Behaviour of concrete composite slabs with truss type shear connectors of different orientation angle

Abstract

Concrete composite slab system with pre-cast bottom segment and cast-in-situ top segment is gaining importance due to wide-ranging applications in buildings, bridges and industrial flooring, etc. These slabs may also be subjected to repeated loading during its service life and hence, it is essential to evaluate their static and fatigue behaviour under flexural loading. In present study, at first, experimental investigations are conducted to evaluate the performance of composite slabs with truss type shear connectors with two orientation angles, namely, 45° and 60°. The composite slab specimens are subjected to four-point bending. The load-versus-displacement responses obtained for the composite slabs are compared with that of control full slab. It is observed that the composite slab with 45° shear connector showed better performance when compared to composite slab with 60° shear connector in terms of load-displacement response, rotation, ductility and energy absorption and also found to be efficient in terms of area of steel crossing the interface. Further, numerical investigations are also carried out to assess load transfer mechanism of truss type shear connectors at different significant events. The force flow mechanism of truss type shear connectors is brought out clearly. From the numerical investigations on composite slabs with different orientation angles, it is observed that composite slab with 45° shear connector exhibited superior performance over other orientation angles in terms of better force flow mechanism and lesser area of steel crossing the interface. Further, fatigue performance of composite slab with 45° shear connector is evaluated and compared with the fatigue performance of control slab. It could be observed that the composite slab with 45° truss type shear connector showed better fatigue performance when compared to the control slab.

Keywords

concrete composite slab flexural behaviour monotonic loading and fatigue loading orientation angle truss type shear connector

Introduction

Rapid economic growth is closely linked with the development of infrastructure facilities, especially in the developing countries. This huge infrastructural demand could be met by employing fast track construction practices, such as precast and composite construction. Precast construction enables fast track construction, economy in large scale and ensures quality in view of its mechanised industrial setup. The social, environmental and structural benefits of precast concrete construction were very well established by Yee (2001) and Yee and Eng (2001). Concrete composite slabs find wide range of applications like bridge decks (for highways and railways), and floors or roofs for repair and rehabilitation of damaged slabs etc. Most frequently, concrete composite construction is employed to reduce total construction time. Concrete composite slab construction, also popularly known as half-slab construction, has a precast bottom slab segment and a cast-in-place top portion of the slab. Precast bottom slab segment initially acts as formwork and is integrated with cast-in-situ concrete segment using shear connectors for the required bending and shear resistance and thereby reducing the staging requirement.

For the precast and in situ segments to behave monolithically, the interface must remain intact and composite member shall deform as a single unit when loaded. The intact interface will facilitate the transfer of horizontal shear forces across the interface. The full-composite action is reflected in strains remaining essentially linear across the entire slab thickness. However, when a composite slab with a weaker interface is loaded, the interface slips and the slab elements resist a portion of the load as two separate members having two neutral axes. Similarly, if the interface is in such a way that it provides partial composite connection, the strain across the depth varies accordingly. The insulating materials like expanded polystyrene, extruded polystyrene and foam panels are also being provided between the concrete segments to perform the dual function of transferring load and thermal insulation (Amran et al., 2016; Carbonari et al., 2012; Fernando et al., 2017; Huang et al., 2020) and are known as sandwich panels. They are often served as the cladding for the buildings. As there is insulating medium between the two slab layers, the contribution of interface cohesion and interlocking friction will be absent and the entire shear resistance has to be mobilised from the shear connectors. Hence, the shear transfer is often associated with the slip between the slab segments. Thus, in this category of precast concrete sandwich panels, composite action will be partial and the degree of compositeness will generally be less than 100%.

Numerous studies were reported on composite slabs with different shear connectors for slab or wall panel applications. The shear connector varies depending on the type of material, shape, continuous truss or discrete shear connectors, structural action (one-way or two-way) and hybrid connectors. Usage of steel based shear connectors either in discrete or continuous truss form is the most common and also commercially available. The steel shear connectors were used in the form of wires, bent bars, ties, C-clip, steel plate, Z shaped and truss connectors (Ahmad et al., 2013; Amin Eniea et al., 1991; Benayoune et al., 2008; Bush and Stine, 1994; Gara et al., 2012; Kinnane et al., 2020; Lee and Pessiki, 2008; Naito et al., 2012; Yuksel et al., 2020). Non-metallic shear connectors were used by researchers with the view of increasing the thermal efficiency as the metal connectors will conduct heat (Einea et al., 1994; Frankl et al., 2011; Henin et al., 2014; Hopkins et al., 2017; Jiang et al., 2018; Salmon et al., 1997; Tomlinson and Fam, 2015). In general, discrete shear connectors are orthogonal to the interface except for specific shapes such as W-, Z-, M-, N-shaped connectors. Shear connectors in the form of truss with different angles of inclination with the horizontal, such as 45° (Amran et al., 2016; Benayoune et al., 2008; Henin et al., 2014), 50° (Salmon et al., 1997), 54° (Salmon et al., 1997), and 67.5° (Thanoon et al., 2010) were used.

The composite slab systems were also employed for bridge deck systems, for long span basement parking systems and hollow core slab with in situ toppings (Badie et al., 1998; Baran, 2015; Dowell and Auer, 2011; Ibrahim et al., 2016; Lee et al., 2016; Peterman and Ramirez, 1998). Mansour et al. (2015) investigated the flexural behaviour of precast concrete slab with steel fibre reinforced toppings. It was reported that the ductility and toughness improved with the addition of steel fibre reinforced toppings to the precast concrete slab. Mohamed et al. (2020) assessed the flexural performance of semi-precast slabs with in situ concrete slabs with different interface detailing and shear ties. The study reported that the type of surface treatment has pronounced influence on the flexural capacity of composite slab systems tested. Limited studies were reported to study the behaviour of composite slabs under cyclic loading (Bush and Stine, 1994; Choi et al., 2016; Peterman and Ramirez, 1998; Teixeira and Fam, 2017). Bush and Stine (1994) studied behaviour of composite slabs under cyclic loading due to thermal gradient and reported that a decrease in stiffness was observed after 55,000 cycles of loading. Peterman and Ramirez (1998) applied 5 million cycles of service load before testing the composite slabs to failure. It was reported that the ultimate moment capacity of slab did not decrease. Choi et al. (2016) and Teixeira and Fam (2017) studied the effect of wind induced cyclic loading on composite slab panels.

From the reported studies, it is very clear that better structural efficiency could be obtained by metallic shear connectors in the form of continuous truss. But, studies on effectiveness of truss type shear connectors of different orientation angles in achieving the composite action are scarce. Hence, an attempt is made in this study to evaluate the influence of orientation angle of the truss type shear connector on the behaviour of composite slab. Moreover, very few studies were reported in the literature on the fatigue behaviour of composite slabs. With the increasing usage of composite slabs as highway bridge decks, airport runway bridge decks to cross drainage channels where there is absolute necessity to use concrete composite slabs to avoid formwork and staging, and industrial floors, they are subjected to repetitive cyclic loads and thus study on the fatigue behaviour of composite slabs assumes importance. Hence, attempts are also made to assess the fatigue performance of composite slab with identified orientation angle of shear connectors. In present study, experimental investigations are carried out on flexural behaviour of composite slabs with 45° and 60° (representing the different range of orientation angles reported in literature) orientation angle of truss type shear connectors under four-point bending. Numerical investigations are also carried out to ensure repeatability of the experimental results and also to get better insight into the behaviour of composite slabs. Further, numerical studies are also used to bring out the efficiency of shear connectors with different orientation angles. Furthermore, experimental investigations are carried out to evaluate the fatigue performance of composite slabs with identified shear connector orientation angle and compared with that of the control slab to qualify the suitability of composite slab for bridge deck application.

Details of slab specimens chosen for the study

One-way slab of size 2.4 m × 0.8 m × 0.15 m is taken up for the study. Out of the three slab specimens considered for monotonic test, one is the control that is the full slab and the entire depth of the slab specimen is cast in single pour of concrete and the other two are composite slabs and are cast as two half slabs that is in two pours. The slab specimens are reinforced with 8 mm diameter bars at the spacing of 130 mm c/c as main reinforcement and 8 mm diameter at the spacing of 250 mm c/c as distributor steel as shown in Figure 1(a). Truss type shear connectors with 45° and 60° orientation angle as shown in Figure 1(b) are chosen for composite slabs. The detailed design of shear connectors is presented in Appendix A.

Figure 1.

(a) Reinforcement details of the slab specimen and (b) details of truss type shear connector units used for study.

Details of casting of control and composite slab specimens

For casting of slab specimens, concrete mix ratio of Cement (1):sand (1.695):coarse aggregate (3.013) with w/c ratio of 0.5 is used. Clear cover to the main reinforcement is 25 mm. For control slab, the entire depth of 150 mm is cast in single pour and wet cured for a period of 28 days. The composite slab specimens are cast in two stages. The bottom half of the slab is cast on the ground and second half slab is cast over the already cast bottom slab after 7 days.The first pour of concrete is done up to the depth of 75 mm simulating the precast segment and second pour of concrete is done after 7 days to the remaining depth of 75 mm simulating the cast-in-situ portion of concrete slab as shown in Figure 2. After the first pour, the surface of the bottom slab is leveled and the surface of the concrete is not roughened using any special means. Hence, no further measurements are made to quantify the surface roughness of the concrete bottom slab. The composite slab specimens are provided with truss type shear connector of different orientation angles, namely, 45° and 60° with the horizontal (i.e. direction of main reinforcement). The composite slab specimens are cured initially for 7 days after first pour of concrete to half-depth and then cured for a period of 28 days after the second pour of concrete in the balance half-depth of the specimen is completed. The average cylinder compressive strength of concrete cast along with slabs to be tested under static and fatigue loading are found to be 39.2 and 37.22 MPa respectively.The yield and ultimate strength of reinforcement bars are found to be 527.5 and 641 MPa respectively. The yield and ultimate strength of shear connectors are found to be 546.7 and 657.5 MPa respectively.

Figure 2.

Casting sequence of twin layered concrete composite slab specimens.

Experimental investigations on control and concrete composite slab specimens under monotonic loading

The effective span of the slab is 2.1 m with the constant bending moment zone of 700 mm. The slab specimens are subjected to four-point bending. Simply supported boundary conditions are simulated by supporting one edge of slab over a roller and the other edge over a hinge plate assembly as shown in Figure 3. The deflections of the slab specimen along the span length are measured by means of the LVDTs (linear variable displacement transducer) at centre and quarter span points of the slab. The rotation of the slab is measured using digital tilt meter.The loads are applied using hydraulic jack fixed to the loading frame of 30 t capacity. A calibrated load cell is used for measuring the load applied by the hydraulic jack. The specimen is given an initial load of 15 kN and the load is released to check the functionality of all the sensors. This process is repeated thrice to ensure the repeatability of displacement within the elastic regime. The load is applied gradually till the failure of the specimen. Displacements, load, rotations are monitored and recorded throughout the experiment. The crack patterns observed during the tests are noted for all the tested specimens.

Figure 3.

Typical test set up for composite slab specimen under monotonic flexural loading.

Flexural behaviour of control and composite slab specimens under monotonic loading

Damage progression

In control slab specimen, the first crack has appeared in the middle one-third (in constant bending zone) at the load of 30 kN. Upon further loading, most of the flexural cracks are developed in the constant bending zone and at the loading points. The flexural cracks are developed symmetrically about the centre of the span and propagated to the full width of the slab as shown in Figure 4(a). With the increase in loading, yielding of reinforcement and widening of cracks are observed. Finally, the specimen failed in flexural mode. The ultimate displacement undergone by the control slab is 92 mm.

Figure 4.

(a) Crack patterns observed in control slab specimen, (b) crack pattern observed in composite slab with 45° truss type shear connector, and (c) crack pattern observed in composite slab with 60° truss type shear connector.

In the composite slab specimen with 45° truss type shear connector, first crack has appeared in the constant bending region at the load of 40 kN. With further loading, yielding of the reinforcement occurred and then immediately after that interface cracks are developed at the load of 75 kN on both faces of the slab at constant shear zone. From the overall response of the slab, three distinct cracking patterns could be categorised, (i) continuous flexural cracks developed and grown to the full depth of the slab in constant bending zone, (ii) the discontinuous flexural cracks that is the cracks developed up to first half-slab and later propagated along the interface and developed further on the second half-slab at loading points, (iii) Cracks grown only up to bottom half-slab and not propagated into the second half-slab at constant shear zone as shown in Figure 4(b). The composite slab with 45° shear connector sustained ultimate deformation (97 mm) little more than the full control slab.

Initially, cracks are developed in the constant bending zone for the composite slab specimen with 60° truss type shear connector unit with first crack at the load of 38 kN. With further loading, yielding of reinforcement followed by interface cracking are developed at the load of 68 kN on one face of the slab. In this composite slab, crack pattern similar to that of composite slabs with 45° shear connectors observed in the face where interface cracks are formed. After interface cracking, the vertical crack at the centre became more active when compared to the other cracks and propagated to the top at the time of failure as shown in Figure 4(c). But the ultimate deformation undergone by the composite slab with 60° shear connector is 55 mm.

Load-displacement

The load versus displacement plots obtained from the experimental investigations on control and composite slabs with different orientations of truss type shear connectors are compared in Figure 5. Summary of loads at significant events of control and composite slab specimens are presented in Table 1. All the slab specimens exhibited nearly tri-linear load versus displacement behaviour, linear behaviour till first cracking, followed by stiffness reduction after first cracking and flattening of load-displacement after reinforcement yielding. The stiffness and load carried by the control specimen is slightly lesser than that of the composite slabs. This is due to the presence of additional chord reinforcement in view of shear connector in the composite slab in addition to the main reinforcement at the slab bottom. The composite slab with 45° and 60° truss type shear connectors had one number of 8 mm rebar as chord reinforcement where the 6 mm diameter inclined rods are welded to 8 mm reinforcements. But the overall performance of the composite slabs is similar to that of the control slab in terms of overall load-displacement response and also showed higher first cracking, yield, ultimate load when compared to control slab (Table 1). But there is difference in the damage progression behaviour of control and composite slabs. Interface cracks occurred in both composite slabs just after the reinforcement yielding, as design demand is exceeded in both composite slabs due to material over strength (Table 1). Among the composite slabs, the slab with 45° truss shear connector exhibited better load-displacement behaviour when compared to composite slab with 60° truss shear connector in terms of both load as well as displacement. The composite slab with 60° truss shear connector showed a sudden drop in load at the displacement of around 55 mm, whereas the composite slab with 45° truss type sustained ultimate displacement of 97 mm.

Figure 5.

Load versus displacement for control and composite slabs.

Table 1.

Summary of loads at significant events.

Specimen detail	First crack load (kN)	Load at interface crack (kN)	Yield load (kN)	Peak load (P_p) (kN)	Design shear flow (kN/m)	Shear flow at interface crack (kN/m)
Control full slab	30	–	62	75	–	–
Composite slab (45°)	40	75	73	88	218	325
Composite slab (60°)	38	68	67	77	218	290

The maximum rotation, ductility and energy absorption of control and composite slabs are compared in Table 2. Among the composite slab specimens, the composite slab with 45° truss type shear connector has under gone nearly same amount of rotation as that of control slab before failure. The composite slab with 60° shear connector has undergone nearly one-half of the rotation undergone by the control slab specimen. It is found that the energy absorbed by the composite slab with 45° shear connector is more than that of control slab as the slab carried more load as well as displacement when compared to control slab specimen. Whereas the composite slab with 60° showed lesser energy absorption when compared to control slab specimen even though it carried more load than control slab specimen, it sustained less displacement when compared to the control slab. It could be observed that the ductility of the composite slab with 45° truss connector is more than the control slab whereas the ductility of composite slab with 60° truss connector is lesser than the control slab specimen.Thus, it could be observed that the composite slab with 45° shear connectors is better than the composite slab with 60° shear connector in terms of load-displacement, energy absorbed, rotation, ductility, etc.

Table 2.

Rotation, ductility and energy dissipation of control and composite slabs.

Specimen detail	Maximum rotation (°)	Energy absorbed (kNm)	Yield displacement (mm)	Ultimate displacement (mm)	Ductility
Control full slab	5.7	6.17	11.74	91.533	7.8
Composite slab (45°)	5.39	7.8	9.14	96.758	10.57
Composite slab (60°)	2.75	3.79	8.44	54.58	6.46

In order to ascertain the better behaviour of composite slab with 45° truss type shear connector over composite slab with 60° shear connector, to ensure the repeatability of the experimental results and also to get better insight into force flow mechanism of shear connector used in the composite slabs nonlinear finite element analysis (NLFEA) is carried out. Further, influence of orientation angle of shear connector and its effectiveness in shear resistance could also be established using NLFEA.

Nonlinear finite element analysis

NLFEA of the control and composite slabs are carried out using Atena 3D. The full geometry of the composite slabs is modelled. The details of finite elements used for modelling of different components of the slab are shown in Figure 6. The details of material models and the properties adopted for the study are presented in Table 3. The tested material properties of steel and concrete are used for NLFEA. The coefficient of friction and cohesion specified for smooth interface which is not roughened intentionally as specified in AASHTO LRFD (2014) is used. The tangential stiffness for concrete-concrete interface as specified by Dias-da-Costa et al. (2012) is used. Further, the normal stiffness of the interface is assumed equal to the tangential stiffness of the interface. A mesh size of 25 mm is used for meshing of composite slab as well as supporting steel plates. Simply supported boundary conditions are assigned to the composite slab. The loads are applied in load controlled mode in smaller increments that is load steps.

Figure 6

(a) Elements used for interface, shear connector and reinforcement and (b) elements used for concrete and steel supporting/loading plate.

Table 3.

Detail of material modelling.

Description	Material modelling strategy	Material properties
Concrete	Fracture-plastic model is used for constitutive law incorporating fracture in tension and plasticity in compression and also includes tension stiffening, hardening and softening characteristic of concrete.	Cylinder strength = 31 MPa
		Tensile strength = 2.639 MPa
		Fracture energy = 6.598 × 10⁵ MN/m
		E = 3.28 × 10⁴ MPa
Reinforcement Steel and shear connector	Bilinear material model with strain hardening	Yield, ultimate strength and modulus as obtained from tests
Interface	Mohr-coulomb with tension cut-off	Normal and tangential stiffness = 2 × 10⁴ MN/m³
		Co-efficient of friction = 0.6
		Cohesion = 0.5 MPa
Bond Model	Rigid bond law that is no slip is allowed between steel and surrounding concrete	Perfect bond is assigned between reinforcement and concrete.

NLFEA for simulation of experimental investigation

The composite slabs with shear connector (45° and 60°) along with additional chord reinforcement are modelled as such to simulate that of experimental study and also to get better insight into the force flow in the shear connector at significant events. The load-versus displacement obtained for composite slab from the finite element analysis and experimental investigations are compared in Figure 7(a) and (b). It could be observed that the results obtained from the finite element analysis corroborated well with the load-displacement response of composite slab tested. Further, in order to get better insight into the behaviour of composite slabs, the force flow in the shear connectors are visualised in terms of axial stress in the shear connectors at three instances, namely (i) prior to first cracking of concrete, (ii) Just after the yielding of reinforcement and (iii) at the ultimate condition as shown in Figure 8(a) to (f). From Figure 8, it is clear that before concrete cracking, the force flow in the shear connector follows typical truss action, that is, tension and compression along the diagonals in constant shear zone and zero force in the constant bending zone. After yielding of reinforcement, the diagonals at the constant shear zone showed stress nearly close to the yield stress but the diagonal at the centre carried tensile stress. This is probably due to the bending action and the deflection is more at the centre and diagonals at the centre tend to prevent the separation and get stressed up. At the ultimate condition, the tensile stresses in diagonal in the constant bending region increased further with the increased displacement even though the increase in load is very small between yield to ultimate. Further, it is also observed that in the constant shear zone, the force flow in the shear connector is through truss action and hence, the vertical component of the compression diagonal will not contribute to the shear resistance, as it will not provide clamping force to the surrounding concrete through shear-friction. In 60° shear connector, the majority of the shear resistance is mobilised through vertical component when compared to horizontal component and whereas in 45° shear connector, both components are same. The better performance of the composite slab with 45° truss type shear connector in terms of load as well as displacement capacity when compared to composite slab with 60° truss shear connector is attributed to the load resistance mechanism of the shear connector. Thus, the study brings out clearly behaviour of composite slab in terms of force flow in the shear connector.

Figure 7.

(a) Load-displacement curves of composite slab (45° shear connector) and (b) load-displacement curves of composite slab (60° shear connector).

Figure 8

(a) Shear connector stress in composite slab (45°) prior to first crack, (b) shear connector stress in composite slab (45°) after yielding, (c) shear connector stress in composite slab (45°) at ultimate, (d) shear connector stress in composite slab (60°) prior to first crack, (e) shear connector stress in composite slab (60°) after yielding, and (f) shear connector stress in composite slab (60°) at ultimate.

Performance of composite slab in the absence of additional chord reinforcement

In order to ensure the performance of composite slab in the absence of additional chord reinforcement as that of the control slab, NLFEA are carried out on composite slabs without additional chord reinforcement. The load-displacement behaviour obtained from finite element analysis are compared with the experimental load-displacement response of control slab in Figure 9(a). It could be observed from Figure 9(a) that in the absence of additional chord reinforcements, the stiffness of composite slabs is same as that of the control slab before flattening of load-displacement curve. Further, composite slab with 45° truss type shear connector, showed slightly higher yield and nearly same ultimate loads when compared to control full slab. Whereas ultimate loads of the composite slab with 60° shear connector is slightly lower than control slab. Further, the ultimate displacement of the composite slab with 60° shear connector is much lower than the control slab. Thus, it is observed that the composite slab with 45° shear connector showed better performance when compared to composite slab with 60° shear connector.

Figure 9.

(a) Load-displacement response of control and composite slab, (b) load-displacement of composite slab with different orientation angles, and (c) load-displacement responses of composite slabs without interface cohesion.

Influence of orientation angle on efficiency of shear connector

To evaluate the influence of orientation angles on efficiency of shear connector in terms of area of steel crossing the interface, composite slabs with different orientation angles of shear connector such as 30, 45, 60 and 90 are chosen. The design of shear connector is presented in the Appendix A. From Table A1, it is clear that the area of steel required for truss type shear connector is much lesser when compared to shear connector with 90° orientation angle. The load-displacement curves obtained for composite slabs with different orientation angles are compared with control slab in Figure 9(b). From Figure 9(b), it is observed that composite slabs with inclined shear connectors and control slab have same yield load. The ultimate load of composite slab with 45° shear connector is same as that of control slab whereas the other composite slabs with inclined shear connectors showed ultimate load slightly lower than that of control slab. It could be observed that the composite slab with 90° shear connector carried lower yield and ultimate load when compared with the control slab, even though the area of shear connector steel provided in the composite slab with 90° shear connector is more when compared to the inclined shear connectors. In the case of slab with inclined shear connectors, the full flexural capacity could be achieved as that of the control slab. Thus, the composite slabs with inclined shear connectors are more efficient in terms of area of steel crossing the interface.

Load resistance of composite slabs with different orientation angles of shear connectors in the absence of interface cohesion

In order to assess the load resistance of composite slabs with different orientation angles of shear connectors without interface cohesion, composite slabs with 30, 45, 60 and 90° orientation angle are taken up. The interface between the slab segments is assumed to have no cohesion. This is done so by assuming interface cohesion as 0.001 MPa in the numerical model. This will represent the case of composite slab when the interface is cracked or insulation layer is present between the slab segments. The area of shear connector steel as specified in Table A1 is provided for composite slabs with different orientation angles, that is, full shear connection is provided by shear connectors by assuming no contribution from the interface cohesion while designing the shear connection. The load-displacement responses of the composite slabs and full control slab are shown in Figure 9(c). From Figure 9(c), it is observed that the composite slabs with 30° and 45° shear connectors showed nearly same load carrying capacity as that of the control slab but the initial stiffness is slightly lower than control slab. In the case of composite slab with 60° shear connectors showed substantial reduction in load carrying as well as initial stiffness when compared to control slab. The composite slab with 90° shear connector, showed poor flexural performance when compared with control slab as well as composite slab with inclined shear connectors by showing lesser load carrying capacity and increased deflection. Even though all the composite slabs are designed for full shear connection ignoring interface cohesion, full composite action could not be achieved in the case of composite slabs with 60° and 90° shear connectors.All the composite slabs showed reduced ultimate deflection capacity when compared to that of the control slab due to interface failure. Further, the composite slabs showed more deformation after first cracking when compared to control slab. For example, at the load of 50 kN, the control and composite slab with 30°, 45°, 60°, 90° showed a displacement of 6.97, 8.9, 8.76, 10.77 and 34.86 mm respectively. The composite slab with 30° and 45° shear connector showed much better flexural performance when compared to composite slabs with other orientation angles of shear connector in the absence of interface cohesion. Hence, orientation angle of the shear connector plays a major role in determining the behaviour of composite slabs and also horizontal shear resistance is improved with the decrease in orientation angle with the horizontal in the absence of interface cohesion.

Experimental investigations on control and composite slab specimens under fatigue loading

From the experimental and numerical investigations,it is noted that the composite slab specimen with 45° orientation angle exhibited better performance under monotonic loading. Hence, composite slab with 45° truss type shear connector is chosen for investigating its performance under fatigue loading. The slab specimens, viz., full slab (control specimen) and concrete composite slab with 45° truss type shear connector unit, are taken up for investigation under fatigue loading. The test set up is similar to that used for testing of slabs under monotonic loading. The fatigue load is applied by means of the hydraulic actuator of 50 t capacity. The load ratio of 0.1 is maintained between the minimum and maximum load. The maximum load of 60% of ultimate load and minimum load of 6% of ultimate load of control specimen are used for cyclic loading. Hence, the load range is 54% of the ultimate load carried under static loading. The loading frequency of 1.5 Hz is maintained. The fatigue performance of the control and composite slabs are compared in terms of load-displacement hysteresis at selected cycles and stiffness degradation.

The full slab (control specimen) is the regular slab specimen where the concreting is done in single pour. The specimen is subjected to cyclic load varying from 4.2 to 42 kN till the failure of the specimen. The load range is kept constant throughout the loading history. Upon loading the specimen, flexural cracks appeared symmetrically on either side of the centre of the specimen in the constant bending moment zone as shown in Figure 10(a). With the increase in number of cycles, one of the flexural cracks in constant bending zone showed a predominant growth with increased crack width. With further increase in number of cycles, this predominant flexural crack propagated to nearly full depth of the slab and opened-up completely resulting in sudden fatigue failure of the specimen as shown in Figure 10(b). The specimen sustained 1, 64,149 cycles before the failure. The load-displacement curves obtained at different cycles of loading are depicted in Figure 11(a). It may be noted that the displacements at both maximum and minimum load, increase with increase in the number of cycles, indicating loss of stiffness. It may also be noted that the increase in displacement during the maximum load is higher than the increase in the displacement at minimum load till 150,000 cycles. After 150,000 cycles, both maximum and minimum displacements increased drastically just before the failure of the specimen.

Figure 10.

(a) Flexural cracks at initial cycles of control slab, (b) growth of predominant flexural crack at fatigue failure of control slab, (c) growth on one predominant crack in composite slab, and (d) closer view of flexural crack at fatigue failure of composite slab.

Figure 11.

Comparison of load versus displacement curves and stiffness degradation of control and composite slab: (a) control slab (full slab) specimen, (b) composite slab specimen, and (c) stiffness degradation of control and composite slab specimens.

Further, concrete composite slab specimen with 45° truss type shear connector is tested under fatigue loading. The specimen is subjected to fatigue load varying from 4.2 to 42 kN till the failure of the specimen. The composite slab with 45° truss type shear connector also showed similar fatigue behaviour as that of the control slab. Growth of one of the flexural cracks became predominant when compared to the other cracks (Figure 10(c)) and with the increase in the number of cycles, this crack had grown to full depth of the slab and resulted in sudden fatigue failure. The closer view of the flexural crack at the time of fatigue failure is as shown in Figure 10(d). No interface separation cracks are observed in the composite slab specimen as the load level (4.2–42 kN) applied is lower than the load at which interface crack (75 kN) developed in static test. The specimen sustained 362,510 cycles before the failure. The load-displacement curves at different load cycles are shown in Figure 11(b). The displacement undergone by the composite slab specimen is smaller than that of the control specimen throughout the history of the loading cycles. At failure, the loading and unloading paths exhibited huge difference when compared to that of the initial cycles of loading. Even though the displacement at minimum load increases with the number of cycles, the increase in the displacement at maximum load is found to be more predominant. The maximum deflection of control specimen at fatigue failure is 12.5 mm compared to the maximum deflection of 8.65 mm in the case of composite slab specimen at its fatigue failure.

The stiffness degradation undergone by control and composite slabs with 45° shear connectors are shown in Figure 11(c). The stiffness degradation behaviour of control and composite slab are similar with initial drop in stiffness due to cracking, then nearly uniform stiffness degradation indicating the crack propagation phase followed by sudden decrease in stiffness. The fatigue failure is manifested through increased deflection and sudden stiffness degradation. From Figure 11(c), it is also noted that the stiffness degradation undergone by the composite slab is lesser when compared to that of the control specimen. Thus, the composite slab showed better load-displacement hysteresis and stiffness degradation behaviour under fatigue loading. This clearly demonstrates the fatigue performance of composite slab specimen and qualifies for application in repeated loading scenarios particularly for bridge deck slab application where due to vehicle movement, the deck slabs are subjected to repeated loading during its service life.

Conclusions

From the study on full slab (control specimen) and concrete composite slabs with truss type shear connectors of different orientation angles, the following conclusions are drawn:

The composite slab specimens with truss type shear connectors could be designed to carry the ultimate load equal to or higher than that of the control specimen. Hence, the capacity of the composite slabs could be assumed as that of the full slab for design purpose and general moment equation prescribed by the code could be used for evaluating the moment capacity of the slab.

Experimental investigation clearly brings out the superior performance of composite slab with 45° truss type shear connector when compared to the composite slab with 60° truss type shear connector in terms of better load-displacement, rotation, ductility, energy absorption and also better damage progression behaviour. Results of numerical investigations on composite slabs with 45° and 60° shear connectors corroborated well with the experimental results and showcased the better performance of composite slab with 45° over composite slab with 60° shear connector. Further, it is essential to highlight the fact that 45° truss shear connector is efficient when compared to 60° truss type shear connector in terms of area of shear reinforcement crossing the interface and better load-transfer mechanism.

Numerical investigations on composite slab with different orientation angle such as 30°, 45°, 60° and 90° showed that for same flexural performance, the inclined shear connector needs lesser area of steel when compared to that of the composite slab with 90° shear connector. It is observed that the efficiency of shear connector increases with decrease in orientation angle in terms of area of steel crossing the interface.

In the absence of interface cohesion for shear resistance, the composite slabs showed reduced ultimate deflection capacity when compared to that of the control slab, full composite action could not be achieved in the case of composite slabs with 60° and 90° shear connectors.Further, among the composite slabs, composite slab with 30° and 45° shear connector showed much better flexural performance when compared to composite slabs with other orientation angles of shear connector. Hence, orientation angle of the shear connector plays a major role in determining the behaviour of composite slabs in the absence of interface cohesion.

The study on fatigue behaviour of composite slab reveals that as the number of fatigue cycles increase, stiffness of the specimen decreases. The increase in displacement during the maximum load is found to be higher than the increase in the displacement at minimum load. It is also noted that both maximum and minimum displacements increase drastically just before fatigue failure of the specimen. The fatigue failure is often associated with the sudden increase in displacement before fatigue failure.

The composite slab specimen with truss types shear connector showcased better fatigue performance and endured more fatigue cycles when compared to the control slab. This highlights the efficacy of composite slab with truss type shear connectors for bridge deck application.

Footnotes

Appendix A: Design of shear connectors

The shear connectors are designed for shear demand that would arise when the slab reaches its ultimate moment capacity. The rectangular parabolic stress block is assumed for the evaluation of section resistance. From the force equilibrium, the depth of neutral axis is obtained using the expression given below:

0.36 f_{ck} x_{u} b = 0.87 f_{y} A_{st}

The ultimate moment of resistance (M_u) of the section due to yielding of slab bottom reinforcement is given by

M_{u} = 0.87 f_{y} A_{st} (d - 0.42 x_{u})

Where, f_ck is the cube compressive characteristic strength of concrete; x_u is neutral axis depth; f_y is the yield strength of main steel reinforcement; A_st is the area of main steel reinforcement; d is the effective depth.

As the slab is loaded under two point loads, the vertical shear demand at the ultimate moment capacity could be estimated as follows:

M_{u} = Va

Where “a” is the distance from the centre of support to the point of application of point load.

For the design of shear connectors, the horizontal shear flow (V_h) could be evaluated by,

v_{h} = \frac{V}{d_{v}}

where,

V = flexural shear

d_v = distance between compressive and tensile forces

Assuming that the shear flow (v_h) is uniform over the length of the panel, the number of truss type shear connectors required are calculated based on equilibrium equation as given below. The notations used are as shown in Figure A1. The contribution of interface cohesion for shear resistance is ignored for the design of shear connectors as per ACI 318 (2019). The criterion to ensure full shear transfer at interface is as follows:

T_{u} = n A_{s} f_{yb} (\cos θ + μ \sin θ)

Where, T_u = v_h·l

T_u = ultimate capacity of the shear connector; v_h = shear flow; l = horizontal projection of individual truss element; n = no. of truss type shear connectors; A_s = Area of shear connector; f_yb = yield strength of the shear connector; θ = inclination of the shear connector with the horizontal; µ is coefficient of friction taken as 0.6 (ACI 318 (2019)). As the shear connectors are inclined, the shear resistance is mobilised by horizontal component parallel to interface and due to shear friction of reinforcement crossing the interface.In present study, continuous truss type shear connector is used for interface shear resistance between the slab segments, hence the longitudinal spacing of the diagonal rebarsdepends on the orientation angle of the shear connector. Since the truss is continuous, the diagonals will be repeated continuously along the length of the slab. The shear resistance is directly proportional to the diameter of the rebarand number of truss type shear connectors to be provided along the width of the slab. Thus, depending on the shear demand, the sizes of shear connector could be arrived.

The details of shear connector used for experimental and numerical investigations are shown in Table A1. The shear connectors are welded to the chord reinforcement in order to hold the shear connectors in position while pouring the concrete for bottom slab. This chord reinforcement will also contribute to the moment of resistance, hence for composite slabs tested in the experiments the shear demand is evaluated considering the chord reinforcements also. For numerical investigations, the chord reinforcements are not provided and shear demand is evaluated accordingly. For numerical investigations one truss type shear connector is provided at the centre for all the composite slabs and diameter of the shear connectors are calculated accordingly and hence, diameter of the bars is expressed in terms of decimal numbers. The area of shear reinforcements is provided exactly to meet the shear demand. For composite slab with 90° truss type shear connectors, the legs are placed at the spacing of 100 mm and the area of shear connectors are calculated accordingly and hence, the diameter is large when compared to other shear connectors.

Acknowledgements

Authors gratefully acknowledge the Grant-in-aid provided by Department of Science and Technology, New Delhi, for the project entitled “Segmental Composite slabs for bridge decks.”

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Department of Science and Technology, New Delhi, India.

ORCID iD

A Kanchana Devi

References

AASHTO LRFD (2014) Bridge design specifications. American Association of State Highway and Transportation Officials, Washington, DC.

ACI 318 (2019) Building Code Requirements for Structural Concrete (ACI 318M-19) and Commentary. Reported by ACI Committee, 319.

Ahmad

Goh

Samsuddin

, et al. (2013) Structural behaviour of precast lightweight foamed concrete sandwich panel (PLFP) with double shear truss connectors under eccentric load: Preliminary result. In Zaki

Faizul

Darus

(eds) Advanced materials research, vol. 795. Trans Tech Publications Ltd, pp.414–418. Switzerland.

Amin Einea

Salmon

Fogarasi

, et al. (1991) State-of-the-art of precast concrete sandwich panels. PCI Journal 36(6): 78–98.

Amran

Ali

Rashid

, et al. (2016) Structural behavior of axially loaded precast foamed concrete sandwich panels. Construction and Building Materials 107: 307–320.

Badie

Baishya

Tadros

(1998) NUDECK-a new prestressed stay-in-place concrete panel for bridge decks. In:1998 transportation conference proceedings. Iowa State University, Iowa, USA.

Baran

(2015) Effects of cast-in-place concrete topping on flexural response of precast concrete hollow-core slabs. Engineering Structures 98: 109–117.

Benayoune

Samad

Trikha

, et al. (2008) Flexural behaviour of pre-cast concrete sandwich composite panel–experimental and theoretical investigations. Construction and Building Materials 22(4): 580–592.

Bush

Stine

(1994) Flexural behavior of composite precast concrete sandwich panels with continuous truss connectors. PCI Journal 39(2): 112–121.

10.

Carbonari

Cavalaro

SHP

Cansario

, et al. (2012) Flexural behaviour of light-weight sandwich panels composed by concrete and EPS. Construction and Building Materials 35: 792–799.

11.

Choi

Kim

You

(2016) Effect of cyclic loading on composite behavior of insulated concrete sandwich wall panels with GFRP shear connectors. Composites Part B: Engineering 96: 7–19.

12.

Dias-da-Costa

Alfaiate

Júlio

ENBS

(2012) FE modeling of the interfacial behaviour of composite concrete members. Construction and Building Materials 26: 233–243.

13.

Dowell

Auer

(2011) Flexure-shear behavior of precast concrete deck panels with cast-in-place concrete topping. PCI Journal 56(3): 60–74.

14.

Einea

Salmon

Tadros

, et al. (1994) A new structurally and thermally efficient precast sandwich panel system. PCI Journal 39(4): 90–101.

15.

Fernando

PLN

Jayasinghe

MTR

Jayasinghe

(2017) Structural feasibility of Expanded Polystyrene (EPS) based lightweight concrete sandwich wall panels. Construction and Building Materials 139: 45–51.

16.

Frankl

Lucier

Hassan

, et al. (2011) Behavior of precast, prestressed concrete sandwich wall panels reinforced with CFRP shear grid. PCI Journal 56(2): 42–54.

17.

Gara

Ragni

Roia

, et al. (2012) Experimental behaviour and numerical analysis of floor sandwich panels. Engineering Structures 36: 258–269.

18.

Henin

Morcous

Tadros

(2014) Precast/prestressed concrete sandwich panels for thermally efficient floor/roof applications. Practice Periodical on Structural Design and Construction 19(3): 04014013.

19.

Hopkins

Chen

Yossef

(2017) Static and dynamic analyses of insulated concrete sandwich panels using a unified non-linear finite element model. Engineering Structures 132: 249–259.

20.

Huang

Jiang

Chong

, et al. (2020) Experimental study on precast concrete sandwich panel with cross-shaped GFRP connectors. Magazine of Concrete Research 72(3): 149–162.

21.

Ibrahim

Elliott

Abdullah

, et al. (2016) Experimental study on the shear behaviour of precast concrete hollow core slabs with concrete topping. Engineering Structures 125: 80–90.

22.

Jiang

Guo

Liu

(2018) Composite behavior of sandwich panels with W-shaped SGFRP connectors. KSCE Journal of Civil Engineering 22(5): 1889–1899.

23.

Kinnane

West

Hegarty

(2020) Structural shear performance of insulated precast concrete sandwich panels with steel plate connectors. Engineering Structures 215: 110691.

24.

Lee

Pessiki

(2008) Experimental evaluation of precast, prestressed concrete, three-wythe sandwich wall panels. PCI Journal 53(2): 95–115.

25.

Lee

Yoon

Kang

THK

(2016) Combined half precast concrete slab and post-tensioned slab topping system for basement parking structures. Journal of Structural Integrity and Maintenance 1(1): 1–9.

26.

Mansour

Bakar

Ibrahim

, et al. (2015) Flexural performance of a precast concrete slab with steel fiber concrete topping. Construction and Building Materials 75: 112–120.

27.

Mohamed

Thamboo

Jeyakaran

(2020) Experimental and numerical assessment of the flexural behaviour of semi-precast-reinforced concrete slabs. Advances in Structural Engineering 23(9): 1865–1879.

28.

Naito

Hoemann

Beacraft

, et al. (2012) Performance and characterization of shear ties for use in insulated precast concrete sandwich wall panels. Journal of Structural Engineering 138(1): 52–61.

29.

Peterman

Ramirez

(1998) Behavior and strength of bridges with full-span prestressed concrete form panels. PCI Journal 43(2): 80–91.

30.

Salmon

Einea

Tadros

, et al. (1997) Full scale testing of precast concrete sandwich panels. Structural Journal 94(4): 354–362.

31.

Teixeira

Fam

(2017) Fatigue behavior of partially composite-insulated concrete sandwich walls. ACI Structural Journal 114(1): 125–136.

32.

Thanoon

Yardim

Jaafar

, et al. (2010) Structural behaviour of ferrocement–brick composite floor slab panel. Construction and Building Materials 24(11): 2224–2230.

33.

Tomlinson

Fam

(2015) Flexural behavior of precast concrete sandwich wall panels with basalt FRP and steel reinforcement. PCI Journal 60(6): 51–71.

34.

Yee

(2001) Social and environmental benefits of precast/prestressed concrete construction. PCI Journal 46(3): 14–19.

35.

Yee

Eng

PHD

(2001) Structural and economic benefits of precast/prestressed concrete construction. PCI Journal 46(4): 34–43.

36.

Yuksel

Gullu

Durgun

, et al. (2020) Z-type shear connector for interface of hollow-core slab and cast-in-place topping concrete. Engineering Structures 214: 110563.