Thin-Walled Structures 42 (2004) 16311644

www.elsevier.com/locate/tws

Behaviour of high strength rectangular

concrete-lled steel hollow section columns
under eccentric loading
Dalin Liu

School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798,
Singapore
Received 10 September 2003; received in revised form 23 June 2004; accepted 29 June 2004

Abstract
This paper presents an experimental study on the behaviour of 12 high strength rectangular concrete-lled steel hollow section columns subjected to eccentric loading. The primary
test parameters were the cross-sectional aspect ratio, slenderness and load eccentricity. The
specimens with cross-sectional aspect ratios of 1.0, 1.5 and 2.0 were fabricated from high
strength materials (fy 550 MPa; fcu 70:8 and 82.1 MPa). The slenderness ratios of the
specimens were 20 and 50, while the load eccentricity ratios (e=B; e is the load eccentricity, B
is the breadth of cross-section) varied between 0.17 and 0.40. Favourable ductility performance was observed for all specimens during the test. The experimental ultimate capacities of
the specimens were compared with the design strengths predicted by the codes. Comparison
of results showed that Eurocode 4 overestimated the ultimate capacities of the columns with
a dierence of 3%. ACI and AISC, on the other hand, conservatively predicted the failure
loads by 11% and 25%, respectively.
# 2004 Elsevier Ltd. All rights reserved.
Keywords: Steel; Concrete; Eccentric loading; High strength material; Concrete-lled steel hollow section
column

1. Introduction
The use of concrete-lled steel hollow section (CFSHS) columns for the construction of high-rise buildings, bridges, warehouses, etc. has become widespread in

Tel.: +65-67905086; fax: +65-67921650.

E-mail address: [email protected] (D. Liu).

0263-8231/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tws.2004.06.002

1632

D. Liu / Thin-Walled Structures 42 (2004) 16311644

Nomenclature
Ac
As
B
e
Ec
Es
fc0
fcu
fu
fy
H
Ic
Is
k
L
M
Me
N
Ne
rs
SR
t
a

cross-sectional area of steel hollow section

cross-sectional area of concrete core
cross-sectional breadth of column
load eccentricity
modulus of elasticity of concrete
modulus of elasticity of steel
cylinder strength of concrete
cube strength of concrete
tensile strength of steel
yield stress of steel
cross-sectional height of column
moment of inertia of un-cracked concrete core
moment of inertia of steel hollow section
moment amplier factor
buckling length of column
applied bending moment
experimental ultimate end bending moment
applied axial force
experimental ultimate axial force
radius of gyration of steel hollow section
slenderness ratio of column
thickness of steel hollow section wall
volumetric steel-to-concrete ratio

recent decades. This is mainly due to their combination of the advantages of both
steel and concrete, namely the high strength, light weight, and speed of construction of steel and large damping and economy of concrete. The enhanced mechanical properties of CFSHS columns can be explained in terms of composite action
between the steel hollow section and the concrete core. Due to the discrepancy of
Poissons ratios between steel and concrete, the volume increase of the concrete
core due to the propagation of cracks is constrained by the exterior steel hollow
section. Consequently, both the strength and ductility of the concrete are
enhanced. On the other hand, the inward local buckling of the steel hollow section
is prevented by the in-lled concrete.
CFSHS columns can manifest more architectural and economical benets if they
are fabricated from high strength materials. High strength CFSHS columns require
a smaller cross-sectional size to withstand the loads, which is very much appreciated by modern designers and building engineers. However, the use of high
strength CFSHS columns in the construction industry is still scarce throughout the
world, because little is known about their structural behaviour. A research need

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1633

thus exists to investigate the behaviour of high strength CFSHS columns under
various types of loading so that guidelines can be developed for design.
In the current experimental study, the behaviour of high strength rectangular
CFSHS beam-columns is examined. The main objective is to evaluate the specications in the codes for the design of high strength rectangular CFSHS beam-columns. Three commonly used design codes in practice, namely Eurocode 4 (EC4)
[1], American Concrete Institute (ACI) [2] and American Institute of Steel Construction (AISC) [3], are considered for comparison.

2. Previous research
Many research projects have been conducted since the 1960s to investigate the
behaviour of CFSHS columns over a variety of cross-sectional shapes, slenderness,
material properties and loadings. Furlong [4] concluded from tests on square
CFSHS columns that the stability performance of the steel hollow section was signicantly enhanced by the concrete inll. Experiments on 268 CFSHS columns
reported by Tomii et al. [5] highlighted two failure modes during the concentric
loading tests. They were the overall buckling for slender columns and the crushing
of concrete for stub columns. Additionally, the post-yield behaviours of circular and
octagonal CFSHS columns were either strain hardening or perfectly plastic. The
square CFSHS columns, on the other hand, exhibited strain softening behaviour
after the failure load had been reached.
The eect of slenderness and load eccentricity on the behaviour of circular
CFSHS columns was investigated by Rangan and Joyce [6]. Based on the test
results of nine columns with slenderness ratios ranging from 32 to 91, a simplied
method was developed to evaluate the ultimate capacity of the CFSHS columns.
However, the maximum discrepancy between the proposed method and the test
results was as high as 60%. Schneider [7] concluded from his tests on 14 CFSHS
stub columns that the axial load behaviour of CFSHS columns was signicantly
aected by the cross-sectional shape and breadth-to-thickness (B=t) ratio. For circular CFSHS column with B=t ratios between 60 and 220, OShea and Bridge [8]
concluded from the tests that load and bond conditions signicantly inuenced the
axial load behaviour of the columns. In addition, the stability performance of the
steel hollow section was not improved by the in-lled concrete, because the predominant outward buckling remained unaected by the concrete core. Uy [9] carried out experiments on 19 square CFSHS columns under axial loading and
bending. The steel yield stress of the specimens was 750 MPa, while the cylinder
strengths of concrete varied from 28 to 32 MPa. Test results revealed that the EC4
approach overestimated the strength of the columns subjected to pure compression
or pure bending. Han [10] provided test results of 24 normal strength rectangular
CFSHS columns under concentric loading and commented that the strength
increase of the concrete core due to the connement by the steel hollow section was
inuenced by the cross-sectional aspect ratio, material properties and conning
factor.

1634

D. Liu / Thin-Walled Structures 42 (2004) 16311644

In recent years, much attention has been paid to the behaviour of CFSHS column made from high strength steel and concrete. Fujimoto et al. [11] reported an
extensive set of test on square CFSHS stub columns subjected to combined compression and bending. A total of 32 specimens with steel yield stresses of 262, 617
and 834 MPa and concrete cylinder strengths varying from 25 to 80 MPa were tested. Test results suggested that the strength of the CFSHS beam-columns was signicantly aected by the B=t ratio and the axial load level. In addition, the
specimens made from high strength steel exhibited worse ductility performance
than those made from normal strength steel. Varma et al. [12,13] concluded from
the tests on eight square CFSHS beam-columns that the moment capacity of the
columns could be well predicted using the ACI provisions. The test specimens had
uniform concrete cylinder strength of 110 MPa and steel yield stresses ranging
from 269 to 660 MPa. Kilpatrick and Rangan [14] conrmed from the tests on 41
high strength circular CFSHS columns that the slenderness and load eccentricity
signicantly aected the strength of columns. After the review of the previous
research projects pertaining to CFSHS columns, it was concluded that the experimental investigation on high strength rectangular CFSHS columns was still insufcient. Thus, the current study was carried out to investigate the behaviour of high
strength rectangular CFSHS columns subjected to eccentric loading and to calibrate the provisions in the codes for the design of high strength rectangular
CFSHS columns.

3. Experimental investigation
A total of 12 rectangular CFSHS columns were constructed and tested under
monotonic eccentric loading to investigate the structural behaviour. The test parameters of the specimens were the material strength, cross-sectional aspect ratio
(B=H), volumetric steel-to-concrete ratio (a), load eccentricity ratio (e=B) and slenderness ratio (SR). All these parameters were selected according to the ranges of
construction practice.
3.1. Material properties
The steel hollow sections were fabricated by welding four component at plates
together, as shown in Fig. 1. All the component steel plates were cut from a high
strength (A572 Grade 65) steel sheet with a nominal yield stress of 460 MPa. In
order to obtain the actual material properties, six steel coupons were ame cut
from the steel sheet and tested under tension. Among the six steel coupons, three
were prepared and tested according to the requirements of the American Standards
for Testing and Materials (ASTM) [15], while the other three conformed with the
specications of the Chinese testing standards GB2975 [16]. The measured values
of yield stress (fy), tensile strength (fu) and modulus of elasticity (Es) are summarized in Table 1. As the discrepancies of results between the two codes were minimal, the measured values determined according to GB2975 were used in the study.
The average yield stress, tensile strength and modulus of elasticity were determined

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1635

Fig. 1. Cross-section of CFSHS specimen.

as 550, 608 and 207,000 MPa, respectively. A typical stressstrain curve for the
steel coupons is illustrated in Fig. 2.
Two dierent mix designs were used for the high strength concrete. Three concrete
cubes (150  150  150 mm) were prepared for each mix design and crushed to failure to determine the actual compressive strength. The average cube strengths (fcu)
of the concrete were determined as 70.8 and 82.1 MPa. In order to use the code
specications to predict the failure loads of the specimens, the cube strengths were
converted into the equivalent cylinder strengths (fc0 ). Following the tables proposed
by Chen et al. [17], the equivalent cylinder strengths of concrete were determined as
60.8 and 72.1 MPa, respectively. The moduli of elasticity of the two grades of
concrete (Ec) were assumed to be 39  103 and 40  103 MPa, respectively.
3.2. Test specimen
A 10-mm-thick base plate was welded to the steel hollow section before concrete
was vertically cast. After the concrete had been cast into the steel hollow section,
Table 1
Material properties of steel
Testing standards

Coupon No.

fy (MPa)

fu (MPa)

fy =fu

Es (GPa)

ASTM

1
2
3
Average

519
555
551
542

601
612
621
611

0.86
0.91
0.89
0.89

202
203
203
203

GB2975

4
5
6
Average

532
558
559
550

582
623
619
608

0.91
0.90
0.90
0.90

206
206
210
207

1636

D. Liu / Thin-Walled Structures 42 (2004) 16311644

Fig. 2. A typical stressstrain curve for the steel coupons.

the specimen was left with top open to the air for two weeks for the concrete to
set. A layer of high strength cement mortar was subsequently added to ush the
concrete core with the steel hollow section at the top before a 10-mm-thick steel
plate was welded to the top of the specimen. The geometric and material properties
of the specimens are given in Table 2.
3.3. Test setup
All the specimens were tested under eccentric loading in a 5000-kN-capacity universal testing machine at the Harbin Institute of Technology, China. Fig. 3(a) gives
a front view of a specimen in the testing machine. Since the standard accessories of

Table 2
Geometric and material properties of specimens
No.

B
(mm)

H
(mm)

B=H

t
(mm)

L
(mm)

SR

e
(mm)

e=B

fcu
(MPa)

fy
(MPa)

E01
E02
E03
E04
E05
E06
E07
E08
E09
E10
E11
E12

150
150
150
150
180
180
120
120
200
200
160
160

150
150
150
150
120
120
80
80
100
100
80
80

1.0
1.0
1.0
1.0
1.5
1.5
1.5
1.5
2.0
2.0
2.0
2.0

4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18
4.18

870
870
2170
2170
1040
1040
1740
1740
1150
1150
2310
2310

20
20
50
50
20
20
50
50
20
20
50
50

30
30
30
60
30
70
20
20
60
40
60
30

0.20
0.20
0.20
0.40
0.17
0.39
0.17
0.17
0.30
0.20
0.38
0.19

70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1

550
550
550
550
550
550
550
550
550
550
550
550

0.12
0.12
0.12
0.12
0.13
0.13
0.20
0.20
0.14
0.14
0.18
0.18

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1637

Fig. 3. Test setup. (a) Photograph; (b) diagrammatic view; (c) section A-A: strain gauge locations.

the testing machine could not produce eccentric loading, special end assemblages
were constructed which allowed the load from the testing machine to be applied at
given eccentricities to the specimen. A detailed drawing of the end assemblages is
given in Fig. 3(b). Knife edges were employed at both the bottom and top of the
specimen to simulate pinpin boundary conditions. The knife edges consisted of a
male edge and a female counterpart. The male edge was bolted to the top (bottom)
platen of the testing machine. The female edge was positioned to the top (bottom)
end plate of the specimen. A short steel rebar was welded to the centre of the end

1638

D. Liu / Thin-Walled Structures 42 (2004) 16311644

plates in order to align the female edges to the specimen and to prevent the slippage between the female edges and the specimen. During the test, the deection in
the plane of loading over the height of the specimen was recorded by ve linear
voltage displacement transducers (LVDTs). The out-plane deection, as well as the
displacement of the two platens of the testing machine were monitored by another
ve LVDTs. Twelve strain gauges were bonded at the four exterior surfaces of the
specimen at mid height to measure the longitudinal and transverse strains during
loading, as shown in Fig. 3(c).
3.4. Test procedure and results
The specimens were tested to failure under combined compression and bending
about the major axis. Prior to the actual test, a load level of 20 kN was applied so
that the platens of the testing machine were rmly attached to both ends of the
specimen. The axial load was slowly applied to the specimen by careful manipulation of the loading and unloading valves. During the test, the longitudinal and
the transverse strains as well as the in-plane and out-plane deections of the specimen were recorded at a load increment of 50 kN. The experimental ultimate axial
load (Ne) and the ultimate end bending moment (Me Ne e) are summarized in
Table 3. During the test, the out-plane deection of the specimen was less than
1 mm, thus the specimen was conrmed to be under compression combined with
uniaxial bending.
Figs. 4(a) and (b) give the correlation of the axial force and the mid height
deection for the specimens with slenderness ratio of 20 and 50, respectively. The
axial load decreased slowly after the failure load had been reached, indicating the
favourable ductility performance in the post-yield behaviour of the specimens. It
can be concluded from the comparison of behaviours between the specimens, such
as specimens E01 and E02 that the strength of concrete had a positive eect on the
ultimate capacity of the columns. On the other hand, the slenderness and load
eccentricity demonstrated an adverse inuence on the strength of the specimens. In
Fig. 5, the axial forces were plotted against the strain values of the compression
and tension sides in the mid height cross-section. The strain on the compression
side of the specimen at the failure load was higher than 0.003, indicating the yielding of the steel hollow section before the critical load was reached.
Three types of failure mode were observed for the specimens during the test. The
rst type of failure, which occurred for specimens E01 and E06, was identied as
the crushing of concrete and local buckling of steel hollow section close to the end
of the specimen, as shown in Fig. 6(a). Fig. 6(b) gives the second type of failure
mode for all the other specimens with slenderness ratio of 20. Local buckling was
observed near the quarter height of the specimens. The third type of failure mode
was noticed for all the specimens with slenderness ratio of 50. As illustrated
in Fig. 6(c), the failure loads of these specimens were reached followed by local
buckling of steel hollow section on the compression side at the mid height of the
specimen.

150
150
150
150
180
180
120
120
200
200
160
160

E01
E02
E03
E04
E05
E06
E07
E08
E09
E10
E11
E12

Average
SD
COV

B
(mm)

Specimen

150
150
150
150
120
120
80
80
100
100
80
80

H
(mm)

70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1
70.8
82.1

fcu
(MPa)

30
30
30
60
30
70
20
20
60
40
60
30

e
(mm)

Table 3
Comparison of results between the test and the codes

20
20
50
50
20
20
50
50
20
20
50
50

SR

1678
1850
1330
1020
1950
1140
660
855
1310
1800
670
1020

Ne
(kN)

50.3
55.5
39.9
61.2
58.5
79.8
13.2
17.1
78.6
72.0
40.2
30.6

Me
(kN m)

49.5
54.1
44.6
71.8
50.7
90.9
16.3
17.3
76.2
65.4
43.1
31.1

Mc
(kN m)

EC4

0.972
0.102
0.105

1.017
1.026
0.895
0.852
1.154
0.878
0.810
0.988
1.031
1.101
0.933
0.984

Me =Mc
46.6
50.5
37.5
55.7
47.8
76.3
14.3
15.0
67.6
60.7
35.0
26.0

1.112
0.080
0.072

1.080
1.099
1.064
1.099
1.224
1.046
0.923
1.140
1.163
1.186
1.149
1.177

39.5
41.4
34.0
50.6
41.2
65.9
13.4
13.8
60.4
51.2
34.4
24.2

Mc
(kN m)

Mc
(kN m)
Me =Mc

AISC

ACI

1.249
0.117
0.094

1.274
1.341
1.174
1.209
1.420
1.211
0.985
1.239
1.301
1.406
1.169
1.264

Me =Mc

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1639

1640

D. Liu / Thin-Walled Structures 42 (2004) 16311644

Fig. 4. Loaddeection curves for the specimens. (a) SR 20; (b) SR 50.

4. Comparison of results
In this section, the design philosophies adopted in EC4, ACI and AISC were
commented and employed to predict the ultimate capacities of the test specimens.
The calculated values were subsequently compared with the failure loads obtained
from the test.
4.1. Code provisions
4.1.1. Eurocode 4 (EC4)
To determine the cross-sectional moment resistance of a CFSHS beam-column,
the EC4 provisions assume plastic stress distribution in the cross-section and full
compatibility between the steel and concrete. The second-order eect is taken into
account by applying a moment amplier method. Eq. (1) shows the moment amplier factor for CFSHS column under axial compression combined with equal end

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1641

Fig. 5. Loadstrain curves for the specimens. (a) SR 20; (b) SR 50.

bending moments.
k

1:1
1  N=Ncr

in which N is the applied axial force; Ncr p2 EI=L2 , EI Es Is 0:6Ec Ic , Is is the

moment of inertia of steel hollow section, Ic is the moment of inertia of un-cracked
concrete core, L is buckling length of column.
4.1.2. American Concrete Institute (ACI)
According to ACI, the strength of a CFSHS beam-column is evaluated on the
basis of several assumptions: (1) a linear strain distribution exists in the crosssection; (2) there is no slippage between the steel and concrete; and (3) the
maximum allowable compressive strain of concrete is 0.003. In the calculation of
cross-sectional moment resistance, the ultimate strength of concrete is taken as fc0
instead of 0:85fc0 to reect that the concrete core inside the steel hollow section
does not split and is environmentally well protected [18]. For slender CFSHS
columns with equal end bending moments, a moment amplier factor is adopted to

1642

D. Liu / Thin-Walled Structures 42 (2004) 16311644

Fig. 6. Typical failure modes of the specimens.

consider the second-order eect, as shown in Eq. (2).

k

1
1  N=0:75Ncr

in which Ncr p2 EI=L2 , EI Es Is 0:2Ec Ic .

4.1.3. American Institute of Steel Construction (AISC)
The AISC approach for the design of CFSHS columns is essentially identical to
that for steel columns. The cross-sectional axial capacity and moment resistance of
a CFSHS beam-column are determined from the axial forcebending moment
interaction diagrams, as in Eqs. (3) and (4).
N
8M

1:0
N0 9Mn
N
M

1:0
2N0 Mn

N
0:2
N0
N
for
< 0:2
N0
for

3
4

in which N0 As fcr , fcr 0:658kc fmy for kc

1:5, fcr 0:877=k2c fmy for kc < 1:5,
p
fmy fy 0:85fc0 Ac =As , kc L=rs p fmy =Es 0:4Ec Ac =As , As is the cross-sectional area of steel hollow section, Ac is the cross-sectional area of concrete core, rs
is the radius of gyration of steel hollow section.
In Eqs. (3) and (4), the nominal exural strength Mn is the cross-sectional
moment capacity at zero axial load level. It is determined based on plastic stress
distribution for both the steel and the concrete in the cross-section. However, the
extent to which the concrete core contributes to the cross-sectional moment resistance depends on the axial load level. For N=N 0 0:3, full composite action is
assumed, while for N=N 0 < 0:3, Mn is determined by straight line transition

D. Liu / Thin-Walled Structures 42 (2004) 16311644

1643

between the nominal strength at N=N 0 0:3 and the nominal strength at N 0
[3]. Similar to EC4 and ACI, for slender CFSHS columns with equal end bending
moments, a moment magnier factor as in Eq. (5) applies to consider the secondorder eect.
k

1
1  N=Ncr

in which Ncr p2 Em Is =L2 , Em Es 0:4Ec Ac =As .

4.2. Comparison of results between the tests and the codes
The experimental ultimate end bending moment for each specimen was compared with that calculated from the design codes, as in Table 3. SD and COV in
Table 3 denote the standard deviation and coecient of variation, respectively. The
comparison of results suggested that the EC4 approach was slightly unsafe to
evaluate the strengths of the specimens with a dierence of 3%. The reason
accounting for the un-conservative prediction was considered to be the rigid-plastic
assumption made in EC4. For the high strength rectangular CFSHS columns tested in the current study, the steel hollow sections might not have fully yielded at
the failure load. The specications in ACI, however, underestimated the critical
loads of the specimens by 11%. This revealed that the assumption of the strain
limit of 0.003 for the concrete at the ultimate capacity was too strict for high
strength CFSHS columns, because at this strain limit only the out-most steel in the
cross-section could yield. The AISC provisions, on the other hand, signicantly
underestimated the strengths of the specimens by 25%. Since the design equations
in AISC are essentially identical to those for steel columns, modications are thus
required to be incorporated into the code so that the contribution of the concrete
core to the moment resistance of the composite section can be more precisely
considered and hence a better design can be achieved for high strength CFSHS
columns.

5. Conclusion
An experimental investigation into the behaviour of 12 high strength rectangular
CFSHS columns subjected to eccentric loading has been presented. Favourable ductility performance was observed for the specimens during the test. Three types of
failure mode were identied. For specimens with slenderness ratio of 20, local buckling occurred either close to the end or near the quarter height of the specimens. For
specimens with slenderness ratio of 50, the failure loads were reached with local
buckling at the mid height. A comparison of the test strengths with the design
strengths has been presented. The design strengths were calculated using the EC4,
ACI and AISC provisions for CFSHS columns. The comparison of results shows
that EC4 is slightly unsafe to predict the ultimate capacity of the high strength rectangular CFSHS columns by 3%. ACI and AISC, on the other hand, conservatively
predict the failure loads by 11% and 25%, respectively. The discrepancies of results

1644

D. Liu / Thin-Walled Structures 42 (2004) 16311644

between the tests and the codes indicate the need for an extensive investigation into
the behaviour of high strength rectangular CFSHS columns, so that a better
understanding on the structural behaviour and an accurate design can be achieved.
Acknowledgements
The assistance from Dr. Jiepeng Liu, Yuyin Wang and Jie Yuan in the experimental work is gratefully appreciated. The author would especially like to thank
Emeritus Prof. Shantong Zhong for his advice and supervision on the research
programme.
References
[1] British Standard Institution. Eurocode 4, ENV 1994-1-1. Design of composite steel and concrete
structures, Part 1.1: general rules and rules for buildings. London, 1994.
[2] American Concrete Institute. Building code requirements for reinforced concrete and commentary.
ACI 318-95/ACI 318R-95, ACI Committee 318, Detroit, 1995.
[3] American Institute of Steel Construction. Load and resistance factor design specication for structural steel buildings. Chicago, 1994.
[4] Furlong RW. Strength of steel-encased concrete beam columns. Journal of Structural Engineering
Division 1967;93(5):11324.
[5] Tomii M, Yoshimura K, Morishita Y. Experimental studies on concrete lled steel tubular stub columns under concentric loading. Proceedings of the International Colloquium on Stability of Structures under Static and Dynamic Loads, Washington. 1977, p. 71841.
[6] Rangan BV, Joyce M. Strength of eccentrically loaded slender steel tubular columns lled with
high-strength concrete. ACI Structural Journal 1992;89(6):67681.
[7] Schneider SP. Axially loaded concrete-lled steel tubes. Journal of Structural Engineering
1998;124(10):112538.
[8] OShea MD, Bridge RQ. Design of circular thin-walled concrete lled steel tubes. Journal of Structural Engineering 2000;126(11):1295303.
[9] Uy B. Strength of short concrete lled high strength steel box columns. Journal of Constructional
Steel Research 2001;57:11334.
[10] Han LH. Tests on stub columns of concrete-lled RHS sections. Journal of Constructional Steel
Research 2002;58:35372.
[11] Fujimoto T, Nishiyama I, Mukai A, Baba T. Test results of eccentrically loaded short columns
square CFT columns. Proceedings of the Second Joint Technical Coordinating Committee Meeting
on Composite and Hybrid Structures, Hawaii. 1995.
[12] Varma AH, Ricles JM, Sause R, Lu LW. Experimental behaviour of high strength square concretelled steel tube beam-columns. Journal of Structural Engineering 2002;128(3):30918.
[13] Varma AH, Ricles JM, Sause R, Lu LW. Seismic behaviour and modelling of high-strength composite concrete-lled steel tube (CFT) beam-columns. Journal of Constructional Steel Research
2002;58:72558.
[14] Kilpatrick AE, Rangan BV. Tests on high-strength concrete-lled steel tubular columns. ACI Structural Journal 1999;96(2):26875.
[15] American Society for Testing and Materials. Test methods for tension testing of metallic materials.
Pennsylvania: West Conshohocken; 1997.
[16] GB2975: Test standards for metallic materials. Beijing: Chinese Planning Press; 1982 [in Chinese].
[17] Chen Z, Zhu J, Wu P. High strength concrete and its application. Beijing: Tsinghua University
Press; 1996 [in Chinese].
[18] Zhang W, Shahrooz BM. Strength of short and long concrete-lled tubular columns. ACI Structural Journal 1999;96(2):2308.