed
in specimen CFRP-1-D160 is approximately 42% compared to RB-D75, that the application of different CFRP strengthened techniques was
CFRP-1-D75 and CFRP-1-D110. The reason for this about the presence unsuccessful with a D/H ratio of 0.64 and above.
of the large web openings decreased the compression capacity of the The ratio of the opening diameter to the beam height (D/H) was
beam web, and crushing took place under the diagonal strut forces 0.30, 0.44 and 0.64. The rigidity, ductility and energy dissipation ca­
(Fig. 8c). In other words, the hole diameter increase in these beams pacity of each specimen were also calculated from the

1375
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 10. Parameters considered in the evaluation of the beams.

Table 2
The results of the specimens.
Test specimens Pmax (kN) Pu (0.85Pmax) (kN) δy (mm) δu (mm) Ductility (δu/ δy) Initial stiffness (kN/mm) Failure Type Energy dissipation capacity (kJ)

1.RB-D75 117.4 99.80 — 29.00 — 21.82 Shear 3.37

2.RB-D110 78.00 66.30 — 10.40 — 16.73 Shear 0.83
3.RB-D160 39.00 33.15 4.00 14.00 3.50 11.88 Flexure 0.76
4.CFRP-1-D75 117.00 99.45 15.70 56.20 3.57 17.24 Flexure 6.07
5.CFRP-1-D110 117.10 99.50 — 19.50 — 21.93 Shear 1.55
6.CFRP-1-D160 68.40 58.10 — 13.10 — 14.62 Shear 0.86
7.CFRP-2-D75 113.00 96.10 13.20 58.00 4.39 24.13 Flexure 6.40
8.CFRP-2-D110 128.50 109.2 13.80 61.70 4.46 17.62 Flexure 8.78
9.CFRP-2-D160 85.60 72.80 — 19.50 — 15.17 Shear 1.30

— No yield displacement was obtained in specimens failed due to shear damage without observing bending damage.

load–displacement curve. The ductility ratio is calculated as the ratio of

Table 3 the displacement value (δu/δy) corresponding to 85% of the maximum
Concrete damage plasticity parameters (Özkılıç et al. [64]). load-bearing capacity (Fig. 10a). It is reported in the literature that the
Ψ ε σb0/ σb0 K µ minimum value of this ratio for an under-balanced ductile design beam
31 0.1 1.16 0.667 0.0005
is around 4 ~ 5 [19]. The initial stiffness (rigidity) was calculated from
the load–displacement graph in Fig. 10b. In addition, the energy dissi­
where the dilation angle (Ψ), the flow potential eccentricity (ε), the ratio of pation capacity was calculated as the area under the vertical
initial biaxial compressive yield stress to initial uniaxial compressive yield stress
load–displacement curve (Fig. 10c).
(σ b0/ σb0), the ratio of the second stress invariant on the tensile meridian to that
Tables 2 and 3 were created based on the definitions given in Fig. 10,
on the compressive meridian (K) and the viscosity parameter (µ).
where Pmax is the maximum load-carrying capacity, δu and δy are the

Fig. 11. Mesh configuration.

Fig. 12. Boundary conditions and interactions.

1376
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Table 4 Table 7
Calculation of tensile and compressive stress–strain curves (Özkılıç et al. [64]). Material properties of CFRP composite.
Description Equations Properties of CFRP composites Value

Compressive strain and stress curve (ε ) Longitudinal elasticity modulus (GPa) E1 100
n
c
σc εco Transversal elasticity modulus (GPa) E2 8
= ( ε )n
fc Poisson ratio, v12 0.3
(n − 1) +
c
εco Shear modulus (GPa) G12 3
n = 2.758x10− 6 fc + 1.0 Shear modulus (GPa) G13 3
Tensile strength σt = 0.3fc(2/3)
′ Shear modulus (GPa) G23 2
Tensile strength in fiber direction (MPa) XT 1450
Tensile cracking strain ′
ft Compressive strength in fiber direction (MPa) XC 1550
εcr =
Ec Tensile strength in matrix direction (MPa) YT 100
Compressive strength in matrix direction (MPa) YC 100
where σ c is compressive stress, εc is compressive strain, fc is uniaxial compressive Longitudinal shear strength (MPa) SL 100
strength and εco is uniaxial compressive strain. Ec is the slope of initial tangent of Transverse shear strength (MPa) ST 100
compressive stress and strain curve. Energy required to fully damage a ply using fiber tension only (mJ/ Gft 75
mm2)
Energy required to fully damage a ply using fiber compression only Gfc 75
(mJ/mm2)
Energy required to fully damage a ply using matrix tension only Gmt 1
Table 5 (mJ/mm2)
Energy required to fully damage a ply using matrix compression Gmc 1
Calculation of failure parameters (Özkılıç et al. [64]).
only (mJ/mm2)
Concrete Failure Parameter Equations

Compression (dc) σc E−c 1

dc = 1 − − 1
beam utilized in the experimental study were adopted. Nonlinear ge­
εpl (1/bc − 1) + σc Ec
ometry and material were implemented.
c
Tension (dt) σt E−c 1
dt = 1 − pl
εt (1/bt − 1) + σt E−c 1
Concrete was modeled using the 3-D element called C3D8R (A
general-purpose eight-node linear brick element with reduced integra­
where bc and bt are taken as 0.7 and 0.4, respectively. tion), whereas the reinforcements were modeled utilizing a 2-D element
These parameters can take values from zero to one depend on the severity of the
called T3D2 (2-node linear displacement truss element). On the other
damage. The value of 0 stands for the undamaged material, while 1 stands for
hand, CFRP composite was represented using a four-node general-pur­
fully damaged material.
pose shell with reduced integration and hourglass control (S4R). The
numerical convergence, efficiency and accuracy were examined. Several
Table 6
models having different mesh sizes were investigated to determine the
Hashin Failure Criteria (Özkılıç et al. [64]). optimum mesh size of the numerical models, which results in the highest
accuracy and least convergence problem and computational cost. Based
Mode description Equations
on the mesh sensitivity analyses, a mesh size of 20 mm was selected for
Tensile fiber mode (σ ) 2 (τ )2
Ftf =
11
+
12 CFRP wrapping and concrete, whereas a mesh size of 10 mm was
XT SL preferred for the reinforcements. Mesh configuration of the numerical
Fiber compressive mode (σ )2
Fcf =
11 models is demonstrated in Fig. 11.
XC
(σ )2 (τ ) 2 The beams were subjected to four-point loading in the experiments.
Tensile matrix mode
The loadings were implemented into the numerical models utilizing
22 12
Ftm = +
YT SL
Compressive matrix mode )2 [( C ) 2 ] ) displacement-controlled loading. The loadings were applied to the areas
(σ
22 Y σ22 (τ12 2
Fcm = + − 1 + where the rollers contact the beams. The beams were placed to the steel
2ST 2ST YC SL
support. These supports were modeled as elastic in the numerical
where σ 11 and σ 22 are components of the effective stress tensor, XT and XC are the
models. Fixed boundary conditions were applied to the bottom of the
tensile and compressive strength in the direction of fiber, YT and YC are the
supports. A frictional interaction was defined between steel support and
tensile and compressive strength in the direction of matrix, SL and ST are the
shear strength of longitudinal and transverse direction, respectively. the beam. The friction coefficient was selected as 0.7 based on the
recommendation of Aksoylu et al. 0.7 [65]. The reinforcements were
embedded in the concrete using the “embedded constrain” option, which
displacement value at Pu (Pu = 0.85Pmax), the displacement value at the
constrains the translation degrees of freedom of the reinforcements to
yielding. Table 2 demonstrates Pmax, Pu, δy, δu and δy, ductility, initial
the interpolated values of the corresponding degrees of freedom of the
stiffness, and energy dissipation capacities.
concrete [66]. Fig. 12 demonstrates the boundary conditions and
interactions.
3. Numerical study A perfect bond was assumed through the use of the tie constrain
between the concrete and surrounding CFRP composite. When the per­
The experimented specimens were further investigated in the present fect bond is utilized, no failure at the interaction between the concrete
numerical analyses. A total of 95 reinforced concrete beams with and CFRP should be expected. In other words, the failure modes of
openings were designed in order to carry out a comprehensive para­ delamination and debonding should not be expected. The concept of the
metric study. 45 of these beams were strengthened by a CFRP composite. perfect bond was successfully adopted by many researchers, including
To this end, three-dimensional nonlinear finite element models were the previous studies of the Authors [67–71]. The studies [72,73] showed
developed. The numerical models were generated using the finite that the models with perfect bonding give slightly conservative results.
element tool, ABAQUS. Implicit solver was utilized for the analyses. The interactions and boundary conditions are demonstrated in Fig. 12.
Static type of step analysis was selected. The exact dimensions of the

1377
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

relationship is defined in concrete damaged plasticity model. The

calculation of these failure modes is given in Table 5.
Hashin Failure criteria [74,75] were utilized to model the material
behavior of CFRP composite. Many researchers successfully utilized
Hashin Failure criteria to simulate the behavior of FRP composite in the
reinforced concrete structures [76,77]. Four failure modes are defined in
the Hashin Failure criteria: tensile fiber mode, fiber compressive mode,
tensile matrix mode, and compressive matrix mode. These failure modes
can be computed from the equations given in Table 6. The material
properties of the CFRP composite shown in Table 7 were implemented
based on the coupon tests (mechanical properties of the FRP materials
are determined by tensile coupon test methods). Similar material prop­
erties are also reported in the study of Özkılıç et al. [64].
A typical stress–strain curve given in Özkılıç et al. [64] was utilized
in this present study. Poisson ratio and elasticity modulus are 0.3 and
200 GPa, respectively. In order to implement strain and stress relation to
ABAQUS, engineering stress and strain were converted to true stress and
strain.
Nine experimented specimens were simulated to verify the numeri­
cal model described in the previous sections. Load-displacement curves
of the experimental and numerical results are compared in Fig. 13. It is
seen that the numerical model accurately simulated the initial stiffness
of the experimented specimens. Approximately after the first crack,
numerical models exhibited a slightly higher stiffness than those of
experimental results. The reason of relatively high stiffness can be
related to two possible reasons: firstly, unmodeled microcracks, and
secondly, the interactions between the concrete and CFRP and between
the concrete and reinforcements, which were modeled as perfect
bonding. The numerical models also successfully captured both brittle
(shear) failure and ductile (bending) failure, which is essential for this
study except for D160.
Fig. 14 depicts the failure modes of the experimental and numerical
findings in terms of DAMAGET (tension damage). The red color in­
dicates fully damage while blue color represents no damage. A general
trend was observed in numerical results that bending cracks decrease as
the diameter of the opening increases. These results are in agreement
with the experimental results. The numerical models also captured
bending and shear failures. The tensile cracks observed in the experi­
mented specimens are also simulated by the numerical results. It can be
concluded that the numerical models can accurately simulate the
behavior of the beams with openings.

4. Parametric study

In this part, a series of numerical studies were carried out considering

different parameters after the experiments performed by the authors
were verified in ABAQUS. The parametric studies were performed to
examine the effects of concrete compressive strength, opening location,
Fig. 13. Experimental and numerical comparison of specimens. stirrup spacing, ply orientations, the number of layers, material prop­
erties of CFRP composite on the behavior of shear deficient beam
strengthened by CFRP strip composites. The selected variables are
The inelastic behavior of the concrete was modeled by the concrete
shown in Table 8. Although these parameters in reinforced concrete
damaged plasticity model. Table 3 demonstrates the adapted parameters
mechanics are known, it is essential to examine these variables on the
used in the concrete damaged plasticity model. The tensile and
beams strengthened by CFRP. Therefore, a comprehensive parametric
compressive stress–strain curves adopted by Özkılıç et al. [64] were
study was conducted to investigate the effects of the variables
utilized in this present study (Table 4). For the tensile behavior of
mentioned above.
concrete, tension stiffening is considered. The equations to compute the
stress–strain curves are depicted in Table 5.
Failure modes of concrete in terms of tension damage (DAMAGET) 4.1. The effects of concrete compressive strength
and compression damage (DAMAGEC) are available in the library of
ABAQUS. To predict these failure modes constitutive material In this section, only the effect of the variation of concrete compres­
sive strength on the behavior of the beams has been examined (Table 9).

1378
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 14. Experimental and numerical damage view of specimens.

Table 8
Selected parameters.
Selected Parameters Experimental Values Parametric Values

The parameters related to RC Concrete compressive strength (MPa) 35 25 16

Steel yield strength (MPa) 440 440 440
Longitudinal tensile reinforcement ratio (ρ) 0.0149 0.0149 0.0149
Longitudinal compressive reinforcement ratio (ρ’) 0.0035 0.0035 0.0035
Opening location 1.5H D/2 H 2H
Stirrup spacing (mm) 250 62.5 125
The parameters related to CFRP strengthening CFRP ply orientation [90◦ ] [±45◦ ] [0◦ /90◦ ]
Wrapping type of CFRP CFRP-1CFRP-2 CFRP-3 CFRP-4
The number of CFRP layer 1 2 4

D: Circular opening diameter; H: Beam height; CFRP-1 and CFRP-2 used in the experimental study; CFRP-3: Side bonding; CFRP-4: U-shaped wrapping.

Table 9
The results of the specimens.
Test specimens Pmax (kN) Pu (0.85Pmax) (kN) δy (mm) δu (mm) Ductility (δu/ δy) Pmax Stiffness (kN/mm) Failure Type Energy dissipation capacity (kJ)

1. RB-D75- C35 117.40 99.80 — 29.00 — 6.41 Shear 3.37

2.RB-D75- C25 116.98 99.44 14.30 28.25 1.98 5.54 Shear 4.18
3.RB-D75- C16 104.64 88.95 14.45 26.09 1.81 4.94 Shear 3.31
4.CFRP-1-D75-35 117.00 99.45 15.70 56.20 3.57 6.14 Bending 6.07
5.CFRP-1-D75-25 123.45 104.93 12.69 57.35 4.52 5.08 Bending 7.97
6.CFRP-1-D75-16 114.01 96.91 13.80 47.30 3.43 5.39 Shear + Bending 3.92
7.CFRP-2-D75-C35 113.00 96.10 13.20 58.00 4.39 4.44 Bending 6.40
8.CFRP-2-D75-C25 120.78 102.66 13.52 46.12 3.41 5.45 Shear + Bending 4.04
9.CFRP-2-D75-C16 110.46 93.89 14.30 36.13 2.53 5.65 Shear 2.57
10. RB-D110- C35 78.00 66.30 — 10.40 — 9.48 Shear 0.83
11.RB-D110- C25 72.96 62.01 5.40 9.18 1.70 9.93 Shear 0.12
12.RB-D110- C16 58.34 49.59 4.61 8.40 1.82 8.71 Shear 0.29
13.CFRP-1-D110-C35 117.10 99.50 — 19.50 — 6.61 Shear 1.55
14.CFRP-1-D110-C25 89.38 75.97 9.87 16.04 1.63 6.75 Shear + Bending 0.82
15.CFRP-1-D110-C16 71.25 60.56 8.41 13.63 1.62 6.05 Shear 0.81
16.CFRP-2-D110-C35 128.50 109.2 13.80 61.70 4.46 4.12 Bending 8.78
17.CFRP-2-D110-C25 133.85 113.78 13.70 45.52 3.32 4.70 Shear + Bending 4.12
18.CFRP-2-D110-C16 124.48 105.81 14.53 29.18 2.01 5.92 Shear 1.76
19. RB-D160- C35 39.00 33.15 4.00 14.00 3.50 4.86 Bending 0.76
20.RB-D160- C25 37.76 32.10 3.54 6.48 1.83 7.90 Shear 0.21
21.RB-D160- C16 30.53 25.95 3.09 5.74 1.86 6.65 Shear 0.18
22.CFRP-1-D160-C35 68.40 58.10 — 13.10 — 5.46 Shear 0.86
23.CFRP-1-D160-C25 51.56 43.83 6.01 10.66 1.77 5.31 Shear 0.35
24.CFRP-1-D160-C16 39.54 33.61 5.14 9.35 1.82 5.63 Shear 0.31
25.CFRP-2-D160-C35 85.60 72.80 — 19.50 — 5.09 Shear 1.30
26.CFRP-2-D160-C25 65.71 55.86 7.92 13.34 1.68 5.27 Shear 0.34
27.CFRP-2-D160-C16 54.55 46.37 7.13 12.35 1.73 4.94 Shear 0.27

— No yield displacement was obtained in specimens failed due to shear damage without observing bending damage.

1379
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 15. The effects of material properties of concrete.

Other parameters were kept constant. The effects of concretes with CFRP-1 type strengthening was not successful. With the hole diameter of
compressive strength (fc) of 35 MPa, 25 MPa, and 16 MPa were inves­ 160 mm, it has been understood that the strengthening types of CFRP-1
tigated. In the experiment, the concrete compressive strength of 35 MPa and CFRP-2 cannot be effective no matter how high the concrete class is.
was used. It was observed in Fig. 15 that a decrease in compressive This situation directly affected the damage modes and energy con­
strength (especially for 16 MPa) leads to a significant decrease in the sumption capacities of the beams.
load-carrying capacity of the specimens. In other words, the decrease in
the compressive strength (35 MPa to 16 MPa) significantly affected the
4.2. The effects of the opening location
structural behavior of the specimens. The increase in concrete
compressive strength with a decrease in the hole diameter causes the
The diameter and location of holes were considered as the main
ductile plateau to become more pronounced. It can be seen in Fig. 15
parameters. The other parameters were the same as the experiments.
that D75 and D110 strengthened beams exhibited a ductile plateau for
Within this scope, a total of twelve unstrengthened reference RC beams
all compressive strength of beams. Moreover, the decrease in compres­
with opening diameters of 75, 110, and 160 mm and the distance of the
sive strength leads to a decrease in initial stiffness for all specimens.
opening from end support of D/2, H, 1.5H, and 2H were analyzed.
The ductility ratio of a typical bending beam with under-balanced
Fig. 16 shows the load–displacement curves of the beams with different
reinforcement ratio is around 4 ~ 5. When Table 9 is examined, RB-
opening diameters and opening locations. All of the holes are located
D75 beams exhibited brittle behavior for all concrete classes. Howev­
within the shear spans (distance from support as 2H) no matter if the
er, as a result of their CFRP-1 and CFRP-2 type strengthening, beams
distance of the hole center from the support is 2H, 1.5H, H, and D/2.
with C25 and C30 class concrete tended to be ductile, while brittle
Since maximum moment occurs in the middle span of the beam, the
behavior was observed in beams with C16 class. This shows that as the
stresses around the hole started to increase as the diameter of the holes
concrete strength increases, the shear capacity will increase as a result of
moved away from the supports. However, the load-carrying capacity for
the bending behavior. Reference beams with a hole diameter of 110 mm
the D75 is approximately similar. This showed that there would be no
exhibit brittle behavior before being strengthened. While CFRP-2 type
significant reduction in load-carrying capacity for D/H ≤ 0.3. It is
strengthening of these beams led to ductile behavior, it showed that
clearly observed that the decrease in load-carrying capacity reaches

1380
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 16. Comparison of the numerical results for 75, 110, and 160 mm opening diameters.

Table 10
Numerical results of the RB specimens.
Test specimens Pmax (kN) Displacement at Pmax (mm) Pmax Stiffness (kN/mm) Failure Type

1. RB-D75- C35-D/2 123.48 21.00 5.88 Shear

2. RB-D75- C35-H 122.33 25.08 4.87 Shear
3. RB-D75- C35-1.5H 121.72 21.55 5.64 Shear
4. RB-D75- C35-2H 123.54 20.85 5.92 Shear
5. RB-D110- C35-D/2 82.15 6.31 13.02 Shear
6. RB-D110- C35-H 72.59 7.32 9.91 Shear
7. RB-D110- C35-1.5H 86.20 7.99 10.78 Shear
8. RB-D110- C35-2H 71.30 6.57 10.85 Shear
9. RB-D160- C35-D/2 46.13 5.03 9.17 Shear
10. RB-D160- C35-H 30.37 4.35 6.98 Shear
11. RB-D160- C35-1.5H 43.90 5.16 8.50 Shear
12. RB-D160- C35-2H 35.14 8.23 4.26 Shear

severe levels as the hole diameter increases and moves away from the than in other locations (Table 10).
support. Since the point where the single load is applied is at a distance
of 2.92H from the support, the hole location remains in the shear span.
Regardless of the hole diameter, the reduction in shear capacity due to 4.3. The effects of stirrup spacing
the arching effect in holes that come very close to the support area is less
Three different stirrup spacing (62.5 mm, 125 mm, and 250 mm)

1381
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Table 11
Numerical results of the comparison of RB specimens stirrup spacing.
Test specimens Pmax Displacement at Pmax Stiffness Failure
(kN) Pmax (mm) (kN/mm) Type

1. RB-D75- 123.24 25.89 4.76 Shear

C35-s62.5
2. RB-D75- 124.42 20.92 5.94 Shear
C35-s125
3. RB-D75- 121.72 21.55 5.64 Shear
C35-s250
4. RB-D110- 112.48 9.29 12.10 Shear
C35- s62.5
5. RB-D110- 85.09 7.60 11.19 Shear
C35- s125
6. RB-D110- 86.20 7.99 10.78 Shear
C35- s250
7. RB-D160- 44.01 5.17 8.51 Shear
C35- s62.5
8. RB-D160- 43.86 5.12 8.56 Shear
C35- s125
9. RB-D160- 43.90 5.16 8.50 Shear
C35- s250

spacing. While the D75 beam with 62.5 mm stirrup spacing shows the
best ductility performance, its load-carrying capacity is identical with
125 mm and 250 mm stirrup spacing. It is clearly understood in Fig. 17
that load-carrying capacity does not change for D/H ≤ 0.3 beams if the
stirrup spacing is different. Moreover, a significant increase in both
ductility and energy dissipation capacity was observed for the D75 beam
with a stirrup spacing of 62.5 mm. On the other hand, the load-carrying
contribution gradually decreases for the stirrup spacing of 125 and 250
mm for D110 specimen. For example, in stirrup beams placed with 62.5
mm intervals in D110 specimen, stirrups compared spacing significantly
increased the load-carrying capacity, while the decrease in 125 and 250
mm stirrup beams was more pronounced. Yu and Bazant [78] stated that
no significant effect was observed on the shear capacity, although the
stirrup spacing significantly affected the inclination of the diagonal
shear crack. As a result of the numerical analysis, the damage views of
each model are given in Table 11 and Fig. 18. The figure shows the
distribution of von misses stress caused by different stirrup spacings and
different hole diameters.

4.4. The effects of the number of CFRP layers and ply orientation

As it is known, the selection of CFRP plies orientation and number of

layers are critical in the strengthened application. Therefore, the
configuration and number of layers of fibers are some of the most crucial
features that should be considered (Table 12). When the beam is
strengthened by CFRP, two types of damage are considered at the
reinforced concrete: shear and bending damage. Shear and bending
damages occur at 45◦ and 90◦ , respectively. Therefore, when fibers are
parallel to the direction of the force, they exhibit their maximum ca­
pacity. In order to prevent shear and bending cracks, 45◦ and 0◦ fibers
should be used, respectively. Multiple ply orientations were simulated to
investigate the behavior of the beams with different CFRP ply orienta­
tions and to prove the conceptual design of FRP. The comparison of the
numerical models with different lay-up orientations for CFRP-1-D110,
CFRP-1-D160, and CFRP-2-D160 beam wrapped only at shear zone is
given in Fig. 19. All numerical models of the beam with [90◦ ] lay-up
Fig. 17. Comparison of the models with stirrup spacing (62.5, 125, 250 mm).
sequence coincide with the experimental result of CFRP-1-D110 and
CFRP-1-D160. In CFRP-2–160 with [±45]4 lay-up sequence, the fibers
were considered. The effect of stirrup spacing on load-carrying capacity
are strong to shear cracks, and the fibers at the shear zone do allow
shows a different effect for different diameters. In the experiment, 250
ductile behavior since [− 45◦ ]4 lay-up sequence are perpendicular to the
mm stirrup spacing was utilized for beams. For instance, the strength at
shear force. Therefore, CFRP-2-D160 with [±45]4 lay-up sequence
the lowest level is observed in the D160 beam for a different stirrup
showed the maximum load-carrying capacity. On the other hand, the

1382
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 18. Comparison of the models with von misses stress value.

Table 12
Numerical results of the comparison of RB specimens number of layer and ply orientation.
Test specimens Pmax (kN) Displacement at Pmax (mm) Pmax Stiffness (kN/mm) Failure Type

1. CFRP-1-D110-C35-[90◦ ] 104.73 13.90 7.53 Shear

2. CFRP-1-D110-C35-[0◦ /90◦ ] 103.88 13.72 7.57 Shear
3. CFRP-1-D110-C35-[0◦ /90◦ ]2 107.27 13.83 7.75 Shear
4. CFRP-1-D110-C35-[0◦ /90◦ ]4 109.21 13.81 7.90 Shear
5. CFRP-1-D110-C35-[±45◦ ] 100.88 13.43 7.51 Shear
6. CFRP-1-D110-C35-[±45◦ ]2 102.52 13.52 7.58 Shear
7. CFRP-1-D110-C35-[±45◦ ]4 101.16 12.98 7.79 Shear
8. CFRP-1-D160-C35-[90◦ ] 56.69 8.15 6.95 Shear
9. CFRP-1-D160-C35-[±45◦ ]4 77.58 8.82 8.79 Shear
10. CFRP-1-D160-C35-[0◦ /90◦ ]4 63.19 8.50 7.43 Shear
11. CFRP-2-D160-35-CFRP-2 71.95 10.09 7.13 Shear
12. CFRP-2-D160-35-[0◦ /90◦ ] 78.86 12.02 6.56 Shear
13. CFRP-2-D160-35-[0◦ /90◦ ]2 86.54 11.82 7.32 Shear
14. CFRP-2-D160-35-[0◦ /90◦ ]4 100.71 14.15 7.11 Shear
15. CFRP-2-D160-35-[±45◦ ] 93.01 13.76 6.75 Shear
16. CFRP-2-D160-35-[±45◦ ]2 103.21 17.04 6.05 Shear
17. CFRP-2-D160-35-[±45◦ ]4 121.83 19.11 6.37 Shear + Bending

1383
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 19. Comparison of the models with a number of layer and ply orientation.

lowest load carrying capacities are observed in CFRP-2-D160 with [0◦ / 2-D160 strengthening is quite effective when the hole diameters are 160
90◦ ] lay-up orientations (except [90◦ ]). The reason for this is that 0◦ /90◦ mm. As a result, [±45◦ ] fiber orientation and complete wrapping around
fibers are 45◦ relative to the shear cracks and fibers are weak when they the hole (CFRP-2-D160) are recommended to prevent or delay shear
are subjected to the force of 45◦ . All models with different lay-up se­ damage. The maximum load, displacement at Pmax, stiffness at Pmax and
quences increased load-carrying capacity but did not show ductile damage failure of the models are given in Table 12. Especially in CFRP-
behavior except for [±45]4. It is observed herein that the applied CFRP- 2-D160, the number of layers and ply orientation seriously affected the

1384
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 20. Wrapping types used in the models.

load-carrying capacity. On the other hand, when CFRP-1 strengthening orientations did not change the beam damage behavior. However, the
is utilized, the beams could not reach ductile behavior even if four layers beams having openings of 110 mm with [90◦ ] CFRP-4 strengthening
of CFRP are utilized. This indicates the importance of wrapping around could not achieve ductile behavior. When [±45◦ ] and [0◦ /90◦ ] ply
the openings. [0◦ /90◦ ], [0◦ /90◦ ]2, [0◦ /90◦ ]4, [±45◦ ], [±45◦ ]2 and orientations were used, CFRP-4 also exhibited ductile behavior. This
[±45◦ ]4 increased by 9.6%, 20.2%, 39.9%, 29.2%, 43.4% and 69.3% shows the importance of the selection of the ply orientation as the hole
compared to the beam CFRP-2-D160, respectively. diameter increases. If the hole diameter is 160 mm, the choice of a
number of layers and ply orientation comes to the fore even more. As can
be seen from Fig. 22 and Table 14, the ply orientation of [90◦ ] with the
4.5. The effects of wrapping type
fully wrapped CFRP-2 reached a similar load–displacement value. In
CFRP-4–160 models, an increase in the number of layers increased the
Fig. 20 demonstrates the wrapping types used in this study. CFRP-1
load-carrying capacity, but partly ductile behavior was observed only at
and CFRP-2 were used in the experimental study. In CFRP-3 wrapping,
[±45◦ ]4. Similar results were obtained when CFRP-2 was utilized.
CFRP was applied only to two sides of the beam. No CFRP was applied at
the bottom and top of the beams. On the other hand, CFRP-4 represents
5. Conclusion
U-shape wrapping where CFRP was applied on two sides and the bottom
of the beam.
In this study, the behavior of RC beams with circular openings in the
The results of CFRP-3 are illustrated in Fig. 21. It has been observed
shear zone and the behavioral changes resulting from the strengthening
that the behavior changes with the number of layers and the change in
of these beams with CFRP are parametrically investigated. The ultimate
ply orientation. However, in side bonding application, although there
load capacity, ductility, stiffness, energy dissipation, and failure modes
are slight differences in maximum load–displacement values depending
of the beams are modeled with the help of the ABAQUS program. The
on the number of layers, typical ductile behavior was not observed. This
results of the experimental study carried out by the authors were used as
is a sign that the side bonding application cannot show sufficient
the validation set, and then a comprehensive data set was obtained from
contribution in terms of ductility. It is also seen that the [±45◦ ]4 type
95 RC beam models created with the selected parameters, and the results
wrapping exhibits the closest behavior to the CFRP-2 type wrapping. In
were examined comparatively. In the study, the diameter of the circular
Table 13, the load, displacement, stiffness and damage failure values
opening/beam height ratio, the concrete compressive strength, the
determined by considering the number of different layers and ply ori­
spacing of the stirrups in the beam, the position of the circular opening
entations are given.
relative to the beam support, the applied CFRP strengthening type, CFRP
The load–displacement graph of the beams strengthened by U-sha­
ply orientation, and the number of wraps was selected as parameters.
ped wrapping (CFRP-4 type) is shown in Fig. 22. Wrapping the models
The main findings obtained from the study are as follows;
with a relatively small hole diameter (75 mm) in different ply

1385
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

capacity gain and undergoes shear failure with both types of

strengthening (CFRP-1 and CFRP-2).
3) On the other hand, the change in concrete compressive strength was
not very effective in strengthened beams of D75 and D110 as the
behavior of D75 and D110 beams was dominated by bending and a
relatively ductile plateau was formed. It is also seen that concrete
compressive strength positively affects ductility, especially in D75
and D110 beams, which were strengthened with CFRP-2 and have
ductile behavior. From this, it is seen that the second type of
strengthening should be utilized and the hole diameter/beam height
ratio should not exceed 0.44 (maximum hole diameter is 110 mm) in
order for the concrete strength to be an effective parameter. As an
increase in concrete elasticity modulus increases the shear and
bending stiffness of the section, the increase in concrete compressive
strength causes a slight increase in the initial stiffness.
4) Based on the point where the single load is applied, the hole location
remains in the shear span in each alternative. Regardless of the hole
diameter, the reduction in span capacity due to the arching effect in
holes that are very close to the support area is less than in other
locations.
5) It is seen that small diameter holes will not cause a significant
decrease in ductility in beams wrapped in accordance with rein­
forced concrete structures codes (this range corresponds to 62.5 cm
(2.5H) in this study). However, as the hole diameter increases, the
stirrup tightening does not have any benefit in terms of ductility.
Especially in large holes (D160), stirrup tightening did not have any
effect on the capacity. The shear stresses increasing with the decrease
in the concrete section cause a very large increase in the principal
tensile stresses and the beam reaches its shear capacity with shear
failure.
6) It has been observed that CFRP-2 will significantly enhance ductility,
especially in large diameter holes (such as D160). For this reason, it
is extremely important to strengthen beams with large openings
using a single piece of CFRP-2 instead of segmented CFRP-1 in order
to prevent shear failure.
7) It is seen that CFRP type and the number of windings do not have a
significant effect when D/H is below 0.44. Similarly, if a partial
strengthening is to be applied in large diameter holes, it is theoret­
Fig. 21. Comparison of the models with wrapping type. ically necessary to choose the number of layers as 4 in terms of
considerable ductility and capacity gain. However, an increasing
1) Finite elements models generated by using ABAQUS are capable of number of layers may result in difficulties in applying beams in the
predicting the behavior of the beams having openings with and site.
without CFRP strengthenings. The models successfully captured the 8) In CFRP-1, especially in beams where D/H is less than 0.44, the
load–displacement curves and failure modes. number of CFRP layers and ply orientation do not have a significant
2) Concrete compressive strength has a significant effect on the effect on the behavior. On the other hand, the number of CFRP layers
behavior of beams. Especially since the beams without CFRP and ply orientation significantly affected the behavior of the beams
strengthening reached the capacity by shear failure, the change of with CFRP-2. The beams with side bonding exhibited the worst case
concrete strength in these beams is more effective on the behavior. among the beams with CFRP-2, CFRP-3 and CFRP-4. The beams with
This is more evident in the D160 beam, which has the weakest

Table 13
Numerical results of the comparison of CFRP-3 specimens number of layer and ply orientation.
Test specimens Pmax (kN) Displacement at Pmax (mm) Pmax Stiffness (kN/mm) Failure Type

1. CFRP-3-D75-C35-CFRP-2 123.87 25.71 4.81 Shear + Bending

2. CFRP-3-D75-C35-[90◦ ] 116.07 16.71 6.94 Shear
3. CFRP-3-D75-C35-[0◦ /90◦ ] 122.31 24.86 4.92 Shear
4. CFRP-3-D75-C35-[0◦ /90◦ ]2 122.67 25.01 4.90 Shear
5. CFRP-3-D75-C35-[±45◦ ] 121.82 24.81 4.91 Shear
6. CFRP-3-D75-C35-[±45◦ ]2 122.81 25.10 4.89 Shear
7. CFRP-3-D110-C35-CFRP-2 136.77 27.59 4.95 Bending
8. CFRP-3-D110-C35-[90◦ ] 109.61 14.59 8.51 Shear
9. CFRP-3-D110-C35-[0◦ /90◦ ] 124.31 16.48 7.54 Shear
10. CFRP-3-D110-C35-[0◦ /90◦ ]2 129.98 16.97 7.65 Shear
11. CFRP-3-D110-C35-[0◦ /90◦ ]4 134.94 24.94 5.41 Shear + Bending
12. CFRP-3-D110-C35-[±45◦ ] 134.34 20.36 6.59 Shear
13. CFRP-3-D110-C35-[±45◦ ]2 134.40 25.00 5.37 Shear + Bending
14. CFRP-3-D110-C35-[±45◦ ]4 134.71 24.91 5.40 Shear + Bending

1386
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Fig. 22. Comparison of the models with wrapping type.

1387
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

Table 14
Numerical results of the comparison of CFRP-4 specimens number of layer and ply orientation.
Test specimens Pmax (kN) Displacement at Pmax (mm) Pmax Stiffness (kN/mm) Failure Type

1. CFRP-4-D75-C35-CFRP-2 123.87 25.71 4.81 Bending

2. CFRP-4-D75-C35-[90◦ ] 123.85 26.62 4.65 Bending
3. CFRP-4-D75-C35-[0◦ /90◦ ] 123.98 25.79 4.80 Bending
4. CFRP-4-D75-C35-[±45◦ ] 124.00 26.11 4.75 Bending
5. CFRP-4-D110-C35-CFRP-2 136.77 27.59 4.95 Shear
6. CFRP-4-D110-C35-[90◦ ] 135.64 18.46 7.34 Shear
7. CFRP-4-D110-C35-[0◦ /90◦ ] 136.78 26.95 5.05 Bending
8. CFRP-4-D110-C35-[±45◦ ] 137.04 25.10 5.46 Bending
9. CFRP-4-D160-C35-CFRP-2 71.95 10.09 7.13 Shear
10. CFRP-4-D160-C35-[90◦ ] 68.99 8.07 8.54 Shear
11. CFRP-4-D160-C35-[0◦ /90◦ ] 108.33 16.79 6.45 Shear
12. CFRP-4-D160-C35-[0◦ /90◦ ]2 121.40 17.18 7.06 Shear
13. CFRP-4-D160-C35-[0◦ /90◦ ]4 132.68 18.32 7.24 Shear
14. CFRP-4-D160-C35-[±45◦ ] 114.93 17.47 6.57 Shear
15. CFRP-4-D160-C35-[±45◦ ]2 134.14 18.46 7.26 Shear
16. CFRP-4-D110-C35-[±45◦ ]4 134.29 20.85 6.44 Bending

CFRP-4 behaved similar to the beam with CFRP-2; therefore, U [15] Amiri S, Masoudnia R, Pabarja AA. The study of the effects of web openings on the
concrete beams. Aust J Basic Appl Sci 2011;5:547–56.
wrapping can be recommended instead of using full wrapping.
[16] Yang K-H, Eun H-C, Chung H-S. The influence of web openings on the structural
behavior of reinforced high-strength concrete deep beams. Eng Struct 2006;28(13):
It is seen that the openings to be drilled in the beams have quite 1825–34.
different effects depending on the selected parameters. CFRP applica­ [17] Aykac B, Kalkan I, Aykac S, Egriboz YE. Flexural behavior of RC beams with
regular square or circular web openings. Eng Struct 2013;56:2165–74.
tion, which is one of the most practical strengthening methods, can [18] Aykac B, Aykac S, Kalkan I, Dundar B, Can H. Flexural Behavior and Strength of
significantly increase the shearing capacity of the beam and, accord­ Reinforced Concrete Beams with Multiple Transverse Openings. ACI Struct J 2014;
ingly, its ductility. This study has shown that many factors affecting the 111.
[19] Aksoylu C, Yazman Ş, Özkılıç YO, Gemi L, Arslan MH. Experimental analysis of
beam behavior must be considered in order to solve the capacity prob­ reinforced concrete shear deficient beams with circular web openings strengthened
lem of beams with holes drilled afterward, which is frequently by CFRP composite. Compos Struct. 2020;249:112561.
encountered in practice. Apart from the parameters discussed in this [20] El Ame F, Mwero JN, Kabubo CK. Openings Effect on the Performance of
Reinforced Concrete Beams Loaded in Bending and Shear. Eng Technol Appl Sci
study, many different factors such as shear span, loading type, hole Res 2020;10:5352–67.
geometry need to be tested in future studies. [21] Allam SM, Ebeido TI. Retrofitting of RC beams predamaged in shear using CFRP
sheets. Alexandria Eng J 2003;42:87–101.
[22] Elamary AS, Sharaky IA, Alqurashi M. Flexural behaviour of hollow concrete
beams under three points loading: Experimental and numerical study. Structures
Declaration of Competing Interest 2021;32:1543–52.
[23] Torunbalci N. Behaviour and design of small square openings in reinforced
concrete beams. Architect Sci Rev 2000;43(4):201–10.
The authors declare that they have no known competing financial [24] Amiri S, Masoudnia R. Investigation of the opening effects on the behavior of
interests or personal relationships that could have appeared to influence concrete beams without additional reinforcement in opening region using fem
the work reported in this paper. method. Aust J Basic Appl Sci 2011;5:617–27.
[25] Ashour AF, Rishi G. Tests of reinforced concrete continuous deep beams with web
openings. Struct J 2000;97:418–26.
References [26] Torunbalci N. Behaviour and design of large rectangular openings in reinforced
concrete beams. Architect Sci Rev 2002;45(2):91–6.
[27] Ojaimi MF. Experimental Study on Shear Strengthening of Reinforced Concrete
[1] Aksoylu C, Sezer R. Investigation of precast new diagonal concrete panels in
Beams Using Different Techniques of Concrete Jacketing. Basrah J Eng Sci 2021;21
strengthened the infilled reinforced concrete frames. KSCE J Civ Eng 2018;22(1):
(2):53–61.
236–46.
[28] Aksoylu C. Experimental analysis of shear deficient reinforced concrete beams
[2] Aksoylu C, Kara N. Strengthening of RC frames by using high strength diagonal
strengthened by glass fiber strip composites and mechanical stitches. Steel Compos
precast panels. J Build Eng 2020;31:101338.
Struct 2021;40:267.
[3] Kaltakci M. Y., Arslan M.H., Yilmaz U.S., Arslan H.D.. A new approach on the
[29] Aksoylu C, Özkılıç YO, Yazman Ş, Gemi L, Arslan MH. Experimental and Numerical
strengthening of primary school buildings in Turkey: An application of external
Investigation of Load Bearing Capacity of Thinned End Precast Purlin Beams and
shear wall. Building and Environment 2008;43(6):983–90. https://doi.org/
Solution Proposals. Teknik Dergi 2021;32(3):10823–58.
10.1016/j.buildenv.2007.02.009.
[30] Gemi L, Aksoylu C, Yazman Ş, Özkılıç YO, Arslan MH. Experimental investigation
[4] Wojdak R. Structural analysis of a failured RC beam with openings in a building
of shear capacity and damage analysis of thinned end prefabricated concrete
under construction. Czasopismo Techniczne Budownictwo 2013;110:157–68.
purlins strengthened by CFRP composite. Compos Struct 2019;229:111399.
[5] Somes N, Corley W. Circular openings in webs of continuous beams. Special
[31] Arslan MH, Yazman Ş, Hamad AA, Aksoylu C, Özkılıç YO, Gemi L. Shear
Publication 1974;42:359–98.
strengthening of reinforced concrete T-beams with anchored and non-anchored
[6] Mansur MA, Lee YF, Tan KH, Lee SL. Tests on RC continuous beams with openings.
CFRP fabrics. Structures 2022;39:527–42.
J Struct Eng 1991;117(6):1593–606.
[32] Madenci E, Özkılıç YO, Gemi L. Buckling and free vibration analyses of pultruded
[7] Mansur M. Deflections of reinforced concrete beams with web openings. ACI Struct
GFRP laminated composites: Experimental, numerical and analytical
J: Citeseer 1992.
investigations. Compos Struct 2020;254:112806.
[8] Mansur M. Design of reinforced concrete beams with web openings. In:
[33] Madenci E, Özkılıç YO, Gemi L. Experimental and theoretical investigation on
Proceedings of the 6th Asia-Pacific structural engineering and construction
flexure performance of pultruded GFRP composite beams with damage analyses.
conference (ASPEC 2006): Kuala Lumpur; 2006. p. 5–6.
Compos Struct 2020;242:112162.
[9] Mansur MA. Effect of openings on the behaviour and strength of R/C beams in
[34] Madenci E, Özkılıç YO, Gemi L. Theoretical investigation on static analysis of
shear. Cem Concr Compos 1998;20(6):477–86.
pultruded GFRP composite beams. Acad Platform J Eng Sci 2020;8:483–90.
[10] Mansur M, Tan K-H. Concrete beams with openings: analysis and design. CRC
[35] Özkılıç YO, Madenci E, Gemi L. Tensile and compressive behaviors of the pultruded
Press; 1999.
GFRP lamina. Turkish J Eng 2020;4:169–75.
[11] Mansur M, Tan K-H, Wei W. Effects of creating an opening in existing beams. Struct
[36] Panjehpour M, Abang Ali AA, Aznieta FN. Energy absorption of reinforced concrete
J 1999;96:899–905.
deep beams strengthened with CFRP sheet. Steel Composte Struct 2014;16(5):
[12] Tan K-H, Mansur MA. Design procedure for reinforced concrete beams with large
481–9.
web openings. Struct J 1996;93:404–11.
[37] Bae S-W, Murphy M, Mirmiran A, Belarbi A. Behavior of RC T-beams strengthened
[13] Tan K-H, Mansur MA, Huang L-M. Reinforced concrete T-beams with large web
in shear with CFRP under cyclic loading. J Bridge Eng 2013;18(2):99–109.
openings in positive and negative moment regions. Struct J 1996;93:277–89.
[38] Belarbi A. Design of FRP systems for strengthening concrete girders in shear.
[14] Allam SM. Strengthening of RC beams with large openings in the shear zone.
Transp Res Board 2011.
Alexandria Eng J 2005;44:59–78.

1388
Y.O. Özkılıç et al. Structures 41 (2022) 1369–1389

[39] Belarbi A, Acun B. FRP systems in shear strengthening of reinforced concrete [59] Salih R, Abbas N, Zhou F. Experimental and Numerical investigations on the cyclic
structures. Procedia Eng 2013;57:2–8. load behavior of beams with rectangular web openings strengthened using FRP
[40] Azam R, Soudki K, West JS, Noël M. Strengthening of shear-critical RC beams: sheets. Structures 2021;33:655–77.
Alternatives to externally bonded CFRP sheets. Constr Build Mater 2017;151: [60] Mansour W. Numerical analysis of the shear behavior of FRP-strengthened
494–503. continuous RC beams having web openings. Eng Struct. 2021;227:111451.
[41] Soudki K, El-Salakawy E, Craig B. Behavior of CFRP strengthened reinforced [61] Özkılıç YO, Aksoylu C, Arslan MH. Numerical evaluation of effects of shear span,
concrete beams in corrosive environment. J Compos Constr 2007;11(3):291–8. stirrup spacing and angle of stirrup on reinforced concrete beam behaviour. Struct
[42] Al-Salloum YA, Elsanadedy HM, Alsayed SH, Iqbal RA. Experimental and Eng Mech Int J 2021;79:309–26.
numerical study for the shear strengthening of reinforced concrete beams using [62] Özkılıç YO, Aksoylu C, Arslan MH. Experimental and numerical investigations of
textile-reinforced mortar. J Compos Constr 2012;16(1):74–90. steel fiber reinforced concrete dapped-end purlins. J Build Eng 2021;36:102119.
[43] Elsanadedy HM, Abbas H, Al-Salloum YA, Almusallam TH. Prediction of [63] Bengi Aykac SAIKBD, Husnu C. Flexural Behavior and Strength of Reinforced
intermediate crack debonding strain of externally bonded FRP laminates in RC Concrete Beams with Multiple Transverse Openings. ACI Struct J. 111.
beams and one-way slabs. J Compos Constr 2014;18(5):04014008. [64] Özkılıç YO, Yazman Ş, Aksoylu C, Arslan MH, Gemi L. Numerical investigation of
[44] Norris T, Saadatmanesh H, Ehsani MR. Shear and flexural strengthening of R/C the parameters influencing the behavior of dapped end prefabricated concrete
beams with carbon fiber sheets. J Struct Eng 1997;123(7):903–11. purlins with and without CFRP strengthening. Constr Build Mater 2021;275:
[45] Saadatmanesh H, Malek AM. Design guidelines for flexural strengthening of RC 122173.
beams with FRP plates. J Compos Constr 1998;2(4):158–64. [65] Aksoylu C, Özkılıç YO, Arslan MH. Damages on prefabricated concrete dapped-end
[46] Osman BH, Wu E, Ji B, Abdelgader SAM. A state of the art review on reinforced purlins due to snow loads and a novel reinforcement detail. Eng Struct 2020;225:
concrete beams with openings retrofitted with FRP. Int J Adv Struct Eng 2016;8: 111225.
253–67. [66] Dere Y, Koroglu MA. Nonlinear FE modeling of reinforced concrete. Int J Struct
[47] Pimanmas A. Strengthening R/C beams with opening by externally installed FRP Civil Eng Res 2017;6:71–4.
rods: Behavior and analysis. Compos Struct 2010;92(8):1957–76. [67] Kachlakev D, McCurry DD. Behavior of full-scale reinforced concrete beams
[48] Gemi L, Madenci E, Özkılıç YO. Experimental, analytical and numerical retrofitted for shear and flexural with FRP laminates. Compos B Eng 2000;31(6-7):
investigation of pultruded GFRP composite beams infilled with hybrid FRP 445–52.
reinforced concrete. Eng Struct 2021;244:112790. [68] Lesani M, Bahaari MR, Shokrieh MM. Numerical investigation of FRP-strengthened
[49] Elsanadedy HM, Al-Salloum YA, Almusallam TH, Alshenawy AO, Abbas H. tubular T-joints under axial compressive loads. Compos Struct 2013;100:71–8.
Experimental and numerical study on FRP-upgraded RC beams with large [69] Lesani M, Bahaari MR, Shokrieh MM. Experimental investigation of FRP-
rectangular web openings in shear zones. Constr Build Mater 2019;194:322–43. strengthened tubular T-joints under axial compressive loads. Constr Build Mater
[50] Abdalla HA, Torkey AM, Haggag HA, Abu-Amira AF. Design against cracking at 2014;53:243–52.
openings in reinforced concrete beams strengthened with composite sheets. [70] Lesani M, Bahaari MR, Shokrieh MM. FRP wrapping for the rehabilitation of
Compos Struct 2003;60(2):197–204. Circular Hollow Section (CHS) tubular steel connections. Thin-Walled Struct 2015;
[51] Nie X, Zhang SS, Teng JG, Chen G. Experimental study on RC T-section beams with 90:216–34.
an FRP-strengthened web opening. Compos Struct 2018;185:273–85. [71] Rasheed HA, Larson KH, Amiri SN. Analytical Solution of Interface Shear Stresses
[52] Salih R, Zhou F, Abbas N, Khan MA. Experimental Investigation of Reinforced in Externally Bonded FRP-Strengthened Concrete Beams. J Eng Mech 2013;139(1):
Concrete Beam with Openings Strengthened Using FRP Sheets under Cyclic Load. 18–28.
Materials 2020;13:3127. [72] Arduini M, Nanni A, Di Tommaso A, Focacci F. Shear response of continuous RC
[53] Khalaf MR, Al-Ahmed AHA, Allawi AA, El-Zohairy A. Strengthening of Continuous beams strengthened with carbon FRP sheets. In: Non-Metallic (FRP) Reinforcement
Reinforced Concrete Deep Beams with Large Openings Using CFRP Strips. for Concrete Structures, Proceedings of the Third Symposium; 1997. p. 459–66.
Materials 2021;14:3119. [73] Obaidat YT, Heyden S, Dahlblom O. The effect of CFRP and CFRP/concrete
[54] Mahmoud AM. Strengthening of concrete beams having shear zone openings using interface models when modelling retrofitted RC beams with FEM. Compos Struct
orthotropic CFRP modeling. Ain Shams Eng J 2012;3(3):177–90. 2010;92(6):1391–8.
[55] Morkhade SG, Kshirsagar M, Dange R, Patil A. Analytical study of effect of web [74] Hashin Z. Failure criteria for unidirectional fiber composites. J Appl Mech 1980;47:
opening on flexural behaviour of hybrid beams. Asian J Civil Eng 2019;20(4): 329–34.
537–47. [75] Hashin Z, Rotem A. A fatigue failure criterion for fiber reinforced materials.
[56] Almusallam T, Al-Salloum Y, Elsanadedy H, Alshenawy A, Iqbal R. Behavior of J Compos Mater 1973;7(4):448–64.
FRP-Strengthened RC Beams with Large Rectangular Web Openings in Flexure [76] Maziz A, Tarfaoui M, Gemi L, Rechak S, Nachtane M. A progressive damage model
Zones: Experimental and Numerical Study. Int J Concr Struct Mater 2018;12:47. for pressurized filament-wound hybrid composite pipe under low-velocity impact.
[57] El-Basiouny AM, Askar HS, El-Zoughiby ME. Experimental and numerical study on Compos Struct. 2021;276:114520.
the performance of externally prestressed reinforced high strength concrete beams [77] Maziz A, Tarfaoui M, Rechak S, Nachtane M, Gemi L. Finite Element Analysis of
with openings. SN Appl Sci 2021;3:37. Impact-Induced Damage in Pressurized Hybrid Composites Pipes. Int J Appl Mech
[58] Sayed AM. Numerical study using FE simulation on rectangular RC beams with 2021;13(07).
vertical circular web openings in the shear zones. Eng Struct. 2019;198:109471. [78] Yu Q, Bažant ZP. Can stirrups suppress size effect on shear strength of RC beams?
J Struct Eng 2011;137(5):607–17.

1389