- Email: [email protected]
Experimental study on seismic performance of partially precast steel reinforced concrete columns
Engineering Structures 175 (2018) 63–75
Contents lists available at ScienceDirect
Engineering Structures journal homepage: www.elsevier.com/locate/engstruct
Experimental study on seismic performance of partially precast steel reinforced concrete columns
T
⁎
Yong Yang, Yicong Xue , Yunlong Yu, Fengqi Gao School of Civil Engineering, Xi’an University of Architecture & Technology, Xi’an, Shaanxi 710055, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Steel reinforced concrete columns Precast structures Cyclic loading tests Reactive powder concrete Seismic performance
This paper aims to develop two kinds of innovative precast steel reinforced concrete columns, which are partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns. Both the two kinds of composite columns have precast reactive powder concrete (RPC) shells, and the PPSRC column has a cast-in-place column core. In this paper, a series of cyclic loading tests on 10 column specimens subjected to combined static axial loading and cyclic lateral loading were carried out to explore their seismic performances. All specimens were evaluated by the failure modes, hysteresis characteristics, strength and stiffness degradation, energy dissipation capacity and ductility. The effects of section shape, stirrup spacing, axial compression and concrete strength of cast-in-place inner-part were investigated in details. The experiment results indicated that the PPSRC columns exhibited more satisfactory seismic behavior than the HPSRC columns in terms of hysteretic behavior, strength degradation, ductility and energy dissipation, while their bearing capacities were almost identical under low axial compression. Steel fibers could effectively prevent the cracked concrete from spalling, therefore, the encased steel shape was efficiently confined by the surrounding concrete during the entire test process. Higher stirrup ratio and lower axial compression of the column leaded to more satisfactory energy dissipation capacity, stiffness degradation and higher ductility. Based on the plastic stress theory, the seismic bending moment capacity analysis was conducted, and the results obtained form the formulas agreed well with those from the experiments.
1. Introduction During the past few decades, steel-concrete composite structural systems have been widely used in constructing new buildings and retrofitting existing structures all over the world. This system combines the rigidity and formability of reinforced concrete members with the strength and convenience of construction associated with structural steel to produce an economic and robust structure [1–4]. As an important part of steel-concrete composite structures, steel reinforced concrete structures (SRCs), have been widely applied in marine, largespan and high-rise structures due to the high bearing capacity, great stiffness, great durability and outstanding ductility performance[5–7]. Nevertheless, cast-in-place SRC structures usually involve the construction procedures both of steel structures and concrete structures, therefore, the application of cast-in-place SRC structures has been limited in conventional buildings due to complex construction process. At the same period, applications of precast structures have increased because of the efficiency and high quality in construction, and it can also provide an advisable solution to the problem of wasted resources
⁎
for on-site activities [8–10]. With the aim to facilitate the construction process of SRC structures, some researchers have suggested the combination of SRC members and precast concrete members. As a good example, the Fujita steel and reinforced precast concrete system has been adopted in Japan [10]. It indicates that the application of entire precast SRC members seems to be a solution to the problem aforementioned, however, the heavy deadweight and the vulnerable joint area could become new problems to the structure engineers. On the other hand, the builders must employ special large vehicles and cranes to lift the entire precast SRC members, which would directly increase the building cost. As an alternative method to reduce the deadweight of precast SRC members, Hong et al. [11–14] proposed a new kind of partially precast steel reinforced concrete system and named it as MHS (modularized hybrid system). However, the optimal combination mode, section shape and the structural integrity in the partially precast SRC members must be further investigated for better structural capacity and constructability. Furthermore, the type and strength of concrete in the precast part and cast-in-place part should be altered to match different
Corresponding author at: No. 13, Yanta Road, Xi'an, Shaanxi 710055, China. E-mail address: [email protected] (Y. Xue).
https://doi.org/10.1016/j.engstruct.2018.08.027 Received 20 February 2018; Received in revised form 24 July 2018; Accepted 11 August 2018 0141-0296/ © 2018 Elsevier Ltd. All rights reserved.
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Fig. 1. Schematic diagram of the PPSRC column and HPSRC column.
building with the limited loss of rigidity. Meanwhile, the dimensions of the PPSRC and HPSRC columns can be unified along the building height to facilitate the conventional variable section design. As an essential issue in composite members, the connecting and bonding behaviors should be emphasized [20,21]. Because there are two kinds of concrete in the same cross-section of PPSRC columns, it is necessary to examine the integrity of PPSRC columns under seismic loads, therefore, a series of cyclic loading tests on 10 column specimens subjected to combined static axial loading and cyclic lateral loading were carried out. The crack patterns, failure modes, hysteretic loops, skeleton curves, strength and stiffness degradation, energy dissipation capacity and ductility of the specimens were critically examined. The effects of section shape, stirrup spacing, axial compression and concrete strength of cast-in-place inner-part on the seismic performance of the specimens were also explored in details. Based on the experimental observations and test results, an analytic model for calculating seismic bending moment capacity is also presented later in this paper.
purposes. With the aim to promote the applications of SRC structures and solve the problems mentioned above, an innovative partially precast steel reinforced concrete structure system (PPSRC) was proposed by the authors, as illustrated in Fig. 1. The PPSRC frame structure is composed of PPSRC beams, PPSRC columns and PPSRC joint, and the mechanical performance of PPSRC beams has been thoroughly investigated in reference paper [15,16]. In this paper, two innovative partially precast SRC columns, which are partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns, are presented here. In the PPSRC column, the precast outerpart, which is composed of a cruciform steel shape, high-performance concrete, continuous rectangle stirrups and longitudinal rebar, can be prefabricated in factory, and the cast-in-place inner-part can be cast by conventional strength concrete, lightweight aggregate concrete or recycled aggregate concrete on the construction site after the outer-part erected. For the HPSRC columns, the outer-part is the same as that of PPSRC columns, but the column core is kept hollow to further reduce the deadweight or can be filled with heat insulating material to enhance the anti-fire performance. Meanwhile, the HPSRC columns can be regarded as formwork of the inner-part of PPSRC columns. With the aim to further improve the performance of precast outerpart, the reactive powder concrete (RPC) is applied here. As a kind of ultra-high performance concrete, reactive powder concrete has been widely studied due to some favorable characteristics, such as high durable and mechanical performance [17,18]. Especially, the steel fibers in RPC can significantly show down the crack growth in concrete [19]. Continuous rectangle stirrups are also adopted here to avoid the invalidation of stirrup hooks in severe earthquakes. For the inner-part, it can be cast by conventional strength concrete with the beam core of PPSRC beams and adjacent slabs at the same time to ensure strutural integrity of PPSRC structures and to save the expensive high-performance concrete. As shown in Fig. 1, the PPSRC columns and HPSRC columns can be composite in a high-rise building vertically. For the columns in the ground floor, the PPSRC columns with inner concrete will meet the needs of great axial capacity and sufficient ductility under proper design, and for the columns in upper floors, the use of HPSRC columns with hollow core can efficiently reduce the deadweight of the entire
2. Experimental program 2.1. Test specimens Ten column specimens were designed and constructed, including six PPSRC columns, and four HPSRC columns. The key parameters are summarized in Table 1. These ten specimens were similar in cross section shape and dimensions, except for the HPSRC columns with hollow cores. As shown in Fig. 2, the columns were 300 mm × 300 mm square in cross section, with a height of 900 mm from the point of lateral loading to the top of the stub. The specimens were designed to represent the structural columns of lower stories of high-rise structures and were designed at one second of full scale to match the capacity of the specimen with that of the test device. A reinforced concrete stub footing with a cross section of 500 mm by 550 mm was cast together with the specimen, representing a relative rigid foundation. The steel shape in all specimens was cruciform and weld by two HN175 × 90 × 5 × 8 steel of Q235 grade according to the Chinese standards. The total height and the width of the steel shape were 175 mm and 90 mm, respectively, and the thickness of the web and flange were 5 mm and 8 mm, respectively. As shown in Fig. 2, the steel 64
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Table 1 Matrices of test specimens. ID
λ
Section shape
N (kN)
Concrete
Reinforcements
Steel shape
fc,out (MPa)
fc,in (MPa)
ρsv
ρs
Type
ρss
PPSRC-1 PPSRC-2 PPSRC-3 PPSRC-4 PPSRC-5 PPSRC-6
3.0 3.0 3.0 3.0 3.0 3.0
Solid Solid Solid Solid Solid Solid
1500 2000 2000 2000 2000 2000
96.8 96.8 96.8 96.8 96.8 96.8
24.3 24.3 47.6 24.3 24.3 96.8
1.26%(D8@65) 2.00%(D8@40) 1.26%(D8@65) 1.26%(D8@65) 0.68%(D8@120) 1.26%(D8@65)
3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18)
2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8)
5.0% 5.0% 5.0% 5.0% 5.0% 5.0%
HPSRC-1 HPSRC-2 HPSRC-3 HPSRC-4
3.0 3.0 3.0 3.0
Hollow Hollow Hollow Hollow
1500 2000 2000 2000
96.8 96.8 96.8 96.8
– – – –
1.26%(D8@65) 1.26%(D8@65) 0.68%(D8@120) 2.00%(D8@40)
3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18)
2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8)
5.0% 5.0% 5.0% 5.0%
Note: λ is the aspect ratio of the specimens; N is the constant axial load on the column top; fc,out is the compressive strength of the reactive powder concrete; fc,in is the compressive strength of the cast-in-place concrete; ρsv is the volumetric stirrup ratio; ρs is the longitudinal reinforcement ratio; ρss is the structural steel ratio.
in this step. In the first step, before the concrete cast, tear plates were assembled between the adjacent steel flanges by dot welding as the formwork of the outer-part. The effect of the thin tear plate on the cyclic performance of entire specimen can be ignored due to its weak stiffness, but the tear plates and high-strength bolts in the outer-part could play important roles to enhance the bonding performance between the precast concrete, cast-in-place concrete and steel shape. In the practical applications, reusable inflated rubber formwork can be applied in the outer-part to further prefabricate the outer-part efficiently.
flanges were drilled and high-strength bolts were then assembled at the bottom and top of the steel flange as shear connectors to enhance the bonding performance between the precast outer-part and cast-in-place inner-part. Because the head of the high-strength bolt was in the precast outer-part and the rear of the high-strength bolt was in the cast-in-place inner part, both the precast concrete and cast-in-place concrete could benefit from the shear connectors, as illustrated in Fig. 3(c). All the reinforcing rebar and stirrups of the specimens are illustrated in detail in Fig. 2. 2.2. Specimens fabrication
2.3. Materials All the PPSRC and HPSRC columns were constructed in two steps. As shown in Fig. 3, in the first step, the outer-part of the specimens was molded and precast. The outer part was composed of a cruciform steel shape, reactive powder concrete, continuous rectangle stirrups and longitudinal reinforcements, and the concrete in the outer-part was cast at the same height of the steel shape, as indicated in Fig. 3(b). The second step was started after the precast concrete hardened, and the inner-part, namely the column core, was cast by conventional concrete
The concrete in the outer-part is a kind of reactive powder concrete, which is designed as C100 grade. The mixed proportions for the cement-based matrix of the RPC used in this study are summarized in Table 2. The components of RPC include 52.5P.O. cement, silica fume, fly ash, mineral powder, quartz sand, river sand, water reducer, water and steel fiber. A 2% volume incorporation of steel fibers is used in the RPC and the mechanical properties of the steel fibers are shown in
Fig. 2. Details of specimens (Unit: mm). 65
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Continuous rectangle stirrup
(b) The outer part of specimen
(c) Shear connectors
Fig. 3. Configuration of the specimens.
Table 3. The average tested compressive strength of the RPC coupons is 96.8 MPa, and the average tested tensile strength was 6.83 MPa. The strength grades of concrete in the inner-part were designed at three different levels, and the tested compressive strengths at 28 days were 24.3 MPa, 47.6 MPa and 96.8 MPa, respectively. Table 4 summarizes the mechanical properties of steel rebar used in this study, where fy is the yield strength and fu is the ultimate strength.
Table 3 Properties of steel fiber. Type
Diameter (mm)
Length (mm)
Tensile strength (MPa)
Straight
0.2
13.0
2850
Table 4 Mechanical properties of steel reinforcement.
2.4. Test setup and instrumentations
Reinforcement
Type
Steel grade
fy (MPa)
fu (MPa)
The test setup and details of instrumentations are schematically shown in Fig. 4. The vertical compressive loading was applied by a 5000 kN hydraulic jack, and the cyclic lateral loading was applied on the load point by a 1000 kN electro-hydraulic servo machine after the axial compression force on the top of the specimens stabilized. The reaction devices on the ground and two steel beams on the footing stub of the specimens were employed to prevent the foundation of columns from sliding during the test. During the test process, the horizontal deflections at the loading point and the footing stub and the vertical deflections at the both sides of the plastic hinge area were measured using linear variable differential transformers (LVDTs). A considerable number of electrical-resistance strain gauges were arranged on the web and flanges of the steel shape and the reinforcing bars to monitor the strain response. A total of 14 strain gauges were adhered on the steel shape and 7 strain gauges for the reinforcement cage. The layouts of the strain gauges and LVDTs are illustrated in Fig. 5.
Longitudinal rebar Stirrup
D18 D8
HRB400 HPB300
443.5 393.2
598.0 562.3
Steel shape
HN175 × 90 × 5 × 8
Q235 Q235
311.7 272.5
438.3 520.0
Flange Web
3. Experimental results and analysis 3.1. Experimental observations The failure modes and damage patterns of the specimens are shown in Fig. 7 and the main test results are recorded in Table 5. (1) Based on the failure patterns of the 6 PPSRC columns and 4 HPSRC columns, two typical failure modes of the specimens could be observed, namely flexural failure and flexural-shear failure, as shown in Fig. 7, in which only the local pictures, about 1/2 height of the entire specimen, are presented for better illustration of the crack patterns and its propagation. The flexural failure was characterized that the transverse cracks initiated and developed after longitudinal rebar and steel flange yield, and the flexural-shear failure was defined as a combination with some diagonal cracks, and yet it still failed because of the propagation of main transverse cracks. Based on the above analysis, flexural failure was only found in specimen PPSRC-2 with sufficient stirrups and specimen PPSRC-6 with the highest-strength core. Slight or moderate shear failure characteristic of other specimens might cause by the lack of coarse aggregate in the reactive powder concrete, which should weaken the shear resistance of aggregate interlock. The similar conclusions could be found in other research about reinforced RPC or other high-performance concrete columns [17,18,22]. As shown in Fig. 7, abundant cracks appeared and propagated in the region where the distance from the column bottom was about 300 mm to 350 mm (6 grids to 7 grids, 50 mm per grid), indicating that the plastic length of all the specimens was approximate 350 mm, namely 1/3 height of the entire column. (2) As shown in Fig. 7(a)–(f), bonding cracks were captured on the
2.5. Test procedure At the beginning of the tests, an initial loading procedure was followed to make sure that there was no significant eccentricity by checking the data of the LVDTs. Then the predetermined axial compressive load was applied with the speed of 2 kN/s and the constant axial load was maintained during testing. A small lateral force was also applied several times at this stage in order to stabilize the test system, then the lateral force was cycled under a lateral displacement control mode. Six single cycles with peak drift ratios of Δ/L = 0.1%, 0.2%, 0.3%, 0.6%, 0.8%, 1.0% were initially applied. Cycles with peak drift ratios of Δ/L = 1.5%, 2.0%, 2.5%…were then applied, with the cycles repeated three times. Fig. 6 shows the load procedure of the cyclic lateral force. The test terminated when the lateral force fell by more than 25% of the maximum experienced force.
Table 2 Designed mix proportions of concrete matrix. 52.5P.O.Cement
Silica fume
Fly ash
Mineral powder
Quartz sand
River sand
Super plasticizer
W/B
Steel fiber
1
0.15
0.2
0.1
0.5
0.8
0.003
0.22
0.02
Note: the ratio of steel fiber is the volume ratio, and ratios of other components are weight ratios. 66
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Schematic diagram
(b) Photo Fig. 4. Test device.
observed sides (perpendicular to the loading direction) in some specimens. It can be attributed to the relatively small concrete cover thickness of the bolts on the steel flanges perpendicular to the loading direction, but the development of these bonding cracks seemed to affect the cyclic performance of the specimens limitedly. (3) Steel fibers in the specimens played an important role to prevent bottom concrete covers from spalling. Although lots of steel fibers failed because of pulling out from concrete, a closer observation to the cracks indicated that some fibers still bridged the two sides of a crack (Fig. 7(g)). Meanwhile, as shown in Fig. 7(h), some specimens were broken after the tests, and no obvious buckling was found both in longitudinal rebar and steel shape, which indicated that the steel reinforcements were well confined by the existing surrounding concrete. However, once the fibers were removed from high-strength concrete, significant spalling of bottom cover concrete and severe damage in the plastic region would cause the buckling of steel shape and longitudinal rebar [23].
Fig. 6. Loading protocol of lateral force.
in Fig. 8. For better comparison, the skeleton curves of the specimens are plotted with bold lines in Fig. 8. The occurrences of special events such as cracking of concrete, initial yielding of longitudinal rebar and steel flange and the ultimate condition (concrete crushing) are marked in the figures. When drifts were no greater than approximate 0.9%, all
3.2. Hysteretic loops and skeleton curves The measured lateral force versus top displacement hysteretic relationships and skeleton curves, which can be obtained by connecting the peak points of the hysteretic loops, for all specimens are presented
(a) Setup of strain gauges (Reinforcement cage)
(b) Setup of strain gauges (Steel shape) Fig. 5. Layout of strain gauges and LVDTs. 67
(c) Setup of LVDTs
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC-1
(b) PPSRC-2
(c) PPSRC-6
(d) HPSRC-1
(e) HPSRC-2
(f) HPSRC-4
(g) Close observation of fiber bridges
(h) Photo of a broken specimen
Fig. 7. Failure modes and close observations. Table 5 Summary of the experimental results. ID
Failure mode
Δy (mm)
Δu (mm)
μ
θu (%)
Esum (kN·m)
Me (kN·m)
Mc (kN·m)
Mc /Me
PPSRC-1 PPSRC-2 PPSRC-3 PPSRC-4 PPSRC-5 PPSRC-6 HPSRC-1 HPSRC-2 HPSRC-3 HPSRC-4
F-S F F-S F-S F-S F F-S F-S F-S F-S
15.8 13.5 18.0 13.5 15.8 15.8 13.5 13.5 15.8 13.5
45.0 49.5 45.0 45.0 36.0 49.5 40.5 36.0 36.0 36.0
2.9 3.7 2.5 3.3 2.3 3.1 3.0 2.7 2.3 2.7
5.0 5.5 5.0 5.0 4.0 5.5 4.5 4.0 4.0 4.0
217.5 276.6 218.0 188.9 150.4 250.9 149.1 116.7 106.2 130.0
430.50 419.25 416.89 418.65 416.93 437.31 413.19 380.66 413.72 451.73
371.59 378.68 389.9 378.68 378.68 406.03 365.92 370.76 370.76 370.76
0.86 0.90 0.94 0.90 0.91 0.93 0.89 0.97 0.90 0.82
Note: F-S denotes flexural-shear failure, and F denotes flexural failure; Δy is the yielding displacement; Δu is the ultimate displacement; θu is the ultimate drift ratio; Esum is the sum of energy consumption; Me is the tested moment capacity; Mc is the calculated moment capacity.
propagate after the drift ratio of steel shape yielded. It was noticeable that the slope of skeleton curves slowed down for almost specimens after steel shape yielded, and the yielding displacement of the specimens was therefore defined as the displacement of steel shape yielding. Generally, the hysteretic loops of the PPSRC specimens are more robust than those of the HPSRC specimens, indicating the better
the specimens behaved elastically without any cracking of concrete. It indicated that the existence of steel fibers delayed the crack load compared with the results of steel reinforced ultra-high concrete columns in reference paper [23]. The initial yielding of steel shape and longitudinal rebar occurred in cycles at drift ratios at 1.5% to 2.0%. For the specimens suffered flexural-shear failure, diagonal cracks would 68
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC-1
(b) PPSRC-2
(c) PPSRC-3
(d) PPSRC-4
(e) PPSRC-5
(f) PPSRC-6 Fig. 8. Hysteretic loops.
compression (N = 1500 kN, Fig. 8(a) and (g)) showed relatively stable hysteretic characteristic and good energy dissipation capacity than those of the specimens with a high axial compression (N = 2000 kN, Fig. 8(d) and (h)).
ductility and energy dissipation capacity in PPSRC series. Among them, the specimen PPSRC-2 with abundant stirrups showed the best hysteretic behavior, followed by PPSRC specimens with moderate stirrup ratio, and then the HPSRC specimens. For the specimens with different axial loads, the hysteretic loops for the specimens with a low axial 69
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) HPSRC-1
(b) HPSRC-2
(c) HPSRC-3
(d) HPSRC-4 Fig. 8. (continued)
3.3. Strain results
not dominant by shear.
Fig. 9 showed the strain-displacement relationships of the specimens. The stirrup strain could directly reflect the development of inclined cracks of the columns. As shown in Fig. 9(a), the stirrup did not yield before the peak load, indicating that the formation and propagation of the inclined cracks in the core concrete were restricted in a considerable extent. It was also noticeable that stirrups provided great confinements to the core concrete in the ultimate stage of the test because the stirrup yielded after the peak load. Fig. 9(b) recorded the strain-displacement relationships of the specimen PPSRC-3, it was obvious that the steel flange had yielded at the peak load, indicating that the steel shape worked compatibly with the concrete part. Fig. 9 (c) and (d) presented the shear strains obtained by the strain rosettes of the steel web of the specimen HPSRC-2 and PPSRC-3, respectively. The results showed that the steel web slightly yielded in shear both in the PPSRC and HPSRC specimens at the end stage of the test, which was consistent with the phenomenon of observed flexuralshear failure. However, the maximum shear strain of HPSRC-2 was relatively greater than that of the PPSRC-3, indicating that the solid column core could supply more rigid and capacity than the hollow column core. Compared with the strain result of steel flange in Fig. 9(b), the slight yielding of steel web also denoted that the final failure was
3.4. Stiffness and strength degradation The stiffness degradation, which can be caused by cracking, yielding of reinforcements, bond-slipping between steel and concrete, is an important index to reflect the level of damage of the columns. In this paper, as illustrated in Fig. 10, stiffness is defined as the slope of the line connecting the extreme loading points in both directions of the lateral load- displacement curve for a given cycle, and can be calculated with the following equation:
Ki + =
+Pi +Δi
and Ki − =
−Pi −Δi
(1)
where Ki+ is the secant stiffness at the ith drift level in the positive half cycle, and Ki- is the secant stiffness at the ith drift level in the negative half cycle. As plotted in Fig. 11, the early stiffness decreased due to a plenty of cracks, and then a slow degradation of stiffness could be observed because of the plastic deformation. Therefore, an almost identical “fast followed by slow” degradation law in stiffness could be found for all the specimens. What is noticeable was that the PPSRC columns provided relatively higher values of initial stiffness than HPSRC columns due to the solid column core. Although the HPSRC specimens applied hollow column cores, the steel shape and the continuous stirrups could still 70
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Strain of the transverse stirrup (PPSRC-3)
(b) Strain of the steel flange (PPSRC-3)
(c) Shear strain of the steel web (HPSRC-2)
(d) Shear strain of the steel web (PPSRC-3)
Fig. 9. Strain analysis.
cycle to that in the first cycle, determined by Eq. (2),
γi =
F3i F1i
(2)
Fig. 12 presented the strength degradation of different specimens. It can be seen from the figure that the specimens of PPSRC series suffered a stable strength degradation during the test, and the specimen PPSRC2 with abundant stirrups exhibited the mildest decline among the specimens of PPSRC series, indicating that the beneficial effect of steel fibers and abundant continuous rectangle stirrups becomes more distinctive. Compared with the PPSRC series, the specimens of HPSRC series suffered a dramatic decrease in capacity at the end of the test, and the hollow column core should shoulder the responsibility because the confined concrete in the column core could significantly enhance the deformability of the steel shape [24] and the specimens with a hollow core might suffer a reduced rigid and capacity both in load and deformation. The stirrup volume played a significant role on the strength degradation of all the specimens. The specimens with abundant stirrups (PPSRC-2 and HPSRC-4) showed the best hysteretic behaviors, followed by conventional specimens (PPSRC-4 and HPSRC-2) and then the specimens with light stirrups (PPSRC-5 and HPSRC-3). As the axial load increased, the strength degradation became more obvious, which could be found in the curves of PPSRC-1, PPSRC-4,
Fig. 10. Diagram for result analysis.
confine the precast RPC to prevent an abrupt loss in stiffness, and the precast concrete could also avoid outward buckling of the steel flanges. In this section, the strength degradation factor γi, at the ith drift level, can be defined by the ratio of the bearing capacity in the third 71
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC series
(b) HPSRC series Fig. 11. Stiffness degradation.
It was clear that most of the PPSRC specimens exhibited greater ductility and ultimate displacement than the HPSRC specimens, as discussed in the former section, the weaker deformability could be attributed to the elimination of inner-concrete. It also can be seen that both stirrup spacing, strength of cast-in-place concrete and axial load affected the ductility of the specimens. Generally, the ductility and ultimate drift ratio increased with the decreasing of the stirrup spacing, axial load, and strength of cast-in-place concrete. The area enclosed by the load-displacement hysteretic loops represents the energy dissipated during various cycles of loading. The total energy dissipated during the test can be determined as: m2
Esum =
m1
∑∑ j
Eij (4)
i=1
where m1 is the number of repeated cycles for each drift level; m2 is the number corresponding to the failure drift; Ei j is the area enclosed by the load-displacement hysteretic loop by a cycle i of the jth drift level, as shown in Fig. 10. The cumulative energy dissipation of each specimen is plotted in Fig. 13, where it can be seen that the trend for Esum is similar to that for ductility. Under the same design parameters, greater energy dissipation capacity could be found in the PPSRC specimens compared with the HPSRC specimens. Meanwhile, both in the PPSRC series and HPSRC series, weaker energy dissipation capacity could be obtained in the
Fig. 12. Strength degradation.
HPSRC-1 and HPSRC-2. This indicated that the increasing of axial compression had a negative effect on the cyclic behavior of the specimens. Meanwhile, the strength of cast-in-place concrete did not affect the strength degradation obviously. 3.5. Ductility and energy dissipation capacity Ductility and energy dissipation capacity are the basic indexes used to quantify the seismic response of the column specimens. To evaluate the deformability of the specimens, the inelastic deformation is quantified by the displacement ductility factor μ and the ultimate drift ratio θu. As shown in Fig. 6, the factor θu is defined as the lateral displacement dividing the distance from the lateral loading point to the top of the footing. As illustrated in Fig. 10, the factor μ in terms of the displacement was defined as:
μ=
Δ−u ⎞ 1 ⎛ Δ+u ⎜ + + ⎟ 2 ⎝ Δy Δ−y ⎠
(3)
where the ultimate displacement Δu is taken as the displacement when the remaining capacity drops to 75% of the maximum load. The yield displacement Δy in this paper is determined as the displacement when the steel flange yielded based on the data collected by the strain gauges. The values of Δy, Δu, θu and μ of the specimens were recorded in Table 5.
Fig. 13. Cumulative dissipated energy. 72
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
4. Seismic bending moment capacity 4.1. Basic assumptions According to the experimental results, it is noticeable that the PPSRC and HPSRC specimens both behaved as integral members, indicating that the specimens were sufficiently composite and the steel shape worked together well with the precast outer-part and cast-inplace inner-part. All the final failure modes of the specimens were dominant by flexure, as discussed in Section 3.1. Several researchers have proposed calculation models or methods to predict the flexural capacity of composite members [25,26]. In this paper, with the aim to simplify the calculation procedure, the plastic stress theory could be used here to determine the seismic bending moment capacity of the specimens. In the plastic stress theory, some simple yet rational assumptions can be listed as follows: (1) The plane section assumption was satisfied (2) The part of the steel shape and the longitudinal reinforcements at the tension zone were considered as tensed to yield, and the rest of the steel shape and the longitudinal reinforcements at the compression zone were considered as compressed to yield. (3) The tensile strength of the conventional concrete could be neglected, but that of the reactive powder concrete should be taken into account.
Fig. 14. Calculation method for equivalent viscous damping ratio.
specimens under greater axial compression or with lighter transverse reinforcement. In order to further analyze the energy dissipation capacity of the specimens, the equivalent viscous damping ratio is also employed here, as illustrated in Fig. 14. The equivalent viscous damping factor (ξeq) can be used to evaluate the pinching degree of the hysteretic loop. The definition of ξeq is expressed as:
ξeq =
1 S(DBC+DAC) 2π S(OBE + OAF)
4.2. Calculation methods Based on the above assumptions, a model for determining seismic bending moment capacity of PPSRC and HPSRC column is proposed here. In order to simplify the design procedure of reactive powder concrete members, some researchers [27,28] proposed the rectangle stress block method for reactive powder concrete, and it was employed here to calculate the contribution of precast outer-part. As shown in Fig. 16, based on the above analysis, the height of the concrete compression zone x and the seismic bending moment capacity M, which can be calculated by Eqs. (6)–(10), can be determined from the equilibrium conditions of internal force in the cross section.
(5)
Fig. 15 showed the variation of equivalent viscous damping ratio of the specimens. In general, equivalent viscous damping ratio of each test specimen increases gradually as the displacement increases. The ξeq of PPSRC specimens is higher than that of HPSRC specimens at the ultimate stage, indicating that the hysteretic loops of the PPSRC specimens are more robust. Under the same design parameters, the higher ξeq can be found in the specimens with abundant stirrups, and the strength of cast-in-place concrete has a limited effect on ξeq.
∑N = 0 ′ + fys Asf′ + Couter + Cinner−Touter−fy As −fys Asw −fys Asf N = fy As′ + fys Asw −fys (2Asf2 + Asw2 ) (6)
∑M = 0 ′ M = fy As′ (x −as) + fys Asf′ (x −ass) + fys Asw +
(x − a ) Cinner 2 ss
+
(h − x ) Touter 2
(x − ass) 2
(
( )( ) + N( )
+ fy As (h−as−x ) + fys Asw
+ fys Asf (h−ass−x ) + fys (2Asf2 + Asw2
βx
+ Couter x − 2
h −x 2
)
hw − x − ass 2
)
h −x 2
(7)
Couter = αfc,outer βxb−αfc,outer h w (x −ass)
(8)
Cinner = fc,inner h w (x −ass)
(9)
Touter = kft b (h−x )−h w (h−ass−x )
(10)
where x is the height of concrete compression zone; h is the column height; b is the column width; hw is the height of steel web; as is the thickness of cover concrete; ass is the distance from the steel flange to the column edge; As and As′ are the areas of tensile longitudinal rebar and compressive longitudinal rebar, respectively; Asw and Asw′ are the areas of tensile steel web and compressive steel web, respectively; Asf and Asf′ are the areas of tensile steel flange and compressive steel flange, respectively; Asf2 and Asw2 are the areas of steel flange and steel
Fig. 15. Equivalent viscous damping ratio. 73
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Fig. 16. Calculation diagram of normal section at the ultimate limit state.
columns can be much higher than those of the specimens in this paper, therefore, the deformability, ductility ratios and energy dissipation capacity can be significantly enhanced [23], indicating that large-scale HPSRC columns might exhibit better hysteretic behaviors. Therefore, more large-scale tests should be conducted in the future.
web perpendicular to the loading direction; fc,outer is the compressive strength of reactive powder concrete; fc, inner is the compressive strength of cast-in-place concrete; fys and fy are the yield strengths of steel shape and longitudinal rebar; α and β are factors of rectangle stress block method of RPC members, and k is the factor to determine the contribution of tensile strength of RPC to the bearing capacity, according to references [27,28], α = 0.9, β = 0.77 and k = 0.25 here. Based on the equations established above, the seismic bending moment capacities of all the specimens were calculated. All the calculated results were listed and plotted in Table 5 and Fig. 17, respectively. As can be seen, the calculated values agreed reasonably well with the test results, and an average ratio of 0.90 and a coefficient of variation of 0.04 were observed here.
5. Conclusions The seismic behaviors of partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns have been investigated in this paper. A total of 10 column specimens were tested under combined axial compression and lateral cyclic load. The main conclusions drawn from this paper are as follows: (1) Two typical failure modes were observed in the specimens, namely flexural and flexural-shear failure. The flexural failure occurred in the specimens with abundant stirrups and the specimens with highstrength solid column core, and the other specimens all failed in flexural-shear mode. The specimens of PPSRC series exhibited better behaviors than those of HPSRC series both in ductility, strength degradation and energy dissipation, indicating that the solid column core played an important role in seismic behavior of the whole member. In the same series, higher stirrup ratio and lower axial compression of the column leaded to more satisfactory energy dissipation capacity, stiffness degradation and higher ductility. In the PPSRC specimens, the effect of strength of cast-in-place concrete was insignificant on the hysteretic behavior. (2) Due to the high-strength bolts and tear plates, the steel shape, precast concrete and cast-in-place concrete of both the PPSRC specimens and HPSRC specimens bonded well during the entire test process. Although some bonding cracks were found at the end of the test, the final failure was not dominant by bonding invalidation. Attribute to the steel fibers in precast shells, the bottom concrete covers of all the specimens bulged but did not spall during the test. The steel shape and longitudinal rebar were well confined by the continuous rectangle stirrups and surrounding concrete, so no obvious buckling was found in steel reinforcements. (3) Based on the plastic stress theory, a set of calculation formulas for determining seismic bending moment capacity of the PPSRC and HPSRC columns were proposed here. The results obtained form the formulas agreed well with those from the experiments. From the calculated axial capacity-moment capacity interaction curves, it can be seen that the seismic bending moment capacity of the HPSRC column was close to that of the PPSRC column in the low level of axial compression, indicating that when a structural column is subjected to low compressive load, such as columns of upper floors, the HPSRC column without inner concrete can be applied as a permanent column.
4.3. Axial capacity-moment capacity interaction curves The axial capacity - moment capacity interaction curves could be obtained by the similar procedures as the former section discussed. In Fig. 17, the load-carrying capacity of the hollow section (HPSRC) was compared with that of the solid section (PPSRC). As shown in this figure, the seismic bending moment capacity of the HPSRC column was close to that of the PPSRC column in the low level of axial compression. This result indicated that when a structural column is subjected to low compressive load, such as columns of upper floors, the HPSRC column without inner concrete can be applied as a permanent column. Because of the minimum thickness requirement of the concrete cover of steel shapes, the height of the steel web is limited here, which leads to a relatively low steel ratio. In the practical applications where full-scale columns are applied, the steel ratios of the PPSRC and HPSRC
Fig. 17. Test results and N-M interaction curves. 74
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Acknowledgment [14]
The experiments are sponsored by the National Natural Science Foundation of China (Grant No. 51578443, 51778525) and National Key Research and Development Program of China (Program No. 2017YFC0703404). The supports are greatly appreciated.
[15] [16]
References
[17]
[1] Hajjar JF. Composite steel and concrete structural systems for seismic engineering. J Constr Steel Res 2002;58(5):703–23. [2] Johnson RP. Composite structures of steel and concrete: beams, slabs, columns, and frames for buildings. Blackwell Pub; 2004. [3] Griffis LG. Some design considerations for composite-frame structures. Eng J 1986;23(2):59–64. [4] Foraboschi P. Versatility of steel in correcting construction deficiencies and in seismic retrofitting of RC buildings. J Build Eng 2016;8:107–22. [5] El-Tawil S, Deierlein GG. Strength and ductility of concrete encased composite columns. J Struct Eng 1999;125(9):1009–19. [6] Chen C, Wang C, Sun H. Experimental study on seismic behavior of full encased steel-concrete composite columns. J Struct Eng 2014;140(6):1–10. [7] Xue J. Design principles of steel-concrete composite structures. China Architecture and Building Press; 2012 [in Chinese]. [8] Tam VWY, Tam CM, Zeng SX, et al. Towards adoption of prefabrication in construction. Build Environ 2007;42(10):3642–54. [9] Park R. Seismic design and construction of precast concrete buildings in New Zealand. PCI J 2002;47(5):60–75. [10] Zhu B. Introduction of the prefabricated structures. China Architecture and Building Press; 2012. [in Chinese]. [11] Hong WK, Park SC, Kim JM, et al. Composite beam composed of steel and precast concrete (Modularized Hybrid System, MHS). Part I: Experimental investigation. Struct Des Tall Special Build 2008;19(3):275–89. [12] Hong WK, Kim JM, Park SC, et al. Composite beam composed of steel and precast concrete. (Modularized Hybrid System, MHS) Part II: Analytical investigation. Struct Des Tall Special Build 2009;18(8):891–905. [13] Hong WK, Park SC, Lee HC, et al. Composite beam composed of steel and precast
[18] [19] [20] [21] [22]
[23]
[24]
[25] [26] [27] [28]
75
concrete (modularized hybrid system). Part III: Application for a 19-storey building. Struct Des Tall Special Build 2009;19(6):679–706. Hong WK, Kim SI, Park SC, et al. Composite beam composed of steel and precast concrete (modularized hybrid system). Part IV: Application for multi-residential housing. Struct Des Tall Special Build 2010;19(7):707–27. Yang Y, Xue Y, Yu Y, et al. Experimental study on flexural performance of partially precast steel reinforced concrete beams. J Constr Steel Res 2017;133:192–201. Yang Y, Yu Y, Guo Y, et al. Experimental study on shear performance of partially precast Castellated Steel Reinforced Concrete (CPSRC) beams. Steel Compos Struct 2016;21(2):289–302. Xu S, Wu C, Liu Z, et al. Experimental investigation of seismic behavior of ultra-high performance steel fiber reinforced concrete columns. Eng Struct 2017;152:129–48. Zheng W, Lv X. Literature review of reactive powder concrete. J Build Struct 2015;36(10):45–58. (in Chinese). Foraboschi P. Analytical model to predict the lifetime of concrete members externally reinforced with FRP. Theor Appl Fract Mech 2015;75(1):137–45. Focacci F, Foraboschi P, Stefano MD. Composite beam generally connected: analytical model. Compos Struct 2015;133:1237–48. Foraboschi P. Layered plate with discontinuous connection: exact mathematical model. Compos B 2013;47(3):365–78. Huang L, Xu L, Chi Y, Xu H. Experimental investigation on the seismic performance of steel-polypropylene hybrid fiber reinforced concrete columns. Constr Build Mater 2015;87:16–27. Zhu W, Jia J, Gao J, et al. Experimental study on steel reinforced high-strength concrete columns under cyclic lateral force and constant axial load. Eng Struct 2016;125:191–204. Han LH, Li W, Bjorhovde R. Developments and advanced applications of concretefilled steel tubular (CFST) structures: Members. J Constr Steel Res 2014;100:211–28. Foraboschi P. Three-layered plate: elasticity solution. Compos B 2014;60(1):764–76. Eurocode 4: design of composite steel and concrete structures, part 1-1, general rules and rues for building. European Committee for Standardization; 2004. Li L. Mechanical behavior and design method for reactive powder concrete beams [Ph.D. Dissertation]. Harbin Institute of Technology; 2010 [in Chinese]. Zheng W, Li L, Lu S. Experimental research on mechanical performance of normal section of reinforced reactive powder concrete. J Build Struct 2011;32(6):125–34. [in Chinese].
Contents lists available at ScienceDirect
Engineering Structures journal homepage: www.elsevier.com/locate/engstruct
Experimental study on seismic performance of partially precast steel reinforced concrete columns
T
⁎
Yong Yang, Yicong Xue , Yunlong Yu, Fengqi Gao School of Civil Engineering, Xi’an University of Architecture & Technology, Xi’an, Shaanxi 710055, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Steel reinforced concrete columns Precast structures Cyclic loading tests Reactive powder concrete Seismic performance
This paper aims to develop two kinds of innovative precast steel reinforced concrete columns, which are partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns. Both the two kinds of composite columns have precast reactive powder concrete (RPC) shells, and the PPSRC column has a cast-in-place column core. In this paper, a series of cyclic loading tests on 10 column specimens subjected to combined static axial loading and cyclic lateral loading were carried out to explore their seismic performances. All specimens were evaluated by the failure modes, hysteresis characteristics, strength and stiffness degradation, energy dissipation capacity and ductility. The effects of section shape, stirrup spacing, axial compression and concrete strength of cast-in-place inner-part were investigated in details. The experiment results indicated that the PPSRC columns exhibited more satisfactory seismic behavior than the HPSRC columns in terms of hysteretic behavior, strength degradation, ductility and energy dissipation, while their bearing capacities were almost identical under low axial compression. Steel fibers could effectively prevent the cracked concrete from spalling, therefore, the encased steel shape was efficiently confined by the surrounding concrete during the entire test process. Higher stirrup ratio and lower axial compression of the column leaded to more satisfactory energy dissipation capacity, stiffness degradation and higher ductility. Based on the plastic stress theory, the seismic bending moment capacity analysis was conducted, and the results obtained form the formulas agreed well with those from the experiments.
1. Introduction During the past few decades, steel-concrete composite structural systems have been widely used in constructing new buildings and retrofitting existing structures all over the world. This system combines the rigidity and formability of reinforced concrete members with the strength and convenience of construction associated with structural steel to produce an economic and robust structure [1–4]. As an important part of steel-concrete composite structures, steel reinforced concrete structures (SRCs), have been widely applied in marine, largespan and high-rise structures due to the high bearing capacity, great stiffness, great durability and outstanding ductility performance[5–7]. Nevertheless, cast-in-place SRC structures usually involve the construction procedures both of steel structures and concrete structures, therefore, the application of cast-in-place SRC structures has been limited in conventional buildings due to complex construction process. At the same period, applications of precast structures have increased because of the efficiency and high quality in construction, and it can also provide an advisable solution to the problem of wasted resources
⁎
for on-site activities [8–10]. With the aim to facilitate the construction process of SRC structures, some researchers have suggested the combination of SRC members and precast concrete members. As a good example, the Fujita steel and reinforced precast concrete system has been adopted in Japan [10]. It indicates that the application of entire precast SRC members seems to be a solution to the problem aforementioned, however, the heavy deadweight and the vulnerable joint area could become new problems to the structure engineers. On the other hand, the builders must employ special large vehicles and cranes to lift the entire precast SRC members, which would directly increase the building cost. As an alternative method to reduce the deadweight of precast SRC members, Hong et al. [11–14] proposed a new kind of partially precast steel reinforced concrete system and named it as MHS (modularized hybrid system). However, the optimal combination mode, section shape and the structural integrity in the partially precast SRC members must be further investigated for better structural capacity and constructability. Furthermore, the type and strength of concrete in the precast part and cast-in-place part should be altered to match different
Corresponding author at: No. 13, Yanta Road, Xi'an, Shaanxi 710055, China. E-mail address: [email protected] (Y. Xue).
https://doi.org/10.1016/j.engstruct.2018.08.027 Received 20 February 2018; Received in revised form 24 July 2018; Accepted 11 August 2018 0141-0296/ © 2018 Elsevier Ltd. All rights reserved.
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Fig. 1. Schematic diagram of the PPSRC column and HPSRC column.
building with the limited loss of rigidity. Meanwhile, the dimensions of the PPSRC and HPSRC columns can be unified along the building height to facilitate the conventional variable section design. As an essential issue in composite members, the connecting and bonding behaviors should be emphasized [20,21]. Because there are two kinds of concrete in the same cross-section of PPSRC columns, it is necessary to examine the integrity of PPSRC columns under seismic loads, therefore, a series of cyclic loading tests on 10 column specimens subjected to combined static axial loading and cyclic lateral loading were carried out. The crack patterns, failure modes, hysteretic loops, skeleton curves, strength and stiffness degradation, energy dissipation capacity and ductility of the specimens were critically examined. The effects of section shape, stirrup spacing, axial compression and concrete strength of cast-in-place inner-part on the seismic performance of the specimens were also explored in details. Based on the experimental observations and test results, an analytic model for calculating seismic bending moment capacity is also presented later in this paper.
purposes. With the aim to promote the applications of SRC structures and solve the problems mentioned above, an innovative partially precast steel reinforced concrete structure system (PPSRC) was proposed by the authors, as illustrated in Fig. 1. The PPSRC frame structure is composed of PPSRC beams, PPSRC columns and PPSRC joint, and the mechanical performance of PPSRC beams has been thoroughly investigated in reference paper [15,16]. In this paper, two innovative partially precast SRC columns, which are partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns, are presented here. In the PPSRC column, the precast outerpart, which is composed of a cruciform steel shape, high-performance concrete, continuous rectangle stirrups and longitudinal rebar, can be prefabricated in factory, and the cast-in-place inner-part can be cast by conventional strength concrete, lightweight aggregate concrete or recycled aggregate concrete on the construction site after the outer-part erected. For the HPSRC columns, the outer-part is the same as that of PPSRC columns, but the column core is kept hollow to further reduce the deadweight or can be filled with heat insulating material to enhance the anti-fire performance. Meanwhile, the HPSRC columns can be regarded as formwork of the inner-part of PPSRC columns. With the aim to further improve the performance of precast outerpart, the reactive powder concrete (RPC) is applied here. As a kind of ultra-high performance concrete, reactive powder concrete has been widely studied due to some favorable characteristics, such as high durable and mechanical performance [17,18]. Especially, the steel fibers in RPC can significantly show down the crack growth in concrete [19]. Continuous rectangle stirrups are also adopted here to avoid the invalidation of stirrup hooks in severe earthquakes. For the inner-part, it can be cast by conventional strength concrete with the beam core of PPSRC beams and adjacent slabs at the same time to ensure strutural integrity of PPSRC structures and to save the expensive high-performance concrete. As shown in Fig. 1, the PPSRC columns and HPSRC columns can be composite in a high-rise building vertically. For the columns in the ground floor, the PPSRC columns with inner concrete will meet the needs of great axial capacity and sufficient ductility under proper design, and for the columns in upper floors, the use of HPSRC columns with hollow core can efficiently reduce the deadweight of the entire
2. Experimental program 2.1. Test specimens Ten column specimens were designed and constructed, including six PPSRC columns, and four HPSRC columns. The key parameters are summarized in Table 1. These ten specimens were similar in cross section shape and dimensions, except for the HPSRC columns with hollow cores. As shown in Fig. 2, the columns were 300 mm × 300 mm square in cross section, with a height of 900 mm from the point of lateral loading to the top of the stub. The specimens were designed to represent the structural columns of lower stories of high-rise structures and were designed at one second of full scale to match the capacity of the specimen with that of the test device. A reinforced concrete stub footing with a cross section of 500 mm by 550 mm was cast together with the specimen, representing a relative rigid foundation. The steel shape in all specimens was cruciform and weld by two HN175 × 90 × 5 × 8 steel of Q235 grade according to the Chinese standards. The total height and the width of the steel shape were 175 mm and 90 mm, respectively, and the thickness of the web and flange were 5 mm and 8 mm, respectively. As shown in Fig. 2, the steel 64
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Table 1 Matrices of test specimens. ID
λ
Section shape
N (kN)
Concrete
Reinforcements
Steel shape
fc,out (MPa)
fc,in (MPa)
ρsv
ρs
Type
ρss
PPSRC-1 PPSRC-2 PPSRC-3 PPSRC-4 PPSRC-5 PPSRC-6
3.0 3.0 3.0 3.0 3.0 3.0
Solid Solid Solid Solid Solid Solid
1500 2000 2000 2000 2000 2000
96.8 96.8 96.8 96.8 96.8 96.8
24.3 24.3 47.6 24.3 24.3 96.8
1.26%(D8@65) 2.00%(D8@40) 1.26%(D8@65) 1.26%(D8@65) 0.68%(D8@120) 1.26%(D8@65)
3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18)
2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8)
5.0% 5.0% 5.0% 5.0% 5.0% 5.0%
HPSRC-1 HPSRC-2 HPSRC-3 HPSRC-4
3.0 3.0 3.0 3.0
Hollow Hollow Hollow Hollow
1500 2000 2000 2000
96.8 96.8 96.8 96.8
– – – –
1.26%(D8@65) 1.26%(D8@65) 0.68%(D8@120) 2.00%(D8@40)
3.4%(12D18) 3.4%(12D18) 3.4%(12D18) 3.4%(12D18)
2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8) 2(HN175 × 90 × 5 × 8)
5.0% 5.0% 5.0% 5.0%
Note: λ is the aspect ratio of the specimens; N is the constant axial load on the column top; fc,out is the compressive strength of the reactive powder concrete; fc,in is the compressive strength of the cast-in-place concrete; ρsv is the volumetric stirrup ratio; ρs is the longitudinal reinforcement ratio; ρss is the structural steel ratio.
in this step. In the first step, before the concrete cast, tear plates were assembled between the adjacent steel flanges by dot welding as the formwork of the outer-part. The effect of the thin tear plate on the cyclic performance of entire specimen can be ignored due to its weak stiffness, but the tear plates and high-strength bolts in the outer-part could play important roles to enhance the bonding performance between the precast concrete, cast-in-place concrete and steel shape. In the practical applications, reusable inflated rubber formwork can be applied in the outer-part to further prefabricate the outer-part efficiently.
flanges were drilled and high-strength bolts were then assembled at the bottom and top of the steel flange as shear connectors to enhance the bonding performance between the precast outer-part and cast-in-place inner-part. Because the head of the high-strength bolt was in the precast outer-part and the rear of the high-strength bolt was in the cast-in-place inner part, both the precast concrete and cast-in-place concrete could benefit from the shear connectors, as illustrated in Fig. 3(c). All the reinforcing rebar and stirrups of the specimens are illustrated in detail in Fig. 2. 2.2. Specimens fabrication
2.3. Materials All the PPSRC and HPSRC columns were constructed in two steps. As shown in Fig. 3, in the first step, the outer-part of the specimens was molded and precast. The outer part was composed of a cruciform steel shape, reactive powder concrete, continuous rectangle stirrups and longitudinal reinforcements, and the concrete in the outer-part was cast at the same height of the steel shape, as indicated in Fig. 3(b). The second step was started after the precast concrete hardened, and the inner-part, namely the column core, was cast by conventional concrete
The concrete in the outer-part is a kind of reactive powder concrete, which is designed as C100 grade. The mixed proportions for the cement-based matrix of the RPC used in this study are summarized in Table 2. The components of RPC include 52.5P.O. cement, silica fume, fly ash, mineral powder, quartz sand, river sand, water reducer, water and steel fiber. A 2% volume incorporation of steel fibers is used in the RPC and the mechanical properties of the steel fibers are shown in
Fig. 2. Details of specimens (Unit: mm). 65
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Continuous rectangle stirrup
(b) The outer part of specimen
(c) Shear connectors
Fig. 3. Configuration of the specimens.
Table 3. The average tested compressive strength of the RPC coupons is 96.8 MPa, and the average tested tensile strength was 6.83 MPa. The strength grades of concrete in the inner-part were designed at three different levels, and the tested compressive strengths at 28 days were 24.3 MPa, 47.6 MPa and 96.8 MPa, respectively. Table 4 summarizes the mechanical properties of steel rebar used in this study, where fy is the yield strength and fu is the ultimate strength.
Table 3 Properties of steel fiber. Type
Diameter (mm)
Length (mm)
Tensile strength (MPa)
Straight
0.2
13.0
2850
Table 4 Mechanical properties of steel reinforcement.
2.4. Test setup and instrumentations
Reinforcement
Type
Steel grade
fy (MPa)
fu (MPa)
The test setup and details of instrumentations are schematically shown in Fig. 4. The vertical compressive loading was applied by a 5000 kN hydraulic jack, and the cyclic lateral loading was applied on the load point by a 1000 kN electro-hydraulic servo machine after the axial compression force on the top of the specimens stabilized. The reaction devices on the ground and two steel beams on the footing stub of the specimens were employed to prevent the foundation of columns from sliding during the test. During the test process, the horizontal deflections at the loading point and the footing stub and the vertical deflections at the both sides of the plastic hinge area were measured using linear variable differential transformers (LVDTs). A considerable number of electrical-resistance strain gauges were arranged on the web and flanges of the steel shape and the reinforcing bars to monitor the strain response. A total of 14 strain gauges were adhered on the steel shape and 7 strain gauges for the reinforcement cage. The layouts of the strain gauges and LVDTs are illustrated in Fig. 5.
Longitudinal rebar Stirrup
D18 D8
HRB400 HPB300
443.5 393.2
598.0 562.3
Steel shape
HN175 × 90 × 5 × 8
Q235 Q235
311.7 272.5
438.3 520.0
Flange Web
3. Experimental results and analysis 3.1. Experimental observations The failure modes and damage patterns of the specimens are shown in Fig. 7 and the main test results are recorded in Table 5. (1) Based on the failure patterns of the 6 PPSRC columns and 4 HPSRC columns, two typical failure modes of the specimens could be observed, namely flexural failure and flexural-shear failure, as shown in Fig. 7, in which only the local pictures, about 1/2 height of the entire specimen, are presented for better illustration of the crack patterns and its propagation. The flexural failure was characterized that the transverse cracks initiated and developed after longitudinal rebar and steel flange yield, and the flexural-shear failure was defined as a combination with some diagonal cracks, and yet it still failed because of the propagation of main transverse cracks. Based on the above analysis, flexural failure was only found in specimen PPSRC-2 with sufficient stirrups and specimen PPSRC-6 with the highest-strength core. Slight or moderate shear failure characteristic of other specimens might cause by the lack of coarse aggregate in the reactive powder concrete, which should weaken the shear resistance of aggregate interlock. The similar conclusions could be found in other research about reinforced RPC or other high-performance concrete columns [17,18,22]. As shown in Fig. 7, abundant cracks appeared and propagated in the region where the distance from the column bottom was about 300 mm to 350 mm (6 grids to 7 grids, 50 mm per grid), indicating that the plastic length of all the specimens was approximate 350 mm, namely 1/3 height of the entire column. (2) As shown in Fig. 7(a)–(f), bonding cracks were captured on the
2.5. Test procedure At the beginning of the tests, an initial loading procedure was followed to make sure that there was no significant eccentricity by checking the data of the LVDTs. Then the predetermined axial compressive load was applied with the speed of 2 kN/s and the constant axial load was maintained during testing. A small lateral force was also applied several times at this stage in order to stabilize the test system, then the lateral force was cycled under a lateral displacement control mode. Six single cycles with peak drift ratios of Δ/L = 0.1%, 0.2%, 0.3%, 0.6%, 0.8%, 1.0% were initially applied. Cycles with peak drift ratios of Δ/L = 1.5%, 2.0%, 2.5%…were then applied, with the cycles repeated three times. Fig. 6 shows the load procedure of the cyclic lateral force. The test terminated when the lateral force fell by more than 25% of the maximum experienced force.
Table 2 Designed mix proportions of concrete matrix. 52.5P.O.Cement
Silica fume
Fly ash
Mineral powder
Quartz sand
River sand
Super plasticizer
W/B
Steel fiber
1
0.15
0.2
0.1
0.5
0.8
0.003
0.22
0.02
Note: the ratio of steel fiber is the volume ratio, and ratios of other components are weight ratios. 66
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Schematic diagram
(b) Photo Fig. 4. Test device.
observed sides (perpendicular to the loading direction) in some specimens. It can be attributed to the relatively small concrete cover thickness of the bolts on the steel flanges perpendicular to the loading direction, but the development of these bonding cracks seemed to affect the cyclic performance of the specimens limitedly. (3) Steel fibers in the specimens played an important role to prevent bottom concrete covers from spalling. Although lots of steel fibers failed because of pulling out from concrete, a closer observation to the cracks indicated that some fibers still bridged the two sides of a crack (Fig. 7(g)). Meanwhile, as shown in Fig. 7(h), some specimens were broken after the tests, and no obvious buckling was found both in longitudinal rebar and steel shape, which indicated that the steel reinforcements were well confined by the existing surrounding concrete. However, once the fibers were removed from high-strength concrete, significant spalling of bottom cover concrete and severe damage in the plastic region would cause the buckling of steel shape and longitudinal rebar [23].
Fig. 6. Loading protocol of lateral force.
in Fig. 8. For better comparison, the skeleton curves of the specimens are plotted with bold lines in Fig. 8. The occurrences of special events such as cracking of concrete, initial yielding of longitudinal rebar and steel flange and the ultimate condition (concrete crushing) are marked in the figures. When drifts were no greater than approximate 0.9%, all
3.2. Hysteretic loops and skeleton curves The measured lateral force versus top displacement hysteretic relationships and skeleton curves, which can be obtained by connecting the peak points of the hysteretic loops, for all specimens are presented
(a) Setup of strain gauges (Reinforcement cage)
(b) Setup of strain gauges (Steel shape) Fig. 5. Layout of strain gauges and LVDTs. 67
(c) Setup of LVDTs
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC-1
(b) PPSRC-2
(c) PPSRC-6
(d) HPSRC-1
(e) HPSRC-2
(f) HPSRC-4
(g) Close observation of fiber bridges
(h) Photo of a broken specimen
Fig. 7. Failure modes and close observations. Table 5 Summary of the experimental results. ID
Failure mode
Δy (mm)
Δu (mm)
μ
θu (%)
Esum (kN·m)
Me (kN·m)
Mc (kN·m)
Mc /Me
PPSRC-1 PPSRC-2 PPSRC-3 PPSRC-4 PPSRC-5 PPSRC-6 HPSRC-1 HPSRC-2 HPSRC-3 HPSRC-4
F-S F F-S F-S F-S F F-S F-S F-S F-S
15.8 13.5 18.0 13.5 15.8 15.8 13.5 13.5 15.8 13.5
45.0 49.5 45.0 45.0 36.0 49.5 40.5 36.0 36.0 36.0
2.9 3.7 2.5 3.3 2.3 3.1 3.0 2.7 2.3 2.7
5.0 5.5 5.0 5.0 4.0 5.5 4.5 4.0 4.0 4.0
217.5 276.6 218.0 188.9 150.4 250.9 149.1 116.7 106.2 130.0
430.50 419.25 416.89 418.65 416.93 437.31 413.19 380.66 413.72 451.73
371.59 378.68 389.9 378.68 378.68 406.03 365.92 370.76 370.76 370.76
0.86 0.90 0.94 0.90 0.91 0.93 0.89 0.97 0.90 0.82
Note: F-S denotes flexural-shear failure, and F denotes flexural failure; Δy is the yielding displacement; Δu is the ultimate displacement; θu is the ultimate drift ratio; Esum is the sum of energy consumption; Me is the tested moment capacity; Mc is the calculated moment capacity.
propagate after the drift ratio of steel shape yielded. It was noticeable that the slope of skeleton curves slowed down for almost specimens after steel shape yielded, and the yielding displacement of the specimens was therefore defined as the displacement of steel shape yielding. Generally, the hysteretic loops of the PPSRC specimens are more robust than those of the HPSRC specimens, indicating the better
the specimens behaved elastically without any cracking of concrete. It indicated that the existence of steel fibers delayed the crack load compared with the results of steel reinforced ultra-high concrete columns in reference paper [23]. The initial yielding of steel shape and longitudinal rebar occurred in cycles at drift ratios at 1.5% to 2.0%. For the specimens suffered flexural-shear failure, diagonal cracks would 68
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC-1
(b) PPSRC-2
(c) PPSRC-3
(d) PPSRC-4
(e) PPSRC-5
(f) PPSRC-6 Fig. 8. Hysteretic loops.
compression (N = 1500 kN, Fig. 8(a) and (g)) showed relatively stable hysteretic characteristic and good energy dissipation capacity than those of the specimens with a high axial compression (N = 2000 kN, Fig. 8(d) and (h)).
ductility and energy dissipation capacity in PPSRC series. Among them, the specimen PPSRC-2 with abundant stirrups showed the best hysteretic behavior, followed by PPSRC specimens with moderate stirrup ratio, and then the HPSRC specimens. For the specimens with different axial loads, the hysteretic loops for the specimens with a low axial 69
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) HPSRC-1
(b) HPSRC-2
(c) HPSRC-3
(d) HPSRC-4 Fig. 8. (continued)
3.3. Strain results
not dominant by shear.
Fig. 9 showed the strain-displacement relationships of the specimens. The stirrup strain could directly reflect the development of inclined cracks of the columns. As shown in Fig. 9(a), the stirrup did not yield before the peak load, indicating that the formation and propagation of the inclined cracks in the core concrete were restricted in a considerable extent. It was also noticeable that stirrups provided great confinements to the core concrete in the ultimate stage of the test because the stirrup yielded after the peak load. Fig. 9(b) recorded the strain-displacement relationships of the specimen PPSRC-3, it was obvious that the steel flange had yielded at the peak load, indicating that the steel shape worked compatibly with the concrete part. Fig. 9 (c) and (d) presented the shear strains obtained by the strain rosettes of the steel web of the specimen HPSRC-2 and PPSRC-3, respectively. The results showed that the steel web slightly yielded in shear both in the PPSRC and HPSRC specimens at the end stage of the test, which was consistent with the phenomenon of observed flexuralshear failure. However, the maximum shear strain of HPSRC-2 was relatively greater than that of the PPSRC-3, indicating that the solid column core could supply more rigid and capacity than the hollow column core. Compared with the strain result of steel flange in Fig. 9(b), the slight yielding of steel web also denoted that the final failure was
3.4. Stiffness and strength degradation The stiffness degradation, which can be caused by cracking, yielding of reinforcements, bond-slipping between steel and concrete, is an important index to reflect the level of damage of the columns. In this paper, as illustrated in Fig. 10, stiffness is defined as the slope of the line connecting the extreme loading points in both directions of the lateral load- displacement curve for a given cycle, and can be calculated with the following equation:
Ki + =
+Pi +Δi
and Ki − =
−Pi −Δi
(1)
where Ki+ is the secant stiffness at the ith drift level in the positive half cycle, and Ki- is the secant stiffness at the ith drift level in the negative half cycle. As plotted in Fig. 11, the early stiffness decreased due to a plenty of cracks, and then a slow degradation of stiffness could be observed because of the plastic deformation. Therefore, an almost identical “fast followed by slow” degradation law in stiffness could be found for all the specimens. What is noticeable was that the PPSRC columns provided relatively higher values of initial stiffness than HPSRC columns due to the solid column core. Although the HPSRC specimens applied hollow column cores, the steel shape and the continuous stirrups could still 70
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) Strain of the transverse stirrup (PPSRC-3)
(b) Strain of the steel flange (PPSRC-3)
(c) Shear strain of the steel web (HPSRC-2)
(d) Shear strain of the steel web (PPSRC-3)
Fig. 9. Strain analysis.
cycle to that in the first cycle, determined by Eq. (2),
γi =
F3i F1i
(2)
Fig. 12 presented the strength degradation of different specimens. It can be seen from the figure that the specimens of PPSRC series suffered a stable strength degradation during the test, and the specimen PPSRC2 with abundant stirrups exhibited the mildest decline among the specimens of PPSRC series, indicating that the beneficial effect of steel fibers and abundant continuous rectangle stirrups becomes more distinctive. Compared with the PPSRC series, the specimens of HPSRC series suffered a dramatic decrease in capacity at the end of the test, and the hollow column core should shoulder the responsibility because the confined concrete in the column core could significantly enhance the deformability of the steel shape [24] and the specimens with a hollow core might suffer a reduced rigid and capacity both in load and deformation. The stirrup volume played a significant role on the strength degradation of all the specimens. The specimens with abundant stirrups (PPSRC-2 and HPSRC-4) showed the best hysteretic behaviors, followed by conventional specimens (PPSRC-4 and HPSRC-2) and then the specimens with light stirrups (PPSRC-5 and HPSRC-3). As the axial load increased, the strength degradation became more obvious, which could be found in the curves of PPSRC-1, PPSRC-4,
Fig. 10. Diagram for result analysis.
confine the precast RPC to prevent an abrupt loss in stiffness, and the precast concrete could also avoid outward buckling of the steel flanges. In this section, the strength degradation factor γi, at the ith drift level, can be defined by the ratio of the bearing capacity in the third 71
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
(a) PPSRC series
(b) HPSRC series Fig. 11. Stiffness degradation.
It was clear that most of the PPSRC specimens exhibited greater ductility and ultimate displacement than the HPSRC specimens, as discussed in the former section, the weaker deformability could be attributed to the elimination of inner-concrete. It also can be seen that both stirrup spacing, strength of cast-in-place concrete and axial load affected the ductility of the specimens. Generally, the ductility and ultimate drift ratio increased with the decreasing of the stirrup spacing, axial load, and strength of cast-in-place concrete. The area enclosed by the load-displacement hysteretic loops represents the energy dissipated during various cycles of loading. The total energy dissipated during the test can be determined as: m2
Esum =
m1
∑∑ j
Eij (4)
i=1
where m1 is the number of repeated cycles for each drift level; m2 is the number corresponding to the failure drift; Ei j is the area enclosed by the load-displacement hysteretic loop by a cycle i of the jth drift level, as shown in Fig. 10. The cumulative energy dissipation of each specimen is plotted in Fig. 13, where it can be seen that the trend for Esum is similar to that for ductility. Under the same design parameters, greater energy dissipation capacity could be found in the PPSRC specimens compared with the HPSRC specimens. Meanwhile, both in the PPSRC series and HPSRC series, weaker energy dissipation capacity could be obtained in the
Fig. 12. Strength degradation.
HPSRC-1 and HPSRC-2. This indicated that the increasing of axial compression had a negative effect on the cyclic behavior of the specimens. Meanwhile, the strength of cast-in-place concrete did not affect the strength degradation obviously. 3.5. Ductility and energy dissipation capacity Ductility and energy dissipation capacity are the basic indexes used to quantify the seismic response of the column specimens. To evaluate the deformability of the specimens, the inelastic deformation is quantified by the displacement ductility factor μ and the ultimate drift ratio θu. As shown in Fig. 6, the factor θu is defined as the lateral displacement dividing the distance from the lateral loading point to the top of the footing. As illustrated in Fig. 10, the factor μ in terms of the displacement was defined as:
μ=
Δ−u ⎞ 1 ⎛ Δ+u ⎜ + + ⎟ 2 ⎝ Δy Δ−y ⎠
(3)
where the ultimate displacement Δu is taken as the displacement when the remaining capacity drops to 75% of the maximum load. The yield displacement Δy in this paper is determined as the displacement when the steel flange yielded based on the data collected by the strain gauges. The values of Δy, Δu, θu and μ of the specimens were recorded in Table 5.
Fig. 13. Cumulative dissipated energy. 72
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
4. Seismic bending moment capacity 4.1. Basic assumptions According to the experimental results, it is noticeable that the PPSRC and HPSRC specimens both behaved as integral members, indicating that the specimens were sufficiently composite and the steel shape worked together well with the precast outer-part and cast-inplace inner-part. All the final failure modes of the specimens were dominant by flexure, as discussed in Section 3.1. Several researchers have proposed calculation models or methods to predict the flexural capacity of composite members [25,26]. In this paper, with the aim to simplify the calculation procedure, the plastic stress theory could be used here to determine the seismic bending moment capacity of the specimens. In the plastic stress theory, some simple yet rational assumptions can be listed as follows: (1) The plane section assumption was satisfied (2) The part of the steel shape and the longitudinal reinforcements at the tension zone were considered as tensed to yield, and the rest of the steel shape and the longitudinal reinforcements at the compression zone were considered as compressed to yield. (3) The tensile strength of the conventional concrete could be neglected, but that of the reactive powder concrete should be taken into account.
Fig. 14. Calculation method for equivalent viscous damping ratio.
specimens under greater axial compression or with lighter transverse reinforcement. In order to further analyze the energy dissipation capacity of the specimens, the equivalent viscous damping ratio is also employed here, as illustrated in Fig. 14. The equivalent viscous damping factor (ξeq) can be used to evaluate the pinching degree of the hysteretic loop. The definition of ξeq is expressed as:
ξeq =
1 S(DBC+DAC) 2π S(OBE + OAF)
4.2. Calculation methods Based on the above assumptions, a model for determining seismic bending moment capacity of PPSRC and HPSRC column is proposed here. In order to simplify the design procedure of reactive powder concrete members, some researchers [27,28] proposed the rectangle stress block method for reactive powder concrete, and it was employed here to calculate the contribution of precast outer-part. As shown in Fig. 16, based on the above analysis, the height of the concrete compression zone x and the seismic bending moment capacity M, which can be calculated by Eqs. (6)–(10), can be determined from the equilibrium conditions of internal force in the cross section.
(5)
Fig. 15 showed the variation of equivalent viscous damping ratio of the specimens. In general, equivalent viscous damping ratio of each test specimen increases gradually as the displacement increases. The ξeq of PPSRC specimens is higher than that of HPSRC specimens at the ultimate stage, indicating that the hysteretic loops of the PPSRC specimens are more robust. Under the same design parameters, the higher ξeq can be found in the specimens with abundant stirrups, and the strength of cast-in-place concrete has a limited effect on ξeq.
∑N = 0 ′ + fys Asf′ + Couter + Cinner−Touter−fy As −fys Asw −fys Asf N = fy As′ + fys Asw −fys (2Asf2 + Asw2 ) (6)
∑M = 0 ′ M = fy As′ (x −as) + fys Asf′ (x −ass) + fys Asw +
(x − a ) Cinner 2 ss
+
(h − x ) Touter 2
(x − ass) 2
(
( )( ) + N( )
+ fy As (h−as−x ) + fys Asw
+ fys Asf (h−ass−x ) + fys (2Asf2 + Asw2
βx
+ Couter x − 2
h −x 2
)
hw − x − ass 2
)
h −x 2
(7)
Couter = αfc,outer βxb−αfc,outer h w (x −ass)
(8)
Cinner = fc,inner h w (x −ass)
(9)
Touter = kft b (h−x )−h w (h−ass−x )
(10)
where x is the height of concrete compression zone; h is the column height; b is the column width; hw is the height of steel web; as is the thickness of cover concrete; ass is the distance from the steel flange to the column edge; As and As′ are the areas of tensile longitudinal rebar and compressive longitudinal rebar, respectively; Asw and Asw′ are the areas of tensile steel web and compressive steel web, respectively; Asf and Asf′ are the areas of tensile steel flange and compressive steel flange, respectively; Asf2 and Asw2 are the areas of steel flange and steel
Fig. 15. Equivalent viscous damping ratio. 73
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Fig. 16. Calculation diagram of normal section at the ultimate limit state.
columns can be much higher than those of the specimens in this paper, therefore, the deformability, ductility ratios and energy dissipation capacity can be significantly enhanced [23], indicating that large-scale HPSRC columns might exhibit better hysteretic behaviors. Therefore, more large-scale tests should be conducted in the future.
web perpendicular to the loading direction; fc,outer is the compressive strength of reactive powder concrete; fc, inner is the compressive strength of cast-in-place concrete; fys and fy are the yield strengths of steel shape and longitudinal rebar; α and β are factors of rectangle stress block method of RPC members, and k is the factor to determine the contribution of tensile strength of RPC to the bearing capacity, according to references [27,28], α = 0.9, β = 0.77 and k = 0.25 here. Based on the equations established above, the seismic bending moment capacities of all the specimens were calculated. All the calculated results were listed and plotted in Table 5 and Fig. 17, respectively. As can be seen, the calculated values agreed reasonably well with the test results, and an average ratio of 0.90 and a coefficient of variation of 0.04 were observed here.
5. Conclusions The seismic behaviors of partially precast steel reinforced concrete (PPSRC) columns and hollow precast steel reinforced concrete (HPSRC) columns have been investigated in this paper. A total of 10 column specimens were tested under combined axial compression and lateral cyclic load. The main conclusions drawn from this paper are as follows: (1) Two typical failure modes were observed in the specimens, namely flexural and flexural-shear failure. The flexural failure occurred in the specimens with abundant stirrups and the specimens with highstrength solid column core, and the other specimens all failed in flexural-shear mode. The specimens of PPSRC series exhibited better behaviors than those of HPSRC series both in ductility, strength degradation and energy dissipation, indicating that the solid column core played an important role in seismic behavior of the whole member. In the same series, higher stirrup ratio and lower axial compression of the column leaded to more satisfactory energy dissipation capacity, stiffness degradation and higher ductility. In the PPSRC specimens, the effect of strength of cast-in-place concrete was insignificant on the hysteretic behavior. (2) Due to the high-strength bolts and tear plates, the steel shape, precast concrete and cast-in-place concrete of both the PPSRC specimens and HPSRC specimens bonded well during the entire test process. Although some bonding cracks were found at the end of the test, the final failure was not dominant by bonding invalidation. Attribute to the steel fibers in precast shells, the bottom concrete covers of all the specimens bulged but did not spall during the test. The steel shape and longitudinal rebar were well confined by the continuous rectangle stirrups and surrounding concrete, so no obvious buckling was found in steel reinforcements. (3) Based on the plastic stress theory, a set of calculation formulas for determining seismic bending moment capacity of the PPSRC and HPSRC columns were proposed here. The results obtained form the formulas agreed well with those from the experiments. From the calculated axial capacity-moment capacity interaction curves, it can be seen that the seismic bending moment capacity of the HPSRC column was close to that of the PPSRC column in the low level of axial compression, indicating that when a structural column is subjected to low compressive load, such as columns of upper floors, the HPSRC column without inner concrete can be applied as a permanent column.
4.3. Axial capacity-moment capacity interaction curves The axial capacity - moment capacity interaction curves could be obtained by the similar procedures as the former section discussed. In Fig. 17, the load-carrying capacity of the hollow section (HPSRC) was compared with that of the solid section (PPSRC). As shown in this figure, the seismic bending moment capacity of the HPSRC column was close to that of the PPSRC column in the low level of axial compression. This result indicated that when a structural column is subjected to low compressive load, such as columns of upper floors, the HPSRC column without inner concrete can be applied as a permanent column. Because of the minimum thickness requirement of the concrete cover of steel shapes, the height of the steel web is limited here, which leads to a relatively low steel ratio. In the practical applications where full-scale columns are applied, the steel ratios of the PPSRC and HPSRC
Fig. 17. Test results and N-M interaction curves. 74
Engineering Structures 175 (2018) 63–75
Y. Yang et al.
Acknowledgment [14]
The experiments are sponsored by the National Natural Science Foundation of China (Grant No. 51578443, 51778525) and National Key Research and Development Program of China (Program No. 2017YFC0703404). The supports are greatly appreciated.
[15] [16]
References
[17]
[1] Hajjar JF. Composite steel and concrete structural systems for seismic engineering. J Constr Steel Res 2002;58(5):703–23. [2] Johnson RP. Composite structures of steel and concrete: beams, slabs, columns, and frames for buildings. Blackwell Pub; 2004. [3] Griffis LG. Some design considerations for composite-frame structures. Eng J 1986;23(2):59–64. [4] Foraboschi P. Versatility of steel in correcting construction deficiencies and in seismic retrofitting of RC buildings. J Build Eng 2016;8:107–22. [5] El-Tawil S, Deierlein GG. Strength and ductility of concrete encased composite columns. J Struct Eng 1999;125(9):1009–19. [6] Chen C, Wang C, Sun H. Experimental study on seismic behavior of full encased steel-concrete composite columns. J Struct Eng 2014;140(6):1–10. [7] Xue J. Design principles of steel-concrete composite structures. China Architecture and Building Press; 2012 [in Chinese]. [8] Tam VWY, Tam CM, Zeng SX, et al. Towards adoption of prefabrication in construction. Build Environ 2007;42(10):3642–54. [9] Park R. Seismic design and construction of precast concrete buildings in New Zealand. PCI J 2002;47(5):60–75. [10] Zhu B. Introduction of the prefabricated structures. China Architecture and Building Press; 2012. [in Chinese]. [11] Hong WK, Park SC, Kim JM, et al. Composite beam composed of steel and precast concrete (Modularized Hybrid System, MHS). Part I: Experimental investigation. Struct Des Tall Special Build 2008;19(3):275–89. [12] Hong WK, Kim JM, Park SC, et al. Composite beam composed of steel and precast concrete. (Modularized Hybrid System, MHS) Part II: Analytical investigation. Struct Des Tall Special Build 2009;18(8):891–905. [13] Hong WK, Park SC, Lee HC, et al. Composite beam composed of steel and precast
[18] [19] [20] [21] [22]
[23]
[24]
[25] [26] [27] [28]
75
concrete (modularized hybrid system). Part III: Application for a 19-storey building. Struct Des Tall Special Build 2009;19(6):679–706. Hong WK, Kim SI, Park SC, et al. Composite beam composed of steel and precast concrete (modularized hybrid system). Part IV: Application for multi-residential housing. Struct Des Tall Special Build 2010;19(7):707–27. Yang Y, Xue Y, Yu Y, et al. Experimental study on flexural performance of partially precast steel reinforced concrete beams. J Constr Steel Res 2017;133:192–201. Yang Y, Yu Y, Guo Y, et al. Experimental study on shear performance of partially precast Castellated Steel Reinforced Concrete (CPSRC) beams. Steel Compos Struct 2016;21(2):289–302. Xu S, Wu C, Liu Z, et al. Experimental investigation of seismic behavior of ultra-high performance steel fiber reinforced concrete columns. Eng Struct 2017;152:129–48. Zheng W, Lv X. Literature review of reactive powder concrete. J Build Struct 2015;36(10):45–58. (in Chinese). Foraboschi P. Analytical model to predict the lifetime of concrete members externally reinforced with FRP. Theor Appl Fract Mech 2015;75(1):137–45. Focacci F, Foraboschi P, Stefano MD. Composite beam generally connected: analytical model. Compos Struct 2015;133:1237–48. Foraboschi P. Layered plate with discontinuous connection: exact mathematical model. Compos B 2013;47(3):365–78. Huang L, Xu L, Chi Y, Xu H. Experimental investigation on the seismic performance of steel-polypropylene hybrid fiber reinforced concrete columns. Constr Build Mater 2015;87:16–27. Zhu W, Jia J, Gao J, et al. Experimental study on steel reinforced high-strength concrete columns under cyclic lateral force and constant axial load. Eng Struct 2016;125:191–204. Han LH, Li W, Bjorhovde R. Developments and advanced applications of concretefilled steel tubular (CFST) structures: Members. J Constr Steel Res 2014;100:211–28. Foraboschi P. Three-layered plate: elasticity solution. Compos B 2014;60(1):764–76. Eurocode 4: design of composite steel and concrete structures, part 1-1, general rules and rues for building. European Committee for Standardization; 2004. Li L. Mechanical behavior and design method for reactive powder concrete beams [Ph.D. Dissertation]. Harbin Institute of Technology; 2010 [in Chinese]. Zheng W, Li L, Lu S. Experimental research on mechanical performance of normal section of reinforced reactive powder concrete. J Build Struct 2011;32(6):125–34. [in Chinese].