ARTICLE IN PRESS

Thin-Walled Structures 47 (2009) 12721280

Contents lists available at ScienceDirect

Thin-Walled Structures
journal homepage: www.elsevier.com/locate/tws

Concrete-lled aluminum circular hollow section column tests

Feng Zhou, Ben Young 
Department of Civil Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong, China

a r t i c l e in f o a b s t r a c t

Article history: An experimental investigation of concrete-lled aluminum circular hollow section (CHS) stub columns
Received 4 March 2009 is presented in this paper. A series of tests was conducted to investigate the effects of the geometric
Received in revised form dimension of the aluminum CHS and concrete strength on the behavior and strength of concrete-lled
31 March 2009
aluminum CHS stub columns. The structural performance of the concrete-lled aluminum CHS stub
Accepted 31 March 2009
Available online 8 May 2009
columns was investigated using different concrete cylinder strengths of 40, 70 and 100 MPa. The CHS
tubes were fabricated by extrusion using 6061-T6 heat-treated aluminum alloy having nominal 0.2%
Keywords: proof stress of 240 MPa. The diameter-to-thickness ratio of the CHS tubes ranged from 9.7 to 59.7. The
Aluminum tubes column lengths were chosen so that the length-to-diameter ratio generally remained at a constant value
Circular hollow sections
of 3 to prevent overall column buckling. The concrete-lled aluminum CHS specimens were subjected to
Composite columns
uniform axial compression. The column strengths, load-axial shortening relationship, load-axial strain
Concrete
Experimental investigation relationship and failure modes of columns were presented. The test strengths were compared with the
Tubular structures design strengths calculated using the American specications and Australian/New Zealand standards
for aluminum and concrete structures. It is shown that the design strengths are generally conservative
for concrete-lled aluminum CHS stub columns.
& 2009 Elsevier Ltd. All rights reserved.

1. Introduction aluminum alloy. The overall depth-to-thickness ratio of the tube

It is well known that concrete-lled steel composite columns
have the advantages of high-bearing capacity and ductility, easy
construction and cost saving [14]. Similarly, aluminum tube
columns lled with concrete can effectively take advantages of
these two materials to provide both high strength and high
stiffness. Furthermore, the aluminum tubes surrounding the
concrete core eliminate permanent formwork; hence, construc-
tion time can be reduced. However, little research has been
carried out on concrete-lled aluminum tube composite columns.
Hence, there is a need to investigate the structural performance of
concrete-lled aluminum tube columns.
Experimental investigation of concrete-lled aluminum square
and rectangular hollow sections (SHS and RHS) composite
columns was reported by Zhou and Young [5]. A series of tests
was conducted to investigate the effects of the section size of the
aluminum tubes, plate thickness and concrete strength on the
behavior and strength of concrete-lled aluminum tube columns,
using square and rectangular hollow sections. The structural
performance of the concrete-lled aluminum tube columns was
investigated using different concrete cylinder strengths. The
tubes were fabricated by extrusion using 6061-T6 heat-treated
 Corresponding author. Tel.: +852 2859 2674; fax: +852 2559 5337.
E-mail address: young@hku.hk (B. Young).
sections ranged from 8.2 to 63.8. It was shown that the stiffness of
the composite columns improves compared with the aluminum
SHS and RHS tube columns without concrete inll. Local buckling
of the aluminum SHS and RHS tubes was found for specimens
with slender sections. Generally, the composite columns failed by
the aluminum tubes splitting near the corner of the SHS and RHS
sections. Therefore, the design strengths predicted using the
American and Australian/New Zealand (AS/NZS) specications are
generally unconservative for most of the concrete-lled alumi-
num SHS and RHS tube columns.
Circular hollow section (CHS) tube provides much better
conning effect to concrete core, especially, when the diameter-
to-thickness (D/t) ratio is small compared with SHS and RHS tubes
[6]. Furthermore, splitting of aluminum CHS tube is not likely to
occur.
The purpose of this paper is to investigate the structural
behavior and strength of concrete-lled aluminum CHS stub
columns by testing. A series of tests was performed on aluminum
CHS tubes with concrete inll of different strengths. The stub
composite columns were subjected to uniform axial compression.
The dimensions of the aluminum tube cross-sections were
chosen, so that they include both compact and relatively slender
sections. The test strengths were compared with the design
strengths calculated using the general design guidelines specied
in the American specications [7,8] and Australian/New Zealand
standards [9,10], for aluminum and concrete structures.

0263-8231/$ - see front matter & 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tws.2009.03.014
 ARTICLE IN PRESS
F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280
Nomenclature
The following symbols are used in this paper:
Aa full cross-section area of aluminum CHS tube
Ac area of concrete
D outer diameter of CHS
E0 initial Youngs modulus
FL limit state stress of aluminum tube
fc measured concrete cylinder strength
kc coefcient for compression members
L length of column specimen
P design strength
PAA1 nominal axial strength calculated using the American
specications according to design approach 1
PAA2 nominal axial strength calculated using the American
specications according to design approach 2
2. Experimental investigation
2.1. Test specimens
A series of tests was conducted to study the structural
performance of concrete-lled aluminum circular hollow section
stub columns. The CHS tubes were fabricated by extrusion using
6061-T6 heat-treated aluminum alloy, hence the residual stresses
of the tubes are very small and can be ignored. The test program
consisted of ten test series of CHS tubes (CHS1CHS10). The
nominal section sizes (D  t) of series CHS1, CHS2, CHS3, CHS4,
CHS5, CHS6, CHS7, CHS8, CHS9 and CHS10 are 38  4, 50  3,
60  2.5, 76  2, 100  2, 120  2.5, 150  2.5, 150  5, 160  4 and
180  3.5 mm2, respectively, where D is the diameter and t is the
thickness of the sections, as shown in Fig. 1. The measured
diameter-to-thickness ratio of the CHS tubes ranged from 9.7 to
59.7. The column lengths (L) were chosen so that the length-to-
diameter ratio (L/D) for the concrete-lled aluminum CHS stub
columns generally remained at a constant value of 3 to prevent
overall column buckling. The column specimens were tested using
nominal concrete cylinder strengths of 40, 70 and 100 MPa. The
aluminum CHS stub columns without concrete inll were also
tested for reference purposes. The measured dimensions of the
concrete-lled aluminum CHS stub column test specimens are
shown in Table 1.
Aluminum tube
t length were used to measure the longitudinal strain. A data
Concrete
D relatively slender section of CHS9 are shown in Figs 2(a) and (b),
Fig. 1. Denition of symbols for concrete-lled aluminum circular hollow section
(CHS) specimens.
1273
PAS/NZS1 nominal axial strength calculated using the Austra-
lian/New Zealand standards according to design
approach 1
PAS/NZS2 nominal axial strength calculated using the Austra-
lian/New Zealand standards according to design
approach 2
P3 nominal axial strength calculated according to design
approach 3
PExp test ultimate load (test strength)
t thickness of aluminum CHS tube
ef elongation (tensile strain) after fracture based on
gauge length of 50 mm
s0.2 static 0.2% proof stress
su static ultimate stress
D axial shortening of specimens
2.2. Specimen labeling
In Table 1, the column test specimens are labelled such that the
shape of the aluminum tube and nominal concrete cylinder
strength could be identied from the label. For example, the label
CHS6C70-R denes the following specimen:
 The rst four letters indicate the shape of aluminum tube and
the test series, where CHS is the circular hollow section. This
specimen label denes the specimen belonged to test series
CHS6, with the nominal dimensions of the CHS of
120  2.5 mm2.
 The following notation C70 indicates the nominal concrete
cylinder strength in MPa, where C70 indicates 70 MPa.
 If a test was repeated, then -R indicates the repeated test.
2.3. Material properties of aluminum tubes
The material properties of the aluminum tube specimens were
determined by tensile coupon tests as well as stub column tests.
The tensile coupons were taken from the curved faces of the CHS
tubes in the longitudinal direction. The coupon specimens of
6 mm wide with a gauge length of 25 mm were extracted from the
CHS tubes. Circular holes with a diameter of 8.5 mm were drilled
near both ends of the curved coupons, and the coupons were
tested between two pins in a MTS displacement-controlled testing
machine. This avoids the bending stress that could be introduced
from the singly-symmetric-shaped coupons during the tests. Two
strain gauges and a calibrated extensometer of 25 mm gauge
acquisition system was used to record the load and strain at
regular intervals during tests. The static load was obtained by
pausing the applied straining for 1.5 min near the 0.2% tensile
proof stress and near the ultimate tensile strength. This allowed
stress relaxation associated with plastic straining to take place.
The material properties obtained from the tensile coupon tests are
summarized in Table 2, which includes the static 0.2% tensile
proof stress (s0.2), static tensile strength (su), initial Youngs
modulus (Eo) and elongation after fracture (ef) based on a gauge
length of 50 mm. The typical stressstrain curves obtained from
the tensile coupon tests for compact section of CHS2 and
respectively.
The stub column tests of the aluminum CHS tubes were
conducted to determine the material properties of the complete
 ARTICLE IN PRESS
1274 F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280

Table 1
Measured test specimen dimensions of circular hollow sections (CHS).

Specimen Diameter, D (mm) Thickness, t (mm) D/t Length, L (mm) L/D Aluminum area, Aa (mm2) Concrete area, Ac (mm2)

CHS1C0 38.0 3.91 9.7 114 3.0 419 0

CHS1C40 38.0 3.89 9.8 114 3.0 417 717
CHS1C70 38.0 3.90 9.7 114 3.0 418 716
CHS1C100 38.0 3.92 9.7 114 3.0 420 715
CHS2C0 50.0 3.13 16.0 150 3.0 461 0
CHS2C40 50.0 3.13 16.0 150 3.0 461 1503
CHS2C70 50.0 3.12 16.0 150 3.0 460 1504
CHS2C100 50.0 3.13 16.0 150 3.0 461 1503
CHS3C0 60.0 2.52 23.8 180 3.0 455 0
CHS3C40 60.0 2.55 23.5 180 3.0 460 2368
CHS3C70 60.0 2.54 23.6 180 3.0 459 2369
CHS3C100 59.9 2.53 23.7 180 3.0 456 2362
CHS4C0 76.1 2.05 37.1 228 3.0 477 0
CHS4C40 76.1 2.06 36.9 228 3.0 479 4070
CHS4C70 76.0 2.06 36.9 228 3.0 479 4058
CHS4C100 76.0 2.05 37.1 228 3.0 476 4061
CHS5C0 99.9 2.02 49.5 300 3.0 621 0
CHS5C0-R 99.8 2.00 49.9 299 3.0 615 0
CHS5C40 99.7 2.02 49.4 300 3.0 620 7188
CHS5C70 99.8 2.06 48.4 300 3.0 633 7191
CHS5C100 100.0 2.05 48.8 300 3.0 631 7224
CHS6C0 119.7 2.55 46.9 360 3.0 939 0
CHS6C40 119.8 2.49 48.1 360 3.0 918 10356
CHS6C70 120.0 2.55 47.1 360 3.0 941 10370
CHS6C70-R 119.6 2.48 48.2 360 3.0 913 10323
CHS6C100 119.9 2.48 48.3 360 3.0 915 10377
CHS7C0 149.8 2.51 59.7 449 3.0 1162 0
CHS7C40 150.1 2.53 59.3 450 3.0 1173 16524
CHS7C70 150.1 2.54 59.1 451 3.0 1178 16520
CHS7C100 149.9 2.53 59.2 450 3.0 1171 16479
CHS8C0 150.2 4.99 30.1 448 3.0 2277 0
CHS8C40 150.2 5.03 29.9 450 3.0 2294 15427
CHS8C70 150.2 5.04 29.8 450 3.0 2299 15422
CHS8C100 150.2 5.03 29.9 450 3.0 2294 15427
CHS9C0 160.2 4.01 40.0 480 3.0 1968 0
CHS9C40 160.1 4.03 39.7 480 3.0 1976 18158
CHS9C70 160.5 4.07 39.4 480 3.0 2000 18234
CHS9C100 160.5 4.06 39.5 480 3.0 1996 18239
CHS10C0 180.2 3.75 48.1 540 3.0 2079 0
CHS10C40 180.0 3.71 48.5 540 3.0 2055 23395
CHS10C70 180.4 3.69 48.9 540 3.0 2049 23515
CHS10C100 180.5 3.75 48.1 540 3.0 2083 23509

Table 2 the applied load and the readings of displacement transducers as

Measured material properties obtained from tensile coupon tests. well as the strain gauge readings at regular intervals during the
tests. Table 3 shows the measured material properties obtained
Test series Section D  t (mm) s0.2 (MPa) su (MPa) Eo (GPa) ef (%)
from the stub column tests of the aluminum CHS tubes, which
CHS1 38  4 242.4 278.5 65.0 8.1 includes the 0.2% proof stress (s0.2), tensile strength (su) and
CHS2 50  3 238.4 259.1 66.1 8.9 initial Youngs modulus (Eo). The values of the s0.2 for CHS5, CHS7
CHS3 60  2.5 237.8 261.4 69.8 8.4 and CHS10 were not obtained. This was due to the strain at
CHS4 76  2 237.0 256.0 64.9 8.2
CHS5 100  2 244.3 259.5 65.6 7.4
ultimate load less than the 0.2% stain offset. The typical
CHS6 120  2.5 253.1 264.7 66.5 5.0 stressstrain curves obtained from the stub column tests for
CHS7 150  2.5 267.9 282.9 64.9 10.1 compact section of CHS2 and relatively slender section of CHS9
CHS8 150  5 216.9 251.9 65.8 9.5 are also shown in Figs 2(a) and (b), respectively.
CHS9 160  4 254.2 272.9 66.6 8.4
CHS10 180  3.5 264.9 287.1 68.7 6.2

2.4. Material properties of concrete

cross-section. The stub column test strengths were also obtained.
The measured cross-section dimension and column length of
the stub column specimens without concrete inll are shown in
Table 1. A servo-controlled hydraulic testing machine was used to
apply compressive axial force to the test specimens. Displacement
control was used to drive the hydraulic actuator at a constant
speed of 0.2 mm/min for all test specimens. The use of dis-
placement control allowed the tests to be continued into the
post-ultimate range. A data acquisition system was used to record
The material properties of concrete were determined from
standard cylinder tests. The concrete cylinder dimensions and test
procedure conformed to the American Specication [8] for
concrete testing. The three nominal concrete strengths of C40,
C70 and C100 were produced using commercially available
materials with normal mixing and curing techniques. The
concrete mix design is shown in Table 4. A total of twenty-nine
concrete cylinder tests were conducted. The mean values of the
measured compressive concrete cylinder strength are 44.8, 70.2
 ARTICLE IN PRESS
F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280 1275

and 106.0 MPa with the corresponding coefcients of variation

350 (COV) of 0.033, 0.018 and 0.039 for nominal concrete strengths of
C40, C70 and C100, respectively. Table 5 summarizes the mea-
300 sured concrete cylinder strengths and the number of tests. The
concrete cylinder tests were conducted at the time of the
250 concrete-lled aluminum tube column tests.
Stress (MPa)

200
2.5. Column testing procedure

150
A typical concrete-lled aluminum CHS stub column test is
shown in Fig. 3. A servo-controlled hydraulic testing machine was
100
Tensile coupon test used to apply compressive axial force to the column specimens.
Stub column test Prior to testing, both ends of the columns were milled at, and
50
then strengthened by ber reinforced polymer (FRP). Hence, the
column failure would not occur at the ends of the columns.
0
0 2 4 6 8 10 Strengthening the ends of the columns by FRP has been used by
Strain (%) Zhou and Young [5] for concrete-lled aluminum SHS and RHS
columns. The ends of the columns were then cast in plaster to
ensure the load was applied uniformly on the column specimens.
300 The load was applied on the columns by uniform axial
compression over the concrete and aluminum tube, as shown in
250 Fig. 3. Displacement control was used to drive the hydraulic
actuator at a constant speed of 0.2 mm/min. This allowed the
tests to be continued in the post-ultimate range. Three
200 displacement transducers were used to measure the axial
Stress (MPa)

shortening of each column. The axial shortening was obtained

from the average readings of the three displacement trans-
150
ducers for each column specimen. For test series of CHS2 with
compact section and CHS7 with relatively slender section, ve
100 strain gauges (one strain gauge near both ends and three
strain gauges in the middle height of the columns) were also
Tensile coupon test used to monitor the axial strain of the aluminum CHS tubes.
50 All strain gauges were placed on the outside surface of the
Stub column test
aluminum CHS tubes. The top and bottom gauges were located at
a distance of 35 mm from both ends of the columns. The three
0
0 2 4 6 8 10 middle strain gauges were equally spaced in a quarter portion
Strain (%) of the columns. During the tests, a data acquisition system was
used to record the applied load, the readings of displacement
Fig. 2. Typical stressstrain curves: (a) compact section of CHS2 and (b) relatively transducers as well as the readings of strain gauges at regular
slender section of CHS9.
intervals.

Table 3
Measured material properties obtained from stub column tests. 2.6. Test results

Test series Section D  t (mm) s0.2 (MPa) su (MPa) Eo (GPa)

The test strengths, load-axial shortening relationship and load-
CHS1 38  4 185.2 273.3 66.9 axial strain relationship were measured for the column speci-
CHS2 50  3 264.0 306.3 66.0 mens. The test strengths (PExp) of the concrete-lled aluminum
CHS3 60  2.5 242.2 266.6 66.7 CHS stub columns are shown in Table 6. The load-axial shortening
CHS4 76  2 230.6 237.7 73.6
relationship of the concrete-lled aluminum CHS stub columns
CHS5 100  2 262.0 66.7
CHS6 120  2.5 280.1 281.7 66.4 for series CHS2 and CHS7 obtained from the displacement trans-
CHS7 150  2.5 244.3 70.9 ducer readings are shown in Figs 4(a) and 5(a), respectively.
CHS8 150  5 184.7 230.9 70.2 The initial parts of the load-axial shortening curves of the
CHS9 160  4 221.0 231.8 69.8
concrete-lled aluminum CHS tube columns have larger slopes
CHS10 180  3 232.2 66.5
compared with the aluminum CHS tube columns without

Table 4
Concrete mix design.

Nominal concrete strength (MPa) Water/cement ratio Mix proportions (to the weight of cement)

Cement Water Fine aggregate 10 mm aggregate CSFa SPb

C40 0.55 1.0 0.55 2.04 3.06 0.00 0.004

C70 0.37 1.0 0.37 1.80 1.80 0.00 0.007
C100 0.27 1.0 0.27 1.60 1.98 0.18 0.019

a
CSF condensed silica fume.
b
SP super plasticizer.
 ARTICLE IN PRESS
1276 F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280

Table 5 Table 6
Measured concrete cylinder strengths. Comparison of test strengths with design strengths for concrete-lled aluminum
circular hollow section (CHS) columns.
Nominal concrete Mean value of measured Coefcient of Number of
strength (MPa) concrete strength (MPa) variation (COV) concrete cylinder Specimen PExp (kN) PExp/PAA1 PExp/PAS/NZS1 PExp/PAA2 PExp/PAS/NZS2 PExp/P3
tests
CHS1C0 114.5 1.13 1.26 1.48 1.65 1.00
C40 44.8 0.033 6 CHS1C40 158.9 1.27 1.39 1.56 1.70 1.14
C70 70.2 0.018 11 CHS1C70 167.2 1.16 1.26 1.39 1.50 1.06
C100 106.0 0.039 12 CHS1C100 171.5 1.06 1.13 1.24 1.32 0.98
CHS2C0 141.2 1.28 1.44 1.16 1.30 1.00
CHS2C40 217.0 1.35 1.45 1.26 1.36 1.13
CHS2C70 238.9 1.20 1.28 1.13 1.21 1.04
CHS2C100 327.5 1.38 1.45 1.31 1.39 1.22
CHS3C0 121.3 1.12 1.26 1.10 1.23 1.00
CHS3C40 244.1 1.29 1.37 1.27 1.36 1.21
CHS3C70 292.4 1.17 1.23 1.16 1.22 1.11
CHS3C100 412.6 1.33 1.39 1.33 1.38 1.28
CHS4C0 113.4 1.04 1.12 1.06 1.15 1.00
CHS4C40 329.9 1.33 1.38 1.34 1.39 1.31
CHS4C70 415.7 1.19 1.21 1.19 1.22 1.17
CHS4C100 611.4 1.35 1.37 1.35 1.38 1.33
CHS5C0 162.7 1.17 1.20 1.00
CHS5C0-R 160.4 1.17 1.20 0.99
CHS5C40 543.6 1.42 1.43 1.34
CHS5C70 712.0 1.25 1.26 1.21
CHS5C100 995.8 1.32 1.32 1.28
CHS6C0 264.5 1.20 1.25 1.10 1.13 1.00
CHS6C40 822.8 1.45 1.47 1.40 1.41 1.33
CHS6C70 1010.3 1.21 1.22 1.18 1.19 1.15
CHS6C70-R 1004.0 1.21 1.22 1.18 1.19 1.14
CHS6C100 1388.7 1.27 1.28 1.24 1.25 1.21
CHS7C0 283.9 1.04 1.04 1.00
CHS7C40 1111.1 1.33 1.33 1.31
CHS7C70 1496.4 1.19 1.19 1.18
CHS7C100 2057.8 1.23 1.23 1.22
CHS8C0 525.8 1.07 1.19 1.25 1.40 1.00
CHS8C40 1481.9 1.45 1.53 1.56 1.64 1.41
CHS8C70 1740.6 1.23 1.28 1.30 1.34 1.21
CHS8C100 2666.1 1.48 1.52 1.54 1.58 1.45
CHS9C0 456.1 0.96 1.02 1.09 1.17 1.00
CHS9C40 1494.1 1.37 1.40 1.44 1.48 1.39
CHS9C70 1974.4 1.26 1.28 1.31 1.33 1.28
CHS9C100 2797.3 1.38 1.40 1.42 1.44 1.39
CHS10C0 482.8 0.95 0.98 1.00
CHS10C40 1690.2 1.30 1.32 1.32
CHS10C70 2274.2 1.20 1.21 1.21
CHS10C100 3139.2 1.25 1.26 1.27

Mean 1.24 1.29 1.29 1.36 1.17

COV 0.102 0.099 0.112 0.114 0.122
Minimum 0.95 0.98 1.06 1.13 0.98
Maximum 1.48 1.53 1.56 1.70 1.45

Fig. 3. Test setup of concrete-lled aluminum CHS specimen.

Generally, local buckling of aluminum CHS tubes was observed

concrete inll. It is shown that the stiffness of the composite
columns improves. It is also shown that the ductility of concrete-
lled aluminum CHS stub columns decrease with the increase of
the concrete strength, especially for the relatively slender section
of CHS7, as shown in Figs. 4 and 5.
The load-axial strain relationship of the concrete-lled alumi-
num CHS stub columns for series CHS2 and CHS7 obtained from
strain gauge readings are also shown in Figs 4(b) and 5(b),
respectively. For relatively slender CHS, it is shown that the
concrete-lled aluminum CHS stub columns reach the ultimate
loads after the aluminum CHS tubes reach the 0.2% proof stress
(yield stress) for stub columns with nominal concrete cylinder
strengths of 40 and 70 MPa. However, this is not the case for
nominal concrete cylinder strength of 100 MPa. The concrete-
lled aluminum CHS stub columns reach the ultimate loads
before the aluminum CHS tubes reach the yield stress, as shown in
Fig. 5(b). Therefore, the aluminum CHS tubes are not fully utilized
for concrete-lled aluminum slender CHS tube columns with
nominal concrete cylinder strength of 100 MPa.
for relatively slender sections. Fig. 6 shows the failure mode of
column specimen CHS5C40 having a circular hollow section of
100  2 mm2 and nominal concrete cylinder strength of 40 MPa.
The failure mode of column specimen CHS5C70 having a nominal
concrete cylinder strength of 70 MPa is shown in Fig. 7. For this
specimen the concrete crushing was clearly observed.
3. Design rules
3.1. General
The concrete-lled aluminum CHS stub column test strengths
(PExp) are compared with unfactored design strengths predicted
using the general design guidelines specied in the American
specications [7,8] and the Australian/New Zealand standards
[9,10] for aluminum and concrete structures. The design strengths
of concrete-lled aluminum CHS stub columns were obtained by
determining the strength of the aluminum tube (AaFL) using the
 ARTICLE IN PRESS
F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280 1277

350 2500
CHS7C0
300
CHS7C40
2000 CHS7C70
250 CHS7C100
Load (kN)

200 1500

Load (kN)
150
1000
100 CHS2C0
CHS2C40
50 CHS2C70 500
CHS2C100
0
0 2 4 6 8 10 12 0
(mm) 0 2 4 6 8 10 12 14
(mm)
350
2500
300
CHS7C0
CHS7C40
250 2000 CHS7C70
CHS7C100
Load (kN)

200
1500
Load (kN)

150

100 CHS2C0 1000

CHS2C40
CHS2C70
50
CHS2C100 500
0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
Strain (%) 0
0.0 0.5 1.0 1.5 2.0
Fig. 4. The concrete-lled aluminum column test curves for compact section of
CHS2: (a) load versus axial shortening and (b) load versus strain. Strain (%)

Fig. 5. The concrete-lled aluminum column test curves for relatively slender
section of CHS7: (a) load versus axial shortening and (b) load versus strain.

specications [7,9] for aluminum structures as well as the

strength of concrete inll (0.85Acfc) using the specications
[8,10] for concrete structures, as shown in Eq. (1)

P Aa F L 0:85Ac f c (1)
where Aa is the full cross-section area of aluminum tube, FL is the
limit state stress calculated using Sections 3.4.7 through 3.4.10
and Sections 4.7.2 and 4.7.4 of the American (AA) Specication [7],
and Sections 3.4.8 through 3.4.11 and Sections 4.7.2 and 4.7.4
of the Australian/New Zealand Standard [9], Ac is the area of
concrete and fc is the concrete cylinder strength. The design rules
in the AS/NZS Standard [9] for calculating the design strengths of
aluminum columns are generally identical to those in the AA
Specication [7], except that the AS/NZS Standard reduces the
yield load of the column using a coefcient kc which is not
included in the AA Specication. The American Specication [8]
and the Australian Standard [10] for concrete structures generally
use the same formula to calculate the strength of the concrete
inll.
Three design approaches were investigated in the calculation
of the concrete-lled aluminum CHS tube column strengths using
Eq. (1). The calculated unfactored design strengths are denoted by
P1, P2 and P3 for the three design approaches. Fig. 6. Failure mode of column specimen CHS5C40.
 ARTICLE IN PRESS
1278 F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280

mean values of the PExp/PAS/NZS1 and PExp/PAS/NZS2 ratios are 1.29

and 1.36 with the corresponding COV of 0.099 and 0.114 for the
concrete-lled aluminum CHS stub columns. For design approach

1.8

1.6

1.4

1.2

PExp / P
1.0 AA-1
AS/NZS-1
0.8 AA-2
AS/NZS-2
0.6 Approach 3

Fig. 7. Failure mode of column specimen CHS5C70. 0.4

0 20 40 60 80 100 120
Concrete strength (MPa)
3.2. Design approach 1 (P1) Fig. 8. Comparison of test strengths with design strengths for concrete-lled
specimens of series CHS1.
The American and the Australian/New Zealand design
strengths (PAA1 and PAS/NZS1) are calculated using the material
properties obtained from the tensile coupon tests for aluminum 1.8
CHS tubes in the calculation of the term AaFL in Eq. (1). The
measured material properties obtained from the tensile coupon 1.6
tests are shown in Table 2. The calculation of the strength of the
concrete inll for the term 0.85Acfc in Eq. (1) is carried out using 1.4
the measured concrete cylinder strengths, as shown in Table 5.
1.2
PExp / P

3.3. Design approach 2 (P2) 1.0 AA-1

AS/NZS-1
The design approach 2 calculates the term AaFL in Eq. (1), using 0.8 AA-2
the material properties obtained from the stub column tests of the
AS/NZS-2
aluminum tubes, as shown in Table 3. The American and the 0.6
Approach 3
Australian/New Zealand design strengths calculated using Eq. (1)
are denoted by PAA2 and PAS/NZS2, respectively. The strength of 0.4
the concrete inll is calculated in the same way as the design 0 20 40 60 80 100 120
approach 1. Concrete strength (MPa)

Fig. 9. Comparison of test strengths with design strengths for concrete-lled

3.4. Design approach 3 (P3) specimens of series CHS2.

In this approach, the term AaFL in Eq. (1) is replaced by the test
strengths of the aluminum CHS tubes without concrete inll for 1.8
each test series. Hence, the strengths of the aluminum tubes are
114.5, 141.2, 121.3, 113.4, 162.7, 264.5, 283.9, 525.8, 456.1 and 1.6
482.8 kN for series CHS1, CHS2, CHS3, CHS4, CHS5, CHS6, CHS7,
CHS8, CHS9 and CHS10, respectively. Once again, the strength of 1.4
the concrete inll is calculated in the same way as the design
PExp / P

approach 1. The design strengths calculated using Eq. (1) is 1.2

denoted by P3 for design approach 3.
1.0 AA-1
AS/NZS-1
4. Comparison of test strengths with design strengths 0.8 AA-2
AS/NZS-2
The comparison of concrete-lled aluminum CHS stub column 0.6
Approach 3
test strengths (PExp) with design strengths (P1, P2 and P3) is
shown in Table 6. For the American specications [7,8], the mean 0.4
0 20 40 60 80 100 120
values of the PExp/PAA1 and PExp/PAA2 ratios are 1.24 and 1.29
with the corresponding coefcients of variation of 0.102 and 0.112 Concrete strength (MPa)
for the concrete-lled aluminum CHS stub columns, as shown in Fig. 10. Comparison of test strengths with design strengths for concrete-lled
Table 6. For the Australian/New Zealand standards [9,10], the specimens of series CHS3.
 ARTICLE IN PRESS
F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280 1279

1.8 1.8

1.6 1.6

1.4 1.4
PExp / P

1.2 1.2

PExp / P
1.0 1.0
AA-1 AA-1
AS/NZS-1 AS/NZS-1
0.8 0.8
AA-2 AA-2
AS/NZS-2 AS/NZS-2
0.6 0.6
Approach 3 Approach 3
0.4 0.4
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Concrete strength (MPa) Concrete strength (MPa)

Fig. 11. Comparison of test strengths with design strengths for concrete-lled Fig. 14. Comparison of test strengths with design strengths for concrete-lled
specimens of series CHS4. specimens of series CHS7.

1.8 1.8

1.6 1.6

1.4 1.4

1.2 1.2
PExp / P

PExp / P

1.0 1.0
AA-1 AA-1
AS/NZS-1 AS/NZS-1
0.8 0.8
AA-2 AA-2
AS/NZS-2 AS/NZS-2
0.6 0.6
Approach 3 Approach 3
0.4 0.4
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Concrete strength (MPa) Concrete strength (MPa)
Fig. 12. Comparison of test strengths with design strengths for concrete-lled Fig. 15. Comparison of test strengths with design strengths for concrete-lled
specimens of series CHS5. specimens of series CHS8.

1.8 1.8

1.6 1.6

1.4 1.4

1.2 1.2
PExp / P

PExp / P

1.0 1.0
AA-1 AA-1
AS/NZS-1 AS/NZS-1
0.8 0.8
AA-2 AA-2
0.6 AS/NZS-2 0.6 AS/NZS-2
Approach 3 Approach 3
0.4 0.4
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Concrete strength (MPa) Concrete strength (MPa)

Fig. 13. Comparison of test strengths with design strengths for concrete-lled Fig. 16. Comparison of test strengths with design strengths for concrete-lled
specimens of series CHS6. specimens of series CHS9.

3, the mean value of the PExp/P3 ratio is 1.17 with the The comparison of the composite column test strengths (PExp)
corresponding COV of 0.122 for the concrete-lled aluminum with design strengths (P1, P2 and P3) are also plotted against
CHS stub columns. the measured concrete cylinder strengths, as shown in Figs. 817.
 ARTICLE IN PRESS
1280 F. Zhou, B. Young / Thin-Walled Structures 47 (2009) 12721280

1.8 American specications and Australian/New Zealand standards

for aluminum and concrete structures. Three design approaches
1.6 were used in the calculation of the column strengths. The material
properties of the aluminum CHS tube specimens obtained from
1.4 the tensile coupon tests and stub column tests were used to
calculate the design strengths. The test strengths of aluminum
1.2 tubes without concrete inll were also used in the calculation of
PExp / P

the design strengths. It is shown that the three design approaches

1.0
AA-1
0.8 AS/NZS-1
AA-2
0.6 AS/NZS-2
Approach 3
0.4
0 20 40 60 80 100
Concrete strength (MPa)
Fig. 17. Comparison of test strengths with design strengths for concrete-lled
specimens of series CHS10.
The PExp/P ratio is shown on the vertical axis, while the measured
concrete cylinder strength is shown on the horizontal axis of Figs.
817. It is shown that the three design approaches using American
and AS/NZS specications are generally conservative for the
concrete-lled aluminum CHS stub columns having nominal
concrete cylinder strengths of 40, 70 and 100 MPa with the D/t
ratio of the CHS tubes ranging from 9.7 to 59.7.
5. Conclusions
A test program on concrete-lled aluminum circular hollow
section (CHS) stub columns has been presented. A series of tests
was performed subjected to uniform axial compression to
investigate the structural performance of the concrete-lled
aluminum CHS stub columns. The diameter-to-thickness ratio of
the aluminum CHS tubes ranged from 9.7 to 59.7. Various concrete
cylinder strengths of 40, 70 and 100 MPa were investigated. A total
of 42 column specimens were tested. The column strengths, load-
axial shortening relationship, load-axial strain relationship and
failure modes of the columns have been presented.
The concrete-lled aluminum CHS stub column test strengths
were compared with the design strengths predicted using the
using the American and the Australian/New Zealand specications
are generally conservative for the concrete-lled aluminum CHS
stub columns.
Acknowledgements
120
The authors gratefully acknowledge the Asia Aluminum
Manufacturing Company for supplying the test specimens. The
authors are also grateful to Mr. Hoi-Yuen Chan and Mr. Kam-Hung
Chan for their assistance in the experimental program as part of
their nal year undergraduate research projects at The University
of Hong Kong.
References
[1] Yu Q, Tao Z, Wu YX. Experimental behaviour of high-performance concrete-
lled steel tubular columns. Thin-Walled Structures 2008;46:36270.
[2] Han LH, Yao GH, Chen ZB, Yu Q. Experimental behaviours of steel tube
conned concrete (STCC) columns. Steel and Composite Structures 2005;5(6):
45984.
[3] Han LH, Liu W, Yang YF. Behaviour of concrete-lled steel tubular stub
columns subjected to axially local compression. Journal of Constructional
Steel Research 2008;64(4):37787.
[4] Schneider SP. Axially loaded concrete-lled steel tubes. Journal of Structural
Engineering, ASCE 1998;124(10):112538.
[5] Zhou F, Young B. Tests of concrete-lled aluminum stub columns. Thin-
Walled Structures 2008;46:57383.
[6] Hu HT, Huang CS, Wu MH, Wu YM. Nonlinear analysis of axially loaded
concrete-lled tube columns with connement effect. Journal of Structural
Engineering, ASCE 2003;129(10):13229.
[7] Aluminum Association. Aluminum design manualPart I: specication for
aluminum structures. Washington, D.C.; 2005.
[8] ACI. Building code requirements for structural concrete and commentary. ACI
318-08. Detroit (USA): American Concrete Institute; 2008.
[9] Australian/New Zealand Standard. Aluminium structuresPart 1: limit state
design. AS/NZS 1664.1:1997, Standards Australia, Sydney, Australia, 1997.
[10] Australian Standard AS3600. Concrete structures. AS3600-2001. Sydney
(Australia): Standards Australia; 2001.