Construction and Building Materials 120 (2016) 382–392

Contents lists available at ScienceDirect

Construction and Building Materials

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

Comparative performance of channel and angle shear connectors in high

strength concrete composites: An experimental study
M. Shariati a, N.H. Ramli Sulong a,⇑, A. Shariati a, A.B.H. Kueh b
a
Department of Civil Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Construction Research Centre, Institute for Smart Infrastructure and Innovative Construction, Faculty of Civil Engineering, Universiti Teknologi Malaysia, 81310 Johor Bahru, Malaysia

h i g h l i g h t s

Channel and angle connectors in high strength concrete composites are studied.

Connectors shear behaviors are tested under static (S) and cyclic (C) loads.

Channel connectors are more ductile and more affected by crack than those of angle.

6.8–30% and 0–18.5% less strength are seen in angles under S and C loads respectively.

Greater load is taken by higher connector; more concrete crack is seen in that longer.

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

Article history: Structural performance of steel-concrete composite structures relies deeply on the characteristics of the
Received 3 November 2015 shear connectors. Recently, great attention has been focused on the implementation of C-shaped connec-
Received in revised form 7 May 2016 tors due to numerous advantages when used in composite beams. However, very little information is
Accepted 12 May 2016
available in regard to the response of such connectors when embedded in the high strength concrete
(HSC). To address this research gap, sixteen experiments on push-out specimens were conducted to com-
pare the performance of channel and angle shear connectors embedded in HSC. The shear resistance and
Keywords:
ductility of the connectors were primarily investigated by applying static and cyclic loadings. Results
Channel shear connector
Angle shear connector
were also compared with the cases when using normal reinforced concrete. Furthermore, the evaluation
Push-out test of the available equations suggested by the American and Canadian codes for estimating these connec-
Monotonic loading tors’ capacities when using HSC was carried out. In general, channel connectors exhibited 6.8–30.1% more
Cyclic loading shear strength than those of angle under monotonic loading, and up to 18.5% more when subjected to
High strength concrete cyclic loading. Angle connectors were also less ductile than channel connectors and did not satisfy the
ductility criteria specified in the codes’ requirements. Connector fracture mode of failure was recorded
for both connector types.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Many types of shear connectors are available commercially [1].

Composite structures are well-recognized worldwide because
of their high strength and reliable structural behavior. The benefits
and versatility of composite beams have attracted and encouraged
designers to use these structures in numerous skyscrapers in
modern-day high-rise constructions. In detail, the strength and
ductility of shear connectors play a vital role in the composite
beam design. The successful design of shear connectors requires
a careful estimation of the relationship between the transfer of
shear force and slope provided at the steel and concrete boundary.
⇑ Corresponding author.
E-mail address: hafizah_ramli@um.edu.my (N.H. Ramli Sulong).
Headed studs [2,3], Perfobond [4–8], and C-shaped sections [9–13]
are the most commonly used. Headed studs provide relatively high
automation in the workplace. Popular use of headed shear connec-
tors comes from their performance and ease of installation by
means of welding gun. Even though headed shear studs are popu-
lar, their application and installation come with reliability prob-
lems. Unless great care is given to the installation process,
welding strength and performance can be affected substantially
by the weather, coating material of the steel and surface condi-
tions. Because of their small load capacity, headed shear connec-
tors are commonly installed in a large amount. This creates a
crowded and unsafe working environment. The disadvantage of
using headed studs continues with their comparatively poor

http://dx.doi.org/10.1016/j.conbuildmat.2016.05.102
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392
performance under fatigue loading. Moreover, these connectors
demand high energy consumption and a particular welding
arrangement during installation [14]. This specific weld type can
easily initiate crack development under fatigue loading [15], which
leads to a girder design with partial interaction [16]. In addition,
when reinforcement bars have to pass the connector openings,
providing a good arrangement of the bottom reinforcement in
the slab may infer further difficulty [17]. Due to these drawbacks,
researchers and construction companies are continuously in pur-
suit of other better performing shear connectors.
In comparison, C-shaped or channel shear connectors are in
many aspects exceeding the headed studs’ performance. While
studs need special equipment like welding guns, conventional
welding equipment is enough for channel shear connectors. Con-
ventional welding has a proven performance and is more reliable
than the welding gun used to install the headed studs. Channel
shear connector’s load carrying capacity is at least twice that of
the headed shear studs. This results in the use of fewer shear con-
nectors. Fewer shear connectors means a substantial decrease in
the labor time, offering a safer work environment as well. Also,
conventional welding and channel connectors are more robust so
that rough handling can be better tolerated.
The principal disadvantage of Perfobond shear connector lies in
the placement of the slab transversal bottom reinforcement, which
is often laborious and difficult. This drawback is efficiently over-
come by C-shaped connectors, attributed to their higher con-
structability advantages. The bottom reinforcement is also
relatively easy to adjust in the slab. Moreover, these shear connec-
tors are commercially available in different sizes and necessitate
only little cutting of steel sections into C-shaped profiles to make
them ready for use. This relatively reduces expenditures and the
manufacturing time. Given these advantages, C-shaped connectors
are popularly utilized in composite beams in developing countries.
This study aims to examine and compare the performance of
channel and angle shear connectors when embedded in high
strength concrete (HSC). The profiles of the angle and channel
shear connectors are shown in Fig. 1. The angle connectors differ
from those of channel in that they do not have a bottom flange,
such that they consume less steel material and require little weld-
ing process. Thus, these connectors provide a more cost-effective
solution for composite beam design.
In predicting the strength of shear connectors, Slutter et al. [2],
Pashan [18], and Viest et al. [19] performed different tests to esti-
mate the performance quality of C-shaped shear connectors. These
researchers proposed different equations to calculate the strength
of the channel shear connectors [20]. Maleki and Bagheri [21,22]
investigated the strength of the channel connectors used in
different concrete types under different load types. Maleki and
Mahoutian [23] tested the channel connectors implanted in
Fig. 1. Typical (a) angle and (b) channel shear connectors.
383
polypropylene concrete. Two different equations that determine
the strength of channel connectors in metal and solid deck slabs
were introduced by Pashan and Hosain [24]. Baran and Topkaya
[25] proposed a new equation that predicts the strength of C-
shaped connectors, which improves the previous equations pro-
vided by different standards.
The American Institute of Steel Construction (AISC) [20] sug-
gests that the nominal strength of a channel connector surrounded
in concrete slab can be determined by the following:
p 0
Q n ¼ 0:3ðt f þ 0:5t w ÞLc ðf c Ec Þ ð1Þ
where Qn is the nominal strength (N), tf and tw are the thicknesses of
the flange and web, respectively, Lc is the beam length, and Ec and f0 c
are the modulus of elasticity and compressive strength of concrete
(MPa), respectively. The National Building Code (NBC) [26] of
Canada suggests the following equation for the same purpose:
p 0
Q n ¼ 36:5ðtf þ 0:5t w ÞLc f c ð2Þ
To date, there exists a number of studies exploring the behavior of
angle shear connectors. Shariati et al. [27–29] determined different
design characteristics of angle connectors surrounded with normal
and high-strength reinforced concrete under different load condi-
tions. Kiyomiya et al. [30] calculated the strength and deformation
in different shear connector types through push-out tests. These
researchers declared that the failure of a connector significantly
depends on the geometry and location of the connectors and con-
crete strength. The fatigue strength of the weld between the bottom
plate in the composite slab and angle shear connectors was exam-
ined by Choi [31] through experimental and numerical investiga-
tions. Fukazawa et al. [32] further investigated the possibility of
using angle shear connectors under moving loads. The specimens
exhibited satisfactory stiffness and durability.
Saidi et al. [33] explored the use of T- and C-shaped angle con-
nectors in a sandwich beam in terms of the relationship between
the transmitted shear force and consequent deformation by means
of experiment and numerical approaches. They provided a formula
to predict the transferred shear force at a sudden decrease of the
equivalent stiffness of the shear connector. Ros and Shima [34]
observed that the capacity of the shear connector can be largely
affected by the line of action of the shear force applied to the
connector.
Kiyomiya and Yokota [35] and Yamada and Kiyomiya [36]
established the following formula to identify the strength of C-
shaped connectors:
p 0
P ¼ 65Lc ðf c tw Þ ð3Þ
where P is the load carrying capacity (kgf). A relationship was pro-
duced by Ros [37] to quantify the ultimate shear capacity of angle
shear connectors that depends on the failure type:
384 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392

Table 1
Specimens’ geometric properties (mm).

Property C10050 C7550 C10030 C7530 A10050 A7550 A10030 A7530

Height 100 75 100 75 100 75 100 75
Length 50 50 30 30 50 50 30 30
Web thickness 6 5 6 5 6 5 6 5
Flange thickness 8.5 7.5 8.5 7.5 8.5 7.5 8.5 7.5

p 0
V u ¼ kLc h f c ; k ¼ 63t w =h þ 1:60 ð4Þ scarcely available. So, this research is vital for promoting better
performance of mentioned shear connectors to be widely used in
where h is the height of the connector (mm). composite systems.
Shariati et al. [38] compared the performance of C-shaped chan-
nel and angle shear connectors embedded in normal strength con-
crete subjected to different load conditions. The requirement of
satisfactory strength and durability coupled with a cost-effective
product has resulted in an abrupt increase in the use of HSC ele- Table 2
ments in the construction industry [39]. HSC promotes the use of Particle size analysis of silica sand (SS) based on BS 822: Clause 11 [45].

slender structures in the composite construction by increasing Sieve size Sieve WSS + WS WS WSS Ret. Cum. Ret. Pass
their load carrying capacity, especially in structures under fatigue (lm) No. (g) (g) (g) (%) (%) (%)
[40]. It significantly varies the ratio between the deformation 4750 3/16 in 409.9 408.3 1.6 0.32 0.032 99.68
capacity of connection and maximum slip requirement [41]. Simi- 2360 No. 7 462.3 375.7 86.6 17.33 17.65 82.35
lar to other applications, HSC is also widely used to embed shear 1180 No. 14 437.2 343.0 94.2 18.85 36.5 63.50
600 No. 25 450.7 316.2 134.5 26.93 63.42 36.58
connectors in composite beams [41,42].
300 No. 52 379.1 288.7 90.4 18.09 81.51 18.49
This study explores the performance of various geometries of 150 No. 100 322.1 274.8 47.3 9.47 90.99 9.02
channel and angle shear connectors surrounded in HSC of different 75 No. 200 309.9 275.2 34.7 6.94 97.92 2.08
strength levels under static and fatigue loading in terms of their Pan – 250.8 240.4 10.4 2.08 – 0.00
Total 499.7 388.31
shear strength and ductility. It is noteworthy to realize that at pre-
sent there is no existing literature on the comparative study of Fineness modulus = 388.31/100 = 3.88; Water absorption for silica sand is 0.93%;
experimental and analytical works on both channel and angle WSS = Silica sand weight; WS = Sieve weight; Cum. Ret = Cumulative retained.
shear connectors especially when embedded in HSC as well as
under static and cyclic loading environment. Hence, their behav-
Table 3
iors when embedded in reinforced normal concrete [38] are used Grain size distribution for granite gravels based on BS 882: 1992.
as control for comparison. Since the loading environment plays a
significant role in influencing the performance of concrete struc- Sieve size Sieve size WG + WS WS (g) WG (g) Ret. Cum. Ret. Pass
(lm) (in) (g) (%) (%) (%)
tures [43,44], it is greatly relevant that additionally to the use of
high-strength concrete, the current study examines the structural 19 3/4 1626.6 1616.1 10.5 0.42 0.42 99.58
12.7 1/2 2181.6 1398.8 782.8 31.32 31.74 68.26
behavior not only in static load but also that of cyclic. The load
9.5 3/8 2271.3 1378.4 892.9 35.72 67.46 32.54
capacities of both C-shaped channel and angle connectors in HSC 4.75 3/16 2170.5 1397.4 773.1 30.93 98.39 1.61
are also compared with the current available equations to deter- Pan – 886.2 846.0 40.2 1.61 – 0.00
mine their reliability in predicting the respective shear capacities. Total FM = 798.01/100 = 7.98 2499.5 100 600
+ 198.01
The application of HSC in structures with composite systems is

Fig. 2. (a) Test specimen and (b) test configuration.

 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392 385

2. Push-out test program
2.1. Details of specimens and push-out set-up
Sixteen specimens categorized into two push-out test series
were tested under static and fully-reversed fatigue loading. One
test series was conducted for the channel specimens, and the other
series was for those of angle. Each series comprises further two dif-
ferent groups, in which each group has eight specimens. The first
group was assessed via a gradually increased monotonic loading,
whereas the second group was tested under fully-reversed cyclic
loading. Results achieved from both groups were compared. Chan-
nel specimens of four different dimensions were tested: 75 mm
and 100 mm heights as well as 30 mm and 50 mm lengths. To
ensure a matching comparison, legs of the angle specimens were
cut to achieve the length equivalent to the channel flange. The
angle specimen thicknesses were milled to be identical to those
of channel. The specimen details for both loading types are pre-
sented in Table 1. It is noted that the push-out test inflicts only a
pure shear loading on the specimens. For more realistic condition,
either simply-supported or continuous steel-concrete composite
beam floor system is recommended for its behavior examination
in a combined action of bending and shear. Since the current paper
emphasizes on the comparative performance of channel and angle
shear connections, only the elementary pure shear behavior was
explored.
For the overall structure, the flanges of the steel I-beam were
connected to two concrete slabs and welded to the connector as
shown in Fig. 2. The I-beam flange was welded to the angle speci-
men legs and to all sides of the channel bottom flange. Two layers
of 10 mm diameter steel
bars were arranged in two perpendicular directions for the
slabs. The test configuration was similar to that of Maleki et al.
[21,23] as shown in Fig. 2. High-strength reinforced concrete with
compression strengths of 63 and 82 MPa was used in this study.
The aggregates used in the concrete were air dried. Approximately
4.75 mm silica sand was used as fine aggregate, and 10 mm
crushed granite was used as coarse aggregate. The details of the
fine aggregate are provided in Table 2 [46,47]. Table 3 presents
the grain size distribution for the crushed granite. Ordinary Port-
land cement that corresponds to ASTM C150 type II [48] cement
was used in each mix. The cement chemical properties are shown
in Table 4 [46]. The workability of the mixes was enhanced by
using a Rheobuild 1100 superplasticizer with a pH ranges between
6.0 and 9.0 [49]. Its color was dark brown and its specific gravity
was 1.195. The concrete slabs were cast in a horizontal direction
following the similar casting of the composite beams in practice.
The mix properties of the high-strength concrete materials are pre-
sented in Table 5.
The compressive strength was assessed in the forms of 150 mm
diameter and 300 mm long cylinders as well as 100 mm cubes, cast
simultaneously with the push-out testing specimens. All prepared

Table 4
Composition of cementitious materials for OPC (% by mass).

P2O5 SiO2 Al2O3 MgO Fe2O3 CaO MnO K2O TiO2 SO3 CO2 LOI
0.068 18.47 4.27 2.08 2.064 64.09 0.045 0.281 0.103 4.25 4.20 1.53

Table 5
Material properties of constituents of high strength concrete.

Mix No Cement Coarse aggregate Fine aggregate Water Silica fume SP (%) W/C Modulus of Compressive
(kg/m3) (kg/m3) (kg/m3) (kg/m3) (kg/m3) elasticity (GPa) strength (MPa)
H1Series 460 910 825 168 40 1 0.37 39 80
H2Series 360 940 870 180 – 1 0.50 32 63

Fig. 3. Load-slip curves of the specimens under monotonic loading.

386 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392

objects were cured for 28 days. For consistency, all specimens,

cylinders, and cubes were tested simultaneously. The compressive

H1C10050- H1C10050- H1A10050- H1A10050- H1C7550- H1C7550- H1A7550- H1A7550- H2C10030- H2C10030- H2A10030- H2A10030- H2C7530- H2C7530- H2A7530- H2A7530-
strength testing was carried out in accordance with ASTM C39 [45].

97.6
2.0
A particular symbolization was allocated to each specimen for

C
identification. H1 and H2 represent the higher and lower strength
slabs, respectively. C and A represent the channel and angle con-
nector types, respectively. The following set of numbers is the

1.5
1.4
99.0
M height, whereas the latter set of numbers is the connector length
embedded in concrete. M and C represent respectively the mono-
tonic and cyclic loadings. For example, H1C10050 means a higher
74.6
6.5
strength slab with channel connectors that is 100 mm high and
C

50 mm long.
91.1

18.1
4.0

2.2. Testing procedure

M

Half of the specimens in each group were tested for monotonic

loading, whereas the other half were examined by applying fully-
1.5
105.1

reversed fatigue loading. A specific support was prepared to apply

C

the load through a universal testing machine (UTM) with 600 kN

load capacity. The fatigue loading condition was applied similar
1.5
0.1
105.2

to that of Civjan and Singh [50] and Maleki and Bagheri [21]. The
M

pseudo-dynamic loading involved three cycles (six half cycles) at

±1/3 M, ±2/3 M and ±M, where M is the static yield capacity of
the control specimen achieved from the load-slip curve of mono-
90.8
7.0

tonic loading (Fig. 3). The monotonic loading was gradually

C

increased up to the specimen failure. The loading rate used for

all specimens was 0.04 mm/s. To apply the reversed cyclic loading,
the specimens were turned and the loading was applied to the
109.6

17.1
6.5

upper concrete faces.

M

The UTM was used to record the load-slip relation at each time
step in each half cycle of the test during the loading of the speci-
2.5
150.8

mens. The recorded data was used to develop a hysteresis loop his-
C

tory for the cyclic fatigue test.

152.9

1.4
1.0

3. Results and discussion

M

3.1. Failure mode

165.8
7.0

The most familiar modes of push-out specimen failure are con-

C

nector fracture, crushing and splitting [21]. The failure modes of

both angle and channel connectors were almost similar. Both con-
186.3
6.5
11.0

nector types failed because of connector fracture. However, the

M

channel connectors’ failure was more ductile than the angles under
Failure load capacity, maximum slip, and strength degradation of specimens.

monotonic loading. Notably, the failure mode of the connectors

embedded in the HSC resembled that embedded in the normal
175.8
1.0

concrete, which was tested earlier [38]. The measured maximum

C

slip and shear load capacity of all specimens are summarized in

Table 6 based on Fig. 3, which illustrates the load-slip curves of
specimens under monotonic loading.
1.4
178.3
0.5

Further loading of specimens creates a longitudinal crack along

M

the slab that progressed and eventually opened due to the load
increment. Connectors that yield at higher loads were observed
7.5
174.3

in most of the experiments. Fig. 4 shows a typical failure mode

C

for both the channel and angle connectors.

3.2. Effect of connector geometries

191.3

8.9
7.0
M

Table 6 lists the failure load, maximum slip, and percentage of

Strength degradation
Maximum slip (mm)

strength degradation of the specimens. Strength degradation of

Failure load (kN)

8.9–18.1% and 0.1–1.4% occurred in the channel and angle connec-

tors, respectively, subjected to fully-reversed fatigue loading. The
angle shear connectors had considerably less shear capacity degra-
Property

dation compared with those channel in the case of low cyclic load-
(%)
Table 6

ing. The 100 mm high connectors had a higher load capacity than
the 75 mm high connectors in both channel and angle connectors.
 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392 387

(a) Fractured channel attached to steel I beam (b) Fractured channel in the slab

(c) Fractured angle attached to steel I beam (d) Fractured angle in the slab

Fig. 4. Typical fracture of (a, b) channel and (c, d) angle connectors.

Table 7
Comparison of shear strength capacity and relative slip: Channel vs angle.

Specimen Monotonic failure Channels/angles monotonic strength Relative slip Cyclic failure load Channels/angles cyclic strength Relative slip
load (kN) degradation (%) (mm) (kN) degradation (%) (mm)
H1C10050 191.3 6.8 7.0 174.3 0.0 7.5
H1A10050 178.3 0.5 175.8 1.0
H1C7550 186.3 17.9 6.5 165.8 9.0 7.0
H1A7550 152.9 1.0 150.8 2.5
H2C10030 109.6 24.2 6.5 90.8 11.9 7.0
H2A10030 83.1 1.5 80.0 1.5
H2C7530 91.1 30.1 4.5 74.6 18.5 6.5
H2A7530 63.7 2.0 60.8 2.5

This is because the shorter connector covers a comparatively smal-
ler loading area. Higher connectors were also more flexible. The
maximum slip was 6.5–7.0 mm for the 100 mm high channels
compared with 4.0–6.5 mm for the 75 mm high channels. The cor-
responding slip ranged from 0.5 mm to 1.5 mm for both the
100 mm high and 75 mm high angles.
The 50 mm long connectors had a higher load capacity than the
30 mm long connectors in both the channel and angle connectors.
The specimens with 50 mm long connectors were more flexible
than those with the 30 mm long connectors. The slip at the ulti-
mate load level was 6.5–7.0 mm for specimens with the 50 mm
long channel connectors compared with 4.0–6.5 mm for those with
30 mm. The slip was recorded at 0.5–1.5 mm for specimens with
50 mm long angle connectors compared with 1.5 mm for those
with 30 mm. Previous studies on these connectors showed that
they were more flexible in HSC when using longer connectors in
contrast to when embedded in a normal strength concrete.
The longer connectors experienced more cracks that run paral-
lel throughout the longer edge of the top surface of the slab. The
slabs with longer connectors experienced concrete cracking on
the sides of the slab when connector fracture occurred. Slabs with
shorter connectors were affected by the development of small
cracks around the slab center. The channel connectors exhibited
more cracks compared with the angle connectors regardless of
388 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392

Table 8 Table 10
Comparison of shear strength capacity and relative slip: Connectors with the same Ductility factor for each specimen.
length.
Specimen Dmax Dy l lChannel/lAngle
Specimen Monotonic Monotonic Cyclic Channels/angles cyclic
H1C10050-M 13.5 4.5 3.0 1.4
failure load strength failure strength degradation
H1A10050-M 8.5 4.0 2.1
(kN) degradation (%) load (kN) (%)
H1C10050-C 11.5 3.0 3.8 1.9
H1C10050 191.3 2.6 174.3 4.9
H1A10050-C 6.0 3.0 2.0
H1C7550 186.3 165.8
H1C7550-M 9.5 5.0 1.9 1.0
H1A10050 178.3 14.3 175.8 14.2
H1A7550-M 6.0 3.0 2.0
H1A7550 152.9 150.8
H1C7550-C 8.0 3.0 2.7 1.2
H2C10030 109.6 16.9 90.8 17.8
H1A7550-C 4.5 2.0 2.3
H2C7530 91.1 74.6
H2C10030-M 11.5 4.0 2.9 1.2
H2A10030 83.1 23.4 80.0 24.0
H2A10030-M 6.0 2.5 2.4
H2A7530 63.7 60.8
H2C10030-C 12.5 5.0 2.5 1.7
H2A10030-C 4.5 3.0 1.5
H2C7530-M 9.0 4.0 2.3 0.9
Table 9 H2A7530-M 5.0 2.0 2.5
Comparison of shear strength capacity and relative slip: Connectors with same height. H2C7530-C 9.0 3.5 2.6 1.2
H2A7530-C 5.5 2.5 2.2
Specimen Monotonic Monotonic Cyclic Channels/angles cyclic
failure load strength failure strength degradation
(kN) degradation (%) load (kN) (%)
H1C10050 191.3 42.7 174.3 47.9 angle connectors have a significant strength degradation compared
H2C10030 109.6 90.8 with channels when considering only the same connector length or
H1A10050 178.3 53.4 175.8 54.5 height in similar loading conditions.
H2A10030 83.1 80.0
H1C7550 186.3 51.1 165.8 55.0
H2C7530 91.1 74.6 3.3. Assessment of monotonic loading
H1A7550 152.9 58.4 150.8 59.7
H2A7530 63.7 60.8 The load-slip curves for an accurate one-to-one comparison of
specimens with the same dimensions of angle and channel connec-
tors were readily offered in Fig. 3. The curve was terminated unex-
pectedly in each specimen. Conventionally, the strength of
concrete and shear connectors are directly proportional to concrete
failure [3,51–53]. The connector yields and then fractures near the
fillet weld between the connector and steel beam during connector
failure [21].
The slip tends to occur between the concrete slab and steel I-
beam in both monotonic and cyclic load conditions. The monotonic
curves show that when the slip is higher than 4.0 mm, consider-
able ductility exists in all channel shear connectors. Less than
2.5 mm of slip was observed in all angle connector specimens
under both monotonic and cyclic load conditions.
Fig. 5 shows a general elasto-plastic load-slip curve replicating
those observed in the measurement, from which the ductility fac-
tor of shear connectors, l, is defined. The initial slope and horizon-
tal straight line in the load-slip curve are used to determine the
connection stiffness and strength, respectively. The displacement
up to the juncture of these lines represents the corresponding yield
Fig. 5. Definition of ductility factor. displacement, Dy. The deflection where the connector failure
occurs is represented as the maximum deflection, Dmax. Subse-
quently, the expression, Dmax/Dy, defines the ductility factor. A
either embedded in the HSC or normal concrete. More concrete more ductile composite system has a higher ductility factor with
cracks were observed for longer connectors. Table 7 shows that an increased inelastic redistribution of the applied load [54]. The
channel shear connectors have a higher shear strength compared calculated ductility factor for each specimen is presented in
with those angle. The comparison of both connectors with similar Table 10, which shows that more ductility was found in channel
dimensions and loading conditions indicates that angle connectors connectors compared with those of angle under cyclic load because
experienced a strength degradation of 6.8–30.1% and 0.0–18.5% the ratio lChannel/lAngle is in general more than 1.
under static and low cyclic fatigue loadings, respectively. Almost all channel specimens had a yield plateau, which indi-
The monotonic loading curve analysis reveals that the connec- cates that the slip increases as soon as the load achieves its ulti-
tor’s shear resistance and slip can be affected by changes in the mate level. The yield plateau was not obviously observed in the
height and length of the connectors in both the channel and angle angle connectors (Fig. 3).
connectors. Tables 8 and 9 show the enhancement of shear Fig. 6 and Table 7 show that the angle shear connectors under
strength corresponding with the increased height and length of static load experienced a strength degradation of 6.8–30.1% more
the connectors, respectively. Such an increased in strength is a than the channel shear connectors. The comparison ranges, when
direct result of a greater interaction of connectors with the con- normal reinforced concrete was used, were 7.5–36.4% less shear
crete, owing to larger contact area. These tables also show that strength for the angle connectors than the channel under static
 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392 389

Fig. 6. Load-slip curves comparison between channel and angle connectors of the same dimensions.

Fig. 7. Typical load-slip curve of specimen with channel connectors under cyclic loading.

loading [38]. Even with a better ultimate shear capacity, the angle
connectors could not fulfill the ductility requirements recom-
mended by Eurocode 4 [55]. The angle connector’s resistance
dropped below 90% of the maximum load for a slip of less than
2.5 mm. This reduction was more abrupt for the higher angle con-
nectors. These observations were similarly seen in the behavior of
both connectors when embedded in the normal reinforced con-
crete [38].
3.4. Assessment of reversed cyclic loading
Distinctive load-slip relations for the channel and angle speci-
mens under cyclic loading are shown in Figs. 7 and 8, respectively.
A comparison of static and fatigue testing of similar specimens for
both channel and angle connectors show that the introduction of
fully-reversed repeated loading lowered the connector capacity
immediately after a specific number of load cycles. The specific
390 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392
Fig. 8. Typical load-slip curve of specimen with angle connectors under cyclic
loading.
number of load cycles can be considered as the shear capacity of
the connector to resist failure under fatigue loading. This stress
level demands a cyclic testing to assure considerable fatigue [21].
Collectively, both channel and angle connectors efficiently sus-
tained 1.5 cycles at one-third and two-thirds of the failure load for
specimens tested under static loading (M), but failed during 1.5
cycles when the full value of M was applied. Each angle shear con-
nector experienced cyclic failure during the first half cycle of M.
The comparison was readily presented in Table 6. The angle shear
connectors show 0.0–18.5% strength degradation compared with
those of channel under fully-reversed cyclic loading. This indicates
that the shear strength of the channel connectors was 81.5–100%
greater than that of the angle connectors under fully-reversed cyc-
lic loading depending on their dimensions. These comparative
ranges were 23.6–49.2% when the connectors were embedded in
the normal reinforced concrete [38]. Thus, all the channel speci-
mens resisted 81.9–91.1% of their monotonic capacity compared
with the angle shear connectors, which resisted 98.6–99.9% of their
monotonic capacity. The fully-reversed fatigue loading caused the
concrete to be crushed near the bottom surface of the connectors
Fig. 9. Details of connectors’ specifications.
on the compressive face of the slabs. This was also attributed to
the slip and plastic yielding in this region.
3.5. Assessment of existing design codes
Table 7 represents the measured monotonic failure loads for all
connectors. These loads were compared with the formulas for the
ultimate strength capacity suggested by the AISC [20] (Eq. (1)) and
NBC [26] (Eq. (2)). Fig. 9 shows the details of the connectors’ spec-
ifications described by the equations. A comparison between the
resulting actual shear capacities of the angle shear connectors with
those obtained from Eqs. (3) and (4) was also performed. The ratios
from this comparison are shown in Table 11.
Comparatively, the Canadian Code (Eq. (2)) shows more conser-
vative results compared with the equation suggested by the Amer-
ican Code (Eq. (1)) for channel connectors. Several previous studies
on connectors embedded in the normal concrete also obtained
these results [21,38]. Kiyomiya and Yokota [35] (Eq. (3)) proposed
a more conservative formula than that presented by Ros [37] (Eq.
(4)) in determining the shear capacity of the angle shear connec-
tors. Similar results were found in the case of normal concrete [38].
4. Conclusions
A comparative study for composite structures of steel and high
strength concrete materials bonded with channel and angle shear
connectors under monotonic and cyclic loads was successfully con-
ducted. The results reveal that the composite structures with angle
connectors showed 6.8–30.1% and 0.0–18.5% less shear strength
compared to those with channel, when subjected to static and
fully-reversed fatigue loading, respectively. The comparison
ranges, when normal reinforced concrete was used, was 7.5–
36.4% less shear strength for the angle shear connectors than those
of channel under monotonic loading, and 23.6–49.2% less under
fully-reversed cyclic loading from previous studies. Other principal
conclusions drawn from the current study include:

Both the angle and channel shear connectors experienced con-
nector fracture mode of failure in all tests.

Channel connectors exhibited higher ductility compared with
those of angle.

Greater connector heights enabled them to sustain larger loads
with more flexibility.

The channels had a yield plateau, whereas the angles did not
exhibit an obvious yield plateau.

The angle connectors were less affected by cracking than the
channel connectors under the same loading environment.
 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392 391

Table 11
Comparison of test results with code prediction.

Specimen Failure load (kN) Eq. (1) (kN) Test/Eq. (1) Eq. (2) (kN) Test/Eq. (2) Eq. (3) (kN) Test/Eq. (3) Eq. (4) (kN) Test/Eq. (4)
H1C10050 191.3 233.6 0.8 187.7 1.02
H1A10050 178.3 72.6 2.5 240.6 0.7
H1C7550 186.3 203.1 0.9 163.2 1.1
H1A7550 152.9 66.3 2.3 194.5 0.8
H2C10030 109.6 124.4 0.9 99.9 1.1
H2A10030 83.1 38.7 2.1 128.1 0.7
H2C7530 91.1 108.2 0.8 86.9 1.1
H2A7530 63.7 35.3 1.8 103.6 0.6

Longer connectors resulted in greater concrete cracking.

The Canadian Code (Eq. (2)) estimates the shear capacity of
channel connectors more accurately and conservatively com-
pared with the American Code (Eq. (1)).

The shear capacity of the angle connectors can be conserva-
tively estimated by Kiyomiya et al. [35] compared with that rec-
ommended by Ros [37].
Acknowledgements
The authors would like to acknowledge the support of the
University of Malaya Higher Impact Research (HIR) Grant: UM.C/
HIR/MOHE/ENG/57.
References
[1] A. Shariati, N.H. Ramli Sulong, M. Suhatril, M. Shariati, Various types of shear
connectors in composite structures: a review, Int. J. Phys. Sci. 7 (22) (2012)
2876–2890.
[2] R. Slutter, G.C. Driscoll, Flexural strength of steel-concrete composite beams, J.
Struct. Eng. ASCE. (1965) 71–99.
[3] I. Viest, Investigation of Stud Shear Connectors for Composite Concrete and
Steel T-Beams, ACI, 1956.
[4] W. Zellner, Recent designs of composite bridges and a new type of shear
connectors, in: Proceedings of the IABSE/ASCE Engineering Foundation
Conference on Composite Construction, 1987, pp. 240–252. Henniker, NH.
[5] M. Veldanda, M. Hosain, Behavior of Perfobond rib shear connectors: push-out
tests, Can. J. Civ. Eng. 19 (1) (1992) 1–10.
[6] E. Oguejiofor, M. Hosain, Behaviour of Perfobond rib shear connectors in
composite beams: full-size tests, Can. J. Civ. Eng. 19 (2) (1992) 224–235.
[7] E.C. Oguejiofor, Perfobond rib shear connectors for composite beams Ph.D.
thesis, University of Saskatchewan, 1994.
[8] E. Oguejiofor, M. Hosain, A parametric study of Perfobond rib shear connectors,
Can. J. Civ. Eng. 21 (4) (1994) 614–625.
[9] I. Viest, Full-scale tests of channel shear connectors and composite T-beams,
Univ. Illinois Eng. Exp. Station Bull. no 405 (1951).
[10] I. Viest, C. Siess, J. Appleton, N. Newmark, Studies of slab and beam highway
bridges: Part IV. Full scale tests of channel shear connectors and composite T-
beams, University of Illinois Engineering Excerpt, Station Bulletin, 1952, p.
405.
[11] L. Dallam, Push-Out Tests of Stud and Channel Shear Connectors in Normal-
Weight and Lightweight Concrete Slabs, University of Missouri-Columbia,
Columbia, 1968.
[12] Y. Hiroshi, O. Kiyomiya, Load Carrying Capacity of Shear Connectors made of
Shape Steel in Steel-Concrete Composite Members. Structures Division
Subaqueous Tunnels and Pipelines Laboratory, 1987. PARI Technical Note
0595.
[13] A. Shariati, N.H. Ramli Sulong, M. Suhatril, M. Shariati, Investigation of channel
shear connectors for composite concrete and steel T-beam, Int. J. Phys. Sci. 7
(11) (2012) 1828–1831.
[14] J.D.C. Vianna, L.F. Costa-Neves, P.D.S. Vellasco, S.A.L. Andrade, Experimental
assessment of Perfobond and T-Perfobond shear connectors’ structural
response, J. Constr. Steel Res. 65 (2) (2009) 408–421.
[15] P. Chromiak, J. Studnicka, Computer model of Perfobond connector,
Proceedings of SDSS Conference, Lisbon, Portugal, 2006.
[16] J.P.C. Rodrigues, L. Laím, Behaviour of Perfobond shear connectors at high
temperatures, Eng. Struct. 33 (10) (2011) 2744–2753.
[17] G.S. Veŕıssimo, M.I.B. Valente, J.L.R. Paes, P.J.S. Cruz, R.H. Fakury, in: Design and
experimental analysis of a new shear connector for steel and concrete
composite structures, 3rd International Conference on Bridge Maintenance,
Safety and Management, Porto, Portugal, 2006.
[18] A. Pashan, Behaviour of Channel Shear Connectors: Push-Out Tests M.S. thesis,
Department of Civil Engineering, University of Saskatchewan, Canada, 2006.
[19] I. Viest, C. Siess, J. Appleton, N. Newmark, Full-scale tests of channel shear
connectors and composite T-beams, Bull. Series (1952) 405.
[20] AISC, Specification for Structural Steel Buildings, American Institute of Steel
Construction. Inc, Chicago (IL), 2005.
[21] S. Maleki, S. Bagheri, Behavior of channel shear connectors, Part I:
experimental study, J. Constr. Steel Res. 64 (2008) 1333–1340.
[22] S. Maleki, S. Bagheri, Behavior of channel shear connectors, Part II: analytical
study, J. Constr. Steel Res. 64 (2008) 1341–1348.
[23] S. Maleki, M. Mahoutian, Experimental and analytical study on channel shear
connectors in fiber-reinforced concrete, J. Constr. Steel Res. 65 (8–9) (2009)
1787–1793.
[24] A. Pashan, M. Hosain, New design equations for channel shear connectors in
composite beams, Can. J. Civ. Eng. 36 (9) (2009) 1435–1443.
[25] E. Baran, C. Topkaya, An experimental study on channel type shear connectors,
J. Constr. Steel Res. 74 (2012) 108–117.
[26] NBC, National Building Code of Canada, 2005.
[27] M. Shariati, N.H. Ramli Sulong, M. Suhatril, A. Shariati, M.M. Arabnejad
Khanouki, H. Sinaei, Behaviour of C-shaped angle shear connectors under
monotonic and fully reversed cyclic loading: an experimental study, Mater.
Des. 41 (2012) 67–73.
[28] A. Shariati, M. Shariati, N.H. Ramli Sulong, M. Suhatril, M.M. Arabnejad
Khanouki, M. Mahoutian, Experimental assessment of angle shear connectors
under monotonic and fully reversed cyclic loading in high strength concrete,
Constr. Build. Mater. 52 (2014) 276–283.
[29] A. Shariati, M. Shariati, N.H. Ramli Sulong, M. Suhatril, M.M. Arabnejad
Khanouki, Fatigue energy dissipation and failure analysis of angle shear
connectors embedded in high strength concrete, Eng. Fail. Anal. 41 (2014)
124–134.
[30] O. Kiyomiya, H. Yokota, M. Suzuki, T. Chiba, Strength Properties of Shear
Connectors by Shape Steel, Transactions of the Japan Concrete Institute, 1986.
[31] S.M. Choi, Fatigue resistance of angle shape shear connector used in steel-
concrete composite slab Ph.D. thesis, Graduate School of Engineering, Nagoya
University, Japan, 2011.
[32] K. Fukazawa, M. Sakai, N. Sudou, K. Kobayashi, Fatigue durability of steel-
concrete composite slab, MELAB and application to continuous composite
steel girder bridge, Mitsui Zosen Tech. Rev. 6 (2002) 8–18.
[33] T. Saidi, H. Furuuchi, T. Ueda, The transferred shear force-relative displacement
relationship of the shear connector in steel-concrete sandwich beam and its
model, Doboku Gakkai Ronbunshuu E. 64 (1) (2008) 122–141.
[34] S. Ros, H. Shima, A New Beam Type Test Method for Load-Slip Relationship of
L-Shape Shear Connector, The 8th Symposium on Research and Application of
Hybrid and Composite Structures, Japan Society of Civil Engineers, 2009, pp. 1–
8.
[35] O. Kiyomiya, H. Yokota, Strength of shear connector by shape steel in
composite member with steel and concrete, Proc. Symp. Res. Appl. Compos.
Constr., JSCE (1986) 113–118.
[36] M. Yamada, O. Kiyomiya, Experimental Study on the Loading Capacity of L-
Shape and Headed Stud Shear Connectors for Composite Structures, Port and
Harbour Research Institute Ministry of Transport Japan, 1986.
[37] S. Ros, Formulation for shear force-relative displacement relationship of L-
shape shear connector in steel-concrete composite structures Ph.D. thesis,
Kochi University of Technology Academic Resource Repository, Japan, 2011.
[38] M. Shariati, N.H. Ramli Sulong, M. Suhatril, A. Shariati, M.M. Arabnejad
Khanouki, H. Sinaei, Comparison of behaviour between channel and angle
shear connectors under monotonic and fully reversed cyclic loading, Constr.
Build. Mater. 38 (2013) 582–593.
[39] A.A. Shah, Y. Ribakov, Recent trends in steel fibered high-strength concrete,
Mater. Des. 32 (8–9) (2011) 4122–4151.
[40] J.M. Portolés, M.L. Romero, J.L. Bonet, F.C. Filippou, Experimental study of high
strength concrete-filled circular tubular columns under eccentric loading, J.
Constr. Steel Res. 67 (4) (2011) 623–633.
[41] S. Bullo, R. Di Marco, A simplified method for assessing the ductile behaviour
of stud connectors in composite beams with high strength concrete slab, J.
Constr. Steel Res. 60 (9) (2004) 1387–1408.
[42] L. An, K. Cederwall, Push-out tests on studs in high strength and normal
strength concrete, J. Constr. Steel Res. 36 (1) (1996) 15–29.
[43] L.Z. Chen, G. Ranzi, S.C. Jiang, F. Tahmasebinia, G.Q. Li, Behaviour and design of
shear connectors in composite slabs at elevated temperatures, J. Constr. Steel
Res. 115 (2015) 387–397.
392 M. Shariati et al. / Construction and Building Materials 120 (2016) 382–392
[44] M.A.R. Bhutta, T. Maruya, K. Tsuruta, Use of polymer-impregnated concrete
permanent form in marine environment: 10-year outdoor exposure in Saudi
Arabia, Constr. Build. Mater. 43 (2013) 50–57.
[45] ASTM C39 (2004), Standard test method for compressive strength of
cylindrical concrete specimens, Annual Book of ASTM Standards (2005).
[46] F. Sajedi, H.A. Razak, Effects of curing regimes and cement fineness on the
compressive strength of ordinary Portland cement mortars, Constr. Build.
Mater. 25 (4) (2010) 2036–2045.
[47] F. Sajedi, H.A. Razak, Comparison of different methods for activation of
ordinary Portland cement-slag mortars, Constr. Build. Mater. 25 (1) (2011) 30–
38.
[48] ASTM C150, Standard Specification for Portland Cements, American Society
for Testing and Materials, ASTM International, West Conshohocken, PA, 2003.
[49] F. Sajedi, H.A. Razak, The effect of chemical activators on early strength of
ordinary Portland cement-slag mortars, Constr. Build. Mater. 24 (10) (2010)
1944–1951.
[50] S.A. Civjan, P. Singh, Behavior of Shear Studs Subjected to Fully Reversed Cyclic
Loading, J. Struct. Eng. – ASCE. 129 (11) (2003) 1466–1474.
[51] C. Davies, Small-scale push-out tests on welded stud shear connectors,
Concrete 1 (9) (1967) 311–316.
[52] R. Slutter, G. Driscoll, Test results and design recommendations for composite
beams, Lehigh Univ. Fritz Eng. Laborat. Rep. 279 (10) (1962).
[53] J. Ollgaard, R. Slutter, J. Fisher, Shear strength of stud connectors in lightweight
and normal-weight concrete, J. Struct. Eng. – ASCE. 8 (2) (1971) 55–64.
[54] G. Kwon, M.D. Engelhardt, R.E. Klingner, Parametric studies and preliminary
design recommendations on the use of post-installed shear connectors for
strengthening non-composite steel bridges, J. Bridge Eng. 1 (2011) 162.
[55] Eurocode, Design of Composite Steel and Concrete Structures. Part 1. 1:
General Rules and Rules for Buildings, 1994.