Journal of Constructional Steel Research 198 (2022) 107585

Contents lists available at ScienceDirect

Journal of Constructional Steel Research

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

Modelling of demountable steel-concrete composite connections:

Validation of finite element model and parametric study
Abdolreza Ataei *
Department of Civil Engineering, University of Isfahan, Isfahan, Iran

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

Keywords: A three-dimensional finite element modelling and parametric analysis of a composite connection having pre­
Bolt shear connectors fabricated concrete slabs and preload bolted shear connectors are presented in this study. The load-slip rela­
Finite element model tionship for this type of shear connection subjected to monotonically loading with the effect of different
Deconstructability
parameters including bolted shear connector size, bolt grade, hole size in precast concrete slab and steel beam
Sustainable construction
Steel-concrete composite connection
top flange, concrete strength, and bolt preloading force is studied by means of ABAQUS software. A finite
Composite beam element model of the composite connection considering the non-linearities of the geometry, interfaces between
Push-out tests the components, and materials is developed and verified against the results of the experimental study. An
extensive parametric study is conducted, and the influences of different parameters on the structural behaviour
of demountable steel-concrete composite shear connection are discussed in detail. In addition, a comparison of
the estimated values obtained from design equations proposed by current standards, including EC4 and AISC for
headed stud shear connectors, and the design recommendations provided in other studies for bolted shear
connectors are conducted. According to the calculation result, the accuracy of different design methods is
evaluated. Finally, a regression analysis of the data produced by the FE analysis has been employed to propose a
new design equation to determine the ultimate strengths for the post-installed preload bolt shear connectors. It is
concluded that the ultimate shear load capacity values predicted by EC4 and AISC provisions for composite
connections having prefabricated concrete slabs and preload bolted shear connectors found to be relatively less
accurate but conservative.

1. Introduction by means of bolt shear connectors. Accordingly, almost all components

of the steel-concrete composite beams and joints can be dismantled and
For many decades, steel-concrete composite floor system has been reused at the end of their service life. Fig. 1 shows a schematic layout of
known as an excellent choice to be used in the construction of bridges demountable composite frame building skeleton, and Figs. 2 and 3 show
and buildings. This system has superior strength and stiffness compared two generic composite beams and beam-to-column composite joint,
with bare steel beams. Mechanical shear connectors such as headed stud respectively. In this new system, the precast concrete slabs are attached
shear connectors which have been widely used in composite structure to the steel girders by using bolt shear connectors to gain shear
constructions to resist vertical uplift forces and horizontal shear. Headed connection in the direction of the secondary beam (see Fig. 2). The
stud shear connectors are commonly used worldwide due to their high precast concrete planks can be interconnected by shear keys, as shown in
degree of automation on the site or in the workshop. However, steel- Fig. 4, to maintain robustness and to carry the shear forces mobilised
concrete composite beams having welded headed stud shear connec­ between the concrete slab and steel beam. In order to prevent rotation of
tors cannot be disassembled, and their components cannot be reused at the precast concrete slab around the steel beam (Fig. 2 (a)), post-
the end of their service life. tensioned bars can be used. The secondary beams can be connected to
In recent years, many experimental and analytical researches have primary beams using a simple bolted connection. Furthermore, a bolted
been conducted on different types of new shear connectors to facilitate flush end-plate joint (see Fig. 3) for a primary beam to a supporting
the demountability of steel-concrete composite beams. In recently pro­ column connection can be used to facilitate deconstructability at the end
posed methods, concrete slabs are generally connected to the steel beam of structure service life or replacement of the structural components

* Corresponding author at: School of Civil and Environmental Engineering, UNSW Sydney, NSW 2052, Australia.
E-mail address: a.ataei@eng.ui.ac.ir.

https://doi.org/10.1016/j.jcsr.2022.107585
Received 7 May 2022; Received in revised form 12 September 2022; Accepted 20 September 2022
Available online 28 September 2022
0143-974X/© 2022 Elsevier Ltd. All rights reserved.
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

during the life of the structure. and bolts in the push-out tests [8]. It was concluded that headed studs
Several studies examined significant issues in using bolted shear and bolted shear connectors have similar shear resistance, while headed
connectors in composite connection, beam, and beam to column joint. studs are more ductile. Ataei et al. [9–15] conducted extensive experi­
Steel-concrete composite connections made of reinforced concrete slabs mental studies and numerical modelling of steel-concrete composite
connected to the top flange of steel beams by means of adhesive anchors, connections, beams, and beam to column joints having precast concrete
double-nut bolt shear, and fiction-grip bolt shear connectors have been slab with a variety of post-installed preload bolt shear connectors. The
proposed by Kwon et al. [1,2]. In this study, this type of composite test results confirmed that all composite connections, beams and joints
connection was experimentally evaluated to study the structural have significant ductility when subjected to static loading. It was also
behaviour in steel-composite composite bridge girders. Fiction-grip bolt confirmed that after loading within the service load range, all composite
shear connectors appeared to have a very high fatigue strength members could be deconstructed easily, and all components including
compared with headed stud shear connectors. Two different types of precast concrete slab, steel beam, and steel column can be reused again.
blind bolts (Hollo bolts and AJAX fasteners) were used by Mirza et al. [3] More recently, Zhang et al. [16] have conducted push-out tests to
to connect concrete slabs to the steel beam. The test results showed evaluate the failure modes, load-slip, and stiffness of bolted shear con­
almost the same shear resistance capacity for both bolt shear connectors. nectors in steel-concrete composite beams. It was concluded that the
Six push-out tests using two different types of blind bolts (three with strength capacity of the composite connection has been significantly
M20 Hollo-bolts and three with M20 AJAX fasteners) as shear connec­ influenced by the bolt size, bolt grade, and concrete strength. The au­
tors were carried out and reported by Pathirana et al. [4]. These con­ thors also developed a finite element model of this type of composite
nectors were proposed as alternatives for headed stud shear connectors connection, and the same results was observed. A new removable
to facilitate retrofitting and strengthening of existing composite bridge friction-based shear connector for precast steel-concrete composite
decks. The test results showed that the initial stiffness of the AJAX fas­ bridges was proposed by Karavasilis et al. [17]. In this study, the
teners is very low and this can lead to low composite action between the friction-based shear connector passes through countersunk holes drilled
concrete slab and the steel beam, and subsequently increase the ductility on the top flange of the steel beam. Eleven push tests were conducted to
as well as deflection of the composite beams under service-load condi­ investigate the shear load-slip displacement behaviour on this new type
tion. Pathirana et al. investigated the load-slip behaviour of blind-bolt of shear connector. The test results showed that the shear resistance and
and welded stud connectors in grout under retrofitted conditions. The slip capacity of this type of shear connector were much higher than those
test and FE analyse results provide an extended understanding of the for welded studs. An experimental study was conducted by Karavasilis
feasibility of utilising blind-bolts for retrofitting substandard composite et al. [18] to assess the flexural behaviour of a precast steel-concrete
beams. An experimental study on steel-concrete composite connections composite beam equipped with the friction-based shear connectors. It
using demountable shear connectors was conducted by Lam et al. [5]. was proved that the composite beam with the friction-based shear
The superior ductility of demountable steel-concrete composite con­ connectors was effective. Chen et al. [19] proposed an innovative and
nections has been confirmed by the test results obtained from this study advanced bolt shear connector in steel-concrete composite structures. In
when compared with that of headed stud shear connectors. Push-out this research, 22 push-out tests were conducted and the effect of
tests using machining headed stud shear connectors were conducted different parameters including bolt diameter, slab hole dimension,
by Dai et al. [6] to investigate the static behaviour of bolt shear con­ reserved hole constraint condition, and hole size were considered as the
nectors. A shear resistance capacity of about 84% of that of welded variables. It was concluded that bolt connectors with corrugated pipe
headed shear connectors with a 6 mm slip between the concrete slab and had higher shear stiffness than normal reserved hole types and cast-in-
steel beam was observed during the tests. Pavlovic et al. [7] conducted place slab specimens. Wang et al. [20] proposed a steel-concrete com­
four push-out tests using M16 single-nut embedded bolted shear con­ posite connection system that consists of precast concrete slabs con­
nectors to characterise the behaviour of this type of shear connector and nected to the top flange of steel beam by means of clamping bolt shear
the results were compared with push-out tests conducted on stud shear connectors. In this study, all composite connections were subjected to
connectors by other researchers. It was concluded that the shear resis­ low-cycle high amplitude cyclic loading. It was demonstrated that the
tance and stiffness of single-nut embedded bolted shear connectors were energy dissipation and ductility of the composite connection using
respectively about 95% and 50% of those of welded stud shear con­ clamping bolt shear connectors are excellent compared to those of the
nectors. An experimental study was conducted on welded headed studs composite connection with headed shear stud. More recently, a lockbolt

Fig. 1. Schematic layout of sustainable and deconstructable composite frame.

2
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

demountable shear connector was proposed by He et al. [21,22] for capable of simulating composite connection subject to monotonic
steel-concrete composite beams. The mechanical behaviour of this novel loading and to obtain accurate and precise results, a finite element
shear connector was investigated by four pushout tests. It was concluded model comprised of all composite connection components was used
that high shear resistance and high stiffness was achieved by this lock­ [25]. The steel beam, precast concrete slab, shear connectors, steel
bolt demountable shear connector when compared with other reinforcing bars and steel washers were considered in the finite element
demountable shear connector. The results obtained from experimental modelling. As the interaction between different components has a sig­
and numerical studies summarised above demonstrated that the appli­ nificant effect on the behaviour of the composite connection, non-
cation of prefabricated concrete panels connected to the steel beam by linearity interaction was modelled properly. In addition, geometric
means of bolted shear connectors can improve the demountability and nonlinearity and material nonlinearity were considered during the FE
sustainability of steel-concrete composite structures. However, it can be modelling. In the following sections, the details of the model develop­
concluded that the experimental and numerical studies on the structural ment including geometric model, constitutive law for all materials used
performance of demountable bolt shear behaviour in composite beams in the experimental study [25], element type and meshes, interaction
are limited, and further studies should be conducted to highlight the and constraint conditions, load application, load application, and
significant benefits that may be achieved with the use of demountable boundary conditions are described in detail.
bolt shear behaviour in the hogging moment regions of composite
beams. In addition, previous researches conducted on the demountable 2.1. Geometric model, element type and mesh
steel-concrete composite shear connection did not considered all the
effective parameters on the structural behaviour of this type of shear As the specimens were symmetric in terms of geometry, boundary
connector. Moreover, the feasibility of several design equations pro­ conditions, material properties, and to decrease the computational ef­
vided by different standards for demountable bolt shear connectors forts and time, one-quarter of the composite connection was simulated,
should be assessed in detail. as shown in Fig. 5. The prefabricated concrete slabs, steel beam, bolt
The existence of bolted shear connectors in the hogging moment shear connectors, and square washer were modelled using 8-node solid
regions of the composite beams is necessary to distribute the longitu­ elements with reduced integration (C3D8R) available in ABAQUS
dinal shear forces and to develop composite action between the concrete whereas, the steel reinforcement was simulated by means of two-node
slab and steel beam. Accordingly, the behaviour of the bolted shear linear three-dimensional truss elements (T3D2). In order to compro­
connectors has a significant effect on the behaviour of the overall frame. mise between the computational efficiency and accuracy and to choose
It is aimed in this study that the mechanical and structural behaviour of appropriate mesh size for all connection components, a sensitivity study
deconstructable bolted shear connectors subjected to a monotonic was carried out for the finite element simulation of the composite
loading in composite beams are investigated through a three- connection so that the model can predict local and global responses. A
dimensional finite element model. The effect of different parameters, fine mesh was employed for the region around the holes, steel washers,
including bolted shear connector size, bolt grade, hole size in the precast and bolts (see Fig. 6) to decrease the computational time and to improve
concrete slab and steel beam top flange, concrete strength, and bolt the accuracy of the results. Fig. 6 illustrates the outline of the mesh for
preloading force, is studied by means of ABAQUS software. A finite composite connection components.
element model of the composite connection considering the non-
linearities of the geometry and interfaces between the components 2.2. Material constitutive relationship
and materials is developed and verified against the results of the
experimental study. An extensive parametric study is carried out, and The accuracy of the constitutive regulations used to express the
the influences of different parameters on the structural behaviour of mechanical performance of the materials has significant influences on
demountable steel-concrete composite shear connection are discussed in the precision of the finite element model. In the following sections, the
detail. In addition, a comparison of the estimated values obtained from general constitutive regulations to define the stress-strain curves of the
design equations proposed by current standards including EC4 [23] and materials used in this study are outlined.
AISC [24] for headed stud shear connectors and the design recommen­
dations provided in other studies for bolted shear connectors are con­ 2.2.1. Concrete
ducted. Finally, a regression analysis of the data produced by the FE For the concrete, the material behaviour in both the compressive and
analysis has been employed to propose a new design equation to tensile state was modelled. The stress-strain relationship for the concrete
determine the ultimate strengths for the post-installed preload bolt shear in compression proposed by Carreira and Chu [26] was used in the
connectors. model developed in this study. According to Carreira and Chu's model,
the full stress-strain curves for short-term loading under uniaxial
2. Finite element model compression compose of two distinctive parts including a non-linear
parabolic range and an elastic range represented by the following
In this study, the finite element program ABAQUS was used to equation.
simulate the composite connection having demountable bolt shear
connectors and prefabricated concrete slabs. To construct a FE model

Fig. 2. Schematic layout of sustainable and deconstructable secondary composite beam; (a) Cantilever method, (b) Simply supported method.

3
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

⎧ ′
⎪ Eε, x ≤ 0.35 fc
⎪
⎪
⎪
⎪ ( )
⎨ ε
σc = fc γ
⎪ εc ′
⎪
⎪ ( )γ , x > 0.35 fc
⎪
⎪ ε
⎩γ − 1 +
εc
Fig. 4. Schematic outline of the shear keys for cantilever method.
where εc = 0⋅002, γ = (fc/32 ⋅ 4)3 + 1 ⋅ 55, fc is the mean cylinder
compressive strength (in MPa) and σ c and ε are the stress and strain of
pretension and main external load). In the experimental study, pre­
the core concrete respectively. For concrete in tension, it was assumed
fabricated concrete slabs were connected to the steel beam using preload
that the tensile stress was increased up to 10% of its compressive
bolted shear connectors. According to EC4 [23], the minimum bolt
strength and then dropped to 0.0 at a strain of about ten times of its
tension was applied to the all bolt shear connectors. Therefore, before
failure strain.
applying the main load on the composite specimen, the bolt was sub­
jected to a pretension load same as what was applied in the tests. The
2.2.2. Steel components
“BOLT LOAD” function available in ABAQUS [30] was used for this step
Fig. 7 shows the generalized stress-strain relationships representing
to apply the pretension load to the bolt. For the second step, the vertical
the properties of steel materials, including yield strength, ultimate strain
load was applied in a displacement-control mode.
and strength, and modulus of elasticity used in the FE model. The values
were standardized from the material laboratory tests which were re­
ported in the study conducted by Ataei et al. [25]. The experimental 2.5. Boundary conditions
mechanical properties of all the steel components are summarised in
Table 1. The boundary conditions applied to the FE model have significant
influences on the precision of the finite element model. Therefore, to
2.3. Constraint and interaction conditions match the boundary condition considered in the test set-up [25], it was
tried to simulate the boundary conditions as close as possible to the test
Different contacts were considered in this study to simulate the in­ set-up. As mentioned before, a double vertical symmetry boundary
teractions between components (viz steel beam, steel washers, shear condition was considered in this study, and only one-quarter of the
connector, reinforcements, and precast concrete slab) of the composite composite connection was simulated. Same as the test set-up, all nodes
connection. Between two components that interrelate with each other of surface 1 (the bottom face of the precast concrete slab) were
(for example: between the steel beam and precast concrete slab), an restrained from translating in X, Y and Z directions as shown in Fig. 8. In
appropriate interaction available in ABAQUS should be applied. The order to obtain the load for drawing load-slip curves, the vertical reac­
contact interfaces technique named the surface-to-surface contact and tion forces of this surface (i.e. surface 1) were used. On the symmetry
provided in ABAQUS was used to simulate the contact between all the plane (surfaces 2) located on the precast concrete slab and steel beam,
interfaces. In this technique, any two components were not allowed to all the nodes were restrained to translate in the Y direction and to rotate
penetrate to each other but to separate from each other. For the in the X and Z directions (i.e. UY=URX = URZ = 0). Moreover, on the
tangential and normal behaviour between two surfaces, the “PENALTY” symmetry plane (surfaces 3) located on the precast concrete slab and
and “HARD” options were employed, respectively. Friction coefficients steel beam, all the nodes were restrained to translate in X direction and
of 0.45 and 0.25 were considered for the steel-concrete and steel-steel to rotate in the Y and Z directions (i.e. UX = URY=URZ = 0).
interfaces, respectively [27–29]. In order to simulate the contact inter­
action between the reinforcements and prefabricated concrete slabs, 3. Experimental study
embedded constraints were employed, in which the reinforcing bars
were assumed to be the embedded region whereas, the prefabricated Twelve push-out tests were carried out by the author to study the
concrete slab was considered as the host region. By considering the structural behaviour of composite connections having prefabricated
Embedded features, it is assumed that the relative slip and debonding concrete slabs and bolt shear connectors [25]. The influence of bolt
between the reinforcements and concrete slab is zero and can be diameter and strength, concrete strength, and hole size in the concrete
assumed to be ignored. slab on the behaviour of composite connection in steel-concrete com­
posite beams were investigated in this experimental study. Results ob­
2.4. Loading and analysis method tained from this experimental study were used to validate the finite
element model and to recommend an accurate practical design method.
The Riks analysis method in ABAQUS was employed to solve the Fig. 9 shows the details of push-out test specimens designed [25]. As
nonlinear problem comprised material and geometric nonlinear anal­ shown in Fig. 9, two prefabricated concrete slabs, a standard universal
ysis. The loading process of the specimen included two steps (vis. Bolt IPE270 steel beam, and bolt shear connectors were employed to

Fig. 3. Schematic outline of the deconstructable beam-to-column composite joint; (a) with and (b) without concrete slab.

4
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 5. FE models geometry and model assemblage.

Fig. 6. Finite element mesh of FE model: (a) concrete slab, (b) steel section, (c) square washer and (d) model mesh overview.

construct each composite specimen. Concrete slabs were connected to 22 mm) and in steel beam (16, 20, and 25 mm) were considered in this
the steel beam flanges using two pairs of preload bolt shear connectors study. An assembled specimen ready to be tested is shown in Fig. 10.
to form the composite connection. The details of the push-out test The test set-up and loading arrangement for the tests are shown in
specimens are summarised in Table 2. As shown in Table 2, different Fig. 10. Steel plates with different thicknesses were placed underneath
concrete strength (i.e. 25 and 40 MPa), different bolt sizes (i.e. 12, 16 of the concrete slabs to align the specimens vertically and to bed the
and 20 mm) with two various bolt strength (i.e. Grade 8.8 and Grade concrete slabs. In addition, a thick steel plate was welded over the top of
12.9), and various hole diameters in precast concrete slab (14, 18, and the steel beam to produce a uniformly distributed load in the steel beam

5
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

was measured on both sides of the composite specimen. The separation

of concrete slabs from the steel beam was also measured by one LVDT, as
shown in Fig. 10.
The test results [25] showed that the general load-slip relationship of
deconstrcutable steel-concrete composite bolted shear connectors under
the monotonic loadings includes three distinctive stages including a
region of full shear interaction, a region with almost zero shear inter­
action between the steel beam and prefabricated concrete slabs, and a
partial shear interaction stage. At the first stage, the shear load increased
rapidly with almost zero slip, which indicated a very high stiffness due
to the pretensioning of the bolted shear connectors and producing me­
chanical friction between the steel beam and precast concrete slab. The
second stage started when the friction at the steel-precast concrete slab
was overcome. At the second stage, significant slip between the steel
Fig. 7. Material properties of steel components. beam and precast concrete slab occurred due to the clearance between
the bolted shear connectors and the holes in the precast concrete slab/
steel beam. Towards the end of this stage, a loud “bang” was heard when
Table 1
the increasing applied vertical load finally overcame the friction at the
Material properties of steel components.
interface. After the second stage, the applied load increased significantly
specimen Yield strength Tensile strength Elastic modulus
due to the bearing mechanism and the composite action between the
(MPa) (MPa) (GPa)
steel beam and prefabricated concrete slab is provided by bearing of the
Flange 352 531 201 bolts and precast concrete slab. The bolts then started to bear against the
Steel beam
Web 398 551 205
surface of the holes in the steel beam and precast concrete slab. The
8.8 830 955 201
Bolt behaviour of bolted shear connector at this stage was almost similar to
12.9 1115 1319 202
Reinforcement 420 610 198 welded headed stud shear connectors.
The test results showed that three failure modes including the shear
connector fracture, concrete crashing close to the bolted shear connector
and steel beam flange deformation around the oversized holes in the
steel beam were detected for the connections of the steel-concrete
composite beams. Fig. 12 shows the concrete crashing close to the bol­
ted shear connector and concrete cracks in the precast concrete slab for
the specimens PT7 and PT9. The failure modes of the specimens PT5 and
PT7 tested in this research are shown in Fig. 13 and Fig. 14, respectively.

4. Verification of finite element model

Verifications of the developed model against experimental results are

essential to determine the accuracy and reliability of the model.
Accordingly, in this section, results obtained from the FE models are
illustrated and compared to the test results to validate the FE models
developed in this study. As mentioned before, an extensive finite
element modelling was carried out to investigate the behaviour of
Fig. 8. Boundary conditions. demountable composite connections having prefabricated concrete
slabs and steel beams. The finite element model was firstly constructed
cross-section. As recommended by Oehlers [31], Oehlers and Bradford and compared against to the results obtained from the experimental
[32] and Loh et al. [33], the one slab was placed on a support that study. Then, models were constructed using validated models based on
prevented lateral movement, and the other slab rested on a roller sup­ the experimental results for conducting a systematic parametric study.
port of plate and rollers. It was expected to offer reasonable and yet Ductility and shear resistance criterions for demountable bolt shear
conservative results for the mean strength of the shear connectors on connectors are achieved using the parametric finite element models.
both sides of the slabs. It should be noted that any lateral movement of Force-slip relationships obtained from the FE models were compared
the hydraulic jack was effectively prevented by adequate bracing chains with those obtained from the tests, as shown in Fig. 15. In addition to
using steel chains placed on the sides of the hydraulic jack as shown in load-slip curves, essential parameters including the shear load at the first
Fig. 10. In order to produce a shear load in the bolt shear connectors, a bearing (Pf), the maximum value of slip in slipping stage (Sa), the
hydraulic jack with a maximum capacity of 1000 kN was used to apply a maximum shear load (Pm) and its corresponding slip (Pm), and the
vertical downward load to the steel beam. maximum slip (Su) obtained from the FE model and tests are summarised
Loading regime (Fig. 11) adapted from EC4 [23] was considered in in Table 3. Comparison the experimental results with the FEM pre­
this experimental study. As can be seen in Fig. 11, firstly, the specimen dictions in terms of different parameters was also made as shown in
was loaded to 40% of expected failure load (Fu). Then, force-controlled Fig. 16. The failure modes obtained from the tests and the FE prediction
cyclic load ranging from 5% to 40% of the expected failure load was are also compared to evaluate the ability of the model to predict the
applied 25 cycles. After the cyclic loading, the specimen was loaded behaviour of the composite connection in term of failure modes. The
monotonically in a displacement-control mode until the failure of the simulation prediction for specimen PT1 is illustrated in Fig. 17 (a). Based
specimen (Fig. 11). The last step loading was conducted so that failure of on the results obtained from the simulation, it can be demonstrated that
the specimen was not appeared in <15 min. the developed FE models are able to predict the behaviour of the com­
Each test specimen was equipped with four LVDTs (Linear Variable posite connection at the local and global level.
Differential Transducers) to measure the longitudinal slip between the The model developed also able to predict the failure pattern in the
precast concrete slab and steel beam. Longitudinal slip at the interface precast concrete slab close to the hole (Fig. 17 (b)). However, for some of

6
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 9. Details of push-out test specimens (unit: mm).

Table 2
The details of push-out test specimens.
Specimen Bolt type Bolt steel Grade ds dc Concrete strength

mm mm MPa

PT1 M12 8.8 14 16 25

PT2 M12 8.8 14 20 25
PT3 M12 12.9 14 16 25
PT4 M12 12.9 14 20 25
PT5 M16 8.8 18 20 25
PT6 M16 8.8 18 25 25
PT7 M16 12.9 18 20 25
PT8 M16 12.9 18 25 25
PT9 M20 8.8 22 25 25
PT10 M16 12.9 18 20 40
PT11 M16 12.9 18 25 40
PT12 M20 8.8 22 25 40

Notes: dc = hole diameter in concrete slab, ds = hole diameter in steel beam. Fig. 11. Loading protocol used in the test.

Fig. 10. Test set-up and arrangement of instrumentations for push-out test.

7
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 12. Crack propagation of specimens.

5. Parametric study

5.1. General

The proposed finite element model has been verified with test results
described in the previous sections. Accordingly, a systematic parametric
study was carried out by means of the model adequately validated to
further investigate the behaviour of demountable bolted shear connec­
tors having prefabricated concrete slabs and to achieve data for practical
design rules development. The parametric study concerning of: bolted
shear connector size, bolt grade, hole size in precast concrete slab and
steel beam top flange, concrete strength, and bolt preloading force is
made to assess their effect on ductility and shear resistance of
demountable bolted shear connectors. Table 4 summaries the variable
considered in this study and the full details of the parametric values. The
ranges of parameters considered in the parametric study are deemed to
Fig. 13. Failure modes of the specimen PT7 with Grade 12.9 M16 bolts. be appropriate for real composite connections.

5.2. Effect of bolt size

The effect of the bolt size on the behaviour of the bolted shear con­
nectors was studied by changing the bolt diameter in this section. Six
different types of bolts, including M8, M10, M12, M16, M20, and M24
were considered in the parametric study. All bolts were used in this
section had the same tensile strength (Grade 8.8). The influence of
varying the bolt size on the behaviour of the composite connection is
shown in Fig. 18 and the variation of the ultimate shear resistance and
the slip capacity of the composite connection with respect to the bolt size
is shown in Fig. 19. It can be seen that the trend of the load-slip curves is
evident that as the size of the bolt increases, almost all main parameters
(i.e. ultimate shear resistance and slip capacity) of the composite
connection are changed as a result. An increase in bolt diameter was
demonstrated to increase the shear resistance and its corresponding slip.
For example, as shown in Figs. 18 and 19, using M24 bolts as the shear
Fig. 14. Shear failure mode - Specimen PT5 with Grade 8.8 M16 bolts. connector in composite connection increases the shear load capacity and
its corresponding slip by about 560% and 72%, respectively when
the specimen (specimen PT1for example), the prediction of the model in compared with the composite connection with M8 bolts. The bolt sizes
terms of the maximum value of slip in slipping stage and the shear load chosen as M8 to M24 in this parametric study are commonly used in real
for the first bearing are poor when compared to the test results. The structures.
reason for the difference in the value of the maximum value of slip in
slipping stage may be because the bolts were not positioned at the center 5.3. Effect of bolt grade
of the holes in the top flange of the steel beam and in the precast con­
crete slab in the experimental study. Furthermore, a few discrepancies In the analytical study, a parametric study of bolt tensile strength
between the test and model results in term of the shear load for the first was carried out to investigate its effect on load-slip behaviour of com­
bearing (specimen PT1) may be because of the higher preloading force posite connection. Three different bolt Grade including Grades 8.8, 10.9,
exerted to the bolts in the test. and 12.9 were considered in this study. FE model results are illustrated

8
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 15. Comparison of the FE predictions with the experimental results.

increased with the increment of bolt tensile strength (Fig. 21). For
Table 3 example, the ultimate shear load capacity and its corresponding slip of
Comparisons of the FEM predictions with the experimental results.
the model with M16 Garde 12.9 bolts is about 40% and 20% higher than
Test Pf (kN) Sa (mm) Pm (kN) Sam (mm) Su (mm) those of the model with M16 Grade 8.8 bolts, respectively as can be seen
Test/FE Test/FE Test/FE Test/FE Test/FE in Fig. 20. As shown in Fig. 20, the initial stiffness and the maximum
Mean 1.01 0.93 1.06 0.83 0.82
value of slip in slipping stage of the composite specimens with different
S.D. 0.22 0.20 0.10 0.15 0.13 bolt tensile strength was not affected by enhancing the bolt size as
expected.

in Fig. 20 as load-slip relationship of bolted shear connectors with 5.4. Bolt preloading force
different bolt Grades. In addition, the variation of the ultimate shear
resistance and slip capacity with respect to the bolt grade are shown in Results obtained for the experimental study showed that bolt pre­
Fig. 21. It is shown that the load-slip relationship is highly dependent on loading force has a significant effect on the load-slip relationship of the
the bolt tensile strength. Shear resistance and maximum slip are shear connectors. Therefore, to investigate the effect of bolt preloading

9
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 16. Comparisons of the FEM predictions with the experimental results: (a) slip at the first bearing, (b) load for the first bearing, (c) ultimate load capacity and,
(d) slip corresponding to the maximum shear load.

Fig. 17. Failure patterns: (a) bolts, (b) precast concrete slab for PT1 (unit: MPa).

force on the behaviour of bolt shear connectors, a parametric study of influence of varying the bolt preloading force on the behaviour of the
bolt preloading force was carried out in this section. Six different bolt composite connection is shown in Fig. 22 and the variation of the ulti­
preloading forces, including 10%, 20%, 30%, 40%, 50%, and 60% of the mate shear resistance and the slip capacity of the composite connection
ultimate tensile strength of bolt were considered in this study. The with respect to the bolt preloading force is shown in Fig. 23. It can be

10
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Table 4 and the steel beam on the behaviour of the composite connection is
Parameters considered for parametric studies. shown in Fig. 24. The variation of the ultimate shear resistance and the
Variable Range of variable selected slip capacity of the composite connection with respect to the clearances
in the concrete slab, and the steel beam are shown in Figs. 25 and 26,
Bolt size (mm) M8 M12 M16 M20 M24
respectively. It can be seen that by increasing the bolt clearance in the
Bolt Grade 8.8 10.9 12.9 prefabricated concrete slab and the steel beam, the maximum slip ca­
Pretension load (%) 10 20 30 40 50 60
Clearance in steel beam (mm) 0.5 1.0 1.5 2.0 2.5
pacity of shear connectors increases, as observed in experimental study
Clearance in concrete slab (mm) 1 2 3 4 5 [25]. However, the maximum shear load capacity of the connector is not
Concrete strength (MPa) 20 30 40 50 60 70 influenced significantly by changing the hole clearance in the concrete
Notes: Values in bold adopted for standard joint. slab and steel beam as expected. For example, the ultimate slip (i.e. Su)
corresponding to the maximum shear load for the model with 10 mm
clearance in the prefabricated concrete slab is about 41% bigger than
that of the model with 2 mm clearance. For these two models, a 2 mm
clearance was considered in the top flange of the steel beam. The
maximum shear load for these two models is 102.2 KN and 100.6 KN
respectively, confirming that the value of the clearance does not have
significant influence on the maximum shear capacity of the connection.
In addition, Fig. 24 shows that the maximum value of slip in slipping
stage is significantly depended on the size of the oversized holes in the
precast concrete slab and steel beam, the clearance between the bolted
shear connector and concrete slab and the clearance between the bolted
shear connectors and the hole wall in the top flange of the steel beam. A
schematic outline of the preload bolted shear connector and its behav­
iour before and after the slip in slipping stage is shown in Fig. 27. It can
be observed that the value of the slip in slipping stage is approximately
equal to the sum of the bolt clearance in the concrete slab and the top
Fig. 18. Influence of different bolt size on load-slip response for compos­
ite connection.
flange of steel beam (see Figs. 24 and 27). The results of the parametric
study of clearance in the concrete slab and steel beam show that the
ductility of the composite connection can be significantly improved by
seen that as the bolt pretension load increase, the maximum value of slip
increasing the clearance in the connection between the components.
in slipping stage occurred at a higher load. In other words, the friction
Fig. 24 also shows that for all the specimens with different clearances,
between the precast concrete slab and steel beam was increased by
the slip in slipping stage was occurred at the same level of load (at
increasing the bolt preloading force, that in turn leads to an increase in
around 20 KN for the models considered in this parametric study). This
the force needed to overcome the friction. For example, the shear load
is because the pre-tensioning load applied to the bolts for all the models
required to overcome the friction for specimen with a pretension load of
60% of the bolt ultimate tensile load is about 395% higher than that of
the model with a pretension load of 10% the bolt ultimate tensile load.
However, the bolt pretension load does not have a significant influence
on the maximum shear load capacity and the corresponding slip, as can
be seen in Figs. 22 and 23.

5.5. Effect of clearance

The effect of the clearance between the bolt shank and the hole wall
in the steel beam/concrete slab on the behaviour of the bolted shear
connectors was studied by changing the hole size in the top flange of the
steel beam and the precast concrete slab. Five different clearances in the
prefabricated concrete slab (i.e. 2, 4, 6, 8, and 10 mm) and five different
clearances in top flange of the steel beam (i.e. 1, 2, 3, 4, and 5 mm) were
Fig. 20. Influence of different bolt grade on load-slip performance of the
considered in the modelling of the composite connection. The influence
composite connection.
of varying the value of the clearance in the prefabricated concrete slab

Fig. 19. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to bolt diameter.

11
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Fig. 21. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to bolt tensile strength.

ultimate slip of model with higher concrete strength decreases about

40% compared with model with 20 MPa concrete strength. Based on the
results, it is evident that composite connection with higher concrete
compressive strength is able to achieve slightly higher shear load ca­
pacity but at the expense of a reduced ductility and maximum slip
capacity.

6. Design models and recommendations

The validated model proposed in this study is able to simulate the

performance of the composite connection using concrete slabs and
bolted shear connectors under monotonic loading. However, the
developed model is not applicable in a design perspective and therefore
a simple model for practising engineers is required. In addition, there is
Fig. 22. Influence of different bolt pretension on load-slip performance of the no design equations of the behaviour of bolted shear connectors avail­
composite connection. able in EC 4 [23] and AISC [24]. Therefore, existing simple design
models to predict the shear load capacity of preload bolted shear con­
was the same. nectors are reviewed herein and their applicability and suitability to be
incorporated in design practice are discussed. In addition, based on the
5.6. Effect of concrete compressive strength results, a new design equation to determine the ultimate strengths is
proposed.
Different concrete compressive strength (i.e. 20, 30, 40, 50, 60, 70
MPa) was considered in order to study the effect of this parameter on the
behaviour of composite connection. The influence of varying the value 6.1. Available design equations
of the compressive strength of concrete on the behaviour of the com­
posite connection is shown in Fig. 28 and the variation of the ultimate Rare design models to predict the performance of bolted shear con­
shear resistance and the slip capacity of the composite connection with nectors in term of the shear load capacity are available in the literature.
respect to the clearances in the concrete slab and steel beam are shown The proposed equations available in the literature are presented and
in Fig. 29. It can be seen that as the compressive strength of concrete summarised herein. The shear load capacities are calculated using these
increase, the maximum shear resistance of the composite connection equations and then compared to the shear load capacities achieved from
slightly increases but with a much-reduced maximum slip and ductility. the tests and the FE models. As mentioned before, design equations of
This may be because the deformation of the bolt and damage of the the bolted shear connectors are not provided by the two most popular
concrete slab around the bolt hole decreases due to using higher con­ provisions (i.e. AISC [24] and EC4 [23]). Both provisions proposed the
crete strength. For example, the maximum shear load capacity of design methods only for the headed stud shear connectors. AISC [24]
connection with a concrete compressive strength of 70 MPa is about 6% suggested the following equation to estimate the shear load capacity of
higher than that of the model with 20 MPa compressive strength, but the headed stud shear connectors:

Fig. 23. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to bolt pretension.

12
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

√̅̅̅̅̅̅̅̅̅
(1)
′
Qu = 0.7Asc f c Ec ≤ Asc fu

where fc′ = cylinder compressive strength of concrete (in MPa), Ec =

modulus elasticity of concrete (in MPa), Asc = cross-sectional area of
shear connector (in mm2), fu = tensile strength of shear connector (in
MPa). Based on the EC4 [23] definition, the shear load capacity of
headed stud shear connectors, Qu, can be determined as,
/
Qu,c = 0.29α d2 (Ec fck )0.5 γ v (2)

/
Qu,s = 0.8Asc fu γv (3)

in which, α = 0.2(hsc/d + 1) for 3≤hsc/d ≤ 4 and α = 1 for hsc/d > 4, γ v=

Fig. 24. Influence of different clearance in (a) concrete slab and (b) steel beam
on load-slip performance of the composite connection.

Fig. 27. Behaviour of preload bolted shear connector before and after slip in
slipping stage [34].

Fig. 25. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to clearance in steel beam.

Fig. 26. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to clearance in steel beam.

13
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

recommended by AISC [24] and EC4 [23] for predicting the shear ca­
pacity of the demountable shear connectors. It was concluded that the
design method provided by EC4 [23] is more accurate than the methods
recommended by AISC [24].

6.2. Design recommendation

A regression analysis of the data produced by the FE analysis has

been employed to propose a new design equation to determine the ul­
timate strengths (Qu,RD) for the post-installed preload bolt shear con­
nectors. The design equation can be written as:
Qu,RD = 0.6345 Asc fu (9)

Fig. 28. Influence of different concrete strength on load-slip performance for The ultimate shear load capacity, Qu attained from the proposed
composite connection. design equations and the design formulas of the welded headed shear
connector recommended by EC4 [23] and AISC [24] and the Eq. (9)
1.25. The minimum value achieved from the Eqs. (7) and (8) was proposed in this study for the post-installed preload bolt shear connec­
considered as the ultimate shear load capacity (Qu) of the shear tors, have been compared with the results obtained from the experi­
connectors. mental study in Table 5. In addition, the ultimate shear load capacities,
The calculation equation of the ultimate shear load capacity of a Qu (in N) attained from all the design equations have been compared
bolted shear connector, Qu (in N) is expressed by Kwon et al. [1] as, with those achieved from the FE models in Table 6.
It can be seen from Table 5 and Table 6 that the prediction of the
Qu = 0.5Asc Fu (4) ultimate shear load capacity values by EC4 and AISC provisions are
found to be relatively less accurate but conservative. It can be also
where Fu = tensile strength of the bolt (in MPa) and Asc = cross-sectional illustrated that the equations proposed by Zhang et al. [16] is found to
area of the bolted shear connector (in mm2). Liu et al. [27] modified the predict the ultimate shear resistance of composite connections more
above equation based on the test results and proposed the following accurately than the equations proposed by Kwon [1] and Liu [27]. Ta­
formula to represent the ultimate shear load capacity of composite bles 4 and 5 also show that the design equation proposed in this study is
connections using bolted shear connectors: found to be relatively more accurate.
Qu = 0.66Asc Fu (5)
7. Conclusion
Based on an experimental study conducted by Zhang et al. [16], the
following equations were proposed to calculate the ultimate shear load
The mechanical and structural behaviour of deconstructable bolted
capacity of the bolted shear connector in composite connections.
shear connectors subjected to a monotonic loading in composite beams
√̅̅̅̅̅̅̅̅̅̅
Qu,c = 0.7Asc fck Ec (6) are investigated through a three-dimensional finite element model. The
effect of different parameters, including bolted shear connector size, bolt
Qu,s = 0.62Asc fu (7) grade, hole size in precast concrete slab and steel beam top flange,
concrete strength, and bolt preloading force is studied by means of
in which, fck = cylinder compressive strength of concrete. The minimum ABAQUS software. A finite element model of the composite connection
value achieved from these equations was considered as the shear load considering the non-linearities of the geometry, interfaces between the
capacity of the bolt shear connector. components and materials is developed and verified against the results
The calculation equation of the ultimate shear load capacity of a of the experimental study. An extensive parametric study is conducted
bolted shear connector, Qu (in N) was proposed based on both experi­ and the influences of different parameters on the structural behaviour of
mental and numerical results by Chen et al. [19] as, demountable steel-concrete composite shear connection is discussed in
detail. Shear load capacity and slip corresponding to the maximum load
Qu = 0.23d1.78 fcu 0.29 (0.0007fs + 0.53), (8) are also evaluated. Furthermore, a comparison of the estimated values
obtained from design equations proposed by current standards including
where fcu= cubic compressive strength of concrete (in MPa), fs= yield
EC4 [23] and AISC [24] for headed stud shear connectors and the design
strength of shear connector (in MPa) and d = diameter of bolt shank (in
recommendations provided by other studies for bolted shear connectors
mm).
are conducted. Finally, a regression analysis of the data produced by the
Dai et al. [6] assessed the feasibility of design calculation methods
FE analysis has been employed to propose a new design equation to

Fig. 29. Variation of (a) shear load capacity and (b) slip corresponding to maximum shear load with respect to concrete compressive strength.

14
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

Table 5
Comparisons of the experimental results with the calculated ultimate shear capacity.
Specimen Qu,test (kN) Qu,test Qu,test Qu,test Qu,test Qu,test Qu,test Qu,test/Pu,RD
Qu,AISC Qu,EC4 Qu,Kwon Qu,Liu Qu,Zhang Qu,Chen

PT1 82.0 1.544 2.612 1.612 1.221 0.769 1.47 1.140

PT2 79.8 1.502 2.542 1.569 1.188 0.790 1.43 1.150
PT3 83.0 1.563 2.644 1.224 0.927 0.896 1.19 0.964
PT4 83.5 1.572 2.660 1.231 0.933 0.891 1.20 0.970
PT5 118.0 1.250 2.114 1.305 0.989 0.950 1.26 1.028
PT6 130.1 1.378 2.331 1.439 1.090 0.862 1.39 1.134
PT7 161.5 1.710 2.893 1.339 1.015 0.819 1.39 1.055
PT8 183.7 1.945 3.291 1.524 1.154 0.720 1.58 1.150
PT9 180.0 1.220 2.064 1.274 0.965 0.973 1.30 1.004
PT10 165.2 1.335 2.259 1.370 1.038 0.905 1.28 1.080
PT11 189.2 1.530 2.588 1.569 1.189 0.790 1.47 1.050
PT12 196.2 1.015 1.717 1.389 1.052 0.893 1.27 1.094
Mean 1.464 2.476 1.404 1.063 0.855 1.35 1.068
Standard deviation 0.234 0.396 0.132 0.100 0.074 0.12 0.065

Table 6
Comparisons of ultimate shear load capacity from FE model and recommended design equations.
Variable Specimen Qu,FEM (kN) Qu,FEM Qu,FEM Qu,FEM Qu,FEM Qu,FEM Qu,FEM Qu,FEM
Qu,AISC Qu,EQ4 Qu,Kown Qu,Liu Qu,Zhang Qu,Chen Qu,RD
Pretension load
M16–10% 101.1 1.070 1.811 1.257 0.952 0.986 1.145 0.99
M16–20% 101.4 1.074 1.817 1.261 0.956 0.983 1.149 0.99
M16–30% 101.5 1.075 1.819 1.263 0.957 0.982 1.150 1.00
M16–40% 101.4 1.074 1.817 1.262 0.956 0.983 1.149 0.99
M16–50% 101.9 1.079 1.826 1.268 0.961 0.978 1.155 1.00
M16–60% 101.5 1.075 1.819 1.263 0.957 0.982 1.150 1.00
Bolt size M8–8.8 37.7 1.597 2.702 1.876 1.421 0.661 1.467 0.94
M10–8.8 55.5 1.506 2.547 1.769 1.340 0.701 1.453 1.18
M12–8.8 75.9 1.429 2.417 1.679 1.272 0.739 1.435 1.32
M16–8.8 126.5 1.340 2.267 1.574 1.192 0.788 1.434 1.15
M20–8.8 180.5 1.223 2.070 1.437 1.089 0.863 1.375 1.13
M24–8.8 248.7 1.171 1.980 1.375 1.042 0.902 1.369 1.08
Bolt grade M16–8.8 101.8 1.078 1.824 1.267 0.960 0.979 1.154 1.00
M16–10.9 126.5 1.340 2.267 1.259 0.954 0.985 1.209 0.99
M16–12.9 149.9 1.588 2.686 1.243 0.942 0.882 1.292 0.98
Clearance in concrete 1 mm 101.7 1.077 1.821 1.265 0.958 0.980 1.152 1.00
2 mm 102.2 1.082 1.830 1.271 0.963 0.976 1.158 1.00
3 mm 102.3 1.084 1.833 1.273 0.964 0.974 1.159 1.00
4 mm 102.1 1.081 1.829 1.270 0.962 0.976 1.157 1.00
5 mm 102.2 1.083 1.832 1.272 0.964 0.975 1.159 1.00
Clearance in steel 0.5 mm 99.7 1.056 1.787 1.241 0.940 0.999 1.130 0.98
1.0 mm 100.4 1.063 1.799 1.249 0.946 0.993 1.138 0.98
1.5 mm 101.5 1.075 1.818 1.262 0.956 0.982 1.150 0.99
2.0 mm 102.2 1.082 1.830 1.271 0.963 0.976 1.158 1.00
2.5 mm 103.2 1.092 1.848 1.283 0.972 0.966 1.169 1.01
Concrete strength C20 95.6 1.373 2.322 1.190 0.901 1.020 1.219 0.94
C30 97.5 1.032 1.746 1.212 0.918 1.023 1.104 0.96
C40 98.7 0.843 1.426 1.228 0.931 1.010 1.029 0.97
C50 100.4 0.725 1.226 1.249 0.946 0.993 0.981 0.98
C60 101.6 0.639 1.082 1.263 0.957 0.981 0.941 1.00
C70 102.0 0.635 0.991 1.269 0.961 0.977 0.904 1.00
Mean 1.121 1.893 1.327 1.005 0.942 1.187 1.02
Standard deviation 0.228 0.391 0.164 0.124 0.092 0.138 0.08

determine the ultimate strengths for the post-installed preload bolt shear 5) As the bolt pretension load increase, the slip in slipping stage
connectors. Based on the FE analysis and parametric study, the following occurred at a higher load. In other words, the friction between the
conclusions can be made. precast concrete slab and steel beam was increased by increasing
the bolt preloading force, that in turn leads to an increase in the
1) Bolt size, bolt grade and concrete compressive strength have force required to overcome the friction. However, the bolt pre­
significant effect on the ultimate shear load capacity of the bolt tension load does not have significant influence on the maximum
shear connection. shear load capacity and the corresponding slip.
2) Clearance between the bolt shank and the hole wall in the steel 6) By increasing the bolt clearance in the prefabricated concrete slab
beam/concrete slab has significant effect on the ductility of the and the steel beam, the maximum slip capacity of shear connec­
bolt shear connection. tors increases. However, the maximum shear load capacity of the
3) An increase in bolt diameter was demonstrated to increase the connector is not influenced significantly by changing the hole
shear resistance and its corresponding slip. clearance in the concrete slab and steel beam as expected.
4) Shear resistance and maximum slip are increased with increment
of bolt tensile strength.

15
A. Ataei Journal of Constructional Steel Research 198 (2022) 107585

7) The value of the slip in slipping stage is approximately equal to [8] M. Pavlović, M. Spremić, Z. Marković, M. Veljković, Headed shear studs versus
high-strength bolts in prefabricated composite decks, in: Composite Construction in
the sum of the bolt clearance in the concrete slab and the top
Steel and Concrete VII, 2016, pp. 687–702.
flange of steel beam. [9] A. Ataei, PhD thesis SoCaEE. A Low-carbon Deconstructable Steel–concrete
8) The ductility of the composite connection can be significantly Composite Framed System with Recyclable Beam and Slab Components, 2016.
improved by increasing the clearance in the connection between [10] A. Ataei, M.A. Bradford, Numerical study of deconstructable flush end plate
composite joints to concrete-filled steel tubular columns, Structures. 8 (2016)
the components. 130–143.
9) It can be seen that as the compressive strength of concrete in­ [11] A. Ataei, M.A. Bradford, X. Liu, Experimental study of composite beams having a
crease, the maximum shear resistance of the composite connec­ precast geopolymer concrete slab and deconstructable bolted shear connectors,
Eng. Struct. 114 (2016) 1–13.
tion slightly increases but with a much-reduced maximum slip [12] A. Ataei, M.A. Bradford, X. Liu, Experimental study of flush end plate beam-to-
and ductility. column composite joints with precast slabs and deconstructable bolted shear
10) The ultimate shear load capacity values predicted by EC4 and connectors, Structures. 7 (2016) 43–58.
[13] A. Ataei, M.A. Bradford, H.R. Valipour, Experimental study of flush end plate
AISC provisions are found to be relatively less accurate but beam-to-CFST column composite joints with deconstructable bolted shear
conservative. connectors, Eng. Struct. 99 (2015) 616–630.
11) The design methods proposed in this study is found to predict the [14] A. Ataei, M.A. Bradford, H.R. Valipour, Finite element analysis of HSS semi-rigid
composite joints with precast concrete slabs and demountable bolted shear
ultimate strength of the post-installed preload bolt shear con­ connectors, Finite Elem. Anal. Des. 122 (2016) 16–38.
nectors more accurately [15] A. Ataei, M.A. Bradford, H.R. Valipour, X. Liu, Experimental study of sustainable
high strength steel flush end plate beam-to-column composite joints with
deconstructable bolted shear connectors, Eng. Struct. 123 (2016) 124–140.
Authorship statement [16] Y. Zhang, B. Chen, A. Liu, Y.-l. Pi, J. Zhang, Y. Wang, et al., Experimental study on
shear behavior of high strength bolt connection in prefabricated steel-concrete
Abdolreza (Reza) Ataei: Everything: Conceptualization, Methodol­ composite beam, Compos. Part B 159 (2019) 481–489.
[17] A.S.H. Suwaed, T.L. Karavasilis, Removable shear connector for steel-concrete
ogy, Software, Data curation, Writing- Original draft preparation and so composite bridges, Steel Compos. Struct. 29 (2018) 107–123.
on. [18] A.S.H. Suwaed, T.L. Karavasilis, Demountable steel-concrete composite beam with
full-interaction and low degree of shear connection, J. Constr. Steel Res. 171
(2020), 106152.
Declaration of Competing Interest [19] J. Chen, W. Wang, F.X. Ding, P. Xiang, Y.J. Yu, X.M. Liu, et al., Behavior of an
advanced bolted shear connector in prefabricated steel-concrete composite beams,
Materials (Basel). 12 (2019).
The authors declare that they have no known competing financial [20] L. Wang, M.D. Webster, J.F. Hajjar, Pushout tests on deconstructable steel-concrete
interests or personal relationships that could have appeared to influence shear connections in sustainable composite beams, J. Constr. Steel Res. 153 (2019)
618–637.
the work reported in this paper.
[21] J. He, A.S.H. Suwaed, G. Vasdravellis, S. Wang, Behaviour and design of the
The authors declare the following financial interests/personal re­ ‘lockbolt’ demountable shear connector for sustainable steel-concrete structures,
lationships which may be considered as potential competing interests: Structures. 44 (2022) 988–1010.
Dr. Abdolreza Ataei. Centre for Infrastructure Engineering and [22] J. He, A.S.H. Suwaed, G. Vasdravellis, Horizontal pushout tests and parametric
analyses of a locking-bolt demountable shear connector, Structures. 35 (2022)
Safety, School of Civil and Environmental, Engineering, UNSW Sydney, 667–683.
NSW 2052, Australia. [23] British Standards Institution, Eurocode 4: Design of Composite Steel and Concrete
Structures: Part 1. 1 General Rules and Rules for Buildings, BS EN 1994-1-1, BSI,
London, 2006.
Data availability [24] AISC, Specification for Structural Steel Buildings, American Institute of Steel
Construction, Chicago, Illinois, 2016.
No data was used for the research described in the article. [25] A. Ataei, M. Zeynalian, A study on structural performance of deconstructable
bolted shear connectors in composite beams, Structures. 29 (2021) 519–533.
[26] D.J. Carreira, CKS-srfpcicASJ 82 (6) (2022) 797–804.
References [27] X. Liu, M.A. Bradford, M.S.S. Lee, Behavior of high-strength friction-grip bolted
shear connectors in sustainable composite beams, J. Struct. Eng. 141 (2015)
[1] G. Kwon, M.D. Engelhardt, R.E. Klingner, Behavior of post-installed shear 04014149.
connectors under static and fatigue loading, J. Constr. Steel Res. 66 (2010) [28] S.S.M. Lee, M.A. Bradford, Sustainable composite beams with deconstructable
532–541. shear connectors, in: 5th International Conference on Structural Engineering,
[2] G. Kwon, M.D. Engelhardt, R.E. Klingner, Experimental behavior of bridge beams Mechanics and Computation, Cape Town, South Africa, 2013.
retrofitted with Postinstalled shear connectors, J. Bridg. Eng. 16 (2011) 536–545. [29] X. Liu, M.A. Bradford, A. Ataei, Flexural performance of innovative sustainable
[3] O. Mirza, X.Q. Zhu, B. Uy, Retrofitting/strengthening strategy for existing bridges composite steel-concrete beams, Eng. Struct. 130 (2017) 282–296.
using blind bolts as shear connectors, in: 4th Australian Small Bridges Conference [30] ABAQUS, 6.141 [Computer software]. Providence R, Dassault Systemès Simulia,
201, 10-11th May, 2011. 2022.
[4] S.W. Pathirana, B. Uy, O. Mirza, X. Zhu, Strengthening of existing composite steel- [31] D.J. Oehlers, Methods of estimating the fatigue endurances of stud shear
concrete beams utilising bolted shear connectors and welded studs, J. Constr. Steel connectors, in: IABSE Periodica 3, IABSE Proceedings P-145/90, 1990, pp. 65–84.
Res. 114 (2015) 417–430. [32] D.J. Oehlers, M.A. Bradford, Composite Steel and Concrete Structural Members:
[5] D. Lam, X. Dai, E. Saveri, Behaviour of demountable shear connectors in Fundamental Behaviour, Pergamon, Oxford, 1995.
steelconcrete composite beams, in: Proceedings of the 2013 Composite [33] H.Y. Loh, B. Uy, M.A. Bradford, The effects of partial shear connection in composite
Construction in Steel and Concrete VII. Queensland, Australia, 2013. flush end plate joints Part I — experimental study, J. Constr. Steel Res. 62 (2006)
[6] X.H. Dai, D. Lam, E. Saveri, Effect of concrete strength and stud collar size to shear 378–390.
capacity of demountable shear connectors, J. Struct. Eng. 141 (2015) 04015025. [34] Ahmed Suwaed, Development of Novel Demountable Shear Connectors for Precast
[7] M. Pavlović, Z. Marković, M. Veljković, D. Buđevac, Bolted shear connectors vs. Steel- Concrete Composite Bridges, School of Engineering, University of Warwick,
headed studs behaviour in push-out tests, J. Constr. Steel Res. 88 (2013) 134–149. 2017. PhD thesis.

16