Experimental and numerical study on seismic performance of steel reinforced recycled concrete frame structure under low-cyclic reversed loading

Abstract

This article mainly focused on the seismic performance of steel reinforced recycled concrete frame structure under low-cyclic reversed loading. To evaluate seismic performance of steel reinforced recycled concrete frame structure, a two-span three-storied steel reinforced recycled concrete frame was conducted at civil engineering laboratory of Xi’an University of Architecture and Technology. Experimental and numerical studies were implemented to investigate the crack status, failure modes, hysteresis loops, skeleton curves, energy dissipation capacity, load–displacement curves, P-Δ effect, and the influence of recycled concrete strength under low-cyclic reversed loading. Results indicate that the steel reinforced recycled concrete frame structure has good seismic behavior during test, and the spindle-shaped hysteresis loops illustrate that the frame has relatively high energy dissipation capacities. The design of steel reinforced recycled concrete frame satisfied the requirements of strong column weak beam, strong shear weak bending, and strong joint weak components. Finally, the simulated results obtained by OpenSees software agree well with the test, which verify the rationality and reliability of the proposed model. The conclusions of this article will be helpful for the design of steel reinforced recycled concrete structures in seismic regions.

Keywords

low-cyclic reversed loading OpenSees finite-element simulation recycled aggregate concrete seismic performance steel reinforced recycled concrete

Introduction

Recycling of waste concrete is beneficial and necessary from the viewpoint of environmental preservation and effective utilization of resources (Xiao et al., 2006). The development of concrete structure is resource consumption-dependent. Non-renewable resources were over exploited seriously due to rapid urbanization. Simultaneously, large amounts of demolition waste were produced every year. It is believed that over 260 million tons of solid waste were produced by demolishing old buildings, and over 140 million tons solid waste were generated with the construction of new buildings. These wastes had a significant impact on the environment and mostly contribute to landfill saturation. To alleviate these negative effects, it is necessary to recycle the waste concrete. It is an effective way of resources utilization and helpful for environmental preservation. The new method is a good way to solve this problem by using the recycled aggregate concrete (RAC), which is used to completely or partially replace the natural coarse aggregate (NCA) in concrete mixtures.

To promote the usage of RAC, the structural behavior of RAC should be investigated. A considerable number of researches have been done at national (Chinese) and abroad. Most of these studies focused on the mechanical properties of RAC, such as the elastic models of RAC structures, compressive strength, and bond strength. The stress–strain relationship was especially important among the above studies. Many investigations on stress–strain relationship have been performed worldwide. Günçan (1995) obtained the complete stress–strain curves of RAC with different replacement ratio of 0%, 30%, 50%, 70%, and 100%. He found that the values of compressive strength, elastic energy, and elastic modulus decrease with the increase of replacement ratio. Bairagi et al. (1993) investigated the stress–strain relationship and found that the stress–strain curves with different recycled concrete aggregate replacement ratio followed similar trends but the curvature of each curve progressively improved with the increase of replacement ratio. Some achievements have already been summarized by Nixon (1978), Hansen (1986), American Concrete Institute (ACI) Committee 555 (2002), Tabsh and Abdelfatah (2009), and Xiao et al. (2012a) that some mechanical properties of RAC may be lower than that of natural concrete but the RAC is still sufficient for some practical applications in construction industry. The experimental test designed by Paine et al. (2009) and Hao et al. (2010) indicated that the elastic modulus of RAC is about 60%–80% of ordinary concrete when natural aggregates were totally substituted by recycled aggregates (Xiao et al., 2011). In recent years, numerous investigations were also performed to explore the seismic performance of recycled reinforced concrete beams (Ivan et al., 2013; Maruyama et al., 2004; Xiao et al., 2012b), columns (Ajdukiewicz and Kliszczewicz, 2007; Choi and Yun, 2012; Konno et al., 1997), joints (Corinaldesi et al., 2011; Fathifazl et al., 2009), and frame structures (Xiao et al., 2006). Previous results in different literatures are consistent and meaningful that the failure patterns of RAC structures are similar to the ordinary concrete, while the bearing capacity is lower than that of ordinary concrete to some extent (Buck, 1977; Ravindrarajah and Tam, 1985).

Xiao et al. (2006) tested four 1:2-scaled reinforced recycled concrete frame specimens under low-cyclic lateral load with different replacement ratio for recycled concrete aggregate (i.e. 0, 30%, 50%, and 100%). It is concluded that the seismic behavior of frame declined, followed by the increase of replacement ratio. However, the positive effect of adding shape steel to reinforced recycled concrete frame has not been considered. The shape steel can restrain the core concrete and improve the deformation and bearing capacity of the frame. Taking the good seismic performance and high bearing capacity of shape steel into account, combined with the energy saving and environmental protection of recycled concrete, a new steel reinforced recycled concrete (SRRC) structure was put forward in this article.

In this study, experiments were performed to provide a comprehensive understanding and analytical evaluation on the seismic performance of SRRC frame under low-cyclic reversed loading with constant vertical actions. The article particularly focused on the analysis of crack status, failure modes, hysteresis loops, skeleton curves, energy dissipation capacity, and P-Δ effect. Numerical simulations were performed by OpenSees finite-element software to study the mechanism of SRRC frame at last. The results presented in this article are significant to popularize the application of SRRC frame in practice.

Experimental investigations

Material properties

In this test, $\underline{ϕ}$ 8 ribbed reinforcing bars were used for the longitudinal reinforcements in beams and exterior columns. $\underline{ϕ}$ 12 ribbed reinforcing bars were used for the longitudinal reinforcements in interior columns. ϕ6 smooth steel bars were used for the transverse stirrups. The elevation view and dimensions of cross section were presented in Figure 1. As shown in Figure 2, the shape steel frame was prefabricated in factory, No.16 I-shaped steel of Q235B were adopted in columns, No.10 I-shaped steel were adopted in beams of top layers, while the other beams adopt No.12 I-shaped steel. Table 1 lists the mechanical properties of I-shaped steel and steel bars used in this experiment, in which f_y, f_u, ε_y, and E_s stands for yield strength, ultimate strength, yield strain, and elastic modulus, respectively.

Figure 1.

Dimensions of SRRC test specimen: (a) elevation view of SRRC frame, (b) 1-1 cross section, (c) 2-2(3-3) cross section, (d) 4-4 cross section, and (e) 5-5 cross section.

Figure 2.

Front view of shape steel frame: (a) joints of shape steel frame and (b) shape steel skeleton.

Table 1.

Mechanical properties of steel bars and I-shaped steel.

Steel and steel bars		f _y (MPa)	f _u (MPa)	ε _y (10⁻⁶)	E _s (10³ MPa)
Steel bars	ϕ6	398.4	518.6	1677	237.5
	$\underline{ϕ}$ 8	433.9	664.2	2187	199.0
	$\underline{ϕ}$ 12	490.0	662.7	2290	214.7
Shape steel	No.10 I-steel	312.9	441.3	1560	204.0
	No.12 I-steel	300.8	438.2	1482	203.0
	No.16 I-steel	283.4	440.7	1402	202.0

The recycled concrete aggregate used in this investigation was provided by Xi’an Hongcheng Building Materials Co., Ltd. As shown in Figure 3, 4 –8 mm diameter recycled concrete aggregate can be obtained after the demolished prefabricated concrete slabs were crushed and screened. Considering the water absorption capacity of recycled concrete aggregate, the recycled concrete aggregate was presoaked in water before mixing. The content of cement, recycled concrete aggregate, sand, water, and water reducer are 488, 1158, 527, 205, and 5 kg/m³, respectively. Three 150 mm × 150 mm × 150 mm concrete cube was reserved and tested after casting specimen, and the designed 28-day cube compressive strength was approximately 52 MPa.

Figure 3.

Recycled coarse aggregates: (a) screening of recycled concrete aggregate and (b) drying of recycled concrete aggregate.

Figure 4.

Test setup and loading procedure: (a) schematic diagram of test devices, (b) completed SRRC frame structure, and (c) loading procedure of test specimen.

Figure 5.

Cracks appeared at elastic stage: (a) cracks at the end of column, (b) beam end of first story, and (c) cracks at beam end (west).

Figure 6.

Cracks appeared at plastic stage: (a) inclined cracks of joint, (b) the cover concrete spalled, and (c) stirrups exposed.

Figure 7.

Failure modes of SRRC frame structure: (a) ends of side column (west), (b) beam ends of second story, (c) beam ends of second story, (d) beam ends of first story, (e) beam ends of second story, and (f) middle joint of first story.

Figure 8.

Subsequence of plastic hinges of SRRC frame: (a) loading at positive direction and (b) loading at negative direction.

Description of the specimen

A two-span three-storied SRRC frame structure with 100% replacement ratio of recycled concrete aggregate was designed and constructed to investigate the seismic performance under low-cyclic reversed loading. The SRRC frame is at 1:2.5 scale which is confined by the laboratory equipment capabilities. Detail dimensions of test specimen are illustrated as Figure 1: the frame height is 4900 mm, and both beam spans are 2400 mm. The column cross section is 200 × 240 mm, while the beam cross section is 140 × 210 mm. The concrete cover of both beams and columns is 10 mm. The specific procedure of casting SRRC specimen is shown as follows: manufacture of shape steel → weld shape steel → bind foundation beam reinforcement → bind frame steel bars → install foundation template → fix the steel-reinforcement frame → pour concrete of foundation beam → install frame template → cast the whole frame.

Loading procedure

Figure 4 showed the test setup and loading procedure of SRRC specimen. The axial loads, which represented the dead and live loads of each floor, were 400 kN for external columns and 800 kN for middle column. After the vertical loads applied, the cyclic lateral loads will be applied on the lateral load points of SRRC frame by using a 1000-kN electro-hydraulic servo actuator.

The low-cyclic reversed loading procedure involves two stages:

Load control stage: In this stage, three electro-hydraulic servo actuators were applied to beam centroid of each story to provide 10:7:4 inverted triangular load from top to bottom. The increment of cyclic load reversals (loading along positive direction was performed at first) was 4 kN until obvious stiffness degradation could be observed in hysteresis curves of SRRC frame. At this time, the reinforcement steel bars have already been yielded, and the corresponding lateral drifts of top beam ends were defined as Δ_y.

Displacement control stage: After the longitudinal steel bars yielded, the loading procedure was controlled by displacement. The increment of displacement is denoted as the yield displacement Δ_y, and each displacement level was loaded three cycles until the lateral load of SRRC frame drops sharply to 85% of peak value. All displacement levels in successive sets of cycles here, namely, +Δ_y, −Δ_y, +2Δ_y, −2Δ_y, +3Δ_y, −3Δ_y, …, were multiples of Δ_y.

To detect the lateral displacement of test specimen, six linear variable displacement transducers (LVDTs) were placed along the external column. Two LVDTs were also installed to measure the horizontal and vertical displacement of foundation beam.

Results and analysis

Crack status and failure modes

To describe the test observations clearly, the pushing direction was denoted as positive (+) and the pulling direction was denoted as negative (−). Based on the failure characteristics of SRRC frame, the failure process was divided into three stages, that is, elastic stage, plastic stage, and failure stage. The detail failure process of SRRC test specimen can be summarized as following:

In the early stage of loading, the SRRC frame was at elastic stage and no cracks occurred on the surface. The strains of longitudinal steel bars were small at this time. As the lateral load added to 20 kN, some minor cracks initiate at the ends of beams and columns. With the increase of lateral load, the numbers of cracks gradually increased and the cracks that already existed extended longer and wider. When the lateral load reached 24 kN, some cracks even extended to the whole section of beams and with a maximum width of 1 mm. Some typical crack patterns in elastic stage were depicted in Figure 5.

As the reinforcement steel bars yielded and the corresponding displacement reached to Δ_y, the loading procedure would be controlled by displacement. During the reversed cyclic loading of ±80 mm, the previous cracks continued to develop slowly, and some cracks penetrated across the full depth of beam and column sections. As the lateral displacement continuously increased (±120 mm), horizontal and inclined cracks could be observed at the core of beam–column joints. Some cover concrete was spalled at the end of beam and column. Meanwhile, concrete was crushed at the west of column ends followed by the bucking of longitudinal steel bars. With the reversed cyclic displacement added to ±140 mm, the concrete was crushed more seriously and the phenomenon of concrete spalling was more obvious. The stirrups of middle column were exposed. Even the SRRC frame was damaged seriously this time; the test specimen was not collapsed at once, which indicated that the frame structure still has the ability to resist limited lateral displacement. Figure 6 showed the crack modes of different regions of SRRC frame in plastic stage.

In the failure stage of test specimen, the lateral displacement finally reached ±180 mm during the process of low-cyclic reversed loading, large area concrete of beams and columns spalled off and crushed, the longitudinal steel bars were exposed, while diagonal cracks appeared on the surfaces of beam-column joints. The final typical failure modes of test specimen are described in Figure 7.

The subsequences and positions of plastic hinges of SRRC test frame are presented in Figure 8, where No.1–No.15 stands for the first and last hinges formed, respectively. The first plastic hinge occurred at the beam end (east) of second story in positive direction, while in negative direction, plastic hinge occurred at the beam end (east) of first story. With the increase of lateral load, many flexural cracks appeared at each end of beam and the plastic hinges developed at the beam end of first and second story. The damage degree of beam end was the most serious.

For the columns, damage concentrated on the bottom ends, and the damage degree was not as serious as beams. Plastic hinge first formed at the bottom of middle column (first story), then at the external columns (first story), which were mainly due to the larger vertical loading at the top of middle column and the moment transferred from beams at both sides. In the upper stories, there was no plastic hinge formed at column end.

The damage degree of beam-column joints was slight compared with beam and column as shown in Figure 7(f). The damage of joints usually occurred at last. In general, the SRRC test specimen exhibited good seismic performance and failed in a partial-beam sideway mechanisms. In this failure mechanism, plastic hinges formed at beam ends first, while the whole frame finally failed because of formation of plastic hinges at column bases, which is the most expected failure mechanism by the designer. The failure mechanism preformed in this article was in conformance with the seismic design principle of “strong column weak beam.” The failure process of SRRC frame presented here is consistent with the principle of Eurocode 8. The principle of Eurocode 8 (2005) is to reduce earthquake damage, protect people’s life, and ensure that important buildings are still available after earthquake.

Hysteresis curves

For the seismic design of structures or structural members, hysteresis curves are an important parameter because they reveal the mutual relationship between lateral cycle loads and the corresponding displacements during earthquakes (Ma et al., 2015). The hysteresis curves of SRRC frame are presented in Figure 9; seen from Figure 9, some conclusions can be drawn as follows:

On the whole, the hysteresis curves are approximately straight lines prior to cracking in the early stage of loading, which indicates that the SRRC frame is still at elastic stage and there is little damage accumulation at this state. Besides, the area of hysteresis curves is narrow and small at elastic state. With the increase of lateral loads, the slopes of hysteresis curves begin to decline gradually as a few cracks emerged at beam ends. It is suggested that the SRRC frame is in elastic–plastic state until an obvious stiffness degradation can be observed during the cyclic loading.

After the SRRC frame yielded, the unloading curve cannot return along the same route due to residual deformation. With damage accumulated on test specimen, the peak load of the last cycle is lower than that of the early cycles, indicating that the bearing capacity and lateral stiffness decreased gradually.

All hysteresis curves of SRRC frame showed spindle shape and no obvious pinch effects exhibited in hysteresis curves. This indicates that the SRRC frame has a relatively significant energy dissipation capacity.

Figure 9.

Story shear force-displacement hysteresis curve of SRRC frame: (a) hysteresis curve of whole model, (b) hysteresis curve at bottom story, (c) hysteresis curve of middle story, and (d) hysteresis curve of top story.

Skeleton curves

Skeleton curve is an important index to reflect the mutual relationship between the peak loads and corresponding displacement during each loading stage. Skeleton curves can be obtained by connecting the peak points of hysteresis curves. The skeleton curves of SRRC frame structure (skeleton curve of whole model; skeleton curve of each story) used in this article are shown in Figure 10. Based on the data from Figure 10, conclusions can be drawn as follows:

Totally three stages—elastic stage, plastic stage, and failure stage—are experienced by SRRC frame from the beginning of loading to failure. In the elastic stage, the skeleton curve is almost a straight line. The stiffness of test specimen is basically unchanged, and the lateral load is linearly increased.

After the frame cracks, the skeleton curves became nonlinear, which suggests that the SRRC frame was in plastic state and the stiffness would decrease gradually. With the increase of lateral loads, more plastic hinges appeared at beam ends and column ends, and the stiffness degradation became more obvious. When the lateral loads reached their peak point, the descending segment of skeleton curves was relatively flat, indicating that the SRRC test specimen had good ductility and deformation capacity.

Figure 10.

Skeleton curves of SRRC frame structure: (a) skeleton curve of whole model and (b) skeleton curves of each story.

Energy dissipation capacity

During the investigation on seismic performance of frame structure, the equivalent viscous damping coefficient h_e is adopted to quantify the seismic energy absorption capacity of SRRC frame. The equivalent viscous damping coefficient h_e is calculated by the area surrounded by hysteresis curve as shown in equation (1)

h_{e} = \frac{1}{2 π} \cdot \frac{S_{(ABCDA)}}{S_{(OBE + ODF)}}

(1)

As presented in Figure 11, S_(ABCDA) represents the area surrounded by the hysteresis curve ABCDA, $S_{OBE + ODF}$ represents the area surrounded by the triangles of OBE and ODF.

Figure 11.

Calculation diagram of equivalent viscous damping coefficient h_e.

The equivalent viscous damping coefficient h_e of SRRC frame is obtained by equation (1), in which h_ey = 0.10 represents the yield point, h_ep = 0.18 represents the peak point, and h_eu = 0.28 represents the ultimate point. Literature (Sun et al., 2006) discussed the seismic performance of RAC frame and presented the equivalent viscous damping coefficient h_e. Conclusions can be drawn out that the h_e is between 0.16 and 0.18 at the peak point and 0.22–0.30 at the ultimate point. It can be concluded that the SRRC frame structure has a better energy dissipation capacity compared with the RAC frame.

Numerical simulation

Unit selection and section partition

A numerical simulation was conducted to provide further insights about the hysteretic capability of SRRC frame structure under cyclic loading. In this article, OpenSees (Open system for Earthquake Engineering Simulation) finite-element program was adopted to make a further research on the influence of other design parameters on seismic behavior of SRRC frame. Displacement-Based Beam-Column Element was adopted to simulate all beams and columns in the finite-element model. Displacement of rod element was calculated by the corresponding node displacement; then the deformation can be also calculated according to the displacement interpolation function. Corresponding section resistance and tangent stiffness matrix can be obtained through the sectional restoring force relationship. Finally, the stiffness and resistance matrix of the whole unit can be calculated by Gauss–Legendre integral. This element is a distributed plastic model which allows the stiffness change along the section.

In order to reduce the analysis error and ensure the accuracy of analysis results, the columns of the frame were divided into two units, the beams were divided into four units, and five integral points were set in each unit. More details about the finite-element model were shown in Figure 12. Considering the P-Δ effect and the additional bending moment caused by the lateral displacement, P-Delta Transformation and Linear Transformation method were adopted to transform the column and beam element from local coordinate to overall coordinate system.

Figure 12.

Unit selection and section partition of frame structure: (a) partition of the frame, (b) setting of the integral point, and (c) division of fiber section.

As shown in Figure 12(c), based on the different confinement degrees of stirrups and steel to recycled concrete, sections of SRRC beams and columns were divided into three areas: (1) strong confined regions, (2) weak confined regions, and (3) no confined regions. Each recycled concrete area was divided into 5mm × 5mm recycled concrete fiber, then the corresponding material constitutive relationship was given to the fiber to simulate the hysteretic behavior of SRRC frame under cyclic load.

Constitutive model

1. Constitutive model of recycled concrete

Concrete02 Material was chosen as the recycled concrete constitutive model in OpenSees program. This command was used to construct a uniaxial concrete material object with tensile strength and linear tension softening. The constitutive model of recycled concrete was shown in Figure 13.

Figure 13.

Constitutive model of recycled concrete: (a) Concrete02 material-material parameters and (b) stress–strain curve of recycled concrete.

The skeleton curve of recycled concrete under compression adopted Kent–Park model which was modified by Scott et al. (1982). The compression skeleton curve can be expressed by equations (2) to (4), as shown below

σ_{c} = K f'_{c} [(\frac{2 ε_{c}}{ε_{0}}) - {(\frac{ε_{c}}{ε_{0}})}^{2}] (ε \leq ε_{0})

(2)

σ_{c} = K f'_{c} [1 - Z_{m} (ε_{c} - ε_{0})] (ε_{0} < ε \leq ε_{u})

(3)

σ_{c} = 0.2 K f'_{c} (ε > ε_{u})

(4)

where

K = \frac{1 + ρ_{sv} f_{yv}}{{f'}_{c}}

(5)

and

Z_{m} = \frac{0.5}{((3 + 0.29 {f'}_{c}) / (145 {f'}_{c} - 1000)) + 0.75 ρ_{sv} \sqrt{(h_{c} / s_{h})} - 0.002 K}

(6)

Where K is the amplifying coefficient of concrete compressive strength caused by the constraint of stirrups and shape steel, $ε_{0}$ is peak strain of concrete, $ε_{u}$ is ultimate compression strain of concrete, Z_m is slope of strain-softening in descending stage, $f'_{c}$ is concrete compressive cylinder strength, $s_{h}$ is center-to-center spacing of hoop sets (mm), f_yv is yield strength of stirrups, $ρ_{sv}$ is volumetric percentage of stirrups, and h_c is width of concrete core measured to outside of peripheral stirrups (mm).

In terms of related literature, there are some differences of physical and mechanical properties between the recycled concrete and ordinary concrete. Therefore, the stress–strain relationship of ordinary concrete cannot be directly applied into recycled concrete materials. Xiao (Xiao, 2007; Xiao and Li, 2005) suggested that the stress–strain curves of recycled concrete were similar to those of the ordinary concrete, but the values at the characteristic points were different. In this article, the constitutive relationship of recycled concrete was determined by adjusting the peak stress, peak strain, ultimate stress and ultimate strain in the compression stress–strain curves of ordinary concrete.

2. Constitutive model of steel

The constitutive model of shape steel and steel bars selected Steel02 Material, which was originally proposed by Menegotto and Pinto and modified by Filippou and others (Filippou et al., 1983). This command was used to construct a uniaxial Giuffre–Menegotto–Pinto steel material object with isotropic strain hardening, and the Bauschinger effect was also estimated simultaneously. Figure 14 showed the details of the constitutive model of shape steel and steel bars.

Figure 14.

Constitutive model of shape steel and steel bars: (a) skeleton curve of shape steel, (b) stress–strain curve of shape steel, and (c) Menegotto–Pinto steel model.

The model proposed by Menegotto and Pinto can be expressed as follows

σ^{*} = b ε^{*} + \frac{(1 - b) ε^{*}}{{(1 + ε^{* R})}^{1 / R}}

(7)

where

ε^{*} = \frac{ε - ε_{r}}{ε_{0} - ε_{r}}

(8)

and

σ^{*} = \frac{σ - σ_{r}}{σ_{0} - σ_{r}}

(9)

Equation (7) represents a curved transition from a straight line asymptote with slope E₀ to another asymptote with slope E₁ (line (a) and (b), respectively, in Figure 14(c)). σ₀ and ε₀ are the stress and strain at the point where the two asymptotes of the branch under consideration meet (point B in Figure 14).

Boundary conditions and load definition

Loads were defined by using the pattern command in OpenSees software. Plain pattern was selected to apply vertical loads to corresponding nodes. The axial load was applied to the top of three columns after 10 steps and then kept as constant. The cyclic lateral loads were applied to the center of each layer beam. In this analysis, a lateral displacement cycle (positive and negative) was imposed at the corresponding nodes. The imposed displacements were applied by using a displacement-control integrator. The initial increment step of horizontal displacement was 1.0% of the maximum displacement step.

Fix command was used to construct homogeneous single-point boundary constraints. As presented in Figure 8, in which 0 and 1 represent the state of unconstrained and constrained, respectively, nodes 1, 2, and 3 were fully fixed in three directions (X, Y, Z) and six degree-of-freedom, while the other nodes released the translation in X direction and rotation in Z direction.

Results of finite-element simulation

In this part, comparison is made between finite-element simulation and the results of test to verify the accuracy of simulation. After that, a further research about the influence of other parameters to the hysteretic behavior of SRRC frame has been carried out by OpenSees software, including the influence of axial compression ratio, the influence of recycled concrete strength, and the influence of shape steel strength.

1. Load–displacement curves

Both simulation and test results of load-displacement hysteretic curves are shown in Figure 15. It can be seen that the simulation results anastomose the test ones. Under positive loading, the peak displacement value of simulation is slightly less than the test result. While under negative loading, after reaching the peak load, the bearing capacity of frame is less than the simulation result. The main reasons of difference between simulation and test are probably as follows:

There is a preload effect on frame beam which is carried out by electro-hydraulic servo, while simulation test is under ideal conditions and the preloading effect is not considered.

In the simulation model, fiber element model is adopted for beams and columns of SRRC frame, neglecting the bond-slip effect between shape steel and recycled concrete.

Figure 15.

Load–displacement curves of simulation and test: (a) hysteresis curves and (b) skeleton curves.

Skeleton curves reflect the important mechanical properties, such as crack load, yield load, peak load, and failure load. Both simulation and test skeleton curves are presented in Figure 15(b). It is clear that the simulation results of skeleton curves are similar to the test ones. This implies that the simulation has a high accuracy.

2. P-Δ effect on hysteretic performance of SRRC frame

Under the interaction of lateral and vertical loads, the horizontal displacement and vertical load will have an additional moment on the structure, called P-Δ effect. Technical specification for concrete structures of tall building stipulates that the adverse effect of this second-order effect on buildings must be taken into consideration under some conditions. In OpenSees finite-element program, when the unit is converted from the local coordinate to the whole coordinate, appropriate parameters can be selected to consider P-Δ effect. In this simulation, Linear Transformation is chosen for beam unit during coordinate transformation, and the Linear Transformation is selected without considering the P-Δ effect. Considering the P-Δ effect, P-Delta Transformation is chosen for column unit.

As presented in Figure 16, the hysteresis and skeleton curves in which the two cases considering P-Δ effect or not are plotted with two different lines. It can be seen that the bearing capacity of the model considering P-Δ effect is lower than that without considering P-Δ effect. Before the frame yielded, there is little difference between the two load–displacement curves for the two cases. After the frame yielded, the declined rate of bearing capacity becomes slow as the horizontal displacement increased, if P-Δ effect is not considered.

Figure 16.

Load–displacement curves considering and no considering P-Δ effect: (a) hysteresis curves and (b) skeleton curves.

Table 2 shows the specific values of the characteristic points in hysteresis curves of two cases. If P-Δ effect is not considered, the positive and negative loads decrease to 93.0% and 94.0% of the peak load, respectively, when the frame reaches its maximum displacement. While considering P-Δ effect, the positive and negative loads decrease to 72.0% and 73.0%. The peak bearing capacity in positive and negative directions increase by 12.1% and 10.8% compared with the situation without considering P-Δ effect. Results show that the ductility of structure develops more sufficiently when considering P-Δ effect.

Table 2.

P-Δ effect on hysteretic behavior of SRRC frame.

Category	Loading direction	Peak point		Limit point		P_u/P_m	h_e
		Displacement (mm)	P _m (kN)	Displacement (mm)	P _u (kN)	P_u/P_m	h_e
Category 1	Positive	59.21	314.94	180.03	225.94	0.72	0.351
Category 1	Negative	−52.41	−284.77	−179.01	−208.85	0.73	0.351
Category 2	Positive	80.62	353.11	180.03	329.39	0.93	0.245
Category 2	Negative	−65.48	−315.47	−179.01	−295.48	0.94	0.245
Differ	Positive	36.20	12.10	0.00	45.80	–	–

SRRC: steel reinforced recycled concrete.

Category 1—considering P-Δ effect; Category 2—without considering P-Δ effect; Differ = (Category 2 − Category 1)/Category 1 × 100%.

3. Influence of recycled concrete strength on hysteresis behavior

The influences of different recycled concrete strength on hysteresis behavior of SRRC frame have been presented in Figure 17. The improvement in recycled concrete strength brings about a little increase in the lateral bearing capacity of the SRRC frame. Load–displacement curves are linear before cracks appear on the beam ends. This indicates that the SRRC frame is in elastic range. The hysteresis loops of test specimen become nonlinear after cracks form on beam ends, and the reinforcement of beams and columns yield as the applied horizontal loads increase. Meanwhile, the stiffness decreases, residual deformation increases because of the cumulative damage in concrete, and the hysteresis areas increase correspondingly. Hysteresis curves of SRRC frame in three cycles are quite similar with closer maximum load at the same displacement stage. Also, the maximum loads of first cycle are obviously larger than those of the latter two cycles. Conclusions can also be drawn from Figure 17, and the differences between load capacity in pushing and pulling directions are significantly high. The lower negative load-carrying capacity of SRRC frame is primarily caused by the concrete cracking. The lateral loading is first loaded on the specimen in positive direction, and the concrete cracks may cause some damage in structure. Because of the damage that already existed, the test specimen would achieve lower load-carrying capacity in the negative direction. It can be seen from the skeleton curves in Figure 17, the slope of skeleton curves increases slowly with the increase of recycled concrete strength before yield.

Figure 17.

Load–displacement curves with different recycled concrete strength: (a) hysteresis curves and (b) skeleton curves.

Summary and conclusions

This article primarily focused on the seismic resistance capacity of SRRC frame structure under low-cyclic reversed loading. Based on the results and analysis in this article, major conclusions are summarized as follows:

Several main seismic performances, such as load, displacement, hysteresis curve, and skeleton curve, are obtained through the experimental study on a two-span three-storied SRRC frame under low-cyclic reversed loading. It reveals that the SRRC frame has good seismic behaviors and meets the structure design principle of “strong column weak beam” and “strong joint weak component.”

Under the same conditions, the bottom of middle column was first destroyed and formed a plastic hinge compared with external columns, though the middle column’s I-shaped steel flange had already been welded with 4 mm thickness steel plate on both sides. The damage degree of middle column was much more serious than that of the exterior columns. It is necessary to increase the dimension of cross section and steel configuration ratio of middle column to retard the yielding of the column bottom in design.

The hysteresis curves of SRRC test specimen showed spindle shape and no obvious pinch effects occurred in hysteresis curves, indicating that the SRRC structure has good seismic resistance and a relatively significant energy dissipation capacity.

The predicted results of SRRC frame under low-cyclic loading are reasonably close to the measured trends of simulation, which indicates that the existed analytical models and code-based procedures for conventional steel reinforced concrete frame can also be applied to SRRC frame structures.

The finite-element calculation results of SRRC frame under low-cyclic reversed loading corresponded well with the experimental results, which indicates that the observations of simulation are reliable and the OpenSees software is suitable for simulation of SRRC frame structure.

P-Δ effect has a great influence on the hysteresis behavior of SRRC frame structure. Considering P-Δ effect, the horizontal bearing capacity of SRRC frame is much lower than that of the case without considering P-Δ effect. Meanwhile, SRRC frame shows a better plastic development when considering P-Δ effect.

Outlook

From the point of economical view, there is no economic profit for producing recycled concrete aggregate. In China, for C30 natural concrete, the cost of cement, river sand, gravel, and water reducing agent is 121.9, 40.1, 57.75, and 19.92 RMB per ton, respectively. For C30 recycled concrete, the cost of cement, river sand, recycled coarse aggregate, and water reducing agent is 130.9, 39.9, 39.18, and 21.4 RMB per ton, respectively (Yang et al., 2013). In terms of material cost only, the cost of C30 natural concrete and recycled concrete per cubic meter is almost the same. Nevertheless, the production process of recycled aggregate is very complicated. The recycled coarse aggregate has to be sorted, broken, screened, soaked, and cleaned before it is used. The huge cost of labor generated by the complicated manufacture processes is the main reason for high cost of producing recycled concrete. But the application of recycled concrete brings great environmental and social benefits. Therefore, the validity evaluation of structure cannot be simply measured by production cost. Especially for multi-story buildings, using shape steel can reduce the size of the component cross section. By this way, the using area and net height of each layer can be increased, which has great social benefits. In the long run, SRRC is a structure with great potential application in the future.

Footnotes

Acknowledgements

In addition, the author would like to acknowledge the kind help of Gang Wang, Manman Cui, and Boyu Ze. Also, the writer would like to thank the anonymous reviewers for their valuable comments and suggestions.

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: Financial supports from the National Natural Science Foundation of China (Grant No. 51608435) and Scientific Research Project of Shaanxi Province (Grant No.2016JQ5113) are greatly appreciated.

ORCID iD

Xin Zhang

References

Ajdukiewicz

Kliszczewicz

(2007) Comparative tests of beams and columns made of recycled aggregate concrete and natural aggregate concrete. Journal of Advanced Concrete Technology 5(2): 259–273.

American Concrete Institute (ACI) Committee 555 (2002) Removal and reuse of hardened concrete. ACI Materials Journal 99(3): 300–325.

Bairagi

Ravande

Pareek

(1993) Behaviour of concrete with different proportions of natural and recycled aggregates. Resources, Conservation and Recycling 9(1–2): 109–126.

Buck

(1977) Recycled concrete as a source of aggregate. Journal of the American Concrete Institute 74: 212–219.

Choi

Yun

(2012) Compressive behavior of reinforced concrete columns with recycled aggregate under uniaxial loading. Engineering Structures 41(3): 285–293.

Corinaldesi

Letelier

Moriconi

(2011) Behaviour of beam–column joints made of recycled-aggregate concrete under cyclic loading. Construction and Building Materials 25(4): 1877–1882.

Eurocode 8 (2005) Design of structures for earthquake resistance—part 1: general rules, seismic actions and rules for buildings. Available at: https://www.saiglobal.com/PDFTemp/Previews/OSH/IS/EN/2005/I.S.EN1998-1-2005.pdf

Fathifazl

Razaqpur

Isgor

et al . (2009) Flexural performance of steel-reinforced recycled concrete beams. ACI Structural Journal 106(6): 858–867.

Filippou

Popov

Bertero

(1983) Modeling of R/C joints under cyclic excitations. Journal of Structural Engineering 109(11): 2666–2684.

10.

Günçan

(1995) Using waste concrete as aggregate. Cement and Concrete Research 25(7): 1385–1390.

11.

Hansen

(1986) Recycled aggregates and recycled aggregate concrete second state-of-the-art report developments 1945–1985. Materials and Structures 19(3): 201–246.

12.

Hao

Liu

(2010) Study on complete stress-strain curves of recycled concrete. In: Proceeding of 2nd international conference on waste engineering and management, Bagneux, France: RILEM Publications SARL, 13–15 October, pp. 506–512.

13.

Ivan

Snežana

Zoran

et al . (2013) Flexural behavior of reinforced recycled aggregate concrete beams under short-term loading. Materials and Structures 46(6): 1045–1059.

14.

Konno

Sato

Kakuta

et al . (1997) The property of recycled concrete column encased by steel tube subjected to axial compression. Transactions of the Japan Concrete Institute 19: 231–238.

15.

Xue

Liu

et al . (2015) Cyclic loading tests and shear strength of steel reinforced recycled concrete short columns. Engineering Structures 92: 55–68.

16.

Maruyama

Sogo

Sogabe

et al . (2004) Flexural properties of reinforced recycled concrete beams. Materials and Structures 8(11): 525–535.

17.

Nixon

(1978) Recycled concrete as an aggregate for concrete—a review. Matériaux Et Construction 11(5): 371–378.

18.

Paine

Collery

Dhir

(2009) Strength and deformation characteristics of concrete containing coarse recycled and manufactured aggregates. In: Proceedings of the 11th international conference on non-conventional materials and technologies, University of Bath, Bath, 6–9 September.

19.

Ravindrarajah

Tam

(1985) Properties of concrete made with crushed concrete as coarse aggregate. Magazine of Concrete Research 37(130): 29–38.

20.

Scott

Park

Priestley

MJN

(1982) Stress-strain behavior of concrete by overlapping hoops at low and high strain rates. ACI Structural Journal 79(1): 13–27.

21.

Sun

Xiao

Zhou

et al . (2006) An experimental study on the seismic behavior of recycled concrete frames. China Civil Engineering Journal 39(5): 9–15.

22.

Tabsh

Abdelfatah

(2009) Influence of recycled concrete aggregates on strength properties of concrete. Construction and Building Materials 23(2): 1163–1167.

23.

Xiao

(2007) Experimental investigation on complete stress-strain curve of recycled concrete under uniaxial loading. Journal of Tongji University 35(11): 1445–1449.

24.

Xiao

(2005) Study on relationships between strength indexes of recycled concrete. Journal of Building Materials 8(2): 197–201.

25.

Xiao

Huang

Shen

(2012a) Seismic behavior of semi-precast column with recycled aggregate concrete. Construction and Building Materials 35(10): 988–1001.

26.

Xiao

Huang

Yang

et al . (2011) Mechanical properties of confined recycled aggregate concrete under axial compression. Construction and Building Materials 26(1): 591–603.

27.

Xiao

Fan

et al . (2012b) An overview of study on recycled aggregate concrete in china (1996–2011). Construction and Building Materials 31(6): 364–383.

28.

Xiao

Sun

Falkner

(2006) Seismic performance of frame structures with recycled aggregate concrete. Engineering Structures 28(1): 1–8.

29.

Yang

Yan

Zheng

(2013) Advantages and economic analysis of recycled concrete. Construction Economy 7(3): 85–88.