Engineering Structures 28 (2006) 716–728

www.elsevier.com/locate/engstruct

Design and behaviour of concrete-filled cold-formed stainless steel

tube columns
Ehab Ellobody a , Ben Young b,∗
a Department of Structural Engineering, Faculty of Engineering, Tanta University, Tanta, Egypt
b Department of Civil Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong

Received 24 April 2005; received in revised form 22 September 2005; accepted 27 September 2005
Available online 15 November 2005

Abstract

This paper presents a nonlinear finite element model to investigate the behaviour and design of axially loaded concrete-filled cold-formed high
strength stainless steel tube columns. The study was conducted over a wide range of concrete cylinder strengths from 20 to 100 MPa. The depth
of flat portion-to-plate thickness (d/t) ratio of the square and rectangular hollow sections ranged from 16 to 96, covering compact and slender
sections. The columns had different lengths so that the length-to-depth ratio remained a constant of 3. Nonlinear material models for confined
concrete and stainless steel tubes were used in the finite element model. The results obtained from the finite element analysis were verified against
experimental results. An extensive parametric study was conducted to investigate the effects of cross-section geometry and concrete strength
on the behaviour and strength of the columns. The column strengths obtained from the finite element analysis were compared with the design
strengths calculated using the American specifications and Australian/New Zealand standards. A design equation was proposed for concrete-filled
stainless steel tube columns. It is shown that the design strengths obtained using the proposed modified equation are more accurate compared to
the American and Australian/New Zealand predictions.
c 2005 Elsevier Ltd. All rights reserved.

Keywords: Composite columns; Concrete; Cold-formed; Finite element; Modeling; High strength; Stainless steel tubes; Square hollow sections; Rectangular hollow
sections; Structural design

1. Introduction steel and high strength steel tube columns using circular, square
and rectangular hollow sections. There are limited test data on
Concrete-filled steel tube columns have many advantages in concrete-filled stainless steel tube columns. The behaviour of
terms of their high strength, high ductility, high stiffness and stainless steel sections is different from that of carbon steel
full usage of construction materials. In recent years, stainless sections. Stainless steel sections have a rounded stress–strain
steel tube members have become popular due to their high curve with no yield plateau and low proportional limit
corrosion resistance, ease of construction and maintenance stress compared to carbon steel sections. Recent experimental
as well as aesthetic appearance. However, investigations of investigations of stainless steel columns without concrete
concrete-filled stainless steel tube columns are rarely found infilled were conducted by Young and Hartono [12], Young and
in the literature, especially using high strength stainless steel Liu [13], Liu and Young [14], and Gardner and Nethercot [15,
tubes. Tests of concrete-filled carbon steel tube columns were 16], while experimental investigation of high strength stainless
conducted by Schneider [1], Uy [2–4], Huang et al. [5], Han and steel columns were conducted by Young and Lui [17,18].
Yao [6], Mursi and Uy [7], Liu et al. [8], Uy [9], Sakino et al. Available design guides for concrete-filled stainless steel tube
[10], Giakoumelis and Lam [11], and many other researchers. columns are limited to the general design guides specified in the
These tests were carried out on concrete-filled carbon mild American specifications [19,20] and Australian standards [21,
22] for cold-formed stainless steel and concrete structures.
∗ Corresponding author. Tel.: +852 2859 2674; fax: +852 2559 5337. Experimental investigation of concrete-filled cold-formed high
E-mail address: young@hku.hk (B. Young). strength stainless steel columns has been conducted by Young

0141-0296/$ - see front matter 

c 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.engstruct.2005.09.023
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 717

and Ellobody [23]. A series of tests was performed with depth-

Nomenclature to-plate thickness ratio of the tube sections varied from 25.7 of
Ac Cross-sectional area of concrete compact sections to 55.8 of relatively slender sections.
Ae Effective cross-sectional area of stainless steel Numerical modeling can be used to understand the
tube behaviour of concrete-filled stainless steel tube columns.
As Cross-sectional area of stainless steel tube However, developing an accurate model requires careful
B Overall width of cross-section (smaller dimen- presentation of the concrete confinement, stainless steel tube
sion) and stainless steel tube–concrete interface. Hu et al. [24]
C Proposed factor for design equation developed a nonlinear finite element (FE) model using the
D Overall depth of cross-section (larger dimension) ABAQUS program to simulate the behaviour of concrete-filled
d Depth of flat portion of cross-section carbon steel tube columns. The concrete confinement was
E cc Young’s modulus of confined concrete achieved by matching the numerical results by trial and error
Eo Initial Young’s modulus of stainless steel tube via parametric study. Roufegarinejad et al. [25] described a
Et Tangent modulus theoretical model for the behaviour and design of concrete-
Fn Design stress filled stainless steel columns subjected to axial compression
f Equivalent uniaxial stress and bending. Ellobody et al. [26] developed a nonlinear finite
fc Unconfined compressive cylinder strength of
element model to investigate the behaviour of normal and
concrete
high strength concrete-filled compact carbon steel tube circular
f cc Confined compressive strength of concrete
f cu Unconfined compressive cube strength of con- columns. The concrete confinement was considered in the
crete model.
fl Lateral confining pressure The main objective of this study is to develop a nonlinear
Fy Yield stress of stainless steel tube (Fy = σ0.2 ) finite element model to investigate the behaviour and strength
K Ratio of flow stress in triaxial tension to that in of concrete-filled cold-formed high strength stainless steel
compression tube columns. Both square and rectangular hollow sections
k1 Coefficient for confined concrete were investigated. The finite element program ABAQUS [27]
k2 Coefficient for confined concrete was used in the analysis. The effects of concrete strength
k3 Coefficient for confined concrete and concrete confinement were considered in the analysis.
L Length of column Nonlinear stress–strain curves of stainless steel tubes were
n Exponent in Ramberg–Osgood expression used. The interface between the concrete and the stainless steel
PACI/ASCE-1 Nominal axial strength calculated using tube was also modeled. The results obtained from the model
the American specifications (unfactored design were verified against the test results conducted by Young and
strength according to approach one) Ellobody [23]. A parametric study was performed to investigate
PACI/ASCE-2 Nominal axial strength calculated using the effects of concrete strength and cross-section geometry on
the American specifications (unfactored design the behaviour and strength of axially loaded concrete-filled
strength according to approach two) cold-formed high strength stainless steel tube columns. The
PACI/ASCE-3 Nominal axial strength calculated using
column strengths obtained from the parametric study were
the American specifications (unfactored design
compared with the design strengths calculated using the general
strength according to approach three)
design guides specified in the American specifications [19,
PFE Ultimate load obtained from finite element
20] and Australian/New Zealand standards [21,22] for stainless
analysis
PTest Test ultimate load (test strength) steel and concrete structures.
R Coefficient for confined concrete
RE Coefficient for confined concrete 2. Summary of experimental investigation
Rε Coefficient for confined concrete
Rσ Coefficient for confined concrete The experimental investigation of concrete-filled cold-
r Reduction factor for confined concrete formed high strength stainless steel tube columns performed by
ri Inside corner radius of stainless steel tube Young and Ellobody [23] provided the experimental ultimate
t Plate thickness of stainless steel tube loads of columns subjected to uniform axial compression. The
ε Equivalent uniaxial strain columns had square hollow sections (SHS) and rectangular
εc Unconfined concrete strain hollow sections (RHS) cold-rolled from flat strips of duplex
εcc Confined concrete strain and high strength austenitic stainless steel material. The test
εf Elongation (tensile strain) after fracture based on program consisted of five test series that included two series
gauge length of 50 mm of concrete-filled SHS tubes (SHS1 and SHS2) and three series
β Material angle of friction of concrete-filled RHS tubes (RHS1, RHS2 and RHS3). The
υcc Poisson’s ratio of confined concrete nominal section sizes (D× B ×t) of series SHS1, SHS2, RHS1,
σ0.2 Static 0.2% proof stress
RHS2 and RHS3 are 150×150×6, 150×150×3, 200×110×4,
σu Static ultimate strength
160 × 80 × 3 and 140 × 80 × 3 mm, respectively, where D
is the overall depth, B is the overall width and t is the plate
718 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

(a) SHS. (b) RHS.

Fig. 1. Definition of symbols for concrete-filled square and rectangular hollow section specimens.

Table 1
Measured and predicted material properties

Test series Section D × B × t (mm) Position σ0.2 (MPa) σu (MPa) E o (GPa) ε f (%)
SHS1 150 × 150 × 6 Flat 497 761 194 52
Corner 618 822 214 23
SHS2 150 × 150 × 3 Flat 448 699 189 52
Corner 557 741 214 23
RHS1 200 × 110 × 4 Flat 503 961 200 36
Corner 635 1105 207 34
RHS2 160 × 80 × 3 Flat 536 766 208 40
Corner 667 887 214 23
RHS3 140 × 80 × 3 Flat 486 736 212 47
Corner 605 804 214 23

Table 2
Comparison between test and finite element results
PTest
Specimen Test FE PFE
D/t d/t ri (mm) L (mm) PTest (kN) PFE (kN)
SHS1C0 25.7 21.9 5.3 601 1927.4 1910.0 1.01
SHS1C40 25.8 22.0 5.3 450 2768.1 2850.0 0.97
SHS1C60 25.8 22.0 5.3 450 2972.0 3040.0 0.98
SHS1C80 25.8 22.0 5.3 450 3019.9 3200.0 0.94
SHS2C0 53.8 48.5 4.6 600 408.6 424.3 0.96
SHS2C40 54.1 48.8 4.6 450 1381.5 1310.0 1.05
SHS2C60 54.1 48.8 4.6 450 1620.0 1570.0 1.03
SHS2C80 54.2 48.9 4.6 450 1851.3 1930.0 0.96
RHS1C0 48.9 42.4 9.1 600 957.0 925.0 1.03
RHS1C40 48.5 42.1 9.1 600 1627.2 1610.0 1.01
RHS1C80 49.4 42.9 9.1 600 2180.0 2190.0 1.00
RHS2C0 55.8 49.4 6.3 600 537.3 536.0 1.00
RHS2C40 55.0 48.7 6.3 480 881.5 907.0 0.97
RHS2C60 55.4 49.1 6.3 480 1014.5 1030.0 0.98
RHS2C80 55.4 49.0 6.3 480 1280.1 1260.0 1.02
RHS3C0 45.5 39.0 7.0 600 558.2 549.0 1.02
RHS3C40 45.2 38.7 7.0 420 1048.7 996.0 1.05
RHS3C60 45.2 38.7 7.0 420 1096.9 1100.0 1.00
RHS3C80 45.3 38.7 7.0 420 1258.8 1310.0 0.96
Mean – – – – – – 1.00
COV – – – – – – 0.032

thickness in mm, as shown in Fig. 1 and Table 1. The measured depth of the flat portion-to-thickness (d/t) ratio and column
inner corner radius (ri ), overall depth-to-thickness (D/t) ratio, length (L) are summarized in Table 2. The specimens were
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 719

tested using nominal concrete cylinder strengths of 40, 60

and 80 MPa. The concrete-filled high strength stainless steel
tube column test specimens are labeled such that the shape of
stainless steel tube and concrete strength could be identified
from the label. For example, the label “SHS1C40” indicates that
the specimen has square hollow section that belonged to test
series SHS1, and the letter “C” indicates the concrete strength
followed by the value of concrete strength in MPa (40 MPa).
Five tests were conducted on stainless steel tube columns
without concrete infilled, denoted by “C0” in each series. The
measured cross-section dimensions of the test specimens are
detailed in Young and Ellobody [23].
The material properties of the stainless steel tube specimens
were determined by tensile coupon tests of flat and corner
portions. The tensile coupon test specimens of the flat portion
were taken from the center of the plate at 90◦ from the
weld in the longitudinal direction of the flat portion of the
untested specimens. The coupon dimensions conformed to the
Australian Standard AS 1391 [28] for the tensile testing of
metals using 12.5 mm wide coupons of gauge length 50 mm.
The initial Young’s modulus (E o ), static 0.2% proof stress
(σ0.2 ), static ultimate tensile strengths (σu ) and elongation after
Fig. 2. Finite element mesh of concrete-filled square hollow section specimens.
fracture (ε f ) based on a gauge length of 50 mm were measured.
The Ramberg–Osgood [29] parameter (n) that describes the main three components of these columns have to be modeled
shape of the stress–strain curve was 3, 4, 4, 5 and 6 for properly. These components are the confined concrete, the
series SHS1, SHS2, RHS1, RHS2 and RHS3, respectively. stainless steel tube and the interface between the concrete and
The tensile coupon tests of the flat coupons are detailed in the stainless steel tube. In addition to these parameters, the
Young and Ellobody [23]. The material properties of the corner choice of the element type and mesh size that provide reliable
portion of the specimens were also determined by tensile results with reasonable computational time is also important in
coupon tests for series RHS1 and RHS2, as detailed in Young
simulating structures with interface elements.
and Lui [17]. The material properties of the corners of series
SHS1, SHS2 and RHS3 were extrapolated from the material 3.2. Finite element type and mesh
properties of series RHS2 according to the measured 0.2%
tensile proof stress (σ0.2 ) in the flat portion for each series. The 4-noded doubly curved shell element with reduced
Table 1 shows the measured and predicted material properties integration, S4R, is used to model the buckling behaviour
of the flat and corner portions for series SHS1, SHS2, RHS1, of the stainless steel tube. The S4R element has six degrees
RHS2 and RHS3. The material properties of concrete were of freedom per node and provides an accurate solution
determined from standard cylinder tests. The concrete cylinder to most applications. The element allows for transverse
dimensions and test procedures conformed to the American shear deformation, which is important in simulating thick
Specification [20] for concrete testing. The mean compressive shell elements. A fine mesh of three dimensional 8-noded
cylinder strengths of the concrete were obtained as 46.6, 61.9 solid elements (C3D8) is used to model the concrete
and 83.5 MPa for nominal concrete cylinder strengths of C40, infilled. Different mesh sizes have been tried to choose
C60 and C80, respectively. The concrete cylinder tests are a reasonable mesh that provides both reliable results and
detailed in Young and Ellobody [23].
less computational time. It is found that a mesh size of
A 4600 kN capacity 815 Rock Mechanics Test System
1(length):1(width):1.25(depth), for most of the elements, can
servo-controlled hydraulic testing machine was used to apply
achieve accurate results. Fig. 2 shows the finite element mesh
compressive axial force to the column specimens. The load was
of the concrete-filled cold-formed stainless steel tube column
applied to the columns in axial uniform compression over the
of series SHS1.
concrete and steel tube. The concrete-filled cold-formed high
strength stainless steel tube column tests are detailed in Young 3.3. Boundary conditions and load application
and Ellobody [23].
Following the testing procedures conducted by Young and
3. Finite element model
Ellobody [23], the top and bottom surfaces of the concrete-
3.1. General filled cold-formed high strength stainless steel tube columns
were restrained against all degrees of freedom, except for the
In order to simulate the actual behaviour of concrete-filled displacement at the loaded end, which is the top surface, in the
cold-formed high strength stainless steel tube columns, the direction of the applied load, as shown in Fig. 2. Other nodes
720 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

were free to displace in any directions. The load was applied

in increments using the modified RIKS method available in the
ABAQUS library. The load was applied as static uniform load
using the displacement control at each node of the loaded top
surface, which is identical to the experimental investigation.

3.4. Material modeling of high strength stainless steel tubes

The measured stress–strain curves obtained from tensile

coupon tests presented by Young and Lui [17] for square and
rectangular cold-formed high strength stainless steel tubes have
been used in the finite element analysis. The material behaviour
provided by ABAQUS [27] (using the PLASTIC option) allows
Fig. 3. Equivalent uniaxial stress–strain curves for confined and unconfined
for a nonlinear stress–strain curve has been used. The first concrete.
part of the nonlinear curve representing the elastic part up to
the proportional limit stress, with measured values of Young’s To define the full equivalent uniaxial stress–strain curve for
modulus and Poisson’s ratio equals to 0.3, were used in the confined concrete as shown in Fig. 3, three parts of the curve
finite element model. have to be identified. The first part is the initially assumed
elastic range to the proportional limit stress. The value of the
3.5. Modeling of confined concrete proportional limit stress is taken as 0.5( f cc ) as given by Hu
et al. [24]. The initial Young’s modulus of confined concrete
Concrete-filled high strength stainless steel tube columns (E cc ) is reasonably calculated using the empirical Eq. (3) given
with a high value of d/t ratio provide inadequate confinement by ACI [20]. The Poisson’s ratio (υcc ) of confined concrete is
for the concrete. This is attributed to the premature failure of taken as 0.2.
the columns due to local buckling of stainless steel tubes. On 
E cc = 4700 fcc MPa. (3)
the other hand, concrete-filled high strength stainless steel tube
columns with a small value of d/t ratio provide remarkable The second part of the curve is the nonlinear portion starting
confinement for the concrete. In this case, the concrete from the proportional limit stress 0.5( f cc ) to the confined
strength is considerably improved and the concrete model concrete strength ( fcc ). This part of the curve can be
can be assumed as confined concrete model. Fig. 3 shows determined from Eq. (4), which is a common equation proposed
equivalent uniaxial stress–strain curves for both unconfined by Saenz [32]. This equation is used to represent the multi-
and confined concrete, where fc is the unconfined concrete dimensional stress and strain values for the equivalent uniaxial
cylinder compressive strength, which is equal to 0.8( f cu ), and stress and strain values. The unknowns of the equation are the
f cu is the unconfined concrete cube compressive strength. uniaxial stress ( f ) and strain (ε) values defining this part of the
The corresponding unconfined strain (εc ) is taken as 0.003, curve. The strain values (ε) are taken between the proportional
as recommended by the ACI Specification [20]. The confined strain, which is equal to (0.5 f cc /E cc ), and the confined strain
concrete compressive strength ( f cc ) and the corresponding (εcc ), which corresponds to the confined concrete strength. The
confined stain (εcc ) can be determined from Eqs. (1) and (2), stress values ( f ) can be determined easily from Eq. (4) by
respectively, proposed by Mander et al. [30]. assuming the strain values (ε).

f cc = f c + k1 fl (1) E cc ε

 f =    2  3
ε ε ε
fl 1 + (R + R E − 2) − (2R − 1) +R
εcc = εc 1 + k2 (2) εcc εcc εcc
fc (4)
where fl is the lateral confining pressure imposed by the steel
tube. The lateral confining pressure ( fl ) depends on the D/t where R E and R values are calculated from Eqs. (5) and (6),
ratio and the steel tube yield stress. The approximate value of respectively:
( fl ) can be obtained from empirical equations given by Hu et al. E cc εcc
[24], where a wide range of D/t ratios ranging from 17 to 150 RE = (5)
f cc
were investigated. The value of ( fl ) has a significant effect for R E (Rσ − 1) 1
steel tubes with a small D/t ratio. On the other hand, the value R= − (6)
(Rε − 1) 2 R ε
of ( fl ) is equal to zero for steel tubes with D/t ratios greater
than or equal to 29.2. The factors (k1 ) and (k2 ) are taken as 4.1 while the constants Rσ and Rε are taken to be equal to 4, as
and 20.5, respectively, as given by Richart et al. [31]. Knowing recommended by Hu and Schnobrich [33].
( fl ), (k1 ) and (k2 ), the values of equivalent uniaxial confined The third part of the confined concrete stress–strain curve is
concrete strength ( f cc ) and the corresponding confined strain the descending part from the confined concrete strength ( f cc )
(εcc ) can be determined using Eqs. (1) and (2). to a value lower than or equal to r k3 f cc with the corresponding
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 721

strain of 11εcc . The reduction factor (k3 ) depends on the D/t

ratio and the steel tube yield stress ( f y ). The approximate value
of k3 can be calculated from empirical equations given by Hu
et al. [24]. The reduction factor (r ) was introduced by Ellobody
et al. [26], based on the experimental investigation conducted
by Giakoumelis and Lam [11], to account for the effect of
different concrete strengths. The value of r is taken as 1.0 for
concrete with cube strength ( f cu ) equal to 30 MPa. The value
of r is taken as 0.5 for concrete with f cu greater than or equal
to 100 MPa. Linear interpolation is used to determine the value
of r for concrete cube strength between 30 and 100 MPa.
The yielding part of the confined stress–strain curve
for concrete, which is the part after the proportional limit
stress, is treated by the Drucker Prager yield criterion model
available in the ABAQUS material library. The model is Fig. 4. Load versus axial strain for column specimen SHS1C40.
used to define the yield surface and flow potential parameters
for materials subjected to triaxial compressive stresses. The from approximately 36% to 60% of the yield strength [34].
two parameters (*DRUCKER PRAGER and *DRUCKER Therefore, the lower proportional limits would affect the
PRAGER HARDENING) are used to define the yield stage of buckling behaviour of stainless steel structural members. The
confined concrete. The linear Drucker Prager model is used load versus axial strain curves observed from the tests were
with associated flow and isotropic rule. The material angle compared with the finite element analysis predictions. Fig. 4
of friction (β) and the ratio of flow stress in triaxial tension shows the load versus axial strain curves of the column
to that in compression (K ) are taken as 20 degrees and 0.8, specimen SHS1C40 obtained experimentally and numerically.
respectively, as recommended by Hu et al. [24]. Generally, good agreement between the experimental and
numerical curves is achieved. The axial strain observed in the
3.6. Modeling of concrete–stainless steel tube interface test was 1.56% at the ultimate load of 2768.1 kN, compared
with the axial strain of 1.38% at the ultimate load of 2850 kN
The contact between the stainless steel tube and the concrete obtained from the finite element analysis.
is modeled by interface elements. The interface elements The deformed shapes of the columns after failure observed
consist of two matching contact faces of stainless steel tube from the tests were also compared with the finite element
and concrete elements. The friction between the two faces is analysis predictions. The ABAQUS viewer [27] has been used
maintained as long as the surfaces remains in contact. The to plot the deformed shapes for all columns. It was found that
coefficient of friction between the two faces is taken as 0.25 good agreement exists between the experimental and numerical
in the analysis. The interface element allows the surfaces to deformed shapes of the columns. Fig. 5 shows a comparison
separate under the influence of tensile force. However, the between the deformed shapes observed experimentally and
contact elements are not allowed to penetrate each other. numerically for column specimen RHS2C40. The specimen
had a rectangular hollow section of nominal depth 160 mm,
4. Verification of finite element model nominal width 80 mm, nominal stainless steel tube plate
thickness 3 mm, and column length 480 mm. The column had a
In order to verify the finite element model, a comparison nominal concrete cylinder strength of 40 MPa. The failure mode
between the experimental results and finite element results was observed experimentally confirmed the numerical prediction,
carried out. The ultimate loads obtained from the tests (PTest ) which was mainly local buckling of the high strength stainless
and finite element analysis (PFE ), as well as the deformed steel tube.
shapes after failure, have been investigated. Table 2 shows a
comparison of the ultimate loads of the concrete-filled cold- 5. Parametric study
formed high strength stainless steel tube columns obtained
experimentally and numerically using the finite element model. A total of 40 columns were analyzed in the parametric
It can be seen that good agreement has been achieved between study, and the dimensions as well as the material properties of
both results for most of the columns. A maximum difference of the columns with SHS and RHS are summarized in Tables 3
6% was observed between experimental and numerical results and 4, respectively. The columns are labeled such that the
for column specimen SHS1C80. The mean value of PTest/PFE high strength stainless steel hollow section shape and concrete
ratio is 1.00 with the corresponding coefficient of variation cylinder strength can be identified from the label. For example,
(COV) of 0.032, as shown in Table 2. the label “S1C40” defines the concrete-filled cold-formed high
The mechanical properties of stainless steel are significantly strength stainless steel tube column with square hollow section
different from those of carbon steel. For carbon and low-alloy indicated by the letter “S”, and the letter “C” defines the
steels, the proportional limit is assumed to be at least 70% of the concrete strength followed by the value of the concrete cylinder
yield point, but for stainless steel the proportional limit ranges strength in MPa (40 MPa). Four series of SHS (S1, S2, S3
722 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

(a) Experimental. (b) FE Analysis.

Fig. 5. Comparison of experimental and finite element analysis failure mode of specimen RHS2C40.

Table 3
Specimen dimensions and material properties of concrete-filled SHS stainless steel tube columns for parametric study

Specimen Dimensions Material properties

D B t ri L D/t d/t As Ac Concrete f c Tube Fy = σ0.2
(mm) (mm) (mm) (mm) (mm) (mm2 ) (mm2 ) (MPa)
Flat (MPa) Corner (MPa)
S1C20 300 300 15.0 15.0 900 20.0 16.0 16 521 72 707 20 503 635
S1C40 300 300 15.0 15.0 900 20.0 16.0 16 521 72 707 40 503 635
S1C60 300 300 15.0 15.0 900 20.0 16.0 16 521 72 707 60 503 635
S1C80 300 300 15.0 15.0 900 20.0 16.0 16 521 72 707 80 503 635
S1C100 300 300 15.0 15.0 900 20.0 16.0 16 521 72 707 100 503 635
S2C20 300 300 4.5 4.5 900 66.7 62.7 5267 84 664 20 503 635
S2C40 300 300 4.5 4.5 900 66.7 62.7 5267 84 664 40 503 635
S2C60 300 300 4.5 4.5 900 66.7 62.7 5267 84 664 60 503 635
S2C80 300 300 4.5 4.5 900 66.7 62.7 5267 84 664 80 503 635
S2C100 300 300 4.5 4.5 900 66.7 62.7 5267 84 664 100 503 635
S3C20 450 450 5.5 5.5 1350 81.8 77.8 9701 192 695 20 503 635
S3C40 450 450 5.5 5.5 1350 81.8 77.8 9701 192 695 40 503 635
S3C60 450 450 5.5 5.5 1350 81.8 77.8 9701 192 695 60 503 635
S3C80 450 450 5.5 5.5 1350 81.8 77.8 9701 192 695 80 503 635
S3C100 450 450 5.5 5.5 1350 81.8 77.8 9701 192 695 100 503 635
S4C20 450 450 4.5 4.5 1350 100.0 96.0 7967 194 464 20 503 635
S4C40 450 450 4.5 4.5 1350 100.0 96.0 7967 194 464 40 503 635
S4C60 450 450 4.5 4.5 1350 100.0 96.0 7967 194 464 60 503 635
S4C80 450 450 4.5 4.5 1350 100.0 96.0 7967 194 464 80 503 635
S4C100 450 450 4.5 4.5 1350 100.0 96.0 7967 194 464 100 503 635

and S4) and four series of RHS (R1, R2, R3 and R4) were high strength stainless steel tube columns (PFE ) obtained from
studied. The values of the overall depth-to-width (D/B) ratio the parametric study using the verified finite element model
were 2.0, 1.5, 2.0 and 1.5 for series R1, R2, R3 and R4, for SHS and RHS, respectively. The test strengths (PTest ) are
respectively. The columns were designed for different (d/t) also shown in Tables 5 and 6 for the purpose of comparing the
ratios, which are different to that used in the experimental design strengths.
investigation conducted by Young and Ellobody [23], as shown
in Tables 3 and 4. The length of the columns was taken as 6. Comparison with design guides
three times the overall depth of the sections. The nonlinear
stress–strain curves of the flat and corner portions of series 6.1. General
RHS1 tested by Young and Lui [17] were used in the parametric
study. The five columns investigated in each series had concrete The strengths (PFE ) of the concrete-filled cold-formed
cylinder strengths of 20, 40, 60, 80 and 100 MPa. Tables 5 high strength stainless steel tube columns obtained from the
and 6 show the strengths of the concrete-filled cold-formed parametric study using the finite element model are compared
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 723

Table 4
Specimen dimensions and material properties of concrete-filled RHS stainless steel tube columns for parametric study

Specimen Dimensions Material properties

D B t ri L D/t d/t As Ac Concrete f c Tube Fy = σ0.2
(mm) (mm) (mm) (mm) (mm) (mm2 ) (mm2 ) (MPa)
Flat (MPa) Corner (MPa)
R1C20 300 150 15.0 15.0 900 20.0 16.0 12 021 32 207 20 503 635
R1C40 300 150 15.0 15.0 900 20.0 16.0 12 021 32 207 40 503 635
R1C60 300 150 15.0 15.0 900 20.0 16.0 12 021 32 207 60 503 635
R1C80 300 150 15.0 15.0 900 20.0 16.0 12 021 32 207 80 503 635
R1C100 300 150 15.0 15.0 900 20.0 16.0 12 021 32 207 100 503 635
R2C20 300 200 4.5 4.5 900 66.7 62.7 4367 55 564 20 503 635
R2C40 300 200 4.5 4.5 900 66.7 62.7 4367 55 564 40 503 635
R2C60 300 200 4.5 4.5 900 66.7 62.7 4367 55 564 60 503 635
R2C80 300 200 4.5 4.5 900 66.7 62.7 4367 55 564 80 503 635
R2C100 300 200 4.5 4.5 900 66.7 62.7 4367 55 564 100 503 635
R3C20 450 225 5.5 5.5 1350 81.8 77.8 7226 93 920 20 503 635
R3C40 450 225 5.5 5.5 1350 81.8 77.8 7226 93 920 40 503 635
R3C60 450 225 5.5 5.5 1350 81.8 77.8 7226 93 920 60 503 635
R3C80 450 225 5.5 5.5 1350 81.8 77.8 7226 93 920 80 503 635
R3C100 450 225 5.5 5.5 1350 81.8 77.8 7226 93 920 100 503 635
R4C20 450 300 4.5 4.5 1350 100.0 96.0 6617 128 314 20 503 635
R4C40 450 300 4.5 4.5 1350 100.0 96.0 6617 128 314 40 503 635
R4C60 450 300 4.5 4.5 1350 100.0 96.0 6617 128 314 60 503 635
R4C80 450 300 4.5 4.5 1350 100.0 96.0 6617 128 314 80 503 635
R4C100 450 300 4.5 4.5 1350 100.0 96.0 6617 128 314 100 503 635

Table 5
Comparison of column strengths and design strengths for concrete-filled SHS stainless steel tube columns
PFE or PTest PFE or PTest PFE or PTest
Specimen d/t PFE & PTest (kN) PACI/ASCE-1 (kN) PACI/ASCE-2 (kN) PACI/ASCE-3 (kN) PACI/ASCE-1 PACI/ASCE-2 PACI/ASCE-3

S1C20 16.0 10 573.1 9545.9 9825.8 9545.9 1.11 1.08 1.11

S1C40 16.0 11 210.2 10 781.9 11 061.8 10 781.9 1.04 1.01 1.04
S1C60 16.0 12 183.4 12 017.9 12 297.8 12 017.9 1.01 0.99 1.01
S1C80 16.0 13 247.2 13 253.9 13 533.8 13 253.9 1.00 0.98 1.00
S1C100 16.0 14 609.2 14 489.9 14 769.9 14 489.9 1.01 0.99 1.01
SHS1C40 22.0 2768.1 2399.4 2435.9 2399.4 1.15 1.14 1.15
SHS1C60 22.0 2972.0 2651.2 2687.6 2651.2 1.12 1.11 1.12
SHS1C80 22.0 3019.9 3004.4 3040.8 3004.4 1.01 0.99 1.01
SHS2C40 48.8 1381.5 1324.5 1335.9 1324.5 1.04 1.03 1.04
SHS2C60 48.8 1620.0 1596.9 1608.3 1596.9 1.01 1.01 1.01
SHS2C80 48.9 1851.3 1984.4 1995.8 1984.4 0.93 0.93 0.93
S2C20 62.7 3378.0 2874.3 2899.5 3276.7 1.18 1.17 1.03
S2C40 62.7 4997.0 4313.5 4338.7 4917.4 1.16 1.15 1.02
S2C60 62.7 6678.3 5752.8 5778.0 6558.2 1.16 1.16 1.02
S2C80 62.7 8328.7 7192.1 7217.3 8199.0 1.16 1.15 1.02
S2C100 62.7 10 026.3 8631.4 8656.6 9839.8 1.16 1.16 1.02
S3C20 77.8 6447.5 5479.2 5516.8 6245.1 1.18 1.17 1.03
S3C40 77.8 10 345.1 8755.0 8792.7 9980.7 1.18 1.18 1.04
S3C60 77.8 14 068.7 12 030.8 12 068.5 13 715.1 1.17 1.17 1.03
S3C80 77.8 17 963.1 15 306.7 15 344.3 17 449.6 1.17 1.17 1.03
S3C100 77.8 21 888.0 18 582.5 18 620.1 21 184.1 1.18 1.18 1.03
S4C20 96.0 6485.7 4812.2 4837.4 5485.9 1.35 1.34 1.18
S4C40 96.0 9766.0 8118.1 8143.3 9254.6 1.20 1.20 1.06
S4C60 96.0 13 695.0 11 424.0 11 449.2 13 023.4 1.20 1.20 1.05
S4C80 96.0 17 590.8 14 729.9 14 755.1 16 792.1 1.19 1.19 1.05
S4C100 96.0 21 527.7 18 035.8 18 061.0 20 560.8 1.19 1.19 1.05
Mean – – – – – 1.13 1.12 1.04
COV – – – – – 0.082 0.087 0.048
724 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

Table 6
Comparison of column strengths and design strengths for concrete-filled RHS stainless steel tube columns
PFE or PTest PFE or PTest PFE or PTest
Specimen d/t PFE & PTest (kN) PACI/ASCE-1 (kN) PACI/ASCE-2 (kN) PACI/ASCE-3 (kN) PACI/ASCE-1 PACI/ASCE-2 PACI/ASCE-3

R1C20 16.0 7927.3 6593.9 6873.8 6593.9 1.20 1.15 1.20

R1C40 16.0 8249.9 7141.4 7421.3 7141.4 1.16 1.11 1.16
R1C60 16.0 8465.3 7688.9 7968.8 7688.9 1.10 1.06 1.10
R1C80 16.0 8830.0 8236.4 8516.3 8236.4 1.07 1.04 1.07
R1C100 16.0 9341.7 8783.9 9063.8 8783.9 1.06 1.03 1.06
RHS3C40 38.7 1048.7 939.2 959.0 939.2 1.12 1.09 1.12
RHS3C60 38.7 1096.9 1067.4 1087.3 1067.4 1.03 1.01 1.03
RHS3C80 38.7 1258.8 1247.8 1267.6 1247.8 1.01 0.99 1.01
RHS1C40 42.1 1627.2 1742.6 1780.1 1742.6 0.93 0.91 0.93
RHS1C80 42.9 2180.0 2329.5 2366.3 2329.5 0.94 0.92 0.94
RHS2C40 48.7 881.5 1018.3 1037.0 1018.3 0.87 0.85 0.87
RHS2C60 49.1 1014.5 1159.0 1177.5 1159.0 0.88 0.86 0.88
RHS2C80 49.0 1280.1 1366.6 1385.0 1366.6 0.94 0.92 0.94
R2C20 62.7 2757.2 2323.1 2348.3 2648.3 1.19 1.17 1.04
R2C40 62.7 3590.6 3267.7 3292.9 3725.2 1.10 1.09 0.96
R2C60 62.7 4724.0 4212.3 4237.4 4802.0 1.12 1.11 0.98
R2C80 62.7 5863.6 5156.8 5182.0 5878.8 1.14 1.13 1.00
R2C100 62.7 6965.8 6101.4 6126.6 6955.6 1.14 1.14 1.00
R3C20 77.8 4233.9 3663.0 3700.6 4175.8 1.16 1.14 1.01
R3C40 77.8 6152.5 5259.6 5297.3 5995.9 1.17 1.16 1.03
R3C60 77.8 7948.5 6856.3 6893.9 7816.2 1.16 1.15 1.02
R3C80 77.8 9862.1 8452.9 8490.6 9636.3 1.17 1.16 1.02
R3C100 77.8 11 741.1 10 049.6 10 087.2 11 456.5 1.17 1.16 1.02
R4C20 96.0 4465.8 3652.0 3677.2 4163.3 1.22 1.21 1.07
R4C40 96.0 7156.9 5833.3 5858.5 6650.0 1.23 1.22 1.08
R4C60 96.0 9344.0 8014.7 8039.9 9136.8 1.17 1.16 1.02
R4C80 96.0 11 914.6 10 196.0 10 221.2 11 623.4 1.17 1.17 1.03
R4C100 96.0 14 343.4 12 377.3 12 402.5 14 110.1 1.16 1.16 1.02
Mean – – – – – 1.10 1.08 1.02
COV – – – – – 0.094 0.098 0.073

with unfactored design strengths predicted using the general strength. In the calculation of the strength of the stainless steel
guidelines specified in the American specifications [19,20] tubes, it was found that the values of the design stresses (Fn )
and Australian/New Zealand standards [21,22] for cold-formed are equal to the yield stresses (Fy ) for all columns. This is due
stainless steel and concrete structures. These specifications to the short column lengths. In this study, the yield stress (Fy )
consider the effect of local buckling of stainless steel is taken as the measured static 0.2% proof stress (σ0.2 ).
tubes using the effective width concept. The American The American and Australian/New Zealand specifications
specifications and Australian/New Zealand standards for cold- use the same Winter effective width equations in calculating the
formed stainless steel and concrete structures generally use the effective area (Ae ) of stainless steel tube cross-sections. In the
same formula to calculate the design strengths. The design calculation of buckling stress (Fn ), the design rules specified
strengths (PACI/ASCE ) for the concrete-filled stainless steel tube in the American Specification are based on the Euler column
columns were obtained by determining the strength of the strength that requires the calculation of tangent modulus (E t )
stainless steel tube (Ae Fn ) using the specifications [19,21] of using an iterative design procedure. The design rules specified
cold-formed stainless steel structures as well as the strength of in the Australian Standard adopt the Euler column strength
the concrete infilled (0.85 Ac f c ) using the specifications [20,22] or alternatively the Perry curve that needs only the initial
of concrete structures, as shown in Eq. (7). Young’s modulus (E o ) and a number of parameters to calculate
the design stress. In this study, the Euler column strength
PACI/ASCE = Ae Fn + 0.85 Ac f c (7)
method is used in the calculation of the design strengths for the
where Ae is the effective cross-sectional area of the stainless Australian/New Zealand standard. Hence, the design strengths
steel tube, which is equal to the full cross-sectional area (As ) in calculated using Eq. (7) are identical for both the American
the case of compact cross-sections and less than As in the case and Australian/New Zealand specifications. The columns were
of slender cross-sections due to the effect of local buckling, designed as concentrically loaded compression members. The
Fn is the flexural buckling stress determined according to term 0.85 Ac f c in Eq. (7) represents the contribution of the
Sections 3.4.1 and 3.4.2 of the American Specification [19] concrete infilled in the calculation of the column design
and Australian/New Zealand Standard [21], respectively, Ac is strengths. Three design approaches were investigated in the
the area of concrete, and f c is the measured concrete cylinder calculation of the column design strengths using Eq. (7).
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 725

The calculated unfactored design strengths are denoted by

PACI/ASCE-1 , PACI/ASCE-2 and PACI/ASCE-3 . The three design
approaches are presented in the following sections.

6.2. Design approach 1 (PACI/ASCE-1 )

The design strengths (PACI/ASCE-1 ) are calculated using the

material properties obtained from the tensile coupon test of
the flat portion for high strength stainless steel tube of series
RHS1 in the calculation of the term Ae Fn in Eq. (7). The
measured material properties obtained from the tensile coupon
test of the flat portion for series RHS1 are tabulated in Table 1.
Fig. 6. Comparison of FE strengths with design strengths for concrete-filled
The calculation of the strength of the concrete infilled for the specimens of series S1.
term 0.85 Ac fc in Eq. (7) is carried out using the measured
concrete cylinder strengths tabulated in Tables 3 and 4. The 6.5. Comparison of column strengths with design strengths
design strengths (PACI/ASCE-1 ) for the parametric study and
tests for the concrete-filled cold-formed stainless steel SHS and The comparison of the column strengths obtained from
RHS columns are shown in Tables 5 and 6, respectively. the parametric study (PFE ) and tests (PTest ) with the
design strengths (PACI/ASCE-1 , PACI/ASCE-2 and PACI/ASCE-3 ) is
6.3. Design approach 2 (PACI/ASCE-2 ) shown in Tables 5 and 6 for the concrete-filled cold-formed
stainless steel tube columns of square and rectangular hollow
The design strengths (PACI/ASCE-2 ) are calculated using both sections, respectively. The mean values of PFE /PACI/ASCE-1 ,
the material properties obtained from the tensile coupon tests PFE /PACI/ASCE-2 and PFE /PACI/ASCE-3 ratios, as well as
of the flat and corner portions for high strength stainless steel PTest /PACI/ASCE-1 , PTest /PACI/ASCE-2 and PTest /PACI/ASCE-3
tube of series RHS1 in the calculation of the term Ae Fn in ratios, with the corresponding coefficients of variation (COV)
Eq. (7). The measured material properties obtained from the are shown in Tables 5 and 6. It is shown that the design
tensile coupon tests of the flat and corner portions for series strengths calculated using the American and Australian/New
RHS1 are tabulated in Table 1. The strength of the concrete Zealand specifications are generally conservative for concrete-
infilled is calculated in the same way as design approach 1. filled cold-formed stainless steel square and rectangular hollow
Tables 5 and 6 summarize the design strengths (PACI/ASCE-2 ) section columns with d/t ratios less than 60, whereas the
of the concrete-filled cold-formed stainless steel SHS and RHS design strengths were more conservative for columns with
columns, respectively. d/t ratios greater than 60. Hence, modification to the design
rules specified in the American and Australian/New Zealand
specifications was proposed for concrete-filled cold-formed
6.4. Design approach 3 (PACI/ASCE-3 )
stainless steel square and rectangular hollow section columns
with d/t ratios greater than 60.
In this approach, the design strengths (PACI/ASCE-3 ) are The finite element strengths (PFE ) and the design strengths
calculated in the same way as design approach 1 (PACI/ASCE-1 ) (PAC I / ASC E−1 , PACI/ASCE−2 and PACI/ASCE-3 ) are plotted
for both the stainless steel tubes and concrete, and the design against the measured concrete strengths, as shown in
strengths are multiplied by a factor C, as shown in Eq. (8). Figs. 6–13. Figs. 6 and 10 show the comparison of the finite
PACI/ASCE-3 = C(Ae Fn + 0.85 Ac f c ) (8) element strengths with the design strengths for concrete-filled
cold-formed stainless steel tube columns of compact square and
where rectangular hollow sections, respectively. It can be seen that
 PACI/ASCE-1 and PACI/ASCE-2 , as well as the proposed design
1.0 for 16 ≤ d/t < 60
C= (9) strengths PACI/ASCE-3 , are generally conservative for compact
1.14 for 60 ≤ d/t ≤ 96.
square and rectangular sections with a d/t ratio of 16 for series
The factor C was calibrated with the column strengths S1 and R1. The comparison of the test strengths (PTest ) with the
obtained from the finite element analysis (PFE ) and tests (PTest ) design strengths (PACI/ASCE-1 , PACI/ASCE-2 and PACI/ASCE-3 ) for
for different d/t ratios. It is found that using C = 1.14 in the SHS columns with average d/t ratios of 22 and 48.8, shown in
calculation of PACI/ASCE-3 provides accurate design strengths Table 5, and RHS columns with average d/t ratios of 38.7, 42.5
for concrete-filled cold-formed stainless steel SHS and RHS and 48.9, shown in Table 6, are generally conservative, except
columns for 60 ≤ d/t ≤ 96, while using C = 1.0 in the for series RHS1 and RHS2.
calculation of PACI/ASCE-3 provides accurate design strengths Figs. 7–9 and 11–13 and Tables 5 and 6 show the
for 16 ≤ d/t < 60. Tables 5 and 6 summarize the proposed comparison of the finite element strengths with the design
design strengths (PACI/ASCE-3 ) for the parametric study and strengths for concrete-filled cold-formed stainless steel tube
tests. columns of slender square hollow sections with d/t ratios of
726 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

Fig. 7. Comparison of FE strengths with design strengths for concrete-filled Fig. 10. Comparison of FE strengths with design strengths for concrete-filled
specimens of series S2. specimens of series R1.

Fig. 8. Comparison of FE strengths with design strengths for concrete-filled Fig. 11. Comparison of FE strengths with design strengths for concrete-filled
specimens of series S3. specimens of series R2.

Fig. 9. Comparison of FE strengths with design strengths for concrete-filled Fig. 12. Comparison of FE strengths with design strengths for concrete-filled
specimens of series S4. specimens of series R3.

62.7, 77.8 and 96 (series S2, S3 and S4) and slender rectangular strengths of concrete-filled cold-formed stainless steel tube
hollow sections with d/t ratios of 62.7, 77.8 and 96 (series R2, columns for slender square and rectangular sections with d/t
R3 and R4). It can be seen that the design strengths PACI/ASCE-1 ratios between 60 and 96. Hence, it is recommended that the
and PACI/ASCE-2 are generally conservative for all the columns proposed modified Eq. (8) can be used for concrete-filled cold-
compared with the PFE and PTest . This is because the column formed stainless steel SHS and RHS columns.
strengths of the stainless steel slender square and rectangular Figs. 14 and 15 plotted the relationship between the column
hollow sections are improved by the presence of the concrete strengths obtained from finite element analysis (PFE ) and tests
infilled, which is not considered in the current design rules. (PTest ) versus the proposed design strengths (PACI/ASCE-3 ) for
In this study, it can be seen from Figs. 7–9 and 11–13 and SHS and RHS, respectively. Generally, it can be seen that the
Tables 5 and 6 that the use of a factor C of 1.14 in the proposed proposed design strengths are accurately predicted for concrete-
design strengths PACI/ASCE-3 accurately predicted the column filled cold-formed stainless steel square and rectangular tube
 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728 727

7. Conclusions

A nonlinear finite element model for the analysis of

concrete-filled cold-formed stainless steel tube columns has
been presented. The material nonlinearities of high strength
stainless steel tubes and confined concrete have been accurately
introduced. The comparison between the finite element results
and the experimental results of the columns with different
cross-section geometries and different concrete strengths
showed good agreement in predicting the strengths of the
columns. The column strengths and deformed shapes of the
columns have been predicted using the finite element model and
compared well with the experimental results. Parametric study
Fig. 13. Comparison of FE strengths with design strengths for concrete-filled
specimens of series R4. of 40 concrete-filled cold-formed stainless steel tube columns
of square and rectangular hollow sections with a depth of the
flat portion-to-plate thickness (d/t) ratio ranged from 16 to 96,
as well as different concrete cylinder strengths ranged from 20
to 100 MPa, was performed using the finite element analysis.
The column strengths obtained from the finite element
analysis were compared with the design strengths predicted
using the American and Australian/New Zealand specifications
for cold-formed stainless steel and concrete structures. The
material properties of the high strength stainless steel tube
specimens obtained from tensile coupon tests of flat and
corner portions were used to calculate the deign strengths.
It is shown that the design strengths calculated using the
American and Australian/New Zealand specifications are
generally conservative for concrete-filled cold-formed stainless
steel square and rectangular hollow section columns with
d/t ratio less than 60, whereas the design strengths were
more conservative for columns with d/t ratio greater than
60. Hence, modification to the design rules specified in
Fig. 14. PFE and PTest versus PACI/ASCE-3 for concrete-filled SHS stainless the American and Australian/New Zealand specifications was
steel tube columns. proposed for concrete-filled cold-formed stainless steel square
and rectangular hollow section columns with d/t ratio greater
than 60. The design strengths predicted using the proposed
modified equation are more accurate compared to the design
strengths calculated using the American and Australian/New
Zealand specifications.

References

[1] Schneider SP. Axially loaded concrete-filled steel tubes. Journal of

Structural Engineering, ASCE 1998;124(10):1125–38.
[2] Uy B. Local and post-local buckling of concrete filled steel welded box
columns. Journal of Constructional Steel Research 1998;74(1–2):47–72.
[3] Uy B. Static long-term effects in short concrete-filled steel box columns
under sustained loading. ACI Structural Journal 2001;98(1):96–104.
[4] Uy B. Strength of short concrete filled high strength steel box columns.
Journal of Constructional Steel Research 2001;57(2):113–34.
[5] Huang CS, Yeh YK, Hu HT, Tsai KC, Weng YT, Wang SH et al.
Axial load behavior of stiffened concrete-filled steel columns. Journal of
Structural Engineering, ASCE 2002;128(9):1222–30.
Fig. 15. PFE and PTest versus PACI/ASCE-3 for concrete-filled RHS stainless
[6] Han LH, Yao GH. Influence of concrete compaction on the strength
steel tube columns.
of concrete-filled steel RHS columns. Journal of Constructional Steel
Research 2003;59(6):751–67.
[7] Mursi M, Uy B. Strength of concrete filled steel box columns
columns with different d/t ratios of compact and slender cross- incorporating interaction buckling. Journal of Structural Engineering,
sections. ASCE 2003;129(5):626–39.
728 E. Ellobody, B. Young / Engineering Structures 28 (2006) 716–728

[8] Liu D, Gho WM, Yuan J. Ultimate capacity of high-strength rectangular [22] Australian Standards AS3600. Concrete structures. Sydney (Australia):
concrete-filled steel hollow section stub columns. Journal of Construc- Standards Australia, AS3600-1994; 1994.
tional Steel Research 2003;59(12):1499–515. [23] Young B, Ellobody E. Experimental investigation of concrete-filled
[9] Uy B. High-strength steel–concrete composite columns for buildings. cold-formed high strength stainless steel tube columns. Journal of
Structures & Buildings 2003;156:3–14. Constructional Steel Research [in press].
[10] Sakino K, Nakahara H, Morino S, Nishiyama I. Behavior of centrally [24] Hu HT, Huang CH, Wu MH, Wu YM. Nonlinear analysis of axially
loaded concrete-filled steel-tube short columns. Journal of Structural loaded concrete-filled tube columns with confinement effect. Journal of
Engineering, ASCE 2004;130(2):180–8. Structural Engineering, ASCE 2003;129(10):1322–9.
[11] Giakoumelis G, Lam D. Axial capacity of circular concrete-filled tube [25] Roufegarinejad A, Uy B, Bradford MA. Behaviour and design of concrete
columns. Journal of Constructional Steel Research 2004;60(7):1049–68. filled steel columns utilizing stainless steel cross sections under combined
[12] Young B, Hartono W. Compression tests of stainless steel tubular actions. In: Proceedings of the 18th Australasian conference on the
members. Journal of Structural Engineering, ASCE 2002;128(6):754–61. mechanics of structures and materials. 2004. p. 159–65.
[13] Young B, Liu Y. Experimental investigation of cold-formed stainless steel [26] Ellobody E, Young B, Lam D. Behaviour of normal and high strength
columns. Journal of Structural Engineering, ASCE 2003;129(2):169–76. concrete-filled compact steel tube circular stub columns. Journal of
[14] Liu Y, Young B. Buckling of stainless steel square hollow section Constructional Steel Research 2005 [in press].
compression members. Journal of Constructional Steel Research 2003; [27] ABAQUS Standard User’s Manual, vol. 1–3. Version 6.4. USA: Hibbitt,
59(2):165–77. Karlsson and Sorensen, Inc.; 2004.
[15] Gardner L, Nethercot DA. Experiments on stainless steel hollow sections- [28] AS. Methods for tensile testing of metals. Australian Standard, AS
Part 1: Material and cross-sectional behaviour. Journal of Constructional 1391–1991. Sydney (Australia): Standards Association of Australia;
Steel Research 2004;60:1291–318. 1991.
[16] Gardner L, Nethercot DA. Experiments on stainless steel hollow [29] Ramberg W, Osgood WR. Description of stress strain curves by three
sections-Part 2: Member behaviour of columns and beams. Journal of parameters. Tech. note no 902. Washington (DC): National Advisory
Constructional Steel Research 2004;60:1319–32. committee for Aeronautics; 1943.
[17] Young B, Lui WM. Behavior of cold-formed high strength stainless [30] Mander JB, Priestley MJN, Park R. Theoretical stress–strain model for
steel sections. Journal of Structural Engineering, ASCE 2005;131(11): confined concrete. Journal of Structural Engineering, ASCE 1988;114(8):
1738–45. 1804–26.
[18] Young B, Lui WM. Experimental investigation of cold-formed high [31] Richart FE, Brandzaeg A, Brown RL. A study of the failure of concrete
strength stainless steel compression members. In: Proceedings of the 6th under combined compressive stresses. Bull. 185. Champaign (Illinois,
international conference on tall buildings. 2005. p. 657–65. USA): Univ. of Illinois Engineering Experimental Station; 1928.
[19] ASCE. Specification for the design of cold-formed stainless steel [32] Saenz LP. Discussion of ‘Equation for the stress–strain curve of concrete’
structural members. Reston (Virginia): American Society of Civil by P. Desayi, and S. Krishnan. Journal of the American Concrete Institute
Engineers, SEI/ASCE-8-02; 2002. 1964;61:1229–35.
[20] ACI. Building code requirements for structural concrete and commentary. [33] Hu HT, Schnobrich WC. Constitutive modeling of concrete by using
Detroit (USA): American Concrete Institute, ACI 318-95; 1999. nonassociated plasticity. Journal of Materials in Civil Engineering 1989;
[21] AS/NZS. Cold-formed stainless steel structures. Australian/New 1(4):199–216.
Zealand Standard, AS/NZS 4673:2001. Sydney (Australia): Standards [34] Yu WW. Cold-formed steel design. 3rd ed. New York: John Wiley and
Australia; 2001. Sons, Inc.; 2000.