Experimental study on rectangular CFT columns with high-strength concrete

Journal of Constructional Steel Research 63 (2007) 37–44 www.elsevier.com/locate/jcsr

Experimental study on rectangular CFT columns with high-strength concrete Dung M. Lue ∗ , Jui-Ling Liu, Tsong Yen Department of Civil Engineering, National Chung-Hsing University, 250 Kuo-Kuang Road, Taichung 402, Taiwan Received 3 November 2005; accepted 17 March 2006

Abstract In the 1999 AISC-LRFD, the in-filled concrete strength ( f c0 ) of concrete-filled tube (CFT) columns is limited to a maximum value of 55 MPa (N/mm2 ). That limiting value is raised to 70 MPa in the 2005 AISC-LRFD. This study aims to assess if the LRFD CFT column formulas are applicable to intermediate to long rectangular columns with higher concrete strengths. Twenty four specimens with f c0 varying between 29 and 84 MPa were tested. Various formulas and relevant provisions for CFT columns as specified in the major design codes including AISC-LRFD, EC 4, AS-5100, and CSA S16-01 were examined and compared. The design CFT strength (Pu ) predicted by the AISC-LRFD formulas and the test results (Ptest ) were found to be in good agreement. The higher f c0 limiting value of 70 MPa proposed in the 2005 AISC-LRFD appears acceptable. The test results reveal that the 1999 AISC-LRFD design strengths are conservative and tend to penalize these CFT columns with higher concrete strength. c 2006 Elsevier Ltd. All rights reserved.

Keywords: Concrete-filled tube; CFT; LRFD; High-strength concrete; Column

1. Introduction 1.1. Current design codes Concrete-filled tube (CFT) columns are being increasingly used as structural members since filling the steel section with concrete results in an increase in both strength and ductility without increasing the section size. Design criteria for CFT columns are available in various major design codes such as the AISC-LRFD [1], the ACI 318-05 [2], the Architectural Institute of Japan [3], the European Code EC 4 [4], the British Standards BS 5400 [5] and the Australian Standards AS-5100.6 [6]. In the U.S., prior to the publication of the 1986 AISC-LRFD, the design of composite columns referred to the ACI Code. In 1979, the SSRC Task Group 20 [7] proposed a specification for the design of steel–concrete composite columns to be included in the AISC Specifications. Their report was written in the format of the ultimate strength method rather than the original allowable stress method as proposed by Furlong [8]. The 1986 AISC-LFRD specifies strength limits on both steel and concrete. The upper limit for steel strength (Fy ) was set at 380 MPa, which corresponds to a concrete strain of around ∗ Corresponding author.

E-mail address: [email protected] (D.M. Lue). c 2006 Elsevier Ltd. All rights reserved. 0143-974X/$ - see front matter doi:10.1016/j.jcsr.2006.03.007

0.18%. The code also requires that a compressive concrete strength ( f c0 ) of 21–55 MPa be specified. The 1999 AISCLRFD raised the Fy value to 415 MPa. The new 2005 AISCLRFD [9] raises the material strengths even higher to f c0 = 70 MPa and Fy = 525 MPa. It also modifies the minimum steel wall thickness and contains design provisions for both round and rectangular shapes. For the first time, the 2005 AISCLRFD adopts the concept of effective stiffness with different adjustment coefficients depending on the situation. It keeps the minimum steel ratio at 4% for regular encased columns, but lowers it to 1% for CFT columns, compared with the 1999 AISC-LRFD. Moreover, it revises the resistance factor to 0.75 from 0.85. The newer design specifications included in the 2005 AISC-LRFD appears meriting from both the 2005 ACI 318 and the 2004 EC 4. 1.2. Previous related studies A number of experimental and theoretical studies on the related subject have been carried out since the early 1960s. Test variables considered by the previous investigators include sectional dimensions, width-to-thickness ratio (B/t), material strengths (steel and concrete), structural stiffness, percentage of steel area, residual stress, effective length, slenderness ratio, effect of confinement, loading type, and load eccentricity.

38

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

Nomenclature Ac Ag As b D Es Ec EI eff f c0 fc Fcr Fy Ic Is K L Ptest Po Pn Pu r t wc λc φc

Area of concrete Gross area Area of the steel section Width in rectangular section Outside diameter of circular hollow section Modulus of elasticity of steel (E) Modulus of elasticity of concrete Effective moment of inertia rigidity of composite section Specified compressive strength of concrete at 28 days Compressive concrete stress before ultimate load is reached Critical stress Specified minimum yield stress of steel Moment of inertia of the concrete section Moment of inertia of steel shape Effective length factor for prismatic member Laterally unbraced length of member at the point of load Test failure load Composite compressive strength based on ACI 318-05 Nominal axial strength based on AISC-LRFD (1999, 2005) Design axial strength based on AISC-LRFD (1999, 2005) Radius of gyration Wall thickness of hollow structural steel section Weight of concrete per unit volume Column slenderness parameter Resistance factor for compression

Knowles and Park [10] studied 12 circular and 7 square columns by considering various D/t and l/d ratios. Their results demonstrated that concrete confinement increases the structural capacity of circular tubes. However, this beneficial effect was not noted in square or rectangular shapes. The SSRC Task Group 20 [7] proposed a design specification for steel–concrete composite columns to be included in the AISC Specifications, which was based on the ultimate strength design format. Shakir-Khalil [11–13] conducted a series of tests of CFT rectangular columns on compression and bending from 1989 to 1996. The experimental results indicated that BS 5400 gave safe capacity prediction for the CFT columns subjected to axial loading or uniaxial bending about the major axis. Kenny et al. [14] examined the limiting steel yield stress (Fy ) of 380 MPa as specified in the 1986 LRFD Specification. They reported that the 380 MPa yield limit could be increased to 550 MPa. Bradford [15] proposed a design model for calculating the design strength of slender rectangular CFT columns. Schneider [16] tested 14 specimens with various shapes and investigated the effects of steel tube shape and wall thickness on the ultimate strength of composite columns.

They also addressed the concrete confinement effect. Kilpatrick and Rangan [17] tested 41 CFT columns with high-strength steel tubes (Fy > 400 MPa) filled with concrete with a compressive strength of 58 and 96 MPa to investigate the single and double curvature bendings. Zhang and Shahrooz [18] provided the measured results from past studies and from their own specimens, where they gave the comparison between ACI and AISC for CFT columns. Shanmugam and Lakshmi [19] gave a review of the research carried out on composite columns with emphasis on experimental and analytical work. Han [20] studied 24 short axially-loaded CFT rectangular specimens considering the two major parameters, constraining factor and tube width limit. By comparison, he reported that the loading capacity of the concrete-filled rectangular stub columns could be conservatively predicted by using the AISC-LRFD, AIJ, EC 4, and GJB4142 [21] recommendations. Mursi and Uy [22] studied 3 slender CFT square columns filled with high strength concrete ( f c0 = 65 MPa), and 3 HSS columns. They made a comparison of the design recommendations for the strength evaluation of slender composite columns with thin-walled steel sections. Liu et al. [23] studied 22 CFT rectangular stub columns (Fy = 550 MPa, and f c0 = 70–82 MPa) subjected to concentric loading. They further compared the ultimate loads obtained from experiments with the values calculated based on the EC 4, the AISC-LRFD and the ACI. Their comparison shows that the EC 4 closely predicts the ultimate load with a difference of 6%, while the AISC-LRFD and the ACI underestimated the critical load by more than 14%. Viest [24] reviewed the historical development of the design requirements of composite structures made of steel and concrete. Melcher and Karmazinova [25] presented the analysis of CFT columns with high strength concrete. Sakino et al. [26] studied 114 specimens to investigate the behavior of centrallyloaded short CFT columns, and proposed the formulae for estimating the ultimate axial compression capacities of CFT columns. Liu [27] studied 22 CFT rectangular stub columns filled with high-strength concrete. His comparison indicates that the current ACI and AISC specifications conservatively estimate the failure loads of the specimens by 9% and 11%, respectively. The EC 4 method gives a close and conservative estimate of the ultimate capacities with a difference of 1%. 1.3. Current study The recently developed high strength concrete has lured industrial enterprises and researchers into the field of high strength composite construction. The high strength concrete offers benefits in both strength and stiffness. The majority of the previous studies focus on circular and square sections with lower concrete strength. Research work related to rectangular sections with higher concrete strength is still limited and therefore deserves further investigation. This study examines the ultimate strength of CFT rectangular columns under axial compressive loading experimentally. The following various concrete compressive strengths ( f c0 ) were considered: 29, 63, 70, and 84 MPa, while the average steel yield stress was kept at

39

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

379 MPa. The objectives of this study are: (1) to experimentally investigate the behaviour of rectangular CFT columns with high strength concrete subjected to axial loading; (2) to compare the test results with those predicted using AISC-LRFD (1999, 2005), 2004 EC 4, 2004 AS-5100, and 2001 CSA S16-01 [28]; (3) to evaluate the current AISC-LRFD CFT column formulas to assess if they are still applicable to the cases with f c0 value exceeding the current limit of 55 MPa or 70 MPa. 2. Brief review of code-specified design strengths for CFT columns

Limitations of the 2005 AISC-LRFD include: • As ≥ 0.01A g , Fy ≤ 525 MPa (75 ksi) • Normal weight concrete strength 21 ≤ f c0 ≤ 70 MPa (3 ≤ f c0 ≤ 10 ksi) • The maximum b/t ratio for a rectangularpHSS used as a composite column shall be equal to 2.26 E/Fy . Higher ratios are permitted when their use is justified by testing or analysis. • The maximum D/t ratio for a round HSS filled with concrete shall be 0.15E/Fy . Higher ratios are permitted when their use is justified by testing or analysis.

For background and familiarity purposes, a brief review of the determination of CFT column strength is provided as follows.

2.3. The 2005 ACI 318-05

2.1. The 1999 AISC-LRFD

where Po is the squash load as specified in the ACI Code.

The design compressive strength, Pu = φc Pn , in which φc = 0.85. Computation of the nominal strength (Pn ) of an axiallyloaded CFT column is similar to a regular steel compression member, except the material yield strength and stiffness need to be modified to account for the steel and concrete components in the CFT column. The nominal strength (Pn ) is given as: (1)

Pn = Fcr As

where Fcr is the critical stress and As the gross area of the steel shape, pipe, or HSS. Limitations of the 1999 AISC-LRFD consist of: • As ≥ 0.04A g , and Fy ≤ 415 MPa (60 ksi) • In-filled concrete strength, 21 ≤ f c0 ≤ 55 MPa (3 ≤ f c0 ≤ 8 ksi) • The minimum wall thickness of the structural p steel pipe or HSS filled with concrete shall be equal to b Fy /3E for each p face of width b in rectangular sections and D Fy /8E for circular sections of outside diameter D. 2.2. The 2005 AISC-LRFD The design compression on strength, Pu = φc Pn , for axially-loaded filled composite columns shall be determined for the limit state of flexural buckling based on column slenderness as follows: φc = 0.75 (1) Where Pe ≥ 0.44Po :    Po Pn = Po 0.658 Pe (2) (2) Where Pe < 0.44Po : Pn = 0.877Pe

(3)

where Po = As Fy + 0.85Ac f c0

(rectangular section)

Pe = π (EI eff )/(K L) EI eff = E s Is + C3 E c Ic   As ≤ 0.9 C3 = 0.6 + 2 Ac + As 2

2

(4) (5) (6) (7)

Po = As Fy + 0.85 f c0 Ac

(8)

2.4. The 2004 EC 4 Po = As Fy /γs + 0.85Ac f c0 /γc ,

γs = 1.1, γc = 1.5

(9)

where γs is the partial safety factor for structural steel (=1.1) and γc the partial safety factor for concrete (=1.5), and Po the squash load as specified in the EC 4 Code. 3. The experimental work Twenty two 1855 mm long rectangular hollow steel sections of 150 × 100 × 4.5 mm with mean yield strength (Fy ) = 379.8 MPa were prepared and divided into four groups. The sections in each group are filled with approximate concrete strength f c0 of 29, 63, 70, and 84 MPa, respectively. Another group including two rectangular Hollow Structural Section (HSS) columns was also made for comparison purposes. Tests were conducted on all of these CFT and HSS specimens. Sectional properties of the specimens are summarized in Table 1, in which the following group notation holds: C0k 21–C0k 2-2 denotes the 2 rectangular HSS columns without in-filled concrete, C4k 4-1–C4k 4-4 denotes the 4 specimens with the in-filled concrete strength f c0 around 29 MPa, C9k 61–C9k 6-6 denotes the 6 specimens with the f c0 value around 63 MPa, and C10k 6-1–C10k 6-6 and C12k 6-1–C12k 6-6 denote the 6 specimens with the f c0 value around 70 and 84 MPa, respectively. The yield stress and elastic modulus of steel were obtained from the average value of 3 coupon tests. The test coupons were cut from the steel section, and the testing was done based on the ASTM-A370 [29] procedure. Table 2 gives the mixtures, proportions, and various cylinder strengths. Four different concrete mixtures were designed with the average compressive cylinder strengths at 28 days of 29, 63, 70 and 84 MPa, respectively. In each batch, 6 standard concrete cylinders were cast and tested to achieve a specific 28-day strength. Concrete strengths and concrete elastic moduli were obtained based on ASTM-C39 [30] and ASTM-C469 [31] procedures. The wall thickness of 4.5 mm used in the specimens satisfies the AISC minimum requirement which calls for 4.07 mm (1999 LRFD) or 3.12 mm (2005 LRFD).

40

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

Table 1 Section properties of CFT specimens Specimen group number and notation

Section dimensionsd

Parameters and properties As /A g (%)

b/t a

K Lb (mm)

K L/r

f c0 (MPa)

λc c

1

C0K 2-1–2

14.5

33.4

1855

45.25

0

0.68

2

C4K 4-1–4e

14.5

33.4

1855

45.25

29

0.69

3

C9K 6-1–6

14.5

33.4

1855

45.25

63

0.77

4

C10K 6-1–6

14.5

33.4

1855

45.25

70

0.78

5

C12K 6-1–6

14.5

33.4

1855

45.25

84

0.82

a b c d

b/t = width-to-thickness ratio. K L/r = slenderness ratio, K = 1.0 is assumed. λc = Modified column slenderness parameter based on AISC-LRFD (1999). Rectangular HSS 150 × 100 × 4.5 mm, specimen length 1855 mm, A g = 150 cm2 , Ac = 128.3 cm2 , As = 21.7 cm2 , thickness t = 4.5 mm, steel average yield stress Fy = 379.8 MPa, tensile stress Fu = 452.0 MPa, elastic modulus E s = 171 736 MPa. e C4K 4-1, specifies the first one of 4 CFT specimens with f 0 around 28 MPa (4 ksi). c Table 2 Concrete mix design and physical properties Design strength

Water

28 MPa 62 MPa 69 MPa 83 MPa

211.8b 186.5 169.9 155.6

Cement 231.6 326.4 424.6 453.8

Fly ash 66.2 93.3 121.3 129.6

Slag 33.1 46.6 60.7 64.8

Sand 806.5 743.9 702.6 702.6

Aggregate 943.2 943.2 890.8 890.8

Super plasticizer 2.0 8.4 17.0 18.2

W/Ba ratio

f c0 28-day

E c 28-day

0.64 0.40 0.28 0.24

29c

23047c 31374 31941 32845

63 70 84

a W/B: water to binder ratio. b (kgf/m3 ). c (MPa).

Two bearing plates (340 × 340 × 20 mm) were welded at the top and bottom ends of each specimen with eight spotwelded stiffeners to provide the rigidity at the specimen ends and to make sure that the plane remains plane at the ends of the specimen when the rotation occurs at the onset of buckling. A hole with a 70 mm diameter was cut at the centre of the top bearing plate from where the concrete was cast into the CFT specimen. Concrete was poured and well compacted by rubber hammers. The CFT specimens were placed upright for curing and after the curing a small amount of harden-setting gypsum was poured on the top of CFT specimen. The in-filled gypsum was used to compensate the minor shrinkage of concrete, made the top end surface of CFT specimen in full contact with the bearing plate of the MTS machine, and ensured that the applied load transferred to the concrete core and hollow steel section uniformly during the test. The CFT specimen tests were conducted through the use of the 600-metric ton MTS machine in the Structures and Materials Laboratory. Figs. 1 and 2 depict the strain gauge locations and test specimen setups, respectively. Four strain gauges were placed at the mid-height section of the specimen to measure the longitudinal strains. Two strain gauges were used to determine

Fig. 1. Schematic locations of strain gauges.

the average longitudinal strain of the specimen. The average longitudinal strains located on the axis of buckling (axis y–y) were used to draw the (P–ε) curves. The remaining two strains

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

41

Photo 1. CFT specimen after test. Fig. 2. Front and side views of CFT specimen.

Fig. 3. P–δ diagram for specimen C10K 6-1.

Photo 2. Failed specimens C9K 6-1–C9K 6-6.

in Figs. 3 and 4, respectively. Representative failed specimens are shown in Photos 1 and 2. 4. Discussion of the test results Fig. 4. P–δ diagram for specimen C12K 6-3.

located in the axis perpendicular to the axis of buckling were used to verify the observed buckling load. The longitudinal deformation was also obtained from the 600-metric tonne MTS machine. Two roller supports were placed at the top and bottom ends of the specimen as shown in Fig. 2. The concentric load, measured by the MTS machine, was slowly applied to the specimen according to the ASTM procedure such that the buckling behaviour of the CFT columns could be observed. Two typical axial load–deflection curves (P–δ) for 2 specimens with concrete strengths 70 and 84 MPa are plotted

4.1. Failure loads Maximum and average failure loads obtained from the tests, (Ptest and avg. Ptest ), are summarised in Table 3. The standard deviation and the coefficient of variation for each group are relatively small, which implies that the experimental results are quite reliable. Three failure modes were observed during the tests: global buckling, local buckling, and mixed global–local buckling. For most of the specimens with normal-strength concrete they failed in global buckling. A few cases with high-strength concrete were found to fail in local buckling, immediately followed by global buckling. Pure local buckling

42

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

Table 3 Test results for axial compression loads/failure loads Specimen notation

Ptest No. 1

C0K 2-1–2 C4K 4-1–4

1060.0c

C9K 6-1–6 C10K 6-1–6 C12K 6-1–6 a b c d

1344.8 1756.1 1894.6 2066.1

No. 2

No. 3

No. 4

No. 5

No. 6

1059.4 1281.3 1702.8 1889.2 2196.4

NAd

NA 1367.6 1737.5 1891.6 2090.1

NA NA 1669.2 1862.3 2006.7

NA NA 1705.8 1889.8 2083.5

1320.2 1762.7 1885.6 2096.1

Ptest (avg.)

Std. Dev.a

C.O.V.b (%)

1059.7c 1328.5 1722.3 1885.5 2089.8

0.30 31.98 32.86 10.73 56.14

0.03 2.41 1.91 0.57 2.69

Std. Dev.—Standard deviation. C.O.V.—Coefficient of variation. (kN). NA—Not available.

Table 4 Comparisons between tests and predicted loads Specimen notation

Ptest a (avg.)

Po1 b (ACI)

Po2 c (EC 4)

Pu1 d (LRFD)

Pu2 e (LRFD)

Pu3 f (EC 4)

Pu4 g (AS)

Pu5 h (CSA)

C0K 2-1–2 C4K 4-1–4 C9K 6-1–6 C10K 6-1–6 C12K 6-1–6

1059.7i 1328.5 1722.3 1885.5 2089.8

822.3 1138.6 1507.4 1582.7 1740.8

747.5 995.6 1284.7 1343.9 1467.8

576.9 791.1 1002.3 1041.0 1119.0

610.8 694.9 878.5 912.1 979.5

641.5 829.8 1010.7 1042.3 1103.7

622.3 778.7 917.6 940.8 983.6

654.8 779.5 891.4 910.5 946.6

Specimen notation

Ptest a (avg.)

Ptest Po1

Ptest Po2

Ptest Pu1

Ptest Pu2

Ptest Pu3

Ptest Pu4

Ptest Pu5

C0K 2-1–2 C4K 4-1–4 C9K 6-1–6 C10K 6-1–6 C12K 6-1–6

1059.7i 1328.5 1722.3 1885.5 2089.8

1.70 1.71 1.88 2.00 2.12

1.62 1.70 1.93 2.07 2.21

a b c d e f g h i

1.29 1.17 1.14 1.19 1.20

1.42 1.33 1.34 1.40 1.42

1.84 1.68 1.72 1.81 1.87

1.73 1.91 1.96 2.07 2.13

1.65 1.60 1.70 1.81 1.89

Ptest —Compressive load from tests. Po1 —Column squash load based on ACI 318-05. Po2 —Column squash load based on EC 4 (2004). Pu1 —Design strength Pu = φ Pn , φ = 0.85 based on LRFD (1999). Pu2 —Design strength Pu = φ Pn , φ = 0.75 based on LRFD (2005), HSS φ = 0.90. Pu3 —Design strength Pu based on EC 4 (2004). Pu4 —Design strength based on AS-5100.6 (2004). Pu5 —Design strength based on CSA S16-01 (2001). (kN).

cases were not detected in the tests. The failure modes for the given specimens are either global buckling or mixed global–local buckling.

Furthermore, the advantages of higher strength concrete were not realised in the design of CFT columns. It seems that f c0 limit can be raised to 84 MPa, a value closer to the actual condition.

4.2. Axial compression strengths

4.3. Ductility capacity

Column squash loads (P0i ) and column design strengths (Pui ) predicted by various codes and their comparisons are summarised in Table 4. The 1999 AISC-LRFD is conservative (average Ptest /Pu1 = 1.68–1.87), the 2005 AISC-LRFD is even more conservative (average Ptest /Pu2 = 1.91–2.13). This is expected because a lower resistance factor (i.e., φ = 0.75) is assumed in the 2005 AISC-LRFD. It is interesting to note that such a degree of conservatism prevails in other major design codes including the 2004 EC 4, the 2004 AS5100.6 and the 2001 CSA S16-01. Comparisons between the current experimental results and previous related studies are summarised in Table 5. The Ptest /Pu1 ratios obtained in this study are consistently higher than those reported by others. This study shows that the 1999 AISC-LRFD restriction on concrete strength of 55 MPa is conservative and may not be warranted.

In design, it is often necessary to determine the ductility of a column. The ductility capacity of specimens was measured by two different methods in this study. The first is based on the criterion proposed by Usami and Ge [32], µ1 = δu /δ y = εu /ε y where δu and εu are respectively the displacement and strain corresponding to the ultimate strength and δ y and ε y are respectively the displacement and strain at first yield or local buckling. Since the yield point could not be well defined on the load–strain curve, the yield displacement (δ y ) is defined as the displacement corresponding to ε y = 0.2% as suggested by Schneider [16]. Another way to measure the ductility is based on the rule proposed by Sun [33], µ2 = ε85.2 /ε85.1 where ε85.1 is the strain corresponding to the 85% of ultimate strength of CFT columns measured before the ultimate strength was reached and ε85.2 is the strain corresponding to the 85% of

43

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44 Table 5 Comparison between experimental data and previous studies Test by authors and others

LRFD (1999) Ptest a /Pu1 b

ACI 318-05 Ptest /Po1 c

Sect. shaped (Quantity)

Column type

C0K 2-1–2 (HSS) C4K 4-1–4 ( f c0 = 29 MPa) C9K 6-1–6 ( f c0 = 63 MPa) C10K 6-1–6 ( f c0 = 70 MPa) C12K 6-1–6 ( f c0 = 84 MPa)

1.84 1.68 1.72 1.81 1.87

1.29 1.17 1.14 1.19 1.20

R (2) R (6) R (6) R (6) R (6)

Medium Medium Medium Medium Medium

SSRC Task Group 20 [7] ( f c0 = 9.6 ksi) Kennny, Bruce, and Bjorhovde [14] Schneider [16] LRFD (1999)/Commentary Sec. I2.2 Han [20] Liu, Gho, and Yuan [23] Han and Yao [34] Sakino et al. [26] Sakino et al. [26] Liu [27]

1.60 1.56 1.41 1.39 1.47 1.36 1.18 NA NA 1.31

1.33 NAe 1.18 NA NA 1.14 NA 1.13 1.01 1.09

C (4) C (68) R, S, and C (14) R, S, and C (170) R (24) R (22) S and R (35) C (36) S (48) R (22)

Short Short/medium Short Short/medium Short Short Short Short Short Short

a b c d e

Ptest —Test result for axial compression load. Pu1 —Design strength based on AISC-LRFD (1999). Po1 —Squash load based on ACI 318-05. C—Circular section, S—Square section, R—Rectangular section. NA—Not available.

Table 6 Ductility capacity of specimens under axial loads Specimen notation

ε0.45 f 0 a (%)

ε85.1 d (%)

ε f 0 b (%)

εu c (%)

ε85.2 d (%)

µ1 = εu /ε y = εu /ε0.002 e

µ2 = ε85.2 /ε85.1 d

C0K C0K C4K C4K C4K C4K

2-1 2-1 4-1 4-2 4-3 4-4

NAf NA 0.051 0.051 0.051 0.051

0.237 0.247 0.200 0.232 0.216 0.201

NA NA 0.221 0.221 0.221 0.221

0.459 0.470 0.367 0.395 0.345 0.317

0.969 0.661 0.445 0.380 0.453 0.494

2.30 2.35 1.83 1.97 1.72 1.59

4.10 2.67 2.23 1.89 2.10 2.13

C9K C9K C9K C9K C9K C9K

6-1 6-2 6-3 6-4 6-5 6-6

0.098 0.098 0.098 0.098 0.098 0.098

0.250 0.223 0.248 0.242 0.232 0.247

0.289 0.289 0.289 0.289 0.289 0.289

0.422 0.342 0.368 0.353 0.343 0.363

0.858 0.357 0.362 0.357 0.382 0.838

2.11 1.71 1.84 1.77 1.71 1.81

3.43 1.60 1.46 1.47 1.65 3.39

c

c

C10K C10K C10K C10K C10K C10K

6-1 6-2 6-3 6-4 6-5 6-6

0.103 0.103 0.103 0.103 0.103 0.103

0.272 0.253 0.255 0.268 0.250 0.255

0.309 0.309 0.309 0.309 0.309 0.309

0.431 0.358 0.379 0.386 0.347 0.355

0.776 0.461 0.888 0.445 0.521 0.334

2.15 1.79 1.89 1.93 1.73 1.77

2.86 1.82 3.49 1.66 2.09 1.31

C12K C12K C12K C12K C12K C12K

6-1 6-2 6-3 6-4 6-5 6-6

0.121 0.121 0.121 0.121 0.121 0.121

0.240 0.263 0.260 0.260 0.242 0.262

0.342 0.342 0.342 0.342 0.342 0.342

0.323 0.368 0.372 0.357 0.334 0.364

0.296 0.3606 0.354 0.824 0.321 0.265

1.62 1.84 1.86 1.78 1.67 1.82

1.23 1.37 1.36 3.17 1.32 1.01

aε 0 0.45 f c0 is the strain at which a concrete cylinder stress corresponding to 0.45 f c is reached. b ε 0 is the strain at which a corresponding concrete strength f 0 is reached. c fc c ε is the strain at which the corresponding ultimate strength of CFT column (P ) is reached. u test d (µ = ε 2 85.2 /ε85.1 ) is suggested by Sun [33], where (ε85.1 ) represents the strain corresponding to the 85% of ultimate strength of columns measured before the ultimate strength was reached and (ε85.2 ) presents the strain corresponding to the 85% of ultimate strength of columns measured after ultimate strength reached. eε 0.002 −yield strain (ε y = 0.2%) as suggested by Schneider [16]. f NA—Not available.

ultimate strength of CFT columns measured after the ultimate strength was reached. Strain values corresponding to various

loadings and the calculated ductility values are summarized in Table 6. The results clearly demonstrate the strain-softening

44

D.M. Lue et al. / Journal of Constructional Steel Research 63 (2007) 37–44

characteristics of the CFT rectangular specimens tested. The computed ductility ratios are 1.59–2.15 (µ1 criterion) or 1.01–3.49 (µ2 rule). It appears that the ductility of CFT columns decreases with increasing f c0 . 5. Conclusions Based on the results concluded from this study, the following interesting findings are offered. 1. The comparisons indicate that the 2005 AISC-LRFD is more conservative than the 1999 AISC-LRFD. Although the formulas between these two specifications look quite different in format, the resulting nominal strengths are virtually the same and the design strengths differ due to the different resistance factors assumed (0.75 versus 0.85). 2. The CFT compressive strength equations given in the 1999 AISC-LRFD are also applicable to CFT rectangular columns with a higher f c0 limit of 63–84 MPa. The test results reveal that the 1999 AISC-LRFD design strengths are conservative and tend to penalize these CFT columns with higher concrete strength of 63–84 MPa. The 2005 AISC-LRFD is more receptive with the adoption of a higher f c0 of 70 MPa. This revision to the 1999 AISC-LRFD is justifiable based on the results concluded from this study. 3. The design strength of a CFT column with high-strength concrete determined by the 2004 EC 4, the 2004 AS-5100 or the 2001 CSA S16-01 is more conservative than that from the AISC-LRFD (1999, 2005). Acknowledgements The authors are grateful to the financial support provided by the National Science Council of Taiwan through Grant No. NSC-91-2211-E005-035. The authors would like to express their gratitude to Dr. Y.F. Chen for reviewing this manuscript. References [1] AISC. Load and resistance factor design specification for structural steel buildings. 1st, 2nd, 3rd ed. Chicago (IL): American Institute of Steel Construction, Inc.; 1986, 1993, 1999. [2] ACI 318-05. Building code requirements for structural concrete. Detroit (MI): American Concrete Institute; 2005. [3] AIJ. Recommendations for design and construction of concrete filled steel tubular structures. Tokyo: Architectural Institute of Japan; 1997. [4] Eurocode 4, ENV 1994-1-1. Design of composite steel and concrete structures, Part 1.1, General rules and rules for building. Commission of European Communities; 2004. [5] BS 5400. Steel, concrete and composite bridges: code of practice for design of composite bridges Part 5. London: British Standards Institution; 2000. [6] AS 5100.6. Bridge design Part 6: Steel and composite construction. Sydney (Australia): Australia Standards; 2004. [7] SSRC Task Group 20. Specification for the design of steel-concrete composite column. Engineering Journal, AISC 1979;16(4):101–15. [8] Furlong RW. AISC column design logic makes sense for composite column, too. Engineering Journal, AISC 1976;13(1):1–7.

[9] AISC. Specification for structural steel buildings. Chicago (IL): American Institute of Steel Construction, Inc.; 2005. [10] Knowles RB, Park R. Strength of concrete-filled steel tubular columns. Journal of the Structural Division, ASCE 1969;95(ST 12):2565–87. [11] Shakir-Khalil H, Zeghiche J. Experimental behavior of concrete-filled rolled rectangular hollow-section columns. The Structural Engineer 1989; 67(19):346–53. [12] Shakir-Khalil H, Mouli M. Further testes on concrete-filled rectangular hollow-section columns. The Structural Engineer 1990;68(20):405–13. [13] Shakir-Shalil H, Al-Rawdan A. Experimental behavior and numerical modeling of concrete-filled rectangular hollow section tubular columns. Composite Construction in Steel and Concrete, ASCE 1996;III:222–35. [14] Kenny JR, Bruce DA, Bjorhovde R. Removal of yield stress limitation for composite tubular columns. Engineering Journal, AISC 1994;31(1):1–11. [15] Bradford MA. Design strength of slender concrete-filled rectangular steel tubes. ACI Structural Journal 1996;93(2):229–35. [16] Schneider SP. Axially loaded concrete-filled steel tubes. Journal of Structural Engineering, ASCE 1998;124(10):1125–38. [17] Kilpatrick AE, Rangan BV. Tests on high-strength concrete-filled steel tubular columns. ACI Structural Journal 1999;96(2):268–74. [18] Zhang W, Shahrooz BM. Comparison between ACI and AISC for concrete-filled tubular columns. Journal of Structural Engineering, ASCE 1999;125(11):1213–23. [19] Shanmugam NE, Lakshmi B. State of the art report on steel-concrete composite columns. Journal of Constructional Steel Research 2001;57: 1041–80. [20] Han L-H. Tests on stub columns of concrete-filled RHS sections. Journal of Constructional Steel Research 2002;58:353–72. [21] GJB 4142. Technical specifications for early-strength model composite structures. Peking, China; 2000. [22] Mursi M, Uy B. Strength of concrete filled steel box columns incorporating interaction buckling. Journal of Structural Engineering, ASCE 2003;129(5):626–39. [23] Liu D, Gho W-M, Yuan J. Ultimate capacity of high-strength rectangular concrete-filled steel hollow section stub columns. Journal of Constructional Steel Research 2003;59:1499–515. [24] Viest IM. Development of design rules for composite construction. Engineering Journal, AISC 2003;40(4):181–8. [25] Melcher JJ, Karmazinova M. The analysis of composite steel-andconcrete compression members with high strength concrete. In: Proceedings of SSRC annual technical session. 2004, p. 223–37. [26] Sakino K, Nakahara H, Morino S, Nishiyama I. Behavior of centrally loaded concrete-filled steel-tube short columns. Journal of Structural Engineering, ASCE 2004;130(2):180–8. [27] Liu D. Tests on high-strength rectangular concrete-filled steel hollow section stub columns. Journal of Constructional Steel Research 2005;61: 902–11. [28] CSA S16-01. Limit states design of steel structures. Toronto (Canada): Canadian Standards Association; 2001. [29] Annual Book of ASTM Standards. Standard test methods and definitions for mechanical testing of steel products. ASTM Standard No. A370; 1993. [30] Annual Book of ASTM Standards. Standard test method for compressive strength of cylindrical concrete specimens. ASTM Standard No. C39; 1993. [31] Annual Book of ASTM Standards. Standard test method for static modulus of elasticity and Poisson’s ratio of concrete in compression. ASTM Standard No. C469; 1993. [32] Usami T, Ge HB. Ductility of concrete-filled steel box columns under cyclic loading. Journal of Structural Engineering, ASCE 1994;120(7): 2021–40. [33] Sun WL. Behavior of stiffened concrete-filled steel beam-columns. Master thesis. Taiwan: Department of Civil Engineering, National Taiwan University; 2000. [34] Han L-H, Yao G-H. Influence of concrete compaction the strength of concrete-filled steel RHS columns. Journal of Constructional Steel Research 2003;59:751–67.