Engineering Structures 153 (2017) 290–301 Contents lists available at ScienceDirect Engineering Structures journal homepage: www.elsevier.com/locate/engstruct Experimental study of cyclic behavior of high-strength reinforced concrete columns with different transverse reinforcement detailing configurations Wen-Cheng Liao a, Wisena Perceka a,b,⇑, Michael Wang b a b Department of Civil Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 10617, Taiwan Center For Earthquake Engineering Research, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 10617, Taiwan a r t i c l e i n f o Article history: Received 29 May 2017 Revised 4 October 2017 Accepted 4 October 2017 Keywords: Confinement Conventional closed-hoops Butt-welded hoops Single closed-hoop Cross-ties High-strength concrete High-strength steel High axial compression loading a b s t r a c t Detailing of transverse reinforcement is essential to ensure ductile behavior of reinforced concrete (RC) members, particularly for columns with large displacement demanding under high axial loading level. Requirements of reinforcing details are even important for high strength concrete due to its brittleness nature. This paper presents experimental study regarding cyclic behavior of high strength RC columns with different transverse reinforcement detailing layouts. The three high strength RC columns consisted of different transverse reinforcement detailing configurations, which were conventional closed-hoops, butt-welded hoops and single closed-hoop with cross-ties, respectively. The columns were made of high-strength concrete with compressive strength of 70 MPa and high-strength steel with yield stresses of 685 MPa and 785 MPa for longitudinal and transverse reinforcement, respectively. Cyclic displacement static tests subjected to high axial compression loading of 0.3Agf0 c were conducted to verify the adequacy of transverse detailing. The performance of the column with conventional closed-hoops and that of the column with butt-welded hoops, which were compliant of ACI 318-14, were almost identical and met the performance criteria required by ACI 374. In contrast, the column with a single closed-hoop and cross-tie that was designed based on ACI 318-11 did not perform well as a ductile RC column. These results prove that the ACI 318-14 minimum requirements for confinement should be followed. In addition, butt-welded hoops are acceptable as a form of transverse reinforcement of a column, since no fracture was observed at the welded location. The current flexural and shear design equations for the RC columns are discussed and compared with the test results. Ó 2017 Elsevier Ltd. All rights reserved. 1. Introduction The Taiwan New Reinforced Concrete (New RC) research project was launched in 2008. One of the main objectives of the New RC research project is to apply high-strength steel and concrete to RC columns in high-rise buildings [1]. The use of high-strength concrete and steel has gained attention recently due to the requirement for more available floor area, which can be achieved by limiting the size of lower-story columns in high-rise buildings [2,3]. By using high-strength concrete and steel, the column section size and reinforcement congestion can be reduced. In Taiwan, advanced technology has led to the development of high-strength concrete with a compressive stress of 70 MPa or greater. Meanwhile, ⇑ Corresponding author at: Department of Civil Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 10617, Taiwan. E-mail address: d02521014@ntu.edu.tw (W. Perceka). https://doi.org/10.1016/j.engstruct.2017.10.011 0141-0296/Ó 2017 Elsevier Ltd. All rights reserved. deformed steel bars SD685 and SD785 with specified yield stresses of 685 MPa and 785 MPa for longitudinal and transverse reinforcement, respectively, are commercially available [2,3]. Increasing the compressive strength, however, renders the concrete more brittle, and thus greater transverse reinforcement ratio, particularly for confinement in a high-strength concrete column is necessary [4,5]. Detailing of transverse reinforcement is essential to ensure ductile behavior of RC members, particularly for columns with large displacement demanding under high axial loading level. Requirements of reinforcing details are even important for high strength concrete due to its brittleness nature. The ACI 318-11 requirements for confinement reinforcement, however, do not specify the limit of concrete strength and axial load [6]. The results of a review of column test data conducted by Elwood et al. confirmed that the previous confinement provisions (ACI 318-11) should be changed to include axial load for designing confinement reinforcement [7]. ACI 318-14 then W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 291 Notations Ag As Ash bc, bw Cc, C00 s d d0 d00 f0 c Fs F0 s fy fsh fyt h k Lc kf kn Mm area of the column section total area of longitudinal reinforcement total area of transverse reinforcement parallel to lateral loading direction side dimension of the column core section and side dimension of the column section, respectively, perpendicular to lateral loading direction compressive force in concrete and compressive force in longitudinal steel reinforcement, respectively, in cross section of the column effective depth of the column section (measured from extreme-compression fiber to centroid of extremelayer of tensile steel reinforcement) distance from extreme-compression fiber to centroid of steel reinforcement in cross section of the column distance from extreme-compression fiber to centroid of steel reinforcement closest to extreme-compression fiber specified compressive strength of the concrete material tension force in the extreme-layer of tensile longitudinal steel reinforcement in cross section of the column compressive or tension force in the longitudinal steel reinforcement in cross section of the column yield stress of longitudinal reinforcement stress of transverse reinforcement used to design shear strength (limited to not exceed 600 MPa) stress of transverse reinforcement used to design confinement (limited to not exceed 700 MPa) height of the column section factor relating the concrete or transverse reinforcement capacity to the displacement ductility [Eqs. 12 and 12(a)] clear height of column specimen concrete strength factor for designing confinement reinforcement [Eqs. 8 and 8(a)] confinement effectiveness factor for designing confinement reinforcement [Eqs. 8 and 8(b)] factored moment modified to account for effect of axial compression load [Eqs. 10 and 10(a)] adopted the results of the study conducted by Elwood et al. to develop the new confinement provision, in which the axial load and high-strength concrete factors are directly accounted for [8]. The new confinement provision also considers the number of longitudinal bars around the perimeter of the column core that are laterally supported by the corner of closed-hoops or cross-ties with a 135-degree hook [9]. A cross-tie is a transverse reinforcement bar with a 135-degree hook at one of its edges and a 90-degree hook at another edge. As is commonly known, the transverse reinforcement detailing may be a conventional closed-hoops model with a seismic hook at one corner or a single closed-hoop model with cross-ties [6]. Both ACI 318-11 and ACI 318-14 limit the yield stress of transverse reinforcement as 420 MPa for designing shear strength, while the use of yield stress of 700 MPa is permitted for designing confinement reinforcement. Those codes also specify not to use butt-welded hoops for ductile RC columns, since it can lead to local embrittlement of steel [6,8,10]. However, there are several advantages of using butt-welded hoops, such as the elimination of the possibility of anchor pull-out failure, and the reduction of reinforcement congestion [11]. It is worth mentioning that the amount and detailing of transverse reinforcement is critical to assure ductile behavior of RC columns under cyclic and high axial compression loading. Therefore, the cyclic behavior of high-strength RC columns using different Mn Mpr Mu nl Pn Po Pu Vc Vn Vs Vn,f, Vpr a a1 b1 k qw X1 nominal moment about an axis perpendicular to lateral loading direction (computed about the centroid of rectangular section) [Eq. 4] maximum moment probable about an axis perpendicular to lateral loading direction moment corresponding to applied axial compression load number of longitudinal bars around the perimeter of the column core that are laterally supported by the corner of closed-hoops or conventional cross-ties with a 135degree hook nominal axial compression force in column section at given eccentricity [Eq. 3] nominal strength of reinforced concrete column at zero eccentricity [Eq. 5] applied axial load shear strength provided by concrete [Eqs. 9,10,11,12, and 13] nominal shear strength [Eqs. 9,10,11,12, and 13] shear strength provided by transverse reinforcement [Eqs. 9,10,11,12, and 13] shear force corresponding to the flexural strength of the column section and shear force corresponding to the maximum moment probable, respectively reduction factor [Eqs. 13 and 13(a)] factor relating the magnitude of uniform stress in an equivalent rectangular compressive stress block to the specified compressive strength of the concrete [Eq. 1] ratio of the depth of the equivalent rectangular compressive stress block to the neutral axis [Eq. 2] factor reflecting the reduced mechanical properties of lightweight concrete (1 for normal concrete) ratio of total area of tension steel reinforcement to bw.d ratio of the mean concrete compressive stress corresponding to the maximum axial load resisted by a concentrically loaded column to the specified compressive strength of concrete transverse reinforcement detailing models is investigated in this study. The high-strength RC columns with high-strength steel were prepared and tested. All specimens had the same longitudinal bar configuration, while the transverse reinforcement spacing of all columns satisfied the minimum requirements of ACI 318-14 for transverse reinforcement spacing. 2. Research significance This study investigates the lateral strength and behavior of New RC columns that use different transverse reinforcement detailing configurations under double-curvature cyclic and high axial compression loading. The three high strength RC column specimens consisted of different transverse reinforcement detailing configurations, which were conventional closed-hoops, butt-welded hoops and single closed-hoop with cross-ties, respectively. The applied axial compression ratio was 0.3. The specified concrete compressive strength, yield stress of the longitudinal reinforcing bar, and yield stress of the transverse reinforcing bar were 70 MPa, 685 MPa, and 785 MPa, respectively. This study is significant because the tests also simulated the behavior of columns in the lower stories of high-rise buildings, in which a column is subjected to high axial compression loading. In addition, research on high-strength RC columns using different transverse 292 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 reinforcement detailing layouts under cyclic and high axial compression loading is still limited. The designed strengths (flexural and shear) of three specimens, which were determined using the current flexural and shear design equations, are compared with the test results, and further analysis and application of design equations for New RC columns are clearly discussed accordingly. 3. Experimental program Three large-scale columns with a square cross section of 60 cm 60 cm and a clear height of 180 cm were prepared and tested under double-curvature cyclic and high axial compression loading. Detailed information about the experimental procedure is presented in the next five sub-chapters. 3.1. Material The cementitious materials used in this study were ASTM Type I Portland Cement and ground granulated blast-furnace slag (GGBS). The maximum size of the coarse aggregate was 19 mm, and the fine aggregate used was local natural sand. A superplasticizer (SP) was also used to improve the workability of high-strength concrete in the fresh state. The design concrete compressive strengths were 70 MPa for all specimens. The concrete compressive strength of each column specimen was obtained by taking the average compressive strength of six 100 mm 200 mm cylinders; the compressive strength of the concrete cylinders was measured by the compression tests based on the ASTM C-39 procedure [12]. Two different sizes of reinforcing bars were used: D29 (6.29 cm2) and D13 (1.27 cm2) with specified yield stresses of 685 MPa and 785 MPa, respectively. At least three reinforcing bars from each specified diameter were subjected to a direct tensile test, and a baseline tensile test determined the yield strength, ultimate strength and bar strain at the yield stress for the control reinforcing bar according to ASTM E8/E8M and ASTM A370 [13,14]. The concrete mix proportion is summarized in Table 1. The average slump flow diameter of fresh concrete was 60 cm. The label of each specimen begins with CF-C, denoting the concrete column having flexural failure, since the columns were designed to have flexural failure. The symbols PT, W, and FT refer to the types of configuration of transverse reinforcement. PT is the specimen with conventional closed-hoops with a 135-degree hook at one Table 1 Concrete mix proportion (kg/m3). Cement Slag Coarse aggregate Fine aggregate Water SP 292 238 859 816 170 5.57 Fig. 2. Specimen elevation. of the corners of each hoop, W is the specimen consisting of butt-welded hoops, and FT is the specimen with a single closedhoop and cross-ties. The value of 0.3 denotes the axial compression ratio. 3.2. Specimen design The details of the specimen design are shown in Figs. 1 and 2. Information regarding the specified and actual concrete strength and steel reinforcement strength is summarized in Table 2. Detailed information about the column design parameters is listed in Table 3. The concrete cylinders and their corresponding specimens were tested on the same day. Fig. 3 shows the compression test setup and typical failure of a concrete cylinder, while Fig. 4 shows the failure of welded steel bars due to tensile testing. Fig. 4 shows that the fracture location on the three steel bars was far away from the welded location. The current design equations for RC column were used to predict the strengths of three specimens. Fig. 1. Square column with different detailing configurations of transverse reinforcement: (a) conventional closed hoops; (b) butt welded-hoops; (c) closed hoop with cross-ties. 293 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Table 2 Specified and actual values of concrete compressive strength, and yield stress of longitudinal and transverse reinforcements. Specimen ID Comp. Strength test (f0 c,specified = 70 MPa) CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 73 72 67 Longitudinal reinforcement SD685 (fy,spec = 685 MPa, fu,spec = 860 MPa) Size D29 (#9) fy,test 698 Transverse reinforcement SD785 (fy,spec = 785 MPa, fu,spec = 930 MPa) fu,test 920 Size fy,test fu,test D13 (#4) 886 868 886 1095 1104 1095 Table 3 Column parameters. Specimen ID CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 Longitudinal reinforcement SD685 Transverse reinforcement SD785 Axial load ratio n-size ql (%) Size, spacing (mm) qt (%) P/Agf0 c 16D29 2.87 D13(#4), 100 0.84 0.30 Fig. 3. Compressive test setup and failure of a concrete cylinder. Fig. 4. Fracture locations of welded steel bars after testing. 3.2.1. Nominal flexural and axial strength A stress–strain model and a equivalent rectangular compressive stress block have been proposed for flexural analysis and design of a RC section [15]. The strain distribution, the rectangular stress block, and the internal forces in the analysis of nominal flexural strength for the high-strength RC column are presented in Fig. 5. The parameters a1 and b1 influence the width and height of the stress block, respectively. The ACI ITG-4.3R-07 code [16] has recommended both parameters a1 and b1 , as shown in Eqs. (1) and (2), respectively. The ultimate compressive strain at the extreme fiber is defined as 0.003 [15,16]. The internal axial force and moment on the column section can be calculated using Eqs. (3) and (4). Eq. (5) gives the nominal axial strength for high-strength RC columns as recommended by ACI ITG-4.3R-07. The ratio of Fig. 5. Strain distribution, equivalent rectangular compressive stress block, and internal forces in analysis of flexural strength for a RC column section. 294 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Axial force ratio, P /A .f' g c 1.2 Interaction diagram Design point (0.138; 0.3) 0.9 0.6 0.3 0.138 0 0 -0.3 0.04 0.08 0.12 Moment ratio, M/b .h 2 c c .f'c 0.16 -0.6 P/A .f' g c Fig. 6. Column interaction diagram with axial compression ratio of 0.3. (a) the mean concrete compressive stress corresponding to the maximum axial load resisted by a concentrically loaded column, to the specified compressive strength of concrete, X1, shall be taken as 0.85 for concrete strengths up to 55 MPa, and shall be reduced continuously at a rate of 0.002 for each MPa of strength for concrete strength beyond 55 MPa. The factor X1, however, cannot be less than 0.7 [16]. Fig. 6 shows the dimensionless column interaction curve. The flexural strength of a column is defined by plotting the designed axial load on the column interaction curve. 0 For a1 : 0:70 6 a1 ¼ 0:85 0:0022ðf c 55Þ 6 0:85 0 ð2Þ P ¼ 0 ) Pn Cc C00s þ F0s þ Fs ¼ 0 ð3Þ h h h 1 00 C00s d CC b1 c Mh ¼ 0 ) Mn Fs d 2 2 2 2 2 ð4Þ For b1 : 0:65 6 b1 ¼ 0:85 0:0073ðf c 27:5Þ 6 0:85 X X ð1Þ 0 Po ¼ X1 f C ðAg As Þ þ f y As ð5Þ 3.2.2. Confinement In accordance with ACI 318-11, the amount of confinement shall be the greater of Eqs. (6) and (7). As mentioned, ACI 318-11 does not specify the limit of concrete strength and axial load in designing confinement reinforcement [6]. The new confinement provision given in ACI 318-14 directly considers the axial compression, high-strength concrete, and the number of longitudinal bars around the perimeter of the column core that are laterally supported by the corner of closed-hoops or cross-ties with seismic hooks [8]. Therefore, the greatest of Eqs. (6)–(8) shall be taken in determining the amount of confinement for a column with an axial compression load ratio greater than 0.3, or with a concrete compressive strength of 70 MPa or greater. Fig. 7 shows the required confining reinforcement ratio using ACI 318-11 and ACI 318-14 for the square column with a sectional dimension of 60 cm 60 cm and a longitudinal reinforcement ratio of 2.87% (16D29). It shall be noted that there were 4 longitudinal bars that were not supported by crossties (Fig. 1). By observing Fig. 7, it can be recognized that there is a significant difference between ACI 318-11 and ACI 318-14 in designing confinement reinforcement for high strength RC column under high axial compression loading. Moreover, an examination of confinement provisions for specimen CF-C-FT-0.3 (Fig. 7(b)) shows that the ACI 318-11 minimum requirements for confinement are questionable, particularly for highstrength concrete and steel in a column under high axial compression loading. P/A .f' g c (b) Fig. 7. Confining reinforcement ratio requirement of ACI 318-11 and ACI 318-14 for specimens: (a) CF-C-PT-0.3 and CF-C-W-0.3, (b) CF-C-FT-0.3. 0 Ash Ag f ¼ 0:3 1 c s bc Ach f yt ð6Þ 0 Ash f ¼ 0:09 c s bc f yt ð7Þ Ash Pu ¼ 0:2kf kn s bc f yt Ach ð8Þ 0 with : kf ¼ kn ¼ fc þ 0:6 P 1:0 175 nl ðnl 2Þ ð8aÞ ð8bÞ 3.2.3. Nominal shear strength The shear strength equations provided in ACI 318-11 [6], proposed by Sezen [17], and proposed by Ou and Kurniawan [3] were used to calculate the shear strength of the column specimens. The use of a yield stress limit of 420 MPa in calculating shear strength provided by transverse reinforcement is clearly specified in ACI 318-11 [6]. As reported by Alrasyid, a stress limit of 600 MPa is acceptable to calculate the shear strength provided by highstrength transverse reinforcement (steel bar with a specified yield stress of 785 MPa) [18]. The following equations present the nominal shear strength, Vn, used in this study. 295 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 ACI 318-11 [6]: Simple equation: with : a ¼ qﬃﬃﬃﬃ Pu Av f sh d 0 k f c bw d þ Vn ¼ Vc þ Vs ¼ 0:17 1 þ s 14Ag ðNÞ ð9Þ The detailed equation is taken as the smaller of Eqs. (10) and (11): Vn ¼ Vc þ Vs ¼ qﬃﬃﬃﬃ Vu d Av f sh d 0 bw d þ 0:16k f c þ 17qw Mm s 4h d P0 with : Mm ¼ Mu Pu 8 sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃ 0:29Pu Av f sh d 0 þ Vn ¼ Vc þ Vs ¼ 0:29k f c bw d 1 þ s Ag ðNÞ ð10Þ ð10aÞ ðNÞ ð11Þ sﬃﬃﬃﬃﬃﬃﬃﬃﬃ! Pu 1 0:85 0 ; Ag f c 06 Pu 0 6 0:6 Ag f c ð13aÞ To avoid shear failure, the designed shear strength must be greater than the shear force corresponding to the maximum moment probable, Mpr, at both ends of the column, Vpr. The total Mpr at both ends of a column divided by the clear height of the column, Lc, is equal to the shear force corresponding to the maximum moment probable, Vpr, (Vpr = 2Mpr/Lc). The Mpr occurs due to the strain-hardening behavior of longitudinal reinforcement. In accordance with the current concrete code [6,8], the yield stress of tensile reinforcement in Mpr calculation shall be multiplied by 1.25; for the New RC columns, the Mpr can be approximated as 1.3Mn [19]. The designed strengths of the specimens are listed in Table 4, in which the specified material strengths are used for all calculations. Sezen [17]: 0 qﬃﬃﬃﬃ 1 0 vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ u 0:5 f 2P Av f sh d cu B u C Vn ¼ Vc þ Vs ¼ k@ t1 þ qﬃﬃﬃﬃ A0:8Ag þ k 0 a=d s f c Ag 3.3. Test setup ðNÞ ð12Þ k ¼ 1 for displacement ductility; with : k ¼ 0:7 for displacement ductility; lD 6 2 lD P 6 ð12aÞ Ou and Kurniawan [3]: vﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ qﬃﬃﬃﬃ u 2Pu Av f sh d u 0 þ Vn ¼ Vc þ Vs ¼ 0:29ka f c bw dt1 þ qﬃﬃﬃﬃ 0 s a f c bc d ðNÞ ð13Þ The Multi-Axial Testing System (MATS) at the National Center for Research on Earthquake Engineering (NCREE) Taiwan was used to test the column specimens. The maximum axial and lateral loads that can be applied by MATS are 60 MN and 7 MN, respectively. The lateral load was applied by hydraulic actuators placed at the bottom of the MATS (Fig. 8(a)) using displacement control loading history (Fig. 8(b)). The test was terminated when a significant loss in axial loading capacity of the column was observed. Significant loss in axial loading capacity was determined to have occurred when there was a rapid reduction in axial loading capacity or the reported displacement from the actuators was suddenly greater than or equal to 20 mm. The tests were conducted under high axial compression, resulting in a noticeable P–D effect, and consequently, correction was required. Correction methodology is Table 4 Designed strengths. Specimen ID Nominal Shear Strength, Vn (kN) Flexural strength CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 Shear force due to Mn Shear force due to Mpr ACI 318-11 Mn (kNm) Mpr (kNm) Vnf (kN) Vpr (kN) Simple Detail Sezen Ou and Kurniawan 2131.33 2770.73 2368.14 3078.60 2272.10 3201.30 3372.02 3032 Fig. 8. (a) Test setup on MATS and (b) loading protocol. 296 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Fig. 9. Methodology of P–D effect correction. Fig. 10. Hysteretic behavior and envelope response for specimens: CF-C-PT-0.3 ((a) and (b)), CF-C-W-0.3 ((c) and (d)), CF-C-FT-0.3 ((e) and (f)). W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 detailed in Fig. 9, and calibrated complete hysteresis loops for the column specimens can be modified accordingly. 4. Test results Test results were examined based on visual observations and the data that were recorded by the MATS and strain gauges. As reported by Elwood et al. [7], a sufficient column should satisfy the seismic performance criteria, namely, the ultimate drift shall be at least 3% with lateral strength corresponding to ultimate drift of 80% maximum lateral strength. Meanwhile, ACI 374-1-0.5 specifies that the acceptance performance criteria for a RC column is when a column reaches the ultimate drift ratio of 0.035 and the lateral strength corresponding to ultimate drift of 75% maximum lateral strength [20]. The following sections present the test results and discussions. 4.1. Lateral strength–displacement relationships The lateral strength–drift ratio relationship and the envelope curve for each column are shown in Fig. 10, in which the P–D effect in every case has been eliminated. The important points for observation are the following: the yield point of longitudinal and transverse reinforcements, the idealized yield point, the peak lateral strength, the ultimate lateral strength and corresponding drift ratio, and ductility. The appearance of the columns post-testing is presented in Fig. 11. The following is method to obtain the idealized yield point: a straight line from zero was draw to intersect the lateral strength-drift curve at 70% of the maximum lateral strength. That straight line was then extended to the intersection (a) CF-C-PT-0.3 (b) CF-C-W-0.3 (c) CF-C-FT-0.3 297 with a horizontal line corresponding to the maximum lateral load, and then projected onto the horizontal axis to obtain the yield displacement (idealized Dy) [17]. Specimen CF-C-PT-0.3 had conventional closed-hoops with a 135-degree hook at one of the corners of each hoop and had longitudinal and transverse reinforcement ratios of 2.87% and 0.84%, respectively. Four longitudinal bars around the perimeter of column core were not laterally supported by cross-ties with seismic hooks. The applied axial compression load was 7861 kN (axial load ratio of 0.3). The specimen behaved elastic until the drift ratio reached 0.83%. The lateral stiffness degraded once drift ratios were greater than 0.83%, and the peak lateral strength of 2888 kN was reached at a drift ratio of 1.82%. The longitudinal and transverse reinforcement yielded at a drift ratio of 2% and 1.5%, respectively. The ultimate lateral strength was 2311 kN and occurred at a drift ratio of 3.17%. The specimen CF-C-PT-0.3 failed at a drift ratio of 4.0% when it lost its axial load-carrying capacity. Both longitudinal reinforcement buckling and anchor pullout failure of the transverse reinforcement were noticeable at the end of the test. The transverse reinforcement detailing of specimen CF-C-W-0.3 was similar to that of specimen CF-C-PT-0.3. As previously mentioned, the use of butt-welded hoops in specimen CF-C-W-0.3 is to reduce the reinforcement congestion in the plastic hinge region and to avoid the possibility of pullout failure of an anchor. The applied axial compression load was 7805 kN. Specimen CF-C-W0.3 behaved elastic at drift ratios up to 0.73%. The degradation of lateral stiffness started once the drift ratio exceeded 0.73%, and a peak lateral strength of 2715 kN was reached at a drift ratio of 1.36%. The yield points of longitudinal and transverse reinforcement were found at a drift ratio of 1.5%. The ultimate lateral strength of 2172 kN corresponded with a drift ratio of 3.05%. The test was stopped at a drift ratio of 5.12% when the axial loadcarrying capacity of specimen CF-C-W-0.3 was lost. At the end of the test, longitudinal reinforcement buckling was found, but no fracture occurred at the welded locations on the transverse reinforcement. The transverse reinforcement for specimen CF-C-FT-0.3 was based on the provisions given in ACI 318-11. The applied axial compression load was 7208 kN. As has been observed in specimen CF-C-PT-0.3 or CF-C-W-0.3, the elastic behavior of specimen CF-CFT-0.3 was found at drift ratios in the range of 0–0.8%. However, a peak lateral strength of 2654 kN was reached at a drift ratio of 1.17%. It was found that the longitudinal and transverse reinforcements yielded at a drift ratio of 1.5%. Finally, an ultimate lateral strength of 2123 kN occurred at a drift ratio of 2.31%, and the specimen lost its axial load-carrying capacity at a drift ratio of 3.00%. Longitudinal reinforcement buckling and anchor pullout failure of the transverse reinforcement were clearly identified. The idealized yield point, the peak lateral strength, the ultimate lateral strength and corresponding drift ratio, and ductility for each column specimen are summarized in Table 5. 4.2. Crack patterns (d) CF-C-PT-0.3 (e) CF-C-W-0.3 (f) CF-C-FT-0.3 Fig. 11. The final appearance of specimens at the end of the test ((a) to (c)) and the appearance of concrete cores ((d) to (f)). Flexural cracking near the top of the column occurred on specimens CF-C-PT-0.3, CF-C-W-0.3, and CF-C-FT-0.3 at drift ratios of 0.25%, 0.25%, and 0.375%, respectively. The crack width increased as drift increased, and Table 6 shows the crack width occurring at each drift ratio. In general, the shear cracking appeared as diagonal cracking developed from flexural cracking at the top and bottom of the column. The first shear cracking occurred in specimens CF-C-PT-0.3 and CF-C-W-0.3 at drift ratios of 0.75% and 0.375%, respectively, while the first shear cracking in specimen CF-C-FT0.3 appeared at a drift ratio of 0.25% (before flexural cracking appeared). For specimen CF-C-PT-0.3, the concrete shell started crushing at the top and bottom of the column at a drift ratio of 298 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Table 5 Results of force–displacement relationships. Specimen ID Idealized yield drift % Peak lateral strength (kN) Ultimate lateral strength (kN) Ultimate drift % Ductility CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 0.57 0.57 0.56 2888 2715 2654 2311 2172 2132 3.17 3.05 2.31 5.56 5.35 4.12 Table 6 Maximum crack widths of specimens. Drift (%) CF-C-PT-0.3 Flexural crack width (mm) Shear crack width (mm) Flexural crack width (mm) Shear crack width (mm) Flexural crack width (mm) Shear crack width (mm) 0 0.25 0.375 0.50 0.75 1.0 1.5 2.0 N.A 0.05 0.08 0.10 0.30 0.60 1.20 2.50 N.A N.A N.A N.A 0.20 0.30 0.60 2.20 N.A 0.10 0.25 0.25 0.60 0.65 0.90 3.00 N.A N.A 0.05 0.05 0.30 0.45 0.65 4.00 N.A N.A 0.10 0.25 0.25 0.40 1.20 2.00 N.A 0.05 0.05 0.10 0.25 0.35 0.70 5.00 Crack angle (Degree) 60 50 CF-C-W-0.3 CF-C-FT-0.3 CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 40 30 20 10 0.25 0.5 0.75 1 1.25 1.5 1.75 2 Drift ratio (%) Fig. 12. Crack angle–drift ratio relationships. 1.5% and the concrete shell gradually spalled off as the drift ratio increased. In addition, the diagonal cracks at the middle of the column started appearing at a drift ratio of 0.75%, and as the drift ratio further increased, several diagonal cracks appeared at the top and bottom of the column. More concrete shell at the bottom of the column spalled off at a drift ratio of 2%. For specimen CF-C-W-0.3, the concrete shell started crushing at a drift ratio of 1%. The diagonal cracks were noticeable once the drift ratio reached 1.5%. More concrete shell crushed and more diagonal cracks appeared at the middle of the column as the drift ratio increased. The specimen CF-C-FT-0.3 showed that the concrete shell at the top and bottom of column crushed at a drift ratio of 1%. Diagonal cracks were seen once the drift ratio reached 2%, and more diagonal cracks appeared as the drift ratio increased. The crack width that occurred at each drift ratio is listed in Table 6, and the crack angle–drift relationships are shown in Fig. 12. length of longitudinal reinforcement in specimen CF-C-PT-0.3 is shorter than that of in specimen CF-C-W-0.3. This is because the number of layers of conventional closed-hoops is slightly greater than that of butt-welded hoops. A small clear spacing of transverse reinforcement improves the effective confinement pressure in the column core. More information regarding confinement mechanisms can be found in existing literature [21]. Regardless of the performance aspects of specimen CF-C-W-0.3, butt-welded hoops can be an alternative transverse reinforcement for high-strength concrete columns in seismic regions. As presented in Fig. 11(e), no fractures on the welded location were found. The test results of specimen CF-C-FT-0.3 prove that the confinement provisions provided in ACI 318-14 should be followed. Although, the clear spacing of transverse reinforcement and transverse reinforcement ratio of specimens CF-C-PT-0.3 and CF-FT-FT-0.3 were identical, the pullout failures of ties with a 90-degree hook were found in specimen CF-C-FT-0.3. According to ACI 318-2014, the cross-ties with a 90-degree hook are not as effective as cross-ties with a 135-degree hook [8]. The test results prove that the number of longitudinal bars laterally supported by crossties with seismic hooks influence the performance of column. Specimens CF-C-PT-03 and CF-C-W-0.3 having 16 longitudinal bars with 4 longitudinal bars laterally un-supported still met the performance criteria required by ACI 374, while specimen CF-C-FT-0.3 did not perform well as a ductile RC column. Specimen CF-C-FT-0.3 might achieve better performance if the provided transverse reinforcement ratio was larger than 0.84% or the provided crossties had a bend of 135°. The smallest and greatest flexural cracks were found on specimen CF-C-FT-0.3 and CF-C-W-0.3, respectively. By contrast, the greatest shear crack was found on CF-C-FT-0.3. It can be seen in Fig. 12 that the trends of the crack angle–drift curves are similar. The crack angle decreases as drift ratio increases. 5. Comparison between measured and designed strengths 4.3. Discussion The results show that behavior of specimen CF-C-PT-0.3 was more desirable than that of specimen CF-C-W-0.3. The peak lateral strength, ultimate lateral strength and corresponding drift ratio, and ductility of specimen CF-C-PT-0.3 were greater than those of specimen CF-C-W-0.3, although the two columns had the same transverse reinforcement ratio. It can be known that the unbraced The final step of this study is to evaluate the current strength design equations for RC column. The peak lateral strength of each column specimen is verified using the ACI 318-11 simplified shear strength equation (Vn,1 in Table 7, by Eq. (9)), ACI 318-11 detailed equations (Vn,2 in Table 7, by Eqs. (10) and (11)), Sezen’s shear strength equation (Vn,3 and Vn,4 in Table 7, by Eq. 12(a) for lD 2 and lD > 2), shear strength proposed by Ou and Kurniawan (Vn,5 in Table 7, by Eq. (13)), and shear force corresponding to 299 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Table 7 Comparison between peak lateral and designed strengths. Specimen Vtest/Vn,1 Vtest/Vn,2 Vtest/Vn,3 Vtest/Vn,4 Vtest/Vn,5 Mtest/Mn Vtest/Vn,f Vtest/Vpr CF-C-PT-0.3 CF-C-W-0.3 CF-C-FT-0.3 1.27 1.19 1.17 0.90 0.85 0.83 0.86 0.81 0.79 0.94 0.83 0.79 0.95 0.90 0.88 1.22 1.15 1.12 1.22 1.15 1.12 0.94 0.88 0.86 (a) (b) (c) Fig. 13. The envelope measured and calculated shear strengths for specimens: (a) CF-C-PT-0.3, (b) CF-C-W-0.3, (c) CF-C-FT-0.3. 300 W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 the nominal flexural strength given by ACI ITG-4.3R-07 (Vn,f in Table 7). Vn,4 was obtained by calculating k corresponding to the ratio of drift at peak lateral load to idealized yield drift (D,peakload/ Dy). Fig. 13 shows the envelope measured and designed shear strengths. The ACI 318-11 simplified shear equation only predicts the shear strength that is smaller than the peak lateral strength of specimens obtained from test results. In addition, ACI 318-11 specified that the maximum yield stress of transverse reinforcement for shear strength design is 420 MPa. Therefore, designing the shear strength of the New RC columns using the ACI 381-11 simplified shear equation may be too conservative. On the other hand, the use of the ACI 318-11 detailed equations tends to overestimate the nominal shear strength. Table 4 shows that nominal shear strengths from the ACI 318-11 detailed equations were higher than both the peak lateral force and the shear force corresponding to the maximum probable flexural strength, although the yield stress used in the calculation was 420 MPa. The shear force corresponding to the flexural strength of a column, which is based on ACI ITG-4.3R-07, was in agreement with the ultimate shear force (shear force corresponding to ultimate drift ratio). By observing Mtest/Mn, it can be concluded that the use of the equivalent rectangular compressive stress block recommended in ACI ITG-4.3R-07 is acceptable for determining the nominal flexural strength of the New RC column. Furthermore, the measured lateral strengths are smaller than the calculated nominal shear strengths based on the shear strength equations proposed by Sezen (Eq. (12)) and Ou and Kurniawan (Eq. (13)). This result occurs because the type of failure in the three column specimens was flexural-shear failure. The shear strength corresponding to the maximum moment probable was close to the nominal shear strength proposed by Sezen (Eq. (12)) and Ou and Kurniawan (Eq. (13)). It should be noted that the shear strength corresponding to the maximum probable moment tends to be rare occurrence in a high-strength column under high axial compression loading. It is well known that the higher axial compression loading can reduce the moment capacity, as has been presented in the axial-moment interaction diagram. The axial load increases the compression stress in compression zone so that the distance between the resultant compression force and the resultant tension force from tension steel reinforcement (level arm) and the height of tension zone decrease. Therefore, the stress of tension longitudinal reinforcement may be equal to or smaller than its yield stress once the flexural capacity of column is reached. In compression zone, the stress of compression reinforcement tends to be far from yield since compression stress acting on compression zone is received by concrete and compression reinforcement. Meanwhile, the test results show that the yielding longitudinal reinforcement almost corresponded with the peak lateral strengths. The post yield lateral force might have increased if the tensile stress in the longitudinal reinforcement increased to its ultimate stress (1.25 times yield stress). Therefore, at high axial loading level, this shows that applying factor 1.25 Vpr does not improve column flexural strength and will overestimate the actual maximum lateral strength corresponding to flexural strength. The nominal shear strength equation proposed by Sezen describes shear strength degradation well. It can be seen that the curves of the shear strengths calculated using Sezen’s equation with k larger than 1, and the experimental curves of the specimens tend to have a similar trend. 6. Conclusions Based on the experimental results and analysis that are presented in this paper, the following conclusions can be drawn: 1. The performance of specimen with butt-welded hoops (CFC-W-0.3) tended to be similar to that of specimen with conventional closed-hoops (CF-C-PT-0.3). Therefore, the use of high strength butt-welded hoops (SD785) for confining reinforcement in the New RC columns is acceptable. 2. The ultimate drift ratio of a specimen having conventional closed-hoop transverse reinforcement and cross-ties with 90-degree and 135-degree hooks (CF-C-FT-0.3) was only 2.31%; therefore, the ultimate drift ratio of CF-C-FT-0.3 did not reach the minimum criteria for a ductile RC column. 3. The use of cross-ties with 90-degree hook increases the effectiveness factor (kn). In accordance with new confinement provisions given in ACI 318-14, the improvement of effectiveness factor (kn) increases the required confinement. As shown in Fig. 7(b), the amount of confinement in CF-CFT-0.3 only satisfy the ACI 318-11 minimum confinement. In addition, ACI 318-11 does not specify clearly the limits of concrete strength and axial compression ratio for designing confinement reinforcement. 4. The examination of confinement equations and test results prove that the confinement provisions given in ACI 318-14 should be followed. 5. The ACI 318-11 simplified shear strength equation only predicts shear strength before achieving the peak lateral force, for all specimens. By contrast, the ACI 318-11 detailed equations may overestimate the shear strength of the New RC columns. 6. Equivalent rectangular compressive stress block recommended in ACI ITG-4.3R-07 is a conservative method for predicting the nominal flexural strength of the New RC columns. 7. The nominal shear strength proposed by Sezen (Eq. (12)) and Ou and Kurniawan (Eq. (13)) were close to the shear strength corresponding to the maximum probable flexural strength, Vpr. Meanwhile, the peak lateral strengths from the test the results were smaller than the nominal shear strengths calculated using Eqs. (12) and (13) due to the three column specimens experiencing flexural-shear failure. 8. The shear corresponding to the maximum probable moment tends to be rare occurrence in columns under high axial compression loading. At high axial loading level, the axial load increases the compression stress in compression zone. This reduces the distance between the resultant compression force and the resultant tension force from tension steel reinforcement (level arm) and the height of tension zone. The stress of tension longitudinal reinforcement can be either equal to or smaller than its yield stress once the flexural strength is column is reached. In addition, the possibility of compression reinforcement reaching its yield stress is low, since compression stress acting on compression zone is received by concrete and compression reinforcement. According to the test results, the yielding longitudinal reinforcement almost corresponded with the peak lateral load. The post yield lateral force might have increased if the stress of the longitudinal reinforcement increased to its ultimate stress (1.25 times yield stress). Therefore, at high axial loading level, applying factor 1.25 does not improve column flexural strength and will overestimate the actual maximum lateral force corresponding to flexural strength. 9. The nominal shear strength equation proposed by Sezen (Eq. (12)) describes shear strength degradation well. As shown, the trend of experimental curves was similar to that of the curves obtained using shear strengths calculated by Sezen’s equation with k larger than 1. W.-C. Liao et al. / Engineering Structures 153 (2017) 290–301 Acknowledgments The research described in this paper was sponsored by the Ministry of Science and Technology (MOST) of Taiwan under Grant No. 105-2221-E-002-057-MY2 and 106R890803, and the National Centre for Research on Earthquake Engineering (NCREE) Taiwan is gratefully acknowledged. The opinions expressed in this paper are those of the authors and do not necessarily reflect the views of the sponsor. The authors would like to thank Professor ShyhJiann Hwang from Department of Civil Engineering at National Taiwan University. The Doctoral scholarships for two academic years (2016/2017 and 2017/2018) received by the second author are provided by Centre for Earthquake Engineering Research (CEER) at National Taiwan University. References [1] Lee HJ, Chen JH. Testing of mechanical splices for grade 685 steel reinforcing bars. NCREE 2014. TTK report; 2014. [2] Ou YC, Kurniawan DP. Shear behavior of reinforced concrete columns with high-strength steel and concrete. ACI Struct J 2015;112(1):35–46. [3] Ou YC, Kurniawan DP. Effect of axial compression on shear behavior of highstrength reinforced concrete columns. ACI Struct J 2015;112(2):209–20. [4] Paultre P, Eid R, Langlois Y, Levesque Y. Behavior of steel fiber-reinforced high strength concrete columns under uniaxial compression. J Struct Eng 2010;136 (10):1225–35. [5] Perceka W, Liao WC, Wang YD. High strength concrete columns under axial compression load: hybrid confinement efficiency of high strength transverse reinforcement and steel fibers. Materials 2016;9(4). [6] ACI Committee 318. Building code requirements for structural concrete (ACI 318-11) and commentary. American Concrete Institute; 2011. 301 [7] Elwood KJ, Maffei J, Riederer KA, Telleen K. Improving column confinement part I: Assessment of design provisions. Concr Int 2009:32–9. [8] ACI Committee 318. Building code requirements for structural concrete (ACI 318-14) and commentary. American Concrete Institute; 2014. [9] Elwood KJ, Maffei J, Riederer KA, Telleen K. Improving column confinement part 2: proposed new provisions for the ACI 318 building code. Concr Int 2009:41–8. [10] Sokoli D, Ghannoum WM. High-strength reinforcement in columns under high shear stresses. ACI Struct J 2016;113(3):605–14. [11] Ou YC, Alrasyid H, Haber ZB, Lee H-J. Cyclic behavior of precast high-strength reinforced concrete column. ACI Struct J 2015;112(6):839–50. [12] ASTM. Standard test method for compressive strength of cylindrical concrete specimens. ASTM International Designation: C39/C39M. [13] ASTM. Standard test methods for tension testing of metalic materials. ASTM International Designation: E8/E8M-09. [14] ASTM. Standard test methods and definitions for mechanical testing of steel products. ASTM International Designation: A370-08a. [15] Ozbakkaloglu T, Saatcioglu M. Rectangular stress block for high-strength concrete. ACI Struct J 2004;101(4):475–83. [16] ACI Innovation Task Group 4. Report on structural design and detailing for high-strength concrete in moderate to high seismic applications (ITG-4.3R07). American Concrete Institute; 2007. [17] Sezen H. Seismic behavior and modeling of reinforced concrete building columns. PhD thesis. Berkeley: Engineering – Civil and Environmental Engineering, University of California; 2002. [18] Alrasyid H. Seismic shear behavior of high-strength reinforced concrete columns. Doctoral dissertation. Department of Civil Engineering-College of Engineering, National Taiwan University of Science and Technology; 2015. [19] National Centre for Research on Earthquake Engineering. Design handbook for high strength reinforced concrete structural members; 2015 [in Chinese]. [20] ACI Committee 374. Acceptance criteria for moment frames based on structural testing and commentary (ACI 374-1-0.5). American Concrete Institute; 2005. [21] Mander JB, Priestley MNJ, Park R. Theoretical stress-strain model for confined concrete. J Struct Eng 1988;114(8):1804–26.

1/--страниц