Structures 28 (2020) 2523–2536

Contents lists available at ScienceDirect

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

Effects of FRP grids on punching shear behavior of reinforced concrete slabs

Navid Jafarian *, Davood Mostofinejad, Ali Naderi
Department of Civil Engineering, Isfahan University of Technology (IUT), Isfahan, Iran

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

Keywords: In recent decades, the functionality of the reinforced concrete (RC) flat slabs has remarkably grown concerning
Reinforced concrete their unique characteristics. Despite the numerous practical and economic merits of the slabs, the punching shear
Flat slab failure is a highly controversial issue. Accordingly, many studies have been targeted at developing new tech­
Strengthening
niques for the strengthening of the slabs against the brittle punching. Nonetheless, there are little-known
FRP grid
Punching shear
research findings on the effects of the external shear reinforcement, most notably fiber-reinforced polymer
Finite element modeling (FRP) grids, on the punching shear behavior of the RC flat slabs. The present study aims to investigate the effects
of FRP grids on the punching shear behavior of the slabs. For this purpose, finite element (FE) modeling was
employed to simulate the slabs strengthened with the FRP grids varying in type, pattern, and spacing. First, a FE
model was introduced and calibrated using three experimental tests from available research. Then, 13 slabs were
incrementally analyzed under the monotonic vertical loading. The slabs had identical geometry and reinforce­
ment details in a manner that the punching shear was the governing failure mode. One slab was taken as a
control specimen without any strengthening while 12 others were strengthened with glass and carbon FRP (GFRP
and CFRP) grids in orthogonal and radial patterns located at three different spacing. Hence, the marked effects of
type, configuration, and spacing of the FRP grids on the shear behavioral characteristics of the slabs were
investigated for the first time. Numerical results proved that the FRP grids significantly enhance the punching
shear resistance and the ductility of the slabs. Besides, it was observed that the performance of CFRP grids and
radial pattern on the characteristics of the punching shear failure of the slabs are more desirable than GFRP grids
and orthogonal pattern.

1. Introduction Nevertheless, the main disadvantage of the RC flat slabs is the con­
centration of shear and moment around the loading area, leading to the
Over the last few decades, the rapid development of reinforced brittle punching failure. As punching is accorded fragile failure, it
concrete (RC) flat slabs has had a significant impact on the construction happens with limited warning signs, resulting in a progressive collapse
field. The RC flat slab is a two-way reinforced concrete structural of the entire building [4]. On the other hand, several reasons including
element, which carries comparatively light floor loads, and transfers changes of occupancy, construction or design errors, code updates, and
them to the columns, without any beams or girders. Generally, flat-slab environmental deterioration of materials may cause to decline in the
systems are most commonly employed in multi-story buildings [1]. The punching shear capacity of the slabs [5]. Thus, the flat slabs quite often
flat slabs have both pros and cons, but, overall, the benefits are stronger need strengthening against the punching shear. For this reason, there
than the drawbacks [2]. Some of the many advantages of flat slabs are as have been marked changes in attitudes towards the shear strengthening
follows [3]: methods of the flat slabs.
In 1994, it was proved that using steel fibers is an effective method to
• Saving in building height; enhance the punching shear strength of the RC flat slabs [6]. Another
• Flexibility in design layout; solution for improving the punching shear behavior of the flat slabs is
• Reduction in construction cost and time; employing layers of fiber-reinforced polymer (FRP). It was observed that
• Ease of installation of mechanical and electrical services. installing carbon FRP (CFRP) strips in the vertical direction of the slabs
as the shear reinforcing elements increase the punching shear strength of

* Corresponding author.
E-mail addresses: [email protected] (N. Jafarian), [email protected] (D. Mostofinejad), [email protected] (A. Naderi).

https://doi.org/10.1016/j.istruc.2020.10.061
Received 7 June 2020; Received in revised form 18 September 2020; Accepted 23 October 2020
2352-0124/© 2020 Institution of Structural Engineers. Published by Elsevier Ltd. All rights reserved.
N. Jafarian et al. Structures 28 (2020) 2523–2536

the slabs [7]. Utilizing CFRP sheets and strips on the tensile and technical aspects of FRP grids externally placed in the slabs to investi­
compressive sides of the slabs with and without openings enhances the gate the corresponding shear characteristics. Thus, the current study was
load-carrying capacity of the slabs [8–12]. It was found that anchoring targeted at elucidating the effects of FRP grids on the punching shear
CFRP layers by grooving at the end of the FRP length would help to behavior of the slabs in such a manner that type, pattern, and spacing of
prevent failure due to deboning; as a result, deboning was postponed these strengthening elements to be precisely probed for the first time.
and accompanied by the rupture of the end anchorage with a consid­ The types of fibers included glass and carbon, and the FRP patterns were
erable amount of concrete peeling [8]. Moreover, the enhancement of orthogonal and radial, where their position was in 30, 40, and 50 mm
the punching shear strength was greater in the slabs of higher spacing. The experimental tests done on the RC flat slabs strengthened
compressive concrete strength but with less reinforcement [9]. Besides, with the FRP grids are few and far between the available researches, as
different arrangements of CFRP strips were investigated on the struc­ are also extortionately expensive. Therefore, proposing a capable FE
tural behavior of the slab-column connections [10]. It was also proved model to accurately obtain the structural properties of FRP grids, is an
that the combined 45◦ and 90◦ arrangement of the CFRP sheets is the innovative approach to develop case study on the slabs. For this purpose,
best pattern to retrofit the slabs with lateral and corner openings in for the first time, an enhanced finite element model was proposed to
comparison with other arrangements [11,12]. Nevertheless, the afore­ investigate the shear behavioral characteristics of the slabs strengthened
said strengthening methods affect indirectly the punching shear with the FRP grids under the monotonic vertical load.
behavior of the RC flat slabs. For this reason, it has made new de­ The strategy of the present research obeys the following steps. First,
velopments in the strengthening methods to directly improve the shear the general behavior of the RC flat slab was precisely described. Then,
behavioral characteristics of the slabs. the procedure of the finite element modeling was introduced as well as
Rapid advances are being made in the investigation of the different the important numerical parameters. The FE model was calibrated by
types of external shear reinforcing elements on the punching behavior of available experimental studies. Then, the enriched FE model proposed
the RC flat slabs. Some kinds of the shear strengthening elements include herein was extended to carry out the parametric study on the RC flat
steel rods [13], steel studs [13], shear bolts [14,15], screws [16], CFRP slabs, in which the significant parameters previously mentioned were
rods [16,17], CFRP grids [18], and CFRP fans [19]. Employing steel rods type, pattern, and spacing of the FRP grids. Finally, the numerical results
and studs illustrated that the load capacity of a non-strengthened were proposed in the form of punching shear strength, displacement at
specimen with an opening is 14% less than that of those strengthened the failure point, and energy dissipation capacity of the slabs; addi­
without openings while the strengthened specimen with opening tionally, the strain measurement was conducted on some of the FRP
exhibited an increase of 27% in load capacity compared to the control grids.
specimen with an opening [13]. Moreover, it was observed that using
shear bolts is an efficient method to increase lateral drift capacity and 2. Behavior of RC flat slabs
ductility of the slab-column connections in such a way that the
orthogonal and radial patterns of these kinds of shear reinforcement 2.1. General
have a quite similar performance on the shear strengthening of the
connections [14]. Furthermore, it was proved that using 8 screws and The structural behavior of the two-way RC slabs is heavily dependent
nuts have the same performance on the punching shear capacity of the upon the percentage of flexural reinforcement. Numerous studies
slabs as using 8 CFRP rods while the amount of the screws was 116% [25–29] have shown that the general behavior of the RC flat slabs can be
more than that of the CFRP rods [16]. Furthermore, using CFRP rods classified according to the light, medium, and high flexural reinforce­
showed an increase of about two times in the ductility of the strength­ ment ratios. This classification leads the behavior of the slabs into three
ened specimen compared to the control one [17]. It was noticeable that different basic types of failure; flexure-induced punching, combined
the maximum increased shear capacities of 67%, 56%, and 91.3% were flexural-shear, and brittle punching shear. Each failure behavior of the
observed in the slabs strengthened with CFRP rods, CFRP grids, and slabs is described as follows [1]:
CFRP fans, respectively, compared to the control specimens [16,18,19].
It should be noticed that the FRP grids are known as the novel external a. The failure behavior of the slabs with a light reinforcement ratio
shear reinforcement whereas there are not adequate scientific pro­ called flexure-induced punching generally illustrates a ductile
nouncements about their performance on the structural behavior of the deformation all over the plastic plateau. The significant character­
slabs. istic of this type of failure mode is the widespread yielding of lon­
Rapid advances are being made in the nonlinear finite element gitudinal reinforcement in the formation of a full yield-line
modeling, leading to the full complement of experimental research. mechanism. Furthermore, tension cracks are deeply manifested
Likewise, experimental studies are replicated by the verified FE models along the depth of the slabs while large plastic deflection is observed
owing to cost reduction and time-saving. Furthermore, the superiority of before the occurrence of brittle punching failure.
the FE analyzes over the experimental investigation is obvious in the b. The failure mode of medium reinforced slabs is widely known as a
evaluation of the strain and stress fields throughout the RC structures combined flexural-shear. This type of failure behavior causes the part
[20]. Employing FE models, some studies [1,2,21–24] have been con­ of flexural reinforcement to yield despite the flexure-induced
ducted in the field of nonlinear analysis of RC flat slabs with or without punching behavior that makes a complete yield-line pattern. Be­
shear reinforcement and opening. The numerical results depicted that sides, it looks as if the yield lines of this failure mode spread through
proposed FE-based models adequately predict the behavior of tested RC the diagonal directions, as regards the amount of longitudinal rein­
flat slabs, such as load-carrying capacity, maximum displacement, and forcement ratio; the spread of yield-line pattern move forward to the
crack propagation. full mechanism as the level of reinforcement ratio decreases.
The advent of new external shear reinforcing elements has probably c. Brittle punching shear behavior is observed in the slabs with a high
had the most significant effect on the punching shear strengthening of level of reinforcement ratio. In this type of failure mode, ultimate
the RC flat slabs. FRP grids are considered as one of the relatively new rupture occurs before the yielding of flexural reinforcement with no
shear strengthening elements, although the effects of these elements on visible signs while the propagation of tension cracking is at the very
the punching shear behavior of the slabs have not been yet clarified. In least amount, as is the plastic deformation.
other words, many researches have been conducted on the effect of
different shear reinforcing elements on the RC flat slabs even though 2.2. Punching failure mechanism
precious few studies have been carried out on the performance of FRP
grids. For this reason, this research shed the light on the most important Punching shear failure of the RC flat slabs as a local mechanism

2524
N. Jafarian et al. Structures 28 (2020) 2523–2536

occurs in the area surrounding the support areas due to the intensive pressures, although damaged-based models are designed for concrete
localized shear and bending stresses. This failure mode is distinguished with failure mode caused by diffuse damage under high confining
by the formation of a truncated cone-shaped element, as well as it causes pressures [12,39].
a sudden and sharp drop in the load-carrying capacity of the slabs [6]. According to the pros and cons of employing the classical plasticity-
The development of the punching failure mechanism is described as the based models and damaged-based ones, a concrete damaged plasticity
following levels (Fig. 1) [30]: (CDP) model is desirable to couple the basic formulae of aforesaid
models [12]. Thus, the CDP model splits the significant difference be­
• Initiation of an approximately crack surrounded the column edge on tween the classical-plasticity models and damaged-based ones in the
the tension surface of the slab, as well as its deep intrusion into the inclusion of damages. The CDP model depicts the inelastic behavior of
depth towards the compression zone; concrete by combing the concepts of isotropic tensile and compressive
• Creation of a novel lateral and flexural crack; plasticity with isotropic damage elasticity [40]. Additionally, this model
• Formation of an inclined shear close to the middle of the slab depth is generated for predicting the ductile and brittle damages in the regions
manifested at roughly the amount between a half and two-thirds of of compressive and tensile states of stress, respectively, towards the
the load-carrying capacity load; failure state of stress [12].
• Propagation of inclined cracks along the tension and compression The CDP model can simulate all structural elements of strengthened
zones at an angle between the 25 and 45◦ ; RC slabs including concrete, steel bars, FRP grids, as are their in­
• Achieving the propagated inclined cracks to the corner of the inter­ teractions [40]. This model may be employed as though the failure mode
section of the slab and the support, observed at the failure punching of retrofitted RC slabs with FRP grids is not permitted to be debonding
load. based on the experimental results. So in this way, the crushing of con­
crete becomes progressively dominant in compression zones by com­
Generally, the punching shear strength of slabs is affected by speci­ parison with the yielding of steel bars and FRP grids in tension zones
fied parameters concerning geometry and material; the size of support [12]. For this reason, a tie constraint was defined for the interaction
cross-section, the depth of the slab, compressive strength of concrete, between steel bars or FRP grids with the concrete; such constraint tie
and longitudinal reinforcement ratio [1]. separates elements with each other in such a way that there is not any
relative motion between their surfaces [32].
3. Finite element modeling The performance of the CDP model implemented in commercially
software package seems to be the default in general purposes. Therefore,
Finite element analysis is widely regarded as one of the best nu­ users may make a lot of efforts by changing the default settings, if
merical methods for solving the mathematical problems of engineering necessary, to calibrate the FE model for each special case study. Besides,
[31]. In this research, finite element modeling is performed by the the characteristics of materials are especially essential to be introduced
available software package, ABAQUS 2020 [32]. In this section, the into the CDP model [40]. The CDP model described herein is based on
selected constitutive material model, named concrete damaged plas­ the background theory of available documentation [32] and studies
ticity, is introduced. Furthermore, extremely important points are [41,42], and also is targeted at improving the proposed FE model by
mentioned about material properties, element specifications, boundary using the calibration of the FE model.
conditions, nonlinear solution, and validation of modeling.
3.1.2. Constitutive laws
As previously mentioned, the inelastic behavior of concrete in the
3.1. Concrete damaged plasticity CDP model is defined by both the isotropic damaged elasticity, and the
isotropic tensile and compressive plasticity. The yield potential function
3.1.1. General presented herein combines the yield function of Lubliner et al. [41] with
The widespread use of constitutive material models has enabled to the modifications proposed by Lee and Fenves [42]. The proposed
define nonlinear characteristics of the reinforced concrete structures function accounts for different strength evolutions under tensile and
[33]. These models may be broadly organized into plasticity models, compressive loadings. According to this model, the general form of the
orthotropic models, endochronic models, micro models, nonlinear yield potential function is governed by [32]:
elastic models, and fracture mechanics models [34]. There have been
significant developments in the constitutive material models of concrete ( ) 1 ( ( ) ) ( )
F σ , ε̃pl = q − 3αp + β ε̃pl 〈 ̂
σ max 〉 − γ〈 − ̂
σ max 〉 − ̂
σ c ε̃plc ⩽0 (1)
over the past decade. Almost two different theories provide a basis for 1− α
deriving the fundamental equations of these models [35]; classical In the above equation, the Macaulay bracket, , is a notation used to
plasticity [36] and damage [37,38]. Classical plasticity-based models define equation〈⋅〉 = 12 (|⋅| + ⋅ ); that, •, is a mathematical quantity. Be­
are perfectly capable of predicting the features of concrete as it exhibits
sides, σ is the effective stress, and ̂ σ max is the algebraically maximum
ductile behavior under triaxial compressive stress state. In contrast,
eigenvalue of σ . In the biaxial compressive stress state with ̂
σ max = 0, the
damage-based models are completely able to simulate the characteris­
yield potential criterion used in the CDP model converts to the well-
tics of concrete as its performance is similar to a brittle material under
known Drucker-Prager one [41,42]. The plastic strain tensor shown by
tensile or uniaxial compressive states of stress. Overall, plasticity-based
models can be employed to simulate the concrete with failure behavior εpl is used in the scalar relations under uniaxial loading. Therefore, ε̃pl is
associated with localized damage mechanism at low confining defined as the equivalent plastic strain to be employed in the multiaxial

Fig. 1. Punching shear failure mechanism of RC flat slab [30].

2525
N. Jafarian et al. Structures 28 (2020) 2523–2536

case [32]. relations [43]:

In Eq. (1), α is a dimensionless material constant, which can be [ ( )2 ]
estimated by using the maximum biaxial and uniaxial compressive σ c = σ cu
2εc
−
εc
(4a)
stresses [32]. According to the experimental studies, the values of α are ε0 ε0
obtained in the range of 0.08 to 0.12 [41,42]. In this study, 0.1 was
selected for α according to the FE model calibration. Also, γ is another (4b)
′
σ cu = ks fc
dimensionless material constant, which is implemented in the yield
where, σ cu is the maximum compressive stress, εc is the nominal
function only if the numerical model is under triaxial compression stress
strain, ε0 is the strain corresponding to the compressive concrete
state with ̂σ max < 0 [32]. In this study, calibration of the FE models
strength, fc , and ks is a coefficient that its value is considered to be 1.0,
′

indicated that γ = 3.0 gives the best results. Also, p is the effective hy­
0.97, 0.95, 0.93, and 0.92 as the value of fc equals to15, 20, 25, 30, and
′

drostatic pressure, q is the Mises equivalent effective stress and β is

greater than or equal to 35 MPa, respectively [43].
considered to be a function of ε̃pl as follows [32]:
The second branch of the compressive stress–strain diagram de­
( )
( ) σ c ε̃pl creases obviously from point (ε0 ,σcu ) to point (εcu ,0.85σ cu ) [43]. The
β ε̃pl = ( plc ) (1 − α) − (1 + α) (2) value of εcu is taken 0.0038 concerning the FE model calibration.
σ t ε̃t
in which, σ c and σ t are the effective compressive and tensile stresses 3.2.2. Uniaxial tensile behavior
at the corresponding equivalent plastic strains, ε̃plc and ε̃plt , respectively. The uniaxial tensile behavior of concrete is defined by a stress–strain
As the monotonic loading is subjected to the numerical model, σ c and relationship shown in Fig. 2(b). The tensile pre-failure behavior of
σ t are respectively equivalent to the uniaxial compressive and tensile concrete is expressed by the linear-elastic curve up to its tensile strength,
ones, σ c and σ t [41,42], which are described in Section 3.2. σ t0 . Besides, the tensile post-failure behavior of concrete is associated
To define the relationship between stress and strain for the non- with the strain-softening; as regards the tension stiffening effect is
linear deformation of concrete, flow rule plays a key role [32]. If the considered. According to the model proposed by software for the tension
orientation of plastic flow is normal to the yield surface, the plastic phase of concrete, the ultimate tensile strain is 10 times the cracking
potential function is the same as the yield function; this assumption is strain, εt0 [32]. It should be noted that the post-cracking capacity of
called the associated flow rule. However, experimental studies demon­ concrete is employed more as the amount of failure tensile strain in­
strate that the aforesaid assumption is highly questionable. Therefore, a creases. The value of εt0 can be estimated according to the following
non-associated flow rule is implemented in the CDP model [41,42]. The relations [32]:
plastic flow potential function is expressed as [32]: σ t0
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ εt0 = 1.8 (5a)
Ec
G = (ησ t0 tanΨ) + q2 − ptanΨ (3)
√̅̅̅̅
(5b)
′
In this equation, Ψ is the angle of dilatation measured in the p − q Ec = 0.043w1.5
c fc
plane at high confining pressure, η is a parameter to illustrate the ec­
where, Ec is Young’s modulus of elasticity, and wc is the specific
centricity of the plastic flow potential function with the asymptote, and
weight of concrete.
σ t0 is the uniaxial tensile stress at the failure point [32]. The flow po­
tential approaches to a straight line as the value of η goes closer and
3.3. Steel reinforcement properties
closer to zero [41,42]. In this study, Ψ and η were obtained 5◦ and 0.52,
respectively, based on the FE model calibration.
Steel reinforcement behaves the same as the elastic–plastic material
associated with the strain hardening (Fig. 3(a)). The steel material
3.2. Concrete properties properties are specified by the value of yield and ultimate strains, εy and
εu , yield and ultimate stresses, σ y and σu , and modulus of elasticity, Es . In
3.2.1. Uniaxial compressive behavior this study, the load transfer mechanism from steel reinforcement to
The uniaxial compressive behavior of concrete used herein is speci­ concrete and vice versa was based on the assumption that the perfect
fied by the stress–strain relationship taken from the modified Hognes­ bond between the reinforcing bars and its surrounding concrete exists.
tad’s model as shown in Fig. 2(a). According to this model, the To prevent numerical instability in the plastic phase, a small positive
stress–strain relationship can be divided into ascending second-order value was added to the slope of the strain–stress diagram in the plastic
parabola and descending first-order line. The constitutive law for the criteria [43].
first branch of the compressive stress–strain curve is defined as following

Fig. 2. Stress–strain relationship for concrete in; a) compression [43]; b) tension [32].

2526
N. Jafarian et al. Structures 28 (2020) 2523–2536

Fig. 3. Stress–strain relationship for; a) steel reinforcement; b) FRP [43].

3.4. FRP materials properties this inaccuracy are both the high-stress concentration localized at the
support interfaces and the corner lifting of the slabs [44]. Therefore, the
It was assumed that FRP materials are orthotropic and transversely slabs were modeled, including the supporting steel plates (stiff supports)
isotropic. It means that their mechanical properties were the same in along the edges. Furthermore, an interaction was implemented between
each direction perpendicular to the fibers. Furthermore, the best me­ the slabs and the supports, in which the penalty-based approach was
chanical properties were in the direction parallel to the fibers. The governed. The basis of the penalty contact stiffness value is dependent
behavior of FRP materials might generally obey the stress–strain curve on the tangential behavior, and it was taken to be rough in this study. An
shown in Fig. 3(b); the important characteristics of these materials infinite coefficient of friction was assumed to prevent the slip between
include ultimate stress and strain, εf and εf , and Young’s modulus, Ef the slabs and the supports [25].
[43]. The concentrated load was monotonically subjected to the center
point of the slabs, where a stiff steel plate was centrally placed on them.
3.5. Element specifications To model the loading scheme, the vertical displacement corresponding
to the maximum load was applied to the slabs, more precisely, the
The element of concrete slab was an eight-node linear brick element displacement control method was implemented to investigate the post-
called C3D8; it was a three-dimensional continuum (solid) element with peak behavior of the slabs.
eight nodes and three degrees of freedom at each node, and it also
employed linear interpolation in each direction. The C3D8 element 3.7. Nonlinear solution
could capture the cracking in concrete under tensile loading and the
rupture of concrete under compressive loading. Using the C3D8 element A general-purpose FE analyzer named Abaqus/Standard was used
provide conditions for the steel reinforcement to be taken as volume herein to employ an implicit integration scheme. This analyzer uses a
fraction in three perpendicular directions. To achieve better conver­ solution technique to solve static and low-speed dynamic problems, in
gence in the nonlinear analysis, the crushed elements were eliminated which the accuracy of the stress solutions is vitally important. In this
from the other ones [25]. study, the Riks method formed the basis of the solution technique in the
For steel reinforcement, FRP rods, and grids, a two-node linear truss nonlinear analysis in such a way that it provides solutions without re­
element was used; it was a three-dimensional link element with two gard to whether the response is stable or not. This method implements
nodes, where each node had three degrees of freedom. This element was the loading magnitude as an additional unknown quantity, which solves
capable of manifesting the mechanical properties of reinforcement in the for loads and displacements at the same time. As the load magnitude is
nonlinear analysis [25]. part of the solving technique, another quantity is introduced as a scalar
The mesh size was selected by the mesh sensitivity analysis; it could to control the solution progression. For this reason, the software em­
be inferred that it was not any considerable fluctuations in the results of ploys the arc length along the static equilibrium path in the space of
the FE model as the size of the chosen mesh is declined to a greater load–displacement. The arc length method is introduced as a capable
degree [25]. In the present research, the mesh of steel reinforcement and numerical technique to solve the highly complex systems of the
FRP elements had the minimum value of 20 × 20, however, the mesh nonlinear equations [25].
size of the concrete was up to the maximum value of 40 × 40. Besides, The nonlinear analysis process was done step-by-step. The maximum
the aspect ratio defined as the ratio of the shortest length of the element displacement value at the loading position was considered as the fin­
to the longest one was equal to 1.0. ishing condition to terminate the analysis. Problem-solving continued
for the number of increments specified by the steps corresponding to the
3.6. Boundary conditions aforesaid value [45].

Herein, all of the RC slabs were simply supported along the four 4. Finite element model validity
edges. Due to the obvious symmetry of the selected slabs, a quarter of
them was taken to be modeled. All FE models had simple support along To prove the validity of the FE models, the results obtained by the
the two exterior edges. Additionally, these models had roller support finite element models were compared with the results of the experi­
along the two interior edges, where the value of shear force was equal to mental tests from available studies [16,18]. For this reason, three RC
zero, and the bending moment had the maximum value. If the aforesaid slabs were analyzed; one slab was served as the control specimen
support conditions are straightforwardly prescribed for the slabs, the without any modification, named CS40-3, while two others were
accuracy of the numerical results may be dissolved. The reasons behind strengthened with 24 strengtheners including CFRP rods and grids,

2527
N. Jafarian et al. Structures 28 (2020) 2523–2536

named FR3-24 and FG3-24, respectively. The comparisons were repre­ Table 1
sented by the load–displacement curves. Accordingly, the values of Mechanical properties of concrete used for experimental slabs [16,18].
punching shear strength of the analyzed slabs were compared to the Name Compressive strength, Tensile strength, Young’s modulus,
corresponding values of the tested ones. Furthermore, the failure modes MPa MPa GPa
of the analyzed slabs and the experimental tests were presented and CS40- 42.40 4.56 30.60
juxtaposed with each other. 3
FR3- 44.10 4.65 31.21
24
4.1. Properties of the experimental RC slabs FG3- 41.70 4.52 30.35
24
All the experimental slabs had identical length, width, and height,
and had the same reinforcement layout. The overall dimensions of the
slabs were 1200 × 1200 × 105 mm (square slabs) whereas the effective Table 2
flexural span was 1000 mm. The bottom and top steel reinforcing bars Mechanical properties of steel reinforcement used for experimental slabs
were taken as 11Φ16 and 11Φ6, respectively. The effective depth of the [16,18].
tested slabs was considered 74.1 mm. Therefore, the tensile and Diameter, Yield Ultimate Modulus of Ultimate
compressive steel reinforcement ratios of the tested slabs were 2.2% and mm strength, strength, MPa elasticity, GPa strain, %
0.35%, respectively. Fig. 4 illustrates the layout and reinforcement ar­ MPa
rangements of these slabs. Additionally, the mechanical properties of Φ16 400 500 210 12
concrete and steel reinforcement of the experimental slabs are described Φ6 400 500 210 12
in Tables 1 and 2, respectively.
The arrangement of the CFRP rods and grids was a combination of
the orthogonal and radial configurations, as can be seen in Fig. 5(a). The rate of 1 mm/min. The slabs were all simply supported. Finite element
CFRP rods discussed herein were produced by employing the pultrusion modeling was performed using the explanatory notes raised in Section 3.
process, composed of pitch-based carbon fibers and epoxy resin, in As an example, the finite element model of CS40-3 and its corresponding
which the volume fraction of fibers was 65%. mesh are shown in Fig. 6.
As shown in Fig. 5(b), the diameter of CFRP rods was 12.5 mm,
which had ultimate strength, ultimate strain, and Young’s modulus of 4.2. Comparison of the results of finite element modeling and
1400 MPa, 0.015, and 120 GPa, respectively. Besides, the CFRP grids experimental tests
given herein were 140 mm in length, with a rectangular section of 5.0
and 3.5 mm, respectively, in width and depth. The grids had two side Fig. 7 demonstrates the load–displacement curves of the FE models
wings with length and width of 8.0 and 5.0 mm, respectively (Fig. 5(c)). and the corresponding tested slabs. The difference between the punch­
Furthermore, the CFRP grids had an ultimate strength of 1400 MPa, an ing shear capacities obtained by finite element modeling and experi­
ultimate strain of 0.015, and Young’s modulus of 100 GPa. mental tests was, therefore, in the value range of 0.1%–5%.
The vertical loading was monotonically subjected to the stiff steel According to the experimental results, brittle punching shear,
plate with a size of 150 × 150 × 30 mm until the occurrence of the flexural-induced punching, and combined flexural-shear were reported,
punching shear failure of the experimental slabs. Furthermore, the respectively, for the failure mode of CS40-3, FR3-24, and FG3-24;
displacement control procedure was used for loading with a constant similar failure modes were observed in the equivalent FE models as

Fig. 4. Geometry and reinforcement details of the experimental slabs (L1 to L5 refer to LVDTs, and C1 to C4, 1 to 10, and SG1 to SG4, respectively, refer to strain
gauges on concrete, on bottom reinforcement, and FRP grids) [16,18].

2528
N. Jafarian et al. Structures 28 (2020) 2523–2536

Fig. 5. Shear strengthening elements used for experimental slabs; a) pattern; b) CFRP rods; c) CFRP grids [16,18].

Fig. 6. Specimen CS40-3; a) FE model; b) mesh utilized.

Fig. 7. Comparison of load–displacement curves of the FE models and corresponding experimental specimens; a) CS40-3; b) FR3-24; c) FG3-24.

well. To specify the failure mode of the FE model, the damaged elements carrying capacity. The propagation of inclined cracks at 45◦ was
were investigated precisely, where the elements with high stiffness proved that the failure mode of CS40-3 is punching shear regarding the
reduction manifest the crack propagation path. In Fig. 8, for instance, basic understanding gained in Section 2.2. It was concluded that the
the stiffness degradation contour of CS40-3 is presented at its load- proposed finite element model is capable to predict the behavior of slabs

2529
N. Jafarian et al. Structures 28 (2020) 2523–2536

Fig. 8. Stiffness degradation contour of specimen CS40-3.

strengthened with vertical reinforcing elements, especially FRP grids. to 100 mm. Thus, the analyzed slabs had tensile and compressive flex­
ural reinforcement ratios of 2.54% and 0.63%, respectively. The failure
5. Specimens under investigation mode of all the strengthened slab specimens was punching shear due to
their relatively high reinforcement ratio. The mechanical properties of
To perform parametric studies, the calibrated finite element model concrete and the reinforcement of the slab specimens are listed in
proposed in Section 4 was employed. Additionally, the important points Table 3.
discussed in Section 3 about finite element modeling were completely It should be emphasized that the shear behavioral characteristics of
implemented in the following case studies. Therefore, 13 simply sup­ the analyzed slabs were presented, in which the results obtained from
ported RC flat slabs were analyzed, in which one was served as the the strengthened slabs were compared to those got from the control one;
control specimen without any strengthening against the punching shear the aforesaid characteristics were defined as punching shear strength,
failure, and 12 others were strengthened with the FRP grids. Strength­ displacement at the failure point, and energy dissipation capacity. The
ened slabs were categorized into two main groups regarding the mate­ energy dissipation capacity of the slabs was calculated by using the area
rial of the fibers including glass and carbon. In each group, half of the under the load–displacement curve.
slab specimens had the orthogonal pattern of the FRP grids whereas
others had the radial pattern. Besides, each configuration of the FRP
5.1. Control slab
grids differed in spacing as 30, 40, and 50 mm.
The geometry and reinforcement layout of all the analyzed slabs
Fig. 10(a) shows the load versus displacement of the control FE
were the same. The dimension of the slabs was 1800 × 1800 mm with a
model of the slab specimen. From zero to 18.4 mm displacement, there is
thickness of 125 mm, and the effective length of the span was taken to be
a marked rise in load. Over the next displacements, there is a sharp
1600 mm (Fig. 9). The tensile and compressive flexural reinforcing bars
decrease. According to Fig. 10(a), the load capacity and energy dissi­
were equal to 18Φ18 and 10Φ12, respectively. The clear concrete cover
pation capacity of the control model were obtained 401.4 kN and
of bars was considered to be 16 mm, where the effective depth was equal
4468.0 mm2, respectively.

5.2. Strengthened slabs

The configuration and geometry of the FRP grids are illustrated in

Fig. 11; the notation “s” defines the spacing or maximum distance be­
tween the vertical shear strengthening elements. Besides, the mechani­
cal properties of FRP grids are described in Table 4.
According to the ACI 318–19 [46], the maximum distance or spacing
from the loading area to the first peripheral line of the shear reinforcing
elements must not exceed d/2, along the perpendicular direction to the
loading area, where d is the effective depth of the slab. Additionally, the
constant spacing between the peripheral lines of the shear reinforcing
elements could not be greater than3d/4, in the perpendicular direction
to the loading area. Furthermore, the spacing between adjacent shear
reinforcing elements on the peripheral line nearest to the loading area
must not more than 2d, through the parallel direction to the loading area
[46]. Following the aforesaid points expressed by ACI 318–19 [46], each
slab group is specified as expressed in Table 5; where, the total area ­
of external shear reinforcing elements is identical for all the FE models.
Fig. 10(b) and (c) present the load–displacement curves of the slabs
strengthened with the GFRP and CFRP grids in the orthogonal and radial
patterns. The graphs show changes in the load since the displacement is
Fig. 9. Geometry of case studies. zero, and forecast trends up until the ultimate displacement. However,

2530
N. Jafarian et al. Structures 28 (2020) 2523–2536

Table 3
Mechanical properties of concrete and steel reinforcement used for the specimens investigated.
Material Compressive strength, MPa Tensile strength, MPa Yield strength, MPa Ultimate strength, MPa Ultimate strain, % Young’s modulus, GPa

Concrete 25 3.5 – – – 23.5

Steel – – 400 500 12 210

Fig. 10. Load-displacement curves of the specimens under investigation; a) control slab; b) first group; c) second group.

the value of load and corresponding displacement varies with the external reinforcement, the greater is the perimeter of the critical sec­
spacing of the FRP grids. The most obvious point is that the load in tion; therefore, punching shear stress decreases as the value of spacing
strengthened slabs was substantially more than that in the non- increases, according to the bigger critical section. Finally, the shear
strengthened slab based on the comparison of Fig. 10. The slope of all characteristics of the RC flat slabs have been improved by the spacing
the load–displacement diagrams in the elastic region was the same. increase.
From zero displacement to the displacement of failure point, there was Regarding FRP arrangements, both slabs strengthened with the
an abrupt rise in the load, followed by a sharp decline. orthogonal and radial configurations followed a similar pattern, how­
The slabs strengthened with the FRP grids at the spacing of 30 mm to ever, the load rate of the slabs strengthened with the radial layout of the
50 mm followed a similar pattern from zero displacements to the FRP grids was consistently higher than those strengthened with the
displacement at failure, which was characterized by a gradual increase. orthogonal one.
It was concluded that the more the value of spacing, the greater are the The increase in load was more rapid when the CFRP grids were used
load capacity and equivalent displacement. There was a substantial rise in comparison to GFRP grids, before dropping dramatically in
in the load capacity and corresponding displacement among the larger displacement at failure. In other words, CFRP implemented an upward
spaces. It should be noted that with the same amount of the vertical trend, overtaking that of GFRP.
shear strengthening elements, the more the value of the spacing of the Table 6 provides some information on the characteristics of the

2531
N. Jafarian et al. Structures 28 (2020) 2523–2536

Fig. 11. FRP grids; a) orthogonal pattern; b) radial pattern; c) geometry; numbers refer to strain measurement.

Table 4 Table 6
Mechanical properties of FRP grids used for the specimens under investigation Shear behavioral characteristics of the specimens under investigation.
[43]. Name Punching shear Displacement at the Energy dissipation
Type Ultimate strength, MPa Ultimate strain, % Young’s modulus, GPa strength, MPa failure point, mm capacity, J

GFRP 1150 2.5 46 GGO30 466.1 21.4 6025

CFRP 1400 1.2 147 GGO40 494.2 24.4 7571
GGO50 548.0 33.3 12,015
GGR30 534.1 29.0 9877
GGR40 545.3 32.5 11,567
Table 5 GGR50 562.6 36.2 13,693
Parametric characteristics of the specimens under investigation. CGO30 505.6 23.8 7283
CGO40 557.0 32.7 11,964
Group Name Pattern Spacing, mm CGO50 597.4 38.9 15,611
First GGO30 Orthogonal 30 CGR30 556.0 34.5 12,671
GGO40 Orthogonal 40 CGR40 584.0 39.7 15,683
GGO50 Orthogonal 50 CGR50 611.4 44.1 18,313
GGR30 Radial 30
GGR40 Radial 40
GGR50 Radial 50 punching shear of the specimens under investigation based on Fig. 10.
Second CGO30 Orthogonal 30
The information covers the value of the increase in punching shear
CGO40 Orthogonal 40 strength of the strengthened FE models in comparison with the control
CGO50 Orthogonal 50 FE model, as well as displacement at the failure point and energy
CGR30 Radial 30 dissipation capacity. The statistics are divided into two main groups
CGR40 Radial 40
based on the material properties of the FRPs.
CGR50 Radial 50
From 30 mm to 50 mm spacing, there was a significant rise in the
punching shear strength increase for orthogonal and radial patterns,

2532
N. Jafarian et al. Structures 28 (2020) 2523–2536

respectively, rising from 466.1 to 597.4 kN and 534.1 to 611.4 kN. This charts show the ratio of the load capacity and energy dissipation ca­
growing trend was continuing for displacement at the failure point and pacity of both slabs groups to the control one. There was a steep ratio
energy dissipation capacity of the slabs strengthened with the orthog­ increase in the load capacity and energy dissipation capacity as the fiber
onal configuration of the FRP grids; the enhancement of the energy changes from glass to carbon. The graph also shows that the slabs
dissipation capacity increased sharply from 6025 J in 21.4 mm strengthened with the radial pattern had far much load capacity and
displacement to 15611 J in 38.9 mm displacement per 30 and 50 mm energy dissipation capacity than the orthogonal one. This trend was
spacing. In this manner, the energy dissipation capacity of the slabs similar for larger spacing in comparison to the smaller one.
strengthened with the radial configuration of the FRP grids grew more In specimen CGR40, the increases in load capacity and energy
than those strengthened with the orthogonal ones; from 9877 J in 29.0 dissipation capacity were 45% and 251%, respectively; while in the
mm displacement to 18313 J in 44.1 mm displacement, respectively, at specimen CGR50, the same increases were 52% and 310%, respectively.
the spacing of 30 and 50 mm. In CGO50, the load and energy dissipation capacities rose to over 37%
Regarding the type of the FRP grids, shear behavioral characteristics and 169%, respectively, and in CGO50 they raised to just over 49% and
of the slabs strengthened with the CFRP grids were overwhelmingly 249%. Specimen CGO30 had the smallest amount of load capacity and
more than the corresponding figures of those strengthened with the energy dissipation capacity increase, with just over 16% and 35%,
GFRP ones. For instance, the values of punching shear strength, corre­ respectively. These parameters rose to almost 26% and 63% in specimen
sponding displacement, and energy dissipation capacity in orthogonal CCO30, respectively, which show a large improvement.
layout at 40 mm spacing decline, respectively, from 494.2 kN, 24.4 mm, To evaluate the functionality of the FRP grids, their strain values
and 75,751 J to 557 kN, 32.7 mm, and 11,964 J, as the type of grids were extracted under the implemented displacements. It was not rec­
changes from GFRP to CFRF; as well as these values in radial layout ommended to analyze all of the FRP grids in each slab due to the sym­
convert from 545.3, 32.5 mm, and 11567 J to 584.0 kN, 39.7 mm, and metry of the strengthened slabs. This symmetry was considered based on
15,683. the configuration of the FRP grids. The slabs strengthened with the
Additionally, the increase in the shear behavioral characteristics of radial configuration of the FRP grids were symmetric to the horizontal
the slabs strengthened with the orthogonal pattern of the FRP grids was and vertical lines passing through the center of the slab sides. Therefore,
slightly less than those strengthened with the radial one. The punching eight FRP grids were taken to be analyzed in the radial configuration as
shear strength of the slabs strengthened with the orthogonal and radial shown in Fig. 11(b). Furthermore, the symmetry of the slabs strength­
patterns of the GFRP grids, respectively, are peaked at 548.0 kN in 33.3 ened with the orthogonal configuration of the FRP grids was defined
mm displacement and 562.6 kN in 36.2 mm displacement. Furthermore, based on the diameters of the slabs in addition to the aforementioned
the maximum value of the energy dissipation capacity of the slabs lines for the orthogonal one. Thus, the number of the FRP grids was four
strengthened with the orthogonal pattern of the GFRP grids was equal to in the orthogonal configuration for the strain measurement (Fig. 11(a)).
12,015 J, as the equivalent value for those strengthened with the radial In Fig. 13, the graphs show the strain change between the zero dis­
one was 13,693 J. By contrast, the values of punching shear strength, placements to the displacements corresponding to the load capacities.
displacement at failure, and energy dissipation capacity of the slabs Throughout the loading, the largest quantity of strain was for FRP grids
strengthened with the orthogonal pattern of the CFRP grids, respec­ with number 1 in the orthogonal configurations, and FRP grids with
tively, were dived towards 505.6 kN, 23.8 mm, and 7283 J at 30 mm numbers 1 and 5 in the radial configurations. However, the strain
spacing as well as 556 kN, 34.5 kN, and 12671 J for radial one. measured in the other FRP grids was minimal up to the displacements at
In Fig. 12, the bar charts provide information from load- the failure points.
displacement curves shown in Fig. 10 for choosing the best perfor­
mance of the strengthening cases among two different groups. The

Fig. 12. Discrepancy between the shear behavioral characteristics of the strengthened slabs and control specimen; a) first group; b) second group.

2533
N. Jafarian et al. Structures 28 (2020) 2523–2536

Fig. 13. Strains in FRP grids for the specimens under investigation; a) GGO30; b) GGO40; c) GGO50; d) GGR30; e) GGR40; f) GGR50; g) CGO30; h) CGO40; i)
CGO50; j) CGR30; k) CGR40; l) CGR50.

2534
N. Jafarian et al. Structures 28 (2020) 2523–2536

6. Conclusions [3] Widianto W. Rehabilitation of reinforced concrete slab-column connections for

two-way shear. PhD Thesis. USA: Civil Engineering Department, University of
Texas at Austin, Texas; 2006.
The purpose of the present research was to investigate the effects of [4] Wood, J.G.M., “Pipers Row car park, Wolverhampton: Quantitative study of the
the FRP grids on the punching shear behavior of the slabs by employing causes of the partial collapse on 20th March 1997”, SS&D Contract Report to HSE,
the FE method. Type, configuration, and the maximum distance of the Northbridge House, Chiddingfold, Surrey GU8 4UU, England, 2010.
[5] Torabian A, Isufi B, Mostofinejad D, Pinho Ramos A. Flexural strengthening of flat
FRP grids were taken altogether as the critical parameters under slabs with FRP composites using EBR and EBROG methods. Eng Struct 2020;211:
investigation. Overall, 13 RC flat slabs were modeled and monotonically 110483.
subjected to the concentric vertical load; one slab as the control spec­ [6] Shaaban AM, Gesund H. Punching shear strength of steel fiber reinforced concrete
flat plates. ACI Struct J 1994;91(4):406–14.
imen and 12 slabs strengthened with the orthogonal and radial patterns [7] Binici B, Bayrak O. Punching shear strengthening of reinforced concrete flat plates
of the GFRP and CFRP grids spaced at 30, 40 and 50 mm. The main shear using carbon fiber reinforced polymers. J Struct Eng 2003;129(9):1173–82.
characteristics of the brittle punching behavior of the FE models, [8] Casadei, P., Nanni, A., and Ibell, T. , “Experiments on two-way RC slabs with
openings strengthened with CFRP laminates”, Center for Infrastructure Engineering
including the punching shear strength, displacement at the failure point, Studies, University of Missouri-Rolla, Rolla, Missouri, United States, Report No. 03-
and energy dissipation capacity, were extracted and compared. Based on 39, 2003.
the results obtained from FE modeling, the following conclusions may be [9] Esfahani MR, Kianoush MR, Moradi AR. Punching shear strength of interior
slab–column connections strengthened with carbon fiber reinforced polymer
drawn. sheets. Eng Struct 2009;31(7):1535–42.
[10] Soudki K, El-Sayed AK, Vanzwol T. Strengthening of concrete slab-column
1. The FRP grids significantly improved the punching shear behavior of connections using CFRP strips. J King Saud Univ. – Eng Sci 2012;24(1):25–33.
[11] Labibzadeh M. The numerical simulations of the strengthened RC slabs with CFRPs
the strengthened slabs. The increase in punching shear strength of
using standard CDP material model of Abaqus code. Eur J Environ Civil Eng 2015;
the slabs strengthened with the FRP grids was in the range of 16% to 19(10):1268–87.
52% in comparison with the control specimen, while the corre­ [12] Labibzadeh M, Elahifar T. An enhanced finite element model for reinforced
sponding increase in the energy dissipation capacity was from 35% concrete two-way slabs strengthened with carbon fiber reinforced polymers. Struct
Eng Int 2015;25(1):81–90.
to 310%. [13] El-Salakawy EF, Polak MA, Soudki KA. New strengthening technique for concrete
2. Predictably, the performance of the CFRP grids on the punching slab-column connections. ACI Struct J 2003;100(3):297–304.
shear behavior of the slabs was better than that of GFRP girds. In the [14] Bu W, Polak MA. Effect of openings and shear bolt pattern in seismic retrofit of
reinforced concrete slab–column connections. Eng Struct 2011;33(12):3329–40.
slabs with CFRP grids, the maximum increase in the punching shear [15] Torabian A, Isufi B, Mostofinejad D, Pinho Ramos A. Shear and flexural
capacity was obtained as 52% compared to that obtained for the strengthening of deficient flat slabs with postinstalled bolts and CFRP composites
control; whereas, this increase was 40% for the slabs with the GFRP bonded through EBR and EBROG. Struct Concr 2020:1–18. https://doi.org/
10.1002/suco.202000236.
grids. Furthermore, the maximum increases in the energy dissipation [16] Meisami MH, Mostofinejad D, Nakamura H. Punching shear strengthening of two-
capacity of the slabs strengthened with the GFRP and CFRP grids in way flat slabs using CFRP rods. Compos Struct 2013;99:112–22.
comparison to the control specimen were 206% and 310%, [17] Li R, Cho YS, Zhang S. Punching shear behavior of concrete flat plate slab
reinforced with carbon fiber reinforced polymer rods. Compos B Eng 2007;38(5):
respectively. 712–9.
3. The radial layout of the CFRP grids was introduced as by far the best [18] Meisami, M.H., Mostofinejad, D., and Nakamura, H., “Punching shear
shear strengthening procedure concerning the increased load ca­ strengthening of two-way flat slabs with CFRP grids”, Journal of Composites for
Construction, Vol. 18, 2, pp. 04013047-1-10, 2014.
pacity and ductility enhancement. Compared with the control slab,
[19] Meisami MH, Mostofinejad D, Nakamura H. Strengthening of flat slabs with FRP
the strengthened slab with the orthogonal pattern of the FRP grids fan for punching shear. Compos Struct 2015;119:305–14.
exhibited a maximum increased punching shear capacity and energy [20] Marques MG, Liberati EAP, Pimentel MJ, de Souza RA, Trautwein LM. Nonlinear
dissipation capacity of 39% and 168% at the spacing of 40 mm, finite element analysis (NLFEA) of reinforced concrete flat slabs with holes.
Structures 2020;27:1–11.
respectively; these values were 45% and 251% for the radial layout [21] Genikomsou AS, Polak MA. Finite element analysis of punching shear of concrete
at the same spacing. slabs using damaged plasticity model in ABAQUS. Eng Struct 2015;98:38–48.
4. The effect of the radial configuration on the improvement of the [22] Shu J, Fall D, Plos M, Zandi K, Lundgren K. Development of modelling strategies for
two-way RC slabs. Eng Struct 2015;101:439–49.
punching shear behavior of the slab specimens strengthened with the [23] Genikomsou, A., S., and Polak, M.A., “Finite-element analysis of reinforced
GFRP grids was greater than that in the specimens strengthened with concrete slabs with punching shear reinforcement”, Journal of Structural
CFRP grids. Hence, as the values of mechanical properties decrease, Engineering, Vol. 142, 12, pp. 04016129, 2016.
[24] Genikomsou AS, Polak MA. 3D finite element investigation of the compressive
the importance of the FRP configuration in the strengthening in­ membrane action effect in reinforced concrete flat slabs. Eng Struct 2017;136:
creases. For instance, at the spacing of 30 mm, it was observed that 233–44.
the difference between the percentages of the maximum punching [25] Stein T, Ghali A, Dilger W. Distinction between punching and flexural failure
modes of flat plates. ACI Struct J 2007;104(3):357–65.
shear capacity increase of the orthogonal and radial configurations of [26] Ebead U, Marzouk H. Fiber-reinforced polymer strengthening of two-way slabs.
the CFRPs (i.e., 13%) was less than that of the GFRPs (i.e., 17%). ACI Struct J 2004;101(5):650–9.
5. The maximum strain values recorded in the grids ranged approxi­ [27] Michel L, Ferrier E, Agbossou A, Patrice H. Flexural stiffness modelling of RC slab
strengthened by externally bonded FRP. Compos Part B-Eng 2009;40:758–65.
mately from 0.006 in GFRPs to 0.012 in CFRPs. Accordingly, the
[28] Elstner RC, Hognestad E. Shearing strength of reinforced concrete slabs. ACI J Proc
superiority of the CFRP grids over the GFRP grids in strengthening 1956;53(7):29–58.
the slabs was proved in the aspect of effective strain capacity as well [29] Ghali A, Gayed RB. Universal design for punching resistance of concrete slabs. ACI
as punching shear strength and ductility. Struct J 2019;116(1):207–12.
[30] Rochdi EH, Bigaud D, Ferrier E, Hamelin P. Ultimate behavior of CFRP
strengthened RC flat slabs under a centrally applied load. Compos Struct 2006;72
Declaration of Competing Interest (1):69–78.
[31] Jahanshahi M, Khoei AR, Heidarzadeh N, Jafarian N. A hierarchical thermo-
mechanical multi-scale technique for modeling of edge dislocations in nano-
The authors declare that they have no known competing financial crystalline structures. Comput Mater Sci 2018;141:360–74.
interests or personal relationships that could have appeared to influence [32] ABAQUS Inc., ABAQUS, theory user manual, version, Dassault Systèmes Simulia
Corp 2020 Providence RI, United States 2020.
the work reported in this paper. [33] Dede T, Ayvaz Y. Comparative study of plasticity models for concrete material by
using different criteria including Hsieh–Ting–Chen criterion. Mater Des 2010;31
References (3):1482–9.
[34] Dede T, Ayvaz Y. Plasticity models for concrete material based on different criteria
including Bresler–Pister. Mater Des 2010;31(1):278–86.
[1] Torabian A, Isufi B, Mostofinejad D, Ramos AP. Behavior of thin lightly reinforced
[35] Babua RR, Benipal GS, Singh AK. Constitutive modeling of concrete: an overview.
flat slabs under concentric loading. Eng Struct 2019;196:109327.
Asian J Civil Eng (Build Housing) 2005;6(4):211–46.
[2] Mostofinejad D, Jafarian N, Naderi A, Mostofinejad A, Salehi M. Effects of openings
[36] Babu RR, Benipal GS, Singh AK. Plasticity-based constitutive model for concrete in
on the punching shear strength of reinforced concrete slabs. Structures 2020;25:
stress space. Latin Am J Solids Struct 2006;3:417–41.
760–73.

2535
N. Jafarian et al. Structures 28 (2020) 2523–2536

[37] Labibzadeh, M., and Sadrnejad, S., “Mesoscopic damage based model for plane [42] Lee J, Fenves GL. Plastic-damage model for cyclic loading of concrete structures.
concrete under static and dynamic loadings”, American Journal of Applied J Eng Mech 1998;124(8):892–900.
Sciences, Vol. 3, pp. 2006. [43] Mostofinejad, D., Reinforced concrete structures, 2th Ed., Arkan Danesh
[38] Sadrnejad SA, Labibzadeh M. A continuum/discontinuum micro plane damage Publications, Isfahan, Iran, 2018; [in Persian].
model for concrete. Int J Civil Eng 2006;4:296–313. [44] Enochsson O, Lundqvist J, Täljsten B, Rusinowski P, Olofsson T. CFRP strengthened
[39] Poinard C, Malecot Y, Daudeville L. Damage of concrete in a very high stress state: openings in two-way concrete slabs – an experimental and numerical study. Constr
Experimental investigation. Mater Struct 2010;43:15–29. Build Mater 2007;21(4):810–26.
[40] Mostofinejad D, Mohammadi Anaei M. Effect of confining of boundary elements of [45] Jahanshahi, M., Khoei, A.R., Jafarian, N., and Heidarzadeh, N., “An atomistic-
slender RC shear wall by FRP composites and stirrups. Eng Struct 2012;41:1–13. continuum multiscale method for modeling the thermomechanical behavior of
[41] Lubliner J, Oliver J, Oller S, Oñate E. A plastic-damage model for concrete. Int J heterogeneous nanostructures”, Vol. 16, 5, pp. 441-464, 2018.
Solids Struct 1989;25(3):299–326. [46] ACI. 318–19, Building code requirements for structural concrete and commentary.
Farmington Hills, MI, USA: American Concrete Institute; 2019.

2536