- Email: [email protected]
Circular concrete filled steel tubular (CFST) columns under cyclic load and acid rain attack: Test simulation
Thin-Walled Structures 122 (2018) 90–101
Contents lists available at ScienceDirect
Thin-Walled Structures journal homepage: www.elsevier.com/locate/tws
Full length article
Circular concrete filled steel tubular (CFST) columns under cyclic load and acid rain attack: Test simulation
T
⁎
Fang Yuan, Mengcheng Chen , Hong Huang, Li Xie, Chao Wang Dept. of Civil Engineering and Architecture, East China Jiaotong Univ., Nanchang 330013, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Concrete filled steel tubular (CFST) columns Cyclic load Acid rain attack Seismic performance Test simulation
Concrete filled steel tubular (CFST) structure attracts increasing engineering applications in earthquake prone regions due to its high section modulus, high strength, and good seismic performance. However, the seismic resistance of CFST columns may be affected by the environmental corrosions, such as acid rain attack. This paper makes an attempt to investigate the performance of CSFT columns with circular sections under both a cyclic load and an acid rain attack. First, the tensile mechanical properties of steel plates with various corrosion rates were tested. Second, a total of 12 columns with different corrosion rates were tested subjected to a reversed cyclic load. It was found that the corrosion leads to not only a loss in wall thickness but also an evident decrease in yield strength, elastic modulus, and tensile strain capacity of the steel coupons, and also to a significant deterioration in the load carrying capacity, ductility, and energy dissipation of the CFST columns. The larger the axial force ratio, the severer deterioration of deformation capacity of the columns.
1. Introduction Concrete filled steel tube (CFST) has an increasing utilization in the earthquake prone regions in China due to its high strength, good ductility, and excellent energy dissipation capacity [1]. The outer steel tube of CFST member is exposed to external environment and is prone to suffer environmental corrosions during the service life span, such as acid rain attack. Worldwide acid rain problems have been worsened by industrial and urban developments and acid rainfall has been reported to cover at least one third of Chinese territory [2–5]. Thus, it is essential to evaluate the seismic behaviors of CFST members that have suffered acid rain corrosion. In the past few decades, a great number of studies have been carried out on the seismic behaviors of CFST members [6–13]. Some literature reviews had been conducted by Nakanishi et al. [12] and Elremaily and Azizinamini [13]. It made a consensus that the CFST members exhibit much higher ductility compared with the hollow steel tubes owing to the composite effect between the core concrete and outer steel tube. Han et al. [14] also tested the cyclic behaviors of concrete filled double skin steel tubular (CFDST) members under combined axial and flexural load. It was reported that the CFDST members show good ductility and excellent energy dissipation capacity even under high levels of axial force ratio above 0.6. Experimental studies on steel structures and CFST members under corrosive environment have also been conducted in recent years. For
⁎
Corresponding author. E-mail address: [email protected] (M. Chen).
http://dx.doi.org/10.1016/j.tws.2017.10.005 Received 3 May 2017; Received in revised form 2 September 2017; Accepted 2 October 2017 Available online 13 November 2017 0263-8231/ © 2017 Elsevier Ltd. All rights reserved.
example, Almusallam [15] studied the effect of sodium chloride corrosion on the properties of reinforcing steel bars and found that reinforcing steel bars with more than 12% corrosion indicates a brittle failure. Qin and Cui [16] studied the effect of corrosion models on the time-dependent reliability of steel plate elements and the advantages and the flexibility of the proposed corrosion model were demonstrated. Melchers [17] studied the influential factors on the corrosion rate of steel in seawater environments. Saad-Eldeen et al. [18] tested the load carrying capacity of a corroded steel box girder. Sultana et al. [19] studied the compressive strength of stiffened panels under pitted corrosion. Karagah et al. [20] tested the steel columns under corrosion and axial compression. Han et al. [21,22] and Hou et al. [23] carried out the experimental studies on 22 beams and 34 stub columns under sustained load and chloride corrosion. The test results showed that the chloride corrosion has great effects on the load carrying capacity of construction steel and CFST members. Simplified calculation methods for the load carrying capacity of CFST beams and stub columns were also proposed based on parametric studies [24]. Previous studies have focused on the static behaviors of corroded CFST members. Few experimental works have been studied the seismic behaviors of corroded CFST members, especially under acid rain attack. This gives rise to the need for more studies of the problem. This paper aims to investigate the seismic behaviors of circular CFST members subjected to acid rain corrosion. The effect of corrosion on the mechanical behaviors of steel tubes is firstly tested and discussed. After
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Nomenclature
N0 Nu P Pm Pu Py ts ts0 η ξ μ εu0 εu Δ Δy Δm Δu
The following symbols are used in this paper Ac As D Es Es0 fck fcu fy fy0 fu fu0 n L
cross-sectional area of core concrete cross-sectional area of steel tube after corrosion column diameter elastic modulus of corroded steel elastic modulus of uncorroded steel characteristic compressive strength of concrete compressive strength of cube concrete yield strength of corroded steel yield strength of uncorroded steel ultimate strength of corroded steel ultimate strength of uncorroded steel axial force ratio column length
axial force applied on the columns axial compressive capacity of the columns lateral load of column peak load of column ultimate load of column yield load of column wall thickness of steel tube after corrosion initial wall thickness of steel tube corrosion rate confinement factor ductility coefficient ultimate elongation of corroded steel ultimate elongation of corroded steel lateral displacement of column yield displacement of column displacement of column at peak load ultimate displacement of column
where N0 is the applied axial force on the specimens, and Nu is the axial load carrying capacity of the columns, which is calculated by the simplified formulas described in [25]. The corrosion rate is defined as:
that, the effects of the corrosion rate and axial force ratio on the seismic behaviors of CFST columns, such as ultimate strength, ductility, energy dissipation ability, et al., are experimentally studied and systematically evaluated.
η= 2. Experimental program
In total, 12 circular column specimens were tested in the present work. All tested specimens have a sectional size of D × ts = 1500 × 114 × 4 mm and a length (L) of 1500 mm, where D is the diameter of outer steel tube sand ts is the wall thickness of steel tube. The CFST specimens were fabricated through the following steps. The steel tubes were segmented from an industrial steel tube. A steel plate was welded into one end of the steel tube. Then, the steel tubes were placed upright for casting. The concrete was cast into the steel tube and vibrated by a poker at the same time. After curing, a small gap between concrete surface and top steel tube was observed due to concrete shrinkage. The longitudinal gap was filled with a high strength epoxy in order to make the concrete surface flush with the top steel tube. Another steel plate was welded onto the top end of the steel tube before testing. The main design parameters were an axial force ratio (n) from 0 to 0.5 and a corrosion rate (η) from 0% to 30%. The axial force ratio herein is defined as:
N0 Nu
(2)
where ts0 is the initial thickness of the steel plate; ts is the remaining thickness after corrosion. The designed corrosion rates are 0, 10%, 20% and 30% respectively. Table 1 shows a summary of the tested specimens, where ξ represents the confinement factor to account for the ‘composite action’ between the steel tube and core concrete, and was defined as follows [26]:
2.1. Preparation of specimens
n=
ts0 − ts × 100% ts0
ξ=
As f y Ac fck
(3)
where As is the cross-sectional area of steel tube after corrosion, fy is the yield strength of steel, Ac is the cross-sectional area of the core concrete, fck is the characteristic compressive strength of concrete. The value of fck is calculated to be 67% of the cube strength of concrete (fcu). The following naming rules are employed to distinguish specimens: 1) the two initial characters ‘CC’ represents the circular column section; 2) the Arabic numerals before hyphen stand for the axial force ratio; 3) the Arabic numerals after hyphen represent the corrosion rate. For example, the specimen ‘CC0.2-10’ stands for the circular column with designed axial force ratio of 0.2 and corrosion rate of 10%.
(1)
Table 1 Information of column specimens. No.
Specimen ID
D (mm)
ts (mm)
L (mm)
fcu (MPa)
n
η (%)
ξ
Yield
Peak
Ultimate
Py (kN)
Δy (mm)
Pm (kN)
Δm (mm)
Pu (kN)
Δu (mm)
1 2 3 4
CC0.2-0 CC0.2-10 CC0.2-20 CC0.2-30
114 114 114 114
4.00 3.62 3.23 2.84
1500 1500 1500 1500
60 60 60 60
0.2 0.2 0.2 0.2
0 9.48 19.25 29.00
1.49 1.18 0.95 0.85
56.4 52.3 45.8 40.7
16.18 15.78 15.67 14.89
73.55 69.40 58.80 51.50
35.85 33.25 29.85 29.50
62.69 58.99 51.00 44.12
– 44.21 37.10 34.04
5 6 7 8
CC0.4-0 CC0.4-10 CC0.4-20 CC0.4-30
114 114 114 114
4.00 3.63 3.21 2.81
1500 1500 1500 1500
60 60 60 60
0.4 0.4 0.4 0.4
0 9.25 19.75 29.75
1.49 1.19 0.94 0.84
53.1 46.3 40.2 34.8
12.82 11.15 10.37 8.85
73.75 65.80 56.85 46.70
33.95 33.55 23.55 17.90
62.69 56.49 48.33 39.70
60.75 36.64 33.23 21.60
9 10 11 12
CC0.5-0 CC0.5-10 CC0.5-20 CC0.5-30
114 114 114 114
4.00 3.60 3.19 2.78
1500 1500 1500 1500
60 60 60 60
0.5 0.5 0.5 0.5
0 10.00 20.25 30.50
1.49 1.18 0.94 0.83
52.6 49.3 46.1 39.3
11.44 12.42 11.60 10.11
70.35 62.65 57.70 48.85
31.50 27.50 24.40 23.45
58.36 53.25 49.05 41.52
39.42 38.66 34.48 25.73
91
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
2.2. Corrosion test configuration
Table 2 Mixture proportions of concrete.
The column specimen test was divided into two stages. In the first stage it is accelerated corrosion test and in the second stage it is ultimate strength test. The accelerated corrosion setup was composed of solution tank, stainless steel plate, power source, and electrolytic solution. The corrosive environment was simulated with acid rain solution and applied DC current. The simulated acid rain solution was made up of Na2SO4, (NH4)2SO4, MgSO4, Ca(NO3)2 and HNO3 et al. in terms of the components of acid rain in southern China [27]. The main cations include H+, NH4+, Mg2+ and Ca2+ while the main anions include SO42- and NO3-. The corrosion of constructional steel is mainly caused by H+ and SO42-. The pH value of the solution was determined by HNO3 and was fixed to be 2.3. In order to make path for the corrosion current, the anode of the power source was connected to the specimens while the cathode of the power source was connected to the stainless steel plate, as shown in Fig. 1. Antirust was applied on the two ends of the CFST specimens. As a result, only the outer surface of the steel tubes corroded (Fig. 1). Besides CFST specimen tests, four groups of steel coupons with corrosion rates from 0% to 30% were also tested by the accelerated corrosion, and tensioned to monitor the deterioration of tensile properties of the steel. The steel coupons were fabricated and tested in accordance with the Chinese standards with respect to the metal materials used. Three parallel coupons were prepared corresponding to each corrosion rate. Only the surface perpendicular to the normal direction of the steel coupons (as shown in Fig. 4) was exposed to the corrosive environment and other surfaces were covered with antirust in order to simulate the practical corrosion condition of the steel tubes. The corrosion rate was controlled by adjusting current intensity and immersed time. To fix the current intensity and immersed time at each of designed corrosion rate, steel coupons were first tested before CFST specimen tests. The thickness loss of steel plate was monitored carefully during the corrosion process. The surfaces of the steel coupons and the CFST columns were cleaned after corrosion. Two kinds of approaches were adopted to measure the corrosion rate: one is to weight the tested specimens and the other is to measure the average thickness of the steel coupons or the diameter of the steel plate. It was found that the thickness losses (Δts) measured by these two methods showed high consistency.
Matrix
Cement
water
Sand
Coarse aggregate
Concrete
1.0
0.4
1.1
2.56
properties are discussed in the following sections. One type of concrete with the mixture proportions shown in Table 2 was designed. The measured cube strength (fcu) at 28 days is 60 MPa. 2.4. Cyclic loading configuration The CFST specimens were tested under a constant axial force and increasing cyclic displacements. The schematic cyclic loading test setup is shown in Fig. 2. The two ends of the columns were attached to the pin supports, which were free to rotate to simulate the pin-pin end conditions. The axial force was first applied through a hydraulic jack installed between one end of the column and a fixed concrete block. Then, the reversed cyclic load was applied via a MTS hydraulic ram fixed on a reaction frame with ultra-high flexural stiffness. The lateral loading history was made up of elastic and inelastic cycles. The elastic cycles were conducted by load cycles at load intervals of 0.25Py, where Py is the estimated lateral yield load corresponding to the lateral yield displacement Δy. After the yield point, the inelastic cycles were controlled by displacement cycles at intervals of Δy. Three cycles were imposed at each inelastic displacement level. The loading history is shown in Fig. 3. The test was terminated when the ultimate state was reached where the lateral load resistance drops to 85% of its peak value. The instrumentation of the test included the measurements of lateral load and displacement at the loading point of the specimen, horizontal load of the hydraulic jack, horizontal displacement of the two ends of the concrete blocks, and the strains in the steel tubes. The loads were recorded by load cells (LC1 and LC2) and the displacements were measured with the linear voltage digital transducers (LVDTs, A, B, and C). Strain gauges were attached to the surface of the steel tubes near the base of each column to monitor the strain variations during test. The detailed measurement arrangements are clearly shown in Fig. 2. 3. Results and discussions
2.3. Material properties
3.1. Effects of corrosion on the mechanical properties of steel
The measured yield strength, ultimate strength and elastic modulus of the uncorroded steel coupons are 383 MPa, 516 MPa and 187 GPa, respectively. A tensile test of the corroded steel coupons with different thickness losses was also carried out and the detailed mechanical
3.1.1. Failure modes of steel coupons The steel coupons with various specified corrosion rates were tested in tension. Fig. 4 shows the failure modes of the specimens. It is clearly seen that apparent necking is observed around the fracture section of
Fig. 1. Accelerated corrosion test setup.
92
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 2. Schematic cyclic loading test setup.
Fig. 3. Cyclic loading history.
uncorroded steel coupons and the fracture section is almost perpendicular to the tensile load direction. However, for the corroded steel coupons, tensile fracture occurred suddenly without necking phenomenon and the directions of the fracture sections were disordered. A
Fig. 5. Effect of corrosion rate (η) on the stress-strain response of steel.
Fig. 4. Failure modes of steel coupons.
93
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Table 3 Mechanical properties of uncorroded and corroded steel coupons. No.
η (%)
Yield strength fy (MPa)
Ultimate strength fu (MPa)
Elastic modulus Es (GPa)
Ultimate elongation εu (%)
Es loss (%)
fu loss (%)
fy loss (%)
εu loss (%)
1 2 3 Average
0 0 0 0
370.17 368.92 410.10 383.06
516.11 505.41 526.88 516.13
190.38 188.96 181.60 187.98
36.04 36.20 36.20 36.15
– – – –
– – – –
– – – –
– – – –
4 5 6 Average
9.89 10.44 11.54 10.62
356.71 333.59 319.88 336.73
448.32 442.18 452.64 447.71
174.80 158.44 176.45 169.90
35.34 29.18 28.80 31.11
– – – 9.62
– – – 13.26
– – – 12.09
– – – 13.94
7 8 9 Average
21.70 23.35 24.45 23.17
270.35 308.24 331.27 303.29
413.68 411.83 427.09 417.53
156.00 136.00 151.51 147.84
26.64 25.98 23.64 25.42
– – – 21.36
– – – 19.10
– – – 20.83
– – – 25.42
10 11 12 Average
27.47 27.75 35.44 30.22
304.36 300.54 322.13 309.01
395.27 409.86 418.72 407.95
140 125.3 138.52 134.61
18.88 15.52 13.34 15.91
– – – 28.39
– – – 20.96
– – – 19.33
– – – 55.98
Fig. 6. Effect of corrosion rate (η) on the mechanical properties of steel.
crack will appear from the largest pit and develop along the thinnest section, leading to an irregular fracture section.
careful observation of the corroded steel coupons showed that a large amount of small corrosion pits formed at the surface of the specimens even though the uniform corrosion was designed and applied for all specimens. The higher the degree of the corrosion, the more significant nonuniformity of the local corrosions. Under the tensile load, an initial
3.1.2. Typical tensile stress-strain curves of the corroded steel coupons The typical tensile stress-strain curves of the steel coupons with 94
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 7. Failure patterns of specimens.
εu are observed as the corrosion rate (η) increases. It indicates that the acid rain corrosion on steel tubes results in not only a loss in wall thickness of the steel tube but also a reduction of the tensile strength and the deformation. However, only the effective loss of thickness of a steel tube has been considered for the derivation of simplified strength models of corroded CFST members in most of existing researches [21–24]. As pointed out by Han et al. [26], the confinement factor (ξ) plays an important role in the mechanical behaviors of CFST columns, stub columns, and beam-columns. It can be inferred from Eq. (3) that not only the cross-sectional area of a steel tube (As) that is dependent on the wall thickness of steel tube (ts) but also the yield strength of the steel has a positive effect on ξ. Therefore, the ignorance of the effect of environmental corrosion on the yield strength of a steel tube will result in non-negligible errors in the prediction of the load carrying capacity of CFST members. In the present work, the variations of mechanical properties of steel with corrosion rate (η) are quantitatively analyzed. By analyzing the variations of Es, fy, fu, and εu, a linear relationship between these factors and η is approximated. Thus, the following models of Es, fy, fu, and εu are obtained through the regression analyses:
Fig. 8. Typical failure modes of core concrete in CFST.
different corrosion rates are shown together in Fig. 5. The tensile stress herein is defined as the ratio of the tensile load to that of remaining cross-sectional area of steel coupons after corrosion. It is clearly seen that both the yield strength and ultimate strength of steel coupons without corrosion are significantly larger than those of corroded steel coupons. It can be also found from Fig. 5 that the yield stage becomes shorter and fracture occurs earlier and earlier as the corrosion rate increases, implying that the fracture process of the steel coupons behaves more brittle with the increasing of corrosion rate. It is attributed to the less significant necking phenomenon of the corroded steel coupons during the tensile failure process, which results in more brittle tensile process and lower fracture strain.
Es = (1 − 0.955η) E0
(4)
f y = (1 − 1.007η) f y0
(5)
fu = (1 − 0.748η) fu0
(6)
εu = (1 − 1.450η) εu0
(7)
where Es, fy, fu, and εu are elastic modulus, yield strength, ultimate strength, and ultimate elongation of corroded specimens, respectively. Es0, fy0, fu0, and εu0 are the elastic modulus, yield strength, ultimate strength, and ultimate elongation of uncorroded specimens, respectively. The ratios of Es, fy, fu, and εu estimated by Eqs. (4)–(7) to the measured values have a mean value of 0.998, 0.965, 1.016, and 1.018, respectively, and a standard deviation (SD) of 5.77%, 9.05%, 4.41%, and 16.48%, respectively. Thus, a good agreement between the proposed models and the test results is obtained. Since the uniform acid
3.1.3. Mechanical properties of corroded steel The mechanical properties, such as elastic modulus (Es), yielding strength (fy), ultimate strength (fu), and ultimate elongation (εu) of the steel coupons are shown in Table 3. Fig. 6 shows the effect of corrosion rate on these tensile properties. Clear decreasing trends of Es, fy, fu, and 95
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 9. Cyclic load-displacement curves of each specimen.
bulge also occurred at the opposite face of the specimen as the lateral cyclic load was reversed. With increasing the lateral displacement, a complete ring formed near the loading section at midspan. Finally, some columns failed due to the tensile fracturing at the bulge location, accompanied by rapid drop in the lateral load. The failure of other specimens without tensile failure was determined by the 15% drop of the lateral load resistance. It can be also found from Fig. 7 that the failure patterns of the corroded CFST specimens were similar to those of the uncorroded specimens, which means that corrosion has nearly no effect on the failure patterns of the circular CFST members. The similar observation
rain corrosion was applied for steel coupons in the present work, Eqs. (4)–(7) are only applicable for the steel material under uniform acid rain attack. 3.2. Failure characteristics of CFST members The failure patterns of the CFST columns are shown in Fig. 7. It can be found that the CFST specimens deformed in a quite ductile way. No sign of local deformation was observed before the yielding initiation of the specimens. Beyond the yielding, an outward buckling was observed close to loading point at the compressive face of the specimen. The 96
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 9. (continued)
3.3. Lateral load (P) versus lateral displacement (Δ) curves
has been reported for CFST members under sustained static load and chloride corrosion tested by Han et al. [22] and Hou et al. [23]. Fig. 8 shows the failure patterns of the core concrete after the test. It can be found that, under reversed tension and compression, an evident sign of cracking and crushing was observed in the vicinity of the column base. Because the maximum moment occurred at the column base, this test observation is quite reasonable for columns under reversed lateral load.
Fig. 9 shows the lateral load versus the mid-span displacement curves (which are also called hysteresis curves) for all specimens. It is easily found that the hysteresis loops of the uncorroded columns are more stable and significantly plumper than those of the corroded columns. The hysteresis loop becomes increasingly smaller as the corrosion rate increases, indicating a lower energy dissipated capacity for column with severer corrosion. This phenomenon can be explained by the following factors. First, the wall thickness of the outer steel tube decreases as the corrosion rate increases, which leads to a reduction of 97
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 10. Effect of corrosion rate (n) on the load-displacement envelop curves of specimens.
Fig. 11. Effect of axial force ratio (n) on the global response of columns.
98
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
an important role in the energy dissipation of steel-concrete composite members. The pinching effect of the hysteresis loops would be more apparent for CFST members with a lower steel ratio. Third, as discussed above, corrosion also leads to the reduction in yield strength of a steel tube, which further reduces the confinement factor (ξ), as indicated in Eq. (3). The confinement effect of the steel tube on the core concrete becomes weaker and the local buckling of steel tube is more prone to occur as ξ decreases, resulting in lower load carrying and deformation capacities, and thus smaller hysteresis loops of the CFST members. Fig. 10 shows the load versus displacement envelop curves of all specimens. The curves of the corroded specimens tend to drop more quickly than the uncorroded specimens due to a premature buckling of the steel tubes. The load carrying capacity decreases and the area encompassed by the curve decreases as the corrosion rate increases at each axial force level. For example, the peak load (Pk) and ultimate displacement (Δu) of the CFST columns with an axial force ratio of 0.4 are reduced by 36.7% and 64.4% as the corrosion rate changes from 0% to 29.75%, respectively, where the ultimate displacement herein is defined as the point where the lateral load resistance drops to 85% of its peak value. This definition of ultimate state was also usually employed by other researchers [28–31]. The reductions of Pk and Δu are attributed to the losses of the wall thickness and yield strength of steel tubes caused by corrosion. The detailed test data for each specimen are shown together in Table 1. The ultimate displacement of specimen CC0.2-0
Fig. 12. Effect of corrosion rate (η) on the ductility of specimens.
the tensile load provided by the steel and a composite action between the outer steel tube and core concrete. Second, as is well-known, steel is a kind of material with excellent plasticity and the outer steel tube plays
Fig. 13. Effect of corrosion rate (η) on the cumulative dissipated energy of specimens.
99
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
modulus Es, yield strength fy, ultimate strength fu, and ultimate elongation εu. A linear relationship between Es, fy, fu and εu of steel coupons and corrosion rate is found and proposed. As a result, the loss of strength and deformation capacities of outer steel tubes caused by corrosion should be taken into consideration in the prediction of the load carrying capacity of corroded CFST members. 2. The failure patterns of uncorroded CFST specimens and corroded specimens are similar to each other. However, the seismic performance of uncorroded CFST columns is evidently superior to that of corroded columns, such as higher load carrying capacity, better ductility and larger energy dissipation capacity. The higher the corrosion rate, the poorer the seismic performance of the CFST specimens. 3. The axial force level has nearly no effect on the ultimate lateral load but an adverse influence on the ultimate displacement of the specimens.
was unfortunately not obtained from the test due to exceeding the maximum range of the LVDT. Fig. 11 shows the effects of the axial force ratio on the peak load and ultimate displacement of the specimens. The four points at each axial force ratio correspond to the values at four corrosion rates, respectively. It can be seen that the axial force level has nearly no effect on the load carrying capacity but great influence on the ultimate displacement of the specimens. This phenomenon is quite often observed in reinforced concrete columns. Generally, the deformation capacity of the columns decreases as the axial force level increases. For CFST columns with a higher axial force level, the compression face tends to buckle more quickly due to the larger compression load. As a result, the lateral load tends to drop more rapidly at a relatively smaller lateral deformation. 3.4. Ductility and energy dissipation The ductility coefficient (μ) is a key parameter for evaluating the ductility performance of the composite members. The ductility coefficient herein is defined as the ratio of the ultimate displacement to the yield displacement of CFST columns (Δu/Δy). The effect of the axial force ratio on the ductility coefficient μ for each specimen is presented in Fig. 12. An evident decreasing trend of the ductility of the CFST columns at each axial force level is observed as the corrosion rate increases. Both the yield displacement and the ultimate displacement of the columns decrease with increasing η due to the reduction of the wall thickness ts and confinement factor ξ caused by corrosion, as shown in Table 1. The deterioration of the ultimate displacement is more substantial than that of the yield displacement, resulting in a lower ratio of Δu/Δy and thus a smaller μ for a larger η. The buckling of the steel tube and the corresponding ultimate state occur earlier for the column with a smaller ξ due to the weaker composite effect between the steel tube and core concrete. Fig. 13 shows the cumulative dissipated energy at each displacement level for each specimen. In the present work, the cumulative dissipated energy is defined as the sum of the areas of each hysteresis loop up to the specific displacement level (Δu/Δy). Since all specimens deform in a ductile manner, the cumulative dissipated energy shows a steady increase until failure for each specimen. It can be also found from Fig. 13 that the energy dissipation ability exhibits an evident decreasing trend with the increase of the corrosion rate. The cumulative dissipated energy levels are close to each other before the yield point at each axial force level. After the yield point, however, the gap in the cumulative energy between the corroded specimens and uncorroded specimens becomes increasingly larger as the displacement level increases. It indicates that the composite action between the outer steel tube and core concrete plays an important role in the energy dissipation capacity. As mentioned above, the confinement factor ξ decreases substantially due to the reduction caused by corrosion of the wall thickness as well as yield strength for the steel tubes. As ξ decreases, the compressive strength of the core concrete is reduced and the buckling of the outer steel tube occurs earlier, both of which have adverse effect on the energy dissipation capacity of CFST columns.
It is noted that only uniform corrosion on the outer surface of CFST columns was taken into consideration in the present work. While the practical corrosion situations, such as localized corrosion, has not been taken into accounted in the present experimental study. This topic will be discussed in near future ongoing researches. Acknowledgement Financial supports of the work by National Natural Science Foundation of China under 51378206 and 51608199, and in part by Natural Science Foundation of Jiangxi Province under 20161BAB216140 as well as the program for Advanced Science and Technology Innovation Team of Jiangxi Province under 20152BCB24006, are gratefully acknowledged. References [1] L.H. Han, Some recent developments of concrete filled steel tubular (CFST) structures in China, in: Proceedings of the 4th International Conference on Steel & Composite Structures, 2010. [2] A.B. Eney, D.E. Petzold, The problem of acid rain: an overview, Environ. Syst. Decis. 7 (7) (1987) 95–103. [3] F. Mansfeld, R. Vijayakumar, Atmospheric corrosion behavior in Southern California, Corros. Sci. 28 (9) (1988) 939–946. [4] T. Larssen, H.M. Seip, A. Semb, J. Mulder, et al., Acid deposition and its effects in China: an overview, Environ. Sci. Policy 2 (1) (1999) 9–24. [5] J. Hill, E.A. Byars, J.H. Sharp, et al., An experimental study of combined acid and sulfate attack of concrete, Cem. Concr. Compos. 28 (8) (2003) 997–1003. [6] K. Sakino, M. Tomii, Hysteretic behavior of concrete filled square steel tubular beam-columns failed in flexure, Trans. Jpn. Concr. Inst. 3 (6) (1981) 439–446. [7] Y. Morishita, M. Tomii, Experimental studies on bond strength between square steel tube and encased concrete core under cyclic shearing force and constant axial force, Trans. Jpn. Concr. Inst. 4 (1982) 363–370. [8] H.B. Ge, T. Usami, Cyclic tests of concrete filled steel box columns, J. Struct. Eng. 122 (10) (1996) 1169–1177. [9] P.F. Boyd, W.F. Cofer, D.I. Mclean, Seismic performance of steel-encased concrete columns under flexural loading, ACI Struct. J. 92 (3) (1995) 355–364. [10] J.F. Hajjar, A. Molodan, P.H. Schiler, A distributed plasticity model for cyclic analysis of concrete-filled steel tube beam-columns and composite frames, Eng. Struct. 20 (4–6) (1998) 398–412. [11] K. Lahlou, M. Lachemi, P.C. Aitcin, Confined high-strength concrete under dynamic compressive loading, J. Struct. Eng. 125 (10) (1999) 1100–1108. [12] K. Nakanishi, T. Kitada, H. Nakai, Experimental study on ultimate strength and ductility of concrete filled steel columns under strong earthquake, J. Constr. Steel Res. 51 (3) (1999) 297–319. [13] A. Elremaily, A. Azizinamini, Behavior and strength of circular concrete-filled tube columns, J. Constr. Steel Res. 58 (12) (2002) 1567–1591. [14] L.H. Han, H. Huang, Z. Tao, X.L. Zhao, Concrete-filled double skin steel tubular (CFDST) beam–columns subjected to cyclic bending, Eng. Struct. 28 (12) (2006) 1698–1714. [15] A.A. Almusallam, Effect of degree of corrosion on the properties of reinforcing steel bars, Constr. Build. Mater. 15 (8) (2001) 361–368. [16] S. Qin, W. Cui, Effect of corrosion models on the time-dependent reliability of steel plated elements, Mar. Struct. 16 (1) (2003) 15–34. [17] R.E. Melchers, Recent progress in the modeling of corrosion of structural steel immersed in seawaters, J. Infrastruct. Syst. 12 (3) (2006) 154–162. [18] S. Saad-Eldeen, Y. Garbatov, C.G. Soares, Effect of corrosion severity on the ultimate strength of a steel box girder, Eng. Struct. 49 (2) (2013) 560–571.
4. Conclusions This paper has investigated the mechanical properties of CFST columns subjected to a cyclic load and an acid rain attack through tests. The tensile properties of steel coupons were first studied in detail. A reversed cyclic load test was subsequently carried out for both corroded and uncorroded CFST specimens. The effects of corrosion rate η and the axial force ratio were taken into consideration. In summary, the following conclusions can be drawn from the present paper: 1. Acid rain corrosion has not only a significant influence on the reduction in wall thickness of a steel tube, but also an adverse effect on the mechanical properties of the steel, such as the elastic 100
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al. [19] S. Sultana, Y. Wang, A.J. Sobey, J.A. Wharton, R.A. Shenoi, Influence of corrosion on the ultimate compressive strength of steel plates and stiffened panels, Thin Wall Struct. 96 (2015) 95–104. [20] H. Karagah, C. Shi, M. Dawood, A. Belarbi, Experimental investigation of short steel columns with localized corrosion, Thin Wall Struct. 87 (2015) 191–199. [21] L.H. Han, C. Hou, Q.L. Wang, Square concrete filled steel tubular (CFST) members under loading and chloride corrosion: experiments, J. Constr. Steel Res. 71 (2012) 11–25. [22] L.H. Han, C. Hou, Q.L. Wang, Behavior of circular CFST stub columns under sustained load and chloride corrosion, J. Constr. Steel Res. 103 (2014) 23–36. [23] C.C. Hou, L.H. Han, Q.L. Wang, C. Hou, Flexural behavior of circular concrete filled steel tubes (CFST) under sustained load and chloride corrosion, Thin Wall Struct. 107 (2016) 182–196. [24] C. Hou, L.H. Han, X.L. Zhao, Full-range analysis on square CFST stub columns and beams under loading and chloride corrosion, Thin Wall Struct. 68 (10) (2013) 50–64. [25] DBJ/T13-51-2010, Technical specification for concrete-filled steel tubular
[26] [27] [28] [29]
[30]
[31]
101
structures, The Construction Department of Fujian Province, Fuzhou, China, 2010. (in Chinese). L.H. Han, X.L. Zhao, Z. Tao, Tests and mechanics model for concrete-filled SHS stub columns, columns and beam-columns, Steel Compos. Struct. 1 (1) (2001) 51–74. M.C. Chen, K. Wang, L. Xie, Deterioration mechanism of cementitious materials under acid rain attack, Eng. Fail. Anal. 27 (1) (2013) 272–285. M.J.N. Priestley, R. Park, Strength and ductility of concrete bridge columns under seismic loading, Acids Struct. J. 84 (1) (1987) 61–76. G. Fischer, V.C. Li, Effect of matrix ductility on deformation behavior of steel-reinforced ECC flexural members under reversed cyclic loading conditions, ACI Struct. J. 99 (6) (2002) 781–790. Y.F. Wu, T. Liu, D.J. Oehlers, Fundamental principles that govern retrofitting of reinforced concrete columns by steel and FRP jacketing, Adv. Struct. Eng. 9 (4) (2006) 507–533. F. Yuan, J.L. Pan, L.T. Dong, C.K.Y. Leung, Mechanical behaviors of steel reinforced ECC or ECC/concrete composite beams under reversed cyclic loading, J. Mater. Civ. Eng. 26 (8) (2014) 04014047.
Contents lists available at ScienceDirect
Thin-Walled Structures journal homepage: www.elsevier.com/locate/tws
Full length article
Circular concrete filled steel tubular (CFST) columns under cyclic load and acid rain attack: Test simulation
T
⁎
Fang Yuan, Mengcheng Chen , Hong Huang, Li Xie, Chao Wang Dept. of Civil Engineering and Architecture, East China Jiaotong Univ., Nanchang 330013, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Concrete filled steel tubular (CFST) columns Cyclic load Acid rain attack Seismic performance Test simulation
Concrete filled steel tubular (CFST) structure attracts increasing engineering applications in earthquake prone regions due to its high section modulus, high strength, and good seismic performance. However, the seismic resistance of CFST columns may be affected by the environmental corrosions, such as acid rain attack. This paper makes an attempt to investigate the performance of CSFT columns with circular sections under both a cyclic load and an acid rain attack. First, the tensile mechanical properties of steel plates with various corrosion rates were tested. Second, a total of 12 columns with different corrosion rates were tested subjected to a reversed cyclic load. It was found that the corrosion leads to not only a loss in wall thickness but also an evident decrease in yield strength, elastic modulus, and tensile strain capacity of the steel coupons, and also to a significant deterioration in the load carrying capacity, ductility, and energy dissipation of the CFST columns. The larger the axial force ratio, the severer deterioration of deformation capacity of the columns.
1. Introduction Concrete filled steel tube (CFST) has an increasing utilization in the earthquake prone regions in China due to its high strength, good ductility, and excellent energy dissipation capacity [1]. The outer steel tube of CFST member is exposed to external environment and is prone to suffer environmental corrosions during the service life span, such as acid rain attack. Worldwide acid rain problems have been worsened by industrial and urban developments and acid rainfall has been reported to cover at least one third of Chinese territory [2–5]. Thus, it is essential to evaluate the seismic behaviors of CFST members that have suffered acid rain corrosion. In the past few decades, a great number of studies have been carried out on the seismic behaviors of CFST members [6–13]. Some literature reviews had been conducted by Nakanishi et al. [12] and Elremaily and Azizinamini [13]. It made a consensus that the CFST members exhibit much higher ductility compared with the hollow steel tubes owing to the composite effect between the core concrete and outer steel tube. Han et al. [14] also tested the cyclic behaviors of concrete filled double skin steel tubular (CFDST) members under combined axial and flexural load. It was reported that the CFDST members show good ductility and excellent energy dissipation capacity even under high levels of axial force ratio above 0.6. Experimental studies on steel structures and CFST members under corrosive environment have also been conducted in recent years. For
⁎
Corresponding author. E-mail address: [email protected] (M. Chen).
http://dx.doi.org/10.1016/j.tws.2017.10.005 Received 3 May 2017; Received in revised form 2 September 2017; Accepted 2 October 2017 Available online 13 November 2017 0263-8231/ © 2017 Elsevier Ltd. All rights reserved.
example, Almusallam [15] studied the effect of sodium chloride corrosion on the properties of reinforcing steel bars and found that reinforcing steel bars with more than 12% corrosion indicates a brittle failure. Qin and Cui [16] studied the effect of corrosion models on the time-dependent reliability of steel plate elements and the advantages and the flexibility of the proposed corrosion model were demonstrated. Melchers [17] studied the influential factors on the corrosion rate of steel in seawater environments. Saad-Eldeen et al. [18] tested the load carrying capacity of a corroded steel box girder. Sultana et al. [19] studied the compressive strength of stiffened panels under pitted corrosion. Karagah et al. [20] tested the steel columns under corrosion and axial compression. Han et al. [21,22] and Hou et al. [23] carried out the experimental studies on 22 beams and 34 stub columns under sustained load and chloride corrosion. The test results showed that the chloride corrosion has great effects on the load carrying capacity of construction steel and CFST members. Simplified calculation methods for the load carrying capacity of CFST beams and stub columns were also proposed based on parametric studies [24]. Previous studies have focused on the static behaviors of corroded CFST members. Few experimental works have been studied the seismic behaviors of corroded CFST members, especially under acid rain attack. This gives rise to the need for more studies of the problem. This paper aims to investigate the seismic behaviors of circular CFST members subjected to acid rain corrosion. The effect of corrosion on the mechanical behaviors of steel tubes is firstly tested and discussed. After
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Nomenclature
N0 Nu P Pm Pu Py ts ts0 η ξ μ εu0 εu Δ Δy Δm Δu
The following symbols are used in this paper Ac As D Es Es0 fck fcu fy fy0 fu fu0 n L
cross-sectional area of core concrete cross-sectional area of steel tube after corrosion column diameter elastic modulus of corroded steel elastic modulus of uncorroded steel characteristic compressive strength of concrete compressive strength of cube concrete yield strength of corroded steel yield strength of uncorroded steel ultimate strength of corroded steel ultimate strength of uncorroded steel axial force ratio column length
axial force applied on the columns axial compressive capacity of the columns lateral load of column peak load of column ultimate load of column yield load of column wall thickness of steel tube after corrosion initial wall thickness of steel tube corrosion rate confinement factor ductility coefficient ultimate elongation of corroded steel ultimate elongation of corroded steel lateral displacement of column yield displacement of column displacement of column at peak load ultimate displacement of column
where N0 is the applied axial force on the specimens, and Nu is the axial load carrying capacity of the columns, which is calculated by the simplified formulas described in [25]. The corrosion rate is defined as:
that, the effects of the corrosion rate and axial force ratio on the seismic behaviors of CFST columns, such as ultimate strength, ductility, energy dissipation ability, et al., are experimentally studied and systematically evaluated.
η= 2. Experimental program
In total, 12 circular column specimens were tested in the present work. All tested specimens have a sectional size of D × ts = 1500 × 114 × 4 mm and a length (L) of 1500 mm, where D is the diameter of outer steel tube sand ts is the wall thickness of steel tube. The CFST specimens were fabricated through the following steps. The steel tubes were segmented from an industrial steel tube. A steel plate was welded into one end of the steel tube. Then, the steel tubes were placed upright for casting. The concrete was cast into the steel tube and vibrated by a poker at the same time. After curing, a small gap between concrete surface and top steel tube was observed due to concrete shrinkage. The longitudinal gap was filled with a high strength epoxy in order to make the concrete surface flush with the top steel tube. Another steel plate was welded onto the top end of the steel tube before testing. The main design parameters were an axial force ratio (n) from 0 to 0.5 and a corrosion rate (η) from 0% to 30%. The axial force ratio herein is defined as:
N0 Nu
(2)
where ts0 is the initial thickness of the steel plate; ts is the remaining thickness after corrosion. The designed corrosion rates are 0, 10%, 20% and 30% respectively. Table 1 shows a summary of the tested specimens, where ξ represents the confinement factor to account for the ‘composite action’ between the steel tube and core concrete, and was defined as follows [26]:
2.1. Preparation of specimens
n=
ts0 − ts × 100% ts0
ξ=
As f y Ac fck
(3)
where As is the cross-sectional area of steel tube after corrosion, fy is the yield strength of steel, Ac is the cross-sectional area of the core concrete, fck is the characteristic compressive strength of concrete. The value of fck is calculated to be 67% of the cube strength of concrete (fcu). The following naming rules are employed to distinguish specimens: 1) the two initial characters ‘CC’ represents the circular column section; 2) the Arabic numerals before hyphen stand for the axial force ratio; 3) the Arabic numerals after hyphen represent the corrosion rate. For example, the specimen ‘CC0.2-10’ stands for the circular column with designed axial force ratio of 0.2 and corrosion rate of 10%.
(1)
Table 1 Information of column specimens. No.
Specimen ID
D (mm)
ts (mm)
L (mm)
fcu (MPa)
n
η (%)
ξ
Yield
Peak
Ultimate
Py (kN)
Δy (mm)
Pm (kN)
Δm (mm)
Pu (kN)
Δu (mm)
1 2 3 4
CC0.2-0 CC0.2-10 CC0.2-20 CC0.2-30
114 114 114 114
4.00 3.62 3.23 2.84
1500 1500 1500 1500
60 60 60 60
0.2 0.2 0.2 0.2
0 9.48 19.25 29.00
1.49 1.18 0.95 0.85
56.4 52.3 45.8 40.7
16.18 15.78 15.67 14.89
73.55 69.40 58.80 51.50
35.85 33.25 29.85 29.50
62.69 58.99 51.00 44.12
– 44.21 37.10 34.04
5 6 7 8
CC0.4-0 CC0.4-10 CC0.4-20 CC0.4-30
114 114 114 114
4.00 3.63 3.21 2.81
1500 1500 1500 1500
60 60 60 60
0.4 0.4 0.4 0.4
0 9.25 19.75 29.75
1.49 1.19 0.94 0.84
53.1 46.3 40.2 34.8
12.82 11.15 10.37 8.85
73.75 65.80 56.85 46.70
33.95 33.55 23.55 17.90
62.69 56.49 48.33 39.70
60.75 36.64 33.23 21.60
9 10 11 12
CC0.5-0 CC0.5-10 CC0.5-20 CC0.5-30
114 114 114 114
4.00 3.60 3.19 2.78
1500 1500 1500 1500
60 60 60 60
0.5 0.5 0.5 0.5
0 10.00 20.25 30.50
1.49 1.18 0.94 0.83
52.6 49.3 46.1 39.3
11.44 12.42 11.60 10.11
70.35 62.65 57.70 48.85
31.50 27.50 24.40 23.45
58.36 53.25 49.05 41.52
39.42 38.66 34.48 25.73
91
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
2.2. Corrosion test configuration
Table 2 Mixture proportions of concrete.
The column specimen test was divided into two stages. In the first stage it is accelerated corrosion test and in the second stage it is ultimate strength test. The accelerated corrosion setup was composed of solution tank, stainless steel plate, power source, and electrolytic solution. The corrosive environment was simulated with acid rain solution and applied DC current. The simulated acid rain solution was made up of Na2SO4, (NH4)2SO4, MgSO4, Ca(NO3)2 and HNO3 et al. in terms of the components of acid rain in southern China [27]. The main cations include H+, NH4+, Mg2+ and Ca2+ while the main anions include SO42- and NO3-. The corrosion of constructional steel is mainly caused by H+ and SO42-. The pH value of the solution was determined by HNO3 and was fixed to be 2.3. In order to make path for the corrosion current, the anode of the power source was connected to the specimens while the cathode of the power source was connected to the stainless steel plate, as shown in Fig. 1. Antirust was applied on the two ends of the CFST specimens. As a result, only the outer surface of the steel tubes corroded (Fig. 1). Besides CFST specimen tests, four groups of steel coupons with corrosion rates from 0% to 30% were also tested by the accelerated corrosion, and tensioned to monitor the deterioration of tensile properties of the steel. The steel coupons were fabricated and tested in accordance with the Chinese standards with respect to the metal materials used. Three parallel coupons were prepared corresponding to each corrosion rate. Only the surface perpendicular to the normal direction of the steel coupons (as shown in Fig. 4) was exposed to the corrosive environment and other surfaces were covered with antirust in order to simulate the practical corrosion condition of the steel tubes. The corrosion rate was controlled by adjusting current intensity and immersed time. To fix the current intensity and immersed time at each of designed corrosion rate, steel coupons were first tested before CFST specimen tests. The thickness loss of steel plate was monitored carefully during the corrosion process. The surfaces of the steel coupons and the CFST columns were cleaned after corrosion. Two kinds of approaches were adopted to measure the corrosion rate: one is to weight the tested specimens and the other is to measure the average thickness of the steel coupons or the diameter of the steel plate. It was found that the thickness losses (Δts) measured by these two methods showed high consistency.
Matrix
Cement
water
Sand
Coarse aggregate
Concrete
1.0
0.4
1.1
2.56
properties are discussed in the following sections. One type of concrete with the mixture proportions shown in Table 2 was designed. The measured cube strength (fcu) at 28 days is 60 MPa. 2.4. Cyclic loading configuration The CFST specimens were tested under a constant axial force and increasing cyclic displacements. The schematic cyclic loading test setup is shown in Fig. 2. The two ends of the columns were attached to the pin supports, which were free to rotate to simulate the pin-pin end conditions. The axial force was first applied through a hydraulic jack installed between one end of the column and a fixed concrete block. Then, the reversed cyclic load was applied via a MTS hydraulic ram fixed on a reaction frame with ultra-high flexural stiffness. The lateral loading history was made up of elastic and inelastic cycles. The elastic cycles were conducted by load cycles at load intervals of 0.25Py, where Py is the estimated lateral yield load corresponding to the lateral yield displacement Δy. After the yield point, the inelastic cycles were controlled by displacement cycles at intervals of Δy. Three cycles were imposed at each inelastic displacement level. The loading history is shown in Fig. 3. The test was terminated when the ultimate state was reached where the lateral load resistance drops to 85% of its peak value. The instrumentation of the test included the measurements of lateral load and displacement at the loading point of the specimen, horizontal load of the hydraulic jack, horizontal displacement of the two ends of the concrete blocks, and the strains in the steel tubes. The loads were recorded by load cells (LC1 and LC2) and the displacements were measured with the linear voltage digital transducers (LVDTs, A, B, and C). Strain gauges were attached to the surface of the steel tubes near the base of each column to monitor the strain variations during test. The detailed measurement arrangements are clearly shown in Fig. 2. 3. Results and discussions
2.3. Material properties
3.1. Effects of corrosion on the mechanical properties of steel
The measured yield strength, ultimate strength and elastic modulus of the uncorroded steel coupons are 383 MPa, 516 MPa and 187 GPa, respectively. A tensile test of the corroded steel coupons with different thickness losses was also carried out and the detailed mechanical
3.1.1. Failure modes of steel coupons The steel coupons with various specified corrosion rates were tested in tension. Fig. 4 shows the failure modes of the specimens. It is clearly seen that apparent necking is observed around the fracture section of
Fig. 1. Accelerated corrosion test setup.
92
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 2. Schematic cyclic loading test setup.
Fig. 3. Cyclic loading history.
uncorroded steel coupons and the fracture section is almost perpendicular to the tensile load direction. However, for the corroded steel coupons, tensile fracture occurred suddenly without necking phenomenon and the directions of the fracture sections were disordered. A
Fig. 5. Effect of corrosion rate (η) on the stress-strain response of steel.
Fig. 4. Failure modes of steel coupons.
93
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Table 3 Mechanical properties of uncorroded and corroded steel coupons. No.
η (%)
Yield strength fy (MPa)
Ultimate strength fu (MPa)
Elastic modulus Es (GPa)
Ultimate elongation εu (%)
Es loss (%)
fu loss (%)
fy loss (%)
εu loss (%)
1 2 3 Average
0 0 0 0
370.17 368.92 410.10 383.06
516.11 505.41 526.88 516.13
190.38 188.96 181.60 187.98
36.04 36.20 36.20 36.15
– – – –
– – – –
– – – –
– – – –
4 5 6 Average
9.89 10.44 11.54 10.62
356.71 333.59 319.88 336.73
448.32 442.18 452.64 447.71
174.80 158.44 176.45 169.90
35.34 29.18 28.80 31.11
– – – 9.62
– – – 13.26
– – – 12.09
– – – 13.94
7 8 9 Average
21.70 23.35 24.45 23.17
270.35 308.24 331.27 303.29
413.68 411.83 427.09 417.53
156.00 136.00 151.51 147.84
26.64 25.98 23.64 25.42
– – – 21.36
– – – 19.10
– – – 20.83
– – – 25.42
10 11 12 Average
27.47 27.75 35.44 30.22
304.36 300.54 322.13 309.01
395.27 409.86 418.72 407.95
140 125.3 138.52 134.61
18.88 15.52 13.34 15.91
– – – 28.39
– – – 20.96
– – – 19.33
– – – 55.98
Fig. 6. Effect of corrosion rate (η) on the mechanical properties of steel.
crack will appear from the largest pit and develop along the thinnest section, leading to an irregular fracture section.
careful observation of the corroded steel coupons showed that a large amount of small corrosion pits formed at the surface of the specimens even though the uniform corrosion was designed and applied for all specimens. The higher the degree of the corrosion, the more significant nonuniformity of the local corrosions. Under the tensile load, an initial
3.1.2. Typical tensile stress-strain curves of the corroded steel coupons The typical tensile stress-strain curves of the steel coupons with 94
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 7. Failure patterns of specimens.
εu are observed as the corrosion rate (η) increases. It indicates that the acid rain corrosion on steel tubes results in not only a loss in wall thickness of the steel tube but also a reduction of the tensile strength and the deformation. However, only the effective loss of thickness of a steel tube has been considered for the derivation of simplified strength models of corroded CFST members in most of existing researches [21–24]. As pointed out by Han et al. [26], the confinement factor (ξ) plays an important role in the mechanical behaviors of CFST columns, stub columns, and beam-columns. It can be inferred from Eq. (3) that not only the cross-sectional area of a steel tube (As) that is dependent on the wall thickness of steel tube (ts) but also the yield strength of the steel has a positive effect on ξ. Therefore, the ignorance of the effect of environmental corrosion on the yield strength of a steel tube will result in non-negligible errors in the prediction of the load carrying capacity of CFST members. In the present work, the variations of mechanical properties of steel with corrosion rate (η) are quantitatively analyzed. By analyzing the variations of Es, fy, fu, and εu, a linear relationship between these factors and η is approximated. Thus, the following models of Es, fy, fu, and εu are obtained through the regression analyses:
Fig. 8. Typical failure modes of core concrete in CFST.
different corrosion rates are shown together in Fig. 5. The tensile stress herein is defined as the ratio of the tensile load to that of remaining cross-sectional area of steel coupons after corrosion. It is clearly seen that both the yield strength and ultimate strength of steel coupons without corrosion are significantly larger than those of corroded steel coupons. It can be also found from Fig. 5 that the yield stage becomes shorter and fracture occurs earlier and earlier as the corrosion rate increases, implying that the fracture process of the steel coupons behaves more brittle with the increasing of corrosion rate. It is attributed to the less significant necking phenomenon of the corroded steel coupons during the tensile failure process, which results in more brittle tensile process and lower fracture strain.
Es = (1 − 0.955η) E0
(4)
f y = (1 − 1.007η) f y0
(5)
fu = (1 − 0.748η) fu0
(6)
εu = (1 − 1.450η) εu0
(7)
where Es, fy, fu, and εu are elastic modulus, yield strength, ultimate strength, and ultimate elongation of corroded specimens, respectively. Es0, fy0, fu0, and εu0 are the elastic modulus, yield strength, ultimate strength, and ultimate elongation of uncorroded specimens, respectively. The ratios of Es, fy, fu, and εu estimated by Eqs. (4)–(7) to the measured values have a mean value of 0.998, 0.965, 1.016, and 1.018, respectively, and a standard deviation (SD) of 5.77%, 9.05%, 4.41%, and 16.48%, respectively. Thus, a good agreement between the proposed models and the test results is obtained. Since the uniform acid
3.1.3. Mechanical properties of corroded steel The mechanical properties, such as elastic modulus (Es), yielding strength (fy), ultimate strength (fu), and ultimate elongation (εu) of the steel coupons are shown in Table 3. Fig. 6 shows the effect of corrosion rate on these tensile properties. Clear decreasing trends of Es, fy, fu, and 95
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 9. Cyclic load-displacement curves of each specimen.
bulge also occurred at the opposite face of the specimen as the lateral cyclic load was reversed. With increasing the lateral displacement, a complete ring formed near the loading section at midspan. Finally, some columns failed due to the tensile fracturing at the bulge location, accompanied by rapid drop in the lateral load. The failure of other specimens without tensile failure was determined by the 15% drop of the lateral load resistance. It can be also found from Fig. 7 that the failure patterns of the corroded CFST specimens were similar to those of the uncorroded specimens, which means that corrosion has nearly no effect on the failure patterns of the circular CFST members. The similar observation
rain corrosion was applied for steel coupons in the present work, Eqs. (4)–(7) are only applicable for the steel material under uniform acid rain attack. 3.2. Failure characteristics of CFST members The failure patterns of the CFST columns are shown in Fig. 7. It can be found that the CFST specimens deformed in a quite ductile way. No sign of local deformation was observed before the yielding initiation of the specimens. Beyond the yielding, an outward buckling was observed close to loading point at the compressive face of the specimen. The 96
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 9. (continued)
3.3. Lateral load (P) versus lateral displacement (Δ) curves
has been reported for CFST members under sustained static load and chloride corrosion tested by Han et al. [22] and Hou et al. [23]. Fig. 8 shows the failure patterns of the core concrete after the test. It can be found that, under reversed tension and compression, an evident sign of cracking and crushing was observed in the vicinity of the column base. Because the maximum moment occurred at the column base, this test observation is quite reasonable for columns under reversed lateral load.
Fig. 9 shows the lateral load versus the mid-span displacement curves (which are also called hysteresis curves) for all specimens. It is easily found that the hysteresis loops of the uncorroded columns are more stable and significantly plumper than those of the corroded columns. The hysteresis loop becomes increasingly smaller as the corrosion rate increases, indicating a lower energy dissipated capacity for column with severer corrosion. This phenomenon can be explained by the following factors. First, the wall thickness of the outer steel tube decreases as the corrosion rate increases, which leads to a reduction of 97
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
Fig. 10. Effect of corrosion rate (n) on the load-displacement envelop curves of specimens.
Fig. 11. Effect of axial force ratio (n) on the global response of columns.
98
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
an important role in the energy dissipation of steel-concrete composite members. The pinching effect of the hysteresis loops would be more apparent for CFST members with a lower steel ratio. Third, as discussed above, corrosion also leads to the reduction in yield strength of a steel tube, which further reduces the confinement factor (ξ), as indicated in Eq. (3). The confinement effect of the steel tube on the core concrete becomes weaker and the local buckling of steel tube is more prone to occur as ξ decreases, resulting in lower load carrying and deformation capacities, and thus smaller hysteresis loops of the CFST members. Fig. 10 shows the load versus displacement envelop curves of all specimens. The curves of the corroded specimens tend to drop more quickly than the uncorroded specimens due to a premature buckling of the steel tubes. The load carrying capacity decreases and the area encompassed by the curve decreases as the corrosion rate increases at each axial force level. For example, the peak load (Pk) and ultimate displacement (Δu) of the CFST columns with an axial force ratio of 0.4 are reduced by 36.7% and 64.4% as the corrosion rate changes from 0% to 29.75%, respectively, where the ultimate displacement herein is defined as the point where the lateral load resistance drops to 85% of its peak value. This definition of ultimate state was also usually employed by other researchers [28–31]. The reductions of Pk and Δu are attributed to the losses of the wall thickness and yield strength of steel tubes caused by corrosion. The detailed test data for each specimen are shown together in Table 1. The ultimate displacement of specimen CC0.2-0
Fig. 12. Effect of corrosion rate (η) on the ductility of specimens.
the tensile load provided by the steel and a composite action between the outer steel tube and core concrete. Second, as is well-known, steel is a kind of material with excellent plasticity and the outer steel tube plays
Fig. 13. Effect of corrosion rate (η) on the cumulative dissipated energy of specimens.
99
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al.
modulus Es, yield strength fy, ultimate strength fu, and ultimate elongation εu. A linear relationship between Es, fy, fu and εu of steel coupons and corrosion rate is found and proposed. As a result, the loss of strength and deformation capacities of outer steel tubes caused by corrosion should be taken into consideration in the prediction of the load carrying capacity of corroded CFST members. 2. The failure patterns of uncorroded CFST specimens and corroded specimens are similar to each other. However, the seismic performance of uncorroded CFST columns is evidently superior to that of corroded columns, such as higher load carrying capacity, better ductility and larger energy dissipation capacity. The higher the corrosion rate, the poorer the seismic performance of the CFST specimens. 3. The axial force level has nearly no effect on the ultimate lateral load but an adverse influence on the ultimate displacement of the specimens.
was unfortunately not obtained from the test due to exceeding the maximum range of the LVDT. Fig. 11 shows the effects of the axial force ratio on the peak load and ultimate displacement of the specimens. The four points at each axial force ratio correspond to the values at four corrosion rates, respectively. It can be seen that the axial force level has nearly no effect on the load carrying capacity but great influence on the ultimate displacement of the specimens. This phenomenon is quite often observed in reinforced concrete columns. Generally, the deformation capacity of the columns decreases as the axial force level increases. For CFST columns with a higher axial force level, the compression face tends to buckle more quickly due to the larger compression load. As a result, the lateral load tends to drop more rapidly at a relatively smaller lateral deformation. 3.4. Ductility and energy dissipation The ductility coefficient (μ) is a key parameter for evaluating the ductility performance of the composite members. The ductility coefficient herein is defined as the ratio of the ultimate displacement to the yield displacement of CFST columns (Δu/Δy). The effect of the axial force ratio on the ductility coefficient μ for each specimen is presented in Fig. 12. An evident decreasing trend of the ductility of the CFST columns at each axial force level is observed as the corrosion rate increases. Both the yield displacement and the ultimate displacement of the columns decrease with increasing η due to the reduction of the wall thickness ts and confinement factor ξ caused by corrosion, as shown in Table 1. The deterioration of the ultimate displacement is more substantial than that of the yield displacement, resulting in a lower ratio of Δu/Δy and thus a smaller μ for a larger η. The buckling of the steel tube and the corresponding ultimate state occur earlier for the column with a smaller ξ due to the weaker composite effect between the steel tube and core concrete. Fig. 13 shows the cumulative dissipated energy at each displacement level for each specimen. In the present work, the cumulative dissipated energy is defined as the sum of the areas of each hysteresis loop up to the specific displacement level (Δu/Δy). Since all specimens deform in a ductile manner, the cumulative dissipated energy shows a steady increase until failure for each specimen. It can be also found from Fig. 13 that the energy dissipation ability exhibits an evident decreasing trend with the increase of the corrosion rate. The cumulative dissipated energy levels are close to each other before the yield point at each axial force level. After the yield point, however, the gap in the cumulative energy between the corroded specimens and uncorroded specimens becomes increasingly larger as the displacement level increases. It indicates that the composite action between the outer steel tube and core concrete plays an important role in the energy dissipation capacity. As mentioned above, the confinement factor ξ decreases substantially due to the reduction caused by corrosion of the wall thickness as well as yield strength for the steel tubes. As ξ decreases, the compressive strength of the core concrete is reduced and the buckling of the outer steel tube occurs earlier, both of which have adverse effect on the energy dissipation capacity of CFST columns.
It is noted that only uniform corrosion on the outer surface of CFST columns was taken into consideration in the present work. While the practical corrosion situations, such as localized corrosion, has not been taken into accounted in the present experimental study. This topic will be discussed in near future ongoing researches. Acknowledgement Financial supports of the work by National Natural Science Foundation of China under 51378206 and 51608199, and in part by Natural Science Foundation of Jiangxi Province under 20161BAB216140 as well as the program for Advanced Science and Technology Innovation Team of Jiangxi Province under 20152BCB24006, are gratefully acknowledged. References [1] L.H. Han, Some recent developments of concrete filled steel tubular (CFST) structures in China, in: Proceedings of the 4th International Conference on Steel & Composite Structures, 2010. [2] A.B. Eney, D.E. Petzold, The problem of acid rain: an overview, Environ. Syst. Decis. 7 (7) (1987) 95–103. [3] F. Mansfeld, R. Vijayakumar, Atmospheric corrosion behavior in Southern California, Corros. Sci. 28 (9) (1988) 939–946. [4] T. Larssen, H.M. Seip, A. Semb, J. Mulder, et al., Acid deposition and its effects in China: an overview, Environ. Sci. Policy 2 (1) (1999) 9–24. [5] J. Hill, E.A. Byars, J.H. Sharp, et al., An experimental study of combined acid and sulfate attack of concrete, Cem. Concr. Compos. 28 (8) (2003) 997–1003. [6] K. Sakino, M. Tomii, Hysteretic behavior of concrete filled square steel tubular beam-columns failed in flexure, Trans. Jpn. Concr. Inst. 3 (6) (1981) 439–446. [7] Y. Morishita, M. Tomii, Experimental studies on bond strength between square steel tube and encased concrete core under cyclic shearing force and constant axial force, Trans. Jpn. Concr. Inst. 4 (1982) 363–370. [8] H.B. Ge, T. Usami, Cyclic tests of concrete filled steel box columns, J. Struct. Eng. 122 (10) (1996) 1169–1177. [9] P.F. Boyd, W.F. Cofer, D.I. Mclean, Seismic performance of steel-encased concrete columns under flexural loading, ACI Struct. J. 92 (3) (1995) 355–364. [10] J.F. Hajjar, A. Molodan, P.H. Schiler, A distributed plasticity model for cyclic analysis of concrete-filled steel tube beam-columns and composite frames, Eng. Struct. 20 (4–6) (1998) 398–412. [11] K. Lahlou, M. Lachemi, P.C. Aitcin, Confined high-strength concrete under dynamic compressive loading, J. Struct. Eng. 125 (10) (1999) 1100–1108. [12] K. Nakanishi, T. Kitada, H. Nakai, Experimental study on ultimate strength and ductility of concrete filled steel columns under strong earthquake, J. Constr. Steel Res. 51 (3) (1999) 297–319. [13] A. Elremaily, A. Azizinamini, Behavior and strength of circular concrete-filled tube columns, J. Constr. Steel Res. 58 (12) (2002) 1567–1591. [14] L.H. Han, H. Huang, Z. Tao, X.L. Zhao, Concrete-filled double skin steel tubular (CFDST) beam–columns subjected to cyclic bending, Eng. Struct. 28 (12) (2006) 1698–1714. [15] A.A. Almusallam, Effect of degree of corrosion on the properties of reinforcing steel bars, Constr. Build. Mater. 15 (8) (2001) 361–368. [16] S. Qin, W. Cui, Effect of corrosion models on the time-dependent reliability of steel plated elements, Mar. Struct. 16 (1) (2003) 15–34. [17] R.E. Melchers, Recent progress in the modeling of corrosion of structural steel immersed in seawaters, J. Infrastruct. Syst. 12 (3) (2006) 154–162. [18] S. Saad-Eldeen, Y. Garbatov, C.G. Soares, Effect of corrosion severity on the ultimate strength of a steel box girder, Eng. Struct. 49 (2) (2013) 560–571.
4. Conclusions This paper has investigated the mechanical properties of CFST columns subjected to a cyclic load and an acid rain attack through tests. The tensile properties of steel coupons were first studied in detail. A reversed cyclic load test was subsequently carried out for both corroded and uncorroded CFST specimens. The effects of corrosion rate η and the axial force ratio were taken into consideration. In summary, the following conclusions can be drawn from the present paper: 1. Acid rain corrosion has not only a significant influence on the reduction in wall thickness of a steel tube, but also an adverse effect on the mechanical properties of the steel, such as the elastic 100
Thin-Walled Structures 122 (2018) 90–101
F. Yuan et al. [19] S. Sultana, Y. Wang, A.J. Sobey, J.A. Wharton, R.A. Shenoi, Influence of corrosion on the ultimate compressive strength of steel plates and stiffened panels, Thin Wall Struct. 96 (2015) 95–104. [20] H. Karagah, C. Shi, M. Dawood, A. Belarbi, Experimental investigation of short steel columns with localized corrosion, Thin Wall Struct. 87 (2015) 191–199. [21] L.H. Han, C. Hou, Q.L. Wang, Square concrete filled steel tubular (CFST) members under loading and chloride corrosion: experiments, J. Constr. Steel Res. 71 (2012) 11–25. [22] L.H. Han, C. Hou, Q.L. Wang, Behavior of circular CFST stub columns under sustained load and chloride corrosion, J. Constr. Steel Res. 103 (2014) 23–36. [23] C.C. Hou, L.H. Han, Q.L. Wang, C. Hou, Flexural behavior of circular concrete filled steel tubes (CFST) under sustained load and chloride corrosion, Thin Wall Struct. 107 (2016) 182–196. [24] C. Hou, L.H. Han, X.L. Zhao, Full-range analysis on square CFST stub columns and beams under loading and chloride corrosion, Thin Wall Struct. 68 (10) (2013) 50–64. [25] DBJ/T13-51-2010, Technical specification for concrete-filled steel tubular
[26] [27] [28] [29]
[30]
[31]
101
structures, The Construction Department of Fujian Province, Fuzhou, China, 2010. (in Chinese). L.H. Han, X.L. Zhao, Z. Tao, Tests and mechanics model for concrete-filled SHS stub columns, columns and beam-columns, Steel Compos. Struct. 1 (1) (2001) 51–74. M.C. Chen, K. Wang, L. Xie, Deterioration mechanism of cementitious materials under acid rain attack, Eng. Fail. Anal. 27 (1) (2013) 272–285. M.J.N. Priestley, R. Park, Strength and ductility of concrete bridge columns under seismic loading, Acids Struct. J. 84 (1) (1987) 61–76. G. Fischer, V.C. Li, Effect of matrix ductility on deformation behavior of steel-reinforced ECC flexural members under reversed cyclic loading conditions, ACI Struct. J. 99 (6) (2002) 781–790. Y.F. Wu, T. Liu, D.J. Oehlers, Fundamental principles that govern retrofitting of reinforced concrete columns by steel and FRP jacketing, Adv. Struct. Eng. 9 (4) (2006) 507–533. F. Yuan, J.L. Pan, L.T. Dong, C.K.Y. Leung, Mechanical behaviors of steel reinforced ECC or ECC/concrete composite beams under reversed cyclic loading, J. Mater. Civ. Eng. 26 (8) (2014) 04014047.